S.G. Development & Regeneration

Seunghee Cha Editor
Salivary Gland
Development and
Regeneration
Advances in Research and

Clinical Approaches to
Functional Restoration
123
Salivary Gland Development
and Regeneration
Seunghee Cha
Editor
Salivary Gland
Development and
Regeneration
Advances in Research and Clinical
Approaches to Functional
Restoration
Editor
Seunghee Cha
Oral and Maxillofacial Diagnostic Sciences
University of Florida
Gainesville, Florida
USA
ISBN 978-3-319-43511-4 ISBN 978-3-319-43513-8 (eBook)

DOI 10.1007/978-3-319-43513-8
Library of Congress Control Number: 2017933465
© Springer International Publishing Switzerland 2017

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software,
or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are
exempt from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in
this book are believed to be true and accurate at the date of publication. Neither the publisher nor
the authors or the editors give a warranty, express or implied, with respect to the material
contained herein or for any errors or omissions that may have been made. The publisher remains
neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
Rome was not built in a day, as the English playwright John Heywood
famously wrote. Innovation and advancement in the field of salivary gland
regeneration is one of the great examples that reflects this sentiment. The first
research article available on this topic through the US National Library of
Medicine dates back to 1950. The article, entitled “Regeneration in the
Submaxillary Gland of the Rat,” by B.B. Milstein in 1950, cites Van
Podwyssozki as the first to describe regeneration of organs of small animals
as long ago as 1886 and regeneration of the salivary glands (Die Regeneration
an den Speicheldrusen) in 1887.
Since the 1950s, an ever-expanding literature and diversified approaches
aimed at functional restorations have mirrored strong interest and attention to
this particular subject of research. Journal articles dealing with autoimmune
Sjögren’s syndrome, effects of radiation, and ductal ligation models in rats
and mice appeared in the early 1980s, followed by research on neural regula-
tion of secretion and effectiveness of epidermal growth factor in wound heal-
ing models and glandular regeneration in the late 1980s through the
mid-1990s.
In 1995, the late Dr. Michael Humphreys published his well-known review
article entitled “Saliva and growth factors: the fountain of youth resides in us
all,” which emphasized the vital importance of growth factors in oral/sys-
temic health and glandular repair/regeneration. Histological analyses of glan-
dular architecture and development were established by Dr. Robert Redman,
the author of Chap. 4 of this volume. With the turn of a new century, molecu-
lar and cellular mechanisms of branching morphogenesis and glandular
development were further investigated and pioneered by Drs. Kenneth
M. Yamada and Matthew P. Hoffman, whose work provided foundations for
the application of tissue engineering concepts and methodologies to salivary
regeneration. Outstanding contributions by Dr. Bruce J. Baum to the field of
tissue engineering and gene therapy have ultimately been solidified in appli-
cations such as clinical trials involving AAV2-mediated human aquaporin-1
delivery in recent years. Investigation of ductal ligation models, irradiation
models, and Sjögren’s syndrome NOD models dominated interest in the field
until around 2010, when stem cell research in vitro and in vivo reignited
research interest and passion in salivary gland regeneration.
In the current era, the authors and coauthors in this book, who are renowned
researchers, dentists, and surgeons in the field, have spearheaded efforts to
discover the underlying pathogenesis of xerostomia and innovative approaches
v
vi Preface
to restore secretory function. I am proud to present their collective efforts and

years of their research outcomes revealed in their book chapters, which will
establish another significant milestone in the history and tradition of studies
on glandular regeneration.
This book begins with the description of fundamental and molecular pro-
cesses occurring during salivary gland organogenesis/branching morphogen-
esis and molecular communications among epithelial, mesenchymal,
endothelial, and neuronal cells for cellular differentiation and organ develop-
ment (Chap. 1, Dr. Lombaert). The importance of understanding the commu-
nications and simulating optimal environments in glandular repair and
regeneration is further discussed under Future Prospects.
With rapidly advancing biotechnology, the application of systems biology
has become an indispensable tool in this field. Chapter 2 discusses the defini-
tion and applications of systems biology for glandular tissues and saliva sam-
ples (Chap. 2, Dr. Larson et al.). Systems biology approaches in conjunction
with traditional approaches unveil the complex molecular, cellular, and phys-
ical processes in development, disease processes, and regenerative medicine
involving the salivary glands.
One of the underappreciated subjects in the field is the important role of a
large family of mucins in oral health. In Chap. 3, the authors summarize the
main structural and functional characteristics of salivary mucins, their expres-
sion patterns during salivary gland development and regeneration, and quali-
tative and quantitative changes in pathological processes in the salivary
glands due to irradiation, autoreactive immune cells, neoplasm, or inflamma-
tion (Chap. 3, Dr. Castro et al.).
Changes due to radiation are not limited to mucin expression profiles but
are also manifested in the parenchymal and stromal structures in the salivary
glands. These changes are detailed in Chap. 4 with photomicrographs and
transmission electron micrographs of rat and human salivary glands (Chap. 4,
Dr. Redman). Understanding the damage occurring in the glands before and
after radiation therapy will expedite the development of intervention strate-
gies to protect the salivary glands from the harmful radiation.
In Chap. 5, Dr. Tran’s group reviews recent advances from the years 2010
to 2015 in the treatment of salivary gland hypofunction with a special empha-
sis on mesenchymal stem cells (Chap. 5, Dr. Tran et al.). This chapter covers
in detail adipose tissue-derived stromal cells, mesenchymal stromal cells
derived from various sources, and finally the authors’ experience with the
soluble contents/factors in bone marrow soup extracted from a whole bone
marrow cell lysate.
As differentiation-inducing factors are crucial for initiating stem cell dif-
ferentiation from the state of quiescence, these extrinsic and intrinsic factors
(transcription factors) involved in pancreas, liver, and salivary gland regen-
eration are further detailed in Chap. 6 with a focus on directed-cell differen-
tiation and transdifferentiation (Chap. 6, Drs. Park and Cha).
Current cell models for bioengineering of the salivary glands are presented
in Chap. 7, along with the pros and cons of utilizing various salivary cell
lines. Practical tips on cell isolation and culture techniques in conjunction
with the use of scaffolds complement the use of stem, progenitor, and acinar
Preface vii
cells for salivary gland regeneration. Current trends in salivary gland bioen-
gineering deliver great promise in functional restoration of the salivary glands
(Chap. 7, Dr. Baker).
To explore further the subject of bioengineering, factors and elements
needed for successful development of a functional salivary gland are dis-
cussed in detail in Chap. 8, emphasizing the dynamic nature of the basement
membrane and the significance of the extracellular matrix and cell polarity
in salivary gland development and reconstruction. In addition, studies utiliz-
ing the salivary-derived stem cells/gland progenitor and three-dimensional
(3D) biomimetic scaffolds encompassing decellularization methods, various
matrices, and polymers are summarized for 3D culture technique, which
underpins current knowledge on bioengineering of the salivary glands
(Chap. 8, Martinez et al.).
3D printing technology creates life-size body parts and tissues using living
cells as the ink. This technology has revolutionized the field of regenerative
and reconstructive medicine, enabling customized and personalized thera-
peutic approaches. In Chap. 9, Dr. Choi et al. describe basic principles and
different types of 3D technologies, patient-specific modeling, bioprinting,
and salivary gland regeneration (Chap. 9, Dr. Choi et al.).
A novel bioengineering method involves epithelial and mesenchymal stem
cell manipulation to generate a bioengineered organ germ. In Chap. 10,
Dr. Ogawa explains that the bioengineered glandular germs demonstrated
reciprocal interactions between epithelial and mesenchymal cells in one day
and invagination of epithelial tissue in three days in vitro. Once the germ was
engrafted into the parotid gland duct of salivary gland-defective mice, the
connection between the germ and the duct was established in a month, and
the mice exhibited restored salivary secretion after transplantation. This inno-
vative approach emphasizes that current advancement in the field promises a
therapeutic intervention for patients suffering from xerostomia (Chap. 10,
Drs. Ogawa and Tsuji).
Currently, functional restoration of the salivary glands is still challenging
to accomplish even with successful reconstruction of salivary cellular compo-
nents. Therefore, understanding the mechanisms of saliva secretion becomes
critical for positive clinical outcomes that we desire. Chapter 11 covers con-
siderations for establishing functional secretion by providing information on
stimuli for secretion, neural connection along with neurotransmitters and
receptors, protein secretion, and studies of neural agonists and antagonists.
The chapter also clarifies myths surrounding this topic with recent research
data (Chap. 11, Drs. Carpenter and Carvalho).
Thought-provoking renderings of the past, current, and future of gene
therapy in salivary gland diseases are provided by Dr. Passineau in Chap. 12.
In this chapter, current challenges in the field of salivary gland gene therapy,
along with the author’s proposals to circumvent or overcome the hurdles, are
forthrightly discussed (Chap. 12, Dr. Passineau).
Last, but not least, the chapter on surgical management of salivary gland
disease reveals the critical considerations for glandular regeneration from the
perspectives of otolaryngologists and surgeons (Chap. 13, Drs. Varadarajan
and Dziegielewski). The extensive description in this chapter includes, but is
viii Preface
not limited to, glandular anatomy, pathology, surgical advances for neoplastic
and nonneoplastic diseases of salivary glands, and recent discoveries in the
field such as salivary gland transfer and salivary duct repositioning. The
importance of understanding the expected sequelae in human patients follow-
ing radiation or surgery cannot be overemphasized as none of the existing
laboratory approaches would come to fruition for patients without such
knowledge.
Based on the cutting-edge information offered in this book, it is undoubt-
able that many more innovative strategies for salivary gland regeneration will
emerge in upcoming years. Research that unlocks the complex processes of
organ development would be fundamental to develop such approaches. With
the current enthusiasm and growing interest in the field, it will just be a matter
of time before we build another Rome.
Gainesville Seunghee Cha, DDS, PhD

FL, USA
Contents
Part I Updates on Salivary Gland Development
1 Implications of Salivary Gland Developmental Mechanisms

for the Regeneration of Adult Damaged Tissues. . . . . . . . . . . . . . . 3
Isabelle M.A. Lombaert
2 Systems Biology: Salivary Gland Development, Disease,
and Regenerative Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Melinda Larsen, Petko Bogdanov, Ravi Sood,
Hae Ryong Kwon, Deirdre A. Nelson, Connor Duffy,
Sarah B. Peters, and Sridar V. Chittur
3 Mucins in Salivary Gland Development,
Regeneration, and Disease. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
Isabel Castro, María-José Barrera, Sergio González,
Sergio Aguilera, Ulises Urzúa, Juan Cortés
and María-Julieta González
Part II Glandular Damage and Cell Replacement Therapy
4 Histologic Changes in the Salivary Glands

Following Radiation Therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
Robert S. Redman
5 Adult Stem Cell Therapy for Salivary Glands,
with a Special Emphasis on Mesenchymal Stem Cells . . . . . . . . . 93
Simon D Tran, Yoshinori Sumita, Dongdong Fang,
and Shen Hu
6 Directed Cell Differentiation by Inductive Signals
in Salivary Gland Regeneration: Lessons Learned
from Pancreas and Liver Regeneration . . . . . . . . . . . . . . . . . . . . 103
Yun-Jong Park and Seunghee Cha
Part III Bioengineering of Salivary Glands
7 Current Cell Models for Bioengineering Salivary Glands . . . . . 133

Olga J. Baker
ix
x Contents
8 Matrix Biology of the Salivary Gland:

A Guide for Tissue Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Mariane Martinez, Danielle Wu, Mary C. Farach-Carson,
and Daniel A. Harrington
9 3D Printing Technology in Craniofacial
Surgery and Salivary Gland Regeneration. . . . . . . . . . . . . . . . . . 173
Jong Woo Choi, Namkug Kim, and Chang Mo Hwang
10 Functional Salivary Gland Regeneration
by Organ Replacement Therapy. . . . . . . . . . . . . . . . . . . . . . . . . . 193
Miho Ogawa and Takashi Tsuji
Part IV Therapeutic Considerations for Restoration of

Salivary Function
11 Regulation of Salivary Secretion. . . . . . . . . . . . . . . . . . . . . . . . . . 207

Guy Carpenter and Polliane Carvalho
12 Salivary Gland Gene Therapy in Experimental
and Clinical Trials. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
Michael Passineau
13 Surgical Management of Salivary Gland Disease . . . . . . . . . . . . 229
Varun V. Varadarajan and Peter T. Dziegielewski
Contributors
Sergio Aguilera, MD INDISA Clinic, Santiago, Chile

Olga J. Baker, DDS, PhD School of Dentistry, University of Utah, Salt
Lake City, UT, USA
María-José Barrera, PhD Institute of Biomedical Sciences, Faculty of
Medicine, University of Chile, Santiago, Chile
Petko Bogdanov, PhD Department of Computer Science, University at
Albany, SUNY, Albany, NY, USA
Guy Carpenter, PhD Mucosal and Salivary Biology Division, King’s
College London Dental Institute, London, UK
Polliane Carvalho, MSc, PhD Mucosal and Salivary Biology Division,
King’s College London Dental Institute, London, UK
Isabel Castro, MSc, PhD Institute of Biomedical Sciences, Faculty of
Seunghee Cha, DDS, PhD Department of Oral and Maxillofacial
Diagnostic Sciences, University of Florida College of Dentistry, Gainesville,
USA
Sridar V. Chittur, PhD Center for Functional Genomics, University at
Albany, SUNY, Albany, NY, USA
Jong-Woo Choi, MD, PhD, MMM Department of Plastic and
Reconstructive Surgery, Ulsan University College of Medicine, Asan
Medical Center, Seoul, South Korea
Juan Cortés, PhD Institute of Biomedical Sciences, Faculty of Medicine,
University of Chile, Santiago, Chile
Connor Duffy Biological Sciences, University at Albany, SUNY, Albany,
NY, USA
Peter T. Dziegielewski, MD, FRCS(C) University of Florida Department
of Otolaryngology, Gainesville, FL, USA
Dongdong Fang, BDS, MSc McGill Craniofacial Tissue Engineering and
Stem Cells Laboratory, Department of Biomedical Sciences, Faculty of
Dentistry, McGill University, Montreal, Quebec, Canada
xi
xii Contributors
Mary C. Farach-Carson, PhD BioScience Research Collaborative,

Department of Biochemistry and Cell Biology, Houston, TX, USA
Departments of BioSciences and Bioengineering, Rice University, Houston,
TX, USA
María-Julieta González, MSc Institute of Biomedical Sciences, Faculty of
Sergio González, MSc Mayor University, Santiago, Chile
Daniel A. Harrington, PhD Departments of BioSciences and
Bioengineering, Rice University, Houston, TX, USA
Shen Hu, PhD, MBA Division of Oral Biology and Medicine, School of
Dentistry, University of California, Los Angeles, CA, USA
Chang Mo Hwang, PhD Biomedical Engineering Research Center,
University of Ulsan College of Medicine, Asan Medical Center, Seoul,
South Korea
Namkug Kim, PhD Department of Convergence Medicine, Biomedical
Engineering Research Center, University of Ulsan College of Medicine,
Asan Medical Center, Seoul, South Korea
Hae Ryong Kwon, PhD Biological Sciences, University at Albany, SUNY,
Albany, NY, USA
Oklahoma Medical Research Foundation, Oklahoma City, OK, USA
Melinda Larsen, PhD Biological Sciences, University at Albany, SUNY,
Albany, NY, USA
Isabelle M.A. Lombaert, PhD University of Michigan, School of
Dentistry, Ann Arbor, MI, USA
Biointerfaces Institute, Ann Arbor, MI, USA
Mariane Martinez Department of BioSciences, Rice University, Houston,
TX, USA
Deirdre A. Nelson, PhD Biological Sciences, University at Albany, SUNY,
Albany, NY, USA
Miho Ogawa, PhD Organ Technologies Inc., Tokyo, Japan
Laboratory for Organ Regeneration, RIKEN Center for Developmental
Biology, Kobe, Hyogo, Japan
Yun-Jong Park Department of Pharmacology, The University of North
Carolina at Chapel Hill, Chapel Hill, NC, USA
Michael Passineau, PhD Gene Therapy Program, Allegheny Health
Network, Pittsburgh, PA, USA
Sarah B. Peters, PhD Biological Sciences, University at Albany, SUNY,
Albany, NY, USA
Department of Cell, Developmental, and Integrative Biology, University of
Alabama at Birmingham, Birmingham, AL, USA
Contributors xiii
Robert S. Redman, DDS Oral and Maxillofacial Diagnostic Science,

University of Florida, Gainesville, FL, USA
Washington DC VA Medical Center, Washington, DC, USA
Ravi Sood Department of Computer Science, University at Albany, SUNY,
Albany, NY, USA
Yoshinori Sumita, DDS, PhD Department of Regenerative Oral Surgery,
Nagasaki University, Nagasaki, Japan
Simon D. Tran, DMD, PhD McGill Craniofacial Tissue Engineering and
Stem Cells Laboratory, Department of Biomedical Sciences, Faculty of
Dentistry, McGill University, Montreal, Quebec, Canada
Takashi Tsuji, PhD Organ Technologies Inc., Tokyo, Japan
Laboratory for Organ Regeneration, RIKEN Center for Developmental
Biology, Kobe, Hyogo, Japan
Ulises Urzúa, PhD Institute of Biomedical Sciences, Faculty of Medicine,
University of Chile, Santiago, Chile
Varun V. Varadarajan, MD University of Florida Department of
Otolaryngology, Gainesville, FL, USA
Danielle Wu, PhD Department of BioSciences, Rice University, Houston,
TX, USA
Part I
Updates on Salivary Gland Development
Implications of Salivary Gland
Developmental Mechanisms 1
for the Regeneration of Adult
Damaged Tissues
Isabelle M.A. Lombaert
Abstract
The convergence of the fields of tissue engineering and regenerative
medicine provides a potential blueprint to repair damaged tissues.
Accordingly, a range of therapeutic applications have emerged that hold
great potential to regenerate branching organs, such as salivary glands.
This unique saliva-secreting organ is required for proper oral health,
lubrication, immunity, and food digestion but is susceptible to damage
either by co-irradiation as a side effect of radiotherapy cancer treatment,
autoimmune-related Sjögren syndrome, disease-related medications, or
surgical resection. This chapter focuses on fundamental cellular and
molecular processes occurring during organ ontogenesis and in develop-
ing branching glands. We cover the growth of the epithelial compart-
ment, which is the major functional component of the gland, but also
how surrounding niches such as mesenchymal, endothelial, and neuronal
cells communicate, intertwine, and influence the formation of glands
and other branching organs. Finally, we highlight how this key informa-
tion has created new regenerative-related approaches and how these
impact future clinical translation.
1.1 Introduction
Increasing our knowledge of how organs develop

has profound implications for the design of thera-
pies to regrow and/or repair injured tissues.
I.M.A. Lombaert, PhD
University of Michigan, School of Dentistry,
Understanding the mechanisms regulating cell
2800 Plymouth Road, Ann Arbor, MI 48109, USA survival, expansion, specification, movement,
Biointerfaces Institute, 2800 Plymouth Road,
communication with neighboring cells, as well as
Ann Arbor, MI 48109, USA how they respond to damage is critical to navigat-
e-mail: [email protected] ing the landscape of future therapy designs.
© Springer International Publishing Switzerland 2017 3

S. Cha (ed.), Salivary Gland Development and Regeneration, DOI 10.1007/978-3-319-43513-8_1
4 I.M.A. Lombaert
In order to appropriately translate informa- o utlined in depth in the following chapters. In

tion gathered from studies on organ develop- this review, we focus on different salivary gland
ment, we need to compare molecular and cellular cell types and their supportive environment that
processes during embryonic development with is needed to form the fully functional secretory
adult homeostasis and when repair initiates and/ branching organ. Subsequently, we outline how
or fails after each damaging event. Each of these this knowledge can render future therapeutic
stages correlates with specific cellular responses, implications and/or what potential complications
activation of specific signaling pathways, and might arise.
accumulation of environmental cues. Thus,
developmental-related information is instrumen-
tal to stimulate regrowth within an existing dam- 1.2 pithelial Growth Driven
E
aged in vivo organ or to initiate de novo growth. by Stem Cells
The majority of our current knowledge on
salivary gland organogenesis derives from exper- Branching organs such as salivary, lacrimal, and
imental animal models, primarily mice and rats. mammary glands are comprised of different cell
While rodent biology is not identical to that of types, including epithelial and the surrounding
humans, many processes and pathways are very mesenchymal, endothelial, and neuronal cells
similar. As such, developmental biologists have (Fig. 1.1). Intertwined within these tissues are
been and continue to be a valuable resource to circulating hematopoietic-related blood and
other disciplines such as engineering, oral sur- immune cells. The major component of develop-
gery, and oncology to translate conceptual ideas ing and adult salivary glands (SGs) is the epithe-
into therapeutic designs. lia, which is responsible for saliva secretion and
The advantages of specific biomaterials, gene transportation to the oral cavity. Here, we
therapy, and surgical in vivo approaches are describe how the epithelial compartment of three
a c
Fig. 1.1 Developing salivary glands in mice. (a) Bright (arrow). Confocal image of E-cadherin stained epithelia
field picture represents E13 submandibular (SMG) and and Tubbulin-3 stained PSG. (c) Different niches sur-
sublingual gland (SLG). The epithelial compartment is rounding the epithelium in adult mouse submandibular
comprised of a distal endbud and proximal duct area. (b) gland. Confocal 30 μm projected image of stained SMG
Epithelia (blue) innervated by the parasympathetic with epithelial marker E-cadherin (blue), neuronal
nerves (PSG, red) during embryonic SMG development. marker Tubbulin-3 (red), and endothelial protein CD31
The PSG releases neurotransmitters via varicosities (green)
1 Implications of Salivary Gland Developmental Mechanisms for the Regeneration of Adult Damaged Tissues 5
major glands, which provide 90 % of total saliva, they are not yet fully functional and must still
becomes established by tightly controlled mecha- undergo specific acinar-lineage maturation so that
nisms of cellular interactions. by birth, in both humans and mice, the organ is
comprised of functional secretory compartments.
1.2.1 Morphological Development

of Salivary Glands 1.2.2 C
ontribution of Stem Cells
and Their Differentiating
Salivary glands originate as an invagination of the Progeny
oral epithelium from a placode at embryonic day
(E) 11.5 in mice or Carnegie stage 18 (~44 days, The stem/progenitor cell theory asserts that all
weeks 6–7) in humans. This thickening epithe- salivary gland cells are initiated from and main-
lium arises on the side of the tongue outside of the tained by stem and progenitor cells. By defini-
lamina dentalis at the anlage of the dental arch. tion, these cells are characterized by their ability
Each major gland initiates at slightly different to expand themselves, i.e., self-renew, as well as
locations: the serous parotid gland (PAR) in the to propagate multiple more defined cell types,
labiogingival sulcus, the mucous sublingual such as acinar cells. Stem cells are further clas-
(SLG) in the paralingual sulcus, and seromucous sified as being more potent than progenitor cells
submandibular gland (SMG) in the linguogingival in their self-renewal and differentiation potential.
sulcus. Even though glands arise in the tongue When both these cellular processes are tightly
area, they grow out during development toward controlled, stem/progenitor cells not only give
the back of the mouth below the ears, floor of the rise to tissues but also maintain and repair organ
mouth near the mandibular bone, and the anterior structures during adulthood. Any deregulation in
floor of the mouth. While in mice, SMG, SLG, this regulatory network during development can
and PAR initiate around E11.5, E12, and E13, lead to malformation/absence of the organ and in
respectively, human SLGs initiate later than SMG the adult may cause cancer formation. Over the
and PAR at the ninth embryonic month. More past years, remarkable progress has been made
detailed descriptions on anatomic locations of wherein multiple stem/progenitors have been
human glands have been recently reviewed [41]. classified based on their ability to (1) form mul-
The developmental origin of each gland has not tiple cell types (mouse genetic lineage tracing,
been clear, with some classifying the PAR as ecto- ex vivo culturing), (2) alternate quiescence with
dermal derived and SMG and SLG as endoder- proliferation (BrdU incorporation or genetically
mal, while genetic experiments in mice suggest labeled DNA tracing), and (3) restore radiation-
they are all ectodermal [87]. induced damaged SGs (in vivo transplantation
Once the epithelial thickening arises, a cell assay). One new consensus gathered from this
population, termed endbud or tip, forms dis- data is that different stem/progenitor cells con-
tally from an elongating stalk (Fig. 1.1a), which tribute to the growing SG and that these cells
developmentally progresses to form major ducts may originate at different time points during
termed Wharton’s (SMG), Bharton’s (SLG), and development. Importantly, this permits the organ
Stensen’s (PAR) ducts. The unique SG branch- to compensate for any losses in specific stem/pro-
ing pattern is created by repetitive clefting of the genitor cells and still allows proper development
initial and subsequently formed endbuds. Ductal [88]. Known stem/progenitors contributing to
structures gradually mature by elongation, lumen SG organogenesis include cells marked by their
formation, and expansion. The clefting endbuds expression of intracellular cytokeratin 5 (CK5,
mature by E16 in mice and 19–24 weeks (7th K5) and CK14 [50, 57], receptors KIT (c-Kit,
month) in humans to form polarized pro-acinar CD117) and FGFR2b [57], and transcription fac-
and pro-myoepithelial cells. While pro-acinar tors SOX2 [3] and ASCL3 [10]. Remarkably,
cells do express some secretory-related proteins, stem/progenitors contributing to development
6 I.M.A. Lombaert
Fig. 1.2 Signaling
pathways influencing
epithelial growth. Cartoon
represents known signaling
pathways that influence SG
epithelial cell survival,
proliferation, expansion,
and differentiation. Green:
expressed by epithelial
cells; orange: expressed by
mesenchyme; red:
expressed by neuronal
cells; black: expression by
multiple compartments
might not serve a similar role during adult to significantly increased saliva levels [58, 62, 72,
homeostasis. Recent studies observed active pro- 103]. This does not, however, exclude the poten-
liferation of cells within specific compartments, tial of other SG-specific epithelial cells, non-SG
such as acini and intercalated, striated, and excre- specific cells, and/or their bioactive cell lysate to
tory ducts, wherein these cells self-duplicate to contribute to the repair of damaged SGs. These
replenish their own entity, as reviewed in [4]. To options will be surveyed in following chapters,
what extent these adult compartmental “reser- and their impact on when to use them in different
voir” cells contribute to recovery after injury is damaging situations has been recently reviewed
a focus of ongoing research. At least after severe [61]. In this chapter, we will further outline our
radiation-induced damage, which leads to irre- current understanding of how SGs are structur-
versible hyposalivation, there is no active repair ally built by various cell types (Fig. 1.1b, c) and
initiated by remaining SG cells. This is often a how their continuous interactions are informing
combinatorial result of (a) drastic loss of acinar the design of current and future therapies.
and duct “reservoir” cells or stem/progenitor
cells, (b) decrease in signaling pathways required
to activate surviving “reservoir” or stem/pro- 1.2.3 L
essons from Developmental
genitor cells, and/or (c) severely damaged cells Regulatory Mechanisms
that can no longer contribute to self-duplication Guiding the Epithelium
or differentiation. In such cases, multiple strat-
egies ranging from constructing a new gland to Often disorders in humans and genetic rodent
gene therapy and stem/progenitor cell transplan- model systems can provide critical information
tations may aid in restoring the functional and on what signaling pathways are essential for epi-
morphological components of the gland. Thus thelial cell survival, proliferation, differentiation,
far, transplantations of cells selected for their and movement (Fig. 1.2). Major examples are
expression of receptor KIT, EPCAM, CD24, and/ Fgf10−/− and Fgfr2b−/− mice, which are related to
or CD29 (Integrin β1, ITGβ1) were shown to human loss-of-function mutations in FGF10 and
restore acinar and ductal compartments, leading FGFR2 that result in hereditary diseases including
lacrimal and SG-related aplasia of lacrimal and branching via epithelial ITGβ1 [84]. MMPs also
salivary glands (ALSG), lacrimo-auriculo-dento- cleave pro-HB-EGF into an N-terminal and
digital (LADD) syndrome, and lung- related C-terminal fragment at the membrane so that the
chronic obstructive pulmonary disease (COPD). latter fragment can move to the nucleus to acti-
In these conditions, SG development is stalled, as vate cell proliferation via cyclin A [95].
FGFR2b+ epithelium no longer receives survival Conversely, another member of the EGF family,
and proliferative cues from the surrounding neuregulin (NRG), binds ERBB3 on endbuds to
FGF10-producing mesenchyme [80]. Thus, when aid in their local expansion [70]. NRG1 is further
invading oral epithelial cells at gland ontogenesis essential for innervation as Nrg1−/− mice are
receive FGF10, they initiate an endbud and duct devoid of nerves and show aberrant duct forma-
formation. From then on, FGFR2b signaling tion and lumenization [73].
expands KIT+ progenitors in the continuously Similar to FGF10, Fgf8 hypomorphic and
clefting endbuds in combination with stem cell Fgfr2c heterozygous mice exhibit hypoplastic
factor (SCF)/KIT signaling [57]. As FGF10 has a glands due to reduced communication between
heparan-binding (HB) core, it evokes and expands FGF8-producing epithelia and FGFR2c-receiving
more rapid responses once it is bound to specific mesenchyme. In both FGF-deficient mice, initial
3-O-sulfated heparin sulfate (3-O-HS). This HS epithelial invagination occurs, but subsequent SG
belongs to a group of heparan sulfate proteogly- growth does not occur. To date, FGF8 has been
cans (HSPGs) located in the basement membrane described as a potential target of the EDA path-
or at cell surfaces. Interestingly, KIT+ endbud way. Human mutations in ectodysplasin-A
progenitors highly express HS3ST3, the 3-O-HS- (EDA) or its receptor EDAR result in hypohi-
specific modifying enzyme 3-O-sulfotransferase, drotic ectodermal dysplasia (HED). Defects in
to rapidly increase their expansion during devel- teeth, hair, sweat, and salivary glands are notice-
opment [80]. A similar function remains during able due to reduced cell proliferation and differ-
adult homeostasis. Regulating this FGFR2b sig- entiation [45, 74]. In SGs, EDA and downstream
naling pathway is of crucial importance so that target NF-kB aid in ductal lumenization and end-
every epithelial cell does not undergo extensive bud branching, presumably by inducing ductal
proliferation. Ductal cells therefore express FGF maturation within the center of endbuds. Early
antagonists, Sprouty 1 and 2, to lower FGFR2b on, mesenchyme-produced EDA is downstream
signaling and upregulate WNT [48]. Both canon- of mesenchymal WNT and upstream of epithelial
ical WNT/β-catenin and noncanonical WNT5b SHH (sonic hedgehog) signaling. As such, SHH
pathways drive ductal formation via upregulation treatment can rescue SGs deprived of EDA [100].
of Tfcp2l1 while inhibiting endbud development. After E13, EDA does not seem to correlate with
In turn, endbuds repress duct development by WNT anymore, based on their different expres-
FGF-mediated Wnt5b repression and secretion of sion pattern located in the epithelial or mesen-
WNT ligand-sequestering protein SFRP1 [78]. chymal compartment [31, 78]. SHH’s important
Evidently, a tight FGF-WNT gradient allows for role in SG development has been confirmed, as
KIT+ progenitor expansion in endbuds, while SHH-deficient SGs are hypoplastic with unpolar-
ductal cells prepare for upcoming lumenization ized epithelial cells and underdeveloped lumen
and maturation. In this process, ERBB1 (EGFR)+ formation [36, 43]. SHH is also linked to FGF8
ductal K5+ progenitors proliferate in response to as both can upregulate each other [43]. Therefore,
HB-EGF to give rise to maturing K19+ cells [50]. FGF8 is able to rescue Hedgehog inhibition but
One mechanism of action is via induction of surprisingly not EDA deficiency [43]. As such,
membrane-type-2 matrix metalloproteinase EDA-FGF8’s precise signaling interaction still
(MT2-MMP) and FGFR expression in epithelial needs to be determined.
cells. MT2-MMP is crucial to release bioactive FGF signaling, in particular via FGFR1 (vari-
NC1 domains from extracellular matrix (ECM) ant b in the epithelium and c in the mesenchyme),
protein collagen IV, which in turn promotes can also upregulate bone morphogenetic protein
8 I.M.A. Lombaert
(BMP) ligands. BMPs are part of the TGFβ sig- spatial proliferation, differentiation, and clefting.
naling family and signal via BMP receptors. Initiation of some of those embryonic signaling
FGFR1 signaling regulates BMP7 directly and pathways has been observed in active repair situ-
BMP4 indirectly to regulate epithelial growth. ations, such as ductal ligation settings where aci-
BMP4, which is mesenchyme specific, inhibits nar atrophy and hyposalivation is temporarily
epithelial branching, while BMP7, released by induced by restricting salivary flow from the
both epithelium and mesenchyme, increases it major duct [17]. We can thereby try to manipu-
[90]. The role of another member of the TGF late the activation and/or repression of specific
family, TGFβ1, is still inconclusive. While developmental pathways to stimulate in vivo
TGFβ1-deficient mice have normal SGs, over- repair of damaged SGs, as is outlined further in
stimulation of TGFβ1 results in acinar loss, elon- this chapter.
gated ducts, and/or fibrosis [35, 42].
Additionally, ECM and epithelial integrin cell
interactions are just as essential for branching 1.3 Environmental Cues
morphogenesis. These ECM molecules line up Patterning Epithelial
the basement membrane (BM) separating the epi- Branching and Maturation
thelium from mesenchyme. Interestingly, iso-
lated epithelial cells can easily grow without the SGs are highly vascularized and innervated, all of
physical presence of mesenchymal cells but not which integrate within a condensed mesenchyme.
without ECM component(s), such as laminin, Developmentally, SG epithelia invade a con-
fibronectin, perlecan, collagen, or mouse densed mesenchymal placode already containing
sarcoma- derived reconstituted BM “Matrigel.” a complex endothelial network and parasympa-
Deposition of unique ECM components along thetic neuronal bodies awaiting cues for innerva-
the clefting endbuds and elongating ducts plays a tion [48]. Both signals for epithelial invasion into
role in correct branching. Impairing these con- the mesenchyme and subsequent branching are
nections will lead to reduced clefting, endbud transmitted via direct cell-cell contact and/or
number, cell movement, and/or growth. Detailed indirect paracrine signaling pathways, which are
descriptions of disruptive ECM cell outcomes discussed below.
were recently reviewed in [79].
Finally, an underexplored area contributing to
SG formation are microRNAs (miRNAs), which 1.3.1 Guiding Neurons
are small, noncoding RNAs that specifically tar-
get mRNAs to globally regulate gene expression. Different cranial nerves innervate the pre- and
Epithelial endbud progenitors highly express postnatal SG where they exert different func-
miR200c to reduce FGFR-dependent prolifera- tions. While the autonomic nervous system regu-
tion. miR200c downregulates the autocrine ree- lates the SG at an unconscious level and in stress
lin/very low-density lipoprotein receptor conditions, sensory neurons respond to mechani-
(VLDLR) pathway, which positively regulates cal, thermic, and light signals.
FGFR signaling [83]. Additionally, it was found For decades, both the parasympathetic and
that EGF can specifically induce mesenchymal sympathetic nervous system have been acknowl-
production of miR-21, which decreases multiple edged as the driving stimulant to release saliva
target mRNA candidates. One of these, RECK, from acinar cells into ducts. While parasympa-
inhibits MMPs, which subsequently influences thetic stimulation results in serous secretion and
ECM degradation to enhance SG branching [37]. ion release, sympathetic activation stimulates
In conclusion, various signaling pathways mucous or protein-containing saliva and can also
instruct different cell types within the epithelial play role in local inflammation and blood flow
compartment and not surprisingly interact with [23, 67]. Both parasympathetic and sympathetic
and regulate each other to safeguard temporal- nerves are part of the autonomic nervous system,
Fig. 1.3 Signaling
pathways driving neuronal
survival and innervations.
Illustrated signaling
pathways for
parasympathetic nervous
system were demonstrated
in prenatal glands,
sympathetic nervous system
in postnatal SGs. Green:
cells; red: expressed by
neuronal cells; blue:
expressed by endothelial
cells; black: undefined
which compartment takes
part in it
and their neuronal guidance is ensured by axons initiating SG epithelia to localize into ganglia
that sprout along unique paths within the tissue. around the primary duct and send out axons
While the former innervates along the epithelia, toward the endbuds [48]. Sympathetic nerves
the latter follows the vasculature. This directional innervate SGs along the blood vessels during
guidance is driven by neurotrophic factors, later stages of development when final epithelial
secreted by cells in the periphery, as well as the maturation is needed. As such, developmental
presence of specific receptors on the axons that experiments can clearly dissect the role of neu-
allow or block their adhesion to the adjacent rotransmitters and neurotrophic factors affecting
ECM. The most notable trophic molecules include the PSG. It is now well appreciated that the PSG
neurotrophins (e.g., NGF, BDNF, NT-3), netrins, establishes a communication loop with specific
semaphorins, ephrins, and myelin inhibitors. epithelial cells to allow outgrowth of both com-
Similarly, axons secrete neurotransmitters in their partments. When the PSG is absent, the pool of
proximity, exerting a variety of effects through K5-expressing epithelial progenitors is signifi-
specific receptors on their target cells (Fig. 1.3). cantly reduced [50], which influences down-
Parasympathetic nerves signal via the cholinergic stream K19+ ductal luminal differentiation and
acetylcholine (ACh) pathway, targeting musca- subsequent epithelial outgrowth. This is medi-
rinic receptors on neighboring cells, as well as ated via a loss in ACh-CHRM1 (muscarinic
water channels such as aquaporin 5 (AQP5). In receptor 1) signaling from the PSG to K5+ cells
contrast, sympathetic nerves release epinephrine and resulting in a subsequent reduction of
and norepinephrine (i.e., noradrenaline, NA) that HB-EGF/EGFR pathway signaling that initiates
bind to β-adrenergic receptors (adrenoceptors) on maintenance and differentiation of K5+ progeni-
acini. Other non-ACh, non-NA neurotransmitters tors. Lumenization, which marks further ductal
can be produced by both parasympathetic and maturation, is also coordinated by the PSG but
sympathetic nerves and may include vasoactive not via the ACh pathway. The neurotransmitter
intestinal peptide (VIP), substance P (SP), neuro- VIP activates a cAMP/protein kinase A (PKA)
peptide Y (NPY), neurokinin A, pituitary adenyl- pathway to induce epithelial duct cell prolifera-
ate cyclase-activating peptide (PACAP), and tion and formation of a single lumen by the fusion
calcitonin gene-related peptide (CGRP). of multiple microlumens. After initial lumen for-
Developmentally, parasympathetic ganglia mation, VIP remains essential to expand the
(PSG) neuron cell bodies migrate along the lumen size via the cystic fibrosis transmembrane
branches of mandibular arteries [91] to cues from (CFTR) pathway [73].
10 I.M.A. Lombaert
Organ development also requires proper bidi- does lead to hypoplasia of the gland [82] and
rectional communication. Feedback signaling thus must involve a direct or indirect role for
from epithelial cells toward the PSG stimulates sympathetic nerves in either epithelial cell main-
cell survival, migration, and innervation. At SG tenance or maturation. In the adult gland, RET
ontogenesis, WNT-producing epithelia, particu- signaling is also known to be essential for sym-
larly K5+ progenitors, maintain PSG neuron sur- pathetic neuron survival, but likely via the ligand
vival and proliferation [48]. At later stages of artemin instead of NRTN. SGs also produce high
branching morphogenesis, the neurotrophic fac- amounts of NGF and genetic ablation of NGF or
tor neurturin (NRTN), which is mainly secreted its TrkA receptor leads to defective sympathetic
by endbud progenitors, not only promotes neuro- innervation, indicating its crucial role in sympa-
nal survival via GFRα2/RET but also maintains thetic neuron survival [22, 27]. Depletion of non-
axon outgrowth along ducts toward the endbuds canonical WNT5a in WNT1-derived neural crest
[49]. In the developing lung, there also appears to cells further leads to incomplete sympathetic
be a link between nerves and blood vessels. innervation and branching in prenatal SGs. While
Denervation, in this case via physical cell abla- the authors suggest this is due to an autocrine
tion, resulted in reduced endothelial prolifera- WNT5a/retinoid-related orphan receptor (ROR)
tion, leading to hypo-vascularized lungs [9]. It is pathway in sympathetic neurons, it doesn’t rule
unclear whether this is a direct or indirect out that epithelial WNT5a-producing cells might
neuronal-endothelial effect and whether similari- be stimulating sympathetic neurons as well [89].
ties exist within the developing SG. Similarly, endothelial-released endothelin 3
Detailed anatomical descriptions of nerves in (EDN3) is also suggested to be a cue for a sub-
adult SGs are outlined in a recent review [40]. It set of EDN receptor A+ sympathetic neurons to
is assumed that similar communication between innervate the prenatal SG along the nascent exter-
nerves and epithelium persists into adulthood as nal carotid arteries [63]. The specific role of other
denervation of SGs, via ductal ligation or neurec- neurotransmitters from the GDNF and NPY fam-
tomy, reduces epithelial content that regenerates ily as well as other neurotrophic factors are still
after ligation removal if the nerve is intact or being explored. While semaphorins are involved
reconnected [46, 55, 65]. A morphological differ- in axon pruning and neuronal migration in the
ence of early development with later stages and central nervous system, they also appear to have a
adulthood is that smaller ganglia are found dis- role in developing SGs. Semaphorin (SEMA) 3A
persed within adult tissue [40], presumably to and 3C bind co-receptors neuropilin and plexin.
reach their target cells more easily as distances Neuropilin is expressed by epithelial endbuds
are much larger compared to embryonic and by activation with SEMA3A and 3C cleft
development. formation is induced without changing prolifera-
Even though tyrosine hydroxylase (TH)- tion and, most likely, by affecting cell movement
expressing sympathetic ganglia are presumed [15]. However, additional FGF7/10 growth-pro-
not to be present at SG ontogenesis, some moting signals from surrounding mesenchyme
mRNA expression levels of its unique recep- were required to mediate this cleft formation.
tor neuropeptide Y receptor 2 (NPY2R) were Whether additional participation of SEMAs on
detected early during development at low levels receptive nerves is required for cleft formation or
that increase before birth [23]. Since NPY2R is SG development still remains unclear.
also present on endothelial cells, some, if not At adulthood, it remains to be determined how
all, of the mRNA expression could be related to sensitive sympathetic nerves are to injuries such
blood vessel formation within the SG. However, as radiation. In rodents, sympathetic nerve func-
TH-expressing neuronal cells were detectable tion was retained after radiation [52], and
by E16.5, which might indicate there is a pre- increased levels of TH as well as NGF/NGFR and
natal presence of sympathetic ganglia [89]. adrenergic receptor 2 (ADRA2B) were detected
Nevertheless, postnatal sympathetic denervation in radiated human SMGs [49]. Whether a
r eestablishment of the balance between parasym-

pathetic and sympathetic nervous system is nec-
essary for regeneration is not known.
Furthermore, it is assumed that sensory neuro-
nal cells are present along the sympathetic and
parasympathetic nerve tracks in adult SGs [51],
although they have not yet been studied in detail.
While sensory neurons can be defined into mul-
tiple subtypes based on different criteria such as
their origin and molecular expression patterns,
they often are loosely classified as unmyelinated
capsaicin-sensitive TRPV1+ receptor expressing
neurons or myelinated glutamate receptor-
expressing neurons. Upon activation, sensory
nerves can secrete various neuropeptides, such as
Substance P and CGRP. At least in the lung and
pancreas, literature indicates that sensory neu-
rons release neurotransmitters in the periphery to
Fig. 1.4 Potential communications initiated by and to
serve as direct mediators for recruiting and acti- endothelial cells. Described pathways were majorly found
vating inflammatory cells [68, 86]. Whether a in other branching organs. Whether they exist in SGs
similar mechanism occurs in salivary glands is needs to be determined. Green: expressed by epithelial
not known. cells; orange: expressed by mesenchyme; blue: expressed
by endothelial cells
In conclusion, innervation plays an essential
role for organ development, homeostasis, and
repair after injury. Studies on SG biogenesis have lial secretory factors, termed angiocrines, which
been highly informative for defining the involve- impact organ development (Fig. 1.4). While
ment of the PSG in branching morphogenesis, research on blood vessels in SGs remains limited,
not only of SGs but also other organs such as much can be learned from other branching organs.
prostate and lungs. The existence and/or loss of Overall, complex cross-communication between
bidirectional communication with epithelial epithelial and endothelial cells appears to regu-
stem/progenitor cells have been reported to occur late both epithelial differentiation and angiogen-
in rodent and human SG homeostasis and postra- esis. The initial cues to form an endothelial cell
diation. The inhibition of parasympathetic neuro- plexus around a condensed mesenchyme do not
nal function influences adult epithelial K5+ require epithelia. This is observed in Fgf10−/−
progenitors [49] but it is not known yet whether mice that don’t form initial SMG epithelia but
postradiation regeneration due to epithelial stem/ where the placode of mesenchyme, blood ves-
progenitor transplantation repairs neuronal func- sels, and neuronal bodies are present [48]. After
tion, even though morphological repair has been SG ontogenesis, however, epithelial- derived
suggested [71]. angiogenic factors, such as VEGF-A, do play a
role as null mutations in Vegf-A or its endothelial-
expressed receptor (Vegfr) show v ascular defects
1.3.2 The Role of Blood Vessels in tissues, reduced epithelial budding, and ulti-
mately embryonic lethality [13, 107]. Another
The vasculature in branching organs develops in epithelial-induced angiogenic mechanism may
close proximity to the epithelia, although its spa- include the vitamin D pathway. The enzymes
tial pattern differs from parasympathetic nerves. CYP27B1/24A1 that activate and catabolize
Not only are endothelia important for mediating vitamin D are highly upregulated just before
gas exchange but also as a source of endothe- birth and in postnatal lung. Exogenous vitamin
12 I.M.A. Lombaert
D positively influences lung growth by inducing laminins. These components in turn served as
maturation in vitamin D receptor-expressing epi- scaffolds for increased axon outgrowth.
thelial cells (VDR) [64]. As VDR is also present In addition to blood vessels, we must not for-
on endothelial cells, this enhanced growth might get the circulating cells within them. White blood
be due to direct effects of vitamin D on endothe- cell monocyte-derived macrophages and den-
lial cells and/or indirect effects from epithelia to dritic cells arise from the bone marrow and colo-
endothelia. Nevertheless, it is clear that epithe- nize tissues via blood vessels to phagocytose
lial-endothelial communication requires a tight cellular debris and help in the innate non-specific
balance as any hyper-vascularization inhibits epi- and specific adaptive immune defense. While
thelial growth [14]. macrophages normally develop in the bone mar-
Apart from endothelial-epithelial cross- row via granulocyte-macrophage colony-
communication, there is also endothelial- stimulating factor (GM-CSF), mesenchymal
mesenchymal communication, as recently cells in tissues can also release GM-CSF to
reviewed [94]. The early endothelial cells pro- induce a similar differentiation effect on circulat-
mote survival of pancreatic mesenchymal cells, ing monocytes. While it is clear that both macro-
which in turn have a pivotal role in organ devel- phages and dendritic cells may be involved in
opment. A similar complex paracrine signaling organ morphogenesis, their exact functions are
network was also found in the lung. Retinoic not always fully understood. Also in adult tissues,
acid (RA), which is produced by endothelial for example, in the lung, there is conflicting data
cells, induces VEGF-A expression in lung epi- on their specific role: antigen-sensing dendritic
thelia. Evidently, endothelial cells are recruited cells might induce different immune responses
via VEGF-A, and thus angiogenesis is stimu- depending on their physical location in the tissue
lated via this endothelial-epithelial communica- while surrounding different epithelial cell types
tion loop. Furthermore, endothelial-released RA [54]. Similarly, various macrophages invade the
also stimulated mesenchymal cells to produce mesenchyme where they can interact with den-
more FGF18 and ECM component elastin, thus dritic cells, lymphocytes, and epithelia to regu-
increasing epithelial alveolar formation [108]. late immunity. Macrophages suppress immune
Other organ-specific angiocrine factors that may responses by inhibiting both dendritic-mediated
follow this paracrine loop include HGF, WNT, T-cell activation and inactive TGFβ production.
NOTCH, and BMP ligands. Mesenchymal cells Subsequent activation of this inactive TGFβ into
also signal back to endothelial cells to stimulate bioactive TGFβ by lung epithelia is essential in
survival, proliferation, migration, and autoph- order to prevent spontaneous inflammation after
agy via production of ECM components, such acute injury. Lung alveolar cells in turn secrete
as the perlecan/heparan sulfate proteoglycan various ligands to receptive macrophages to
(HSPG2) fragment endorepellin, decorin, and ensure this prevention of inflammatory responses.
endostatin [18, 75]. Whether a similar action or disruption in this
While blood vessels and nerves can indepen- communication is occurring in adult SGs after
dently respond to their own set of signaling fac- radiation remains to be determined.
tors, there also seems to be a paracrine connection Apart from immune regulators, macrophages
via epithelial-released VEGF. Even though further shape the branching patterning of organs
VEGFR is absent on nerves and not required for by remodeling the ECM around the ducts to
innervation, VEGF overexpression in pancreas allow outgrowth as well as survival of endbud
not only led to hyper-vascularization but also to stem/progenitor cells [11, 102]. They also regu-
hyper-innervation [85]. Interestingly, endothelial late angiogenesis by instructing endothelial cells
cells did not produce any known neurotrophic to undergo apoptosis via WNT signaling, coun-
factors, but the effect appeared to be related to terbalancing a pro-survival factor produced by
their upregulated expression of basement mem- pericytes, which wrap around endothelial cells to
brane components, such as collagens and influence functions such as blood flow [2].
Fig. 1.5 Described
signaling interactions with
the SG mesenchyme. Most
known communications
are between the SG
mesenchyme and
epithelium. Green:
cells; orange: expressed by
mesenchyme; blue:
expressed by endothelial
cells; black: undefined
which compartment takes
part in it
In sum, blood vessels play important roles 29, 99]. This indicated that SG mesenchyme has
during development in other branching tissues strong and unique multicomponent instructional
not only for oxygen supply but also to maintain properties. One of them is the high production of
essential communication signaling pathways FGF10, which is also essential for lung, lacrimal,
with epithelia and surrounding mesenchyme. The and mammary gland initiation. Notably, non-SG
bone marrow-derived cells circulating in the epithelia only adapted SG-like branching patterns
blood vessels also aid in tissue branching mor- when they were placed in close vicinity to high
phogenesis and evoke or suppress immune FGF10-expressing mesenchymal tissue, confirm-
responses after injury. Whether similar mecha- ing the importance of FGF10’s spatiotemporal
nisms exist in the developing and adult SGs dosage [99]. With this in mind, it is important to
remains to be determined. understand that the SG mesenchymal component
in these experimental conditions contains mesen-
chymal cells as well as ganglia and blood vessels
1.3.3 Supporting Mesenchymal albeit disconnected from the rest of the body. It
Cells can therefore not be excluded that SG-specific
neuronal cells and/or blood vessels may have addi-
Embryonic SG mesenchymal cells are WNT1+ tional contributions to this specific SG patterning.
neural crest-derived cells [44, 105] and provide Interestingly, early-stage E11.5–12.5 SG epithe-
supportive cues such as growth factors, proteases, lia, but not later stages, are able to instruct E10.5
and ECM proteins to guide and activate epithelial, mesenchyme from different sources to produce
neuronal, and endothelial cells (Fig. 1.5). In vitro FGF10. However, not every mesenchyme is as
recombination experiments show that the SG mes- competent to receive this signal as only SG and pha-
enchyme induces an SG-like branching pattern in ryngeal second arch mesenchyme responded and
various epithelia such as a pancreatic, mammary, limb mesenchyme, for example, did not. This indi-
and pituitary gland [96]. This property, however, is cates that this initial signal is exclusively located
not found in mesenchyme from non-SG tissues, as within specific regions of the embryo, likely to
E11.5–13 SG epithelium does not properly branch restrict specific organ outgrowth to the correct loca-
unless it is recombined with SG mesenchyme [28, tion in the body. What this initial epithelial signal is
14 I.M.A. Lombaert
remains a subject of debate. Whereas early limb and in the first step, is exclusively expressed in early
lung epithelia secrete FGF8 or FGF9 to initiate this SG mesenchyme, while RA activity is mainly
process, it is unlikely that FGF8 serves a similar observed in RA receptor+ epithelia. Disruption in
function in SGs [99]. Neither is the signal FGF4, Rdh10 results in early embryonic lethality, often
BMP2, SHH, TGFβ1, or WNT6 [47]. One unex- before SG epithelial invagination. However, when
plored candidate is platelet-derived growth factor E13 SGs were treated with an RAR inhibitor
(PDGF). During development, PDGF-A and reduced branching was observed [101]. In contrast,
PDGF-B ligands are mainly produced by epithelia mesenchymal cells can also secrete signaling
and mesenchyme, respectively [105], while inhibitors to slow down branching. DLK1, a non-
PDGFR-A and PDGFR-B receptors are expressed canonical NOTCH1 ligand produced by the mes-
in the mesenchyme. By adding exogenous PDGF, enchyme, inhibits branching and subsequent
epithelial proliferation can accelerate via upregula- innervation, presumably to modulate cleft forma-
tion of mesenchymal Fgf7, Fgf10, Fgf1, and Fgf3 tion [25]. Furthermore, DLK1 appears to regulate
and downregulation of growth inhibitory factor the epithelial balance of K14+ progenitors,
Fgf2. While this induction was observed during SG although the precise mechanism is unclear [26].
morphogenesis (E14), it has not been confirmed The role of TGFβ1 is also unclear; there is some
that epithelial PDGF-A is a potential FGF10 inducer evidence that epithelial-secreted TGFβ1 enhances
at SG ontogenesis. Nevertheless, once FGF10 collagen production from Coll1α1+ mesenchymal
expression is initiated, it persists and becomes inde- cells to inhibit SG acinar formation [42].
pendent from epithelial cues. It is also interesting to Other mesenchymal signaling pathways that
note that mesenchymal condensation at placode ini- may influence epithelial branching are hepato-
tiation is independent of this FGF10 activation. As cyte growth factor (HGF)/c-MET and SDF1/
such, the mesenchyme can condense around a net- CXCR4 signaling [38]. Additional cellular
work of blood vessels and resting PSG cells before instruction mechanisms also include microRNAs
SG initiation and in the absence of FGF10 as seen in (miRNAs). miRNAs are small noncoding RNA
Fgf10−/− mice. This mesenchyme presumably molecules that function to silence other mRNAs.
awaits signal(s) from the invading oral epithelia to While most research has focused on
initiate FGF10, which in turn promotes SG-specific mesenchymal- epithelial interactions, there may
epithelial growth. also be mesenchymal-endothelial interactions.
Even during branching morphogenesis, mesen- When the mesenchymal factor SDF-1 was specifi-
chymal cells continue to play a part in multiple cally inhibited from binding CXCR7+ endothelial
bidirectional signaling pathways. Early on, WNT/ cells, SG epithelial branching was decreased [38],
β-catenin signaling is exclusively induced in the thus suggesting a tri-directional loop between
mesenchyme before it is expressed in lumenizing mesenchymal, endothelial, and epithelial cells. It
ductal cells [78]. This mesenchymal WNT can is also possible that mesenchymal cells may play a
activate EDA and, at least in part, influence SG role in axonal guidance. As outlined earlier, mul-
morphogenesis via epithelial EDAR [31]. tiple studies have verified that mesenchymal cells
Branching epithelia also regulate local FGF10 aid in cellular migration, clefting, and differentia-
expression to reduce aberrant cell proliferation. tion via regulation of ECM production.
Lung epithelia release BMP ligands as well as
SHH to spatially downregulate FGF10 and specifi-
cally induce secondary bud formation [12]. A 1.4 ranslation into Future
T
recent study further points to an important role for Therapies to Repair
mesenchymal retinoic acid (RA) to enhance Damaged Salivary Glands
branching. RA is a small diffusible hormone-like
molecule generated by a two-step enzymatic oxi- There is a tremendous need for long-term thera-
dation of dietary vitamin A via RDH10 and pies to restore salivary flow. Clinical impacts of
ALDH1A. RDH10, which metabolizes vitamin A dry mouth, or xerostomia, not only include difficulty
with food mastication, swallowing, taste, and epithelial cells still being present in the gland.
speech but also increased risks for dental caries, Suppressing cell apoptosis is another treatment
pain, and oral infections [19]. Depending on the that has had some success in animal models.
type of damage, different therapies could be con- Approaches shown to be effective in mouse mod-
sidered. Moderate damage could be addressed by els include pretreatment with insulin growth fac-
protein and/or gene activation, while cell thera- tor 1 (IGF1) [69], FGF2 [30, 53], Tousled kinase
pies and/or bioengineered tissues to restore the (TLK1B) [77], Pkcδ [1], or roscovitine. The lat-
entire organ may be more suitable for cases of ter is a cyclin-dependent kinase inhibitor that
severe radiation-induced damage. Current bioen- transiently inhibits G2/M cell cycle arrest, allow-
gineering, gene and mesenchymal stem cell ther- ing suppression of apoptosis and DNA repair
apies, as well as SG transfers are emphasized in [66].
the following chapters. Here, we review how dif- Alternatively, epithelial cell proliferation can
ferent therapies could contribute to SG repair by be stimulated via posttreatment with aldehyde
correlating them with the developmental con- activator ALDH3 [5] or pre- and posttreatment
cepts described above. with keratinocyte growth factor (KGF) or FGF7
[60, 109] to ameliorate radiation-induced
hyposalivation in mice. Other signaling pathways
1.4.1 Epithelial Protection shown to enhance adult SG regeneration postra-
and Repair diation or post-ductal ligation in mice are concur-
rent or transient activation of WNT/β-catenin
Since epithelial cells are the major component of [33, 34], SHH [32], and EDA [39]. Similar to
SGs, they are the main target for any type of dam- developmental processes, all of these pathways
age, especially radiation. Therefore, the preven- affect epithelial stem/progenitors and aid in their
tion of SG cell apoptosis and/or membrane expansion and differentiation. Notably, radiation
damage-induced dysfunction of acinar cells is itself does not alter endogenous WNT, EDA, or
clinically attempted by using intensity modulated SHH pathway components, but evidently over-
radiation therapy (IMRT) rather than conventional stimulation reinforces epithelial growth mecha-
radiotherapy. The precise delivery of radiation by nisms required to regenerate the damaged tissues.
IMRT reduces the amount of the SGs being tar- Another approach is to use a cocktail of chemo-
geted, still, 40 % of head and neck cancer patients kines, cytokines, and growth factors obtained by
experience moderate to severe oral dryness [8]. mobilizing or injecting bone marrow-derived
Recently, it was revealed that exclusion of a subre- cells, adipocytes, and/or mesenchymal stem cells
gion of the cranial SG is essential to reduce severe into damaged SGs, as reviewed in [61]. These
loss of organ function [97]. Not surprisingly, this bioactive lysates are proposed to drive SG repair
region appears to harbor the highest number of by reactivating signaling pathways in the
epithelial SG stem/progenitor cells. Superior dose epithelial and endothelial cells that remain.

distribution as delivered by proton therapy is fur- Whether they also stimulate post-damage neuro-
ther expected to improve dose sparing of SGs, spe- nal repair is currently unclear.
cifically in this cranial subregion [56]. Restoration of saliva secretion is further fea-
In the meantime, free radical scavengers (ami- sible by epithelial water channel aquaporin 1
fostine and tempol) and saliva-stimulating sialo- (AQP1) protein gene therapy. After extensive ani-
gogues (pilocarpine) provide relief to some mal research in mice, rats, pigs, and monkeys, the
patients. However, major health-related side safety of an adenovirus-containing human AQP1
effects induced by radical scavengers, such as vector (AdhuAQP1) was studied in human Phase
vomiting and fever, need to be taken in consider- I clinical trials [6]. While there was some efficacy
ation. Also the effectiveness of pilocarpine is reported in some patients, follow-up studies to
related to the severity of tissue damage as this test the effectiveness of this therapy for long-term
muscarinic agonist relies on some functional SG maintenance of increased salivary flow are
16 I.M.A. Lombaert
ongoing. The reader is guided to a following derived vesicles or exosomes, which not only medi-
chapter for more in-depth information. ate transmission of proteins but also mRNA and
SG cell transplantations have proven to be an microRNAs [76]. One study proposes a participat-
effective treatment in rodent models. Several SG ing role for mRNA and microRNAs in neuronal
epithelial specific and non-SG cell types are able to myelination and survival. Oligodendrocytes secrete
integrate within the remaining epithelial compart- exosomes with specific proteins and RNA toward
ment and contribute to its repair and subsequent surrounding neurons in the brain to improve their
homeostasis by differentiating into pools of various viability under stress conditions [24]. While exo-
ductal and acinar cell types. Their potential use in some-influenced neuronal repair has yet not been
the clinic has been reviewed in detail [61]. Briefly, investigated in SGs, neurotrophic protein deliveries
autologous SG cell transplantation could occur in via injection or gene therapy are currently being
patients who still need to undergo radiation and explored. In fetal SG experimental settings, radia-
where a SG biopsy could be taken before radiation tion-induced neuronal apoptosis and denervation
therapy starts. Subsequent post-therapy transplan- were reduced by postradiation delivery of exoge-
tation of biopsy-isolated cells could initiate regen- nous neurturin (NRTN), which binds GFRα2/RET
eration of the radiated SG. Alternatively, other cell receptors on parasympathetic nerves [49]. Adult
types such as mesenchymal or adipose cells could radiated SGs were shown to benefit from exogenous
be transplanted or mobilized post-radiotherapy to GDNF, which binds GFRα1 and aids in epithelial
positively influence epithelial, mesenchymal, and stem/progenitor cell expansion [103]. This likely
endothelial repair. occurs via neuronal communication, although its
In conclusion, the number of remaining stem/ reinnervation pattern has not yet been studied.
progenitor cells and functional epithelial cells will Future efforts will also need to be directed
determine the probability of spontaneous regenera- toward determining the impact of radiation on sym-
tion of the damaged SG, as well as efficiency of pathetic nerves and possible positive influences by
sialogogues, apoptosis protectors, and signaling WNT5a, NGF, or END3, which are known to be
pathway stimulators. Restoring the epithelial com- important during development of the gland.
partment could elicit repair of surrounding blood In sum, functional neuronal repair and rein-
vessels and nerves as well. As indicated from ani- nervation may help in the release of neurotrans-
mal models [32, 59, 72], this can occur via para- mitters to maintain epithelial stem/progenitor
crine communications from epithelial angiocrine cells and induce epithelial regeneration and duc-
and neurotrophic factor (e.g., BDNF, NRTN) tal maturation. Moreover, their reactivation is
release. One important clinical issue to be deter- necessary for proper saliva release from acinar
mined is whether these proposed treatments lead to cells and may also reestablish the communication
undesirable side effects such as radioresistance pathways necessary for repair and subsequent
and/or the acceleration of the patient’s tumor cells. homeostasis. Importantly, the major human SMG
and SLG parasympathetic ganglions are located
outside the gland in contrast with many rodent
1.4.2 I nducing Neuronal Survival models [21]. The prevention or reduction of radi-
and Reinnervation ation to these ganglions, as well as the PAR gan-
glion that is located outside the tissue, could also
Nerves have a limited capacity to regenerate, and improve the efficiency of these therapies.
irradiated human SGs have been shown to have
reduced parasympathetic innervation [49]. The role
of nerves and neurotransmitters during organ devel- 1.4.3 R
estoration of Blood Vessel
opment and homeostasis is being elucidated, but Supply
less is known about their role during gland repair.
Apart from neurotrophic release, innervation can Radiation severely impacts blood vessels to the
also be influenced via cell plasma membrane- point that capillary endothelial cells detach from
the basal lamina, their density reduces, and large stress fiber formation and mature their cell matrix
blood vessels dilate. Reducing damage to blood focal adhesions to turn into myofibroblasts,
vessels will contribute to repair of damaged SGs. thereby producing excessive ECM. Injecting
As iterated above, endothelial cells communicate MSC/adipose-derived stem cells reduces muscu-
not only with epithelial and mesenchymal but lar fibrosis as a consequence of radiation [93]. In
also neuronal cells. Furthermore, circulating the SG, these cells could presumably reduce
bone marrow-derived cells migrate via blood inflammation as well as degrade the ECM and
vessels into damaged tissues to phagocytose the induce phagocytosis of apoptotic myofibroblasts.
damaged environment. Several proposed strate- The timing of cell delivery, however, will be cru-
gies actively contribute to repair of the endothe- cial as delayed therapy after radiation resulted in
lial compartment. Pre-radiation gene delivery of increased lung fibrosis due to the differentiation
angiocrine VEGF or FGF2 could ameliorate of mesenchymal stem cells into myofibroblasts
endothelial damage, resulting in reduced hyposal- [106]. Although TGFb has beneficial roles in
ivation [16]. In vivo mobilization of bone wound healing and inflammatory responses,
marrow-derived cells is also a feasible approach excessive levels of TGFβ1 can result in a number
to increase saliva flow as both their secretory bio- of serious conditions that are characterized by
active lysate as well as the presence of endothe- fibrosis, including chronic hepatitis, glomerulo-
lial progenitors can contribute to stimulation of sclerosis, and postradiation tissue remodeling. As
endothelial survival and proliferation [59, 92]. such, treatment with TGFβ inhibitors or geneti-
cally engineered TGFβR or HGF-expressing
MSCs may be beneficial to attenuate fibrosis, as
1.4.4 I nflammation and Stromal has been observed in radiated lungs [98, 104].
Cell-Induced Fibrosis
Inflammation following radiation is an acute 1.5 Future Prospects

damage response but can persist and become
chronic. While recruitment of macrophages is a From both this chapter and following chapters, it is
necessity to phagocytose apoptotic cells [102], clear that there are a number of different strategies
too much secretion of inflammatory cytokines to repair damaged SGs. Intravenous protein deliv-
and chemokines like IL, CCL2, and TNF can ery, particularly of growth factors or agents that
augment epithelial dysfunction. If not immedi- increase stem/progenitor cell activity, is often clin-
ately controlled, the inflammatory response ically disputed due to concerns regarding activa-
becomes pathogenic, as seen in autoimmune dis- tion of any remaining tumor tissue. Local
eases and tissue fibrosis. In such cases, it is retrograde duct delivery of agents, such as viral
imperative to switch the pro-inflammatory vectors as part of gene therapy, provides a local-
response to an anti-inflammatory state in order to ized delivery and safer strategy as SGs are encap-
allow repair. Various potential mediators for sulated and epithelial cells are easily transfected.
reducing inflammation include IL-4, IL-13, While there is a possibility of vector diffusion into
T-reg, and B1 B-cells, and new evidence has also the bloodstream, this has not been a major issue in
shown the ameliorating effects of stromal (mes- preclinical analysis. Similarly, direct cell delivery
enchymal) stem cells (MSCs) [81]. or mobilization of resident cells will always
In adult tissues, the stromal cell pool harbors require evaluation of possible improper growth
interstitial fibroblasts and adipocytes that con- patterns, and bioengineered tissues will have to
tinue to deposit and remodel the ECM. With integrate with existing tissue and/or connecting
aging patients, more adipose and fibrotic tissue is ducts, nerves, and blood vessels.
apparent that together with acinar cell loss leads Research on gland morphogenesis and repair
to a 30–40 % decrease in parenchymal SG vol- will continue to identify targets from which new
ume [7, 20]. Radiated stromal cells can enhance ways to approach regenerative therapies are being
18 I.M.A. Lombaert
developed. Clearly, reestablishing crosstalk 8. Beetz I, Steenbakkers RJ, Chouvalova O, Leemans

CR, Doornaert P, van der Laan BF, Christianen ME,
between different cell types endogenously
Vissink A, Bijl HP, van Luijk P, Langendijk JA. The
enhances the regeneration process. Thus, by QUANTEC criteria for parotid gland dose and their
improving one compartment, one may also indi- efficacy to prevent moderate to severe patient-rated
rectly improve repair in another compartment. xerostomia. Acta Oncol. 2014;53(5):597–604. doi:1
0.3109/0284186X.2013.831186.
Recreating an optimal environment wherein cells
9. Bower DV, Lee HK, Lansford R, Zinn K, Warburton
can multiply and reconstruct the organ will be key. D, Fraser SE, Jesudason EC. Airway branching
Currently, multiple approaches may be required to has conserved needs for local parasympathetic
improve specific tissue structures within the organ innervation but not neurotransmission. BMC Biol.
2014;12:92. doi:10.1186/s12915-014-0092-2.
and/or simultaneously influence other compart-
10. Bullard T, Koek L, Roztocil E, Kingsley PD, Mirels
ments to regenerate a damaged organ. L, Ovitt CE. Ascl3 expression marks a progeni-
tor population of both acinar and ductal cells in
Acknowledgments I thank Dr. Wendy Knosp and Dr. mouse salivary glands. Dev Biol. 2008;320(1):72–8.
Matthew Hoffman (National Institute for Dental and doi:10.1016/j.ydbio.2008.04.018.
Craniofacial Research, NIH, DHHS, USA) for critical 11. Bussard KM, Smith GH. The mammary gland
proofreading. microenvironment directs progenitor cell fate
in vivo. Int J Cell Biol. 2011;2011:451676.
doi:10.1155/2011/451676.
12. Cardoso WV, Lu J. Regulation of early lung mor-
References phogenesis: questions, facts and controversies.
Development. 2006;133(9):1611–24. doi:10.1242/
1. Arany S, Benoit DS, Dewhurst S, Ovitt dev.02310.
CE. Nanoparticle-mediated gene silencing confers 13. Carmeliet P, Ferreira V, Breier G, Pollefeyt
radioprotection to salivary glands in vivo. Mol Ther. S, Kieckens L, Gertsenstein M, Fahrig M,
2013;21(6):1182–94. doi:10.1038/mt.2013.42. Vandenhoeck A, Harpal K, Eberhardt C, Declercq
2. Armulik A, Genove G, Betsholtz C. Pericytes: devel- C, Pawling J, Moons L, Collen D, Risau W, Nagy
opmental, physiological, and pathological perspectives, A. Abnormal blood vessel development and lethal-
problems, and promises. Dev Cell. 2011;21(2):193– ity in embryos lacking a single VEGF allele. Nature.
215. doi:10.1016/j.devcel.2011.07.001. 1996;380(6573):435–9. doi:10.1038/380435a0.
3. Arnold K, Sarkar A, Yram MA, Polo JM, Bronson R, 14. Carolan PJ, Melton DA. New findings in pancreatic
Sengupta S, Seandel M, Geijsen N, Hochedlinger K. and intestinal endocrine development to advance
Sox2(+) adult stem and progenitor cells are important for regenerative medicine. Curr Opin Endocrinol
tissue regeneration and survival of mice. Cell Stem Cell. Diabetes Obes. 2013;20(1):1–7. doi:10.1097/
2011;9(4):317–29. doi:10.1016/j.stem.2011.09.001. MED.0b013e32835bc380.
4. Aure MH, Arany S, Ovitt CE. Salivary glands: 15. Chung L, Yang TL, Huang HR, Hsu SM, Cheng
stem cells, self-duplication, or both? J Dent Res. HJ, Huang PH. Semaphorin signaling facilitates
2015;94(11):1502–7. doi:10.1177/0022034515599770. cleft formation in the developing salivary gland.
5. Banh A, Xiao N, Cao H, Chen CH, Kuo P, Krakow T, Development. 2007;134(16):2935–45. doi:10.1242/
Bavan B, Khong B, Yao M, Ha C, Kaplan MJ, Sirjani dev.005066.
D, Jensen K, Kong CS, Mochly-Rosen D, Koong 16. Cotrim AP, Sowers A, Mitchell JB, Baum
AC, Le QT. A novel aldehyde dehydrogenase-3 acti- BJ. Prevention of irradiation-induced salivary hypo-
vator leads to adult salivary stem cell enrichment function by microvessel protection in mouse salivary
in vivo. Clin Cancer Res. 2011;17(23):7265–72. glands. Mol Ther. 2007;15(12):2101–6. doi:10.1038/
doi:10.1158/1078-0432.CCR-11-0179. sj.mt.6300296.
6. Baum BJ, Alevizos I, Zheng C, Cotrim AP, Liu S, 17. Cotroneo E, Proctor GB, Carpenter GH. Regeneration
McCullagh L, Goldsmith CM, Burbelo PD, Citrin DE, of acinar cells following ligation of rat subman-
Mitchell JB, Nottingham LK, Rudy SF, Van Waes C, dibular gland retraces the embryonic-perinatal
Whatley MA, Brahim JS, Chiorini JA, Danielides S, pathway of cytodifferentiation. Differentiation.
Turner RJ, Patronas NJ, Chen CC, Nikolov NP, Illei 2010;79(2):120–30. doi:10.1016/j.diff.2009.11.005.
GG. Early responses to adenoviral- mediated trans- 18. Davis GE, Senger DR. Endothelial extracellular
fer of the aquaporin-1 cDNA for radiation- induced matrix: biosynthesis, remodeling, and functions
salivary hypofunction. Proc Natl Acad Sci U S A. during vascular morphogenesis and neovessel
2012;109(47):19403–7. doi:10.1073/pnas.1210662109. stabilization. Circ Res. 2005;97(11):1093–107.
7. Baum BJ, Ship JA, Wu AJ. Salivary gland function doi:10.1161/01.RES.0000191547.64391.e3.
and aging: a model for studying the interaction of 19. Delli K, Spijkervet FK, Kroese FG, Bootsma
aging and systemic disease. Crit Rev Oral Biol Med. H, Vissink A. Xerostomia. Monogr Oral Sci.
1992;4(1):53–64. 2014;24:109–25. doi:10.1159/000358792.
20. Drummond JR, Newton JP, Abel RW. Tomographic induced hyposalivation by preserving salivary
measurements of age changes in the human parotid stem/progenitor cells and parasympathetic inner-
gland. Gerodontology. 1995;12(1):26–30. vation. Clin Cancer Res. 2014;20(1):140–50.
21. Ekstrom J, Khosravani N, Castagnola M, Messana doi:10.1158/1078-0432.CCR-13-1434.
I. Saliva and the control of its secretion. In: Ekberg 33. Hai B, Yang Z, Millar SE, Choi YS, Taketo MM,
O, editor. Dysphagia. Berlin/Heidelberg: Springer; Nagy A, Liu F. Wnt/beta-catenin signaling regulates
2012. doi:10.1007/978-3-642-17887-0. postnatal development and regeneration of the sali-
22. Fagan AM, Zhang H, Landis S, Smeyne RJ, Silos- vary gland. Stem Cells Dev. 2010;19(11):1793–801.
Santiago I, Barbacid M. TrkA, but not TrkC, recep- doi:10.1089/scd.2009.0499.
tors are essential for survival of sympathetic neurons 34. Hai B, Yang Z, Shangguan L, Zhao Y, Boyer A, Liu
in vivo. J Neurosci. 1996;16(19):6208–18. F. Concurrent transient activation of Wnt/beta-catenin
23. Ferreira JN, Hoffman MP. Interactions between devel- pathway prevents radiation damage to salivary glands.
oping nerves and salivary glands. Organogenesis. Int J Radiat Oncol Biol Phys. 2012;83(1):e109–16.
2013;9(3):199–205. doi:10.4161/org.25224. doi:10.1016/j.ijrobp.2011.11.062.
24. Fruhbeis C, Frohlich D, Kuo WP, Amphornrat J, 35. Hall BE, Zheng C, Swaim WD, Cho A, Nagineni
Thilemann S, Saab AS, Kirchhoff F, Mobius W, Goebbels CN, Eckhaus MA, Flanders KC, Ambudkar IS,
S, Nave KA, Schneider A, Simons M, Klugmann M, Baum BJ, Kulkarni AB. Conditional overexpression
Trotter J, Kramer-Albers EM. Neurotransmitter-triggered of TGF-beta1 disrupts mouse salivary gland devel-
transfer of exosomes mediates oligodendrocyte-neuron opment and function. Lab Invest. 2010;90(4):543–
communication. PLoS Biol. 2013;11(7):e1001604. 55. doi:10.1038/labinvest.2010.5.
doi:10.1371/journal.pbio.1001604. 36. Hashizume A, Hieda Y. Hedgehog peptide promotes
25. Garcia-Gallastegui P, Ibarretxe G, Garcia-Ramirez cell polarization and lumen formation in developing
JJ, Baladron V, Aurrekoetxea M, Nueda ML, Naranjo mouse submandibular gland. Biochem Biophys Res
AI, Santaolalla F, Sanchez-del Rey A, Laborda J, Commun. 2006;339(3):996–1000. doi:10.1016/j.
Unda F. DLK1 regulates branching morphogenesis bbrc.2005.11.106.
and parasympathetic innervation of salivary glands 37. Hayashi T, Koyama N, Azuma Y, Kashimata
through inhibition of NOTCH signalling. Biol Cell. M. Mesenchymal miR-21 regulates branching mor-
2014;106(8):237–53. doi:10.1111/boc.201300086. phogenesis in murine submandibular gland in vitro.
26. Garcia-Gallastegui P, Luzuriaga J, Aurrekoetxea M, Dev Biol. 2011;352(2):299–307. doi:10.1016/j.
Baladron V, Ruiz-Hidalgo MJ, Garcia-Ramirez JJ, ydbio.2011.01.030.
Laborda J, Unda F, Ibarretxe G. Reduced salivary 38. Hick AC, van Eyll JM, Cordi S, Forez C, Passante L,
gland size and increased presence of epithelial pro- Kohara H, Nagasawa T, Vanderhaeghen P, Courtoy PJ,
genitor cells in DLK1-deficient mice. Cell Tissue Rousseau GG, Lemaigre FP, Pierreux CE. Mechanism
Res. 2015; doi:10.1007/s00441-015-2344-z. of primitive duct formation in the pancreas and sub-
27. Ghasemlou N, Krol KM, Macdonald DR, mandibular glands: a role for SDF-1. BMC Dev Biol.
Kawaja MD. Comparison of target innervation 2009;9:66. doi:10.1186/1471-213X-9-66.
by sympathetic axons in adult wild type and het- 39. Hill G, Headon D, Harris ZI, Huttner K, Limesand
erozygous mice for nerve growth factor or its KH. Pharmacological activation of the EDA/EDAR
receptor trkA. J Pineal Res. 2004;37(4):230–40. signaling pathway restores salivary gland func-
doi:10.1111/j.1600-079X.2004.00160.x. tion following radiation-induced damage. PLoS
28. Grobstein C. Inductive epitheliomesenchymal inter- One. 2014;9(11):e112840. doi:10.1371/journal.
action in cultured organ rudiments of the mouse. pone.0112840.
Science. 1953a;118(3054):52–5. 40. Holmberg KV, Hoffman MP. Anatomy, biogenesis
29. Grobstein C. Morphogenetic interaction between and regeneration of salivary glands. Monogr Oral
embryonic mouse tissues separated by a membrane Sci. 2014;24:1–13. doi:10.1159/000358776.
filter. Nature. 1953b;172(4384):869–70. 41. Holsinger FC, Bui DT. Anatomy, function, and evaluation
30. Guo L, Gao R, Xu J, Jin L, Cotrim AP, Yan X, Zheng of the salivary glands. In: Myers EN, Ferris RL, editors.
C, Goldsmith CM, Shan Z, Hai B, Zhou J, Zhang Salivary gland disorders. Berlin/Heidelberg: Springer;
C, Baum BJ, Wang S. AdLTR2EF1alpha-FGF2- 2007. p. 1–16. doi:10.1007/978-3-540-47072-4_1.
mediated prevention of fractionated irradiation- 42. Janebodin K, Buranaphatthana W, Ieronimakis
induced salivary hypofunction in swine. Gene Ther. N, Hays AL, Reyes M. An in vitro culture system
2014;21(10):866–73. doi:10.1038/gt.2014.63. for long-term expansion of epithelial and mesen-
31. Haara O, Fujimori S, Schmidt-Ullrich R, Hartmann chymal salivary gland cells: role of TGF-beta1 in
C, Thesleff I, Mikkola ML. Ectodysplasin and Wnt salivary gland epithelial and mesenchymal dif-
pathways are required for salivary gland branching ferentiation. Biomed Res Int. 2013;2013:815895.
morphogenesis. Development. 2011;138(13):2681– doi:10.1155/2013/815895.
91. doi:10.1242/dev.057711. 43. Jaskoll T, Leo T, Witcher D, Ormestad M, Astorga J,
32. Hai B, Qin L, Yang Z, Zhao Q, Shangguan L, Ti X, Bringas Jr P, Carlsson P, Melnick M. Sonic hedgehog
Zhao Y, Kim S, Rangaraj D, Liu F. Transient acti- signaling plays an essential role during embryonic
vation of hedgehog pathway rescued irradiation- salivary gland epithelial branching morphogenesis.
20 I.M.A. Lombaert
Dev Dyn. 2004;229(4):722–32. doi:10.1002/ 54. Kopf M, Schneider C, Nobs SP. The development
dvdy.10472. and function of lung-resident macrophages and
44. Jaskoll T, Zhou YM, Chai Y, Makarenkova HP, dendritic cells. Nat Immunol. 2015;16(1):36–44.
Collinson JM, West JD, Hajihosseini MK, Lee J, doi:10.1038/ni.3052.
Melnick M. Embryonic submandibular gland mor- 55. Kyriacou K, Garrett JR. Morphological changes in
phogenesis: stage-specific protein localization of the rabbit submandibular gland after parasympa-
FGFs, BMPs, Pax6 and Pax9 in normal mice and thetic or sympathetic denervation. Arch Oral Biol.
abnormal SMG phenotypes in FgfR2-IIIc(+/Delta), 1988;33(4):281–90.
BMP7(−/−) and Pax6(−/−) mice. Cells Tissues 56. Langendijk JA, Lambin P, De Ruysscher D, Widder
Organs. 2002;170(2–3):83–98. J, Bos M, Verheij M. Selection of patients for
45. Jaskoll T, Zhou YM, Trump G, Melnick radiotherapy with protons aiming at reduction of
M. Ectodysplasin receptor-mediated signaling is side effects: the model-based approach. Radiother
essential for embryonic submandibular salivary Oncol. 2013;107(3):267–73. doi:10.1016/j.
gland development. Anat Rec A Discov Mol Cell radonc.2013.05.007.
Evol Biol. 2003;271(2):322–31. doi:10.1002/ 57. Lombaert IM, Abrams SR, Li L, Eswarakumar
ar.a.10045. VP, Sethi AJ, Witt RL, Hoffman MP. Combined
46. Kang JH, Kim BK, Park BI, Kim HJ, Ko HM, Yang KIT and FGFR2b signaling regulates epithe-
SY, Kim MS, Jung JY, Kim WJ, Oh WM, Kim SH, lial progenitor expansion during organogenesis.
Kim JH. Parasympathectomy induces morphological Stem Cell Rep. 2013;1(6):604–19. doi:10.1016/j.
changes and alters gene-expression profiles in the rat stemcr.2013.10.013.
submandibular gland. Arch Oral Biol. 2010;55(1):7– 58. Lombaert IM, Brunsting JF, Wierenga PK, Faber H,
14. doi:10.1016/j.archoralbio.2009.11.003. Stokman MA, Kok T, Visser WH, Kampinga HH,
47. Kettunen P, Laurikkala J, Itaranta P, Vainio S, de Haan G, Coppes RP. Rescue of salivary gland
Itoh N, Thesleff I. Associations of FGF-3 and function after stem cell transplantation in irradiated
FGF-10 with signaling networks regulating tooth glands. PLoS One. 2008a;3(4):e2063. doi:10.1371/
morphogenesis. Dev Dyn. 2000;219(3):322–32. journal.pone.0002063.
doi:10.1002/1097-0177(2000)9999:9999<::AID- 59. Lombaert IM, Brunsting JF, Wierenga PK,
DVDY1062>3.0.CO;2-J. Kampinga HH, de Haan G, Coppes RP. Cytokine
48. Knosp WM, Knox SM, Lombaert IM, Haddox treatment improves parenchymal and vascular dam-
CL, Patel VN, Hoffman MP. Submandibular age of salivary glands after irradiation. Clin Cancer
parasympathetic gangliogenesis requires Res. 2008b;14(23):7741–50. doi:10.1158/1078-
sprouty-dependent Wnt signals from epithe- 0432.CCR-08-1449.
lial progenitors. Dev Cell. 2015;32(6):667–77. 60. Lombaert IM, Brunsting JF, Wierenga PK,
doi:10.1016/j.devcel.2015.01.023. Kampinga HH, de Haan G, Coppes RP. Keratinocyte
49. Knox SM, Lombaert IM, Haddox CL, Abrams SR, growth factor prevents radiation damage to salivary
Cotrim A, Wilson AJ, Hoffman MP. Parasympathetic glands by expansion of the stem/progenitor pool.
stimulation improves epithelial organ regenera- Stem Cells. 2008c;26(10):2595–601. doi:10.1634/
tion. Nat Commun. 2013;4:1494. doi:10.1038/ stemcells.2007-1034.
ncomms2493. 61. Lombaert IM, Movahednia MM, Adine C, Ferreira
50. Knox SM, Lombaert IM, Reed X, Vitale-Cross JN. Salivary gland regeneration: therapeutic
L, Gutkind JS, Hoffman MP. Parasympathetic approaches from stem cells to tissue organoids. Stem
innervation maintains epithelial progeni- Cells. 2016. doi:10.1002/stem.2455
tor cells during salivary organogenesis. 62. Maimets M, Rocchi C, Bron R, Pringle S, Kuipers
Science. 2010;329(5999):1645–7. doi:10.1126/ J, Giepmans BN, Vries RG, Clevers H, de Haan G,
science.1192046. van Os R, Coppes RP. Long-term in vitro expansion
51. Kobashi M, Ichikawa H, Kobashi M, Funahashi M, of salivary gland stem cells driven by Wnt signals.
Mitoh Y, Matsuo R. The origin of sensory nerve Stem Cell Rep. 2016;6(1):150–62. doi:10.1016/j.
fibers that innervate the submandibular salivary stemcr.2015.11.009.
gland in the rat. Brain Res. 2005;1060(1–2):184–7. 63. Makita T, Sucov HM, Gariepy CE, Yanagisawa M,
doi:10.1016/j.brainres.2005.08.012. Ginty DD. Endothelins are vascular-derived axo-
52. Kohn WG, Grossman E, Fox PC, Armando I, nal guidance cues for developing sympathetic neu-
Goldstein DS, Baum BJ. Effect of ionizing radiation rons. Nature. 2008;452(7188):759–63. doi:10.1038/
on sympathetic nerve function in rat parotid glands. nature06859.
J Oral Pathol Med. 1992;21(3):134–7. 64. Mandell E, Seedorf GJ, Ryan S, Gien J, Cramer
53. Kojima T, Kanemaru S, Hirano S, Tateya I, Suehiro SD, Abman SH. Antenatal endotoxin disrupts
A, Kitani Y, Kishimoto Y, Ohno S, Nakamura T, Ito lung vitamin D receptor and 25-hydroxyvita-
J. The protective efficacy of basic fibroblast growth min D 1alpha- hydroxylase expression in the
factor in radiation-induced salivary gland dysfunc- developing rat. Am J Physiol Lung Cell Mol
tion in mice. Laryngoscope. 2011;121(9):1870–5. Physiol. 2015;309(9):L1018–26. doi:10.1152/
doi:10.1002/lary.21873. ajplung.00253.2015.
65. Mandour MA, Helmi AM, El-Sheikh MM, 78. Patel N, Sharpe PT, Miletich I. Coordination of epi-
El-Garem F, El-Ghazzawi E. Effect of tympanic thelial branching and salivary gland lumen formation
neurectomy on human parotid salivary gland. by Wnt and FGF signals. Dev Biol. 2011;358(1):156–
Histopathologic, Histochemical, and Clinical Study. 67. doi:10.1016/j.ydbio.2011.07.023.
Arch Otolaryngol. 1977;103(6):338–41. 79. Patel VN, Hoffman MP. Salivary gland develop-
66. Martin KL, Hill GA, Klein RR, Arnett DG, Burd ment: a template for regeneration. Semin Cell
R, Limesand KH. Prevention of radiation-induced Dev Biol. 2014;25-26:52–60. doi:10.1016/j.
salivary gland dysfunction utilizing a CDK inhibitor semcdb.2013.12.001.
in a mouse model. PLoS One. 2012;7(12):e51363. 80. Patel VN, Lombaert IM, Cowherd SN, Shworak
doi:10.1371/journal.pone.0051363. NW, Xu Y, Liu J, Hoffman MP. Hs3st3-modified
67. Mathison R, Davison JS, Befus AD. Neuroendocrine heparan sulfate controls KIT+ progenitor expan-
regulation of inflammation and tissue repair by sion by regulating 3-O-sulfotransferases.
submandibular gland factors. Immunol Today. Dev Cell. 2014;29(6):662–73. doi:10.1016/j.
1994;15(11):527–32. devcel.2014.04.024.
68. McGovern AE, Mazzone SB. Neural regula- 81. Prockop DJ. Inflammation, fibrosis, and modu-
tion of inflammation in the airways and lungs. lation of the process by mesenchymal stem/
Auton Neurosci. 2014;182:95–101. doi:10.1016/j. stromal cells. Matrix Biol. 2016; doi:10.1016/j.
autneu.2013.12.008. matbio.2016.01.010.
69. Mitchell GC, Fillinger JL, Sittadjody S, Avila JL, 82. Proctor GB, Asking B. A comparison between
Burd R, Limesand KH. IGF1 activates cell cycle changes in rat parotid protein composition 1 and
arrest following irradiation by reducing binding of 12 weeks following surgical sympathectomy. Q
DeltaNp63 to the p21 promoter. Cell Death Dis. J Exp Physiol. 1989;74(6):835–40.
2010;1:e50. 83. Rebustini IT, Hayashi T, Reynolds AD, Dillard
70. Miyazaki Y, Nakanishi Y, Hieda Y. Tissue interaction ML, Carpenter EM, Hoffman MP. miR-200c regu-
mediated by neuregulin-1 and ErbB receptors regulates FGFR-dependent epithelial proliferation via
lates epithelial morphogenesis of mouse embryonic Vldlr during submandibular gland branching mor-
submandibular gland. Dev Dyn. 2004;230(4):591–6. phogenesis. Development. 2012;139(1):191–202.
doi:10.1002/dvdy.20078. doi:10.1242/dev.070151.
71. Nanduri LS, Lombaert IM, van der Zwaag M, Faber H, 84. Rebustini IT, Myers C, Lassiter KS, Surmak A,
Brunsting JF, van Os RP, Coppes RP. Salisphere derived Szabova L, Holmbeck K, Pedchenko V, Hudson
c-Kit+ cell transplantation restores tissue homeo- BG, Hoffman MP. MT2-MMP-dependent
stasis in irradiated salivary gland. Radiother Oncol. release of collagen IV NC1 domains regulates
2013;108(3):458–63. doi:10.1016/j.radonc.2013.05.020. submandibular gland branching morphogen-
72. Nanduri LS, Maimets M, Pringle SA, van der Zwaag esis. Dev Cell. 2009;17(4):482–93. doi:10.1016/j.
M, van Os RP, Coppes RP. Regeneration of irradi- devcel.2009.07.016.
ated salivary glands with stem cell marker express- 85. Reinert RB, Cai Q, Hong JY, Plank JL, Aamodt
ing cells. Radiother Oncol. 2011;99(3):367–72. K, Prasad N, Aramandla R, Dai C, Levy SE, Pozzi
doi:10.1016/j.radonc.2011.05.085. A, Labosky PA, Wright CV, Brissova M, Powers
73. Nedvetsky PI, Emmerson E, Finley JK, Ettinger AC. Vascular endothelial growth factor coordi-
A, Cruz-Pacheco N, Prochazka J, Haddox CL, nates islet innervation via vascular scaffolding.
Northrup E, Hodges C, Mostov KE, Hoffman MP, Development. 2014;141(7):1480–91. doi:10.1242/
Knox SM. Parasympathetic innervation regulates dev.098657.
tubulogenesis in the developing salivary gland. 86. Rodriguez-Diaz R, Caicedo A. Neural control of the
Dev Cell. 2014;30(4):449–62. doi:10.1016/j. endocrine pancreas. Best Pract Res Clin Endocrinol
devcel.2014.06.012. Metab. 2014;28(5):745–56. doi:10.1016/j.
74. Nordgarden H, Johannessen S, Storhaug K, Jensen beem.2014.05.002.
JL. Salivary gland involvement in hypohidrotic ecto- 87. Rothova M, Thompson H, Lickert H, Tucker
dermal dysplasia. Oral Dis. 1998;4(2):152–4. AS. Lineage tracing of the endoderm during oral
75. Nussenzweig SC, Verma S, Finkel T. The role of autoph- development. Dev Dyn. 2012;241(7):1183–91.
agy in vascular biology. Circ Res. 2015;116(3):480– doi:10.1002/dvdy.23804.
8. doi:10.1161/CIRCRESAHA.116.303805. 88. Rugel-Stahl A, Elliott ME, Ovitt CE. Ascl3 marks
76. Pagella P, Jimenez-Rojo L, Mitsiadis TA. Roles of adult progenitor cells of the mouse salivary gland.
innervation in developing and regenerating orofacial Stem Cell Res. 2012;8(3):379–87. doi:10.1016/j.
tissues. Cell Mol Life Sci. 2014;71(12):2241–51. scr.2012.01.002.
doi:10.1007/s00018-013-1549-0. 89. Ryu YK, Collins SE, Ho HY, Zhao H, Kuruvilla R. An
77. Palaniyandi S, Odaka Y, Green W, Abreo F, Caldito G, autocrine Wnt5a-Ror signaling loop mediates sympa-
De Benedetti A, Sunavala-Dossabhoy G. Adenoviral thetic target innervation. Dev Biol. 2013;377(1):79–
delivery of Tousled kinase for the protection of sali- 89. doi:10.1016/j.ydbio.2013.02.013.
vary glands against ionizing radiation damage. Gene 90. Steinberg Z, Myers C, Heim VM, Lathrop CA,
Ther. 2011;18(3):275–82. doi:10.1038/gt.2010.142. Rebustini IT, Stewart JS, Larsen M, Hoffman
22 I.M.A. Lombaert
MP. FGFR2b signaling regulates ex vivo submandib- 100. Wells KL, Mou C, Headon DJ, Tucker
ular gland epithelial cell proliferation and branching AS. Recombinant EDA or Sonic Hedgehog res-
morphogenesis. Development. 2005;132(6):1223– cue the branching defect in Ectodysplasin A path-
34. doi:10.1242/dev.01690. way mutant salivary glands in vitro. Dev Dyn.
91. Sugimoto T, Taya Y, Shimazu Y, Soeno Y, Sato K, 2010;239(10):2674–84. doi:10.1002/dvdy.22406.
Aoba T. Three-dimensional visualization of devel- 101. Wright DM, Buenger DE, Abashev TM, Lindeman
oping neurovascular architecture in the craniofacial RP, Ding J, Sandell LL. Retinoic acid regulates
region of embryonic mice. Anat Rec (Hoboken). embryonic development of mammalian submandib-
2015;298(11):1824–35. doi:10.1002/ar.23179. ular salivary glands. Dev Biol. 2015;407(1):57–67.
92. Sumita Y, Liu Y, Khalili S, Maria OM, Xia D, Key S, doi:10.1016/j.ydbio.2015.08.008.
Cotrim AP, Mezey E, Tran SD. Bone marrow-derived 102. Wynn TA, Chawla A, Pollard JW. Macrophage biol-
cells rescue salivary gland function in mice with ogy in development, homeostasis and disease. Nature.
head and neck irradiation. Int J Biochem Cell Biol. 2013;496(7446):445–55. doi:10.1038/nature12034.
2011;43(1):80–7. doi:10.1016/j.biocel.2010.09.023. 103. Xiao N, Lin Y, Cao H, Sirjani D, Giaccia AJ, Koong
93. Sun W, Ni X, Sun S, Cai L, Yu J, Wang J, Nie B, Sun AC, Kong CS, Diehn M, Le QT. Neurotrophic fac-
Z, Ni X, Cao X. Adipose-derived stem cells allevi- tor GDNF promotes survival of salivary stem cells.
ate radiation-induced muscular fibrosis by suppress- J Clin Invest. 2014;124(8):3364–77. doi:10.1172/
ing the expression of TGF-beta1. Stem Cells Int. JCI74096.
2016;2016:5638204. doi:10.1155/2016/5638204. 104. Xue J, Li X, Lu Y, Gan L, Zhou L, Wang Y, Lan
94. Talavera-Adame D, Dafoe DC. Endothelium-derived J, Liu S, Sun L, Jia L, Mo X, Li J. Gene-modified
essential signals involved in pancreas organogenesis. mesenchymal stem cells protect against radiation-
World J Exp Med. 2015;5(2):40–9. doi:10.5493/ induced lung injury. Mol Ther. 2013;21(2):456–65.
wjem.v5.i2.40. doi:10.1038/mt.2012.183.
95. Tanida S, Kataoka H, Mizoshita T, Shimura T, 105. Yamamoto S, Fukumoto E, Yoshizaki K, Iwamoto
Kamiya T, Joh T. Intranuclear translocation signaling T, Yamada A, Tanaka K, Suzuki H, Aizawa S,
of HB-EGF carboxy-terminal fragment and muco- Arakaki M, Yuasa K, Oka K, Chai Y, Nonaka
sal defense through cell proliferation and migration K, Fukumoto S. Platelet-derived growth fac-
in digestive tracts. Digestion. 2010;82(3):145–9. tor receptor regulates salivary gland morphogen-
doi:10.1159/000310903. esis via fibroblast growth factor expression. J Biol
96. Tucker AS. Salivary gland development. Semin Chem. 2008;283(34):23139–49. doi:10.1074/jbc.
Cell Dev Biol. 2007;18(2):237–44. doi:10.1016/j. M710308200.
semcdb.2007.01.006. 106. Yan X, Liu Y, Han Q, Jia M, Liao L, Qi M, Zhao
97. van Luijk P, Pringle S, Deasy JO, Moiseenko VV, RC. Injured microenvironment directly guides the
Faber H, Hovan A, Baanstra M, van der Laan differentiation of engrafted Flk-1(+) mesenchymal
HP, Kierkels RG, van der Schaaf A, Witjes MJ, stem cell in lung. Exp Hematol. 2007;35(9):1466–
Schippers JM, Brandenburg S, Langendijk JA, Wu 75. doi:10.1016/j.exphem.2007.05.012.
J, Coppes RP. Sparing the region of the salivary 107. Yancopoulos GD, Davis S, Gale NW, Rudge JS,
gland containing stem cells preserves saliva produc- Wiegand SJ, Holash J. Vascular-specific growth
tion after radiotherapy for head and neck cancer. Sci factors and blood vessel formation. Nature.
Transl Med. 2015;7(305):305ra147. doi:10.1126/ 2000;407(6801):242–8. doi:10.1038/35025215.
scitranslmed.aac4441. 108. Yun EJ, Lorizio W, Seedorf GJ, Abman SH, Vu
98. Wang H, Yang YF, Zhao L, Xiao FJ, Zhang QW, TH. VEGF and endothelium-derived retinoic acid
Wen ML, Wu CT, Peng RY, Wang LS. Hepatocyte regulate lung vascular and alveolar development.
growth factor gene-modified mesenchymal stem Am J Physiol Lung Cell Mol Physiol:ajplung.
cells reduce radiation-induced lung injury. Hum 2015;00229:02015. doi:10.1152/ajplung.00229.2015.
Gene Ther. 2013;24(3):343–53. doi:10.1089/ 109. Zheng C, Cotrim AP, Rowzee A, Swaim W,
hum.2012.177. Sowers A, Mitchell JB, Baum BJ. Prevention of
99. Wells KL, Gaete M, Matalova E, Deutsch D, Rice D, radiation-
induced salivary hypofunction follow-
Tucker AS. Dynamic relationship of the epithelium ing hKGF gene delivery to murine submandibu-
and mesenchyme during salivary gland initiation: lar glands. Clin Cancer Res. 2011;17(9):2842–51.
the role of Fgf10. Biol Open. 2013;2(10):981–9. doi:10.1158/1078-0432.CCR-10-2982.
doi:10.1242/bio.20135306.
Systems Biology: Salivary Gland
Development, Disease, 2
and Regenerative Medicine
Melinda Larsen, Petko Bogdanov, Ravi Sood,

Hae Ryong Kwon, Deirdre A. Nelson,
Connor Duffy, Sarah B. Peters,
and Sridar V. Chittur
Abstract
There are multiple challenges currently facing clinical salivary gland
research, encompassing a wide range of topics from understanding devel-
opment to understanding disease etiology and from diagnosing disease to
designing more effective, personalized therapies. Systems analysis com-
plements traditional reductionist approaches, and the integration of these
two approaches is starting to provide a more comprehensive understand-
ing of the causal relationships leading to normal and abnormal biology.
Understanding normal developmental processes is critical for understand-
ing development of disease. Morphogenesis and differentiation are com-
plex developmental processes involving orchestrated interactions between
heterotypic cell types that have proven difficult to understand through
reductionist approaches alone. It has become clear that it is not possible to
understand the complex molecular, cellular, and physical process integra-
tion that is required for any developmental or disease process without the
use of systems biology approaches. In this chapter, we demonstrate exam-
ples in the use of systems approaches to better characterize the difference
M. Larsen, PhD (*) • H.R. Kwon, PhD

D.A. Nelson • C. Duffy
Department of Biological Sciences,
University at Albany, SUNY, S.B. Peters, PhD
State University of New York, Department of Biological Sciences, University at Albany,
1400 Washington Ave LSRB 1086, SUNY, State University of New York, 1400 Washington
Albany, NY 12222, USA Ave LSRB 1086, Albany, NY 12222, USA
Oklahoma Medical Research Foundation, Department of Cell, Developmental, and Integrative
Oklahoma, OK, USA Biology, University of Alabama at Birmingham,
e-mail: [email protected] 1918 University Blvd., MCML 643, Birmingham,
AL 35294, USA
P. Bogdanov, PhD • R. Sood
Department of Computer Science, S.V. Chittur, PhD
University at Albany, SUNY, Center for Functional Genomics, University
State University of New York, at Albany, SUNY, State University of New York,
1400 Washington Ave, Health Sciences Campus, One Discovery Drive,
Albany, NY 12222, USA Albany, NY 12144, USA

24 M. Larsen et al.
between embryonic submandibular salivary glands grown in vivo and

ex vivo as 3D organ explants and to identify potential signaling networks
in heterotypic subpopulations of cells that lead to the prediction of a func-
tion for endothelial cells in salivary gland development. We conclude with
examples of how systems biology-based approaches using both tissue
samples and saliva from patients are currently being used in many labora-
tories to make progress in salivary gland disease diagnosis, understanding
disease etiology, and informing therapeutic development for cancer and
regenerative medicine strategies.
2.1 Introduction tems biology with reductionist approaches is

proving to provide a more comprehensive
Knowledge of developmental mechanisms is crit- understanding of the causal relationships leading
ical to inform effective therapies for regeneration to normal and abnormal organogenesis.
and repair of damaged glands [66]. During Animal models that have provided insights
embryogenesis, the information encoded in the relevant to human salivary gland development
genome is converted into a 3D organ, where include the mouse, rat, and the fruit fly, Drosophila
primitive clusters of cells undergo morphogene- melanogaster, but most recent studies have used
sis and differentiation to acquire mature structure the mouse model due to its relative similarity to
and function. Embryonic development of secre- humans and its tractable genetics. The mouse sub-
tory and absorptive organs including the salivary mandibular salivary gland begins its development
glands, kidneys, lungs, mammary glands, pros- at embryonic day 11.5 (E11.5) (with the day of
tate, and lacrimal glands depends on a process coital plug discovery defined as E0) as a thicken-
known as branching morphogenesis [11]. ing of the primitive oral epithelium that grows into
Branching morphogenesis allows organs to max- the first branchial (mandibular) arch mesenchyme
imize epithelial surface area while efficiently to form the solid epithelial placode (reviewed in
minimizing volume within the tissue. [21, 27, 30, 34, 67, 87]). This placode protrudes
Differentiation begins with changes in gene into the mesenchyme by E12, forming a single,
expression during early development that trans- solid mass of immature epithelial cells connected
late into differential protein expression, allowing to the tongue epithelium by a stalk of immature
cells to take on specific functions. Morphogenesis epithelial cells. By E12.5, alterations in the base-
and differentiation are complex processes involv- ment membrane form indentations on the surface
ing orchestrated interactions between multiple of the bud termed clefts. The clefts are essential to
cell types that have proven to be difficult to gland development since, as they progress to sep-
understand through reductionist approaches arate the primary bud into multiple buds [10, 37],
alone, and systems biology approaches are they define the boundary between developing pro-
needed to facilitate understanding of the complex acini and ducts. As cleft progression proceeds, the
molecular, cellular, and physical process integra- epithelium proliferates, and the base of the cleft
tions required for organ development. Systems becomes the nascent ductal structure. This pro-
biology can be defined in many ways, but cess of salivary gland branching morphogenesis is
attempts to integrate molecular-level knowledge repeated multiple times over the course of several
of genes, RNAs, and proteins with biophysical days, continually increasing the complexity of the
and cellular processes into networks of interact- organ. By late E14, the simple one-bud one-duct
ing components are the essence of systems biol- salivary gland has both grown and branched sig-
ogy approaches. Systems analysis complements nificantly, and the main duct begins to lumenize.
traditional approaches, and the integration of sys- The end buds undergo reorganization and begin to
2 Systems Biology: Salivary Gland Development, Disease, and Regenerative Medicine 25
form acini – the main secretory units of the sali- primary tumors of the major salivary glands are
vary gland. By E15–16, lumenization of the main relatively rare, accounting for 11 % of oropharyn-
secretory duct is nearly complete, and by E17, the geal neoplasms in the USA [7]. A distinguishing
forming acini lumenize, so that the gland has a feature of salivary gland neoplasms is that they are
continuous network of lumenized, but still imma- highly histologically variable due to the significant
ture, acini and ducts connecting to the oral cavity. variance in their cellular origins and presumably
Both nerves and blood vessels populate the gland variable etiology. As a result, histological classifi-
in association with the branching epithelium, and cation of salivary gland tumors is challenging and
the nerves are known to be critically involved in subject to misdiagnosis by nonexperts [84, 85].
the elaboration of organ structure [32]. Cellular Although potentially invaluable for tumor classifi-
differentiation occurs in parallel with branching cation, molecular signatures for salivary tumors
morphogenesis through pathways that are at least are lacking, similar to the lack of definitive mark-
partially independent [9] as the glands continue ers for SS.
to mature after birth. The developmental program Other than patients with malignant tumors, the
is complex, with multiple processes occurring most significant problems for patients with sali-
simultaneously. vary gland disease typically result from salivary
Salivary gland function is equally complex in hypofunction, or a decrease in or loss of salivary
the adult as are the mechanisms leading to multi- function. Hyposalivation, or decreased saliva
ple forms of dysfunction. Multiple epithelial cell production, occurs in patients that have SS, and
types are required to make saliva and for regula- also as a side effect following radiation therapy
tion of exocrine secretion involving both sympa- for head and neck cancers. Loss of salivary flow
thetic and parasympathetic innervation [73]. Saliva results in “dry mouth” and affects additional
production begins with the secretion of incomplete processes, such as mastication and swallowing.
saliva by submandibular acinar cells, including Secondary oral disease states can subsequently
contributions from both serous and mucous acinar develop, including sialadenitis, dental caries,
cells in human glands. This primary fluid is modi- periodontal disease, and persistent oral infections
fied by ductal cells as it is transported to the oral [61], leading to a general decline in the quality of
cavity, where it mixes with secretions from other life. The molecular basis for hyposalivation in
salivary glands to generate whole saliva. Adult patients suffering from SS and the mechanism of
salivary glands can be affected by viral and bacte- the selective destruction of salivary acinar cells
rial infections, inflammation, autoimmune disease, leading to hyposalivation as a result of radiation
and tumorigenesis. Infections of the salivary gland, therapy remain poorly understood. There are few
or sialadenitis, can occur from bacterial, viral, fun- satisfactory therapeutic options for these patients.
gal, or other causes. Sjögren’s syndrome (SS) is a There are multiple challenges currently facing
complex systemic autoimmune disease affecting salivary gland research encompassing a wide
primarily the salivary glands and the lacrimal range of topics from understanding development
glands with an elusive etiology [13, 40]. SS is dif- to diagnosing human disease and developing
ficult to diagnose until it is in its later stages due to more effective therapies. Understanding disease
many factors, including a lack of molecular mark- pathogenesis and developing rational therapies
ers. Up to 90 % of SS patients express antibodies will require an understanding of the emergent
targeting autoantigens [52], but these autoantibod- properties of these interactions, which is the
ies are not unique to SS. SS patients may also be strength of systems biology. Since this topic was
affected by other autoimmune diseases, and such last reviewed [38], systems-based approaches
patients are classified as secondary SS patients. have been widely incorporated into many studies
Around 5 % of SS patients typically progress to a addressing basic biological mechanisms as well
non-Hodgkin’s, mucosa-associated lymphoid tis- as translational studies that address salivary gland
sue (MALT) lymphoma [33]. Independent from disease pathogenesis and therapeutic develop-
SS, salivary gland tumors can occur. In general, ment. In this chapter, we will demonstrate
26 M. Larsen et al.
e xamples of how systems-based approaches can regulated proliferation and differentiation of pro-
be used to understand the complexities of sali- genitor cell populations. Growth of salivary gland
vary gland organogenesis and disease. cells in small clusters of cells known as organoids
Specifically, we will demonstrate how we have or “salispheres” is currently being used as a use-
used systems- based methods to evaluate the ful method of preserving cell phenotype through
embryonic mouse submandibular salivary gland homotypic and, in some cases, heterotypic cell
organ explants as a model system for the study of interactions. Although organoids are further
in vivo biology and how data mining of existing removed from the in vivo conditions than organ
databases can be used to generate novel hypoth- explants, they offer advantages for maintaining
esis regarding mechanisms of development. We stem/progenitor cell populations. Both organ
will also review how others have used systems- explants and organoids offer the opportunity to
based profiling and modeling approaches to gain manipulate multiple genes in a moderate-
novel insights into mechanisms involved in glan- throughput manner via pharmacological [36],
dular differentiation and salivary gland diseases RNA-mediated knockdown [10, 22, 78], viral
and how systems-based evaluation of saliva is transduction [23, 28, 80], and other means.
useful potentially not only for diagnosis of sali-
vary diseases but of systemic conditions.
2.2.2 Transgenic Mouse Models
2.2 Current Reductionist Numerous transgenic mouse models have identi-

Approaches to Understand fied critical genetic mediators of salivary gland
Salivary Gland Development development (reviewed in [58]). Recent studies
have additionally employed transgenic mouse
2.2.1 O
rgan Explants, Organoid models encoding fluorescent lineage reporters to
Culture, and Other Primary characterize molecular mediators of heterotypic
Cultures cell interactions and progenitor cell function that
control salivary gland development and regenera-
Traditionally, cell and developmental biologists tion [3, 31, 32, 62], providing important insights
have taken a reductionist approach to understand- for improved cell therapy and regenerative medi-
ing the development and function of salivary cine strategies. The future looks bright for the con-
glands. This approach – rooted in the scientific tinued use of transgenic animals in salivary gland
method itself – has provided useful information research. The introduction of CRISPR/Cas-based
regarding linear signaling pathways. First estab- methods for manipulation of the mouse genome
lished in the 1950s [5, 6, 19, 20], the embryonic promises to make the process of generating trans-
submandibular gland organ explant culture sys- genic animals more efficient [81]. New promoters
tem has provided researchers with a 3D experi- are increasingly becoming available to be able to
mental system that can be used to ask in-depth, target specific salivary gland cell populations [51].
complex questions about the control of morpho- The disadvantage of the transgenic approach is that
genesis. Organ explants have been useful to study only one gene can be manipulated at a time; how-
and identify many signaling pathways that are ever, the utilization of CRISPR/Cas methods prom-
critical during early development since they pre- ises to make the manipulation of multiple genes at
serve heterotypic cell-cell interactions that occur a time [95] much more efficient and cost effective
in vivo and the morphogenesis that occurs closely than using traditional transgenic technologies.
mimics the in vivo process. More recently, it has Manipulation of genes within model organisms
been demonstrated that the early differentiation occurs within the context of the organismal system,
of multiple epithelial cell types occurs in the and such manipulations can have widespread
developing organoids [70]. Embryonic develop- effects. Hence, there is a significant need to apply
ment of the salivary glands depends upon the systems-level analysis to transgenic organisms.
2.3 Systems-Based Approaches on our prior work that demonstrated that devel-
to the Study of Gland opment of the mouse submandibular salivary
Development gland is mechanically sensitive. Both epithelial
morphogenesis and differentiation proceed the
2.3.1 S
alivary Gland Molecular most similar to in vivo when glands are grown
Anatomy Project on a substrate having a stiffness that mimics that
of the in vivo environment [70, 71]. We com-
Gene profiling is a powerful tool for researchers pared the genome-wide differential expression
studying both developmental processes and human of individual mRNAs of embryonic glands
disease to identify the transcriptome. Developed grown in vivo from E13 through E15, as reported
by Matthew Hoffman and Kenneth Yamada in the in SGMAP,1 with E13 e mbryonic organ explants
intramural research program at the NIDCR, the cultured on a 0.5 kPa polyacrylamide gel for 24
Salivary Gland Molecular Anatomy Project or 72 h, as previously described [70, 71]. We
(SGMAP) (http://sgmap.nidcr.nih.gov) provides a quantified mRNA expression using Affymetrix
searchable gene expression database for multiple Mouse Gene ST 2.0 gene arrays. Our microar-
stages of mouse submandibular and sublingual ray data was normalized using GeneSpring
salivary gland development that is an invaluable v12.6, first being quantile normalized using a
tool for researchers. Temporal expression patterns PLIER16 algorithm and baseline transformed to
of genes are provided for salivary glands extracted the median of all samples. The log2 normalized
from mice at embryonic day (E) 11.5 through signal values were filtered to remove entities
adult that were obtained using Agilent mouse that show signal in the bottom 20th percentile
microarrays. The SGMAP website allows for across all samples. We then compared gene
searching of any gene symbol, gene description, or expression levels from the organ explants with
gene ontology (GO) term [2] to obtain information data from SGMAP.
on its developmental expression, expression in Comparison of different datasets that were
epithelium vs mesenchyme at E13, or its spatial obtained from different types of samples and that
expression in specific regions of the gland at spe- were processed independently is challenging
cific developmental time points [59]. with microarray data. Ideally, the datasets should
be compared starting with raw data and then pro-
cessed in parallel [96]. In comparing our micro-
array data with the SGMAP data, both datasets
2.3.2 Comparison of the were first normalized to the same scale to facili-
transcriptome of embryonic salivary tate the comparison. Genes from the in vivo data
glands grown in vivo as compared for which no data were available were omitted
with that of embryonic organ from the comparison, which corresponded to
explants grown ex vivo 10.9 % or 2462 of 22,510 genes. Both datasets
were then standardized to a zero mean and a
Although organ explants have been in use for standard deviation of one, and then the expres-
many years, it has not been clear how closely the sion levels in individual developmental stages
explant culture system mimics in vivo develop- were scaled to the interval [0, 1]. The in vivo and
ment. To address this question, we compared ex vivo datasets were cross-matched, retaining
our own gene expression data obtained from only shared genes for the analysis (13,709
embryonic mouse submandibular salivary genes).
glands grown on a polyacrylamide gel con- We first compared the corresponding in vivo
structed to mimic the compliance of the in vivo and ex vivo overall distributions of differential
environment vs gene expression data obtained expression between the in vivo dataset from E13 to
from glands that were grown in vivo that is
available from SGMAP. This study was founded http://sgmap.nidcr.nih.gov/sgmap/sgexp.html
1
28 M. Larsen et al.
3000
In-vivo
2500 Ex-vivo
Number of genes 2000
1500
1000
500
0
-0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8
Differential expression
Fig. 2.1 Comparison of the differential gene expression differentially expressed in either condition. The ex vivo
of all genes expressed in developing salivary glands from mode peaks higher, indicating that there are a greater
E13 to E15 in vivo vs ex vivo. All genes showing develop- number of genes that are not significantly differentially
mental differential gene expression were plotted. Both expressed as compared to in vivo, while the in vivo distri-
distributions appear to have a single mode near 0, indicat- bution is broader, indicating a greater range of differential
ing that it is most common for a gene not to be s ignificantly expression in comparison with ex vivo growth
E15 and the ex vivo data set grown for 24–72 h. As We summarize the annotations of the top 10 %
shown in Fig. 2.1, the overall shapes of the two differentially expressed genes ex vivo and the top
distributions are similar; however, there are more 10 % in vivo genes (Fig. 2.2). The percentage of
genes having higher differential expression that annotations that occur in the top 10 % ex vivo are
are both upregulated and downregulated in the similar to those that occur in the top 10 % in vivo
in vivo dataset than the ex vivo dataset. This analy- across all level 1 biological process GO terms. In
sis suggests that gene expression levels are gener- other words, at least at a high level, genes involved
ally tempered in glands cultured ex vivo compared in the same processes are changing in expression
to glands grown in vivo. levels both in vivo and ex vivo.
Given that there are notable differences in the To understand which biological processes were
global distributions of gene expression between the most conserved and less conserved in subman-
glands grown in vivo vs ex vivo, we attempted to dibular salivary glands grown ex vivo as compared
identify what groups of genes were different. to in vivo, we performed a Kolmogorov-Smirnov
Differential gene expression for each dataset was (KS) test on the genes within each of the level 1 GO
defined as a significant difference between the categories. The KS test quantifies how dissimilar
24 h and the 72 h culture condition for the ex vivo two sample distributions are, with a smaller KS
culture and a significant difference between the indicating high similarity and a larger KS term indi-
E13 and E15 in vivo gene expression time points. cating lesser similarity. Since the mean KS for all
We compared the functional agreement of the genes was 0.128, we considered biological pro-
most differentially expressed genes based on their cesses with a lower KS than 0.128 and a p-value of
biological process annotations in the gene ontol- p < 0.05 to be conserved. As shown in Fig. 2.3, the
ogy (GO) [2]. We performed this GO term analy- Kolmogorov-Smirnov test demonstrated that the
sis for the 10 % most differentially expressed most conserved biological processes in glands
genes (i.e., those genes whose expression grown ex vivo vs in vivo were included in the GO
decreased or increased most drastically between terms for growth (KS 0.101), metabolic processes
the two time points). Agreement was most signifi- (KS 0.111), developmental processes (KS 0.118),
cant for the top 10 % of genes ranked by differen- cellular processes (KS 0.119), and biological regu-
tial expression ex vivo, where 35 % of such genes lation (KS 0.124). As genes included in these
were also in the top 10 % in vivo (data not shown). categories would be predicted to be essential for
Ex-vivo annotation percentages in

highly differentially expressed genes by term
Reproduction: ex-vivo: 0.0 % in-vivo: 0.0 %
Rhythmic process: ex-vivo: 0.5 % in-vivo: 0.4 %
Reproductive process: ex-vivo: 1.5 % in-vivo:1.5 %
Single-organism process: ex-vivo: 15.7 % in-vivo: 16.1 %
Behavior: ex-vivo: 1.0 % in-vivo: 1.1 %
Growth: ex-vivo: 1.0 % in-vivo: 0.7 %
Signaling: ex-vivo: 0.9 % in-vivo: 0.7 %
Multicellular organismal process: ex-vivo: 6.1 % in-vivo: 7.2 %
Biological regulation: ex-vivo: 15.1 % in-vivo: 15.5 %
Cellular process: ex-vivo: 15.5 % in-vivo: 15.2 %
Cellular component organization or biogenesis: ex-vivo: 6.2 % in-vivo: 5.2 %
Development process: ex-vivo: 8.3 % in-vivo: 8.0 %
In-vivo annotation percentages in Metabolic process: ex-vivo: 9.3 % in-vivo: 9.4 %

highly differentially expressed genes by term Localization: ex-vivo: 5.0 % in-vivo: 5.2 %
Biological phase: ex-vivo: 0.0 % in-vivo: 0.0 %
Response to stimulus: ex-vivo: 5.9 % in-vivo: 6.1 %
Multi-organism process: ex-vivo: 1.1 % in-vivo: 1.1 %
Biological adhesion: ex-vivo: 2.4 % in-vivo: 2.0 %
Cell killing: ex-vivo: 0.1 % in-vivo: 0.1 %
Locomotion: ex-vivo: 2.3 % in-vivo: 2.0 %
Immune system process: ex-vivo: 2.2 % in-vivo: 2.1 %
Cell aggregation: ex-vivo: 0.1 % in-vivo 0.1 %
Fig. 2.2 Comparison of the biological process represen- the relative importance of each process ex vivo is similar
tation in differentially expressed in vivo and ex vivo to the in vivo situation. Only the top 10 % of the most dif-
genes. We considered the known annotations of genes ferentially expressed genes were considered. Genes with
with GO terms at the first level of the GO biological pro- more specific annotations at ontology levels deeper than 1
cess ontology and show the percentage of annotations for were considered to be annotated with the corresponding
each term in vivo and ex vivo. This analysis indicates that ancestor biological process at level one for this analysis
development, this analysis validates the mouse system, it is also not surprising that expressions of
embryonic submandibular salivary gland as a valid genes classified in the categories, “Immune System
model system to evaluate developmental processes. Process” and “Behavior,” are somewhat different
The less-conserved biological processes, as deter- in vivo and ex vivo. This systems-based analysis of
mined by a KS value greater than the mean of 0.128 gland development ex vivo confirms that many gen-
with a p-value of <0.05, are biological processes eral biological processes that are required for devel-
whose expression does not agree as well in ex vivo opment, including cell growth, metabolic processes,
as in vivo (Fig. 2.4). At the top of this list were the and other developmental processes, occur in organ
GO categories, cell killing (KS 0.417), immune sys- explants similarly to how they occur in vivo, while
tem process (KS 0.202), and behavior (KS 0.167). other processes that require interaction with the rest
As the organ explants do not maintain an intact of the body system are somewhat different.
immune system, nor do they remain c onnected to To examine how similar the differential gene
the nervous system, vascular system, or lymphatic expression is for specific genes in vivo vs ex vivo,
30 M. Larsen et al.
Fig. 2.3 Comparison 0.4

of the in vivo and In vivo
ex vivo differential
gene expression in GO Ex vivo
processes of highest
agreement. Genes 0.2
associated with a given
biological process at
the first level of the GO
hierarchy are included
in the comparison for 0.0
each of the
distributions. The pairs
of distributions of all
included processes have
a Kolmogorov-Smirnov −0.2
test distance of 0.13 or
less and associated
p-value <0.05
−0.4
Growth
process
Developmental
process
process
regulation
Localization
Single−organism
process
Cellular
component
organization
or biogenesis
Cellular
Biological
Metabolic
0.4
In vivo
Ex vivo
0.2
Fig. 2.4 Comparison of

the in vivo and ex vivo
differential gene 0.0
expression in GO
processes of lowest
agreement. Genes
associated with a given
biological process at the −0.2
first level of the GO
hierarchy are included
in the comparison for
each of the distributions.
The pairs of −0.4
distributions of all
Cell
killing
Immune
system
process
Behavior
Biological
adhesion
Reproductive
process
Multicellular
organismal
process
Signaling
Response
to stimulus
Locomotion
included processes have

a Kolmogorov-Smirnov
test distance higher than
0.14 and associated
p-value <0.05
Color key
-0.2 0 0.1
Value
Bcl9l
Ascl2
Kit
DII 1
Sox9
Six2
Lrp5
Spi1
Elf5
Tfap2c
Mmp24
Tbx3
Prdm16
lgf1
Cdh2
Mecom
Hmga2
Fgf10
Ldb2
In vivo Ex vivo
Fig. 2.5 Hierarchical clustering of differentially distinct groups including a preferentially upregulated
expressed mRNAs in “Maintenance of Cell Number” GO gene group and a downregulated gene group, suggesting
category. The top 10 % of differentially expressed genes that differentially expressed genes between in vivo and
were selected from the “Maintenance of Cell Number” ex vivo are relatively well conserved. This is consistent
category (GO:0098727), a subcategory of “Developmental with the high agreement of the parent GO class
Process” category (GO:0032502), one of the most con- “Developmental Process” category (GO:0032502) in
served categories (Fig. 2.3), to perform hierarchical clus- Kolmogorov-Smirnov test in Fig. 2.3. However, note that
tering, and displayed as a heatmap using the heatmap.2 individual genes are regulated slightly differently in vivo
function in R. The hierarchical clustering provided two and ex vivo
we examined the top three most conserved GO (Fig. 2.5), demonstrates that on the individual
categories and divided them into subcategories gene level, there are some genes that are regu-
based on the next tier of GO descriptors. As an lated more similarly than others. The analysis we
example, we plotted differences in gene expres- report here of differences between differential
sion of genes annotated with GO process gene expressions in organ explants vs glands
“Maintenance of Cell Number,” a subcategory of grown in vivo is useful in understanding how the
“Developmental Process.” The heat map of the transcriptome is regulated similarly in vivo and
subcategory, “Maintenance of Cell Number” in vitro.
32 M. Larsen et al.
2.3.3 Systems-Level Analysis cleft-enriched genes, genes were partitioned into

to Identify Temporal-Spatial epithelium and mesenchyme by the comparison of
Differences in Gene E13 epithelium and E13 mesenchyme gene expres-
Expression During Salivary sion profiles available in the SGMAP database.
Gland Development Surprisingly, over half of the cleft-enriched mRNAs
(specifically, 71.3 % of the conserved cleft-enriched
Given that during embryonic development, het- genes) showed mesenchymal origins, supporting a
erotypic epithelial and mesenchymal cell sub- hypothesis that mesenchymal genes dynamically
types undergo dynamic communication [63], contribute to epithelial patterning through regula-
knowledge of how specific cell subpopulations tion of cleft formation (Fig. 2.6c). Additional func-
contribute to gland development and how these tional prediction of the cleft-enriched genes was
cell subpopulations are regulated is critical to performed with an available integrated analysis
understand gland development [34]. Cleft forma- tool, the Database for Annotation, Visualization
tion initiates the process of branching and Integrated Discovery (DAVID) system (ver.
morphogenesis and subdivides buds into cells
6.7) (https://david.ncifcrf.gov). Results of gene
that will become nascent secondary ducts and ontology (GO) term and Kyoto Encyclopedia of
cells that will become secretory acini. In an early Genes and Genomes (KEGG) pathway analysis by
systems- level analysis using serial analysis of DAVID [26] confirmed the involvement of many
gene expression (SAGE), fibronectin (FN) was mRNAs known to contribute to clefts and identi-
identified as a critical regulator of cleft formation fied some potential new cellular mechanisms to
[37, 77, 78]. Cleft formation was also found to be consider in cleft formation. This analysis first con-
accompanied by a concomitant loss of adjacent firmed the enrichment of genes involved in focal
E-cadherin-based cell-cell junctions [78]. This adhesions, ECM-receptor interactions, and the
conversion of cell-cell adhesions to cell-matrix actin cytoskeleton (Fig. 2.7 and Table 2.1), consis-
adhesions was found to be regulated transcrip- tent with previous observations by our lab and oth-
tionally through increases in BTB (POZ) domain ers [28, 59, 64, 75, 79].
containing 7 (Btbd7) [64]. Btbd7, which was also Interestingly, GO biological process analysis
identified as a cleft-enriched gene using SAGE, is highlighted the cleft-enriched localization of spe-
reported was suggested to activate a local cialized mesenchymal cell types, including endo-
epithelial-to-
mesenchymal transition (EMT) thelial cells, nerves, and immune cells (Fig. 2.6,
within the epithelial cells deep within the cleft to Tables 2.1, and 2.2, and 2.3). The cleft endothelial/
separate the adjacent cells to allow continuous blood vessel signature predicts a possible contribu-
FN assembly [64] and cleft progression. tion of endothelial cells to salivary gland cleft for-
To identify additional genes that may be mation (Table 2.2), which is supported by the
required for cleft formation, we performed in silico known contribution of endothelial cells to the early
data mining of the publically available SGMAP development of other organs, including the liver,
temporal and spatial data. Temporal and spatial pancreas, and lung, prior to the establishment of
mRNA expression profiles derived from laser cap- perfusion [35, 39, 45, 53, 74]. Recent studies have
ture microdissected regions of embryonic subman- demonstrated that parasympathetic ganglia inner-
dibular salivary glands were used to screen for vation regulates epithelial cytokeratin 5+ (K5+) pro-
cleft-enriched genes at developmental stages E12.5 genitor cell expansion via muscarinic and epidermal
and E13.5 [59] (Fig. 2.6a). Differentially expressed growth factor receptor signaling and controls gland
mRNAs were identified in E12.5 cleft regions tubulogenesis through vasoactive intestinal peptide
compared to the main bud (528 mRNAs) or E13.5 (VIP) secretion [32, 62]. However, a specific con-
clefts compared to the central bud/basement mem- tribution for nerve-derived factors in cleft forma-
brane bud (1576 genes). Of these cleft-enriched tion itself is still unclear but is implicated by this
mRNAs, 317 were common to both E12.5 and data (Table 2.3). Endothelial cells and/or nerves
E13.5 clefts (Fig. 2.6b). Among the putative may be important in early patterning of the clefts
a b
SGMAP database
Data collection 211 317 1083
HTML files
Data extraction (Perl code)
E12.5 cleft E13.5 cleft
Gene expression profiles -enriched -enriched
genes (528) genes (1,576)
1) Temporal and spatial gene expression
(TSGE) profiles with laser capture
microdissection c
2) E13 epithelium and E13 mesenchyme 4.5 % 3.4 % 2.5 %
gene expression (E13GE) profiles 100 %
80 %
1) TSGE profiles 55.4 %
60 % 63.1 % 71.3 %
Data mining
(Perl code & Excel) 40 %
25.3 %
Cleft-enriched genes (Fig. 5B) 20 % 20.1 % 9.1 %
12.3 % 15.9 % 17.0 %
Filtering by 0%
E12.5 E13.5 Expressed
2) E13GE profiles CL CL in both
Mesenchymal (528) (1,576) (317)
Cleft-enriched genes (Fig. 5C)
Epithelium-derived
Data mining
Mesenchyme-derived
(DAVID, GO,
Expressed in both
and KEGG mapper)
No data
Functional prediction
Fig. 2.6 Overview of in silico data mining of SGMAP to and E13.5 central bud and between E13.5 cleft and E13.5
identify cleft-enriched genes in developing salivary basement membrane bud (≥2 FC). (c) The origin of cleft-
glands. (a) Overview of data collection and analysis of enriched genes was predicted by the comparison of E13
SGMAP.nidcr.nih.gov mRNA expression profiles. (b) mesenchyme and E13 epithelium mRNA expression pro-
Cleft-enriched genes were screened from temporal and files that were manually collected from SGMAP. The
spatial mRNA expression profiles from regions of devel- genes were classified into four origin groups including
oping salivary gland that were manually collected using epithelium derived (× ≤ −2 FC), mesenchyme derived (≥
laser capture microdissection. E12.5 cleft-enriched genes 2 FC), expressed in both (− 2< × <2 FC), and no data.
(528 genes) were identified from the comparison between Over half of cleft-enriched genes were predicted to be
E12.5 cleft and E12.5 main bud (≥ twofold change (FC)). mesenchymal genes, suggesting a potential role of mesen-
E13.5 cleft-enriched genes (1576 genes) were conserved chymal signaling in salivary gland early cleft formation
genes in the two sets of comparisons between E13.5 cleft and epithelial patterning
during branching morphogenesis, which merits subpopulations required for mesenchymal-epi-

future investigation. thelial signaling are poorly understood. Cleft
Cleft-enriched mRNAs may provide insight enrichment of gene products that are known to
into signaling that occurs in clefts. Mesenchymal stimulate cellular behaviors such as cell-cell
soluble factors such as fibroblast growth factors adhesion and chemotaxis is consistent with active
and platelet-derived growth factors are known to communication between epithelial and mesen-
provide critical contributions to salivary gland chymal cells and/or mesenchymal cell subtypes in
development [67, 94]. Recent work examined a developing clefts (Table 2.1). Other cleft-enriched
role for mesenchymal gene products in cell-ECM cytokines and chemokines may regulate unde-
interaction and ECM remodeling during epithe- fined aspects of mesenchymal function in pro-
lial branching [28, 71], although the specific cell gressing clefts, such as directed cell migration
34 M. Larsen et al.
# of genes
0 5 10 15 20 25
Cell-cell adhesion
(GO:0016337) 23
Chemotaxis
(GO:0006935) 15
Blood vessel development

(GO:0001568) 12
Sensory organ development

Biological process
9
(GO:0007423)
Hematopoietic or lymphoid
organ development (GO:0048534) 9
Oxygen transport
6
(GO:0015671)
Regulation of locomotion
(GO:0040012) 6
Regulation of neurotransmitter
secretion (GO:0046928) 4
Positive regulation of response

4
to external stimulus (GO:0032103)
Fig. 2.7 Gene ontology of biological processes for mesen- (conserved genes in both E12.5 and E13.5; p-value <0.01)
chymal cleft-enriched genes. Gene ontology (GO) of bio- was analyzed by Database for Annotation, Visualization
logical processes for mesenchymal cleft-enriched genes and Integrated Discovery (DAVID) system (ver. 6.7)
and condensation via growth factor-mediated ferentiation continues late into development and is
mechanical compaction [48, 49, 78, 90]. Taken not complete until after sexual maturity. The signal-
together, our analysis reveals promising novel ing pathways required to induce differentiation of
molecular candidates for regulators of cleft for- salivary acinar cells are poorly understood but are
mation and salivary gland development that can critical to understand for application in future regen-
be pursued in future studies to characterize their erative medicine approaches. Although information
potential roles in salivary gland branching mor- on gene expression is available on SGMAP for late
phogenesis and tissue maintenance/regeneration. developmental stages, how these genes are coordi-
nated to induce differentiation is not understood.
Network-based modeling approaches are one way
2.3.4 Systems-Level Analysis to gain insights into organization of signaling net-
of Salivary Gland works. Building on previous network-based model-
Differentiation Networks ing approaches performed by Jaskoll and Melnick
[55, 88] to investigate signaling networks in late
The initiation of differentiation occurs in develop- stage submandibular glands, a recent study used a
ing salivary glands trailing only slightly behind the combinatorial analysis of mRNA and miRNA
initiation of morphogenesis, but the process of dif- expression profiles to suggest that the regulatory
Table 2.1 Most highly expressed mesenchymal cleft-enriched genes grouped by KEGG pathway
KEGG pathway
mmu04510:Focal adhesion
mmu04512:ECM-receptor interaction
mmu04670:Leukocyte transendothelial migration
Common Both
mmu04062:Chemokine signaling pathway
mmu04810:Regulation of actin cytoskeleton
mmu04060:Cytokine-cytokine receptor interaction
mmu04010:MAPK signaling pathway
E12.5
mmu04610:Complement and coagulation cascades
mmu04360:Axon guidance
Stage- mmu04610:Complement and coagulation cascades
specific mmu04310:Wnt signaling pathway
E13.5 mmu04020:Calcium signaling pathway
mmu04142:Lysosome
mmu04650:Natural killer cell mediated cytotoxicity
mmu04080:Neuroactive ligand-receptor interaction
Cleft (CL)-enriched genes in this table were more highly expressed in mesenchyme than epithelium (≥ twofold change).
Signaling pathways were predicted by Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis of
Database for Annotation, Visualization and Integrated Discovery (DAVID). The common category contained KEGG
pathways that were conserved in both E12.5 and E13.5 (purple). There were also KEGG categories of genes that were
highly expressed only in E12.5 (blue) or E13.5 clefts (red)
Table 2.2 Putative endothelial-expressed cleft-enriched genes

E12.5 E13.5 Both
Blood vessel development FGF18, FLT1, NRP1, CSPG4, ENPEP, RECK, ALDH1A2,
(GO:0001568) PRRX1, KDR, PRRX2, TNFSF12, GJA4, MYO18B, EMCN,
CITED2, ZFP697 WNT2, EDNRA, HAND2, NTRK2, COL3A1,
HMOX1, SOX18, LOX, ROBO4, FGF10,
NR2F2, FGF2, NKX2–5, COL1A1, ARHGAP24,
BMP4, EFNB2, TBX1, THY1, CXCL12, CDH5
ANXA2, GTR1A, ID1, ITGA7
Gas transport F830116E18RIK HBA-A1, HBA-A2,
(GO:0015669) HBB-BH1, HBB-B1,
HBA-X, HBB-Y, CAR2
Endothelial genes shown here were predicted by gene ontology (GO) biological process analysis with DAVID (See Fig. 2.7)
Table 2.3 Putative neuronal-expressed cleft-enriched genes

E12.5 E13.5 Both
GO Regulation of PARK2 ADORA2A, NTRK2,
neurotransmitter secretion SNCA, CAMK2A
(GO:0046928)
KEGG mmu04360: axon guidance SEMA5B, PLXNC1, NRP1, RAC2,
EFNB3, EFNB2, PPP3CC, NFATC4,
ROBO2, LRRC4C, SEMA3A, CXCL12,
EPHB1, EPHA3, SLIT3, NFATC1
mmu04080: neuroactive CALCR, GABRA2, PTGER3,
ligand-receptor interaction GABRA1, GRIK1, ADORA2A,
GABRB2, GRIA3, NTSR1, EDNRA,
EDNRB, P2RY6, AGTR2, CHRM3,
AGTR1A, GRM7, ADRA2A
Neuronal genes shown here were predicted by gene ontology (GO) biological process and KEGG pathway analysis with
DAVID (See Fig. 2.7 and Table 2.1)
36 M. Larsen et al.
networks driving acinar cell terminal differentiation proteins and 914 parotid proteins, of which 252 are
in rat parotid salivary glands are not continuous but submandibular/sublingual specific and 249 are
involve temporally distinct developmental transi- parotid specific and are primarily extracellular or
tions [56]. In this study, mRNA profiling of acinar secretory in nature. The salivary proteome is avail-
cell RNA was performed using tissue collected by able as a tool for identification and diagnosis of
laser capture microdissection (LCM) at nine differ- both systemic and salivary gland diseases. Since
ent stages from E18 to P25. The data grouped into proteomic profiling has demonstrated its utility, it
four distinct clusters showing similarity of differen- is likely that future disease studies will employ
tially expressed genes within close developmental these methods as protein detection methods
time points and revealing that differentiation become increasingly sensitive. However, the pro-
occurred in four temporally distinct phases. cessing of samples for proteomic analysis, the spe-
Clustering analysis extended these observations to cifics of the instrumentation, and analysis of the
suggest distinct functional modes of the temporally data can significantly impact the results of the
distinct differentiation phases, and integrated experiment [8]. Analysis of saliva has recently
mRNA-miRNA expression analysis was used to extended beyond the proteome, and has been dem-
decipher regulatory networks controlled by mRNAs onstrated to provide many insights into normal and
and regulatory miRNAs. Interestingly, repression of diseased functions. With the emerging field of
a stemness-promoting pathway and stimulation of epigenomics, evaluation of saliva promises to be an
an Xbp1-stimulated pathway are required for important player, which is an exciting development
parotid acinar differentiation. Significantly, the data that has been reviewed elsewhere [91]. Saliva has
implicate miRNA control of multiple mRNAs as a also been demonstrated to contain exosomes [57],
major driver of parotid cell differentiation. Future small (30–100 nm diameter) cell-secreted vesicles
analyses that incorporate similar co-regulatory pre- that carry both proteins and miRNAs, short RNAs
dictions as well as predictions of likely signaling that work on groups of expressed mRNAs to down-
pathways will be important to identify important regulate gene expression of mRNAs by interfering
pathways for experimental validation and to iden- with their translation and targeting them for degra-
tify putative therapeutic targets. dation [91]. Since saliva can be obtained noninva-
sively and only small amounts are needed for most
analyses, saliva has significant potential as a diag-
2.4 Systems Approaches nostic fluid for detection and diagnosis of many
to the Study of Salivary diseases, perhaps most obviously for oral cancers
Gland Disease [18, 82, 92]. Evaluation of the saliva transcriptome
is also currently under investigation for assessment
2.4.1 Systems-Based Analysis of health and disease states in infants [50]. The
of Saliva future application of saliva-based diagnostics com-
ing out of systems-level analysis of saliva has sig-
Saliva has been investigated as a surrogate for nificant promise for many diseases and many types
blood as a means to identify differences in normal of patients [8].
and diseased patients. As saliva is a readily acces-
sible bodily fluid that can be collected in a nonin-
trusive manner, many “-omics” studies have 2.4.2 Systems-Based Methods
focused on differences in the levels of various bio- to Diagnose Salivary Gland
molecules in saliva. Proteomic approaches have Diseases: Sjögren’s Syndrome
been applied extensively to saliva, resulting in a
comprehensive catalog of the contents of human Diagnosis of SS is difficult due to a lack of spe-
saliva – the salivary proteome – prepared by a team cific molecular markers for the disease. Many
of researchers from five US universities [14]. This studies have used “-omics” methods in an attempt
study identified 917 submandibular/sublingual to discover reliable diagnostic markers for SS
from saliva samples. In a study comparing whole Similar to proteomics, metabolomics is now
saliva from normal and Sjögren’s syndrome possible as a result of recent improvements to
patients, proteins, peptides, and mRNAs that are mass spectrometry technology. Metabolomics is
differentially expressed in these populations were an emerging field that seeks to identify the full
identified [24]. Another study compared proteins spectrum of chemical metabolites produced dur-
taken from parotid salivary glands between SS ing cellular metabolism since many differences
patients and healthy patients through a technique in metabolism occur between normal and abnor-
known as multidimensional protein identification mal cells [69]. In a recent study, Kageyama et al.
technology, or MudPIT [1], which is an unbiased focused on the different levels of metabolites in
method for rapid and large-scale proteome analy- saliva from primary Sjögren’s syndrome (pSS)
sis by multidimensional liquid chromatography, patients vs normal patients [29]. With this
followed by tandem mass spectrometry, and data- approach, 88 metabolites were discovered by
base searching [89]. Out of the 1246 proteins comparing saliva taken from 12 SS patients with
identified by MudPIT, 529 were only detected in that obtained from 21 healthy patients of similar
either SS or healthy patients, 206 were signifi- ages. Out of these, 41 metabolites were expressed
cantly upregulated by more than twofold, and 34 at lower levels in pSS patients relative to healthy
were downregulated by more than twofold [54]. patients [29]. Further comparative analysis
Mass spectrometry assays conducted with the revealed a loss of metabolite diversity in SS
same samples identified 71 proteins, 58 of which patients, primarily resulting from decreased lev-
were proteins that had also been detected by els of the amino acids glycine and tyrosine, uric
MudPIT [1]. Further analysis of proteins with acid, and fucose. The discovery of these differing
differential expression between SS patients and metabolomic profiles between healthy patients
healthy controls identified 100 pathways of sig- and pSS patients could prove useful in establish-
nificance that were differentially regulated in SS ing more reliable diagnostic criteria for SS.
vs normal patients [1]. A recent study by Deutsche
et al. identified 79 peptides that were expressed in
SS patients at more than a threefold rate relative 2.4.3 S
jogren’s Syndrome Disease
to those in healthy patients. The samples, Progression
obtained from the saliva of 18 female SS patients
and 18 age-matched and gender-matched con- Systems biology-based approaches are beginning
trols, were subjected to protein depletion to to be utilized as a tool to better understand
remove high-abundance proteins that make iden- complex salivary gland diseases. Hu et al. used a
tification of low-abundance proteins difficult, systems approach to identify new disease-hub
prior to semiquantitative two-dimensional gel genes, which they define as promising targets for
electrophoresis and quantitative demethylation therapeutic intervention and diagnosis of SS [25].
liquid chromatography tandem mass spectrome- This study compared parotid tissue from three
try (LC-MS/MS), resulting in a threefold increase classes of patients: patients with primary SS,
in the ability to identify low-abundance proteins patients with primary SS associated with mucosa-
[15]. Bioinformatics analysis of the data identi- associated lymphoid tissue (MALT) lymphoma,
fied proteins with a >threefold increase in SS and patients without primary SS (non-primary SS
patients, including calcium-binding proteins, controls). Microarray profiling to examine gene
defense-response proteins, proteins involved in expression and proteomic analysis to examine
apoptotic regulation, stress-response proteins, protein expression were performed on all sam-
and cell motion-related proteins. The results ples, and weighted gene co-expression network
borne of these proteomic studies offer the poten- analysis was performed on these data to identify
tial to create better tools to diagnose SS as well as disease-hub genes. Computational analysis has
allowing for a better understanding of the pathol- also been employed in the study of animal mod-
ogy of SS at the molecular level. els of SS. In one study, the widely used non-obese
38 M. Larsen et al.
diabetic (NOD) mouse model of SS [47] was within 8–12 weeks after IR treatment. Using
compared with the parent line, Balb/C, in terms microarrays, which can quickly compare RNA
of biomarker expression, autoantibody produc- levels in multiple subjects simultaneously, and
tion, glandular inflammation, and saliva produc- real-time polymerase chain reaction (real-time
tion [12]. Principal component analysis was used PCR) to confirm the microarray data, 95 target
to identify significant positive and negative cor- genes were identified that exhibited significant
relations within the data. Interestingly, in this differences in expression as a result of IR treat-
study each biomarker typically associated exclu- ment. Out of these genes, 81 exhibited a decrease
sively with only one of the other parameters. in activity following treatment and a gradual
These data indicate that SS disease progression recovery to levels near to those prior to treatment.
may not follow a linear trajectory, even within The other 14 genes, most of which are cell cycle
this animal model. In humans, the disease etiol- control genes, exhibited an increase in expression
ogy is likely to be significantly more complex and activity throughout the 12-week period. A
and variable. proteomic assay performed on saliva samples
taken both before and after IR treatment showed
that most of the proteins encoded by the 81 genes
2.4.4 Systems Approaches with reduced activity after IR treatment are
to Understand Radiation involved in protein secretion and saliva produc-
Sensitivity of Salivary Glands tion. That the other 14 genes with increased
Following Radiation activity following IR treatment are mostly associ-
Treatment for Head and Neck ated with the cell cycle suggested that IR treat-
Cancers ment led to DNA damage and impairment of the
surviving salivary gland cells to regenerate.
A wide variety of tumors are classified as head Taken together transcriptomic and proteomic
and neck cancers; affected tissues include, but are data were used in this study to show that the loss
not limited to, the mouth, nose, tongue, pharynx, of saliva production following IR treatment is a
larynx, lymph nodes, salivary glands, and sinuses. result of the salivary glands’ loss of capacity to
The typical treatment for head and neck cancer is both produce and secrete proteins and to regener-
dependent on the patient’s medical history and ate following cell damage.
the stage of the cancer. However, owing to the
close proximity of these tumors to the salivary
glands, treatments often adversely affect a 2.4.5 Systems Approaches
patient’s ability to produce saliva, which hinders to Characterize Salivary Gland
the patient’s capacity to eat, swallow, and/or Tumors
speak. Understanding why the salivary glands are
sensitive to radiation and what happens to the Proteomics can also be used to characterize dif-
glands during radiation could lead to improved ferences in protein expression in glandular tissue
methods to prevent salivary gland damage. and has been used recently to categorize salivary
Systems approaches have been used to better tumors. Tumors of the salivary glands themselves
understand the response of salivary glands to make up roughly 8 % of head and neck cancers.
radiation treatment. A 2012 study by Stiubea- Like most tumors, diagnosis of these tumors
Cohen et al. to characterize the effects of irradia- involves fine-needle aspiration biopsy and scoring
tion on salivary glands, a common treatment for of the tumor type by a pathologist on the basis of
cancers of the head and neck, focused on the tran- the appearance of the tumor from a thin section of
scriptome and saliva proteome of submandibular the biopsy tissue that was stained with hematoxy-
salivary glands of rats [97]. In the rat irradiation lin and eosin (H&E) to highlight tumor structure.
model used in this study, saliva output fell to False negatives (i.e., incorrectly identifying a
roughly 50 % of its output prior to IR treatment malignant tumor as benign) lead to premature
deaths, and false positives (i.e., incorrectly identi- malignant salivary gland tumors. miRNA profil-
fying a benign tumor as malignant) lead to unnec- ing of metastatic SACC vs less metastatic SACC
essarily stringent therapeutic regimes. As a result, identified miR-320a as a metastatic repressor that
biomedical scientists are turning to proteomics in targets the integrin beta 3 mRNA (itgb3), sug-
their search for a more reliable diagnostic tool. In gesting a potential for miR-320a-based therapeu-
a 2013 paper, Donadio et al. sought to compare tics. In addition, miR-320a levels positively
the proteome of one of the most frequent benign correlated with prognosis for SACC patients,
salivary gland tumors: Warthin’s tumor (WT), a suggesting an additional application in disease
cystadenolymphoma, with that of another typi- diagnosis [86]. miRNA profiling of salivary
cally benign but more complex mixed tumor, gland adenoid cystic carcinoma (ACC) cell lines,
pleomorphic adenomas (PAs) [17]. PAs are diffi- a relatively rare malignant tumor with a poor
cult to diagnose as these complex tumors involve long-term survival rate, identified reduced levels
the glandular epithelium, the myoepithelium, and of miR-101-3p [41]. While miR-101-3p has
the stromal compartment, and they have the potential utility as a biomarker, its mechanism
potential to transition to a metastatic state. With was investigated in cell lines where it was found
35 patients that had undergone parotidectomy to inhibit cell growth and invasion by inhibiting
(removal of the parotid salivary glands) as their expression of the Pim-1 serine/threonine kinase,
subjects, the group ran mass spectrometry assays which can function as an oncogene. Interestingly,
and gel electrophoresis on gland samples. ACC cell lines stably expressing miR-101-3p
Through a computational comparison of the data, were found to have enhanced sensitivity to cispl-
a total of 34 proteins with different expression atin, the most common treatment for ACC [41],
patterns between WT and PA were identified, demonstrating the possible therapeutic potential
which was narrowed down to 26 different proof miR-101-3p-based therapeutics for ACC.
teins, of which nine were selected for Western
blot analysis. Between healthy tissue and PA/WT
samples, noticeably, visible increases in protein 2.5 Systems Analysis
expression in all nine proteins for either PA or WT and Regenerative Therapies
were observed. Many of the proteins whose
expression is altered in WT are implicated in 2.5.1 Systems Analysis
autoimmune disorders (e.g., rheumatoid arthritis), in Characterizing Cell
while those proteins with altered expression in PA Populations
are implicated in tumorigenesis events (e.g., pro-
liferation, invasion) [17]. These results demon- Interest in using salivary gland stem/progenitor
strate the utility of using a proteomic approach to cell populations for cell therapy has resulted in
study characteristics of salivary gland tumors, as several reports of salivary gland progenitor cell
the differing protein expression patterns between culture expansion and preclinical cell therapies
the two types of tumors could provide the basis [42, 46, 60, 93]. While these studies have very
for a more reliable tool for differential diagnosis promising translational medicine potential, the
and subclassification. Improvement of diagnosis small number of markers used to characterize cell
accuracy directly affects patient quality of life, as phenotypes combined with different markers
it can help physicians choose the best treatment used in each study provides an incomplete under-
regime for the patient. standing of the cell populations in question, and
Other -omics methods have been used to gain no definitive stem/progenitor cell markers have
additional insights into salivary gland cancer cell yielded long-term functional restoration of
function and to identify potential new biomarkers hyposalivation in an animal model as yet. This is
and therapeutic strategies. Salivary adenoid cys- an area that could greatly benefit from a systems
tic carcinomas (SACCs) are salivary ductal- analysis of functional, therapeutic progenitor cell
derived tumors, which account for 24 % of preparations. A recent deep sequencing study of
40 M. Larsen et al.
the transcriptome of SGC was corroborated and p rofiling studies. Since the enormity of systems-
complemented with quantitative real-time PCR, level datasets pose a challenge for researchers,
immunocytochemistry, and flow cytometry anal- continued development of computational tools to
ysis [72]. Postnatal hepatic stem cell markers, make the data accessible and integration possible
EpCAM, NCAM, c-kit, CD44, and CD133, and for both researchers and clinicians is critical to
other cell markers, CD29, CD49f, K7, K19, Sca- facilitate the current trend toward systems thinking
1, and c-Met, were found to be considerably well and incorporation of systems approaches into
conserved in both SGC and LPC under the isola- research. The use of pathway analysis methods
tion and culture methods performed in this study. and mathematical modeling, which is already at
Most of the markers above were similarly use in many diverse salivary and saliva-based
expressed in both LPCs and SGCs or more highly research projects [4, 43, 44, 56, 65, 68, 76, 83],
expressed in LPC than in SGC, while the expres- will become increasingly necessary to “make
sion levels of EpCAM, CD49f, K14, and K7 sense” of the enormity of data and to integrate
were significantly higher in SGC than heteroge- multiple types of “-omics” data to answer specific
neous LPC [72]. Clinical translation would ben- research questions. Application of systems
efit from further system analyses of functionally approaches in the future promises to deepen our
defined salivary gland stem/progenitor cell popu- understanding of the relationships between the
lations and benefit from similar comparisons to genome and the epigenome, the transcriptome and
known stem/progenitor cell populations used for proteome, environmental factors and disease,
functional restoration in other organs. and the relationship between physiological states
and disease. Integration of systems-level “-omics”
data with the physics of tissue biology is a holy
2.6 Conclusions and Outlook grail that promises to provide crucial insights to
normal biology and to disease development [16].
There is much to be gained from the application of Nevertheless, reductionist approaches will still be
systems biology-based approaches to understand required for testing hypotheses generated from
salivary gland developmental processes, to diag- systems biology approaches to validate predic-
nose disease, and to develop better therapeutic tions. In the future, systems-based analysis and
options for patients. The development of effective computational approaches will be useful in under-
therapeutic options is typically more successful standing disease etiology and inspiring new thera-
the more that is understood regarding the mecha- peutics. As we move into the age of personalized
nisms through which disease develops, which medicine, systems-based approaches will eventu-
requires an understanding of the normal develop- ally play an integral role in assessing the patient’s
ment and homeostasis mechanisms at play in nor- disease state and in defining patient-specific thera-
mal glands. Due to the complexity of the salivary peutic regimes.
gland and the complexity of salivary gland dis-
eases, systems-based approaches are integral to Acknowledgments The authors thank Kenneth Yamada
and Matthew Hoffman for reviewing the manuscript. This
making progress on all of these avenues.
work was sponsored in part by the University at Albany,
Systems-based profiling approaches, including SUNY, and by NIH grants R01DE022467 and C06
profiling the transcriptome, the miRNAome, the RR015464.
genome, the epigenome, the proteome, the metab-
olome, and the oral microbiome, are now in com- References
mon use in the study of both salivary gland
development and disease. To make progress 1. Ambatipudi KS, Swatkoski S, Moresco JJ, Tu
toward a comprehensive understanding of the nor- PG, Coca A, Anolik JH, Gucek M, Sanz I, Yates
3rd JR, Melvin JE. Quantitative proteomics of
mal development and homeostasis of the gland
parotid saliva in primary Sjogren’s syndrome.
and mechanisms through which homeostasis is Proteomics. 2012;12(19–20):3113–20. doi:10.1002/
disrupted will require moving beyond simple pmic.201200208.
2. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Niles R, Park SK, Prakobphol A, Ramachandran P,
Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig Richert M, Robinson S, Sondej M, Souda P, Sullivan
JT, Harris MA, Hill DP, Issel-Tarver L, Kasarskis A, MA, Takashima J, Than S, Wang J, Whitelegge JP,
Lewis S, Matese JC, Richardson JE, Ringwald M, Witkowska HE, Wolinsky L, Xie Y, Xu T, Yu W,
Rubin GM, Sherlock G. Gene ontology: tool for the Ytterberg J, Wong DT, Yates 3rd JR, Fisher SJ. The
unification of biology. The gene ontology consortium. proteomes of human parotid and submandibular/
Nat Genet. 2000;25(1):25–9. doi:10.1038/75556. sublingual gland salivas collected as the ductal secre-
3. Aure MH, Konieczny SF, Ovitt CE. Salivary tions. J Proteome Res. 2008;7(5):1994–2006.
gland homeostasis is maintained through acinar 15. Deutsch O, Krief G, Konttinen YT, Zaks B, Wong DT,
cell self-
duplication. Dev Cell. 2015;33(2):231–7. Aframian DJ, Palmon A. Identification of Sjogren’s syn-
doi:10.1016/j.devcel.2015.02.013. drome oral fluid biomarker candidates following high-
4. Bilgin CC, Ray S, Baydil B, Daley WP, Larsen M, Yener abundance protein depletion. Rheumatology (Oxford).
B. Multiscale feature analysis of salivary gland branch- 2015;54(5):884–90. doi:10.1093/rheumatology/keu405.
ing morphogenesis. PLoS One. 2012;7(3):e32906. 16. Discher D, Dong C, Fredberg JJ, Guilak F, Ingber
doi:10.1371/journal.pone.0032906. D, Janmey P, Kamm RD, Schmid-Schonbein GW,
5. Borghese E. The development in vitro of the sub- Weinbaum S. Biomechanics: cell research and
mandibular and sublingual glands of Mus musculus. applications for the next decade. Ann Biomed Eng.
J Anat. 1950a;84(3):287–302. 2009;37(5):847–59. doi:10.1007/s10439-009-9661-x.
6. Borghese E. Explanation experiments on the influ- 17. Donadio E, Giusti L, Seccia V, Ciregia F, da Valle
ence of the connective tissue capsule on the develop- Y, Dallan I, Ventroni T, Giannaccini G, Sellari-
ment of the epithelial part of the submandibular gland Franceschini S, Lucacchini A. New insight into
of Mus musculus. J Anat. 1950b;84(3):303–18. benign tumours of major salivary glands by proteomic
7. Boukheris H, Curtis RE, Land CE, Dores approach. PLoS One. 2013;8(8):e71874. doi:10.1371/
GM. Incidence of carcinoma of the major salivary journal.pone.0071874.
glands according to the WHO classification, 1992 18. Dumache R, Rogobete AF, Andreescu N, Puiu

to 2006: a population-based study in the United M. Genetic and epigenetic biomarkers of molecu-
States. Cancer Epidemiol Biomarkers Prev Publ Am lar alterations in oral carcinogenesis. Clin Lab.
Assoc Cancer Res cosponsored Am Soc Prev Oncol. 2015;61(10):1373–81.
2009;18(11):2899–2906. 19. Grobstein C. Inductive epitheliomesenchymal inter-
8. Castagnola M, Cabras T, Iavarone F, Fanali C, action in cultured organ rudiments of the mouse.
Nemolato S, Peluso G, Bosello SL, Faa G, Ferraccioli Science. 1953a;118(3054):52–5.
G, Messana I. The human salivary proteome: a critical 20. Grobstein C. Morphogenetic interaction between

overview of the results obtained by different proteomic embryonic mouse tissues separated by a membrane
platforms. Expert Rev Proteomics. 2012;9(1):33–46. filter. Nature. 1953b;172(4384):869–70.
doi:10.1586/epr.11.77. 21. Harunaga J, Hsu JC, Yamada KM. Dynamics of salivary
9. Cutler LS. The dependent and independent relation- gland morphogenesis. J Dent Res. 2011;90(9):1070–
ships between cytodifferentiation and morphogenesis 7. doi:10.1177/0022034511405330.
in developing salivary gland secretory cells. Anat 22. Hoffman MP, Kidder BL, Steinberg ZL, Lakhani S,
Rec. 1980;196(3):341–7. Ho S, Kleinman HK, Larsen M. Gene expression
10.
Daley WP, Gulfo KM, Sequeira SJ, Larsen profiles of mouse submandibular gland development:
M. Identification of a mechanochemical checkpoint FGFR1 regulates branching morphogenesis in vitro
and negative feedback loop regulating branching through BMP- and FGF-dependent mechanisms.
morphogenesis. Dev Biol. 2009;336(2):169–82. Development. 2002;129(24):5767–78.
doi:10.1016/j.ydbio.2009.09.037. 23. Hsu JC, Di Pasquale G, Harunaga JS, Onodera T,
11. Davies J. Branching morphogenesis. Molecular biol- Hoffman MP, Chiorini JA, Yamada KM. Viral gene
ogy intelligence unit, 1st ed. Boston: Springer; 2005. transfer to developing mouse salivary glands. J Dent Res.
12. Delaleu N, Immervoll H, Cornelius J, Jonsson
2012;91(2):197–202. doi:10.1177/0022034511429346.
R. Biomarker profiles in serum and saliva of experi- 24. Hu S, Wang J, Meijer J, Ieong S, Xie Y, Yu T, Zhou H,
mental Sjogren’s syndrome: associations with spe- Henry S, Vissink A, Pijpe J, Kallenberg C, Elashoff D,
cific autoimmune manifestations. Arthritis Res Ther. Loo JA, Wong DT. Salivary proteomic and genomic
2008;10(1):R22. doi:10.1186/ar2375. biomarkers for primary Sjogren’s syndrome. Arthritis
13. Delaleu N, Jonsson R, Koller MM. Sjogren’s syn- Rheum. 2007;56(11):3588–600. doi:10.1002/art.22954.
drome. Eur J Oral Sci. 2005;113(2):101–113. 25. Hu S, Zhou M, Jiang J, Wang J, Elashoff D, Gorr S,
14. Denny P, Hagen FK, Hardt M, Liao L, Yan W,
Michie SA, Spijkervet FK, Bootsma H, Kallenberg
Arellanno M, Bassilian S, Bedi GS, Boontheung P, CG, Vissink A, Horvath S, Wong DT. Systems biol-
Cociorva D, Delahunty CM, Denny T, Dunsmore J, ogy analysis of Sjogren’s syndrome and mucosa-
Faull KF, Gilligan J, Gonzalez-Begne M, Halgand associated lymphoid tissue lymphoma in parotid
F, Hall SC, Han X, Henson B, Hewel J, S H, Jeffrey glands. Arthritis Rheum. 2009;60(1):81–92.
S, Jiang J, Loo JA, Ogorzalek Loo RR, Malamud 26. Huang DW, Sherman BT, Tan Q, Kir J, Liu D,

D, Melvin JE, Miroshnychenko O, Navazesh M, Bryant D, Guo Y, Stephens R, Baseler MW, Lane
42 M. Larsen et al.
HC, Lempicki RA. DAVID bioinformatics resources: 41. Liu XY, Liu ZJ, He H, Zhang C, Wang YL. MicroRNA-
expanded annotation database and novel algorithms to 101- 3p suppresses cell proliferation, invasion and
better extract biology from large gene lists. Nucleic enhances chemotherapeutic sensitivity in salivary
Acids Res. 2007;35(Web Server issue):W169–75. gland adenoid cystic carcinoma by targeting Pim-1.
doi:10.1093/nar/gkm415. Am J Cancer Res. 2015;5(10):3015–29.
27. Jaskoll T, Melnick M. Embryonic salivary gland
42. Lombaert IM, Brunsting JF, Wierenga PK, Faber H,
branching morphogenesis. In: Davies JA, edi- Stokman MA, Kok T, Visser WH, Kampinga HH,
tor. Branching morphogenesis. New York: Landes de Haan G, Coppes RP. Rescue of salivary gland
Bioscience/Springer; 2005. p. 160–75. function after stem cell transplantation in irradiated
28. Joo EE, Yamada KM. MYPT1 regulates contractil- glands. PLoS One. 2008;3(4):e2063. doi:10.1371/
ity and microtubule acetylation to modulate integ- journal.pone.0002063.
rin adhesions and matrix assembly. Nat Commun. 43. Lubkin SR, Li Z. Force and deformation on branching
2014;5:3510. doi:10.1038/ncomms4510. rudiments: cleaving between hypotheses. Biomech
29. Kageyama G, Saegusa J, Irino Y, Tanaka S, Tsuda K, Model Mechanobiol. 2002;1(1):5–16.
Takahashi S, Sendo S, Morinobu A. Metabolomics 44. Lubkin SR, Murray JD. A mechanism for early

analysis of saliva from patients with primary Sjogren’s branching in lung morphogenesis. J Math Biol.
syndrome. Clin Exp Immunol. 2015;182(2):149–53. 1995;34(1):77–94.
doi:10.1111/cei.12683. 45. Magenheim J, Ilovich O, Lazarus A, Klochendler

30. Knosp WM, Knox SM, Hoffman MP. Salivary gland A, Ziv O, Werman R, Hija A, Cleaver O, Mishani
organogenesis. Wiley Interdiscip Rev Dev Biol. E, Keshet E, Dor Y. Blood vessels restrain pancreas
2012;1(1):69–82. doi:10.1002/wdev.4. branching, differentiation and growth. Development.
31. Knosp WM, Knox SM, Lombaert IM, Haddox CL, 2011;138(21):4743–52. doi:10.1242/dev.066548.
Patel VN, Hoffman MP. Submandibular parasym- 46. Maimets M, Rocchi C, Bron R, Pringle S, Kuipers
pathetic gangliogenesis requires sprouty-dependent J, Giepmans BN, Vries RG, Clevers H, de Haan G,
Wnt signals from epithelial progenitors. Dev Cell. van Os R, Coppes RP. Long-term in vitro expansion
2015;32(6):667–77. doi:10.1016/j.devcel.2015.01.023. of salivary gland stem cells driven by Wnt signals.
32. Knox SM, Lombaert IM, Reed X, Vitale-Cross L, Stem Cell Rep. 2016;6(1):150–62. doi:10.1016/j.
Gutkind JS, Hoffman MP. Parasympathetic innerva- stemcr.2015.11.009.
tion maintains epithelial progenitor cells during sali- 47. Makino S, Kunimoto K, Muraoka Y, Mizushima

vary organogenesis. Science. 2010;329(5999):1645–7. Y, Katagiri K, Tochino Y. Breeding of a non-
doi:10.1126/science.1192046. obese, diabetic strain of mice. Jikken Dobutsu.
33. Kovacs L, Szodoray P, Kiss E. Secondary tumours in 1980;29(1):1–13.
Sjogren’s syndrome. Autoimmunity reviews; 2009. 48. Mammoto T, Mammoto A, Jiang A, Jiang E, Hashmi
34. Kwon HR, Larsen M. The contribution of specific B, Ingber DE. Mesenchymal condensation-dependent
cell subpopulations to submandibular salivary gland accumulation of collagen VI stabilizes organ-specific
branching morphogenesis. Curr Opin Gen Dev (in cell fates during embryonic tooth formation. Dev Dyn
press). 2015. doi:10.1016/j.gde.2015.01.007. Off Publ Am Assoc Anatomists. 2015;244(6):713–
35. Lammert E, Cleaver O, Melton D. Role of endothelial 23. doi:10.1002/dvdy.24264.
cells in early pancreas and liver development. Mech 49. Mammoto T, Mammoto A, Torisawa YS, Tat T, Gibbs
Dev. 2003;120(1):59–64. A, Derda R, Mannix R, de Bruijn M, Yung CW, Huh
36. Larsen M, Hoffman MP, Sakai T, Neibaur JC,
D, Ingber DE. Mechanochemical control of mesen-
Mitchell JM, Yamada KM. Role of PI 3-kinase and chymal condensation and embryonic tooth organ for-
PIP3 in submandibular gland branching morphogen- mation. Dev Cell. 2011;21(4):758–69. doi:10.1016/j.
esis. Dev Biol. 2003;255(1):178–91. devcel.2011.07.006.
37. Larsen M, Wei C, Yamada KM. Cell and fibronectin 50. Maron JL. The neonatal salivary transcriptome. Cold
dynamics during branching morphogenesis. J Cell Spring Harb Perspect Med. 2015;6(3):a026369.
Sci. 2006;119(Pt 16):3376–84. doi:10.1101/cshperspect.a026369.
38. Larsen M, Yamada KM, Musselmann K. Systems
51. Maruyama EO, Aure MH, Xie X, Myal Y, Gan L, Ovitt
analysis of salivary gland development and disease. CE. Cell-specific cre strains for genetic manipulation
Wiley Interdiscip Rev Syst Biol Med. 2010;2(6):670– in salivary glands. PLoS One. 2016;11(1):e0146711.
82. doi:10.1002/wsbm.94. doi:10.1371/journal.pone.0146711.
39. Lazarus A, Del-Moral PM, Ilovich O, Mishani E, 52. Mathews SA, Kurien BT, Scofield RH. Oral mani-
Warburton D, Keshet E. A perfusion-independent role festations of Sjogren’s syndrome. J Dent Res.
of blood vessels in determining branching stereotypy 2008;87(4):308–18. doi:87/4/308 [pii].
of lung airways. Development. 2011;138(11):2359– 53. Matsumoto K, Yoshitomi H, Rossant J, Zaret KS. Liver
68. doi:10.1242/dev.060723. organogenesis promoted by endothelial cells prior to
40. Lee BH, Tudares MA, Nguyen CQ. Sjogren’s
vascular function. Science. 2001;294(5542):559–63.
syndrome: an old tale with a new twist. Arch doi:10.1126/science.1063889.
Immunol Ther Exp. 2009;57(1):57–66. doi:10.1007/ 54. Mauri P, Scarpa A, Nascimbeni AC, Benazzi L,

s00005-009-0002-4. Parmagnani E, Mafficini A, Della Peruta M, Bassi
C, Miyazaki K, Sorio C. Identification of proteins 66. Patel VN, Hoffman MP. Salivary gland develop-

released by pancreatic cancer cells by multidimen- ment: a template for regeneration. Semin Cell
sional protein identification technology: a strategy Dev Biol. 2014;25-26:52–60. doi:10.1016/j.
for identification of novel cancer markers. FASEB semcdb.2013.12.001.
J. 2005;19(9):1125–7. 67.
Patel VN, Rebustini IT, Hoffman MP.
55. Melnick M, Jaskoll T. Mouse submandibular gland Salivary gland branching morphogenesis.
morphogenesis: a paradigm for embryonic signal pro- Differ Res Biol Divers. 2006;74(7):349–64.
cessing. Crit Rev Oral Biol Med Off Publ Am Assoc doi:10.1111/j.1432-0436.2006.00088.x.
Oral Biologists. 2000;11(2):199–215. 68. Patterson K, Catalan MA, Melvin JE, Yule DI,

56.
Metzler MA, Venkatesh SG, Lakshmanan J, Crampin EJ, Sneyd J. A quantitative analysis of elec-
Carenbauer AL, Perez SM, Andres SA, Appana S, trolyte exchange in the salivary duct. Am J Physiol
Brock GN, Wittliff JL, Darling DS. A systems biol- Gastrointest Liver Physiol. 2012;303(10):G1153–63.
ogy approach identifies a regulatory network in doi:10.1152/ajpgi.00364.2011.
parotid acinar cell terminal differentiation. PLoS 69. Patti GJ, Yanes O, Siuzdak G. Innovation: metabo-
One. 2015;10(4):e0125153. doi:10.1371/journal. lomics: the apogee of the omics trilogy. Nature Rev.
pone.0125153. 2012;13(4):263–9. doi:10.1038/nrm3314.
57.
Michael A, Bajracharya SD, Yuen PS, Zhou 70. Peters SB, Naim N, Nelson DA, Mosier AP, Cady NC,
H, Star RA, Illei GG, Alevizos I. Exosomes Larsen M. Biocompatible tissue scaffold compliance
from human saliva as a source of microRNA promotes salivary gland morphogenesis and differ-
biomarkers. Oral Dis. 2010;16(1):34–8. entiation. Tissue Eng A. 2014;20(11–12):1632–42.
doi:10.1111/j.1601-0825.2009.01604.x. doi:10.1089/ten.TEA.2013.0515.
58. Miletich I. Introduction to salivary glands: structure, 71. Peters SB, Nelson DA, Kwon HR, Koslow M,

function and embryonic development. Front Oral DeSantis KA, Larsen M. TGFbeta signaling promotes
Biol. 2010;14:1–20. doi:10.1159/000313703. matrix assembly during mechanosensitive embryonic
59. Musselmann K, Green JA, Sone K, Hsu JC,
salivary gland restoration. Matrix Biol. 2015;43:109–
Bothwell IR, Johnson SA, Harunaga JS, Wei Z, 24. doi:10.1016/j.matbio.2015.01.020.
Yamada KM. Salivary gland gene expression 72. Petrakova OS, Terskikh VV, Chernioglo ES, Ashapkin
atlas identifies a new regulator of branching mor- VV, Bragin EY, Shtratnikova VY, Gvazava IG,
phogenesis. J Dent Res. 2011;90(9):1078–84. Sukhanov YV, Vasiliev AV. Comparative analysis
doi:10.1177/0022034511413131. reveals similarities between cultured submandibular sal-
60. Nanduri LS, Lombaert IM, van der Zwaag M, Faber ivary gland cells and liver progenitor cells. Springerplus.
H, Brunsting JF, van Os RP, Coppes RP. Salisphere 2014;3:183. doi:10.1186/2193-1801-3-183.
derived c-Kit+ cell transplantation restores tissue 73. Proctor GB, Carpenter GH. Regulation of salivary
homeostasis in irradiated salivary gland. Radiother gland function by autonomic nerves. Auton Neurosci
Oncol. 2013;108(3):458–63. doi:10.1016/j. Basic Clin. 2007;133(1):3–18. doi:10.1016/j.
radonc.2013.05.020. autneu.2006.10.006.
61. Napenas JJ, Brennan MT, Fox PC. Diagnosis and 74. Rafii S, Butler JM, Ding BS. Angiocrine func-

treatment of xerostomia (dry mouth). Odontol Soc tions of organ-specific endothelial cells. Nature.
Nippon Dent Univ. 2009;97(2):76–83. 2016;529(7586):316–25. doi:10.1038/nature17040.
62. Nedvetsky PI, Emmerson E, Finley JK, Ettinger
75. Ray S, Fanti JA, Macedo DP, Larsen M. LIM kinase
A, Cruz-Pacheco N, Prochazka J, Haddox CL, regulation of cytoskeletal dynamics is required for
Northrup E, Hodges C, Mostov KE, Hoffman MP, salivary gland branching morphogenesis. Mol Biol
Knox SM. Parasympathetic innervation regulates Cell. 2014;25(16):2393–407. doi:10.1091/mbc.
tubulogenesis in the developing salivary gland. E14-02-0705.
Dev Cell. 2014;30(4):449–62. doi:10.1016/j. 76. Ray S, Yuan D, Dhulekar N, Oztan B, Yener B, Larsen
devcel.2014.06.012. M. Cell-based multi-parametric model of cleft pro-
63. Nelson DA, Larsen M. Heterotypic control of base- gression during submandibular salivary gland branch-
ment membrane dynamics during branching morpho- ing morphogenesis. PLOS Computational Biology (in
genesis. Dev Biol. 2015;401(1):103–9. doi:10.1016/j. press). 2013.
ydbio.2014.12.011. 77. Sakai T, Larsen M, Yamada KM. Microanalysis of
64. Onodera T, Sakai T, Hsu JC, Matsumoto K,
gene expression in tissues using T7-SAGE: Serial
Chiorini JA, Yamada KM. Btbd7 regulates epi- Analysis of Gene Expression after high-fidelity
thelial cell dynamics and branching morphogen- T7-Based RNA amplification. In: Current protocols
esis. Science. 2010;329(5991):562–5. doi:10.1126/ in cell biology. New York: John Wiley & Sons ; 2002.
science.1191880. pp. 19.13.11–19.14.10.
65. Palk L, Sneyd J, Patterson K, Shuttleworth TJ, Yule 78. Sakai T, Larsen M, Yamada KM. Fibronectin

DI, Maclaren O, Crampin EJ. Modelling the effects requirement in branching morphogenesis. Nature.
of calcium waves and oscillations on saliva secre- 2003;423(6942):876–81.
tion. J Theor Biol. 2012;305:45–53. doi:10.1016/j. 79. Sakai T, Larsen M, Yamada KM. Morphogenesis and
jtbi.2012.04.009. branching of salivary glands: characterization of new
44 M. Larsen et al.
matrix and signaling regulators. Oral Biosci Med. and mesenchyme during salivary gland initiation:
2005;2(2/3):105–13. the role of Fgf10. Biol Open. 2013;2(10):981–9.
80. Sequeira SJ, Gervais EM, Ray S, Larsen M. Genetic doi:10.1242/bio.20135306.
modification and recombination of salivary gland 91. Wren ME, Shirtcliff EA, Drury SS. Not all biofluids
organ cultures. J Vis Exp JoVE. 2013;71:e50060. are created equal: chewing over salivary diagnostics
doi:10.3791/50060. and the epigenome. Clin Ther. 2015;37(3):529–39.
81. Singh P, Schimenti JC, Bolcun-Filas E. A mouse geneti- doi:10.1016/j.clinthera.2015.02.022.
cist’s practical guide to CRISPR applications. Genetics. 92. Wu CC, Chu HW, Hsu CW, Chang KP, Liu HP. Saliva
2015;199(1):1–15. doi:10.1534/genetics.114.169771. proteome profiling reveals potential salivary bio-
82. Sivadasan P, Kumar Gupta M, Sathe GJ, Balakrishnan markers for detection of oral cavity squamous cell
L, Palit P, Gowda H, Suresh A, Abraham Kuriakose M, carcinoma. Proteomics. 2015;15(19):3394–404.
Sirdeshmukh R. Data from human salivary proteome – doi:10.1002/pmic.201500157.
A resource of potential biomarkers for oral cancer. Data 93. Xiao N, Lin Y, Cao H, Sirjani D, Giaccia AJ, Koong
Brief. 2015;4:374–8. doi:10.1016/j.dib.2015.06.014. AC, Kong CS, Diehn M, Le QT. Neurotrophic fac-
83. Sneyd J, Crampin E, Yule D. Multiscale modelling tor GDNF promotes survival of salivary stem cells.
of saliva secretion. Math Biosci. 2014;257:69–79. J Clin Invest. 2014;124(8):3364–77. doi:10.1172/
doi:10.1016/j.mbs.2014.06.017. JCI74096.
84. Speight PM, Barrett AW. Salivary gland tumours.
94. Yamamoto S, Fukumoto E, Yoshizaki K, Iwamoto T,
Oral Dis. 2002;8(5):229–40. Yamada A, Tanaka K, Suzuki H, Aizawa S, Arakaki
85. Speight PM, Barrett AW. Prognostic factors in
M, Yuasa K, Oka K, Chai Y, Nonaka K, Fukumoto
malignant tumours of the salivary glands. Br J Oral S. Platelet-derived growth factor receptor regulates
Maxillofac Surg. 2009;47(8):587–93. salivary gland morphogenesis via fibroblast growth
86. Sun L, Liu B, Lin Z, Yao Y, Chen Y, Li Y, Chen J, factor expression. J Biol Chem. 2008;283(34):23139–
Yu D, Tang Z, Wang B, Zeng S, Fan S, Wang Y, Li 49. doi:10.1074/jbc.M710308200.
Y, Song E, Li J. MiR-320a acts as a prognostic factor 95. Yang H, Wang H, Shivalila CS, Cheng AW, Shi L,
and Inhibits metastasis of salivary adenoid cystic car- Jaenisch R. One-step generation of mice carry-
cinoma by targeting ITGB3. Mol Cancer. 2015;14:96. ing reporter and conditional alleles by CRISPR/
doi:10.1186/s12943-015-0344-y. Cas-mediated genome engineering. Cell.
87. Tucker AS. Salivary gland development. Semin
2013;154(6):1370–9. doi:10.1016/j.cell.2013.08.022.
Cell Dev Biol. 2007;18(2):237–44. doi:10.1016/j. 96. Zhang A. Advanced analysis of gene expression

semcdb.2007.01.006. microarray data. Singapore: World Scientific; 2006.
88. Waddington CH. The strategy of the genes. London: 97. Stiubea-Cohen R, David R, Neumann Y, Krief G,
George Allen & Unwin; 1957. p. 11–58. Deutsch O, Zacks B, Aframian DJ, Palmon A. Effect
89. Washburn MP, Wolters D, Yates 3rd JR. Large-scale of irradiation on cell transcriptome and proteome
analysis of the yeast proteome by multidimensional of rat submandibular salivary glands. PloS One.
protein identification technology. Nat Biotechnol. 2012;7:e40636
2001;19(3):242–7. doi:10.1038/85686.
90. Wells KL, Gaete M, Matalova E, Deutsch D, Rice D,
Tucker AS. Dynamic relationship of the epithelium
Mucins in Salivary Gland
Development, Regeneration, 3
and Disease
Isabel Castro, María-José Barrera,
Sergio González, Sergio Aguilera, Ulises Urzúa,
Juan Cortés, and María-Julieta González
Abstract
Mucins are large glycoproteins that can be grouped as membrane-bound or
secreted. Membrane-bound mucins are essential contributors of the glyco-
calyx of mucosal surfaces where they play important biological roles in cell
interactions and signaling. Secreted mucins are the main structural compo-
nents of the mucus gel that covers the epithelium and contribute to the
protection of the mucosa surface against allergens, debris, pathogens, dry-
ing, injury, and abrasive stress. MUC1 and MUC4 are plasma membrane-
anchored mucins expressed in oral epithelium and salivary glands and are
ubiquitously located in normal epithelia. MUC5B is the major secreted
polymeric mucin present in human saliva and contains negatively charged
glycans allowing the formation of a hydrophilic gel that hydrates and pro-
tects the oral epithelium. MUC7 is a secreted mucin present in saliva and
has low viscoelasticity and high bacteria-agglutinating properties allowing
clearance of microorganisms from the mouth by swallowing.
The roles of mucins during development of salivary glands remain
unknown. Few studies on mucin expression in human salivary glands'
development and in murine models have been described to date. In sali-
vary gland regeneration, mucins have been evaluated only as markers of
functionality of the glandular acini after damage and/or regeneration ther-
apy. Interestingly, changes in quality and quantity of salivary mucins have
been observed in pathological conditions. Over-expression of specific
mucins induced by pro-inflammatory cytokines and ectopic secretion of
I. Castro, PhD • M.-J. Barrera, PhD • U. Urzúa, PhD

J. Cortés, PhD • M.-J. González, MSc (*)
Institute of Biomedical Sciences, Faculty of
e-mail: [email protected]
S. González, MSc
Mayor University, Santiago, Chile
S. Aguilera, MD
INDISA Clinic, Santiago, Chile

46 I. Castro et al.
mucins to extracellular matrix (ECM) support evidence on a self-

perpetuating mucin-cytokine signaling loop that may facilitate the
maintenance of an inflammatory environment in chronic inflammatory
diseases, such as Sjögren’s syndrome (SS). However, the loss of a mucin
member could predispose to infections and inflammatory diseases in a
mucosa. A reduced sulfation of MUC5B has been observed in salivary
glands of SS patients, aggravating their oral dryness. Tumoral cells of sali-
vary glands express aberrant forms or high amounts of mucins, being pro-
posed as markers of malignant transformation. In this chapter, we
summarize the main structural and functional characteristics of salivary
mucins, their expression during salivary gland development, and regenera-
tion and alterations in mucin quantity and quality in pathological processes
affecting the salivary gland.
3.1 Structure, Biosynthesis mucins. The second class consists of membrane-

and Functions of Salivary tethered mucins and includes MUC1, MUC3A,
Mucins MUC3B, MUC4, MUC12, MUC13, MUC15,
MUC16, MUC17, MUC20, MUC21 and MUC22
Mucins are O-glycoproteins composing the mucus [14–17]. Table 3.1 summarizes the tissue distri-
layer that protects mucosal surfaces from external bution of human mucins.
insult and desiccation. Human salivary mucins are The biosynthesis of polymeric mucins starts at
synthesized by the submandibular, sublingual and the endoplasmic reticulum (ER), where the apo-
minor salivary glands. Although their contribution mucin or polypeptide backbone is translated. The
to the total volume of saliva is low, the minor sali- N-glycosylation and the formation of intramo-
vary glands contribute up to 70 % of total mucin lecular disulfide bonds within the cysteine-rich
found in saliva [1]. Mucins are high-molecular- N- and C-termini and the internal cys domains
weight O-glycoproteins containing at least one occur co-translationally. Dimerization takes place
region of repeated sequences that in some cases at the ER by intermolecular disulfide linkage
include variable numbers of tandem repeats between C-terminal domains of mucin monomers
(VNTR) polymorphisms [2, 3] (Fig. 3.1). These [18, 19]. The O-glycosylation is carried out in the
regions are rich in serine and threonine, residues Golgi apparatus, where UDP-GalNAc:polypeptide
that covalently bond with a variety of O-glycans. N-acetylgalactosaminyl transferases transfer
Importantly, oligosaccharides contribute up to 80 N-acetylgalactosamine to a serine or threonine
% of the mucin mass [9], and many of these gly- residue of the mucin protein backbone [20]. The
cans are sialylated and/or sulfated, conferring O-glycosylated products are further elongated by
hydrophilic and polyanionic properties to mucins sequential action of glycosyltransferases to pro-
[10–12]. The N- and C-terminal regions of mucins duce a wide array of oligosaccharides [21].
are rich in cysteine and are non-glycosylated or Intermolecular disulfide linkage between
sparsely N- and O-glycosylated [13] (Fig. 3.1). N-terminal domains mediates polymerization of
A total of 20 mucins in humans, numbered in mucin dimers in the trans-Golgi network or in
the order of their discovery, have been described secretory granules. High concentrations of Ca+2
to date [14]; they can be structurally categorized and H+ inside the secretory granules contribute to
in two major classes. High-molecular-mass mucin aggregation, reducing repulsive forces
secreted mucins that form polymeric structures among oligosaccharide negative charges [22].
and include MUC2, MUC5AC, MUC5B, MUC6, Polymers are packaged and stored in dehydrated
and MUC19. MUC7 and MUC8 are also secreted form within secretory granules until regulated
but consist of low-molecular-mass non-polymeric exocytosis [23].
3 Mucins in Salivary Gland Development, Regeneration, and Disease 47
a b
Fig. 3.1 Scheme of the structure of membrane-tethered CT33 [5], Ma695 [6], VU4H5 [7], MUC1/DF [8], MUC1/
mucins (a) and polymeric secreted mucins (b). (c) The SEC [7] and BOS6E6 [7]
epitope mapping of several MUC1 antibodies: CT1 [4],
The major polymeric mucin found in human mobility analysis of MUC5B revealed two prod-
saliva is MUC5B (>1 MDa), formerly named ucts of high- and low-charge glycoforms [29,
MG1 [24, 25]. MUC5B comprises 15 % protein, 30]. Studies of salivary mucin preparations have
78 % carbohydrate, and 7 % sulfate by weight shown that the highly charged MUC5B glyco-
[26]. MUC5B-linked carbohydrates constitute a form reacts with the F2 monoclonal antibody,
highly heterogeneous set of neutral, sulfated, and which specifically recognizes the sulfated
sialylated oligosaccharides [27, 28]. Anion carbohydrate epitope SO3Galβ1-3GlcNAc- of

exchange chromatography and electrophoretic sulfo-Lewisa [31]. Negatively charged glycans of
48 I. Castro et al.
Table 3.1 Classification and tissue expression of mucins

Tandem repeat
Mucin Class Chromosome length Tissue distribution
MUC1 Membrane anchored 1q21 20 Most epithelia
with secreted variants
MUC2 Large polymeric 11p15.5 23 Endometrium, jejunum, ileon, colon
mucin
MUC3A Membrane anchored 7q22 17 Small intestine, colon, gallbladder
MUC3B Membrane anchored 7q22 17 Small intestine, colon, gallbladder
MUC4 Membrane anchored 3q29 16 Most epithelia
MUC5AC Large polymeric 11p15.5 8 Conjunctiva, respiratory tract, stomach,
mucin endocervix, endometrium
MUC5B Large polymeric 11p15.5 29 Submandibular, sublingual and minor
mucin salivary glands, respiratory tract,
endocervix
MUC6 Large polymeric 11p15.5 169 Stomach, ileum, gallbladder,
mucin endocervix, endometrium
MUC7 Non-polymeric 4q13-q21 23 Submandibular, sublingual and minor
secreted mucin salivary glands
MUC8 Non-polymeric 12q24.3 13/41 Respiratory tract, uterus, endocervix,
secreted mucin endometrium
MUC9 Non-polymeric 1p13 15 Fallopian tubes
secreted mucin
MUC12 Membrane anchored 7q22 28 Colon, pancreas, prostate, uterus
MUC13 Membrane anchored 3q21.2 27 Colon, trachea, kidney, small intestine
MUC15 Membrane anchored 11p14.3 None Colon, respiratory tract, small intestine,
prostate
MUC16 Membrane anchored 19p13.2 156 Ovary, cornea, conjunctiva, respiratory
tract, endometrium
MUC17 Membrane anchored 7q22 59 Stomach, duodenum, colon
MUC19 Large polymeric 12q12 19 Submandibular, sublingual and minor
mucin salivary glands
MUC20 Membrane anchored 3q29 18 Placenta, colon, respiratory tract,
prostate, liver
MUC21 Membrane anchored 6p21 15 Respiratory tract, thymus, colon
MUC22 Membrane anchored 6p21.33 10 Respiratory tract
MUC5B confer hygroscopic properties, allowing tissues against the demineralization caused by
formation of a hydrophilic gel that hydrates and acids produced by microorganisms [34, 35].
protects the oral epithelium [10, 12]. The visco- MUC5B glycans serve as adhesion receptors
elastic properties of salivary mucins provide local allowing binding of microorganisms including
protection against mechanical forces during eat- Haemophilus parainfluenzae [36] and
ing and speaking [32]. Lubricant properties, Helicobacter pylori [37].
adhesiveness and elasticity of salivary MUC5B Another mucin found in saliva is MUC7, for-
reduce the friction of tooth surfaces during chew- merly named MG2 [38–40]. MUC7 has a molec-
ing and contribute to the formation and swallow- ular weight of ~200 kDa and contains about 30 %
ing of a food bolus [33]. Adsorption of MUC5B protein, 68 % carbohydrate, and 1.6 % sulfate
to tooth surfaces, given its high affinity for [40–42]. The oligosaccharides linked to the
hydroxyapatite, contributes to the formation of MUC7 peptide backbone are mainly fucosylated
the acquired enamel pellicle that protects hard and sialylated trisaccharides [43]. MUC7 has low
viscoelastic and high bacteria-agglutinating GSTAPPAHGVTSAPDTRPAP sequence [54]

properties [44]. MUC7 binds to microorganisms (Fig. 3.1). This domain can be released from the
such as Pseudomona aeruginosa [45], cell surface by proteolytic shedding [58]. The
Agregatibacter actinomycetemcomitans [46], C-terminal cytoplasmic domain of cell-anchored
Escherichia coli [47], and Streptococcus mutans mucins has been implicated in cell signaling in
[48], among others. As a result, bacterial aggre- physiological and pathological conditions [59].
gates are formed and are easily cleared out of the
mouth by swallowing. In addition to MUC7 gly-
cans, for example, sialic acid [45, 46], peptide 3.2 xpression of Mucins
E
domains in MUC7 are involved in the interaction During Salivary Gland
with bacteria, and some regions show broad- Development
spectrum antimicrobial activity[49–51]. The
non-glycosylated regions of mucins have many The specific roles played by mucins in human
structural motifs and domains that allow interac- salivary gland development have not been fully
tion with various salivary mucin-interacting pro- studied, mostly due to ethical and technical restric-
teins (e.g., histatin, statherin, acidic proline-rich tions. The submandibular gland starts to develop
proteins, α-amylase and lactoferrin, among oth- around the 6th week of human gestation, followed
ers) [52]. Such interactions might help to enhance by the parotid gland around the 7th week, and the
protein stability and function in the maintenance sublingual gland around the 8th [60].
of oral physiology [52, 53]. A single study on mucins expression in
Membrane-anchored mucins are initially human salivary gland development was carried
translated as a single polypeptide chain that is out by Lourenço et al. (2011). It consisted of an
cleaved to generate two sub-units, one containing immunohistochemical analysis of minor salivary
the extracellular domain and the other containing glands collected from 20 post mortem human
the transmembrane (TM) and C-terminal cyto- fetuses ranging from 4 to 24 weeks of gestation
plasmic domains of the molecule [54, 55]. These after natural miscarriage. Using the monoclonal
fragments produced in the ER are non-covalently Ma695 antibody directed against a glycosylated
associated and remain linked until insertion into epitope of the human MUC1 VNTR region (Fig.
the plasma membrane [54, 55] (Fig. 3.1). 3.1), these authors observed MUC1 in clusters
Cell-tethered mucins expressed by the oral of epithelial cells at the bud stage, showing that
epithelium and salivary glands include MUC1 MUC1 expression begins early in salivary gland
and MUC4, which are almost ubiquitously found development. At the pseudoglandular stage,
in normal epithelia [13]. These plasma MUC1 was observed in the rudimentary luminal
membrane–anchored mucins have a rigid confor- space and appeared in the luminal space of all
mation, which extends up to 1.5 μm from the cell glandular ductal systems in the canalicular stage.
surface providing a protective barrier against During the terminal bud stage, MUC1 retained
microorganisms and other cytotoxic agents [56]. high expression along the luminal pole of epi-
It has been proposed that MUC1 and MUC4 thelial cells throughout the entire ductal system.
serve as a scaffold that contributes to retention of In certain areas, MUC1 was also observed in the
secretory mucins and other salivary proteins in basal pole of larger excretory ductal cells. In fully
the oral cavity [13]. Many alternative splice vari- developed salivary glands, MUC1 was observed
ants encode several MUC1 isoforms [55]. Some in the cytoplasm of acinar cells and in the luminal
MUC1 isoforms have TM and C-terminal cyto- space border of the ducts [6]. Using the same anti-
solic domains, while others lack the TM domain body, MUC1 was detected in striated ducts and in
and are soluble, e.g., MUC1/SEC [57] (Fig. 3.1). basal cells of excretory ducts in parotid and sub-
The extracellular domain of human MUC1 com- mandibular glands of adult individuals. In acinar
prises the N-terminal signal sequence and the cells, MUC1 was observed in the apical plasma
VNTR domain of 20–100 repeats of the membrane of serous cells, while mucous cells
50 I. Castro et al.
were negative [61]. A similar MUC1 expression opment, MUC3, MUC4, MUC5B, and MUC16
pattern was recently observed in human labial were observed, while MUC2, MUC5AC and
salivary glands from adult subjects [7]. Using the MUC6 were not detected [6]. At the initial bud
VU4H5 antibody directed against the VNTR stage, MUC5B and MUC16 were expressed
region of MUC1 (Fig. 3.1), a staining in the apical weakly in some cells. At the pseudoglandular
pole of serous acini and duct cells was observed. stage, MUC3 was observed in small cytoplasmic
Using the BOS6E6 antibody that recognizes the deposits in epithelial cells and MUC4 was
MUC1/Y isoform, staining was observed mainly strongly detected in the luminal region of devel-
in the apical region of serous acinar cells and oping ductal structures. MUC4 was also detected
throughout the cytoplasm of ductal cells. Using an in the canalicular stage, in the luminal space of
antibody directed against the C-terminal region of well-developed ducts and in the wall of blood
the MUC1/SEC isoform, a staining localized vessels. At the terminal bud stage, MUC5B and
mainly at the apical region of serous acinar cells MUC16 were detected in mucous cells [6]. In
was observed. Mucous cells were negative for adult human salivary glands, MUC5B and
MUC1 with several antibodies used [7]. MUC16 were detected in the luminal border of
Analysis of mucin transcripts in homogenates excretory ducts, and MUC16 was also observed
of adult human salivary glands have shown in acinar cells [6]. However, several studies using
expression of MUC1 (MUC1A, MUC1B, antibodies directed against different epitopes of
MUC1/Y, MUC1/SEC), MUC2, MUC3, MUC4, MUC5B showed expression in mucous acinar
MUC5B, MUC5AC, MUC6, MUC7 and MUC19 cells of fully developed submandibular, sublin-
[7, 13, 29, 62, 63]. In human salivary gland devel- gual, and labial salivary glands [28, 64–66]
a c e
b d f
Fig. 3.2 Sections of labial salivary glands from control action for sulfo-Lewisa in SS patients was observed (d).
individuals (a, c, e) and SS patients (b, d, f). (a, b) (e, f) Double staining with Alcian blue (a, b) and immuno-
Immunohistochemistry of MUC5B. Gland section show- histochemistry for laminin. Positive (a, b) staining was
ing MUC5B in the basal region of mucous acinar cells (a). stronger in control individuals (e) than in SS patients (f).
In SS patients, MUC5B was observed both in basal and Decreased (a, b) staining correlated with decreased lam-
apical cytoplasms of mucous acinar cells (b). (c, d) Sulfo- inin immunoreactivity in SS patients (f). Bars 50 μm, m
Lewisa antigen immuno-detection. Mucous acini showed mucous acini, s serousacini; f: inflammatory focus.
sulfo-Lewisa immunoreactivity in the whole cytoplasm Images reproduced with permission of Castro et al. 137].
(c). Low abundance of mucous acini showing immunore-
(Fig. 3.2 a, c and e). Mucous acini expressing that glucocorticoids are required for mucin
MUC5B showed a mosaic pattern of sulfo-Lewisa expression, and treatment in utero with exoge-
stain, indicating that one and the same salivary nous glucocorticoids induces a significant
gland synthesizes different MUC5B glycoforms increase of embryonic mucin mRNA and protein
[28, 65]. When and how these phenotypes origi- [70]. In addition, they also find that IL-6, but not
nate during gland differentiation is yet unknown. TNF-α signaling, modulates embryonic mucin
Are the mucous acini formed by different types of expression in vitro [71].
acinar cells? Or is there asynchrony in the secre- Denny et al. (1982) prepared a rabbit anti-
tory response, among others? The simultaneous mucin serum using purified mouse-submandibular
detection of MUC5B and MUC7 demonstrated sialomucin from 80 to 90-day–old female
that the staining patterns were non-overlapping mice[72]. In newborn animals, the sialomucin
[64]. While MUC5B is observed in mucous cells, was detected in secretory terminal-tubule cells
MUC7 is expressed in serous acinar cells of fully and in proacinar cells, neither of which is morpho-
developed human submandibular, sublingual, and logically identical to the mature acinar cell [73].
labial salivary glands [28, 64, 66, 67]. The sialomucin was also detected in acinar cells
Most of the studies on mucin expression dur- and in granular intercalating-duct cells of the sub-
ing development have been performed on murine mandibular gland of adult mice [73]. Moreover, a
models, with the submandibular gland the most strong Alcian blue (AB) staining was observed in
studied. In mouse submandibular gland develop- proacinar cell granules. Radioimmunoassays
ment, the initial bud (E12.5) starts out as a solid were carried out for mucin quantitation in the
chord that then develops into a network of solid homogenates of submandibular glands. Mucin
stalks and end buds. Branching morphogenesis concentration was lowest in the newborn group
occurs in the pseudoglandular stage (E13.5). and highest in 20-day-old mice. This increase in
During the canalicular (E15.5) and terminal bud mucin concentration was associated with changes
(E18.5) stages, branches and buds hollow out in in acinar cell size and may reflect their maturation
their center by apoptosis of cells that do not make stage. The mucin detected in all age groups was
contact with the basal lamina [68]. Mucin is the apparently antigenically identical to the mucin
initial marker of epithelial differentiation in purified from adult mice. Together these results
mouse submandibular gland. Jaskoll et al. (1998) showed that mucin expression begins prior to the
demonstrated differences between mRNA and final acinar cell differentiation and that sialomu-
protein expression of mucin in the embryonic, cin is present in the submandibular gland from
neonatal, and adult mouse submandibular gland. birth to adulthood [72, 74, 75].
By northern blot assays, E17 and 1-day-old neo- The expression of Muc1 was analyzed during
nates exhibited two mucin transcripts (1.2 and mouse post-implantation development. Muc1
0.85 kb) which are different in size than the sin- protein localization was determined by immuno-
gle (1.01 kb) adult transcript. Two embryonic histochemistry using CT1, a polyclonal antise-
mucin isoforms of ~110 and 152 kDa were rum directed against the 17 C-terminal amino
detected in comparison to ~136 kDa adult mucin. acids of the cytoplasmic tail of human trans-
The ~152 kDa embryonic isoform persisted in membrane Muc1 isoforms (Fig. 3.1). This epit-
neonatal glands. In situ hybridization showed ope is highly conserved in a variety of tissues
mucin transcripts localized in the branching epi- and species [4]. In salivary glands, Muc1 immu-
thelia by E14. The hybridization signal increased nodetection with CT1 showed signal in ducts of
with age in terminal bud and proacinar cells. 15-day-old embryos and, subsequently, Muc1
Immunofluorescence assays showed mucin pro- was detected in terminal tubules. Muc1 protein
tein from E17 in plasma membranes of terminal expression increased with time during epithelial
bud and proacinar cells [69]. Mucin expression in branching and highest levels were observed in
submandibular development is modulated by dif- lumen, days 15–18 [76]. From these data, a role
ferent factors. Melnick and Jaskoll demonstrated for MUC1 in glandular morphogenesis was
52 I. Castro et al.
s uggested due to the coincident onset of glandu- 3.3.1 R

egeneration of Salivary
lar differentiation with changes of MUC1 Glands Damaged by Radiation
expression pattern during embryonic develop-
ment [76]. Hudson et al. [77]evaluated the poten- Radiation therapy for head and neck cancer
tial role of MUC1 on morphogenesis using cell results in significant side effects in normal sali-
lines cultured on type I collagen gels. MUC1 vary glands, provoking a decreased quality of life
altered the three-dimensional growth pattern of for these patients. The salivary gland is extremely
cells and induced tubular and branching mor- sensitive to radiation, mainly acinar cells [101]
phogenesis. MUC1-expressing cells showed and manifests acute and chronic responses to
altered morphogenesis characterized by reduced radiotherapy. During the first week of treatment,
cellular adhesion and enhanced migration [77]. patients may lose up to 50–60 % of salivary flow
These results suggest that MUC1 may induce [81–83] due to high levels of apoptosis and glan-
changes in tissue architecture in development dular shrinkage [85, 104, 105], which affects
and also in cancer, where it is frequently over- saliva flow and composition [81, 83]. Chronic
expressed, misglycosylated and redistributed responses persist months or years after radiother-
[78–80]. apy, where a great number of patients continue to
show a significant decrease in both stimulated
and non-stimulated salivary flow [82, 83, 106].
3.3 ucin and the Regeneration
M Saliva of the submandibular gland from irradi-
of Salivary Glands ated patients was evaluated in order to determine
a relationship between oral dryness and MUC5B
Antecedents on mucin in salivary gland regenera- concentration [91]. Patients with severe xerosto-
tion processes are extremely scarce. Mucins have mia 12 months after radiation therapy showed a
been only evaluated as markers to address the tendency to decreased levels of MUC5B in saliva,
functionality of the glandular acini after damage compared to patients with mild xerostomia.
and/or regeneration therapy. In these studies, Interestingly, half of the patients (8 individuals)
mucins have been indirectly detected using stain- with severe xerostomia did not have detectable
ing with AB (unspecified pH) or Periodic Acid MUC5B levels by 12 months post-radiotherapy;
Schiffs (PAS). Table 3.2 summarizes the main however, the significance of these data should be
changes observed in murine salivary glands after confirmed with a greater number of cases [91].
damage induced by ligation and radiation. Both groups of patients, with severe and mild
Table 3.2 Summary of the effects of radiotherapy and duct ligation on the salivary gland
Radiotherapy References Ligation References
Glandular Decrease salivary flow [81–83] Decrease salivary flow, [84]
function Secretory dysfunction
Glandular shrinkage [85] Inflammatory cell [84, 86–90]
infiltration
Tendency to decreased levels of [91] Loss of glandular weight [88, 92]
MUC5B in submandibular gland saliva
Ductal cells Proliferation of ductal [90, 93, 94]
cells
Acinar cells Apoptosis and replacement by fibrotic [95] Apoptosis [93, 94, 96]
tissue
Decreased glycoproteins in acinar cells [97–100] Decreased glycoproteins [88, 90, 92]
in acinar cells
Dilated lumens, loss of cell polarity and [101] Loss of secretory [102, 103]
discharge of mucins into the stroma granules
xerostomia, showed comparable volumes of age post irradiation and can be used as a source of
saliva, but they differed in the severity of their cell-based therapy for the restoration of induced
symptoms. These differences may be due to fea- salivary hypofunction [99]. Similar results have
tures directly related to the mucins, such as the been observed with transplants using adipose
amount of MUC5B bound to the oral mucosa, tissue-derived human mesenchymal stem cells
which is a better predictor of dry mouth com- (hAdMSC) for regenerating salivary glands dam-
pared to levels of free or soluble MUC5B in aged by radiation in mice. Systemic administra-
saliva [91]. Apparently, mucin quality is a rele- tion of hAdMSCs improves salivary flow rate at
vant point, as it has been reported that non- 12 weeks post-radiation, and salivary glands
irradiated patients with dry mouth and low show less damage and higher levels of mucin
salivary flow still display MUC5B on their muco- production than untreated irradiated salivary
sal surfaces [107]. This finding suggests that glands [100]. However, these studies require
mucins retained in the oral mucosa of patients molecular analysis on the quality of secretory
with dry mouth may be less hydrated than normal products, especially of mucins.
subjects [91]. Alliende et al. described how Interestingly, salivary gland acini of irradi-
changes of MUC5B post-translational process- ated rats showed dilated lumens, loss of cell
ing in labial salivary glands of SS patients, had polarity, and discharge of mucins into the stroma
specifically reduced levels of sulfation and were and not to the acinar lumen [101]. Comparable
able to decrease water uptake of mucins, thereby changes have been reported in salivary glands
explaining the dry mouth sensation [65]. It is from SS patients. In this case, loss of cell polar-
noteworthy that saliva contains heterogeneous ity is triggered by the alteration of tight junc-
MUC5B glycoforms; therefore, it would be nections [108] inducing apico-basal relocation of
essary to perform studies to determine the great- protein secretion machinery (e.g., SNARE pro-
est number of MUC5B glycoforms. Thus, teins [95] and Rab3D [67]), accounting for ecto-
changes in the quantity and quality of MUC5B pic mucin exocytosis into the extracellular
will provide insight on why patients who recover matrix (ECM) [109]. Recently it was revealed
submandibular gland salivary flow after radio- that such anomalously located mucins in the
therapy still have oral dryness. ECM can induce a pro-inflammatory response,
Glandular hypofunction after radiotherapy has which may participate in the pathogenesis of this
been attributed to a loss of acinar cells followed disease [110].
by replacement of fibrotic tissue [95]. Based on
this evidence, studies on mice and rats revealed
that post-irradiation transplant with salivary stem 3.3.2 Bioengineered Germ
cells can restore normal salivary function evalu- Transplants for Salivary Gland
ated by an improvement in glandular morphol- Regeneration
ogy, increased number of acinar cells, salivary
flow, and glycoproteins [97, 98]. Moreover, The organs are generated from reciprocal inter-
intraglandular transplantation with bone marrow- actions between the epithelium and mesenchyme
derived clonal mesenchymal stem cells of germ layers. Ogawa and Tsuji developed a
(BM-cMSCs) also preserved acinar cells and bioengineering method termed “organ germ”
salivary gland morphology in murine models that can reproduce salivary gland organogenesis
with radiation damage [99]. In these studies, the through the epithelial-mesenchymal interactions
presence of mucin was used as a functional [111]. The structure of the bioengineered sali-
marker of salivary gland acini, observing an vary glands, including the location of myoepi-
increase of positive acini (AB+) number in thelial cells, aquaporin 5 water channel, and
BM-cMSC-treated salivary glands compared to neuronal connections was analogous to control
controls [99]. This result suggests that mouse submandibular glands [111]. At 30 days
BM-cMSCs can counteract salivary gland dam- after transplantation, the bioengineered subman-
54 I. Castro et al.
dibular gland showed a mucous gland phenotype Cotroneo et al. reported that both ligated and
with strong PAS-positive staining, suggesting deligated glands showed an increase in the pro-
the presence of sialomucins [111]. In the future, portion of ducts and presence of abnormal
it is expected that studies of these regeneration branched entities characterized by short, tubular-
models could provide information about the shaped structures terminating with small acini
quality and functionality of the secretory prod- [90]. These structures were more common in the
ucts that are synthesized rather than their mere deligated glands and resembled those structures
detection. that arise during branching morphogenesis in
embryonic development (day 18) of the subman-
dibular gland. In deligated glands, some of the
3.3.3 R
egeneration of Salivary acini at the ends of branched structures showed
Glands Atrophied by Ligation positive AB/PAS staining, which suggest an
increase of glycoproteins [90].
The obstruction of major excretory ducts (rat, Morphological studies have established that
mouse, rabbit, and cat) generates salivary gland after intra- and extra-oral duct ligation, the deli-
atrophy and inflammation and severely affects gation allows regenerating new salivary gland tis-
the secretory function of remnant parenchyma sue. After 3 days of deligation in rats, an increase
[86, 87, 112–115]. The development of animal in submandibular gland weight, acini size, AQP5
models involving the ligation of the major excre- expression, and glycoprotein content (staining
tory ducts of salivary glands has contributed to with AB/PAS) were observed in comparison to
the understanding of inflammation and atrophy ligated submandibular gland [90]. Following 8
[102, 115, 116]. There are two models of excre- weeks of deligation, the submandibular glands
tory duct ligation: extra-oral duct ligation (includ- recovered half of their weight, normal morphol-
ing lingual nerve cord) and intra-oral duct ogy, differentiated ductal and acinar structures,
ligation, the latter being the most appropriate and the presence of glycoproteins (AB/PAS stain)
model to study obstructive diseases of the sali- in their acini [92]. The parasympathetic nerves
vary gland since it does not damage the nerve were able to re-associate with new target cells to
involved in the normal secretion of saliva [88, form functional neuro-effector junctions [92].
117]. The atrophy followed by extra-oral duct Moreover, after a long period of deligation
ligation is characterized by inflammation, loss of (24 weeks), the rat submandibular gland is able to
acini, proliferation of ductal cells, and secretory recover 92 % of its normal size, salivary flow,
dysfunction [89, 93, 112, 115]. secretion of total protein, normal acinar morphol-
Some studies showed that the secretory func- ogy, and content of granules suggestive of mucin
tion of acinar cells decreases significantly within (AB/PAS positive) [88].
24 h of intra-oral duct ligation [84], and that all The atrophy of the salivary gland is observed
secretory granules disappear after 2–3 days [87], in different circumstances, such as SS, irradiation
the time coinciding with the apoptosis peak of therapy and obstructive sialadenitis, among oth-
acinar cells [94]. In rat submandibular glands, ers. In severe atrophy of the rat submandibular
loss of acini and inflammation was observed gland induced by ligation of the excretory duct,
from 7–14 days of ligation [87, 88, 90]. By the most acinar cells disappeared − mainly through
fourth week of ligation, the atrophy of the glands apoptosis − while the ductal cells proliferated
was evident with a loss of over half of their nor- and dedifferentiated early [96]. Moreover, the
mal weight, while AB/PAS staining showed a gland can survive in the atrophic state almost
large decrease in glycoprotein levels in acinar indefinitely, with the capability of full recovery if
cells [87, 90, 92]. Unlike acini, ductal cells begin deligated. Silver et al. reported that approxi-
to proliferate 2–3 days after ligation [94], form- mately 10 % of the acinar cells survive in atrophy
ing long undifferentiated ductal structures not induced by ligation, where activation of mTOR
discernible as striated, granular, or intercalated. and autophagosomal pathways can help preserve
acinar cells during salivary gland atrophy after only some, if any, of these signs, and so this topic
injury [96]. This study also showed an apparent is still a controversial issue in the field [126, 127].
contradiction, because gene expression analyses Additionally, disorganization of the basal lamina
by quantitative real-time PCR and microarray of (BL) of acini and ducts, which correlates with an
ligated glands revealed sustained transcription of altered secretory pole of acinar cells and morpho-
genes specific to acinar cells, while genes spe- logical changes in the secretory granules, has
cific to ductal cells decreased to baseline levels been consistently found in all patients evaluated
[96]. Taking into account the large loss of acinar [128, 129]. The BL establishes molecular inter-
cells, this would suggest that the remaining aci- actions with plasma membrane receptors and,
nar cells still have a robust transcription of secre- through them, activates signaling pathways
tory proteins. However, evidence of mucins involved in proliferation, differentiation, and
expression has not been generated from this type survival [130, 131]. Moreover, the disruption of
of analysis. tight junctions (TJ) induces relocalization of
secretory machinery proteins. Altered localiza-
tion of TJ proteins in the salivary glands of SS
3.4 ucins and Sjögren’s
M patients promotes the redistribution of both
Syndrome apico-basal Rab3D and SNARE proteins [67,
108, 109]. The presence of these proteins in the
Overview of SS Primary SS is an autoimmune basolateral plasma membrane may explain the
exocrinopathy characterized by chronic mono- ectopic presence of mucins in the ECM [109]. As
nuclear cell infiltration into salivary and lach- these mucins are normally exocyted by the apical
rymal glands [118, 119]. Ro/SS-A and La/ pole of acinar cells, its altered localization may
SS-B autoantibodies are frequently found in be an inflammatory trigger [110]. A relevant
sera of primary SS patients. Their presence is function derived from these interactions is the
associated with prolonged disease duration, maintenance of the differentiated state of the sali-
increased frequency of non-exocrine manifes- vary gland [132]. In other words, cell polarity is
tations, and a higher intensity of lymphocytic important for the organization and function of all
infiltrates invading minor salivary glands epithelia, including secretory epithelium.
[118, 119]. Interestingly, an animal model of SS − the NOD-
strain of mice − shows some alterations resem-
bling SS patients in salivary glands prior to
3.4.1 Factors Related lymphocytic infiltration [133].
to the Secretory Activity
of Salivary Glands of SS
Patients 3.4.2 Mucins in SS Patients
The secretory activity of salivary glands of SS As mentioned above, acinar cells synthesize,
patients is compromised by diverse factors lead- modify and secrete proteins [134], glycopro-
ing to severe dryness of the mouth (xerostomia) teins [53] and are also implicated in the transport
and eyes (keratoconjunctivitis sicca) [120]. and release of electrolytes and water [135]. The
Autoantibodies against muscarinic M3 receptors mechanisms involved in acinar cell differentia-
[121], imbalances of cytokine levels [122], glan- tion are fundamental to the turnover and activa-
dular denervation [123], acinar atrophy and tion of the molecular machinery that participates
decrease of glandular parenchyma [124], redistri- in the synthesis, post-translational processing of
bution of aquaporin-5 in the acinar cells [125], proteins, formation of the secretory granules and
and increased levels of cholinesterase [126] are exocytosis events [134]. The acini of human
the most frequent changes associated with dry- labial salivary glands, which are used as a rele-
ness symptoms. However, SS patients present vant marker of SS diagnostic criteria, are of the
56 I. Castro et al.
seromucous type, with both mucous and serous (Fig. 3.2) [65]. Altogether, these results suggest
acinar cells able to synthesize mucins (MG1 and that a loss of function of mucous acinar cells
MG2) [28]. Previous studies showed that a high occurs in SS patients. Given the high water-
MG1 concentration in the resting saliva of SS retaining capacity of sulfated mucins, these
patients could be due to a low water content or to severe and generalized changes in the quantity of
a low water-retaining capacity. It was then sug- sulfated oligosaccharides of MUC5B could
gested that this might explain xerostomia [136]. account for the xerostomia found in SS patients.
More recently, it has been shown that post- Moreover, Saari et al. have shown a high MG1
translational modifications of mucins, rather than concentration in the resting whole saliva of SS
their net amount, can strongly affect their func- patients, and our results rather support their
tion, thus providing an alternative explanation for hypothesis that low water retaining capacity
xerostomia [11, 137]. might explain xerostomia [136].
Indeed, MUC5B mRNA and protein levels In mucins, sulfated and sialic acid residues
were similar between controls and SS patients, interact with Ca2+ and H+, leading to inter-strand
while sulfo-Lewisa antigen levels were lower in cross-links displacing water molecules and com-
glandular extracts from SS patients (Fig. 3.2) pacting the mucin granule. These events take
[65]. This finding correlated with a dramatically place during the biogenesis of such granules and
lower number of sulfo-Lewisa antigen-positive thus the low sulfation of mucins will affect the
mucous acini (Fig. 3.2) [65]. In SS patients, the organization of secretory granules [22, 23]. Later,
MUC5B electrophoretic pattern was heteroge- during exocytosis, these ions are replaced again
neous, with at least two protein variants present by water; however, in the presence of low-sulfated
in all analyzed samples. Both MUC5B and sul- mucins, such an exchange will not occur [22, 23].
fated MUC5B showed similar electrophoretic It is noteworthy that the presence of big, pleomor-
patterns with two broad bands larger than 200 kd phic and low electron density secretory granules
and migrating within the stacking gel [65]. These in acini of labial salivary glands from SS patients
findings are consistent with an earlier description has been previously reported [128]. Initially, these
of MUC5B in whole saliva and salivary glands changes were interpreted as the result of the
where different glycoforms of MUC5B were fusion of granules. However, considering that sul-
shown [28]. Interestingly, no correlation between fation also plays an important role in the biogen-
whole unstimulated salivary flow (USF) and the esis of mucin granules, the enlarged structures
percentage of mucous acini with sulfo-Lewisa observed could also be a result of the low mucin
antigen was found [65]. As mentioned above, sulfation detected [65]. In brief, the mucous acini
mucins, in particular MUC5B, play an important of labial salivary glands of SS patients experience
role in lubrication, since they preserve mucosa mucin desulfation, in particular of MUC5B.
hydration by the interaction of water molecules This change is not related to the volumes of
with mucin hydrophilic moieties, including sul- saliva produced by SS patients, but are actually
fate, sialyl acid, and hydroxyl groups. Thus, inde- linked to the dry mouth sensation that could be
pendent of normal or decreased USF found in SS due to the decreased water binding ability of
patients, the dry mouth sensation observed in all under-sulfated mucins. Thus, post-translational
cases could be explained by a low sulfation modifications of MUC5B play a role in the
degree of MUC5B and other mucins present in salivary hypofunction observed in SS patients
these glands, as determined by the monoclonal and might contribute significantly to xerosto-
antibody F2 and by microdensitometric analysis mia. An important corollary of these studies is
of AB staining in mucous acini, respectively that future treatments for SS patients should
(Fig. 3.2) [65]. A microdensitometric analysis target not only enhanced water production but
confirmed a decrease in the total sulfated oligo- rather an improved capacity of water retention
saccharides in mucous acini of SS patients, which by modulating the synthesis of mucins with ade-
correlated with a strong BL disorganization quate post-translational processing. The major
therapeutic approach to reducing mouth dry- synthesis is altered. To address this, mRNA and
ness in SS patients is using secretagogues, protein levels were determined, as was the activ-
such as cholinergic agonists, which bind to ity of enzymes involved [139]. Synthesis and
muscarinic receptors increasing the salivary elongation of mucin oligosaccharides initiates
flow, mainly by increasing water transport with the transfer of N-acetylgalactosamine to ser-
[138]. However, these treatments neither con- ine or threonine residues of the peptidic mucin
sider the quantity nor quality of the secretory core [20]. The oligosaccharide may be extended
products present in saliva, including mucins, with galactose (Gal), N-acetylglucosamine
which are essential for lubrication of the (GlcNAc), fucose, or sialic acid [21, 140]. Each
oral epithelium. sequential step is catalyzed by a distinct glycos-
yltransferase [21] (Fig. 3.3). Modifications of
mucin oligosaccharides include sulfation of Gal
3.4.3 E
valuation of the Sulfo-Lewisa and GlcNAc, reactions catalyzed by Gal3-O-
Biosynthesis Pathway sulfotransferases (Gal3ST), and GlcNAc-6-
sulfotransferases (GlcNAc6ST), respectively
If a low degree of MUC5B sulfate content were [140]. In the labial salivary glands of SS patients,
not related to decreased levels of MUC5B poly- levels of Gal3ST activity were significantly
peptides, the question would emerge as to decreased, without changes of mRNA and protein
whether the metabolic pathway of sulfo-Lewisa levels [139]. Importantly, glycosyltransferases
Fig. 3.3 O-glycosylation of mucins in salivary acinar can be further extended, branched, and modified to form
cells. In the Golgi complex, ppGalNAcT initiates complex O-glycans. Terminal GlcNAc residues can be
O-glycosylation by adding GalNAc to the hydroxyl group used as the basis for the attachment of Lewis determi-
of either serine (S) or threonine (T) residues of the mucin nants. The Sulfo-Lewisa antigen is shown. Sulfated
protein backbone. C1GalT adds a Gal residue to synthe- mucins retain large amounts of water on the epithelial sur-
size the Core 1 structure that can be branched by C2GnT, face. In SS patients, the hyposulfated mucins lose hygro-
forming Core 2. The Core 3 structure is synthesized by scopic properties contributing to mouth dryness sensation
C3GnT that adds GlcNAc to GalNAc. Core 3 can be (images Reproduced with permission of Castro et al.
branched by C2GnT2 to form Core 4. All core structures [137])
58 I. Castro et al.
enzymatic activities involved in the glycosylation istics described, the loss of apico-basal polarity
pathway of mucins were similar in controls and of salivary acinar cells is relevant [67, 108, 109,
SS patients [139]. An inverse correlation was 148, 149]. Based on this observation, a working
observed between Gal3ST activity and glandular hypothesis attributes a significant role of the sali-
function measured by scintigraphy, but not with vary gland epithelium itself to the initiation and
USF. An inverse correlation between Gal3ST perpetuation of local autoimmune responses.
activity and focus score, as well as with the auto- According to this idea, molecular changes in the
antibodies Ro/SS-A and La/SS-B were also epithelial cells result in recruitment, homing,
detected. The decrease in sulfotransferase activ- activation, proliferation, and differentiation of
ity may explain the observed mucin hyposulfa- inflammatory cells [148, 150]. As mentioned, the
tion in labial salivary glands from SS patients. loss of apico-basal polarity in salivary acinar
Since no difference was found either in Gal3STs cells of SS patients is associated with the redistri-
mRNA or protein levels, decreased activity was bution of the molecular machinery involved in
not a consequence of gene downregulation. the exocytosis of secretory granules [67, 108,
Interestingly, the sulfotransferase activity corre- 109, 128, 149]. Proteins involved in membrane
lated with secretory function, inflammation, and fusion (SNARE proteins) relocate from the apical
autoimmunity [139]. to the baso-lateral region of acinar cells [109],
These results suggest that pro-inflammatory and redistribution of mature secretory granules in
cytokines may modulate Gal3ST activity, thereby the cytoplasm is observed [67, 109]. In addition,
altering mucin quality and leading to mouth dry- exocytic fusion complexes formed by SNAREs,
ness. Data on this topic in other cellular types, usually present in the apical plasma membrane,
like bovine synoviocites exposed to TNF-α, and secretory granules, are found in the basolat-
showed a decrease of sulfotransferase activity eral plasma membrane of salivary acinar cells
[141], demonstrating that elevated levels of pro- from SS patients. Concomitant with these
inflammatory cytokines, as occurring in rheuma- changes, mucins MUC5B and MUC7 are aber-
toid arthritis and SS, could modulate the activity rantly secreted to the ECM [109].
and expression of glycosyltransferases [139, It seems reasonable to hypothesize that mucins
141]. Reduced sulfation of mucins has been also present in the ECM may trigger an inflammatory
described in inflammatory and neoplastic intesti- response by acting as ligands that activate poten-
nal diseases [142]. Mucins in ulcerative colitis tial receptors of epithelial or immune cells. A key
have shorter oligosaccharide chains and lower question arising in this scenario concerns the
sulfate content than normal colon mucosa [142]. type of receptors that could mediate such a
Sulfomucins in colon adenocarcinoma are nota- response. Toll-like receptors (TLRs) are a class
bly lower than those of the adjacent normal of pattern recognition receptors (PRRs) that play
mucosa [142–144]. The synthesis of these sulfo- a key role in innate immunity and trigger a spe-
mucins involves β3Galactosyltransferase-5 cific immune response [151]. TLRs are stimu-
(β3GalT-5) and Gal3ST-2 [145, 146]. Lower lated by a variety of structural signatures found in
activity and reduced expression of these enzymes pathogens, referred to as pathogen-associated
in non-mucinous adenocarcinoma compared to molecular patterns, and TLR activation induces
adjacent normal mucosa, is thought to contribute the production of pro-inflammatory cytokines
to mucin hyposulfation in this pathology [147]. [151]. TLRs expressed on the cell surface, such
as TLR1,2,4,5,6 and 10, recognize outer cell wall
components of bacteria and fungi, whereas TLRs
3.4.4 M
UC5B as an Inducer expressed in the intracellular compartments
of Inflammation (TLR3,7,8 and 9) are involved in recognition of
nucleic acid components [152]. In addition,
Although the underlying cause of SS-pathogenesis TLRs may also be activated by damage-
is not fully understood, among several character- associated molecular patterns (DAMPs) produced
by the cell or the ECM, endogenous molecules TLR4 and TLR2 [161], or, alternatively, using a
released, activated, or secreted by host cells and specific blocking antibody raised against TLR4
tissues undergoing stress, damage, and non-phys- [110]. In summary, alterations of acinar cell
iological cell death [152]. Interestingly, TLR4 polarity led to the loss of innate epithelial barrier
recognizes DAMPs that contain oligossacharides function, triggering a series of changes that result
and are present in the ECM, such as fibronectin, in the anomalous release of mucins to the
hyaluronic acid (HA), and heparan sulfate [152– ECM. Human salivary MUC5B and Sulfo-Lewisa
154]. Termeer et al. (2002) demonstrated that are recognized by epithelial cell TLR4 and initi-
hyaluronan oligosaccharides activate dendritic ate a pro-inflammatory response. These signals
cells via TLR4 [155]. Furthermore, the polysac- originally produced by epithelial cells could
charide portion has been shown to play a role in attract inflammatory cells, thus perpetuating
salmonella lipopolysaccharide-induced activa- inflammation and the development of chronic
tion through human TLR4 [156]. After ligand disease. The findings highlight the importance of
binding, the cytoplasmic Toll/IL-1 receptor (TIR) salivary gland epithelial cell organization in con-
domain of the TLRs associates with the TIR trolling innate immunity, and the etiopathogene-
domain of adaptor proteins MyD88, TIRAP, sis of SS [110].
TRIF and TRAM [157, 158]. Moreover, and
depending on the ligand, TLR4 activation may
occur in a manner either dependent or indepen- 3.4.5 M
UC1/SEC AND MUC1/Y
dent of MyD88. In the first case, MyD88 and Over-Expression Is Associated
TIRAP protein adapters are recruited to the TIR with Inflammation in SS
domain and induce a signaling pathway leading
to the expression of pro-inflammatory cytokines The secreted MUC1 isoform (MUC1/SEC),
and chemokines. Alternatively, the MyD88- which lacks the cytoplasmic and transmembrane
independent pathway engages TRAM and TRIF domains, contains a unique 11 amino-acid pep-
as protein adapters to induce type-I interferon tide at the COOH terminus that is not found in
[157]. other isoforms (Fig. 3.1) [57, 162]. This sequence
Normal salivary acinar cells express TLR4, is referred to as the immuno-enhancing peptide
which is significantly increased in SS patients (IEP) due to its ability to stimulate the immune
[159, 160]. Moreover, chronic inflammation is response [163] probably by STAT-1 up-regulation
evident in the salivary glands of SS patients, [164]. IEP has been proposed to modulate both
although the mechanisms that trigger these pro- the innate and adaptive immune responses [163,
cesses are not known. In this context, the hypoth- 165]. MUC1/SEC may induce over-expression of
esis that was tested was whether or not the cytokines through its IEP and/or via formation of
salivary mucins are involved in the expression of a MUC1/SEC-MUC1/Y complex [163].
pro-inflammatory cytokines, exploring the MUC1/Y is a MUC1 transmembrane protein
molecular sensor involved in this response. without VNTR (Fig. 3.1) [166–168]. The interac-
Furthermore, it was investigated whether mucin tion between MUC1/Y and MUC1/SEC can be
oligosaccharides might act as DAMPs that are compared to a receptor-ligand interaction that
recognized by TLR4, and activate the innate might trigger cytokine production and thereby
immune response. Human salivary epithelial modulate the immune response [166]. In addi-
cells (HSG) were stimulated with purified tion, the formation of a receptor-ligand complex
MUC5B or Sulfo-Lewisa, both inducing a signifi- between MUC1/SEC and MUC1/Y in mammary
cant increase of CXCL8, TNF-α, IFN-α, IFN-β, tumors initiates a cell-signaling response that
IL-6, IL-1β, but not BAFF [110]. Cytokine alters cell morphology [163, 164]. On the other
induction was mediated by TLR4 as shown by hand, MUC1/Y has been associated with tran-
using the TBX2-peptide, an inhibitor of TIRAP, scriptional induction of pro-inflammatory cyto-
which is a signaling molecule downstream of kine genes via NF-κB [169]. Considering these
60 I. Castro et al.
previous findings and the relevance of MUC1 in provide light on the specific function of particular
several pathologies, it was interesting to deter- MUC1 isoforms, such as MUC1/SEC and
mine whether MUC1/SEC and MUC1/Y are MUC1/Y.
expressed in the salivary glands of SS patients The immunohistochemical analysis of sali-
and which cytokines are able to induce their vary glands revealed a significantly higher pro-
expression [7]. portion of acini with MUC1/SEC in the
Significantly higher mRNA and protein levels cytoplasm of SS patients, while for controls, a
of both these variants were found in SS patients significantly higher percentage of acini with
[7]. The MUC1 gene is subject to several control MUC1/SEC in the apical region and low pres-
mechanisms by cytokines (IFN-γ, TNF-α, IL-7) ence in the cytoplasm was observed [7]. This
and epigenetic factors (methylation of promoter’s cytoplasmic distribution of MUC1/SEC observed
CpG islands, histone modifications, miRNA in the acini of SS patients was associated with
effects) [170]. However, regulatory mechanisms the loss of cell polarity, where the increased
involved in differential expression of MUC1 intensity of cytoplasmic MUC1/SEC staining
splice variants, specifically MUC1/SEC and coincided with increased acinar alterations[7].
MUC1/Y, have not yet been reported. MUC1/ These results confirmed observations showing
SEC has been associated with progressive altered distribution and accumulation of MUC1/
tumoral development inhibition and anti-tumoral VNTR in the cytoplasm of acinar and ductal
immune responses supported by increased cells from labial salivary glands of SS patients
STAT-1 expression [164]. Although this mecha- [7]. MUC1/Y is redistributed from the apical
nism has not been fully elucidated, a plausible region of acini in labial salivary glands of con-
explanation is that such effects might be medi- trols to the cytoplasm and nuclei in labial sali-
ated by STAT-1 activation with: (1) over- vary glands from SS patients [7]. Nuclear
expression of IFN-γ responsive signal transducer localization of some MUC1 isoforms has been
and/or (2) activation of pro-apoptotic and pro- previously described, but the mechanism
inflammatory genes [164, 171]. Thus, higher lev- involved is still unknown. Such nuclear function
els of MUC1/SEC and MUC1/Y mRNA and has been related with increased transcription of
protein observed in the salivary glands of SS pro-inflammatory cytokines [169]. In summary,
patients may induce the synthesis of cytokines. the over-expression and aberrant localization of
Interestingly, we have recently demonstrated that MUC1/SEC and MUC1/Y observed in the labial
MUC1/SEC and MUC1/Y mRNA levels are salivary glands of SS patients and their over-
induced by TNF-α and IFN-γ in HSG cells [7], expression induced in vitro in HSG cells by pro-
supporting previous evidence of a self- inflammatory cytokines support previous
perpetuating mucin-cytokine signaling loop in evidence of a self-perpetuating mucin-cytokine
inflammatory conditions [110]. Studies evaluat- signaling loop that may facilitate the mainte-
ing MUC1 function in salivary glands are not nance of an inflammatory environment leading
available; however, in other tissues, such as lung, to the disruption of salivary glandular homeosta-
bowel, and brain, studies using MUC1 knockout sis in SS patients [110].
mice suggest an anti-inflammatory role [172–
174]. In some tissues, a complete loss of all
MUC1 isoforms was reported, and in the bowel, 3.4.6 MUC7 in SS Patients
the anti-inflammatory function of MUC1 was
linked to the mucosa protection provided by MUC7 studies in normal human labial salivary
MUC1 anchored to the membrane [173]. These glands were previously reported by Veerman
results have been reproduced by MUC1 knock- et al., showing cytoplasmic localization in acinar
down with a siRNA targeted to a sequence shared cells of serous acini [28]. Preliminary studies
by all known MUC1 variants[173]. However, it is showed that MUC7mRNA levels are similar in
important to emphasize that these studies do not the salivary glands of SS patients and in controls;
however, protein levels detected by western blot Most salivary gland tumors − both benign and
and immunohistochemistry showed a significant malignant − express MUC1, where positive cells
increase in SS patients [67]. MUC7 cytoplasmic range from 67 to 100 %. However, it has been
localization in serous acini supported previous difficult to establish whether MUC1 over-
findings [28]. More studies are needed to collect expression is higher in malignant than in benign
further information on MUC7 changes in SS salivary tumors, probably due to the use of differ-
patients and to evaluate whether this mucin con- ent MUC1 mAb (Fig. 3.1). Sensitivity of anti-
tributes to the xerostomia that afflicts these bodies depends on differential glycosylation
patients. degrees between MUC1 isoforms, thus affecting
epitope recognition in immunohistochemical
assays. Additionally, factors such as small sample
3.5 Mucins in Salivary Gland sizes and varied scoring systems in various stud-
Tumors ies make it difficult to analyze the data in order to
estimate MUC1 over-expression.
Cancer cells express aberrant forms or high Pleomorphic adenoma (PA), or benign mixed
amounts of mucins, which have been suggested tumors of salivary glands, are the most common
as molecular markers of malignant transforma- benign neoplasms, representing 40–70 % of
tion in several organs and tissues [175]. Altered cases [181]. Etiology and mechanisms involved
MUC1 expression in malignancies occurs in vari- in PA growth are not fully understood [180]. PA
ous modes, including up-regulation, mis- might display a wide range of recurrence rates
localization, and aberrant glycosylation. MUC1 (2.5–32.5 %).
expression is related to aggressive tumor behav- Immunohistochemical markers have been
ior and a poor prognosis for patients with human used to explain this behavior and predict recur-
neoplasms [176]. Secreted mucin expression pro- rence. Primary PA shows scarce MUC1 expres-
files of adenocarcinomas have been associated sion, while recurrent PAs exhibit higher MUC1
with etiology [177], tumor progression [8], prog- levels, suggesting the association of this mucin
nosis [178] and histologic characteristics [61]. with recurrence [181] (Table 3.3). Using a
Salivary gland neoplasms are characterized by MUC1/DF Ab (Fig. 3.1), Hamada et al. found
morphological variability. These tumors imitate high MUC1 expression in patients with PA recur-
the histology, and may arise from epithelial, mes- rence (RPA) and proposed that MUC1 would be
enchymal, and/or lymphoid components. Several an independent risk factor of recurrence [8]. RPA
benign and malignant salivary gland tumors also showed malignant transformation in areas
showed abundant extracellular or intracellular where MUC1 expression was higher. However,
mucins [179]. Brieger et al. found that MUC1 did not correlate
with PA recurrence in parotid glands, suggesting
that MUC1 might actually play a role in cellular
3.5.1 Mucins in Benign Salivary dysfunction [180]. It is accepted that the biologi-
Gland Neoplasms cal behavior of tumors during their progression is
influenced by changes in structure and the distri-
Overall, MUC1 is the mucin most related to can- bution of cell-surface glycoproteins. Soares et al.
cer and it is transcribed as multiple [80] alterna- showed differential MUC1 expression during
tively spliced variants; some of these have been malignant transformation of PAs toward widely
detected in salivary gland tumors. As mentioned, invasive carcinomas [178]. Therefore, low MUC1
MUC1 can be found as plasma membrane or expression in primary PA may indicate a reduced
secreted isoforms, and its over-expression has invasive potential and aggressiveness, while a
been associated with biochemical events that higher expression may be indicative of an aggres-
occur in carcinogenesis and/or tumor invasion [8, sive biological behavior, such as can be observed
61, 178, 180]. in recurrent PAs and carcinoma ex-pleomorphic
62 I. Castro et al.
Table 3.3 Immunoreactivity of MUC1, MUC2, MUC4, MUC5AC, MUC5B and MUC6 in salivary gland tumors
MUC1 MUC2 MUC4 MUC5AC MUC5B MUC6
ACA 19/21(90) 12/13(92) 0/8 0/11 NT 0/11
ACC 60/60(100) 2/26(8) 1/2(50) 0/20 NT 0/20
CA 1/1(100) 0/1 1/1(100) 0/1 0/1 1/1(100)
Ca-ex-PA 11/11(100) NT NT NT NT NT
MEC 91/114(84) 24/112(21) 109/131(83) 33/40(83) 33/40(83) 15/60(25)
PLGA NT NT 1/1(100) NT NT NT
SDC 2/2(100) 6/6(100) 1/1(100) 4/5(80) 4/5(80) 6/6(100)
PA 106/158(67) 27/70(31) 7/51(14) NT NT 19/49(39)
WT 28/29(97) 22/22(100) 3/3(100) NT NT NT
In parentheses are the percentages of positive tumors
ACA Acinic cell adenocarcinoma, ACC Adenoid cystic carcinoma, CA Cystadenoma, Ca-ex-PA Carcinoma ex-
pleomorphic adenoma, MEC Mucoepidermoid carcinoma, PLGA Plymorphous low-grade adenocarcinoma, SDC
Salivary duct carcinoma, PA Pleomorphic adenoma, WT Warthin’s tumor, NT Not tested
adenoma [8, 61, 178]. Moreover, Soares et al. diversity with primary MEC displaying a variety
showed an augmented MUC1 expression in RPA of biologic behaviors. While the low- and
relative to primary PA. Altogether, these findings intermediate-grade MECs are of a benign-like
suggest that MUC1 may be a useful indicator of nature with high survival rates, the high-grade
a potential malignancy [178]. MEC usually has a poor prognosis [184]. Other
MUC2 has also been studied in benign sali- less common mucin-producing tumors are col-
vary gland tumors and its expression associated loid (mucinous) carcinoma (CC), mucinous cyst-
with indolent behavior in both human and animal adenocarcinoma (MCA), salivary duct carcinoma
neoplasms [182]. PAs showed a weak cytoplas- (SDC), signet ring cell carcinoma, adenocarci-
mic MUC2 signal in single cells; these findings noma (NOS), and, occasionally, a metastatic
had no relation to the clinical behavior [61]. tumor [8]. Among membrane-bound mucins,
Furthermore, MUC1 and MUC2 have been MUC1 is frequently overexpressed in carcino-
detected in Warthin’s tumors (WT), a benign, mas, particularly adenocarcinomas, which in
well-encapsulated neoplasm usually originating most tumor types are correlated to an adverse
in the caudal pole of parotid glands. Mannweiler effect on prognosis [175], increased metastatic
et al. found that WTs were characterized by potential, and poor survival rates [175]. MUC1
strong MUC1 and MUC2 expression with 90 % expression is elevated in MECs, and is positively
of tumor cells showing a diffuse cytoplasmic correlated with lymph node metastasis in the clin-
staining. In addition, membranes of luminal cells ical stage; it is also a strong independent prognos-
exhibited higher MUC1 expression compared to tic factor [184]. Besides localizing in the apical
MUC2 [61]. Conversely, Yamada et al. found that membranes of luminal tumoral cells, MUC1 was
MUC1 was restricted to basal tumor cells [183]. also detected in the cytoplasm and sub-cellular
membranes of the epidermoid, intermediate,
mucous, and clear MEC tumor cells. Studies on
3.5.2 M
ucins in Malignant Salivary salivary MECs have shown a relationship between
Gland Neoplasms MUC1 expression in tumor cells and outcome
[66, 185]. MUC1 expression in 5–10 % of MEC
Mucoepidermoid carcinoma (MEC) is the most tumor cells is enough to indicate an adverse prog-
common malignant neoplasm of the salivary nosis, such as recurrence, metastasis, and/or can-
glands, accounting for about 30–40 % of all sali- cer-related death. Alos et al. [66] and Handra- Luca
vary carcinomas. MEC may produce high levels et al. [185] detected MUC1 expression in MEC
of extracellular mucin and possess morphologic using Ma695 monoclonal Ab (Fig. 3.1), but it
remains to be determined whether other MUC1 with apical predominance. Regarding MUC2,
glycoforms behave similarly and what their prog- only 2 of 9 ACC cases were positive [61]. Another
nostic potential in this type of tumor is. malignant, mucin-producing neoplasm is the
MUC2 is highly expressed in mucinous carci- acinic cell adenocarcinoma (ACA) that showed
nomas, such as colon, breast, pancreas, ovary, cytoplasmic MUC1 positivity in all studied
and stomach [186]. However, its expression is tumors. Conversely, ACC immunoreactivity was
low in malignant salivary gland neoplasms. Alos rarely observed in glandular structures [61]. This
et al. detected low MUC2 expression in MEC (5 difference could be explained by alterations in
% of tumors analyzed with 5–10 % having posi- processing and targeting pathways, allowing the
tive cells). Furthermore, MUC2 content was not intracellular accumulation of MUC1 in ACA
related to MEC histologic grade and prognosis [61]. On the other hand, the difference observed
for the patients [66]. In other research, in MUC2 immunoreactivity between ACA and
Mannweiler et al. observed MUC2 positivity in ACC, could be partially explained by different
all cases, but only 5–25 % of the tumor cells were cell populations found in these tumors.
stained [61]. In both studies, the same Ab (NCL- Particularly in ACA, tumor cells laden with
clone Ccp58) was used, and the differences secretory granules are frequent, while ACC
observed may be due to sample processing, e.g., showed a large number of basaloid cells and few
antigen retrieval. Muc2 knockout mice frequently cells with secretory characteristics [61].
developed adenomas in the small intestine that
progressed to invasive adenocarcinomas[187]. Conclusions
Also, MUC2 was highly expressed in non- The specific roles played by mucins in human
invasive tumors of the pancreas and intrahepatic salivary gland development have not been
bile duct, which show more favorable outcome fully studied, mostly due to ethical and techni-
than invasive carcinomas of the pancreas and cal restrictions. Most of the studies on mucin
intrahepatic bile duct [176]. expression during development have been per-
MUC4, MUC5AC, MUC5B, and MUC6 are formed on murine models, showing that mucin
also expressed in MEC [66, 185]. Handra-Luca is the initial marker of epithelial differentiation
et al. found that MUC4 is redistributed from an in the mouse submandibular gland. A role for
apical surface to a basolateral surface in duct MUC1 in glandular morphogenesis was sug-
cells in intermediate and epidermoid tumor cells. gested due to the coincident onset of glandular
However, MUC4 was not related to prognosis differentiation with changes of MUC1 expres-
and seems to be associated with MEC grades sion pattern during embryonic development. It
[185]. MUC5AC was expressed in intermediate has been suggested that MUC1 may induce
cells of the tumor, considered undifferentiated changes in tissue architecture in development
tumor cells, and high grade MEC showed reac- and also in cancer, where it is frequently over-
tive cells [185]. MUC5B was similarly distrib- expressed, misglycosylated, and redistributed.
uted to MUC5AC, mainly in low-grade tumors, Mucin expression and glycosylation have been
but not related with tumoral progression and associated with etiology, tumor progression,
patient outcome. MUC6 was detected in MEC prognosis, and histologic characteristics. The
predominantly in the cytoplasm of mucous cells altered glycosylation of mucins confers a wide
and was not related to the histological grade of range of potential ligands on tumor cells for
the tumor, the tumor progression, or the patient’s interaction with other receptors at the cell sur-
outcome [66] (Table 3.3). In adenoid cystic carci- face, contributing to the survival of these
noma (ACC), Mannweiler et al. observed MUC1 tumor cells during invasion and metastasis.
staining in all tumors, although immunoreactivity Mucins are used as diagnostic markers in can-
was heterogeneous with tumor areas strongly cer, and are being researched as therapeutic
positive and others negative [61]. MUC1 local- targets for this disease. In addition, mucins
ization was preferentially in glandular structures have been evaluated as markers to address the
64 I. Castro et al.
functionality of the glandular acini after dam- The over-expression and aberrant localiza-
age and/or regeneration therapy. In the future, tion of MUC1/SEC and MUC1/Y observed in
it is expected that studies using regeneration the LSG of SS patients and their over-expres-
models as bioengineered submandibular sion induced in vitro in HSG cells by pro-
glands could provide information about the inflammatory cytokines are consistent with
quality and functionality of the mucins that are previous evidence that points toward the exis-
rather than their mere detection. tence of a self-perpetuating mucin-cytokine
A decreased quality of salivary mucins and signaling loop that may facilitate the mainte-
reduced salivary flow lead to xerostomia. nance of an inflammatory environment lead-
These changes produce a variety of oral and ing to disruption of salivary glandular
dental disorders that affect the quality of life of homeostasis in SS patients [7].
SS patients. The main therapeutic approach to In short, although mucins have been studied
reduce mouth dryness in SS patients involves for many years, their exact roles and mecha-
the use of secretagogues. These cholinergic nisms in salivary gland development and dis-
agonists bind to muscarinic receptors and eases are still in the process of being discovered.
increase salivary flow, mainly by enhancing As mentioned, the reasons are various and
water release; the treatments neither consider diverse, such as structural complexity, glyco-
the quantity nor quality of the secretory prod- sylation processing and the intricate and com-
ucts present in saliva, such as mucins, which plex processes in which they participate, among
are complex O-linked glycoproteins with others. Fortunately, today there are more meth-
sialylated and/or sulfated oligosaccharides odological tools to study them, which augur a
attached to their protein backbone. This char- more promising future in this field.
acteristic allows mucins to bind large amounts
of water and lubricate the oral epithelium. In Acknowledgements Supported by Fondecyt-Chile
salivary glands of SS patients, altered traffick- [#1120062] (MJG, SG and SA), Fondecyt [#1160015]
(MJG, SG, SA and IC), PhD fellowship Conicyt-Chile
ing and maturation of salivary mucins were (MJB and JC).
observed. These alterations are likely to con-
tribute to the dryness sensation in SS patients.
A better characterization of the molecular References
mechanisms that cause these alterations will
favor the development of more effective thera- 1. Milne RW, Dawes C. The relative contributions of
pies to treat xerostomia in SS patients [137]. different salivary glands to the blood group activity
In SS patients, the alterations in cell polar- of whole saliva in humans. Vox Sang. 1973;25(4):
298–307.
ity lead to the loss of the innate epithelial bar- 2. Gendler SJ, Burchell JM, Duhig T, Lamport D,
rier function, triggering a series of changes White R, Parker M, et al. Cloning of partial cDNA
that result in the release of mucins to the encoding differentiation and tumor-associated mucin
ECM. Human salivary MUC5B and Sulfo- glycoproteins expressed by human mammary epi-
thelium. Proc Natl Acad Sci U S A. 1987;84(17):
Lewisa (just to be consistent with Sulfo-Lewis 6060–4.
in other sections of the book chapter). are rec- 3. Swallow DM, Gendler S, Griffiths B, Corney G,
ognized by epithelial cell TLR4 and initiate a Taylor-Papadimitriou J, Bramwell ME. The human
pro-inflammatory response. These signals, tumour-associated epithelial mucins are coded by an
expressed hypervariable gene locus PUM. Nature.
initially produced by epithelial cells, could 1987;328(6125):82–4.
attract inflammatory cells, which perpetuate 4. Pemberton L, Taylor-Papadimitriou J, Gendler
inflammation and the development of chronic SJ. Antibodies to the cytoplasmic domain of the
disease. These findings highlight the impor- MUC1 mucin show conservation throughout mam-
mals. Biochem Biophys Res Commun. 1992;185(1):
tance of salivary gland epithelial cell organi- 167–75.
zation in controlling innate immunity, and in 5. Croce MV, Isla-Larrain M, Remes-Lenicov F,
the etiopathogenesis of SS [110, 150]. Colussi AG, Lacunza E, Kim KC, et al. MUC1
c ytoplasmic tail detection using CT33 polyclonal and 20. Ten Hagen KG, Fritz TA, Tabak LA. All in the family:
CT2 monoclonal antibodies in breast and colorectal the UDP-GalNAc:polypeptide N-acetylgalactosami
tissue. Histol Histopathol. 2006;21(8):849–55. nyltransferases. Glycobiology. 2003;13(1):1R–16R.
6. Teshima TH, Ianez RF, Coutinho-Camillo CM, 21. Brockhausen I. Glycodynamics of mucin biosynthe-
Buim ME, Soares FA, Lourenco SV. Development sis in gastrointestinal tumor cells. Adv Exp Med
of human minor salivary glands: expression of Biol. 2003;535:163–88.
mucins according to stage of morphogenesis. J Anat. 22. Verdugo P. Goblet cells secretion and mucogenesis.
2011;219(3):410–7. Annu Rev Physiol. 1990;52:157–76.
7. Sung HH, Castro I, Gonzalez S, Aguilera S, 23. Perez-Vilar J, Hill RL. The structure and assembly
Smorodinsky NI, Quest A, et al. MUC1/SEC and of secreted mucins. J Biol Chem. 1999;274(45):
MUC1/Y over-expression is associated with inflam- 31751–4.
mation in Sjogren’s syndrome. Oral Dis. 24. Nielsen PA, Bennett EP, Wandall HH, Therkildsen
2015;21(6):730–8. MH, Hannibal J, Clausen H. Identification of a major
8. Hamada T, Matsukita S, Goto M, Kitajima S, Batra human high molecular weight salivary mucin (MG1)
SK, Irimura T, et al. Mucin expression in pleomor- as tracheobronchial mucin MUC5B. Glycobiology.
phic adenoma of salivary gland: a potential role for 1997;7(3):413–9.
MUC1 as a marker to predict recurrence. J Clin 25. Wickstrom C, Davies JR, Eriksen GV, Veerman EC,
Pathol. 2004;57(8):813–21. Carlstedt I. MUC5B is a major gel-forming, oligo-
9. Gendler SJ, Spicer AP. Epithelial mucin genes. meric mucin from human salivary gland, respiratory
Annu Rev Physiol. 1995;57:607–34. tract and endocervix: identification of glycoforms
10. Tabak LA, Levine MJ, Mandel ID, Ellison SA. Role and C-terminal cleavage. Biochem J. 1998;334(Pt
of salivary mucins in the protection of the oral cav- 3):685–93.
ity. J Oral Pathol. 1982;11(1):–17. 26. Levine MJ, Reddy MS, Tabak LA, Loomis RE,
11. Chaudhury NM, Shirlaw P, Pramanik R, Carpenter Bergey EJ, Jones PC, et al. Structural aspects of
GH, Proctor GB. Changes in saliva rheological prop- salivary glycoproteins. J Dent Res. 1987;66(2):

erties and mucin glycosylation in dry mouth. J Dent 436–41.
Res. 2015;94(12):1660–7. 27. Thomsson KA, Prakobphol A, Leffler H, Reddy MS,
12. van der Reijden WA, Veerman EC, Amerongen Levine MJ, Fisher SJ, et al. The salivary mucin MG1
AV. Shear rate dependent viscoelastic behav- (MUC5B) carries a repertoire of unique oligosac-
ior of human glandular salivas. Biorheology. charides that is large and diverse. Glycobiology.
1993;30(2):141–52. 2002;12(1):1–14.
13. Offner GD, Troxler RF. Heterogeneity of high- 28. Veerman EC, van den Keijbus PA, Nazmi K, Vos W,
molecular-weight human salivary mucins. Adv Dent van der Wal JE, Bloemena E, et al. Distinct localiza-
Res. 2000;14:69–75. tion of MUC5B glycoforms in the human salivary
14. Corfield AP. Mucins: a biologically relevant glycan glands. Glycobiology. 2003;13(5):363–6.
barrier in mucosal protection. Biochim Biophys 29. Thornton DJ, Khan N, Mehrotra R, Howard M,
Acta. 2015;1850(1):236–52. Veerman E, Packer NH, et al. Salivary mucin MG1 is
15. Hijikata M, Matsushita I, Tanaka G, Tsuchiya T, Ito comprised almost entirely of different glycosylated
H, Tokunaga K, et al. Molecular cloning of two forms of the MUC5B gene product. Glycobiology.
novel mucin-like genes in the disease-susceptibility 1999;9(3):293–302.
locus for diffuse panbronchiolitis. Hum Genet. 30. Sheehan JK, Howard M, Richardson PS, Longwill T,
2011;129(2):117–28. Thornton DJ. Physical characterization of a low-
16. Itoh Y, Kamata-Sakurai M, Denda-Nagai K, Nagai charge glycoform of the MUC5B mucin comprising
S, Tsuiji M, Ishii-Schrade K, et al. Identification and the gel-phase of an asthmatic respiratory mucous
expression of human epiglycanin/MUC21: a novel plug. Biochem J. 1999;338(Pt 2):507–13.
transmembrane mucin. Glycobiology. 2008;18(1): 31. Veerman EC, Bolscher JG, Appelmelk BJ, Bloemena
74–83. E, van den Berg TK, Nieuw Amerongen AV. A
17. Thornton DJ, Rousseau K, McGuckin MA. Structure monoclonal antibody directed against high M(r) sali-
and function of the polymeric mucins in airways vary mucins recognizes the SO3-3Gal beta
mucus. Annu Rev Physiol. 2008;70:459–86. 1-3GlcNAc moiety of sulfo-Lewis(a): a histochemi-
18. Asker N, Axelsson MA, Olofsson SO, Hansson cal survey of human and rat tissue. Glycobiology.
GC. Human MUC5AC mucin dimerizes in the rough 1997;7(1):37–43.
endoplasmic reticulum, similarly to the MUC2 32. Amerongen AV, Bolscher JG, Veerman EC. Salivary
mucin. Biochem J. 1998;335(Pt 2):381–7. mucins: protective functions in relation to their
19. Asker N, Axelsson MA, Olofsson SO, Hansson diversity. Glycobiology. 1995;5(8):733–40.
GC. Dimerization of the human MUC2 mucin in the 33. Tabak LA. In defense of the oral cavity: structure,
endoplasmic reticulum is followed by a biosynthesis, and function of salivary mucins. Annu
N-glycosylation-dependent transfer of the mono- Rev Physiol. 1995;57:547–64.
and dimers to the Golgi apparatus. J Biol Chem. 34. Tabak LA, Levine MJ, Jain NK, Bryan AR, Cohen
1998;273(30):18857–63. RE, Monte LD, et al. Adsorption of human salivary
66 I. Castro et al.
mucins to hydroxyapatite. Arch Oral Biol. c ontains a binding domain for oral Streptococci and
1985;30(5):423–7. exhibits candidacidal activity. Biochem
35. Nieuw Amerongen AV, Oderkerk CH, Veerman J. 2000;345(Pt 3):557–64.
EC. Influence of phytate on the adsorption of human 49. Liu B, Rayment S, Oppenheim FG, Troxler
salivary mucins onto hydroxyapatite. J Biol Buccale. RF. Isolation of human salivary mucin MG2 by a
1988;16(4):203–8. novel method and characterization of its interactions
36. Veerman EC, Ligtenberg AJ, Schenkels LC, with oral bacteria. Arch Biochem Biophys.
Walgreen-Weterings E, Nieuw Amerongen 1999;364(2):286–93.
AV. Binding of human high-molecular-weight sali- 50. Groenink J, Walgreen-Weterings E, Nazmi K,
vary mucins (MG1) to Hemophilus parainfluenzae. Bolscher JG, Veerman EC, van Winkelhoff AJ, et al.
J Dent Res. 1995;74(1):351–7. Salivary lactoferrin and low-Mr mucin MG2 in
37. Veerman EC, Bank CM, Namavar F, Appelmelk BJ, Actinobacillus actinomycetemcomitans-associated
Bolscher JG, Nieuw Amerongen AV. Sulfated gly- periodontitis. J Clin Periodontol. 1999;26(5):269–75.
cans on oral mucin as receptors for Helicobacter 51. Bobek LA, Situ H. MUC7 20-Mer: investigation of
pylori. Glycobiology. 1997;7(6):737–43. antimicrobial activity, secondary structure, and pos-
38. Bobek LA, Tsai H, Biesbrock AR, Levine sible mechanism of antifungal action. Antimicrob
MJ. Molecular cloning, sequence, and specificity of Agents Chemother. 2003;47(2):643–52.
expression of the gene encoding the low molecular 52. Senapati S, Das S, Batra SK. Mucin-interacting pro-
weight human salivary mucin (MUC7). J Biol Chem. teins: from function to therapeutics. Trends Biochem
1993;268(27):20563–9. Sci. 2010;35(4):236–45.
39. Mehrotra R, Thornton DJ, Sheehan JK. Isolation and 53. Iontcheva I, Oppenheim FG, Troxler RF. Human sali-
physical characterization of the MUC7 (MG2) vary mucin MG1 selectively forms heterotypic com-
mucin from saliva: evidence for self-association. plexes with amylase, proline-rich proteins, statherin,
Biochem J. 1998;334(Pt 2):415–22. and histatins. J Dent Res. 1997;76(3):734–43.
40. Prakobphol A, Levine MJ, Tabak LA, Reddy 54. Gendler SJ, Lancaster CA, Taylor-Papadimitriou J,
MS. Purification of a low-molecular-weight, mucin- Duhig T, Peat N, Burchell J, et al. Molecular cloning
type glycoprotein from human submandibular- and expression of human tumor-associated polymor-
sublingual saliva. Carbohydr Res. 1982;108(1): phic epithelial mucin. J Biol Chem. 1990;265(25):
111–22. 15286–93.
41. Ramasubbu N, Reddy MS, Bergey EJ, Haraszthy 55. Ligtenberg MJ, Kruijshaar L, Buijs F, van Meijer M,
GG, Soni SD, Levine MJ. Large-scale purification Litvinov SV, Hilkens J. Cell-associated episialin is a
and characterization of the major phosphoproteins complex containing two proteins derived from a com-
and mucins of human submandibular-sublingual mon precursor. J Biol Chem. 1992;267(9):6171–7.
saliva. Biochem J. 1991;280(Pt 2):341–52. 56. Hattrup CL, Gendler SJ. Structure and function of
42. Reddy MS, Bobek LA, Haraszthy GG, Biesbrock the cell surface (tethered) mucins. Annu Rev Physiol.
AR, Levine MJ. Structural features of the low- 2008;70:431–57.
molecular- mass human salivary mucin. Biochem 57. Smorodinsky N, Weiss M, Hartmann ML, Baruch A,
J. 1992;287(Pt 2):639–43. Harness E, Yaakobovitz M, et al. Detection of a
43. Reddy MS, Levine MJ, Prakobphol secreted MUC1/SEC protein by MUC1 isoform spe-
A. Oligosaccharide structures of the low-molecular- cific monoclonal antibodies. Biochem Biophys Res
weight salivary mucin from a normal individual and Commun. 1996;228(1):115–21.
one with cystic fibrosis. J Dent Res. 1985;64(1):33–6. 58. Thathiah A, Blobel CP, Carson DD. Tumor necrosis
44. Amerongen AV, Veerman EC. Saliva--the defender factor-alpha converting enzyme/ADAM 17 medi
of the oral cavity. Oral Dis. 2002;8(1):12–22. ates MUC1 shedding. J Biol Chem. 2003;278(5):
45. Reddy MS, Levine MJ, Paranchych W. Low- 3386–94.
molecular-mass human salivary mucin, MG2: struc- 59. Singh PK, Hollingsworth MA. Cell surface-
ture and binding of Pseudomonas aeruginosa. Crit associated mucins in signal transduction. Trends
Rev Oral Biol Med. 1993;4(3–4):315–23. Cell Biol. 2006;16(9):467–76.
46. Groenink J, Ligtenberg AJ, Veerman EC, Bolscher 60. Miletich I. Introduction to salivary glands: structure,
JG, Nieuw Amerongen AV. Interaction of the sali- function and embryonic development. In: Tucker
vary low-molecular-weight mucin (MG2) with AS, Miletich I, editors. Salivary glands develop-
Actinobacillus actinomycetemcomitans. Antonie ment, adaptations and disease. Basel/New York:
Van Leeuwenhoek. 1996;70(1):79–87. Karger; 2010.
47. Moshier A, Reddy MS, Scannapieco FA. Role of 61. Mannweiler S, Beham A, Langner C. MUC1 and
type 1 fimbriae in the adhesion of Escherichia coli to MUC2 expression in salivary gland tumors and in
salivary mucin and secretory immunoglobulin non-neoplastic salivary gland tissue. APMIS.
A. Curr Microbiol. 1996;33(3):200–8. 2003;111(10):978–84.
48. Liu B, Rayment SA, Gyurko C, Oppenheim FG, 62. Chen Y, Zhao YH, Kalaslavadi TB, Hamati E,
Offner GD, Troxler RF. The recombinant N-terminal Nehrke K, Le AD, et al. Genome-wide search and
region of human salivary mucin MG2 (MUC7) identification of a novel gel-forming mucin MUC19/
Muc19 in glandular tissues. Am J Respir Cell Mol 77. Hudson MJ, Stamp GW, Chaudhary KS, Hewitt R,
Biol. 2004;30(2):155–65. Stubbs AP, Abel PD, et al. Human MUC1 mucin: a
63. Culp DJ, Robinson B, Cash MN, Bhattacharyya I, potent glandular morphogen. J Pathol.
Stewart C, Cuadra-Saenz G. Salivary mucin 19 gly- 2001;194(3):373–83.
coproteins: innate immune functions in 78. Miles DW, Taylor-Papadimitriou J. Therapeutic
Streptococcus mutans-induced caries in mice and aspects of polymorphic epithelial mucin in adeno-
evidence for expression in human saliva. J Biol carcinoma. Pharmacol Ther. 1999;82(1):97–106.
Chem. 2015;290(5):2993–3008. 79. Gendler SJ, Spicer AP, Lalani EN, Duhig T, Peat N,
64. Nielsen PA, Mandel U, Therkildsen MH, Clausen Burchell J, et al. Structure and biology of a
H. Differential expression of human high-molecular- carcinoma- associated mucin, MUC1. Am Rev
weight salivary mucin (MG1) and low-molecular- Respir Dis. 1991;144(3 Pt 2):S42–7.
weight salivary mucin (MG2). J Dent Res. 80. Brockhausen I, Yang JM, Burchell J, Whitehouse C,
1996;75(11):1820–6. Taylor-Papadimitriou J. Mechanisms underlying
65. Alliende C, Kwon YJ, Brito M, Molina C, Aguilera aberrant glycosylation of MUC1 mucin in breast
S, Perez P, et al. Reduced sulfation of muc5b is cancer cells. Eur J Biochem. 1995;233(2):607–17.
linked to xerostomia in patients with Sjogren syn- 81. Henson BS, Eisbruch A, D’Hondt E, Ship JA. Two-
drome. Ann Rheum Dis. 2008;67(10):1480–7. year longitudinal study of parotid salivary flow rates
66. Alos L, Lujan B, Castillo M, Nadal A, Carreras M, in head and neck cancer patients receiving unilateral
Caballero M, et al. Expression of membrane-bound neck parotid-sparing radiotherapy treatment. Oral
mucins (MUC1 and MUC4) and secreted mucins Oncol. 1999;35(3):234–41.
(MUC2, MUC5AC, MUC5B, MUC6 and MUC7) in 82. Eisbruch A, Kim HM, Terrell JE, Marsh LH,
mucoepidermoid carcinomas of salivary glands. Am Dawson LA, Ship JA. Xerostomia and its predictors
J Surg Pathol. 2005;29(6):806–13. following parotid-sparing irradiation of head-and-
67. Bahamondes V, Albornoz A, Aguilera S, Alliende C, neck cancer. Int J Radiat Oncol Biol Phys.
Molina C, Castro I, et al. Changes in Rab3D expres- 2001;50(3):695–704.
sion and distribution in the acini of Sjogren’s syn- 83. Dirix P, Nuyts S, Van den Bogaert W. Radiation-
drome patients are associated with loss of cell induced xerostomia in patients with head and neck
polarity and secretory dysfunction. Arthritis Rheum. cancer: a literature review. Cancer. 2006;107(11):
2011;63(10):3126–35. 2525–34.
68. Tucker AS. Salivary gland development. Semin Cell 84. Carpenter GH, Osailan SM, Correia P, Paterson KP,
Dev Biol. 2007;18(2):237–44. Proctor GB. Rat salivary gland ligation causes
69. Jaskoll T, Chen H, Denny PC, Denny PA, Melnick reversible secretory hypofunction. Acta Physiol
M. Mouse submandibular gland mucin: embryo- (Oxf). 2007;189(3):241–9.
specific mRNA and protein species. Mech Dev. 85. Robar JL, Day A, Clancey J, Kelly R, Yewondwossen
1998;74(1–2):179–83. M, Hollenhorst H, et al. Spatial and dosimetric vari-
70. Melnick M, Jaskoll T. Mouse submandibular gland ability of organs at risk in head-and-neck intensity-
morphogenesis: a paradigm for embryonic signal modulated radiotherapy. Int J Radiat Oncol Biol
processing. Crit Rev Oral Biol Med. Phys. 2007;68(4):1121–30.
2000;11(2):199–215. 86. Tamarin A. Submaxillary gland recovery from
71. Melnick M, Chen H, Zhou Y, Jaskoll T. Embryonic obstruction. I. Overall changes and electron micro-
mouse submandibular salivary gland morphogenesis scopic alterations of granular duct cells. J Ultrastruct
and the TNF/TNF-R1 signal transduction pathway. Res. 1971;34(3):276–87.
Anat Rec. 2001;262(3):318–30. 87. Tamarin A. Submaxillary gland recovery from
72. Denny PA, Denny PC. Localization of a mouse sub- obstruction. II. Electron microscopic alterations of
mandibular sialomucin by indirect immunofluores- acinar cells. J Ultrastruct Res. 1971;34(3):288–302.
cence. Histochem J. 1982;14(3):403–8. 88. Osailan SM, Proctor GB, McGurk M, Paterson
73. Denny PC, Ball WD, Redman RS. Salivary glands: a KL. Intraoral duct ligation without inclusion of the
paradigm for diversity of gland development. Crit parasympathetic nerve supply induces rat subman-
Rev Oral Biol Med. 1997;8(1):51–75. dibular gland atrophy. Int J Exp Pathol.
74. Denny PA, Denny PC, Jenkins K. Purification and 2006;87(1):41–8.
biochemical characterization of a mouse submandib- 89. Norberg LE, Abok K, Lundquist PG. Effects of liga-
ular sialomucin. Carbohydr Res. 1980;87(2):265–74. tion and irradiation on the submaxillary glands in
75. Denny PA, Pimprapaiporn W, Kim MS, Denny rats. Acta Otolaryngol. 1988;105(1–2):181–92.
PC. Quantitation and localization of acinar cell-specific 90. Cotroneo E, Proctor GB, Paterson KL, Carpenter
mucin in submandibular glands of mice during postna- GH. Early markers of regeneration following ductal
tal development. Cell Tissue Res. 1988;251(2):381–6. ligation in rat submandibular gland. Cell Tissue Res.
76. Braga VM, Pemberton LF, Duhig T, Gendler 2008;332(2):227–35.
SJ. Spatial and temporal expression of an epithelial 91. Dijkema T, Terhaard CH, Roesink JM, Raaijmakers
mucin, Muc-1, during mouse development. CP, van den Keijbus PA, Brand HS, et al. MUC5B
Development. 1992;115(2):427–37. levels in submandibular gland saliva of patients
68 I. Castro et al.
treated with radiotherapy for head-and-neck cancer: 106. Li Y, Taylor JM, Ten Haken RK, Eisbruch A. The
a pilot study. Radiat Oncol. 2012;7:91. impact of dose on parotid salivary recovery in head
92. Carpenter GH, Khosravani N, Ekstrom J, Osailan and neck cancer patients treated with radiation ther-
SM, Paterson KP, Proctor GB. Altered plasticity of apy. Int J Radiat Oncol Biol Phys. 2007;67(3):660–9.
the parasympathetic innervation in the recovering rat 107. Pramanik R, Osailan SM, Challacombe SJ, Urquhart
submandibular gland following extensive atrophy. D, Proctor GB. Protein and mucin retention on oral
Exp Physiol. 2009;94(2):213–9. mucosal surfaces in dry mouth patients. Eur J Oral
93. Walker NI, Gobe GC. Cell death and cell prolifera- Sci. 2010;118(3):245–53.
tion during atrophy of the rat parotid gland induced 108. Ewert P, Aguilera S, Alliende C, Kwon YJ, Albornoz
by duct obstruction. J Pathol. 1987;153(4):333–44. A, Molina C, et al. Disruption of tight junction struc-
94. Takahashi S, Nakamura S, Suzuki R, Islam N, ture in salivary glands from Sjogren’s syndrome
Domon T, Yamamoto T, et al. Apoptosis and mitosis patients is linked to proinflammatory cytokine expo-
of parenchymal cells in the duct-ligated rat subman- sure. Arthritis Rheum. 2010;62(5):1280–9.
dibular gland. Tissue Cell. 2000;32(6):457–63. 109. Barrera MJ, Sanchez M, Aguilera S, Alliende C,
95. Radfar L, Sirois DA. Structural and functional injury Bahamondes V, Molina C, et al. Aberrant localiza-
in minipig salivary glands following fractionated tion of fusion receptors involved in regulated exocy-
exposure to 70 Gy of ionizing radiation: an animal tosis in salivary glands of Sjogren’s syndrome
model for human radiation-induced salivary gland patients is linked to ectopic mucin secretion.
injury. Oral Surg Oral Med Oral Pathol Oral Radiol J Autoimmun. 2012;39(1–2):83–92.
Endod. 2003;96(3):267–74. 110. Barrera MJ, Aguilera S, Veerman E, Quest AF, Diaz-
96. Silver N, Proctor GB, Arno M, Carpenter Jimenez D, Urzua U, et al. Salivary mucins induce a
GH. Activation of mTOR coincides with autophagy Toll-like receptor 4-mediated pro-inflammatory
during ligation-induced atrophy in the rat subman- response in human submandibular salivary cells: are
dibular gland. Cell Death Dis. 2010;1:e14. mucins involved in Sjogren’s syndrome?
97. Lombaert IM, Brunsting JF, Wierenga PK, Faber H, Rheumatology (Oxford). 2015;54(8):1518–27.
Stokman MA, Kok T, et al. Rescue of salivary gland 111. Ogawa M, Tsuji T. Functional salivary gland regen-
function after stem cell transplantation in irradiated eration as the next generation of organ replacement
glands. PLoS One. 2008;3(4):e2063. regenerative therapy. Odontology. 2015;103(3):
98. Jeong J, Baek H, Kim YJ, Choi Y, Lee H, Lee E, 248–57.
et al. Human salivary gland stem cells ameliorate 112. Martinez JR, Bylund DB, Cassity N. Progressive
hyposalivation of radiation-damaged rat salivary secretory dysfunction in the rat submandibular gland
glands. Exp Mol Med. 2013;45:e58. after excretory duct ligation. Arch Oral Biol.
99. Lim JY, Yi T, Choi JS, Jang YH, Lee S, Kim HJ, 1982;27(6):443–50.
et al. Intraglandular transplantation of bone marrow- 113. Tsuda T, Hamamori Y, Fukumoto Y, Kaibuchi K,
derived clonal mesenchymal stem cells for ameliora- Takai Y. Epidermal growth factor increases c-myc
tion of post-irradiation salivary gland damage. Oral mRNA without eliciting phosphoinositide turnover,
Oncol. 2013;49(2):136–43. protein kinase C activation, or calcium ion mobiliza-
100. Lim JY, Ra JC, Shin IS, Jang YH, An HY, Choi JS, tion in Swiss 3 T3 fibroblasts. J Biochem.
et al. Systemic transplantation of human adipose 1986;100(6):1631–5.
tissue-derived mesenchymal stem cells for the 114. Smaje LH. Spontaneous secretion in the rabbit sub-
regeneration of irradiation-induced salivary gland maxillary gland. In: Thorn NA, Petersen OH, edi-
damage. PLoS One. 2013;8(8):e71167. tors. Secretory mechanisms of exocrine glands.
101. Glucksmann A. The effect of radiation on the pro- Copenhagen: Munksgaard; 1974. p. 608–22.
duction of mucin in salivary glands and the female 115. Harrison JD, Garrett JR. The effects of ductal liga-
genital tract of rodents. Ann N Y Acad Sci. tion on the parenchyma of salivary glands of cat
1963;106:631–5. studied by enzyme histochemical methods.
102. Tamarin A. The leukocytic response in ligated rat Histochem J. 1976;8(1):35–44.
submandibular glands. J Oral Pathol. 1979;8(5): 116. Cummins M, Dardick I, Brown D, Burford-Mason
293–304. A. Obstructive sialadenitis: a rat model.
103. Harrison JD, Fouad HM, Garrett JR. The effects of J Otolaryngol. 1994;23(1):50–6.
ductal obstruction on the acinar cells of the parotid 117. Garrett JR. The proper role of nerves in salivary
of cat. Arch Oral Biol. 2000;45(11):945–9. secretion: a review. J Dent Res. 1987;66(2):387–97.
104. Hoebers FJ, Kartachova M, de Bois J, van den 118. Ramos-Casals M, Font J. Primary Sjogren’s syn-
Brekel MW, van Tinteren H, van Herk M, et al. drome: current and emergent aetiopathogenic
99mTc Hynic-rh-Annexin V scintigraphy for in vivo concepts. Rheumatology (Oxford). 2005;44(11):

imaging of apoptosis in patients with head and neck 1354–67.
cancer treated with chemoradiotherapy. Eur J Nucl 119. Mavragani CP, Moutsopoulos HM. Sjogren syn-
Med Mol Imaging. 2008;35(3):509–18. drome. CMAJ. 2014;186(15):E579–86.
105. Hall EGA. Radiobiology for the radiologist. 7th ed. 120. Anaya JM, Talal N. Sjogren’s syndrome comes of
Philadelphia: Lippincott Williams & Wilkins; 2011. age. Semin Arthritis Rheum. 1999;28(6):355–9.
121. Dawson LJ, Stanbury J, Venn N, Hasdimir B, Rogers 135. Turner RJ. Ion transport related to fluid secretions in
SN, Smith PM. Antimuscarinic antibodies in pri- salivary glands. In: Dobrosielski-Vergona K, editor.
mary Sjogren’s syndrome reversibly inhibit the Biology of the salivary glands. Boca Raton: CRC
mechanism of fluid secretion by human submandib- Press Inc; 1993. p. 105–27.
ular salivary acinar cells. Arthritis Rheum. 136. Saari H, Halinen S, Ganlov K, Sorsa T, Konttinen
2006;54(4):1165–73. YT. Salivary mucous glycoprotein MG1 in Sjogren’s
122. Fox RI, Kang HI, Ando D, Abrams J, Pisa syndrome. Clin Chim Acta. 1997;259(1–2):83–96.
E. Cytokine mRNA expression in salivary gland 137. Castro I, Sepulveda D, Cortes J, Quest AF, Barrera
biopsies of Sjogren’s syndrome. J Immunol. MJ, Bahamondes V, et al. Oral dryness in Sjogren’s
1994;152(11):5532–9. syndrome patients. Not just a question of water.
123. Konttinen YT, Sorsa T, Hukkanen M, Segerberg M, Autoimmun Rev. 2013;12(5):567–74.
Kuhlefelt-Sundstrom M, Malmstrom M, et al. 138. Yamada H, Nakagawa Y, Wakamatsu E, Sumida T,
Topology of innervation of labial salivary glands by Yamachika S, Nomura Y, et al. Efficacy prediction of
protein gene product 9.5 and synaptophysin immu- cevimeline in patients with Sjogren’s syndrome.
noreactive nerves in patients with Sjogren’s syn- Clin Rheumatol. 2007;26(8):1320–7.
drome. J Rheumatol. 1992;19(1):30–7. 139. Castro I, Aguilera S, Brockhausen I, Alliende C, Quest
124. Fox RI, Tornwall J, Michelson P. Current issues in AF, Molina C, et al. Decreased salivary sulphotransfer-
the diagnosis and treatment of Sjogren’s syndrome. ase activity correlated with inflammation and autoim-
Curr Opin Rheumatol. 1999;11(5):364–71. munity parameters in Sjogren’s syndrome patients.
125. Delporte C, Steinfeld S. Distribution and roles of Rheumatology (Oxford). 2012;51(3):482–90.
aquaporins in salivary glands. Biochim Biophys 140. Brockhausen I. Sulphotransferases acting on mucin-
Acta. 2006;1758(8):1061–70. type oligosaccharides. Biochem Soc Trans.
126. Dawson LJ, Fox PC, Smith PM. Sjogrens syndrome- 2003;31(2):318–25.
-the non-apoptotic model of glandular hypofunction. 141. Yang X, Lehotay M, Anastassiades T, Harrison M,
Rheumatology (Oxford). 2006;45(7):792–8. Brockhausen I. The effect of TNF-alpha on glyco-
127. Dawson L, Tobin A, Smith P, Gordon sylation pathways in bovine synoviocytes. Biochem
T. Antimuscarinic antibodies in Sjogren’s syndrome: Cell Biol. 2004;82(5):559–68.
where are we, and where are we going? Arthritis 142. Corfield AP, Myerscough N, Bradfield N, Corfield
Rheum. 2005;52(10):2984–95. Cdo A, Gough M, Clamp JR, et al. Colonic mucins
128. Goicovich E, Molina C, Perez P, Aguilera S, in ulcerative colitis: evidence for loss of sulfation.
Fernandez J, Olea N, et al. Enhanced degradation of Glycoconj J. 1996;13(5):809–22.
proteins of the basal lamina and stroma by matrix 143. Yamori T, Kimura H, Stewart K, Ota DM, Cleary
metalloproteinases from the salivary glands of KR, Irimura T. Differential production of high
Sjogren’s syndrome patients: correlation with molecular weight sulfated glycoproteins in normal
reduced structural integrity of acini and ducts. colonic mucosa, primary colon carcinoma, and
Arthritis Rheum. 2003;48(9):2573–84. metastases. Cancer Res. 1987;47(10):2741–7.
129. Molina C, Alliende C, Aguilera S, Kwon YJ, Leyton 144. Yamori T, Ota DM, Cleary KR, Hoff S, Hager LG,
L, Martinez B, et al. Basal lamina disorganisation of Irimura T. Monoclonal antibody against human
the acini and ducts of labial salivary glands from colonic sulfomucin: immunochemical detection of
patients with Sjogren’s syndrome: association with its binding sites in colonic mucosa, colorectal pri-
mononuclear cell infiltration. Ann Rheum Dis. mary carcinoma, and metastases. Cancer Res.
2006;65(2):178–83. 1989;49(4):887–94.
130. Aumailley M, Smyth N. The role of laminins in 145. Salvini R, Bardoni A, Valli M, Trinchera M. beta1,3-
basement membrane function. J Anat. 1998;193(Pt Galactosyltransferase beta 3Gal-T5 acts on the
1):1–21. GlcNAcbeta 1-->3Galbeta 1-->4GlcNAcbeta 1-->R
131. Giancotti FG, Ruoslahti E. Integrin signaling. sugar chains of carcinoembryonic antigen and other
Science. 1999;285(5430):1028–32. N-linked glycoproteins and is down-regulated in
132. Hoffman MP, Kibbey MC, Letterio JJ, Kleinman colon adenocarcinomas. J Biol Chem. 2001;276(5):
HK. Role of laminin-1 and TGF-beta 3 in acinar dif- 3564–73.
ferentiation of a human submandibular gland cell 146. Seko A, Ohkura T, Kitamura H, Yonezawa S, Sato E,
line (HSG). J Cell Sci. 1996;109(Pt 8):2013–21. Yamashita K. Quantitative differences in
133. da Costa SR, Wu K, Veigh MM, Pidgeon M, Ding C, GlcNAc:beta1-->3 and GlcNAc:beta1-->4 galactos-
Schechter JE, et al. Male NOD mouse external yltransferase activities between human colonic ade-
lacrimal glands exhibit profound changes in the exo- nocarcinomas and normal colonic mucosa. Cancer
cytotic pathway early in postnatal development. Exp Res. 1996;56(15):3468–73.
Eye Res. 2006;82(1):33–45.
147. Seko A, Nagata K, Yonezawa S, Yamashita
134. Castle D. Cell biology of salivary protein secretion. K. Down-regulation of Gal 3-O-sulfotransferase-2
In: Dobrosielski-Vergona K, editor. Biology of the (Gal3ST-2) expression in human colonic non-
salivary glands. Boca Raton: CRC Press Inc; 1993. mucinous adenocarcinoma. Jpn J Cancer Res.
p. 81–104. 2002;93(5):507–15.
70 I. Castro et al.
148. Tzioufas AG, Kapsogeorgou EK, Moutsopoulos Differential splicing may generate multiple protein
HM. Pathogenesis of Sjogren’s syndrome: what we forms. Eur J Biochem. 1990;189(3):463–73.
know and what we should learn. J Autoimmun. 163. Herbert LM, Grosso JF, Dorsey Jr M, Fu T, Keydar
2012;39(1–2):4–8. I, Cejas MA, et al. A unique mucin immunoenhanc-
149. Perez P, Aguilera S, Olea N, Alliende C, Molina C, ing peptide with antitumor properties. Cancer Res.
Brito M, et al. Aberrant localization of ezrin corre- 2004;64(21):8077–84.
lates with salivary acini disorganization in Sjogren’s 164. Ilkovitch D, Handel-Fernandez ME, Herbert LM,
Syndrome. Rheumatology (Oxford). 2010;49(5): Lopez DM. Antitumor effects of Mucin 1/sec
915–23. involves the modulation of urokinase-type plasmino-
150. Barrera MJ, Bahamondes V, Sepulveda D, Quest AF, gen activator and signal transducer and activator of
Castro I, Cortes J, et al. Sjogren’s syndrome and the transcription 1 expression in tumor cells. Cancer
epithelial target: a comprehensive review. Res. 2008;68(7):2427–35.
J Autoimmun. 2013;42:7–18. 165. Grosso JF, Herbert LM, Owen JL, Lopez DM. MUC1/
151. Frazao JB, Errante PR, Condino-Neto A. Toll-like sec-expressing tumors are rejected in vivo by a T
receptors’ pathway disturbances are associated with cell-dependent mechanism and secrete high levels of
increased susceptibility to infections in humans. Arch CCL2. J Immunol. 2004;173(3):1721–30.
Immunol Ther Exp (Warsz). 2013;61(6):427–43. 166. Baruch A, Hartmann M, Yoeli M, Adereth Y,
152. Tarang S, Kumar S, Batra SK. Mucins and toll-like Greenstein S, Stadler Y, et al. The breast cancer-
receptors: kith and kin in infection and cancer. associated MUC1 gene generates both a receptor
Cancer Lett. 2012;321(2):110–9. and its cognate binding protein. Cancer Res.
153. Gallo PM, Gallucci S. The dendritic cell response to 1999;59(7):1552–61.
classic, emerging, and homeostatic danger signals. 167. Hartman M, Baruch A, Ron I, Aderet Y, Yoeli M,
Implications for autoimmunity. Front Immunol. Sagi-Assif O, et al. MUC1 isoform specific mono-
2013;4:138. clonal antibody 6E6/2 detects preferential expres-
154. Brennan TV, Lin L, Huang X, Cardona DM, Li Z, sion of the novel MUC1/Y protein in breast and
Dredge K, et al. Heparan sulfate, an endogenous TLR4 ovarian cancer. Int J Cancer. 1999;82(2):256–67.
agonist, promotes acute GVHD after allogeneic stem 168. Zrihan-Licht S, Vos HL, Baruch A, Elroy-Stein O,
cell transplantation. Blood. 2012;120(14):2899–908. Sagiv D, Keydar I, et al. Characterization and molec-
155. Termeer C, Benedix F, Sleeman J, Fieber C, Voith U, ular cloning of a novel MUC1 protein, devoid of tan-
Ahrens T, et al. Oligosaccharides of hyaluronan acti- dem repeats, expressed in human breast cancer
vate dendritic cells via toll-like receptor 4. J Exp tissue. Eur J Biochem. 1994;224(2):787–95.
Med. 2002;195(1):99–111. 169. Cascio S, Zhang L, Finn OJ. MUC1 protein expres-
156. Muroi M, Tanamoto K. The polysaccharide portion sion in tumor cells regulates transcription of proin-
plays an indispensable role in salmonella flammatory cytokines by forming a complex with
lipopolysaccharide-induced activation of NF-kappaB nuclear factor-kappaB p65 and binding to cytokine
through human toll-like receptor 4. Infect Immun. promoters: importance of extracellular domain.
2002;70(11):6043–7. J Biol Chem. 2011;286(49):42248–56.
157. Lu YC, Yeh WC, Ohashi PS. LPS/TLR4 signal trans- 170. Yamada N, Kitamoto S, Yokoyama S, Hamada T,
duction pathway. Cytokine. 2008;42(2):145–51. Goto M, Tsutsumida H, et al. Epigenetic regulation
158. Jenkins KA, Mansell A. TIR-containing adap- of mucin genes in human cancers. Clin Epigenetics.
tors in Toll-like receptor signalling. Cytokine. 2011;2(2):85–96.
2010;49(3):237–44. 171. Ilkovitch D, Carrio R, Lopez DM. Mechanisms of
159. Kawakami A, Nakashima K, Tamai M, Nakamura antitumor and immune-enhancing activities of
H, Iwanaga N, Fujikawa K, et al. Toll-like receptor MUC1/sec, a secreted form of mucin-1. Immunol
in salivary glands from patients with Sjogren’s syn- Res. 2013;57(1–3):70–80.
drome: functional analysis by human salivary gland 172. Choi S, Park YS, Koga T, Treloar A, Kim KC. TNF-
cell line. J Rheumatol. 2007;34(5):1019–26. alpha is a key regulator of MUC1, an anti-
160. Spachidou MP, Bourazopoulou E, Maratheftis CI, inflammatory molecule, during airway Pseudomonas
Kapsogeorgou EK, Moutsopoulos HM, Tzioufas aeruginosa infection. Am J Respir Cell Mol Biol.
AG, et al. Expression of functional Toll-like recep- 2011;44(2):255–60.
tors by salivary gland epithelial cells: increased 173. Sheng YH, Triyana S, Wang R, Das I, Gerloff K,
mRNA expression in cells derived from patients Florin TH, et al. MUC1 and MUC13 differentially
with primary Sjogren’s syndrome. Clin Exp regulate epithelial inflammation in response to
Immunol. 2007;147(3):497–503. inflammatory and infectious stimuli. Mucosal
161. Horng T, Barton GM, Medzhitov R. TIRAP: an Immunol. 2013;6(3):557–68.
adapter molecule in the Toll signaling pathway. Nat 174. Yen JH, Xu S, Park YS, Ganea D, Kim KC. Higher
Immunol. 2001;2(9):835–41. susceptibility to experimental autoimmune encepha-
162. Wreschner DH, Hareuveni M, Tsarfaty I, lomyelitis in Muc1-deficient mice is associated with
Smorodinsky N, Horev J, Zaretsky J, et al. Human increased Th1/Th17 responses. Brain Behav Immun.
epithelial tumor antigen cDNA sequences. 2013;29:70–81.
175. Kufe DW. Mucins in cancer: function, prognosis and 182. Gum Jr JR, Hicks JW, Toribara NW, Siddiki B, Kim
therapy. Nat Rev Cancer. 2009;9(12):874–85. YS. Molecular cloning of human intestinal mucin
176. Yonezawa S, Goto M, Yamada N, Higashi M, (MUC2) cDNA. Identification of the amino terminus
Nomoto M. Expression profiles of MUC1, MUC2, and overall sequence similarity to prepro-von
and MUC4 mucins in human neoplasms and their Willebrand factor. J Biol Chem. 1994;269(4):
relationship with biological behavior. Proteomics. 2440–6.
2008;8(16):3329–41. 183. Yamada K, Tanaka T, Mori M, Tsubura A, Morii S,
177. Nath S, Mukherjee P. MUC1: a multifaceted oncop- Tsubone M, et al. Immunohistochemical expression
rotein with a key role in cancer progression. Trends of MAM-3 and MAM-6 antigens in salivary gland
Mol Med. 2014;20(6):332–42. tumours. Virchows Arch A Pathol Anat Histopathol.
178. Soares AB, Demasi AP, Altemani A, de Araujo 1989;415(6):509–21.
VC. Increased mucin 1 expression in recurrence and 184. Siyi L, Shengwen L, Min R, Wenjun Y, Lizheng W,
malignant transformation of salivary gland pleomor- Chenping Z. Increased expression of MUC-1 has
phic adenoma. Histopathology. 2011;58(3):377–82. close relation with patient survivor in high-grade
179. Yakirevich E, Sabo E, Klorin G, Alos L, Cardesa A, salivary gland mucoepidermoid carcinoma. J Oral
Ellis GL, et al. Primary mucin-producing tumours of Pathol Med. 2014;43(8):579–84.
the salivary glands: a clinicopathological and mor- 185. Handra-Luca A, Lamas G, Bertrand JC, Fouret
phometric study. Histopathology. 2010;57(3): P. MUC1, MUC2, MUC4, and MUC5AC expression
395–409. in salivary gland mucoepidermoid carcinoma: diag-
180. Brieger J, Duesterhoeft A, Brochhausen C, Gosepath nostic and prognostic implications. Am J Surg
J, Kirkpatrick CJ, Mann WJ. Recurrence of pleomor- Pathol. 2005;29(7):881–9.
phic adenoma of the parotid gland--predictive value 186. Rakha EA, Boyce RW, Abd El-Rehim D, Kurien T,
of cadherin-11 and fascin. APMIS. 2008;116(12): Green AR, Paish EC, et al. Expression of mucins
1050–7. (MUC1, MUC2, MUC3, MUC4, MUC5AC and
181. Ferreira JC, Morais MO, Elias MR, Batista AC, MUC6) and their prognostic significance in human
Leles CR, Mendonca EF. Pleomorphic adenoma of breast cancer. Mod Pathol. 2005;18(10):1295–304.
oral minor salivary glands: an investigation of its 187. Velcich A, Yang W, Heyer J, Fragale A, Nicholas C,
neoplastic potential based on apoptosis, mucosecre- Viani S, et al. Colorectal cancer in mice genetically
tory activity and cellular proliferation. Arch Oral deficient in the mucin Muc2. Science.
Biol. 2014;59(6):578–85. 2002;295(5560):1726–9.
Part II
Glandular Damage
and Cell Replacement Therapy
Histologic Changes in the Salivary
Glands Following Radiation 4
Therapy
Robert S. Redman
Abstract
Therapeutic radiation for cancer of the head and neck damages salivary
glands that are situated between the radiation source and the target tumor
and its metastases. With moderate to high radiation exposure, salivary
glands are devastated and regeneration is limited. The resulting severe
reduction in saliva has detrimental effects on the teeth and oral mucosa.
The purpose of this review is to describe some of the salient histologic
features of salivary gland structures and cells, how these are functionally
related to salivary production, and thus how radiation-induced loss and
functional impairment of each type of structure may contribute to reduced
quantity and quality of saliva.
4.1 Introduction 83, 100, 101]. Without diligent professional

and home care, rapidly progressive dental car-
Ionizing radiation, often augmented with ies can ensue, leading to a cascade of infec-
administration of chemotherapeutic agents, is tion and extractions with significant risk of
a staple of management of cancers of the head osteoradionecrosis. In this chapter, I describe
and neck region that are not amenable to suc- the histology of radiation-induced damage to
cessful treatment by surgery alone. The jaw- the salivary glands in the context of how this
bones and salivary glands often suffer moderate affects salivary function. I begin with a review
to severe damage, the latter resulting in dra- of the functional morphology of normal,
matically decreased salivary function [23, 73, mature salivary glands, i.e., their anatomical,
light microscopic, and ultrastructural features,
as these relate to the principal functions of the
R.S. Redman, DDS
Chief, Oral Pathology Research Laboratory,
several parenchymal (epithelial) cell types.
Department of Veterans Affairs Medical Center, This is intended to facilitate understanding
Washington, DC 20422; Adjunct Professor, how radiation-induced physical loss or func-
University of Maryland School of Dentistry, tional impairment of each cell type affects the
Baltimore, MD; Adjunct Scientist, National Institute
of Dental and Craniofacial Research, National
quantity and quality of saliva the gland can
Institutes of Health, Bethesda, MD, USA produce. The review utilizes human and rodent
e-mail: [email protected] glands as models.

76 R.S. Redman
4.2 Salivary Gland 4.3.1 I llustrations of Normal

Nomenclature and Irradiated Glands
Salivary glands are classified by size (major and Light microscopic descriptions and illustrations
minor) and secretory product (serous, watery and are of formaldehyde-fixed, paraffin-embedded,
slightly slippery; mucous, thick, viscous, and mature rat (Figs. 4.1 and 4.5) and human (Figs.
very slippery; and mixed, having both serous and 4.2, 4.6 and 4.7) salivary glands. For transmis-
mucous secretory units). They are named by sion electron microscopy (TEM, Figs. 4.3 and
location. The major glands are the parotid (in 4.4), rat salivary glands were fixed in glutaralde-
front of the ear), sublingual (under the tongue), hyde, post-fixed in 0S04, embedded in epoxy
and submandibular (under the mandible). In resin, sectioned at ca. 70 nm, and stained with
humans, the sublingual and submandibular uranyl acetate and lead citrate. Figure 4.1e is a
glands are separate; in rodents, they are enclosed photomicrograph of a tissue fixed and embed-
in the same capsule. Human minor salivary ded for TEM but sectioned at 1 μm, mounted on
glands are located in groups in the buccal and a glass slide, and stained with methylene blue
labial mucosae, the soft and posterior hard palate and azure II. All illustrations are of material from
including the tonsillar pillars, the posterior dorsal previous studies that had been reviewed by the
and anterior ventral surfaces of the tongue, and appropriate institutional review bodies.
the anterior floor of the mouth (minor sublingual
glands). A few also are located subjacent to the
incisive papilla. All are either mucous or mixed, 4.3.2 Acini
mostly mucous glands except for the serous lin-
gual glands of von Ebner, which secrete into the The acini provide most of the proteins and fluid
foliate furrows and the troughs around the vallateof saliva which moisten and initiate digestion
papillae. The distribution of minor salivary (enzymes), lubricate the oral mucosa, teeth, and
glands in rats and mice is similar to that of food (mucins), maintain the minerals of the teeth
humans except that there is none in the labial (statherin), and modulate the oral flora (peroxi-
mucosa, hard palate, or anterior ventral surface of
dase, histatins) [reviewed by Izutsu [33], Tabak
the tongue. Drawings showing the location and [86], and Redman [73]]. Secretory proteins are
shapes of human major glands can be found in glycosylated, and the amount and type of sugars
standard textbooks of human anatomy, e.g., attached to the core protein determine the nature
Gray’s Anatomy [103]. I have published artists’ of the secretory product. In serous acini, the gly-
drawings showing the distribution of human pal- coproteins are lightly glycosylated, carry a pre-
atal glands [70] and all of the minor salivary dominantly neutral charge, and have a molecular
glands of the rat [69, 74]. weight of less than 100,000 KD. Mucous acini
produce mucins, high molecular weight (1–10
million KD) glycoproteins that are heavily glyco-
4.3 Histology and Ultrastructure sylated with both neutral and acidic sugars. The
seromucous acini of rat submandibular glands
The terminology used here is guided by that of produce a mucin of 100,000 KD. The greater
Young and van Lennep [106]. The three- amount and negative charge of the sugars of
dimensional structure of salivary glands is like mucins attract more water, which is a major rea-
that of a bunch of grapes [47], with the grapes son why mucous secretions are much more vis-
being the secretory endpieces (acini) and the cous and slippery than serous secretions.
“stems” (the pedicels, rachis, branches, and By light microscopy of sections stained with
peduncle) being the ducts in order of increasing hematoxylin and eosin (H&E), serous acini take
size (which roughly correlates with distance from both dyes moderately, resulting in violet to
the acini), terminating in the main duct. blue-violet staining (Fig. 4.2a, b). Mucous acini
4 Histologic Changes in the Salivary Glands Following Radiation Therapy 77
a b
c d
e f
Fig. 4.1 Photomicrographs of mature rat salivary glands. epoxy resin section, methylene blue-azure II stain. The
(a) SL, H&E. The mucous acini are capped with small contents of the secretory granules of the seromucous acini
serous demilunes (arrow). The pink (mainly eosin) tall are empty looking, similar to mucous acini. Those of the
columnar cells of the striated ducts have prominent radial granular convoluted tubules (arrow) stain blue with vary-
striations and a single row of centrally placed nuclei. In ing density, typical of serous granules. (f) SL and SM,
the first order excretory duct, the cytoplasm of the short immunohistochemical localization of Muc 19, hematoxy-
columnar cells also is pink, and there are occasional basal lin counterstain. Only the mucous secretory granules of
cells. (b) SL, AB-H. The mucous acini are heavily stained, the sublingual gland reacted with the antibody (brown
and the striated ducts and serous demilunes are unstained precipitate). Magnification bars: a, b, d = 50 μm; c,
by AB. (c) P, PAS-H. The secretory granules of the serous 200 μm; e, 30 μm; f, 100 μm. Abbreviations for all
acini and first-order intercalated ducts (arrow) are PAS + Figures. P, SL, SM, parotid, sublingual, and submandibu-
(light and moderate magenta, respectively). (d) SM, PAS- lar glands; a acini; id, sd, ed, intercalated, striated, and
H. The secretory granules of the seromucous acini (a) and excretory ducts; m, mitochondria; n, nerve; v, blood ves-
serous granular convoluted tubules (arrow) are moderate sel. Stains: AB, alcian blue; E, eosin; H, hematoxylin;
and light magenta, respectively. (e) SM, semi-thin (1 μm) PAS, periodic acid-Schiff.
stain very lightly, giving a pale, cloudy appearance 4.1b and 4.2d) [57] and mucicarmine [43] (Fig.
(Fig. 4.1a). Stains that are commonly employed to 4.6), for acidic glycoproteins, and periodic acid-
distinguish between the glycoproteins of serous Schiff (PAS) for mainly neutral glycoproteins
and mucous acini include alcian blue (AB, Figs. (Figs. 4.1c, d and 4.2d). Serous acini stain a light
78 R.S. Redman
a b
c d
e f
Fig. 4.2 Photomicrographs of human salivary glands. (a) smooth muscle actin (SMA). Myofibrils of myoepithelial
SM, (b) P, both H&E. Serous acini and demilunes are vio- cells (brown precipitate, arrow) invest the acini but not the
let, and mucous acini (arrow) are very lightly stained. striated ducts. Small circles center right and bottom mark
Empty-looking spaces in (b) are adipose cells. (c) SM, myofibrils of smooth muscle cells around the small arter-
AB–E. Mucous acini stain strongly with AB; serous acini, ies, a built-in positive control. (f) Buccal (minor) gland,
and demilunes not at all with AB. (d) Palatal (minor) H&E. This shows the transition from pseudostratified
gland, PAS-H. The mucous acini and secretory product in columnar to stratified squamous epithelium as the main
the lumen of an excretory duct stained a rich magenta color duct approaches the orifice (arrow) to the oral mucosa.
with PAS. (e) P, immunohistochemical localization of Magnification bars: a, f, 150 μm; b, 100 μm; c–e, 50 μm.
magenta hue with PAS and not at all with AB and Acini have a number of enzymes and struc-
mucicarmine. Mucous acini stain an intense tures that work together to move fluid from
magenta with PAS, blue with AB, and pink with the interstitial tissue into the lumen. These
mucicarmine. Secretory enzymes, e.g., α-amylase include Na+/K+−adenosine triphosphatase,
[104] (Fig. 4.6), carbonic anhydrase [39], and sali- Na+/H+ exchangers 1, 2, and 3, and anion
vary peroxidase [77], histatins [3], and specific exchanger 2, and Na+/K+/2Cl- cotransporter
mucins, e.g. [15] (Fig. 4.1f), are a few of the many [31], aquaporins [38], and numerous mito-
secretory products of acini that can be identified chondria (Fig. 4.3b) commensurate with the
by immunohistochemistry (IHC). energy required.
a b
c d
Fig. 4.3 Transmission electron micrographs (TEM) of late substructure. Myoepithelial cells with processes
rat salivary glands. (a, b) P acini. In (a), electron-dense (double arrow) filled with bands of myofilaments
secretory granules occupy the apical three fourths and bunched in dense bodies (dark spots) are attached to the
extensive rough endoplasmic reticulum the basal fourth acini. One has a typically fusiform (cigar-shaped)
of the acinar cells. Arrow marks an intercellular exten- nucleus. (d) First-order (juxta-acinar) intercalated duct
sion of the lumen, or secretory canaliculus. In (b), rough (id). The secretory granules are also electron dense but
endoplasmic reticulum (r) is rimmed with ribosomes smaller and more homogeneous than those of the acini
(dots on membrane walls), and the lumina are engorged (a). A myoepithelial cell (arrow) lies parallel to the long
with nascent secretory proteins. (c) SL (Reproduced axis of the duct. The lumen of a small blood vessel (v)
from Redman and Ball [75]). The electron-lucent secre- harbors an erythrocyte. Reference scale bar: a,
tory granules of the mucous acini have a finely particu- d = 4.0 μm, b = 0.7 μm, c = 1.2 μm
By transmission electron microscopy (TEM), intercellular extensions of the lumen called secre-
acinar cells are factories for the production, tory canaliculi [106] (Fig. 4.3a). Acini generally
storage, and secretion of secretory proteins. To this have a dual sympathetic innervation, with cholin-
end, they have long segments of rough endoplas- ergic nerves stimulating mostly water secretion
mic reticulum for protein synthesis, a prominent and adrenergic nerves stimulating mostly pro-
Golgi apparatus for glycosylation of the nascent teins [24, 29, 107].
proteins, and secretory granules (membrane-
bounded sacs) for storage of the secretory glyco-
proteins between meals (Fig. 4.3). Serous granules 4.3.3 Intercalated Ducts
may be uniformly electron dense or have patterns
of varying density. Mucous granules are electron- Intercalated ducts (Fig. 4.3d) connect acini with
lucent with a flocculent substructure of specks or larger ducts in rat and human parotid and sub-
filaments that may be uniformly spaced (mono- mandibular glands.
phasic, Fig. 4.3c) or have discrete areas of more or By light microscopy, they are composed of
less closely packed substructure (biphasic), e.g. cuboidal cells with fusiform (cigar-shaped)
[74]. A feature of acini that is not seen in ducts is nuclei and, in cross sections, have a smaller
80 R.S. Redman
diameter than acini and other ducts. The primary [33]. These functions are facilitated by the large
(juxta-acinar) segment has small serous secretory cell surface areas provided by the basolateral infold-
granules that stain moderately to strongly with ings (Fig. 4.4a, b) and, during active secretion, apical
PAS (Fig. 4.1c) but not with any stain for mucins blebbing [54] (Fig. 4.4c) and activity of pinocytotic
in the major human and rat salivary glands. By vesicles. Most of the arteries in salivary glands run
TEM, the granules are uniformly electron dense parallel with the ducts countercurrent to s alivary flow
in rat parotid glands (Fig. 4.3d). Alpha-amylase, (distally) to the acini, and the veins retrace this route
deoxyribonuclease I, and salivary peroxidase [106]. The presence of a rich vasculature in the sur-
have been immunohistochemically localized to rounding stroma facilitates the clearance of resorbed
the acinar, but not intercalated duct, secretory ions from interstitial tissue along the base of the stri-
granules of rat parotid glands [76, 104]. In the rat ated ducts (Fig. 4.4a, b).
submandibular gland, by IHC and several histo- Striated duct cells have small, apically located
logic stains, the granules are identical to the secretory vesicles in rat parotid [30] and human
Types I and III granules in the transitional acini [92] submandibular salivary glands. In rodents
of the neonatal rat [55]. The secondary segment such as hamster, mouse, and rat, a secretory duct
has no secretory granules and a paucity of organ- is interposed between the intercalated and stri-
elles, thus appearing to be relatively undifferentiated ducts (Fig. 4.1e). The secretory granules of
ated. For this and other reasons, the intercalated these granular convoluted tubules store and
duct (and the basal cells of the main excretory secrete proteases and bioactive peptides such as
duct) long have been regarded as sources of pro- epidermal, fibroblast, insulin-like, and nerve
genitor cells in mature salivary glands [18]. There growth factors (EGF, FGF, I-LGF, NGF) [26, 62].
is favorable evidence for this in mouse [41] and These tubules show a sexual dimorphism, being
rat [48] submandibular glands, but not in rat androgen-dependent and thus proportionately
parotid gland [72]. much more extensive in males.
Intercalated ducts also may contribute fluid to In human parotid and submandibular glands,
saliva, though micropuncture studies did not sep- the secretory granules of acini and vesicles in
arate the acinar and intercalated duct sources striated ducts also are immunoreactive to EGF
[51]. Some support for a fluid transport role lies [14, 37]. The vesicles are located in both the api-
in the histochemical localization of aquaporin cal and basal cytoplasm of the striated ducts, sug-
5 in both acini and intercalated ducts in mouse gesting that they secrete not only into the lumen
submandibular gland [4, 38]. but also into the circulation. Most human salivary
EGF is of parotid gland origin and does not differ
by gender [12, 97].
4.3.4 Striated Ducts In both humans and rodents, EGF has been
shown to be secreted in response to oral, esopha-
By light microscopy of H&E-stained sections, the geal, and gastroduodenal ulceration or injury and
smaller striated ducts are composed of a single layer of to aid in healing by stimulating cellular prolifera-
tall columnar cells with centrally located, round nuclei tion [35, 36, 44, 49, 62, 63, 65, 95].
and pink cytoplasm (Fig. 4.1c). By TEM, there are cells
with less and more e lectron-dense cytoplasm called
light and dark cells. The cells have deeply infolded 4.3.5 Excretory Ducts
and interdigitated basolateral cell membranes accom-
panied by numerous large, long mitochondria [89, Excretory ducts follow striated ducts and connect
92] (Fig. 4.4a, b). By light microscopy these appear the glandular lobules and lobes to the oral cavity
as radial lines or striations, for which these ducts are via the main excretory duct.
named. A principal function of the striated ducts is ion By light microscopy, excretory ducts have
exchange with the luminal fluid, Na+ and Cl- being pink cytoplasm in H&E-stained sections (Fig.
resorbed and H CO3- being luminally transported 4.1a), and the main excretory duct transitions
a b
c d
Fig. 4.4 TEM of large ducts in rat parotid glands. (a) seen in the supporting stroma. (c) This striated duct exhibits
Striated ducts (sd). Large, oblong mitochondria run parallel apical blebbing (b), or transient, focal expansions of the
to the deeply infolded and interdigitated basal cell mem- cytoplasm into the lumen, and pinocytotic vesicles (arrow).
branes (arrows). Together these make up the radial striations (d) Large excretory duct (ed). Light, dark, and basal cells are
seen by light microscopy. Large, thin-walled veins (v) are illustrated. Reference scale bar: a, d = 5.9 μm; b, c = 1.6 μm
from pseudostratified columnar to stratified squa- but not parotid, gland resorbs Na+ and K+ from the
mous epithelium as it merges with the oral epi- luminal fluid [51, 105]. The basal cells are dark
thelium (Fig. 4.2f). Some of the excretory ducts (relatively electron dense) because of many cyto-
have short columnar cells and occasional basal keratin filaments attached to multiple hemidesmo-
cells (Fig. 4.1a). Tandler [92] regards these as somes in the basal plasmalemma [79, 93].
part of the striated duct population, as some of The concentrations of the ions in saliva are
the columnar cells have infolded basal cell mem- dependent on the flow rate [81]. In resting
branes and large mitochondria. (unstimulated) saliva, the concentrations of
By TEM, the main and other large excretory Na+, Cl-, and HCO3- are low and increase with
ducts of rodents and human major salivary glands increasing flow rate, as the resorption of Na+
are composed of light, dark, and tuft columnar cells, and Cl- by striated and excretory ducts is unable
basal cells (Fig. 4.4d), and scattered mucous (“gob- to keep pace with the increased volume, and
let”) and ciliated cells [82, 92, 93, 96, 106]. The tuft HCO3- is increasingly transported into the
cells (not illustrated) have groups of long microvilli lumen in exchange for the other ions. On the
projecting into the lumen and secretory granules other hand, the concentration of salivary Ca++
and vesicles, suggesting reception and secretory drops with increasing flow rate. A mechanism
functions [79]. Though the columnar cells have for perception and resorption of Ca++ to main-
many mitochondria, basal membrane infoldings are tain Ca++ within physiological limits in the
not prominent. Micropuncture studies have shown lumina of the larger ducts of salivary glands has
that the main excretory duct of rat submandibular, recently been described [6].
82 R.S. Redman
4.3.6 Myoepithelium (SMA, [27], Fig. 4.2e); only the basal cells of
large ducts, with CK-13 and CK-16 [16]; and
In human and rat major salivary glands, there are both myoepithelium and duct basal cells, with
cells shaped like starfish which are situated CK-14 [16]. All duct cells, but not acini or myo-
between the basal lamina and the acini and inter- epithelium, label with CK-5 [9] and CK-19 [25]
calated but not striated or excretory ducts [59, 71, in human salivary glands and CK-5 in rat sali-
88, 91]. In mouse and rat parotid glands, these vary glands [41]. Antiepithelial membrane anti-
cells invest only the intercalated ducts. Their progen (EMA) labels the luminal membranes of
cesses form a basket-like net around the acini acini and intercalated and striated ducts [94].
(Fig. 4.2e) and are arranged spirally along the
long axis of the intercalated ducts (Fig. 4.3d).
The nuclei of myoepithelial cells investing acini 4.4 alivary Gland Damage
S
are disk shaped [59] but often are sectioned such from Ionizing Radiation
that they appear as fusiform, or cigar-shaped
(Fig. 4.3c), and in myoepithelial cells investing 4.4.1 Principles
ducts, all are fusiform (Fig. 4.3d). Myoepithelial
cells are attached to acini and ducts by desmo- Ionizing radiation (IR) gives up energy when it
somes. The cell processes have bundles of fila- is slowed, redirected, or stopped by molecules
ments resembling the myofibrils of smooth in tissues. The released energy gives rise to
muscle, as they are interspersed with dense bod- free radicals and, if the initial energy is high
ies and anchored to the basement membrane with enough, secondary radiation. The energy of the
attachment plaques [75]. Though these myofi- radiation determines the depth in tissues where
brils can be labeled via IHC for smooth muscle most of it interacts. Therapeutic radiation for
actin (Fig. 4.2e), the intermediate filaments are cancers of the head and neck is mainly elec-
made of cytokeratins [56]. Thus, they have fea- tromagnetic beams such as X-ray and gamma
tures of both epithelial and smooth muscle cells, ray. To achieve interaction with the target tis-
which is why they are named myoepithelial cells. sues and to minimize damage to normal tissues,
When the gland is stimulated to secrete, the myo- the external radiation is directed through por-
epithelial cell processes contract rhythmically in tals from several different directions. However,
tandem with the terminal web to help expel the the distribution of useful portals is limited by
secreted fluid out of the acini and through the the density of minerals such as calcium in the
ducts [71, 90]. jawbones, which impede delivery to the target
tissues and induce secondary radiation in the
process. The result is that the major salivary
4.3.7 E
xamples of Other Salivary glands often receive moderate to heavy radia-
Gland Functional Proteins tion. Computerized radiation dose programs
guided by computed tomography can more pre-
There are many other structural and secretory cisely fit the radiation doses to the site(s) of the
proteins in salivary glands that can be identified cancer and its metastases while minimizing the
by IHC. Though a thorough list is beyond the IR affecting the salivary glands, oral mucosa,
scope of this review, a few examples follow. and other structures not likely to harbor tumor
Acini, myoepithelium, ducts, and cell types cells [32, 46, 58]. Another approach is the use of
within ducts have characteristic cytoskeletal and pharmaceuticals such as amifostine (WR2721,
membrane proteins by which they can be distin- Ethyol®, [102]) which spares normal but not
guished via IHC. For example, only myoepithe- malignant tissues from IR damage.
lium (and smooth muscle in blood vessels) In general, beta particles such as β- (electrons)
labels with antibodies to smooth muscle actin or β+ (positrons) penetrate much shorter
a b
Fig. 4.5 Photomicrographs of submandibular glands of 10 Gy irradiation. Proportionately more stroma and
male rats at age 10 months. (a) Sham-irradiated gland. fewer acini and granular convoluted tubules are present,
Acini (blue secretory granules) are tightly spaced in and the granular convoluted tubules are smaller, than in
large bunches, granular convoluted tubules (g) are (a). Alcian blue and hematoxylin (AB-H). Scale bar =
numerous and large, and the stroma is scanty except in 60 μm (Reproduced from photomicrographs [73] of
septa around the larger ducts. (b) Gland 8 months after slides used in a previous study [64])
d istances into tissues than electromagnetic radia- 4.4.2 R

adiation Damage Is Uneven
tion of equal energy because they have both mass Among and Within Salivary
and charge. To achieve therapeutic radiation, beta Glands
particle-emitting radionuclides may be delivered
via the circulation to tissues where they are pref- In both human and rodent salivary glands, serous
erentially absorbed and concentrated. The radio- acini are the structures that are the most vulnerable
nuclide used most in the head and neck is 131I, to ionizing radiation, many undergoing cell death
which is concentrated more than 1000-fold in the and functional impairment after even moderate
thyroid gland and therefore is used to destroy exposure [2, 10, 19, 60, 68, 84]. Regeneration of
papillary and follicular thyroid cancers. acini is proportional to the dose, being good after
Unfortunately the salivary glands, especially the the minor damage caused by mild (6–20 Gy) doses,
striated ducts, also concentrate 131I much more but after more than 50 Gy, the serous acini of the
than do other tissues [50]. The radiation is parotid and submandibular glands almost disap-
continuous until the radioactive iodine has been pear, and little or no regeneration takes place. Less
removed from the gland by secretion and circula- extensive acinar loss and atrophy occur among the
tion. An inflammatory response quickly ensues, mucous acini of the sublingual and submandibular
and the swelling around the ducts and formation glands, and some regeneration may occur.
of a ductal plug result in salivary retention, pro- The histological effects of mild to moderate
longing the radiation exposure for several days. (10 Gy) IR on a rat submandibular gland are
In many patients, the full extent of damage is not shown in Fig. 4.5. The stromal area is proportion-
reached until a year or more later. Substantial ately increased and the acinar and granular con-
long-term damage to both acini and ducts occurs, voluted tubules decreased compared with
as evidenced by reduced salivary secretion and sham-irradiated glands.
elevated salivary Na+ and Cl− [45]. The effects of heavy (70.63 Gy) IR on a human
In this review, all radiation doses are in gray submandibular gland are illustrated in Figs. 4.6 and
(Gy), which is the international standard unit. It 4.7. Compared with a normal gland, the stromal
is equivalent to 100 rads, i.e., if the dose in a ref- area is greatly increased at the expense of the serous
erenced publication was 4000 rads, it will be acini and demilunes, which have all but disappeared,
listed as 40 Gy here. and the mucous acini are reduced in size. Much of
84 R.S. Redman
a b
c d
e f
g h
Fig. 4.6 (a–d) Normal gland from a 52-year-old male. acini are not identifiable as such, mucous acini are
(e–h) Gland removed from a 62-year-old male 4 weeks diminished in number, size, and staining intensity with
after being subjected to 70.34 Gy radiation in divided mucicarmine, the columnar cells of many of the striated
doses concurrent with 8 weeks of chemotherapy as part and excretory ducts are reduced almost to cuboidal
of a treatment regimen for cancer of the oropharynx. In dimensions, and the adipose cells (large, empty-looking
the normal gland, serous acini contain ample secretory cells) and stroma occupy proportionately more space.
granules that stain darkly with hematoxylin, gray with Large nerves appear to be intact (e). Labels: a serous
mucicarmine, blank with anti-smooth muscle actin acini, ed excretory ducts, m mucous acini, n large nerves,
(SMA), and richly (brown) with anti-amylase. Secretion sd striated ducts, v blood vessels. H&E (a, e), mucicar-
product in mucous acini stains pink with mucicarmine, mine (b, f), and immunohistochemical localization of
and myoepithelial cells (arrows) and smooth muscle in α-amylase (c, g) and SMA (d, h) counterstained with
blood vessels stain brown with anti-SMA. In the irradi- hematoxylin. Scale bar for a, e = 100 μm; for b–d and
ated gland, myoepithelial cells are plentiful, but serous f–h = 60 μm (Reproduced from Redman [73])
a b
c d
e f
Fig. 4.7 Same glands as in Fig. 4.6. Masson’s trichrome localization of c-Kit, a marker for stem cells. (c, e) Normal
stain. (a) Normal gland. Mature collagen fibers (dark gland. Labeled cells (brown cytoplasm) are scattered in
blue) are thin and delicate in the stroma around the acini the excretory ducts (ed), acini (a), and intercalated ducts
and thick and dense in the interlobular septa of (a). (b) and widely scattered in the stroma. Lymphocytes around
Irradiated gland. The collagen fibers are thick and dense the blood vessel (v) are unreactive. (d, f) Irradiated gland.
in the stroma not only in the septa but also among the Positive cells are not found in the large ducts (ed) and
septa and in the lobules once occupied by acini, indicating duct-like structures (remnants of acini and intercalated
extensive fibrosis has occurred. (c–f) Immunohistochemical ducts, arrow) but are common in the stroma
the stroma is occupied by fibrosis (Fig. 4.7b) and tered cells in the stroma, larger ducts, and a few
adipose cells (Fig. 4.6f). In contrast to the acini, acini in the normal gland (Fig. 4.7c, e). No labeled
many of the myoepithelial cells have survived and parenchymal cells were seen in the irradiated gland
invest small duct-like structures (Fig. 4.6h), an (Fig. 4.7d), but labeled cells were common in the
arrangement very similar to that observed in duct- stroma (Fig. 4.7f). All of the tumor cells in the posi-
ligated rabbit ([52] and rat [87] salivary glands. IHC tive control (an adenocarcinoma of the colon) were
for c-Kit, a marker for stem cells [41], labeled scat- strongly labeled (not illustrated).
86 R.S. Redman
The mechanism of IR damage to salivary persistent damage including edema and loss of
glands has been thoughtfully considered by vesicles in secretory nerve endings of rat subman-
Nagler [61], Vissink et al. [101], and Dirix et al. dibular gland through 70 days after administration
[17]. The greater damage to serous acini from IR of single doses of 20, 25, or 30 Gy. This damage
has been theoretically attributed to generation of was interpreted as sufficient to cause loss of stimu-
free radicals via copper, iron, manganese, and lation, thus contributing to the cycle of regenera-
zinc contained in their secretory proteins. tion, engorgement with secretory granules, and
However, when the secretory granules of serous death of acinar cells observed during the 70 days
acini and granular convoluted tubules were dis- after IR. On the other hand, much of the secretory
charged by administration of secretagogues to innervation reportedly survives radiation [1, 13,
rats prior to irradiation with 15 Gy, the damage to 22, 34], and surviving acinar cells from rat salivary
and loss of acinar cells were not noticeably dif- glands subjected to mild (10 Gy) radiation still
ferent from the glands of rats that were not pre- respond normally to secretagogues [64].
treated [13, 67]. Better regeneration of acini and Furthermore, although neuropeptides such as sub-
recovery of salivary flow occurred in the pre- stance P and bombesin were increased transiently
treated rats, but this was attributed to the stimula- in rat submandibular gland ganglia at 10 days
tion of proliferation by the sialogogues. Moderate post-radiation of 30–40 Gy IR (given in daily frac-
to strong secretory stimulation has been shown to tionated doses), values had returned to normal by
induce a significant but transient increase in aci- 180 days. Changes in taste after IR also offer a
nar cell mitosis [80]. Note that in these rat experi- clue to the importance of IR effects on nerves in
ments, the mitoses would have been initiated salivary glands. After IR for cancer of the head and
several hours after the administration of a single neck, the taste threshold increased dramatically at
dose of IR. 1 month but had recovered to normal baseline val-
The dramatic drop in salivary output that ues by 6 months [78]. These results indicate that
occurs during the first few days after a seemingly the neuroepithelial (taste) cells, nerve endings, or
mild single dose or the first few fractionated both were destroyed or functionally disabled by
doses of IR apparently is not due to immediate the IR. Results with experimental animals have
death but to widespread dysfunction of the acinar documented damage and destruction of lingual
cells. Plausible explanations for this phenomenon taste buds by IR [20]. Regeneration of taste cells
include transient damage to the plasmalemma via differentiation from surface epithelial cells has
and receptor-signaling apparatuses [13, 17]. The been shown to be dependent on appropriate inner-
transient increase of salivary amylase in the blood vation [20]. From the foregoing, it seems likely
following an initial dose of radiation [8] may be that any nerve damage caused by IR for head and
related to leakage from the damaged cells. In any neck cancer would have been limited to the proxi-
event, if DNA and other cellular damages are not mal portions of the nerve processes, allowing new
severe enough to cause immediate death of the nerve endings to migrate from the surviving por-
cell, inaccurately repaired DNA damage may be tions of the nerves. Interestingly, taste recovered
sufficient to interfere with future proliferative even when there was marked xerostomia, suggest-
activity or cause delayed death. Thus, as the sur- ing that saliva may be less important to taste per-
viving acinar cells and their precursors in the ception than supposed on theoretical grounds [53].
intercalated and other ducts die during the ensu- There was no functional assessment of damage to
ing weeks, the extent to which the acini can the lingual glands of von Ebner, however, which
regenerate may be severely compromised. serve the majority of taste buds in the troughs of
There may be additional reasons for radiation- the vallate papillae and foliate folds [28]. In a case
induced loss of salivary gland function, such as pertinent to this point, Fajardo [19] illustrated “a
damage to innervation, blood vessels, and stroma. heavily irradiated tongue” in which “all that
Chomette et al. [11], using enzyme histochemistry remained of a lingual salivary gland was a dilated
and transmission electron microscopy, reported and ulcerated duct forming a mucocele.”
Following moderate- to high-dose IR, the s ubmandibular glands of mature rats, proliferative
stroma of human parotid and submandibular activity as determined by uptake of bromodeoxy-
glands has been described as undergoing adiposis uridine (BrdU) slowed for 24 h and then resumed
and fibrosis, respectively [23] (Fig. 4.7). Some in the intercalated ducts after 3 days and in the
stromal fibrosis also has been observed in rat sali- acini, striated ducts, and granular convoluted
vary glands but only after high-dose IR [10, 68, tubules after 6–10 days [66]. By IHC localization
84]. The endothelium of blood vessels is suscepti- of proliferating cell nuclear antigen (PCNA), pro-
ble to radiation damage, and compromised blood liferative activity increased by factors of 12.6, 3.4,
supply has been observed after 131I treatment for and 2.2 in the acinar, intercalated duct, and striated
thyroid cancer [50]. Thickening of extracellular duct cells, respectively, in rat submandibular
matrix components in response to high doses of IR glands 7 days after exposure to a single dose of
also has been reported [7]. These stromal changes 30 Gy [5]. These observations suggest that the
may restrict diffusion of nutrients, essential miner- divided doses used in therapeutic IR for cancer of
als, and oxygen to parenchymal cells and thus may the head and neck in humans compound the dam-
adversely affect late attempts at regeneration and age to salivary glands. Each successive spike in
function by surviving acinar cells. proliferating cells, which are more vulnerable to
Because of the importance of acini as the origin radiation damage compared to nondividing cells,
of the water and most of the organic secretory takes place during the next dose of radiation.
products in saliva, the more dramatic effects of IR C-Kit, a marker of stem cells, has been localized to
on acinar cells overshadow the considerable scattered cells in the excretory and intercalated
effects that moderate to high doses of IR have on ducts of mature human salivary glands [21, 42].
excretory and striated ducts [2, 50] (Fig. 3). The Both cell cycle and c-Kit proteins reportedly were
diminished function of striated and other large reduced in rats 12 weeks post-irradiation [85].
ducts is evident in higher salivary sodium and The scattered c-Kit-labeled parenchymal cells
chloride and lower bicarbonate concentrations and in the normal gland and lack of these in the IR
consequently lower pH, in stimulated saliva [100]. gland in Fig. 4.7 are consistent with the finding of
In this regard, stimulated saliva from irradiated Stiubea-Cohen et al. [85], indicating that the
glands is more like unstimulated saliva from nor- stem cell population of salivary glands is seri-
mal glands. The reduced flow and buffering capac- ously depleted by IR. It is noteworthy, however,
ity and lower pH of saliva from irradiated glands that in the normal gland, not only relatively
are major factors in the rapid development of den- undifferentiated basal cells in large ducts but also
tal caries in the absence of rigorous control mea- well-differentiated columnar cells in large ducts
sures. In addition to their importance in ion and acinar cells were labeled by c-Kit (Fig. 4.7c,
exchange with the luminal fluid as noted above, e). This suggests that well-differentiated salivary
they have been shown to have stem cells with the gland cells also may be stem cells. In addition,
capacity to regenerate acinar cells [41]. the enrichment of labeled cells in the stroma is
A plausible explanation for the poor recovery reminiscent of experiments in which mobilized
of acinar cells after moderate to high doses of IR hematopoietic stem cells entered the stroma but
is that the lifetime proliferative capacity of the not the parenchyma of irradiated mouse subman-
acinar cells and their progenitors is partially dibular glands [40].
depleted by repeated mitoses in attempts to
replace cells lost to radiation and is diminished
further by DNA/chromosomal damage in still 4.5 Perspective
viable cells [13, 61].
Studies in rats have shown that IR induces an Partial restoration of IR-damaged mouse sub-
increase in cellular proliferation among the paren- mandibular glands has been accomplished by
chymal cells in proportion to their loss. After a injecting a mixture of donor cells enriched in pro-
single dose of 15 Gy to the parotid and genitor or stem cells via a needle inserted through
88 R.S. Redman
the capsule [41]. However, these glands were K. Expression and localization of aquaporins, mem-
bers of the water channel family, during development
subjected to only 15 Gy of radiation, much less
of the rat submandibular gland. Pflügers Arch-Eur
than the 50–70 Gy that causes serious damage to J Physiol. 2003;446:641–5.
human salivary glands. Hematopoietic stem cells 5. Ballagh RH, Kudryk KG, Lampe HB, Moriarity B,
are attracted to and become part of the normal Mackay A, Burford-Mason AP, Dardick I. The patho-
biology of salivary gland. III. PCNA-localization of
minor salivary gland and oral epithelia [98, 99],
cycling cells induced in rat submandibular gland by
perhaps because labial and buccal mucosae are low-dose x-irradiation. Oral Surg Oral Med Oral
frequently traumatized. Why, then, do they not Pathol. 1994;77:27–35.
home to salivary gland epithelia badly damaged 6. Bandyopadhyay BC, Swaim WD, Sarkar A, Liu X,
Ambudkar IS. Extracellular Ca (2+) sensing in rat
by IR? As noted above, fair to good preservation
salivary duct cells. J Biol Chem. 2012;287:30305–16.
of salivary glands is being affected by innova- 7. Bartel-Friedrich S, Friedrich RE, Lautenschlager
tions such as computerized IR delivery programs C, Holzhausen HJ. Dose-response relationships and
and sparing agents such as amifostine. Perhaps the effect of age and latency period on the expres-
sion of laminin in irradiated rat mandibular glands.
even partial preservation of salivary gland struc-
Anticancer Res. 2000;20:221–5228.
ture, especially the vulnerable serous acini, from 8. Becciolini A, Giannardi CL, Porciani S, Fallai C,
therapeutic IR damage may allow techniques Pirtoli L. Plasma amylase activity as a biochemical
such as intraglandular injection of salivary or indicator of radiation injury to salivary glands. Acta
Radiol Oncol. 1984;23:353–9.
other stem cells to restore salivary function to
9. Caselitz J, Osborn M, Wustrow J, Seifert G, Weber
nearly normal. K. Immunohistochemical investigations on the epi-
myoepithelial islands in lymphoepithelial lesions.
Acknowledgments I thank Su-Wan Chen, Helen Use of monoclonal keratin antibodies. Lab Invest.
Cormier, Edward Flores, Lyvouch Filkoski, Patricia 1986;44:427–32.
Lewis, Rodney McNutt, and Shirley McLaughlin for pre- 10. Cherry CP, Gluckman A. Injury and repair follow-
paring the tissues for morphologic evaluation. The ing irradiation of salivary glands in male rats. Br
research enabling the illustrations was supported in part J Radiol. 1959;32:596–608.
by grants DE-102, DE-03002, 1-K4-DE-40-019, and 11. Chomette G, Auriol M, Vaillant JM, Bertrand JC,
DE-14995 from the National Institute of Dental and Chenal C. Effects of irradiation on the submandibu-
Craniofacial Research, the National Institutes of Health, lar gland of the rat. Virchows Arch (Pathol Anat).
Bethesda, MD, the Universities of Colorado, Minnesota, 1981;391:291–9.
and Washington, and by Research Career and Merit 12. Christensen ME, Hansesn HS, Poulsen SS, Bretlau P,
Review Awards from the United States Department of Nexo E. Immunochemical and quantitative changes
Veterans Affairs. Figures 4.5 and 4.6, the legends to these in salivary EGF, amylase and haptocorrin follow-
figures, and some of the text are reproduced from Redman ing radiotherapy for oral cancer. Acta Otolaryngol.
[73], with prior permission. 1996;116:137–43.
13. Coppes RP, Zeilstra LJW, Kampinga HH, Konings
AWT. Early to late sparing of radiation damage to
the parotid gland by adrenergic and muscarinic
References receptor agonists. Br J Cancer. 2001;85:1055–63.
14. Cossu M, Perra MT, Piludi M, Lantini
1. Aalto Y, Forsgren S, Kjorell U, Franzen L, MS. Subcellular localization of epidermal growth
Gustafsson H, Henriksson R. Time- and dose-related factor in human submandibular gland. Histochem
changes in the expression of substance P in salivary J. 2000;32:291–4.
glands in response to fractionated irradiation. Int 15. Culp DJ, Latchney LR, Fallon MA, Denny PA,
J Radiat Oncol Biol Phys. 1995;33:297–305. Denny PC, Couwenhoven RI, Chuang S. The gene
2. Abok K, Brunk U, Jung B, Ericsson J. Morphologic encoding mouse Muc19: cDNA, genomic organiza-
and histochemical studies on the differing radio- tion and relationship to Smgc. Physiol Genomics.
sensitivity of ductular and acinar cells of the rat 2004;19:303–18.
submandibular gland. Virchows Arch Cell Pathol. 16. Dardick I, Parks WR, Little J, Brown DL.
1984;45:443–60. Characterization of cytoskeletal proteins in basal cells
3. Ahmad M, Piludu M, Oppenheim FG, Helmerhorst of human parotid salivary gland ducts. Virchows Arch
EJ, Hand AR. Immunocytochemical localization A Pathol Anat Histopathol. 1988;412:525–32.
of histatins in human salivary glands. J Histochem 17. Dirix P, Nuyts S, Van den Bogaert W. Radiation-
Cytochem. 2004;52:361–70. induced xerostomia in patients with head and
4. Akamatsu T, Parvin MN, Murdiastuti K, Kosugi- neck cancer. A literature review. Cancer. 2006;
Tanaka C, Yao CJ, Miki O, Kanamori N, Hosoi 107:2525–34.
18. Eversole LR. Histogenetic classification of salivary on sympathetic nerve function in rat parotid glands.
gland tumors. Arch Pathol. 1971;92:433–43. J Oral Pathol Med. 1992;21:134–7.
19. Fajardo LF. Salivary glands and pancreas. In: Fajardo 35. Konturek SJ, Dembinski A, Warzecha Z, Brzozowski
LF, editor. Pathology of radiation injury. New York: T, Gregory H. Role of epidermal growth factor in
Masson Publishing USA; 1982. p. 77–87. healing of chronic gastroduodenal ulcers in rats.
20. Farbman AI. The taste bud: a model system for Gastroenterology. 1988;84:1300–7.
developmental studies. In: Slavkin HC, Bavetta 36. Konturek PG, Ernst EH, Konturek J, Bobrzynski
LA, editors. Developmental aspects of oral biology. T, Kwiecien N, Faller G, Gedliczka O, Hahn
New York: Academic Press; 1972. p. 109–23. EG. Salivary and gastric luminal release of epi-
21. Feng J, van der Zwaag M, Stokman MA, van Os R, dermal growth factor under basal conditions and
Coppes RP. Isolation and characterization of human after pentagastrin stimulation in healthy and in
salivary gland stem cells for stem cell transplanta- duodenal ulcer patients before and after eradica-
tion to reduce radiation-induced hyposalivation. tion of Helicobacter pylori. J Physiol Pharmacol.
Radiother Oncol. 2009;92:466–71. 1997;47:187–94.
22. Forsgren S, Franzen L, Liang Y, Gustafsson H, 37. Lantini MS, Piludi M, Cossu M. Subcellular local-
Henriksson R. Effects of irradiation on neuropep- ization of epidermal growth factor in human parotid
tide expression in rat salivary gland and spinal cord. gland. Histochem J. 2001;33:427–31.
Histochem J. 1994;26:630–40. 38. Larsen HS, Aure MH, Peters SB, Larsen M,
23. Frank RM, Herdly J, Phillipe E. Acquired dental Messelt EB, Galtung HK. Localization of AQP5
defects and salivary gland lesions after irradiation during development of the mouse submandibular
for carcinoma. J Am Dent Assoc. 1965;70:868–83. salivary gland. J Mol Histol. 2011; doi:10.1007/
24. Garrett JR. The innervation of normal human sub- s10735-010-9308-0.
mandibular and parotid salivary glands. Arch Oral 39. Leinonen J, Parkkila S, Kaunisto K, Koivunen P,
Biol. 1967;12:1417–36. Rajaniemi H. Secretion of carbonic anhydrase iso-
25. Geiger S, Geiger B, Leitner O, Marshak G. enzyme VI (CA VI) from human and rat lingual
Cytokeratin polypeptides expression in different epi- serous von Ebner’s glands. J Histochem Cytochem.
thelial elements of human salivary glands. Virchows 2001;49:657–62.
Arch A Pathol Anat Histopathol. 1987;410:403–14. 40. Lombaert IM, Wierenga PK, Kok T, Kampinga HH,
26. Gresik EW. The granular convoluted tubule (GCT) deHaan G, Coppes RP. Mobilization of bone marrow
of rodent submandibular glands. Microsc Res Tech. stem cells by granulocyte colony-stimulating factor
1994;27:1–24. ameliorates radiation-induced damage to salivary
27. Gugliotta P, Sapino A, Macri L, Skalli O, Gabbiani glands. Clin Cancer Res. 2006;12:1804–12.
G, Busolati G. Specific demonstration of myoepi- 41. Lombaert IMA, Brunsting JF, Wierenga PK, Faber
thelial cells by anti-alpha smooth muscle actin anti- H, Stokman MA, Kok T, Visser WH, Kampinga HH,
body. J Histochem Cytochem. 1988;36:659–63. de Haan G, Coppes RP. Rescue of salivary gland
28. Hand AR. The fine structure of von Ebner’s gland in function after stem cell transplantation in irradiated
the rat. J Cell Biol. 1970;44:340–52. glands. PLoS One. 2008;3:e2063. doi:10.1371/jour-
29. Hand AR. Adrenergic and cholinergic nerve terminal.pone.ooo2603: 1-16.
nals in the rat parotid gland. Electron microscopic 42. Lombaert IMA, Abrams SR, Li L, Eswarakumar VP,
observations on permanganate -fixed glands. Anat Sethi AJ, Witt RL, Hoffman MP. Combined KIT and
Rec. 1972;173:131–40. FGFR2b signaling regulates epithelial progenitor
30. Hand AR. Synthesis of secretory and plasma mem- expansion during organogenesis. Stem Cell Reports.
brane glycoproteins by striated duct cells of rat sali- 2013;1:604–19.
vary glands as visualized by radioautography after 43. Luna LG 1968. Manual of histologic staining meth-
3
H-fucose injection. Anat Rec. 1979;195:317–39. ods of the Armed Forces Institute of Pathology. 3rd
31. He X, Tse C-M, Donowitz M, Alper SL, Gabriel SE, edn. New York: McGraw-Hill. p. 94, 127–128.
Baum BJ. Polarized distribution of key membrane 44. Ma L, Wang WP, Chow JY, Yuen ST, Cho
transport membranes in the rat submandibular gland. CH. Reduction of EGF is associated with the delay
Pflügers Arch-Eur J Physiol. 1997;433:260–8. of ulcer healing by cigarette smoking. Am J Physiol
32. Henson BS, Inglehart MR, Eisbruch A, Ship Gastrointest Liver Physiol. 2000;278:G10–7.
JA. Preserved salivary output and xerostomia-related 45. Maier H, Bihl H. Effect of radioactive iodine
quality of life in head and neck cancer patients therapy on salivary glands. Acta Otolaryngol.
receiving parotid -sparing radiotherapy. Oral Oncol. 1987;103:318–24.
2001;37:84–93. 46. Malouf JG, Aragon C, Henson BS, Eisbruch A, Ship
33. Izutsu KT. Salivary electrolytes and fluid protec- JA. Influence of parotid-sparing radiotherapy on
tion in heath and disease. In: Sreebny LM, editor. xerostomia in head and neck cancer. Cancer Detect
The salivary system. Boca Raton: CRC Press; 1987. Prev. 2003;27:305–10.
p. 95–122. 47. Malpighi M. Epistolae anatomicae. De viscerum
34. Kohn WG, Grossman E, Fox PC, Armando I, structura exercitatio anatomica. In: Opera omnia.
Goldstein DS, Baum BJ. Effect of ionizing radiation London: Robert Littlebury; 1687. p. 51–144.
90 R.S. Redman
48. Man Y-G, Ball WD, Marchetti L, Hand AR. secretion in rat salivary glands is related to decreased
Contributions of intercalated duct cells to the normal acinar volume and not impaired signaling. Radiat
parenchyma of submandibular glands of adult rats. Res. 1999;151:150–8.
Anat Rec. 2001;263:201–14. 65. Oxford GE, Jonsson R, Olofsson J, Zelles T,
49. Mancini DM, Veit BC. Salivary growth factor in Humphreys-Beher MG. Elevated levels of human
patients with and without acid peptic ulcer disease. salivary epidermal growth factor after oral and juxta-
Am J Gastroenterol. 1990;85:1102–4. oral surgery. J Oral Maxillofac Surg. 1999;57:154–9.
50. Mandel SJ, Mandel L. Radioactive iodine and the 66. Peter B, Van Waarde MAWH, Vissink A,
salivary glands. Thyroid. 2003;13:265–71. ’s-Gravenmade EJ, Konings AWT (1994) Radiation-
51. Mangos JA, Braun G, Hamann KF. Micropuncture induced cell proliferation in the parotid and subman-
study of sodium and potassium excretion in the rat dibular glands of the rat. Radiat Res 140: 257–265.
parotid gland. Pflügers Arch ges Physiol. 1966; 67. Peter B, Van Waarde MAWH, Vissink A,
291:88–106. ’s-Gravenmade EJ, Konings AWT (1995) The role
52. Maria OM, Maria SM, Redman RS, Maria AM, of secretory granules in radiation-induced dysfunc-
El-Din TAS, Soussa EF, Tran SD. Effects of double tion of rat salivary glands. Radiat Res 141: 176–182.
ligation of Stensen’s duct on the rabbit parotid gland. 68. Phillipe RM. X-ray-induced changes in function
Biotech Histochem. 2014;82:181–98. and structure of the rat parotid gland. J Oral Surg.
53. Matsuo R. Role of saliva in the maintenance of taste 1970;28:431–7.
sensitivity. Crit Rev Oral Biol Med. 2000;11:216–29. 69. Redman RS. The anterior buccal gland of the rat: a
54. Messelt EB, Berg T. Effect of autonomic nerve mucous salivary gland which develops as a branch of
stimulation on bleb formation in striated duct cells stensen’s duct. Anat Rec. 1972;172:167–78.
of the rat submandibular gland. Acta Odontol Scand. 70. Redman RS. The importance of the minor salivary
1987;45:275–81. glands. Northwest Dent. 1974;53:19–23. 1974)
55. Moreira JE, Hand AR, Ball WD. Localization of neo- 71. Redman RS. Myoepithelium of the salivary glands.
natal secretory proteins in different cell types of the Microsc Res Tech. 1994;27:25–45.
developing rat submandibular gland from embryo- 72. Redman RS. Proliferative activity by cell type in
genesis to adulthood. Dev Biol. 1990;139:370–82. the developing rat parotid gland. Anat Rec. 1995;
56. Mori M. Histochemistry of the salivary glands. Boca 241:529–40.
Raton: CRC Press; 1991. p. 1–85. 73. Redman RS. On approaches to the functional res-
57. Mowry RW. The special value of methods that color toration of salivary glands damaged by therapeutic
both acidic and vicinal hydroxyl groups in the his- irradiation for head and neck cancer, with a review
tochemical study of mucins. With revised directions of related aspects of salivary gland morphology and
for the colloidal iron stain, the use of alcian blue 8GX development. Biotech Histochem. 2008;83:103–30.
and their combinations with the periodic acid-Schiff 74. Redman RS. Morphologic diversity of the minor sal-
reaction. Ann N Y Acad Sci. 1963;106:402–23. ivary glands of the rat: fertile ground for studies in
58. Murdoch-Kinch CA, Russo N, Griffith S, Braun T, gene function and proteomics. Biotech Histochem.
Eisbruch A, D’Silva NJ. Recovery of salivary epi- 2012;87:273–87.
dermal growth factor in parotid saliva following 75. Redman RS, Ball WD. Differentiation of myoepi-
sparing radiation therapy: a proof of principle study. thelial cells in the developing rat sublingual gland.
Oral Surg Oral Med Oral Pathol Oral Radiol Endod. Am J Anat. 1979;156:543–66.
2011;111:64–70. 76. Redman RS, Field RB. Chronology of salivary per-
59. Nagato T, Yoshida H, Yoshida A, Uehara Y. A scan- oxidase activity in the developing rat parotid gland.
ning electron microscope study of myoepithelial cells Anat Rec. 1993;235:611–21.
in exocrine glands. Cell Tissue Res. 1980;209:1–10. 77. Riva A, Puxeddu P, Del Fiacco M, Testa-Riva
60. Nagler RM, Baum BJ, Miller G, Fox PC. Long-term F. Ultrastructural localization of endogenous peroxi-
salivary effects of single-dose head and neck irradia- dase activity in human parotid and submandibular
tion in the rat. Arch Oral Biol. 1998;43:297–303. glands. J Anat. 1978;127:181–91.
61. Nagler RM. The enigmatic mechanism of irradiation- 78. Sandow PI, Hejrat-Yazdi M, Heft MW. Taste loss
induced damage to the major salivary glands. Oral and recovery following radiation therapy. J Dent
Dis. 2002;8:141–6. Res. 2006;85:608–11.
62. Nagy A, Nagashima H, Cha S, Oxford GE, Zelles T, 79. Sato A, Miyoshi S. Fine structure of tuft cells of the
Peck AB, Humphreys-Beher MG. Reduced wound main excretory duct in the rat submandibular gland.
healing in the NOD mouse model for Type I autoim- Anat Rec. 1997;248:325–31.
mune diabetes and its reversal by epidermal growth 80. Schneyer CA. Mitosis induced in adult rat parotid
factor supplementation. Diabetes. 2001;50:2100–4. following normal activity of the gland. Proc Soc Exp
63. Noguchi S, Ohba Y, Oka T. Effect of salivary growth Biol Med. 1970;134:603–7.
factor on wound healing of tongue in mice. Am 81. Schneyer LH, Schneyer CA. Inorganic composition
J Physiol. 1991;260:E620–5. of saliva. In: Code CF, editor. Secretion. Handbook
64. O’Connell AC, Redman RS, Evans RL, Ambudkar of physiology, section 6, vol. 2. Washington, DC:
IS. Irradiation-induced progressive decrease in fluid American Physiological Society; 1967. p. 497–529.
82. Shackleford JM, Schneyer LH. Ultrastructural 96. Teste-Riva F, Puxeddu P, Riva A, Diaz G. The epi-
aspects of the main excretory duct of rat subman- thelium of the excretory duct of the human subman-
dibular gland. Anat Rec. 1971;169:679–96. dibular gland: a transmission and scanning electron
83. Ship JA, Hu K. Radiotherapy-induced salivary dys- microscopic study. Am J Anat. 1981;160:361–93.
function. Semin Oncol. 2004;31(Suppl. 18):29–36. 97. Thesleff I, Vinikka L, Saxén S, Lehtonen E,
84. Sholley MM, Sodicoff M, Pratt NE. Early radia- Perheentupa J. The parotid gland is the main source
tion injury in the rat parotid gland. Reaction of of human salivary epidermal growth factor. Life Sci.
acinar cells and vascular endothelium. Lab Invest. 1988;43:13–8.
1974;31:340–54. 98. Tran SD, Pillemer SR, Dutra A, Barrett AJ,
85. Stiubea-Cohen R, David R, Neumann Y, Krief Brownstein MJ, Keys S, Pak E, Leakan RA, Kingman
G, Deutsch O, Zacks B, Aframian DJ, Palmon A, Yamada KM, Baum BJ, Mezey E. Differentiation
A. Effect of irradiation on cell transcriptome and of human bone marrow-derived cells into buccal
proteome of rat submandibular salivary glands. epithelial cells in vivo: a molecular analytical study.
PLoS One. 2012;7(7):e40636. doi:10.1371/journal Lancet. 2003;361:1084–8.
pone.0040636. 99. Tran SD, Redman RS, Barrett AJ, Pavletic SZ, Key
86. Tabak LA. In defense of the oral cavity. Pediatr S, Liu Y, Carpenter A, Nguyen HM, Sumita Y, Baum
Dent. 2006;28:110–7. BJ, Pillemer SR, Mezey É. Microchimerism in sali-
87. Takahashi S, Nakamura S, Suzuki R, Islam N, vary glands after blood- and marrow-derived stem
Domon T, Yamamoto T, Wakita M. Changing myo- cell transplantation. Biol Blood Marrow Transplant.
epithelial cell distribution during regeneration of rat 2011;17:429–33.
parotid glands. Int J Exp Pathol. 1999;80:283–90. 100. Valdez IH, Atkinson JC, Ship JA, Fox PC. Major
88. Tamarin A. Myoepithelium of the rat submaxillary salivary gland function in patients with radiation-
gland. J Ultrastruct Res. 1966;16:320–38. induced xerostomia: flow rates and sialochemistry.
89. Tamarin A, Sreebny LM. The rat submaxillary gland. Int J Radiat Oncol Biol Phys. 1992;25:41–7.
A correlative study by light and electron microscopy. 101. Vissink A, Jansma J, Spijkervet F, Burlage

J Morphol. 1965;117:295–352. FR, Coppes RP. Oral sequelae of head and
90. Tamarin A, Walker J. A longitudinal study of parotid neck radiotherapy. Crit Rev Oral Biol Med.
secretory kinematics by cinematography and com- 2003a;14:199–212.
puter analysis. Pflügers Arch. 1976;366:101–6. 102. Vissink A, Burlage FR, Spijkervet F, Jansma J,
91. Tandler B. Ultrastucture of the human submaxil- Coppes RP. Prevention and treatment of the con-
lary gland. III. Myoepithelium. Z Zellforsch. 1965; sequences of head and neck radiotherapy. Crit Rev
68:852–63. Oral Biol Med. 2003b;14:213–25.
92. Tandler B. Structure of the human parotid and sub- 103. Williams PL, editor. Gray’s anatomy. New York:
mandibular glands. In: Sreebny LM, editor. The sali- Charles Livingston; 1996.
vary system. Boca Raton: CRC Press; 1987. p. 21–41. 104. Yamashina S. Immunohistochemical study of amy-
93. Tandler B. Structure of the duct system in mam- lase and deoxyribonuclease in rat parotid gland.
malian major salivary glands. Microsc Res Tech. Acta Histochem Cytochem. 1981;14:236–60.
1993;26:57–74. 105. Young JA, Schlögel E. Micropuncture investiga-
94. Tatemoto Y, Kumasa S, Watanabe Y, Mori tion of sodium and potassium excretion in the rat
M. Epithelial membrane antigen as a marker of submaxillary gland. Pflügers Arch ges Physiol.
human salivary gland acinar and ductal cell function. 1966;291:85–98.
Acta Histochem. 1987;82:219–26. 106. Young JA, van Lennep EW. The morphology of sali-
95. Tepperman BL, Kiernan JA, Soper BD. The effect of vary glands. New York: Academic Press; 1978.
sialoadenectomy on gastric mucosal integrity in the 107. Yopilko A, Caillou B. Fine structural localization of
rat: roles of epidermal growth factor and prostaglan- acetylcholinesterase activity in the rat submandibu-
din E2. Can J Physiol Pharmacol. 1989;67:1512–9. lar gland. J Histochem Cytochem. 1989;33:439–45.
Adult Stem Cell Therapy
for Salivary Glands, with a Special 5
Emphasis on Mesenchymal Stem
Cells
Simon D Tran, Yoshinori Sumita, Dongdong Fang,

and Shen Hu
Abstract
Mesenchymal stromal/stem cell (MSC) therapy with the goal of restoring sali-
vary function following irradiation injury or in Sjögren’s syndrome (SS) has
made significant advances within the past 5 years. The majority of studies used
MSCs obtained from the bone marrow or adipose tissue, but MSCs isolated
from the salivary gland, dental pulp, and umbilical cord also demonstrated a
therapeutic efficacy in reestablishing salivary function. Based on the amount of
stimulated saliva secretion as a functional quantitative measure, irradiated
mice/rats that received MSC therapy restored their salivary flow rate (SFR) to
60–90 % of normal age-matched animals, while SFR of irradiated animals
without treatment remained at 35–50 % of secretory function. Thus, there was
25–40 % therapeutic improvement in animals receiving MSC therapy versus
those that did not. This would be clinically significant because patients with
severe salivary hypofunction (dry mouth) due to head and neck irradiation have
no improvement in SFR, if left untreated. In the SS-like disease mouse model,
MSC therapy restored SFR 80–100 % when treatment was given at an initial
phase of SS-like disease, while its effectiveness decreased to 50–60 % when
given at an advanced stage of disease. In SS patients, MSC therapy improved
SFR by 40–50 %. When tested in the rodent model, MSC therapy was success-
ful in restoring/maintaining the gland normal weights and histology (acinar
cells, blood vessels) and upregulated the expression of genes favorable for sali-
vary gland development and regeneration while downregulating inflammation
and cell apoptosis, promising positive effects of MSC therapy.
Y. Sumita, DDS, PhD

S.D. Tran, DMD, PhD (*) • D. Fang, BDS, MSc Department of Regenerative Oral Surgery,
McGill Craniofacial Tissue Engineering and Stem Nagasaki University, Nagasaki, Japan
Cells Laboratory, Faculty of Dentistry, McGill S. Hu, PhD, MBA
University, 3640 University Street, Division of Oral Biology and Medicine, School of
Montreal H3A 0C7, Canada Dentistry, University of California,
e-mail: [email protected] Los Angeles, CA, USA

94 S.D. Tran et al.
5.1 Experimental Approaches the presence of some surviving acinar cells. Gene
for the Functional therapy can target remaining ductal cells. Tissue
Restoration of Salivary engineering and cell therapy-based methods can
Glands theoretically be utilized in the absence of surviv-
ing acinar cells and can be viewed as regenerative
Clinically, there are two severe conditions that methods.
greatly reduce the secretion/output of saliva. The first regenerative approach is building an
First, 40,000 new cases of head and neck cancer artificial SG using tissue engineering [1, 4]. We
are diagnosed each year in the United States. and others have reported feasibility in culturing
Worldwide, this amounts to 640,000 new cases SG epithelial cells for a proposed artificial SG
yearly [12]. Irradiation (IR, radiation therapy, prototype [4, 47]. This strategy generates mainly
radiotherapy) is a key component of therapy for one portion of the SG parenchymal tissue (ductal
these cancer patients. Salivary glands (SGs), par- cells), and it is still challenging to regenerate a
ticularly the acinar cells, in the ionizing radiation fully fluid secretory tissue (both ductal and acinar
field suffer severe damage [49, 50]. These cells cells) [28, 30]. Thus, the second regenerative
are the principal site of fluid secretion in SGs, approach, using (stem) cell-based therapy, has
and such patients cannot produce adequate levels been enthusiastically pursued by several research
of saliva, leading to considerable morbidity and groups, including ours [13]. Currently, there are
extreme discomfort [31]. Second, patients with two types of stem/progenitor cells that have
Sjögren’s syndrome (SS), an autoimmune disor- shown promises as experimental treatments for
der, suffer similar irreversible damage to their SG hypofunction: a) salivary epithelial stem cells
salivary glands. In the United States, there are an and b) multipotent mesenchymal stromal cells
estimated 4 million SS patients, mostly peri- (MSCs) from either the bone marrow (BM) or
menopausal women. from the adipose tissue. Within the past 5 years,
Both IR and SS lead to the destruction of the there have been an increasing number of studies
salivary glandular parenchyma, and saliva pro- reporting the successful use of adult stem/pro-
duction is drastically reduced as a result. Both genitor cell-based therapies in treating SG hypo-
groups of patients experience severe SG hypofunction for both IR- and SS-injured SGs. The
function causing symptoms such as xerostomia main emphasis of this chapter will be on MSC
(dry mouth), dysphagia (impaired chewing and cell-based therapies. The topic of salivary epithe-
swallowing), dental caries, altered taste, oro- lial stem cell therapies against IR-induced SG
pharyngeal infections (candidiasis), mucositis, damage will only be mentioned briefly below, as
pain, and discomfort [5, 10]. These patients suf- there is already an excellent review covering this
fer considerable morbidity as xerostomia leads topic [41].
to reduced nutritional intake and weight loss,
significantly affects general health, and severely
reduces their quality of life [5]. For many SS 5.2 alivary Epithelial Stem Cell
S
and IR patients, in particular those whose sali- Therapy for Irradiation-
vary epithelial cells have been replaced by Injured Salivary Glands
fibrotic tissue, there is no available adequate
treatment. Current pharmacological approaches For the past several years, Robert Coppes’ group
(such as saliva-stimulating drugs) require the has steadily made significant advances in the
presence of some surviving epithelial tissue [3]. study and use of salivary epithelial stem cells in
Experimental approaches tested to date for func- the treatment of IR-injured SGs. This group of
tional restoration of SGs are the use of electro- researchers initially developed a culture system to
stimulation, acupuncture, gene therapy, tissue enrich and characterize salivary stem cells (termed
engineering, and cell therapy [2, 3, 15, 36, 39, as “salispheres”) [26]. The salispheres rescued
48]. Electrostimulation and acupuncture require SG functions after transplantation into IR mice.
5 Adult Stem Cell Therapy for Salivary Glands, with a Special Emphasis on Mesenchymal Stem Cells 95
The same group isolated salispheres (c-Kit+ cells) clinically oriented and has succeeded in incorpo-
from the human SG excretory ducts, which when rating a cell-separation method based on
placed in 3D culture developed into acinar-like magnetic-affinity cell sorting microbeads, as well
cells. Subsequent studies identified the salispheres as demonstrating that cultured single alpha6
with additional stem cell markers such as beta1 cell-originated clones could be cryopre-
CD24(high), CD29(high), and/or CD49f [7, 9, 33, served for 3 years without exhibiting genetic or
34]. Recently, the same researchers reported suc- functional instability when compared to non-
cessful isolation of adult human SG stem/progen- cryopreserved salivary cells.
itor cells at the single-cell level with evidence of
in vitro self-renewal and differentiation into mul-
tilineage organoids. These human SG stem/pro- 5.3 Multipotent Mesenchymal
genitor cells repaired IR-injured SG in mice [40]. Stromal Cell (MSCs) Therapy
This novel and promising approach is currently for Salivary Glands
being tested in a clinical study of patients under-
going IR for their head and neck cancers (personal A second source of adult stem cells that has
communication from RP Coppes). This strategy shown promises in reestablishing salivary func-
was initially thought to be limited for clinical use tion is the bone marrow (BM) mesenchymal stro-
due to an insufficient number of salispheres mal/stem cells (MSCs). More recently, stromal
obtained from patients’ SG biopsies. However, cells from the adipose tissue have demonstrated a
Coppes’ group recently demonstrated that FACS- comparable success for repairing SGs.
sorted CD24high and CD29high SG and salisphere- Transplantation methods for bone marrow- and
derived cells from young and old mice exhibited adipose tissue-derived stromal cells are success-
similar in vitro expansion and in vivo regenera- fully being used in the clinic because these cells
tion potential. Although older mice (22–26 months can now be harvested easily and safely from
old) had a reduced number of SG stem cells, they patients. Considering that methods for harvesting
were indistinguishable in vitro from SG stem cells MSCs from the adipose tissue are less invasive
from younger mice (8–12 weeks old) [27]. Quynh than those used for the BM, the adipose-derived
Thu Le’s group also investigated salispheres by stromal cells may gain more popularity among
isolating murine Lin− CD24+ c-Kit+ Sca1+ sali- clinicians.
vary stem cells [51]. In this chapter, the following definitions, taken
Gene expression analysis revealed that “glial from two positional papers of the International
cell line-derived neurotrophic factor” (GDNF has Society for Cellular Therapy (ISCT), will be used
a role in neuron survival, growth, differentiation, [6, 11]. A “stromal cell” is a connective tissue
and migration) was highly expressed in Lin− cell of any organ. A “stem cell” has the ability to
CD24+ c-Kit+ Sca1+ salivary stem cells. self-renew and is multipotent. The “multipotent
Administration of GDNF improved saliva pro- mesenchymal stromal cells” (MSCs) are
duction and increased the number of acinar cells fibroblast-like plastic-adherent cells, regardless
in submandibular glands of IR mice and enhanced of the tissue from which they are isolated. The
salisphere formation in vitro. These promising term “mesenchymal stem cells” (also MSCs) is
data indicated that the GDNF pathway may have used only for cells that meet specified stem cell
potential therapeutic benefits for IR-induced criteria. The widely recognized acronym, MSC,
xerostomia. Another group of researchers led by may be used for both cell populations, but inves-
Aaron Palmon and Doron Aframian have also tigators must clearly define the scientifically cor-
isolated salivary epithelial progenitor cells using rect designation in their publications [11].
the integrin alpha6 beta1 cell marker [37]. They Adipose tissue, either resected as an intact tissue
demonstrated that SG integrin alpha6 beta1- or aspirated using tumescent liposuction, is
expressing cells contained a subpopulation of minced and digested by enzymes. The released
SG-specific progenitor cells [35]. This group is cells are defined as the adipose tissue-derived
96 S.D. Tran et al.
“stromal vascular fraction” (SVF). The SVF con- rate (SFR) to 70–75 % of non-IR controls with
sists of a heterogeneous population of cells that follow-up times between 5 and 24 weeks post-
includes adipose stromal, hematopoietic stem/ transplantation. Irradiated but non-ASC-treated
progenitor cells, endothelial cells, erythrocytes, mice remained at 35–50 % when compared to
fibroblasts, lymphocytes, monocyte/macro- non-IR mice (i.e., healthy mice). All studies
phages, and pericytes [6]. When SVF cells are also reported that ASC-treated mice increased
cultured, a subset of fibroblast-like plastic- SG weights and preserved their histological
adherent cells appears (as observed with bone morphologies with more acini, a higher cell
marrow MSCs). These cells are defined as “adi- proliferation activity, reduced apoptosis, higher
pose tissue-derived stromal cells” (ASCs). amylase/mucin levels, and higher microvessel
Preclinical and clinical studies on the use of these densities than untreated irradiated SGs. The Lim
adult stromal cell populations (ASCs, SVF cells, and the Xiong studies were of particular interest
bone marrow MSCs) in regenerative medicine because they used human ASCs [23, 52].
have increased in recent years. This trend has Other treatment variables/factors that could be
been followed as well in SG research. Within the observed from these four ASCs studies were that:
past 5 years (2011–2015), there were 14 studies (a) One-time injection of ASCs directly into the
using stromal cells as a therapy for SS-like dis- submandibular glands [20] [52] was as successful
eased (n = 5 studies) or IR-injured SGs (n = 9 as with multiple i.v. injections (1x a week for
studies) [14, 16–18, 20, 22–25, 32, 43, 52–54]. 3 weeks [23], 2x a week for 6 weeks [22]). (b)
The mesenchymal stromal cells were either Injections of ASCs immediately following the IR
unselected from whole bone marrow [16, 18, 25, injury restored SG function (18 Gy IR) in the
32, 43], cultured from BM [17, 24, 53], cultured studies of [22, 52] and 15 Gy in the Lim study
from adipose tissue [20, 22, 23, 52], cultured [23], or as far as 10 weeks post-IR were as equally
from the SG [14], cultured from the dental pulp successful, but the IR dose of the latter study was
[54], or isolated from the umbilical cord [53]. much less (10 Gy) [20]. (c) It is important to note
This book chapter will initially review recent that both mouse and human ASCs could be
advances in the treatment of SG hypofunction injected into an immune-competent rodent
using ASCs, then BM MSCs, and finally the (mouse and rat) model. For instance, the Lim
authors’ experience with the soluble contents/ study [23] and the Xiong study [52] demonstrated
factors from these cells. For the sake of brevity, that human ASCs could repair and engraft into
this chapter will list studies published from year mouse and rat salivary glands, respectively. In
2010 and forward because our group has previ- addition, both of these studies were carried out
ously reviewed cell-based therapies in the treat- until 12–24 weeks posttransplantation of human
ment of xerostomia with studies from 2000 to ASCs, which differ from other studies that typi-
2010 [48]. cally followed up transplanted animals for
6–8 weeks.
5.4 Adipose Tissue-Derived

Stromal Cell (ASC) Therapy 5.5 Multipotent Mesenchymal
for Irradiation-Injured Stromal Cell (MSC) Therapy
Salivary Glands to Restore Irradiation-
Induced Salivary
Four studies presented transplanted adipose Hypofunction
tissue-
derived stromal cells (ASCs) for the
regeneration of IR-damaged SG [20, 22, 23, Within the past 5 years (2011–2015), there were
52]. All four studies indicated that ASCs were five studies using mesenchymal stromal cells as a
effective in improving salivary function. ASC- therapy for IR-injured SGs. The stromal cells
treated mice reestablished their salivary flow were either unselected from whole bone marrow
[25, 43], cultured from BM [24], cultured from the treatment of SG hypofunction following head
SG [14], or cultured from the dental pulp [54]. and neck IR [54]. DPECs were injected into the
Both the Lin [25] and the Sumita studies [43] submandibular glands of 15 Gy irradiated mice
used bone marrow cells (which contained a small at 4 and 14 days postirradiation. At 8 weeks
fraction of stromal cells) and reported successful post-IR, mice injected with DPECs reestablished
functional restoration of SG function to 80–90 % their SFR to 60 % SFR when compared to con-
of non-IR controls. Lim and colleagues built their trol (non-IR) mice, while irradiated PBS-injected
study [24] upon the Lin and the Sumita studies by mice remained at ~40 % SFR. In summary, all
isolating a homogeneous group of MSCs from the five abovementioned studies suggested that
BM through clonal cell culture. These cells were BM or its mesenchymal stromal cells could
named “bone marrow-derived clonal mesenchy- be used for restoration of IR-induced salivary
mal stem cells,” BM-cMSCs, and were positive hypofunction.
for CD44, Sca-1, while negative for CD34,
CD45. Injected BM-cMSCs ameliorated
IR-induced SG in mice. At 12 weeks posttrans- 5.6 Multipotent Mesenchymal
plantation SFR was reestablished to ~80 % of Stromal Cell (MSC) Therapy
nonirradiated controls (analogous to findings for Sjögren’s Syndrome (SS)
from the Sumita study). Thus, all three studies
(the Lin, the Sumita, and then the Lim studies) Within the past 5 years (2010–2015), there were
reported that administration of BM cells (either four studies using either whole BM, spleen cells,
whole BM, co-cultured with salivary epithelial MSCs from BM, or umbilical cord as a therapy to
cells, or clonally cultured as MSC) provided suc- repair SG in SS [16–18, 53]. Our group has been
cessful restoration of SG function to 80–90 % of testing the regenerative capacity of BM cells in
non-IR controls and improved histological fea- SGs of NOD mice [16–18, 32, 44], the most fre-
tures such as preserved acinar cells, less apop- quently used animal model to study SS-like
totic cells, and increased microvessel density. disease. NOD mice display infiltrates of lympho-
Two studies reported the use of MSCs as a cell cytes and a gradual loss of salivary function, and
therapy for IR-damaged SG, but these cells were the reduced saliva output mimics in part the con-
not isolated from the BM or adipose tissue. Rather dition seen in patients with SS [48]. We initially
MSCs were isolated from human SGs [14] or from tested if a therapy that reverses end-stage diabe-
the mouse dental pulp [54]. Jeong and colleagues tes in NOD mice would affect their SS-like dis-
cultured cells digested from human parotid and ease [19, 44]. This therapy had two components.
submandibular glands [14]. These cultured The first component was an injection of complete
adherent fibroblastic-like cells expressed CD44, Freund’s adjuvant (CFA) to induce endogenous
CD49f, and CD90 but not CD105, CD34, and TNF-α to kill disease-causing activated T cells.
CD45. These human salivary MSCs underwent The second component of the therapy was the
adipogenic, osteogenic, and chondrogenic differ- transplantation (reintroduction) of major histo-
entiation as well as differentiation, in vitro, into compatibility complex (MHC) class I-matched
amylase-expressing cells. Intraglandular injec- normal splenocytes. We initially chose spleno-
tions of human salivary MSCs into rat SGs 1 day cytes as these were comparable to BM cells in
following IR were found to reestablish their sali- mice. NOD mice that received CFA + spleno-
vary flow rate to ~ 65 % at the 8 weeks follow-up, cytes therapy could reverse both diabetes and
while irradiated non-MSC-treated mice remained xerostomia in SS [44]. Untreated NOD mice
at ~40 % SFR. Yamamura and colleagues tested showed a continuous decline in salivary flow, fol-
if cultured mouse dental pulp cells when differ- lowed by hyperglycemia and death. In a subse-
entiated in vitro into dental pulp endothelial cells quent study [18], we tested the long-term
(DPECs) could be used as a cell source (for their 52 weeks post-therapy effects of CFA and MHC
capacity to participate in neovascularization) in class I-matched normal BM cells to 7-week-old
98 S.D. Tran et al.
NOD mice (i.e., these mice had not yet developed and Th2 while suppressing Th17 and Tfh
SS-like disease). At week 52 posttreatment, responses. Intravenously infused MSCs migrated
CFA+BM cell-treated mice were normoglycemic toward the inflammatory regions in a stromal
compared to 10 % in the control group. BM cell- cell-derived factor-1 (SDF-1)-dependent man-
treated mice had their SG function (SFR) restored ner, while neutralization of the SDF-1 ligand
both quantitatively and qualitatively, compared to CXCR4 abolished the effectiveness of BM MSC
control NOD mice which continued to have their treatment in NOD mice. Thus, Songlin Wang’s
saliva secretion deteriorate over time. group discovered a critical role of the SDF-1/
In humans, SS is usually not diagnosed until CXCR4 axis (C-X-C chemokine receptor 4 is a
an advanced stage when clinical symptoms (such receptor for SDF-1) in directing MSC trafficking
as a decrease in saliva secretion) are manifested. toward inflammation sites, to exert suppressive
It was unknown if cell-based therapies would be activities and improve SG function.
effective in restoring salivary function when
given at an advanced stage of sialadenitis and
loss of salivary secretory function [16]. Thus, we 5.7 Possible Mechanisms
compared the efficacy of two cell-based therapies for the Observed
(BM versus spleen cells) in halting/reversing sal- Therapeutic Efficacy of MSC
ivary hypofunction at two critical time points of Therapy in Restoring
SS-like, which were (a) at the “initial phase of Salivary Function
SS-like” using 7–8-week-old NOD mice which
had normal saliva output and (b) at the “advanced Four in vivo studies have shown that transplanted
clinical disease phase” using 20-week-old NOD BM cells, MSCs, or ASCs differentiated into sali-
mice with minimal saliva output. Either BM or vary epithelial, acinar, ductal, or endothelial cells,
spleen cell therapies were effective during the and thus have proposed the differentiation of
initial phase of SS-like disease as SFRs were transplanted cells as a possible mechanism of
maintained between 80 and 100 % of presymp- action for the measured therapeutic effect in
tomatic levels (baseline SFR). When cell thera- IR-injured SGs [24, 25, 43, 52]. However, few
pies were given at an advanced phase of SS-like studies quantified this observed phenomenon, and
disease (20 weeks and older), SFRs improved but thus the mechanism of action for the therapeutic
were at best 50 % of presymptomatic levels. efficacy of BM and MSC cell therapies remained
The low immunogenicity and (high) immuno- to be investigated. Two human (observational)
regulatory potential of MSCs offer new treat- studies reported that the frequency of cell differ-
ment possibilities for autoimmune diseases. entiation/chimerism (without an IR injury) in SGs
Songlin Wang’s group reported that the immu- was much lower (~1 %) in patients who received
noregulatory activities of BM MSCs were a BM or hematopoietic stem cell transplant [42,
impaired in SS-like disease in NOD mice and SS 46]. This cell transdifferentiation phenomenon
patients [53]. These authors elegantly demon- was further characterized in vitro by using cell
strated that injections of mouse BM MSCs or of culture models of human MSCs co-cultured with
human umbilical MSCs suppressed the autoim- human salivary cells [29], human ASCs co-cul-
munity and restored SG secretory function in tured with mouse salivary cells [23], mouse MSCs
both the NOD mouse model (SFR was main- co-cultured with mouse salivary cells [38], and
tained at pre-disease levels, 100 % in 6-week-old mouse ASCs co-cultured with mouse salivary
NOD and 50–65 % SFR improvement in cells [21]. These in vitro studies indicated that
16-week-old treated NOD mice) and in SS stromal cells adopted an epithelial phenotype
patients (~40–50 % improvement in unstimu- when co-cultured with salivary epithelial cells or
lated and stimulated SFR during the 12 months with the conditioned media of salivary cells [21].
follow-up time). MSC treatment alleviated dis- There has been progress in the putative molec-
ease symptoms and directed T cells toward Treg ular cues responsible for the restoration of
salivaryfunction observed following salivary tumorigenic. However, the components which

epithelial stem cells, MSCs, or ASCs transplanta- were responsible for these promising therapeutic
tion. Seunghee Cha’s group investigated the reg- actions remained unknown. In a recent study, to
ulatory factors by proteomics that differentiate demonstrate that proteins were the active ingredi-
MSCs into salivary cells in vitro [38]. These ents, we devised a method using proteinase K fol-
authors reported that, of the 58 proteins detected, lowed by heating to deactivate proteins and for
three transcription factors involved in SG safe injections into mice. BM Soup and its “deac-
embryogenesis were selected as potential regula- tivated BM Soup” form were injected into mice
tory molecules in driving the transdifferentiation that had their SGs injured by IR [8]. Results at
of multipotent MSCs into salivary epithelial week 8 post-IR showed the “deactivated BM
cells: ankyrin repeat domain-containing protein Soup” was no better than injections of saline,
56, high mobility group protein 20B, and tran- while injections of native BM Soup restored
scription factor E2a. In mouse IR-injured SGs, saliva flow and protected salivary cells and blood
Quynh Thu Le’s group reported that glial cell vessels from IR damage. We demonstrated that
line-derived neurotrophic factor (GDNF) was the protein components but not the nucleic acids,
highly expressed in Lin− CD24+ c-Kit+ Sca1+ lipids, or carbohydrates in the BM Soup were the
salispheres and the GDNF pathway may have active/therapeutic ingredients for functional sali-
potential therapeutic benefits for IR-induced vary restoration following IR. Protein arrays
xerostomia [51]. In the SS-like NOD model, were used to preliminarily identify important
Songlin Wang’s group discovered the SDF-1/ cytokines and growth factors in the BM Soup.
CXCR4 axis directed MSC trafficking toward The protein arrays detected several angiogenesis-
inflammation sites to exert suppressive activities related factors (CD26, FGF, HGF, MMP-8,
and to improve SG function [53]. MMP-9, OPN, PF4, SDF-1) and cytokines
Our group has pursued a relatively more gen- (IL-1ra, IL-16) in BM Soup.
eral approach in searching for these molecular Our group has also tested BM Soup in the
cues. We tested if a paracrine mechanism (i.e., SS-like disease NOD mouse model [32]. This
transplanted cells secrete cytokines and growth study investigated if injections of BM Soup ver-
factors to repair salivary tissue) could be the main sus whole BM cells would provide comparable
effect behind the reported improvement in sali- improvements to diseased SGs, and the molecu-
vary function following cell transplantation [45]. lar alterations associated with these treatments
Whole BM cells were lysed, and their soluble [32]. BM Soup was found to restore SFR to nor-
intracellular contents, which we coined the term mal levels and significantly reduced the focus
as “bone marrow soup” (BM Soup), were injected scores in SGs of NOD mice. More than 1800 pro-
into mice with IR-injured SGs. We use the term teins in SG cells were quantified by a proteomic
BM Soup to represent all the yet-to-be-identified approach. Many salivary proteins involved in
soluble components of the cell lysate from whole inflammation and apoptosis were found to be
bone marrow cells. Eight weeks post-IR, BM downregulated, whereas those involved in SG
Soup restored salivary flow rates to normal lev- biology and development/regeneration were
els, protected salivary acinar, ductal, myoepithe- upregulated in the BM Soup-treated mice.
lial, and progenitor cells, increased cell
proliferation and blood vessels, and upregulated Conclusions
expression of several tissue remodeling/repair/ In summary, MSC therapy with the goal of
regenerative genes (MMP2, CyclinD1, BMP7, restoring SG function following IR injury or
EGF, NGF). BM Soup was as an efficient thera- in SS is an efficient and promising approach.
peutic agent as injections of live whole BM cells When tested in an animal model and using
(which so far were thought to be essential). stimulated saliva secretion as a functional
Because the BM Soup is an extract from a cell quantitative measure, mice/rats with
lysate, it is theoretically less immunogenic and IR-injured SGs which received the MSC
100 S.D. Tran et al.
therapy restored their SFR to 60–90 % of that cellular and molecular mechanisms that sup-
measured in normal age-matched animals, port the use of MSC in repairing injured SGs
while the SFR of IR animals without treat- will need to be deciphered in greater detail.
ment remained at 35–50 %. Thus, there was a
25–40 % therapeutic improvement between Acknowledgments We thank Professor Bruce J Baum
animals receiving the MSC therapy or not. for his invaluable advice, input, and encouragements dur-
ing these past several years on the research pertaining to
This would be a clinically significant out- bone marrow cells and its cell soup in the repair of sali-
come because patients with severe salivary vary glands. These studies were supported in part by the
hypofunction due to head and neck IR have Canadian Institutes of Health Research (CIHR).
0 % improvement in SFR, if left untreated. In
the NOD mouse model, MSC therapy
restored SFR 80–100 % if the treatment was
given at an initial phase of SS-like disease,
References
while its effectiveness decreased to 50–60 % 1. Aframian DJ, Palmon A. Current status of the devel-
when given at an advanced phase of SS-like opment of an artificial salivary gland. Tissue Eng Part
disease. In SS patients, MSC therapy B Rev. 2008;14:187–98.
improved SFR by 40–50 %. When tested in 2. Alajbeg I, Falcao DP, Tran SD, Martin-Granizo R,
Lafaurie GI, Matranga D, Pejda S, Vuletic L, Mantilla
the rodent model, MSC therapy was success- R, Leal SC, Bezerra AC, Menard HA, Kimoto S, Pan
ful in restoring/maintaining the gland normal S, Maniegas L, Krushinski CA, Melilli D, Campisi G,
weights and histology (acinar cells, blood Paderni C, Mendoza GR, Yepes JF, Lindh L, Koray M,
vessels) and upregulated the expression of Mumcu G, Elad S, Zeevi I, Barrios BC, Lopez
Sanchez RM, Lassauzay C, Fromentin O, Beiski BZ,
genes favorable to SG development and Strietzel FP, Konttinen YT, Wolff A, Zunt SL. Intraoral
regeneration while downregulating inflam- electrostimulator for xerostomia relief: a long-term,
mation and cell apoptosis. The majority of multicenter, open-label, uncontrolled, clinical trial.
studies used MSCs obtained from the bone Oral Surg Oral Med Oral Pathol Oral Radiol.
2012;113:773–81.
marrow and adipose tissue, but MSCs iso- 3. Baum BJ, Kok M, Tran SD, Yamano S. The impact of
lated from the SG and dental pulp also dem- gene therapy on dentistry: a revisiting after six years.
onstrated a therapeutic efficacy in J Am Dent Assoc. 2002;133:35–44.
reestablishing SG function. From that per- 4. Baum BJ, Tran SD. Synergy between genetic and tis-
sue engineering: creating an artificial salivary gland.
spective, MSCs from tissues that can be Periodontol 2000. 2006;2000(41):218–23.
obtained with a less invasive procedure (such 5. Bhide SA, Miah AB, Harrington KJ, Newbold KL,
as adipose tissue versus bone marrow) or Nutting CM. Radiation-induced xerostomia: patho-
from tissues already removed during the sur- physiology, prevention and treatment. Clin Oncol (R
Coll Radiol). 2009;21:737–44.
gical procedure (such as SG or teeth) would 6. Bourin P, Bunnell BA, Casteilla L, Dominici M, Katz
be more advantageous to both the patient and AJ, March KL, Redl H, Rubin JP, Yoshimura K,
the clinician. We were excited that two stud- Gimble JM. Stromal cells from the adipose tissue-
ies reported the successful transplantation of derived stromal vascular fraction and culture expanded
adipose tissue-derived stromal/stem cells: a joint
human adipose tissue-derived MSCs in statement of the International Federation for Adipose
restoring salivary function in immune-com- Therapeutics and Science (IFATS) and the
petent mice and rats and that one study (with International Society for Cellular Therapy (ISCT).
a follow-up of 12 months) reported a thera- Cytotherapy. 2013;15:641–8.
7. Coppes R, Stokman M. Stem cells and the repair of
peutic effect of umbilical cord MSCs infused radiation-induced salivary gland damage. Oral Dis.
to SS patients. We perceived that there were 2011;17:143–53.
major advances within the past 5 years 8. Fang D, Hu S, Liu Y, Quan VH, Seuntjens J, Tran
because all studies using MSC therapies SD. Identification of the active components in bone
marrow soup: a mitigator against irradiation-injury to
were moderately to highly successful in rees- salivary glands. Sci Rep. 2015;5:16017.
tablishing function to IR- and SS-injured 9. Feng J, van der Zwaag M, Stokman MA, van Os R,
SGs. Despite these encouraging results, the Coppes RP. Isolation and characterization of human
salivary gland cells for stem cell transplantation to 23. Lim JY, Ra JC, Shin IS, Jang YH, An HY, Choi JS,
reduce radiation-induced hyposalivation. Radiother Kim WC, Kim YM. Systemic transplantation of
Oncol. 2009;92:466–71. human adipose tissue-derived mesenchymal stem
10. Fox PC. Acquired salivary dysfunction. drugs and cells for the regeneration of irradiation-induced sali-
radiation. Ann N Y Acad Sci. 1998;842:132–7. vary gland damage. PLoS One. 2013a;8:e71167.
11. Horwitz EM, Le Blanc K, Dominici M, Mueller I, 24. Lim JY, Yi T, Choi JS, Jang YH, Lee S, Kim HJ, Song
Slaper-Cortenbach I, Marini FC, Deans RJ, Krause SU, Kim YM. Intraglandular transplantation of bone
DS, Keating A. Clarification of the nomenclature for marrow-derived clonal mesenchymal stem cells for
MSC: the international society for cellular therapy amelioration of post-irradiation salivary gland dam-
position statement. Cytotherapy. 2005;7:393–5. age. Oral Oncol. 2013b;49:136–43.
12. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 25. Lin CY, Chang FH, Chen CY, Huang CY, Hu FC,
2010. CA Cancer J Clin. 2010;60:277–300. Huang WK, Ju SS, Chen MH. Cell therapy for sali-
13. Jensen DH, Oliveri RS, Trojahn Kolle SF, Fischer- vary gland regeneration. J Dent Res. 2011;90:341–6.
Nielsen A, Specht L, Bardow A, Buchwald 26. Lombaert IM, Brunsting JF, Wierenga PK, Faber H,
C. Mesenchymal stem cell therapy for salivary gland Stokman MA, Kok T, Visser WH, Kampinga HH, de
dysfunction and xerostomia: a systematic review of Haan G, Coppes RP. Rescue of salivary gland func-
preclinical studies. Oral Surg Oral Med Oral Pathol tion after stem cell transplantation in irradiated
Oral Radiol. 2014;117:335–42. e1 glands. PLoS One. 2008;3:e2063.
14. Jeong J, Baek H, Kim YJ, Choi Y, Lee H, Lee E, Kim 27. Maimets M, Bron R, de Haan G, van Os R, Coppes
ES, Hah JH, Kwon TK, Choi IJ, Kwon H. Human sali- RP. Similar ex vivo expansion and post-irradiation
vary gland stem cells ameliorate hyposalivation of regenerative potential of juvenile and aged salivary
radiation-damaged rat salivary glands. Exp Mol Med. gland stem cells. Radiother Oncol. 2015;16:443–8.
2013;45:e58. 28. Maria OM, Maria O, Liu Y, Komarova SV, Tran

15. Kagami H. The potential use of cell-based therapies in SD. Matrigel improves functional properties of human
the treatment of oral diseases. Oral Dis. 2015;21: submandibular salivary gland cell line. Int J Biochem
545–9. Cell Biol. 2011a;43:622–31.
16. Khalili S, Faustman DL, Liu Y, Sumita Y, Blank D, 29. Maria OM, Tran SD. Human mesenchymal stem cells
Peterson A, Kodama S, Tran SD. Treatment for salivary cultured with salivary gland biopsies adopt an epithe-
gland hypofunction at both initial and advanced stages lial phenotype. Stem Cells Dev. 2011;20:959–67.
of Sjogren-like disease: a comparative study of bone 30. Maria OM, Zeitouni A, Gologan O, Tran SD. Matrigel
marrow therapy versus spleen cell therapy with a 1-year improves functional properties of primary human sali-
monitoring period. Cytotherapy. 2014;16:412–23. vary gland cells. Tissue Eng Part A. 2011b;17:
17. Khalili S, Liu Y, Kornete M, Roescher N, Kodama S, 1229–38.
Peterson A, Piccirillo CA, Tran SD. Mesenchymal 31. Melvin JE, Yule D, Shuttleworth T, Begenisich

stromal cells improve salivary function and reduce T. Regulation of fluid and electrolyte secretion in sali-
lymphocytic infiltrates in mice with Sjogren’s-like vary gland acinar cells. Annu Rev Physiol. 2005;67:
disease. PLoS One. 2012;7:e38615. 445–69.
18. Khalili S, Liu Y, Sumita Y, Maria OM, Blank D, Key 32. Misuno K, Tran SD, Khalili S, Huang J, Liu Y, Hu
S, Mezey E, Tran SD. Bone marrow cells are a source S. Quantitative analysis of protein and gene expres-
of undifferentiated cells to prevent Sjogren’s syn- sion in salivary glands of Sjogren’s-like disease NOD
drome and to preserve salivary glands function in the mice treated by bone marrow soup. PLoS One.
non-obese diabetic mice. Int J Biochem Cell Biol. 2014;9:e87158.
2010;42:1893–9. 33. Nanduri LS, Baanstra M, Faber H, Rocchi C, Zwart E,
19. Kodama S, Kuhtreiber W, Fujimura S, Dale EA,
de Haan G, van Os R, Coppes RP. Purification and
Faustman DL. Islet regeneration during the reversal of ex vivo expansion of fully functional salivary gland
autoimmune diabetes in NOD mice. Science. stem cells. Stem Cell Reports. 2014;3:957–64.
2003;302:1223–7. 34. Nanduri LS, Lombaert IM, van der Zwaag M, Faber
20. Kojima T, Kanemaru S, Hirano S, Tateya I, Ohno S, H, Brunsting JF, van Os RP, Coppes RP. Salisphere
Nakamura T, Ito J. Regeneration of radiation damaged derived c-Kit+ cell transplantation restores tissue
salivary glands with adipose-derived stromal cells. homeostasis in irradiated salivary gland. Radiother
Laryngoscope. 2011;121:1864–9. Oncol. 2013;108:458–63.
21. Lee J, Park S, Roh S. Transdifferentiation of mouse 35. Neumann Y, David R, Stiubea-Cohen R, Orbach Y,
adipose-derived stromal cells into acinar cells of the Aframian DJ, Palmon A. Long-term cryopreservation
submandibular gland using a co-culture system. Exp model of rat salivary gland stem cells for future ther-
Cell Res. 2015;334:160–72. apy in irradiated head and neck cancer patients. Tissue
22. Li Z, Wang Y, Xing H, Wang Z, Hu H, An R, Xu H, Eng Part C Methods. 2012;18:710–8.
Liu Y, Liu B. Protective efficacy of intravenous trans- 36. Nguyen TT, Mui B, Mehrabzadeh M, Chea Y,

plantation of adipose-derived stem cells for the pre- Chaudhry Z, Chaudhry K, Tran SD. Regeneration of
vention of radiation-induced salivary gland damage. tissues of the oral complex: current clinical trends and
Arch Oral Biol. 2015;60:1488–96. research advances. J Can Dent Assoc. 2013;79:d1.
102 S.D. Tran et al.
37. Palmon A, David R, Neumann Y, Stiubea-Cohen R, 46. Tran SD, Redman RS, Barrett AJ, Pavletic SZ, Key S, Liu
Krief G, Aframian DJ. High-efficiency immunomag- Y, Carpenter A, Nguyen HM, Sumita Y, Baum BJ,
netic isolation of solid tissue-originated integrin- Pillemer SR, Mezey E. Microchimerism in salivary glands
expressing adult stem cells. Methods. 2012;56: after blood- and marrow-derived stem cell transplantation.
305–9. Biol Blood Marrow Transplant. 2011a;17:429–33.
38.
Park YJ, Koh J, Gauna AE, Chen S, Cha 47. Tran SD, Sugito T, Dipasquale G, Cotrim AP,

S. Identification of regulatory factors for mesenchy- Bandyopadhyay BC, Riddle K, Mooney D, Kok MR,
mal stem cell-derived salivary epithelial cells in a co- Chiorini JA, Baum BJ. Re-engineering primary epi-
culture system. PLoS One. 2014;9:e112158. thelial cells from rhesus monkey parotid glands for
39. Patel VN, Hoffman MP. Salivary gland development: use in developing an artificial salivary gland. Tissue
a template for regeneration. Semin Cell Dev Biol. Eng. 2006;12:2939–48.
2014;25-26:52–60. 48. Tran SD, Sumita Y, Khalili S. Bone marrow-derived
40. Pringle S, Maimets M, van der Zwaag M, Stokman cells: a potential approach for the treatment of xero-
MA, van Gosliga D, Zwart E, Witjes MJ, de Haan G, stomia. Int J Biochem Cell Biol. 2011b;43:5–9.
van Os R, Coppes RP. Human salivary gland stem 49. Vissink A, Burlage FR, Spijkervet FK, Jansma J,

cells functionally restore radiation damaged salivary Coppes RP. Prevention and treatment of the conse-
glands. Stem Cells. 2016;34:640–52. quences of head and neck radiotherapy. Crit Rev Oral
41. Pringle S, van Os R, Coppes RP. Concise review: Biol Med. 2003a;14:213–25.
adult salivary gland stem cells and a potential therapy 50. Vissink A, Jansma J, Spijkervet FK, Burlage FR,

for xerostomia. Stem Cells. 2013;31:613–9. Coppes RP. Oral sequelae of head and neck radiother-
42. Souza LN, Faria DR, Dutra WO, Gomes CC, Gomez apy. Crit Rev Oral Biol Med. 2003b;14:199–212.
RS. Microchimerism in labial salivary glands of 51. Xiao N, Lin Y, Cao H, Sirjani D, Giaccia AJ, Koong
hematopoietic stem cell transplanted patients. Oral AC, Kong CS, Diehn M, LE QT. Neurotrophic factor
Dis. 2011;17:484–8. GDNF promotes survival of salivary stem cells. J Clin
43. Sumita Y, Liu Y, Khalili S, Maria OM, Xia D, Key S, Invest. 2014;124:3364–77.
Cotrim AP, Mezey E, Tran SD. Bone marrow-derived 52. Xiong X, Shi X, Chen F. Human adipose tissue

cells rescue salivary gland function in mice with head derived stem cells alleviate radiationinduced xerosto-
and neck irradiation. Int J Biochem Cell Biol. mia. Int J Mol Med. 2014;34:749–55.
2011;43:80–7. 53. Xu J, Wang D, Liu D, Fan Z, Zhang H, Liu O, Ding G,
44. Tran SD, Kodama S, Lodde BM, Szalayova I, Key S, Gao R, Zhang C, Ding Y, Bromberg JS, Chen W, Sun
Khalili S, Faustman DL, Mezey E. Reversal of L, Wang S. Allogeneic mesenchymal stem cell treat-
Sjogren’s-like syndrome in non-obese diabetic mice. ment alleviates experimental and clinical Sjogren syn-
Ann Rheum Dis. 2007;66:812–4. drome. Blood. 2012;120:3142–51.
45. Tran SD, Liu Y, Xia D, Maria OM, Khalili S, Wang 54. Yamamura Y, Yamada H, Sakurai T, Ide F, Inoue H,
RW, Quan VH, Hu S, Seuntjens J. Paracrine effects of Muramatsu T, Mishima K, Hamada Y, Saito
bone marrow soup restore organ function, regenera- I. Treatment of salivary gland hypofunction by trans-
tion, and repair in salivary glands damaged by irradia- plantation with dental pulp cells. Arch Oral Biol.
tion. PLoS One. 2013;8:e61632. 2013;58:935–42.
Directed Cell Differentiation
by Inductive Signals in Salivary 6
Gland Regeneration: Lessons
Learned from Pancreas and Liver
Regeneration
Yun-Jong Park and Seunghee Cha
Abstract
Xerostomia (dry mouth) is a deleterious condition that patients with radia-
tion therapy for head and neck cancer or autoimmune Sjögren’s syndrome
suffer from. Current remedies for this condition provide no substantial
relief of xerostomia. As a result, new alternatives to these palliative reme-
dies, such as artificial salivary glands, gene therapy, or cell-based interven-
tions, are on the horizon. An urgent demand for acquisition of knowledge
on stem cell regulation, which is critical for salivary gland regeneration,
has allowed systematic and mechanistic research endeavor focusing on the
identification of key regulators for cell lineage determination. This book
chapter summarizes the key inductive signals, which include extrinsic fac-
tors secreted from the microenvironment and cell intrinsic factors that
drive differentiation of the stem cells into the cells of the pancreas, liver,
and salivary glands as they share the endodermal origin during develop-
ment. The plethora of information available in pancreas and liver regen-
eration studies provides insight into key signals that govern vital processes
during orchestrated stem cell differentiation for salivary epithelial cells.
Some examples of transdifferentiation between differentiated cells of dif-
ferent organs and in vivo applications of inductive factors offer perspec-
tives on future clinical applications with improved safety and efficacy.
6.1 Introduction
S. Cha, DDS, PhD (*) Saliva contains various components that play criti-
Department of Oral and Maxillofacial Diagnostic cal roles in oral health. The production of saliva
Sciences, University of Florida College of Dentistry, from the salivary glands (SGs) can be significantly
Gainesville, FL 32610, USA
altered by pathological conditions, such as auto-
immune Sjögren’s syndrome (SjS) or radiation
Y.-J. Park
Department of Pharmacology, The University of
therapy for head and neck cancer [1]. SjS is char-
North Carolina at Chapel Hill, Chapel Hill, NC acterized by severe dry eyes (keratoconjunctivitis
27599-3280, USA sicca) and dry mouth (xerostomia) with various

104 Y.-J. Park and S. Cha
complications in multiple organs. Although SjS This will require supplement of in vitro grown
can develop at any age, it is more prevalent in SSPCs, which can be procured prior to irradiation
elderly females than in males. Radiation therapy or SPCC-derived and differentiated SECs into the
for patients with head and neck cancer results in glands. In the second, SSPCs are spared from dam-
acinar cell death in the SGs causing pathological age. However, there is a lack of critical inductive
xerostomia. Roughly 50,000 cases of head and signals for differentiation from SSPCs to SECs in
neck cancer are reported in the United States every the injured SGs. Interventions for the second sce-
year ranking as the fifth most common malignancy nario will require providing necessary inductive
worldwide [2]. Upon diagnosis, the standard of signals to the SSPCs, which include extrinsic fac-
care is dictated by tumor stage, and, for locally tors, intrinsic factors, and molecules that relay the
advanced tumors, surgical resection followed by signals generated by extrinsic and/or intrinsic fac-
radiotherapy is recommended. Due to the posi- tors. In most advanced clinical cases, considering
tioning of many oral tumors, non-affected tissues, the extent of radiation/inflammatory damage and
such as the SGs, are often exposed to therapeutic fibrotic/atropic changes that follow, we can indis-
radiation. Radiation damage to the oral tissues putably assume that neither SSPCs nor inductive
results in significant morbidity and diminished signals may exist in the severely damaged SGs.
quality of life [3–5]. Therefore, orchestration of these signals in a
Recently, studies have reported that SGs are timely and a hierarchical manner in conjunction
amenable to a stem cell-based approach for regen- with replenishing SSPCs can be an attractive option
eration following injury [6, 7]. The emerging confor reprograming of stem cells to achieve functional
cepts from recent studies suggest that different SG regeneration. Currently, information on the
types of stem cells, such as embryonic stem (ES) inductive signals for stem cell differentiation in the
cells, adult stem cells from the bone marrow (BM) field of SG research is extremely scarce, especially
and adipocyte, and SG stem cells/progenitor cells TFs in SG regeneration, in contrast to the plethora
(SSPCs), and induced pluripotent stem cells (iPS) of information available in the pancreas or liver
cells, have great potential for SG regeneration [7, regeneration. In this book chapter, we will present
8]. Of those stem cells, SSPCs have provided extrinsic factors and TFs with an emphasis on pan-
insight into future therapeutics as they have dem- creatic/liver TFs for their importance in lineage
onstrated a notable capacity to differentiate into determination. The latter information provides
salivary epithelial-like cells [9], and BM-derived insights into the overview of stem cell research in
mesenchymal stem cells (MSCs) have added their general and can be utilized to confidently guide
superb value to current research endeavor to restore future applications of SG stem cells. In addition, we
glandular function [10–12]. add some of our recent findings on TFs identified in
In this book chapter, we describe molecules that mouse BM-MSC-derived SPCs in vitro. Some
are known to trigger or promote stem cell differen- examples of in vivo applications of inductive sig-
tiation into SG progenitors (SPCs) or SG epithelial nals to mouse models were described as well. We
cells (SECs) by specifically focusing on factors conclude this chapter with important lessons that
available in the SG microenvironment (extrinsic we can learn from currently available studies.
factors) and transcription factors (TFs) (intrinsic
factors) expressed in SSPCs or BM-MSCs. In addi-
tion, we detail TFs that are known to determine cell 6.2 I nductive Signals for Stem or
fate in the field of pancreas and liver regeneration Nonstem Cell Lineage
to exploit the knowledge available. The rationale Determination
for our focus on these molecules is straightforward. in Endodermal Organs
In real clinical cases, patients with irradiation or
SjS rarely recover their secretory function. In other 6.2.1 Extrinsic Factors
words, once harmful radiation is applied or autore-
active cells infiltrate the glands, the damage is most Understanding a SG microenvironment that
likely permanent. This yields two possible scenar- involves complex processes of cell–cell communi-
ios. In the first, SSPCs are completely depleted. cation, cell–matrix interaction, and cell signal
6 Directed Cell Differentiation by Inductive Signals in Salivary Gland Regeneration 105
transduction within a three-dimensional structure Endodermal-originated organs, such as major

is a critical step toward stem cell reprogramming SG (submandibular and sublingual), pancreas,
and SG reconstruction [13–15]. Extrinsic factors and liver, are regulated by complex processes of
available in or from the SG microenvironment con- development, which involves the balancing of
tribute to SG development or SSPC differentiation interplay among common signaling pathways.
by generating appropriate scaffolds or key induc- Specific growth factors and differentiation fac-
tive signals [16]. Specifically, the extracellular tors are released from the microenvironment,
matrix (ECM) is an important element of the cel- which controls the expression of a set of TFs
lular niche and supplies critical biochemical and resulting in the generation of targeted endoder-
physical signals to initiate cellular functions in the mal cell types. The initial interactions usually
tissue [17]. Therefore, researchers in the field of confer competence to respond to additional
tissue engineering focus on various biomaterial inductive signals that establish organ determina-
scaffolds for tissue regeneration or repair that tion and specification at particular time points
mimic some of the critical properties of the ECM- referred to as competence windows [35]. The
like cellular microenvironment. For instance, the importance of the competence windows is exem-
scaffold material Matrigel®, which contains base- plified during liver and pancreatic development.
ment membrane proteins secreted by mouse sar- For instance, when bone morphogenetic protein
coma cells, has been used to reconstitute (BMP) activity generated by the adjacent
tissue-specific ECM to control stem cell fate [18]. mesoderm- derived structures operates on the
Although varying levels of success have been ventral endoderm, hepatic induction ensues dur-
achieved with this product, there are some limita- ing early embryonic development. With active
tions compared to natural scaffold materials such BMP signaling, ventral pancreas progenitors are
as silk, which has a wide range of elasticity and located away from the midline endoderm,
pore size (allowing tissue-specific scaffold forma- whereas BMP inhibition in the dorsal endoderm
tion and nutrition access), the ability to biodegrade, plays a crucial function in the acquisition of a
and low toxicity and immunogenicity [19, 20]. pancreatic fate [36, 37]. This interesting phenom-
Other studies have suggested the use of polyeth- enon illustrates that despite our limited knowl-
ylene glycol hydrogel (PEG-hydrogel) for SG edge of extrinsic factors controlling patterning,
regeneration [21]. It is inert due to its highly proliferation, and differentiation in the pancreas
hydrated and uncharged structure for introduction development, there is strong evidence observed
of growth factors, ECM proteins, or mimetics to from all vertebrate model organisms that com-
control cell behavior. Furthermore, the stiffness, plex spatiotemporal combinations of common
degradability, and mesh size of PEG-hydrogel can signaling pathways are important in the specifi-
be controlled by the composition and relative cation of endodermal cell fate [38–40]. A list of
amounts of PEG macromers [21–26]. PEG- such important pathways includes the sonic
hydrogel also has been utilized successfully to cul- hedgehog, Wnt, retinoid, and notch pathways,
ture and control the behavior of various cell types, activin/BMP, fibroblast growth factor (FGF), vas-
including MSC [22, 27], pancreatic β-cells [24, 25] cular endothelial growth factor (VEGF), epider-
and neurons [26]. Recent research has demon- mal growth factor (EGF), and hepatocyte growth
strated that PEG-hydrogels are bioinert and factor (HGF) signaling pathways [41]. FGFs,
enhance cell–matrix and cell–cell interactions that such as FGF7, FGF8, and FGF10, are known to
are commonly utilized to maintain survivability of generate MSC-derived SPCs, proving the capac-
sensitive cell types [22, 24, 28, 29]. As cell–cell ity of cell differentiation into c-Kit+ ductal cells
interactions, in particular, play a vital role in SG after transplantation, with the involvement of epi-
cell functions in vitro and during gland develop- thelial sonic hedgehog signaling pathways [8,
ment [30–33], utilization of PEG-hydrogel for SG 42]. Some of these extrinsic and intrinsic factors
cell aggregation into microspheres increases long- are summarized in Table 6.1.
term viability of hydrogel-encapsulated SG cells In a rather unconventional approach as a proof
[34]. The detailed information on these scaffold of concept, BM soup, which consists of the soluble
materials can be found in Chap. 8 of this book. contents of lysed BM cells, was delivered via
Table 6.1 Examples of inducing/promoting factors for target cell differentiation in vitro
106
Expression
Extrinsic Pancreas Liver Salivary glands Functions Ref.
FGFs Dalvi et al. (2009) Si-Tayeb et al. Lombaert et al. β-cell differentiation and cluster formation [8, 42–45]
Hardikar et al. (2003) (2010) (2011) Haploinsufficiency of FGF or its receptor showing severe defects in
Patel et al. the survival and function of SPCs
(2006) Induced hepatic specification when pluripotent stem cells were used
in monolayer culture
EGF Cras-Meneur et al. (2001) Kitade et al. Kashimata et al. High concentration being inhibitory to β-cell differentiation [43, 46–48]
Dalvi et al. (2009) (2013) (2000) Acted as a regulating factor for proliferation and/or differentiation of
liver progenitor cells (LPCs)
Showed important role in SG organogenesis, especially in branching
morphogenesis
Betacellulin Demeterco et al. (2000) Formation of islet-like clusters [49]
Induction of β-cell differentiation
Activin A Demeterco et al. (2000) Increased insulin content [49]
Nicotinamide Chen et al. (2004), Segev et al. Differentiation of stem cells of various origins into iPS cells [50–60]
(2004), Tang et al. (2004), Sun Increased insulin content, DNA content, expression of insulin,
et al. (2007), Sun et al. (2007), glucagon, and somatostatin genes
Wu (2007), Chao et al. (2008),
Gabr et al. (2008), Gao et al.
(2008), Otonkoski et al. (1993),
Gao et al. (2008)
Exendin-4 Tang et al. (2004) Differentiation of murine BM stem cells into iPS cells [52, 55, 57,
Timper et al. (2006) Formation of insulin expression cells generated from adipose 61, 62]
Wu et al. (2007) tissue-derived MSCs
Gabr et al. (2008) Differentiation of BM-MSCs into iPS cells
Aquayo-Mazzucato (2011) Increased insulin release by iPS cells generated from mouse ES cells
HGF Mashima et al. (1996) Onitsuka et al. Differentiation of pancreatic acinar cells into iPS cells [63, 64]
Otonkoski et al. (1996) (2010) Increased number of iPS cells in cultured human islets [45, 65, 66]
Si-Tayeb et al. Induced differentiation into LPCs
(2010)
Ishikawa et al.
(2012)
Gastrin Wang et al. (1993, 1997) Stimulation of islet differentiation and islet growth [67, 68]
Glucose Halban et al. (1987) Low concentration increased insulin content [69]
High concentrations increased β-cell replication
Y.-J. Park and S. Cha
Oncostatin M Kamiya et al. Promoted hepatocyte maturation in vitro by inducing the formation of [70–72]
(1999, 2002) adherent junctions
Matsui et al.
(2002)
Dexamethasone Banas et al. BM-MSC or CD105+ adipose-derived MSC differentiation into [73, 74]
(2007) hepatocyte-like cells
Lee et al. (2004)
Intrinsic Expression
Pancreas Liver Salivary gland Functions Ref.
Ascl3 Arany et al. Characterized as a SG progenitor marker [75]
(2011) Involved in the regeneration of SG cells
C/EBPα/β Courtois et al. Promoted hepatoblast differentiation into mature hepatocyte [76–78]
(1987)
Cereghini et al.
(1988)
Tadashi et al.
(2003)
Foxa2 Gao et al. (2007) Xu et al. (2012) Played a key role in hepatogenesis in early development [79–81]
Gao et al. (2008)
Gata4 Carrasco et al. (2012) Carrasco et al. Highly expressed in early pancreatic budding stage and remained in [82]
(2012) mature acinar cells
Played a key role in liver development during embryogenesis and
regulated Hnf4 expression
Hmg20b Park et al. (2015) Increased expression in cocultured MSCs into SPCs [12]
Hnf1b (Tcf2) Haumaitre et al. (2005) Expressed during pancreatic development [83]
Pancreatic agenesis in E13.5 in Tcf2 deficient mice
Hnf4 Watt et al. Regulated key regulatory genes involved in hepatocyte maturation [84]
6 Directed Cell Differentiation by Inductive Signals in Salivary Gland Regeneration
(2003)
Hnf6 Pierreux et al. (2004) Pierreux et al. Broad expression throughout development and regulate α and acinar [85]
(2004) cells in the liver and pancreas
Isl1 Du et al. (2009) Stimulate essential growth factor for pancreatic development [86]
Initiate expression in mouse pancreatic mesenchyme at E9
MafA Aguayo-Mazzucato et al. (2011) Aguayo- Detected in late developmental stage and involved in stimulation of [61]
Mazzucato mature β-cell differentiation
et al. (2011)
(continued)
107
Table 6.1 (continued)
108
Mist1 Pin et al. (2000) Park et al. Expressed in a wider array of tissues including the acinar cells of SGs [12, 87]
(2015) and the serous-secreting cells found in the stomach and prostate
Defect in mice resulting a loss of correct cellular organization in
pancreas and SGs during embryogenesis
NeuroD1 Imai et al. (2007) Found in all endocrine cell of mature islets [88, 89]
Gu et al. (2010)
Ngn3 Magenheim et al. (2011) Magenheim Required for endocrine cell specification in mice and also initiates [90]
et al. (2011) endocrine commitment in pancreatic and SGECs
Nkx2.2 Papizan et al. (2011) Detected in pancreatic precursor cells and stimulate β-cell [91]
differentiation
Nkx6.1 Taylor et al. (2013) Expressed in early multipotent pancreatic progenitors and have key [92]
role in β-cell differentiation
Mnx1 Flanagan et al. (2014) Identified their expression in early stage of the developing pancreas [93, 94]
Bonnefond et al. (2013) Expressed only in adult β-cells at final developmental stage
Pax6 Hart et al. (2013) Activated in early pancreas and maintained throughout the adult islet [95]
cells
Pdx1 Jennings et al. (2013) Plays a role throughout all phases in development of pancreas and [96–98]
Gu et al. (2002) highly expressed in adult β-cell
Brissova et al. (2005) Decreased expression in both human and rodent model type 2
diabetes islets
Ptf1a Yoshitomi et al. (2004) Park et al. Broad expression in dorsal and ventral pancreatic bud in mice and [12, 99,
Weedon et al. (2014) (2015) their mutation leads to pancreatic agenesis 100]
Highly increased in differentiating MSCs during coculture
Sox2 Lombaert et al. Highly abundant and involved in duct lineage differentiation in the [8]
(2011) SGs
Sox9 Kawaguchi et al. (2013) Mead et al. Mead et al. Essential for the maintaining multipoint progenitor population in [101, 102]
(2013) (2013) mice and have effective role in endodermal cells such as the pancreas,
Kawaguchi lung, liver, SG, and gut
et al. (2013)
Sox17 Kanai-Azuma et al. (2002) Detected for a short time period during development [103]
Necessary in endoderm formation and pancreatic fate
Tcf3 Park et al. Protein and gene expression highly increased in cocultured MSCs [12]
(2015) converting to SPCs
Y.-J. Park and S. Cha
intraglandular delivery or systemic delivery, result- 6.2.2 Intrinsic Factors

ing in restored saliva flow in mice with irradiated (Transcription Factors)
SG [104]. The paracrine effects of MSCs were
reconfirmed in various studies where BM-MSCs Cellular programs regarding proliferation,
exerted the capacity for organ repair by secretion of potency, and cell fate determination can be medi-
cytokines, chemokines, and growth factors [105]. ated by signal transduction events that modulate
In a subsequent study, the same group reported that TFs expression and/or activation. As we previ-
deactivated soup treated with proteinase K diges- ously mentioned, the specification of cell types
tion had no effect on the recovery of secretory under normal development is controlled by
function in irradiated mice at 8 weeks postirradia- extrinsic factors that impose regional characteris-
tion [106]. Protein arrays revealed that native pro- tics on specific progenitor cells at early develop-
teins, especially angiogenesis-related factors such mental stages. Signaling molecules play key
as CD26, FGF, HGF, MMP-8, MMP-9, OPN, PF4, roles by controlling the establishment of distinct
SDF-1, and cytokines IL-1ra and IL-16, in the BM progenitor domains that can be defined by unique
soup were the therapeutic ingredients [106]. The expression profiles of TFs. Some of these TFs
studies once more emphasize the importance of function as more broad region-specifying deter-
extrinsic factors that promote tissue regeneration minants, while others are expressed exclusively
and affirm MSCs as trophic mediators. in individual progenitor domains in which they
Two rare genetic syndromes, aplasia of lacrimal mediate highly selective functions [110].
and salivary glands (ALSG) and lacrimo-auriculo- Many of these factors, but not all, contain a
dento-digital syndrome (Ladd syndrome), in protein structure motif called basic helix-loop-
humans shed light on a growth factor and its recep- helix (bHLH). Proteins of the bHLH superfamily
tor pathway that are essential for progenitor cells to have two highly conserved and functionally dis-
initiate and form SGs. These diseases, caused by tinct domains of the basic domain located at the
mutations resulting in haplo insufficiency of Fgf10 amino-terminal end and of the HLH domain
or its receptor Fgfr2B, are characterized by severe located at the carboxy-terminal end [111]. The
defects in the survival and function of SPCs [107, basic domain binds the TF to DNA at consensus
108]. Embryos do not develop SGs in mice that hexanucleotide sequence known as the E-box,
have both copies of Fgf10 or Fgfr2b deleted and the HLH domain facilitates interactions with
(Fgf10−/− and Fgfr2b−/−). Therefore, SPCs require other protein subunits to form homeodimers or
Fgf10/Fgfr2b signaling to survive and initiate heterodimers. The heterogeneity in the E-box
organogenesis of SGs. It has also been demon- sequence that is recognized and the dimers
strated in vivo that Fgf7, another Fgfr2b ligand, formed by different bHLH proteins determine
has an effect on SPCs [109]. Fgf7 injections before how TFs of the bHLH family control diverse
and after gland irradiation enhanced the number of developmental functions through transcriptional
progenitor cells. As a consequence, a higher num- regulation [112].
ber of progenitor cells remained after radiation, TFs promote lineage specification by cross-
forming more saliva-producing cells that pre- repressive interactions between TFs driving
vented radiation- induced hyposalivation [8]. alternative lineage programs in multipotent pro-
Addressing sequential events in development and genitors and by induction of additional TFs to
differentiation can be challenging due to the fact further execute the differentiation process [113].
that signaling events are tightly associated with Thus, interplay between TFs that act down-
each other and they function redundantly or inter- stream of extracellular stimuli further enhances
mittently at several developmental stages. It is also the complexity of lineage determination. Due to
important to note that the developmental effects such complexity in signal pathways, this chapter
elicited from these common signals can differ mainly explains critical TFs that are known to
greatly according to the developmental stage in be involved in stem cell reprogramming as well
which the signal occurs [8]. as organogenesis toward endodermal organs,
Hnf4a
C/EBPa/b
Gata4 Hepatic
Isl1
MafA Hnf6 progenitors Ptf1a+
NeuroD1 Foxa2 Exocrine
Nkx2.2 Sox9 progenitor
Nkx6.1
Mnx1
Tcf2(Hnf1b)
Pax6 Ptf1a Pancreatic
Endoderm Pdx1 Foxa2 progenitors
cell Sox17 Ngn3
Mist1
Ngn3+
Ascl3 Sox9 Endocrine
Sox2 progenitor
Tcf3
Hmg 20b
Salivary grand
progenitors
Fig. 6.1 Schematic diagram of endodermal cell lineage such as Ptf1a and Pdx1 can drive the cells into pancreatic
differentiation and TFs that are known to be involved in lineage and promote pancreatic exocrine progenitor or
the process. TFs in red are reported to be involved in both endocrine progenitor differentiation in combination with
organs. Studies identified that the expression of some TFs other TFs
such as the liver, pancreas, and the SGs. Some expressed only in the second wave of insulin pos-
of exemplary TFs are listed in Fig. 6.1 and itive cells which ultimately become mature
Tables 6.1 and 6.2. β-cells. MafA is known as a maturation marker
and is crucial for glucose-responsive β-cells by
6.2.2.1 Transcription Factors Involved regulating insulin and glucose transporter 2 [61].
in Pancreas Regeneration Recent studies have found reduction of MafA
Understanding how TFs control early develop- levels in mice and in humans under diabetic con-
ment of pancreatic cells can yield insight into ditions [122]. Type 2 diabetes mellitus islets may
transcriptional regulation of cell fate determina- be significantly related to dysfunctional β-cells
tion and guide protocols for therapeutic stem cell caused by loss of nuclear MafA [123, 124].
reprogramming. TFs that are reported to be NeuroD1 (neurogenic differentiation factor 1)
involved in pancreas cell differentiation and is found in all endocrine cell types of mature
development are listed below and exemplified in islets. Homozygous NeuroD1 mutations are pre-
Fig. 6.1. Whether these pancreatic TFs exert sim- dicted to cause autosomal recessive neonatal dia-
ilar or identical functions in SG cell lineage betes [88]. Without the NeuroD1, β-cells
determination remains unknown. generated are immature with increased glycolytic
Isl1 (ISL LIM homeobox 1) appears to be an gene expression of neuropeptide Y (a hormone
essential growth factor for pancreatic develop- that decreases expression after birth) and elevated
ment in both humans and mice. Isl1 is a pan- basal insulin secretion [89].
endocrine cell marker. When Isl1 is mutated, it Nkx2.2 (NK2 Homeobox 2) is a TF involved
exhibits characteristics of diabetic, impaired islet with β-cell differentiation [125, 126]. Nkx2.2
cell maturation, and reduced postnatal islet mass expression is limited to only pancreatic α- and
expansion [86]. Isl1 is first expressed in wide β-cells along with some portion of pancreatic
range in the pancreatic mesenchyme at E9.0, and precursor cells [91].
expression of Isl1 is maintained in the mature Nkx6.1 (NK6 Homeobox 1) is broadly
hormone-positive endocrine cells [120, 121]. expressed in early multipotent pancreatic pro-
MafA (V-Maf avian musculoaponeurotic genitors [127]. Mice lacking Nkx6.1 have been
fibrosarcoma oncogene homolog A) in mice is shown to have a dramatic reduction in β-cells
detected in late developmental stages and [92]. To maintain the β-cell lineage in conditional
Table 6.2 Examples of inducing/promoting factors for target cell/organ regeneration in animal models
Factors Manipulations Target cell /organ Animal strains Delivery methods Outcomes Ref.
Factors derived from mEES-6 (mouse early SG (SGECs) SMG of 8-week-old female Direct transplantation Reconstituted acinar-like [114]
inducers in coculture ES) cocultured with SCID/Jcl mice or duct like structure
human SG-derived
fibroblast
Ngn3, Pdx1, and Tissue-specific stem Pancreas (β-cell) Rag1−/−, Rag1−/−; NOD Purified adenovirus Ameliorated [115, 116]
MafA cell-like cells directly into the splenic hyperglycemia by
generated by lobe of the dorsal remodeling local
introduction of TFs pancreas vasculature and secreting
insulin (over 20 % of
infected cells positive for
insulin)
Activin A and Treated cells with the Pancreas (β-cell) Streptozotocin-treated neonatal Subcutaneous injection β-cell mass increased by [117]
betacellulin (BTC) factors for 3D culture SD rat 69 % at 2 months
post-injection
Expression of Gata4, Isolated tail fibroblast Liver (Hepatocyte) Fah−/−Rag2−/− mice (liver Direct transplantation iHep cells engrafted into [118]
Hnf1a, Foxa3 & and expressed TFs to failure mice w/ liver sinusoid comprised
inactivation p19Arf induce functional immunodeficiency) between 5 % and 80 %
hepatocyte-like cells of total hepatocytes in
iHep-Fah−/−Rag2−/− liver
and substantially
improved liver functions
Hnf1b & Foxa3 Induced hepatocytic Liver (hepatocyte) Fah−/− mice and Direct transplantation iHepSCs possessed the [119]
6 Directed Cell Differentiation by Inductive Signals in Salivary Gland Regeneration
stem cell (iHepSC) 3,5-diethoxycarbonyl-1,4- capacities of both

from mouse dihydrocollidine (DDC)-fed self-renewal and
embryonic fibroblast mice bipotency of
(mMEF) by TF differentiation into both
transfection hepatocytes and
cholangiocytes
111
Nkx6.1 mutants, specifying endocrine precursors expressed, but expression is restricted in adult
are required by marker protein expression, such β-cells. Pdx1 also regulates the expression of
as Ngn3, Pax4, and Pdx1. In mature pancreas, Ins1 and MafA [96]. Besides its main role in
Nkx6.1 plays a key role in identification of β-cells β-cells, reports about mouse lineage tracing and
[92]. Nkx6.1 expression is severely reduced in Pdx1 expression in mouse acinar tissue demon-
diabetic and obese mice [128]. strated that Pdx1+ cells mark progenitors of all
Mnx1 (motor neuron and pancreas homeobox the mature pancreatic cell types including endo-
1) is expressed in early stages of the developing crine, acinar, and ductal cells [97]. A targeted
pancreas and then decreased to lower levels when disruption causing homozygous inactivating
entering into gestation. Mouse Mnx1 expression mutation in the Pdx1/Ifp1 resulted in pancreatic
was found in the E9.5 endoderm with detailed agenesis [98]. An autosomal recessive mutation
expression analysis [128–130]. The expression in the Pdx1 locus is known to cause permanent
was gradually restricted to the Pax6+ endocrine neonatal diabetes [134]. Pdx1 levels were
population by E15.5. At final stage, it is expressed decreased in both human model and rodent model
only in adult β-cells. A recent study reported that islets with type 2 diabetes mellitus [135].
homozygous mutation within the DNA-binding Sox17 (SRY(Sex determining region Y)-box
homeodomain of Mnx1 caused permanent neo- 17) is a TF for expression of the high mobility
natal diabetes [93]. Similar to the null mouse group (HMG) box. It is only observed for a short
model, patients with a defective Mnx1 gene did time period and excluded during mouse pancre-
not show obvious exocrine defects, but β-cell atic development [136]. Studies found that Sox17
numbers were reduced [94]. expression is necessary in early stages of endo-
Tcf2 (transcription factor 2): During early derm formation, but it later represses the pancre-
stages of embryogenesis, a high level of Tcf2 atic fate [103].
(a.k.a. Hnf1b, hepatocyte nuclear factor 1-β)
expression persists, and homozygous mutations 6.2.2.2 Transcription Factors Involved
cause diabetes [83]. Heterozygous loss-of- in Liver Regeneration
function Tcf2 mutations result in diabetes in Combinatorial protein interactions among liver-
humans, but only homozygous mutations pro- specific TFs are required to achieve synergistic
duced diabetes in mice [83]. This could be due to cellular stimulation of tissue-specific gene
a potentiated single wave of human endocrine expression for liver regeneration. Moreover,
differentiation versus the two phases observed in recent advances in hepatocyte and β-cell transdif-
rodents, rendering these human cells more sensi- ferentiation research have provided valuable
tive to Tcf2 dosage. Tcf2-deficient mice exhibit insight into how to regenerate and restore normal
pancreatic agenesis by E13.5, suggesting that the functions of liver and pancreas under pathologi-
role of Tcf2 in pancreatic development is evolu- cal conditions. The definition and the applica-
tionarily conserved [83]. tions of transdifferentiation are discussed in Sect.
Pax6 (paired box 6) is activated in early pan- 6.3. Hepatocyte-specific TFs that are involved in
creas and maintained throughout the adult islet generating functional hepatocytes are listed
cells [131–133]. Pax6 null mice die at birth from herein, which would provide an invaluable
brain abnormalities, but embryos have other side resource to understand the roles of TFs in cells
effects such as reduced islet cell numbers, within the same developmental lineage.
impaired hormone synthesis, and defective islet Hnf4α (hepatocyte nuclear factor 4α) is capa-
morphogenesis, which indicates a role in alloca- ble of modulating hepatocyte gene expression in
tion and differentiation of endocrine cells [95]. differentiating hepatoma cells [84, 137]. Gene
Pdx1 (pancreatic and duodenal homeobox 1), knockout studies of Hnf4α, whether in fetuses or
which is also known as insulin promoter factor 1, in adults, have been found to disrupt expression
plays a role throughout all phases in development of a large number of genes involved in most
of pancreas. In early stages, Pdx1 is highly aspects of mature hepatocyte function [137, 138].
These functions include control of energy metab- Sox2 (SRY(sex-determining Y)-box 2) is

olism, xenobiotic detoxification, bile acid synthe- essential in ES cell self-renewal [152]. However,
sis, and serum protein production [84]. in SG development, it is differentially expressed
C/EBPα/β (CAAT/enhancer-binding proteins within the K5+ population [153]. Sox2 is
α and β) is a basic leucine zipper TF, which plays expressed by a subpopulation of K5+ SPCs and is
an important role in liver function and develop- highly abundant in the committed K5+K19+ and
ment. C/EBPα is enriched in the liver, governing K5−K19+ duct cells in the SG. This suggests that
expression of liver-specific genes, and downregu- their potential to self-renewal, driven by Sox2,
lated after partial hepatectomy [139, 140]. may be present even as K5+ cells differentiate
Hepatocytes of C/EBPα-deficient mice have along the duct lineage [8].
increased proliferative potential [141, 142]. C/ Tcf3 (transcription factor 3) is the most abun-
EBPβ is upregulated after partial hepatectomy dant Tcf/Lef member in mouse ES cells [154]. It
and also upregulated by TNFα and IL-6 in stel- was reported that heterodimers between Tcf3 and
late cells [143–145]. These findings indicate that tissue-specific bHLH proteins play major roles in
C/EBPα and C/EBPβ are major players in liver determining tissue-specific cell fate during
regeneration. Hepatocytes of mice deficient in C/ embryogenesis [155]. Tcf3 is also known to be
EBPα show a pseudoglandular structure [146]. closely involved in Wnt/β-catenin signing to con-
Hepatocytes lining the structure have potential trol self-renewal and regulates the lineage differ-
for differentiation into both hepatocytes and bile entiation of ES cells [156–158]. Interestingly,
duct epithelial cells, suggesting that C/EBPα pro- Tcf3-β-catenin interaction may indirectly affect
motes differentiation of hepatoblasts to mature submandibular SG during mouse embryogenesis.
hepatocytes [147, 148]. Furthermore, vascular integrity defects in organs
such as the submandibular glands and liver were
6.2.2.3 Transcription Factors Involved reported in the Tcf3 knock-in mutation model,
in Salivary Gland Regeneration which specifically lacks Tcf3-β-catenin interac-
In contrast to TFs investigated in the pancreas tion [159]. However, the exact functions of
and liver, critical roles of SG TFs for cell differ- Tcf3 in the SGs during development and stem
entiation, lineage commitment, and fate determi- cell differentiation remain largely unknown.
nation toward secretory acinar or duct cells are Hmg20b (high mobility group 20B) is known
understudied and poorly understood. In this sec- to be expressed in various tissues. Many research-
tion, we will describe SG TFs, typically identi- ers suggest that breast cancer susceptibility gene
fied as SSPC marker proteins during mouth SG 2 and Hmg20b complex may have a role in cell
development, along with a brief introduction of cycle regulation and affect cell fate determination
TFs that our group reported following the pro- [160]. Previously, our lab identified that Hmg20b
teomic analyses of mouse MSC-derived SPCs in and Tcf3 gene and protein expression was upreg-
coculture. ulated as mBM-MSCs transdifferentiated into
Ascl3 (achaete-scute family bHLH transcrip- SPC in a coculture system [12]. Their exact roles
tion factor 3, a.k.a. Sgn1) was reported as a marker in MSC transdifferentiation are currently under
of SPCs for both acinar and ductal cells in mouse investigation.
SGs [149] and as a determinant of ductal cell lin-
eage in mice [150]. The Ascl3 knockout mouse 6.2.2.4 Transcription Factors Involved
model showed the complexity of SG maintenance in the Pancreas, Liver,
and regeneration. Ascl3 is considered to be an and Salivary Gland
attractive molecule for induction of SECs as it has Regeneration
been found to be involved in the regeneration of Ptf1α (pancreas transcription factor 1α, pancreas
acinar, ductal, and myoepithelial salivary cells and SG) is better characterized in mice with
in vitro [75] and known to be expressed in ductal broad expression in dorsal and ventral pancreatic
neonatal progenitor cells [151]. buds, as the name indicates [161]. Ptf1α is known
to be an exocrine lineage determinant in the pan- Foxa2 (forkhead box A2, pancreas and liver)
creas and is predominantly detected in pancreatic is a TF that is expressed throughout the definitive
acinar cells [99]. Ptf1α locus mutation activates endoderm, which persists into adulthood in endo-
autosomal recessive permanent neonatal diabetes dermal derivatives such as the liver, pancreas,
which requires insulin for survival. Similar lung, and thyroid during mouse development. In
behavior has been reported in Ptf1a−/− mice, pancreas development, Foxa2 is a major upstream
which die postnatally with impaired pancreatic regulator of Pdx1 and continues to be active in all
and major SG development. A recent study mature pancreatic cell types of mice and humans
reported that human Ptf1α enhancer mutation [79]. Foxa2 deficiency in a mouse model resulted
leads to pancreatic agenesis [100]. Mutant phe- in the absence of mature α-cells and a reduction
notypes of human and mouse support an evolu- of Pdx1 expression and β-cell differentiation
tionarily conserved role during early pancreatic [165, 166]. Given that Foxa1 is a close homolog
formation. Interestingly, our recent study deter- to Foxa2 and contains an identical DNA-binding
mined that Ptf1α expression was highly increased domain, it is also presumed to control pancreas
in transdifferentiating BM-MSC in coculture development [79]. Moreover, the early activation
although the expression of Ptf1 α in the SGs was of the Foxa2 genes in the hepatogenic region of
never reported before [12]. Nomenclature for this the foregut endoderm, combined with the abun-
molecule may need to be revised once its broader dance of liver-specific Foxa2 target genes, has
role than originally anticipated in cell fate deter- been interpreted as evidence that Foxa2 genes
mination in SG regeneration is explored and play a key role in regulating hepatogenesis [80,
confirmed. 137, 167]. The early lethal phenotype observed in
Gata4 (GATA binding protein 4, pancreas and Foxa2−/− embryos, owing mostly to node and
liver) is expressed during the early pancreatic notochord defects, precluded analysis of Foxa2 at
budding stage but later dramatically decreased in later stages of development, thus necessitating
expression in the pancreatic progenitors and only the derivation of mice harboring a conditional
remained in mature acinar cells [82]. Although Foxa2 allele to directly assess its role during liver
Gata4 mutations are known to be associated with development [168].
congenital heart defects, only the pancreatic phe- Ngn3 (neurogenin 3, pancreas and SG) is
notype is documented with mouse models [162, required for endocrine cell specification in mice
163]. This leads to a belief that there may be [90]. It also initiates endocrine commitment in
another Gata TF in human pancreas [162, 163]. pancreatic and/or SG epithelial cells. Isl1,
Moreover, as briefly shown in Fig. 6.1, the TF NeuroD1, MafB, Nkx2.2, and Pax6 are islet-
plays a key role in the liver development. Each enriched factors that activate downstream of
factor in Fig. 6.1 is important for the expansion of Ngn3 in mice, and these factors are integrated in
the liver bud during embryogenesis but serves late endocrine cell differentiation [169]. Ngn3
redundant functions in hepatic specification. null mutation is a rare mutation with permanent
Based on mouse studies, Gata4 expression is first neonatal diabetes with histologically detectable
detected in the ventral foregut endoderm and car- islets [170]. Ngn3−/− mice completely lack endo-
diac mesoderm at E8.0. Gata4 is required for the crine cells and thus develop diabetes and die only
full expression of select hepatic genes, including few days after birth [171].
albumin and Hnf4 [164]. Mist1 (muscle, intestine, and stomach expres-
Hnf6 (hepatocyte nuclear factor 6, pancreas sion-1, a.k.a. Bhlha15 for basic helix-loop-helix
and liver): mRNA analysis of Hnf6 demonstrated family member A15, pancreas and SG) is a known
that mouse Hnf6 expression at E8.5 has broad bHLH TF with acinar cell-specific expression [87,
expression throughout development and directs 172]. It acts as a regulator of differentiation and
endocrine allocation until before the birth when it morphogenesis of exocrine cells by negative regu-
is restricted by α-cells and acinar cells in the liver lation of bHLH-mediated transcription through an
and pancreas [85]. NH2-terminus repressor domain [87]. Mist1 gene
expression is observed in a wider array of tissues adult stem cells not only differentiated into antic-
including the acinar cells of SG and the serous- ipated and committed lineages but also differenti-
secreting cells found in the stomach, prostate, and ated into cell types beyond the expected lineage
seminal vesicles [173]. An essential function that of the respective stem cells [183]. This plasticity
is shared by all Mist1-positive cells is regulated can be seen in many examples of stem cell
exocytosis, which involves the temporary storage research, including the ones described in Sect.
of zymogen granules at the cell’s apical surface 6.3.2. We use this broadened definition in this
and the establishment of specific signaling path- chapter to cover fascinating studies from the
ways through which external cues induce regu- organs of interest.
lated secretion [87]. Based on knockout studies, This type of conversion can be induced exper-
the primary outcome of its expression defect imentally, or it can normally occur in tissues that
reveals a loss of correct cellular organization in arise from the neighboring regions of the devel-
pancreas and SGs [174, 175]. oping embryo [178]. This ability to interconvert
Sox9 (SRY (sex-determining region Y)-Box9, suggests that the two tissues are related at the
pancreas, liver, and SG) is located in Pdx1+ cells molecular level, sharing some common TFs, but
in early stages of pancreas development [96, 162, in addition there is one or two that are different.
176, 177]. Later, it is excluded from mature Therefore, adjacent tissues presumably differ in
endocrine cells. Sox9 is essential for maintaining the expression of one or a few key TFs [184]. As
the multipoint progenitor population in mice for the experimentally induced conversion, deliv-
[101]. Deletion of Sox9 resulted in islet hypopla- ery of TFs via viral vectors altered cell fate by
sia by unsuccessful maintenance of endocrine transdifferentiating cells into cell types of inter-
progenitors. In Sox knockout mice studies, Sox9 est [101]. For instance, Sox2 expression by lenti-
has been demonstrated to have an effective role in virus in the mouse brain directly converted
endodermal cells such as the pancreas, lung, SG, astrocytes into neuroblast [185]. Human fibro-
kidney, gut, and liver [102]. blasts differentiating into cardiac cells by forced
expression of cardiac TFs with muscle-specific
microRNAs would be another example [102].
6.3 Transdifferentiation Transdifferentiation may occur either directly or
through a de-differentiation step before cells re-
6.3.1 Definition differentiate to a new mature phenotype [186].
The term “transdifferentiation” first surfaced to

describe the conversion of cuticle-producing 6.3.2 Examples
cells into salt-secreting cells in the silkmoth dur-
ing metamorphosis from the larval to the adult Tissue injury can lead to the appearance of mul-
moth [178]. Subsequently, Eguchi and Okada tipotent stem cells for the liver and the pancreas
elegantly demonstrated the switch of pigmented [187, 188]. Dabeva et al. reported that pancre-
epithelial cells to lens fibers in their in vitro atic epithelial progenitor cells isolated from the
clonal cell culture system [179], setting the stan- copper-deficient diet rat pancreas can differenti-
dard for determining a true transdifferentiation ate into hepatocytes when they are transplanted
event [180, 181]. A review by Eberhard and Tosh into the liver [189]. Recent studies have shown
explicitly defines “transdifferentiation” as in that there is a tissue stem cell in the pancreas
“nonstem cell” [182], supporting the well- that can differentiate into hepatocytes in addi-
accepted definition of the conversion of one spe- tion to pancreatic cell types [190–192]. Zulewski
cialized cell type into another without reversion et al. reported that multipotent stem cells from
to pluripotent cells or progenitor cells [179]. rat and human islets differentiated into hepatic
However, in recent years this traditional defini- cell types in culture when the cell density
tion has been broadened by the evidence that increased [193]. In other experiments, a cell line
Intercalated Hepatic
ductal cells cells
?
Salivary grand
cells Pdx1
C/EBPβ
Dexamethasone Ngn3
Endoderm
cell
?
SG progenitor cells
from the duct
Pancreatic
cells
Fig. 6.2 Schematic diagram of transdifferentiation among pancreatic, hepatic, and salivary gland cells and TFs involved
from pancreatic acinar cells differentiated into seems to play a key role in the hepatic-pancreatic
hepatocytes in the presence of dexamethasone cell fate conversion [196]. In mice, Pdx1 expres-
(Fig. 6.2) [194]. Thus, resident stem cells in the sion is known to be critical in the conversion of
pancreas or the liver have demonstrated their hepatocytes into pancreatic β-cell-like cells that
capacity to transdifferentiate into a different secret insulin [197]. Interestingly, Pdx1 and
type of cells in other organs. Ngn3 are known to synergistically induce expres-
Ptf1α, the bHLH TF, was first described in sion of β-cell factors and insulin biosynthesis in
exocrine-specific pancreas development [99] as the liver and drastically ameliorated glucose tol-
mentioned earlier. Recent results have expanded erance in mouse [198].
the role of Ptf1α, with involvement in both exo- The SGs are known to be derived from the
crine and endocrine cell lineages in murine and endoderm and the ectoderm (parotid glands) that
zebrafish models [161]. Kawaguchi et al. demon- participate in organogenesis, although the origins
strated conversion of pancreatic progenitors into of these glands are still controversial [188, 199–
duodenal epithelium through the inactivation of 201]. SGs consist of many cell types such as
Ptf1α [195], thereby suggesting a close develop- three different parenchymal cell types found in
mental relationship between intestine and pan- mice: (1) acinar, (2) intercalated duct and granu-
creas similar to the liver and pancreas. Also, it lar intercalated duct, and (3) granular and striated
provides a possible route for creating new pan- duct cells. The relative proportions of these three
creatic cells following the overexpression of cell types are 43 %, 18 %, and 39 %, respectively
Ptf1α in other cell types such as duodenal epithe- [202]. Intercalated ducts are small ducts connect-
lial cells. It will be interesting to examine if the ing the terminal acini and striated duct, and they
switch between hepatocytes and pancreatic cells are known to contain the tissue stem cells or pro-
is also feasible with the modulation of Ptf1α. genitor cells [202]. Studies have suggested that
There have been reports on cell transdifferentia- the SG might contain stem cells that can differen-
tion of hepatocytes into pancreatic β-cells involv- tiate into the cell types of other endodermal
ing several key pancreatic TFs. Of those, Pdx1 organs, such as the liver and pancreas [199].
Furthermore, the features of SG resemble the Currently, active investigation is underway to

pancreatic exocrine system. Transplantation of define SSPCs and investigate their roles in SG
intercalated ductal cells prepared from damaged regeneration. For instance, Ascl3 is essential for
SG led to identification of epithelium-like cells the determination of cell fate, development, and
that settled in the oval cell response area of the rat differentiation of numerous tissues [75, 149].
liver [201]. Moreover, progenitor cells isolated Transplanted Ascl3-expressing cells were able to
from the duct-ligated SG differentiate into pan- induce differentiation and tissue repair in SGs,
creatic cell types characterized by CD44 expres- indicating that these cells are potent inducers of
sion that is specific for the developing pancreas SG regeneration [153]. Furthermore, Ascl3 was
[201], implying a capacity of transdifferentiation found to be expressed in ductal cells, and they
that SPCs might possess. We have not encoun- contribute to the maintenance of mature SG tis-
tered any studies yet that demonstrated genera- sues [149], pointing to a notion that Ascl3+ cells
tion of SECs from hepatocytes, pancreatic may be a subpopulation of SSPCs.
β-cells, or their progenitors (Fig. 6.2). Other studies demonstrated that cells located
in the striated ducts of the SGs express other stem
cell markers (i.e., CD24, CD49f, CD133, and
6.4 alivary Gland Stem Cells/
S c-Kit+) [205]. Particularly, c-Kit was definitively
Progenitor Cells established as a stem cell marker and therefore
gained the highest priority. However, flow cyto-
Stem cell studies in SG regeneration in recent metric analysis of cells obtained from subman-
years have focused on SSPCs. As explained ear- dibular glands indicated that only a small number
lier, the inability to form new acinar cells in the of salivary cells expressed c-Kit [205]. These
SGs with irradiation or SjS is associated with loss results suggest that c-Kit may not be a practical
of progenitors and/or unfavorable microenviron- marker for stem cell isolation from SGs, as it
ment in the glands. Irradiation-surviving progeni- would not yield workable levels of stem cells. In
tors in vivo are no longer stimulated to form addition, structural protein K5+, which is a cyto-
acinar cells because the loss of parasympathetic keratin to form cytoskeleton intermediate fila-
function and innervation prevents proper regen- ments, has been established as a marker of
eration [153]. Current treatments for hyposaliva- progenitor cells [149, 153]. Initially, K5+ has
tion after radiotherapy include administration of been established as a marker for progenitor cells
saliva substitutes, secretagogues, or palliative in tracheal and lung epithelial cells. Additionally,
approaches. Researchers have proposed an ideal K5+ cells express Sox2, a TF involved in the self-
therapeutic approach involving SSPCs for treat- renewal of stem cells [206]. However, K5+ cells
ing irradiated patients, which requires isolating make up a very small percentage (5–9 %) of Sox2
SSPCs prior to radiation therapy, expanding the cells. These studies also suggest that K5 may not
cells in vitro, and then transplanting them back be a practical marker to isolate SPCs either [207,
into the patients [203]. SSPCs are characterized 208]. Recently, a long-term in vitro expansion of
based on their topographical position in the gland, SG stem cells driven by Wnt signals was reported
expression of TFs and specific marker proteins in a study [209], whose practicality is forthcom-
that are known to be found on stem cells in other ing to be determined.
systems, or unique stem cell/progenitor proper- α6β1 integrin has been found to be a marker
ties. SSPCs are also considered as an important for SPCs in rats. Interestingly, this marker was
source for constructing an artificial SG as they are only expressed after duct ligation [201]. These
known to be involved in the regeneration of aci- α6β1 integrin-expressing cells were used to
nar, ductal, and myoepithelial salivary cells establish an immortalized cell line of rat SPCs
in vitro [8]. In addition, an emerging concept in [210]. This cell line is able to differentiate into
the field is that multiple glandular progenitors both acinar-like and ductal-like structures and
may exist with overlapping functions [204]. has the ability to regulate transdifferentiation
when grown on Matrigel®-based 3D scaffolds. d iabetes patients is not sufficient to mimic the
However, cells grown under these conditions dis- normal function of β-cells. Consequently, diabe-
play uncontrolled growth. Further studies are tes mellitus often progresses and can lead to
needed to determine whether acinar-like and major chronic complications and morbidity
ductal-
like structures generated from this cell [212]. New studies indicate that it may be possi-
line are capable of responding to salivary secre- ble to direct the differentiation of stem cells from
tory agonists. Additionally, the mechanisms of human in ex vivo cell culture to form insulin-
uncontrolled cell growth need to be well under- producing cells for transplantation. Strategies
stood and regulated before these structures can be involving pluripotent stem cells appear to have
used for transplantation in vivo. the highest translational potential [213], allowing
sufficient numbers of pancreatic progenitor cells
with a potential to differentiate into β-cells.
6.5 pplications of Stem Cells
A Differentiated cells within the adult pancreas
for In Vivo Applications retain sufficient plasticity to transdifferentiate
into β-cells as well. The most promising exam-
Predictable and precise controlling cell prolifera- ples are α-cells and acinar exocrine cells that
tion and differentiation requires comprehensive transdifferentiated into β-cells with the introduc-
profiling of the molecular and genetic signals that tion of TFs [115]. However, stem cell-based ther-
regulate cell division and specification. While apeutic approaches have some limitations when it
recent developments with stem cells suggest spe- comes to clinical applications, such as teratoma
cific roles of differentiation factors, techniques formation, ethical issues related to ES cells, and
must be devised to introduce these factors safely safety issues of gene expression associated with
into the cells and induce maximum outcomes iPS. Furthermore, the future success of
without adverse side effects. Stem cells, such as ES-derived β-cell transplantation or regeneration
ES cells, iPS cells, MSC, and tissue adult stem of endogenous β-cells depends on effective pre-
cell, are currently being used to screen and vention of infiltration of autoreactive immune
develop new drugs in preclinical studies [211]. cells and/or alloimmune rejection [213].
To screen drugs effectively, the testing conditions
must be identical for proper comparisons of
drugs, and precise controlling of stem cell fate is 6.5.2 Liver Regeneration
required for consistency and reproducibility. We
summarize published in vivo studies with induc- Most liver diseases lead to hepatic dysfunction
tive signals in the pancreas, liver, and SG regen- with organ failure. Liver transplantation is the
eration in Table 6.2, although the field is still in best curative therapy, but it has some limitations,
its infancy with a diverse range of preliminary such as donor shortage, possibility of rejection,
approaches. A brief discussion of stem cell use and maintenance of immunosuppression [214].
with or without inductive signals is provided New therapies have been actively searched for
herein. over several decades, primarily in the form of
artificial liver-support devices and hepatocyte
transplantation, but both of these modalities
6.5.1 Pancreas Regeneration remain experimental. For this reason, scientific
interests have switched to the use of stem cells to
One of the major pancreatic diseases, diabetes regenerate the liver. Numerous stem cell types
mellitus, is a metabolic disorder caused by an have been used for the differentiation of hepato-
insufficient number of insulin-producing β-cells. cytes both in vivo and in vitro [215]. Plasticity in
Theoretically, replenishment of β-cells by cell adult MSC repopulated liver cells in a liver injury
transplantation can restore normal metabolic animal model, which appeared to improve liver
control. Exogenous insulin administration in function. For instance, multipotent adult progeni-
tor cells (MAPCs) derived from human, mouse, In addition, significant challenges still exist
and rat postnatal BM can differentiate into before these cells can be used in humans. These
hepatocyte-like cells in vitro [216]. The differen- challenges include the lack of consensus about
tiated cells expressed hepatic markers such as the immunophenotype of liver progenitor cells,
Hnf-3β, Gata4, Afp, Alb, and CK-18, and uncertainty about the physiological roles of
obtained functional characteristics of hepatocytes reported candidate stem/progenitor cells, the
as indicated by p450 activity, LDL uptake, ALB practicality of obtaining sufficient quantity of
secretion, urea secretion, and glycogen storing cells for clinical use, and concerns over ethics,
[216]. Historically, MAPCs had been identified long-term efficacy, and safety [228]. Nonetheless,
as a subpopulation of MSCs [217]. Human cells from BM are increasingly recognized as
umbilical cord blood-derived MSCs (UCB- being major therapeutic players for liver fibrosis
MSCs) were also proven to have potential for and regeneration, indicating that the BM posi-
hepatocyte-like cell differentiation in vitro by the tively contributes to liver fibrosis [229].
treatment of special cytokine mixtures [218].
Recently, MSCs were isolated from human thigh
adipose tissue and induced to differentiate into 6.5.3 Salivary Gland Regeneration
hepatocyte-like cells using growth factors, cyto-
kines, and hormones [219–221]. These Two major approaches for xerostomia include (1)
hepatocyte- like cells expressed hepatocyte- utilization of combinations of cells, biomaterials,
specific genes and exhibited hepatocyte functions and biochemical cues for tissue engineering and
[222]. Moreover, a study examined the impact of (2) utilization of cell-based therapy for SG regen-
the coculture with adult liver cells or fetal liver eration. Studies have investigated the role of
cells on hepatic differentiation of rat BM-MSCs MSCs as a therapeutic option for treatment of SjS
[223] and found that liver-specific gene expres- [80, 230]. Investigators used NOD mice with a
sion was induced when treated with gadolinium SjS-like disease to investigate the effect of MSCs
chloride [224]. Although all cells differentiated in reducing lymphocytic infiltrates in the SG and
into hepatocyte-like cells, the cocultured ones restoring salivary function. Authors found that
expressed more hepatic gene markers and exhib- intravenous injection of MSCs reduced lympho-
ited higher metabolic function (P450 activity) cytic infiltrate and inflammation in the SG com-
[216]. These pioneering research studies led to pared to untreated controls, including a tenfold
the application of human BM-derived cells xeno- decrease in the inflammatory cytokine TNF-α.
grafted to damaged liver of Sprague Dawley rats MSC injection also preserved the saliva flow rate
by allylalcohol, resulting in differentiation of over the 14-week posttreatment period; more-
hepatocyte-like cells [225]. over, when MSCs were administered in
Other groups reported plasticity of MSCs conjunction with complete Freund’s adjuvant
after human adipose tissue MSCs were trans- (CFA), the SG regenerative potential increased.
planted into mice with liver damage for hepato- These findings indicate that direct injection of
cyte differentiation [222]. Following the study, MSC for therapy alone reduced inflammation,
human MSCs were transplanted into preimmune but there was additional tissue repair and regen-
fetal sheep, providing further evidence for feasi- eration when administered in conjunction with
bility [226]. As a result, the animals exhibited a CFA [230].
widespread distribution of human hepatocyte- SSPCs are known to be located in intercalated
like cells throughout the liver parenchyma, indi- ducts and exocrine ducts [203, 231, 232]. Stem
cating that MSCs are a valuable source of cells cells isolated from the human SG have revealed
for liver repair and regeneration [227]. Although a certain capacity for in vivo recovery of SG
the hepatogenic capacity of MSCs seems to be function in irradiated rat SGs by differentiating
strongly established, the mechanism by which into amylase-expressing cells [233]. Sixty days
MSCs restore liver functionality is still not clear. after human SSPC were transplanted into the
irradiated glands of rats, the average saliva flow 6.6 onclusion and Future
C
rate of the human SG stem cell-treated group Directions
was twice that of the PBS-treated irradiated
group but was still lower than the undamaged Xerostomia due to SG hypofunction has a severe
group [233]. By using a floating sphere culture, impact on the oral health of patients with SjS and
further in vitro characterization of submandibu- radiation therapy. The regeneration of functional
lar-derived SG stem cells showed that cultured SG tissue is an important therapeutic goal for the
spheres contained acinar and ductal cells charac- field of regenerative medicine and will likely
terized by specific cell marker expression [234]. involve cell therapies such as stem cells and/or
Interestingly, acinar cells mostly disappeared by tissue engineering. Despite research endeavors,
the third day but reappeared within the existing critical information on signaling pathways, mas-
ductal spheres by the fifth day in culture. By day ter regulators, and molecular mechanisms that
ten, acinar cells dominated sphere composition direct stem cell differentiation and tissue engi-
and amylase expression. Intraglandular trans- neering is lacking in the SG research field, com-
plantation of 3-day cultured salispheres into irra- pared to information available in other organ
diated mice resulted in the formation of ductal regeneration. Nevertheless, researchers have iso-
structures near the injection site. There was lated SSPCs from rodent and human SGs and
increased acinar cell surface area in salisphere- demonstrated that SPCs can differentiate into
treated mice compared to the untreated group. both salivary epithelial-like cells and endodermal
Ninety days after irradiation, saliva production progeny, such as pancreatic β-cells and hepato-
in salisphere-treated mice was higher than the cytes [33]. Whether SG cells can be regenerated
untreated counterparts and correlated strongly with hepatic or pancreatic progenitor cells trans-
with acinar surface area [234]. A major limita- planted into the SGs or whether hepatic or pan-
tion for SSPC therapy in the treatment of creatic TFs can contribute to directed
radiation-induced hyposalivation continues to be differentiation of SPCs or SECs is completely
the difficulty with isolating autologous stem unknown. Considering that most TFs studied in
cells from a severely injured gland unless the those organs are critical in determining the lin-
isolation and purification are carried out prior to eage of endocrine cells for enzyme secretion, SG
radiation. As mentioned earlier, a recent research may require a set of distinct TFs that define and
progress by Coppes et al. reported the expansion drive exocrine cells for more organized structural
of SSPCs ex vivo via the stimulation of Wnt sig- differentiation for proper secretion.
naling [209]. Several reports have suggested that the con-
A coculture system of mouse ES cells and version of pancreatic progenitor cells toward the
human SG fibroblasts was developed to facilitate hepatocyte endocrine fate (reciprocal transdiffer-
differentiation of mouse ES cells to SG stem cells entiation) is achievable [213]. Moreover, forced
[114]. After 1 week in coculture, a significant expression of pancreatic TFs elicits insulin
change in cell morphology was found, and expression in the liver and corrects experimental
RT-PCR results showed a sudden appearance of diabetes [235]. Taken together, these findings
amylase and bFGF. These GFP-expressing SG suggest that the adult organs contain cells with
cells were transplanted into SG of normal mice, epigenetic memory of their common embryonic
and histology was performed after 1 month. H&E origin. Based on these studies, SG regeneration
and PAS staining of SG stem cell-treated mice or progenitor cell differentiation might be closely
showed normal formation of ductal and acinar related with pancreas or liver organogenesis and
structures. Even though this method is not lim- stem cell regeneration. Therefore, identification
ited by the need for autologous stem cells from a of intrinsic regulatory factors and external factors
radiation-damaged gland, it is limited by the ethi- for stem differentiation/transdifferentiation and
cal concerns over acquisition of ES cells in clini- the sequential and timely introduction of those
cal settings. factors via viral or nonviral methods, as similar
approaches have shown in pancreas and liver Comprehensive and multidisciplinary research
regeneration, may benefit SG stem cell differen- endeavor focusing on the epigenetic and molecu-
tiation and regeneration. This will ultimately lar regulation of cell fate determination and trans-
allow translational applications for some tissues differentiation of stem cells and differentiated
that normally lack regeneration capacity. cells will allow orchestrated and/or directed
Similarly, mapping and profiling of those factors reprogramming of cells for SG regeneration.
and understanding their critical roles in differen- Harnessing TF-directed differentiation to classi-
tiation will enable us to control cell fate to devise cal approaches of stem cell reprogramming will
therapeutic interventions for xerostomia. expedite clinical translation of cell replacement
As for clinical translation of stem cell therapy, therapy.
safety is the major issue, given that a small num-
ber of contaminating undifferentiated cells could Acknowledgments This work is supported by the NIH/
remain after cell purification and potentially cause NIDCR grants DE019644 and DE025726 (S.C.). The
authors thank Dr. Kathleen M. Berg for critical
teratoma formation. Human iPS cells do not pose proofreading.
ethical concerns unlike ES cells but do present a
potential danger associated with DNA transfer of
cancer-associated genes [213]. Generation of
individualized ES or iPS cell lines to avoid ethical References
issues will be expensive, given the relatively low
efficiency of these procedures. Storing a wide 1. Voulgarelis M, Tzioufas AG. Pathogenetic mecha-
range of human ES and iPS cell lines matched to nisms in the initiation and perpetuation of Sjogren’s
syndrome. Nat Rev Rheumatol. 2010;6:529–37.
the most common HLA haplotypes might be a 2. Seiwert TY, Cohen EE, Haraf DJ, Stenson K, Blair
better option than the generation of patient-spe- EA, et al. A phase I trial of docetaxel based induc-
cific cell lines [236]. Allorejection is another issue tion and concomitant chemotherapy in patients with
to consider, although both issues could theoreti- locally advanced head and neck cancer. Cancer
Invest. 2007;25:435–44.
cally be solved by immunoisolation of the graft 3. Malouf JG, Aragon C, Henson BS, Eisbruch A, Ship
[237]. Immunoisolation could be achieved by JA. Influence of parotid-sparing radiotherapy on
macroencapsulation with or without a biopolymer xerostomia in head and neck cancer patients. Cancer
in devices that prevent the passage of stem cells Detect Prev. 2003;27:305–10.
4. Trotti A, Bellm LA, Epstein JB, Frame D, Fuchs
into the recipient and immune cells into the graft HJ, et al. Mucositis incidence, severity and associ-
[238]. Alternatively, MSCs lessen immunoreac- ated outcomes in patients with head and neck cancer
tivity as they express the human leukocyte antigen receiving radiotherapy with or without chemother-
(HLA)-G, which is a nonclassical HLA class I apy: a systematic literature review. Radiother Oncol.
2003;66:253–62.
molecule that mediates the suppressive effect of 5. Dirix P, Nuyts S, Van den Bogaert W. Radiation-
MSCs through the induction and proliferation of induced xerostomia in patients with head
regulatory T cells, along with other immunosup- and neck cancer: a literature review. Cancer.
pressive properties, which adds to the benefits of 2006;107:2525–34.
6. Baum BJ. Prospects for re-engineering salivary
MSCs for autoimmune SjS [239]. In addition, glands. Adv Dent Res. 2000;14:84–8.
HLA compatibility between a MSC donor and a 7. Feng J, van der Zwaag M, Stokman MA, van Os R,
recipient is not a major concern due to the lack of Coppes RP. Isolation and characterization of human
HLA-DR surface expression [240]. salivary gland cells for stem cell transplantation to
reduce radiation-induced hyposalivation. Radiother
In conclusion, stem cell approaches hold great Oncol. 2009;92:466–71.
promise for future clinical trials for xerostomia 8. Lombaert IM, Knox SM, Hoffman MP. Salivary
despite existing challenges. Increasing knowl- gland progenitor cell biology provides a rationale
edge and information that is currently being gen- for therapeutic salivary gland regeneration. Oral Dis.
2011;17:445–9.
erated from in vitro, ex vivo, and in vivo 9. Pringle S, Van Os R, Coppes RP. Concise review:
applications of stem cells will continue to pro- adult salivary gland stem cells and a potential ther-
vide the foundations for future clinical trials. apy for xerostomia. Stem Cells. 2013;31:613–9.
10. Maria OM, Tran SD. Human mesenchymal stem cells 26. Mahoney MJ, Anseth KS. Three-dimensional
cultured with salivary gland biopsies adopt an epi- growth and function of neural tissue in degrad-
thelial phenotype. Stem Cells Dev. 2011;20:959–67. able polyethylene glycol hydrogels. Biomaterials.
11. Sumita Y, Liu Y, Khalili S, Maria OM, Xia D, et al. 2006;27:2265–74.
Bone marrow-derived cells rescue salivary gland 27. Benoit DS, Schwartz MP, Durney AR, Anseth
function in mice with head and neck irradiation. Int KS. Small functional groups for controlled differen-
J Biochem Cell Biol. 2011;43:80–7. tiation of hydrogel-encapsulated human mesenchy-
12. Park YJ, Koh J, Gauna AE, Chen S, Cha S. mal stem cells. Nat Mater. 2008;7:816–23.
Identification of regulatory factors for mesenchymal 28. Weber LM, Anseth KS. Hydrogel encapsulation
stem cell-derived salivary epithelial cells in a co- environments functionalized with extracellular
culture system. PLoS One. 2014;9:e112158. matrix interactions increase islet insulin secretion.
13. Watt FM, Hogan BL. Out of Eden: stem cells and Matrix Biol. 2008;27:667–73.
their niches. Science. 2000;287:1427–30. 29. Lin CC, Anseth KS. Cell-cell communication
14. Xu YX, Chen L, Wang R, Hou WK, Lin P, et al. mimicry with poly(ethylene glycol) hydrogels for
Mesenchymal stem cell therapy for diabetes enhancing beta-cell function. Proc Natl Acad Sci U
through paracrine mechanisms. Med Hypotheses. S A. 2011;108:6380–5.
2008;71:390–3. 30. Pradhan S, Liu C, Zhang C, Jia X, Farach-Carson
15. Ichim TE, Alexandrescu DT, Solano F, Lara F, MC, et al. Lumen formation in three-dimensional
Campion Rde N, et al. Mesenchymal stem cells cultures of salivary acinar cells. Otolaryngol Head
as anti-inflammatories: implications for treatment Neck Surg. 2010;142:191–5.
of Duchenne muscular dystrophy. Cell Immunol. 31. Walker JL, Menko AS, Khalil S, Rebustini I,
2010;260:75–82. Hoffman MP, et al. Diverse roles of E-cadherin in
16. Sequeira SJ, Larsen M, DeVine T. Extracellular the morphogenesis of the submandibular gland:
matrix and growth factors in salivary gland develop- insights into the formation of acinar and ductal struc-
ment. Front Oral Biol. 2010;14:48–77. tures. Dev Dyn. 2008;237:3128–41.
17. Gattazzo F, Urciuolo A, Bonaldo P. Extracellular 32. Davis MA, Reynolds AB. Blocked acinar develop-
matrix: a dynamic microenvironment for stem cell ment, E-cadherin reduction, and intraepithelial neo-
niche. Biochim Biophys Acta. 2014;1840:2506–19. plasia upon ablation of p120-catenin in the mouse
18. Maria OM, Zeitouni A, Gologan O, Tran salivary gland. Dev Cell. 2006;10:21–31.
SD. Matrigel improves functional properties of pri- 33. Okumura K, Shinohara M, Endo F. Capability of tis-
mary human salivary gland cells. Tissue Eng Part A. sue stem cells to organize into salivary rudiments.
2011;17:1229–38. Stem Cells Int. 2012;2012:502136.
19. Leal-Egana A, Scheibel T. Silk-based materials for 34. Shubin AD, Felong TJ, Graunke D, Ovitt CE, Benoit
biomedical applications. Biotechnol Appl Biochem. DS. Development of poly(ethylene glycol) hydro-
2010;55:155–67. gels for salivary gland tissue engineering applica-
20. Kundu B, Kurland NE, Yadavalli VK, Kundu tions. Tissue Eng Part A. 2015;21:1733–51.
SC. Isolation and processing of silk proteins for 35. Sander, M., Seymour, P., Kopp, J., Shih, B., Patel,
biomedical applications. Int J Biol Macromol. N. & Lim, C. Molecular cues regulating segregation
2014;70:70–7. of pancreatic, hepatic and intestinal lineages. The
21. Anseth KS, Metters AT, Bryant SJ, Martens PJ, FASEB Journal 2011;25, 303.3.
Elisseeff JH, et al. In situ forming degradable 36. Zaret KS. Regulatory phases of early liver develop-
networks and their application in tissue engineer- ment: paradigms of organogenesis. Nat Rev Genet.
ing and drug delivery. J Control Release. 2002;78: 2002;3:499–512.
199–209. 37. Wandzioch E, Zaret KS. Dynamic signaling network
22. Nuttelman CR, Tripodi MC, Anseth KS. In vitro for the specification of embryonic pancreas and liver
osteogenic differentiation of human mesenchymal progenitors. Science. 2009;324:1707–10.
stem cells photoencapsulated in PEG hydrogels. 38. Bonal C, Herrera PL. Genes controlling pancreas
J Biomed Mater Res A. 2004;68:773–82. ontogeny. Int J Dev Biol. 2008;52:823–35.
23. Benoit DS, Durney AR, Anseth KS. The effect of 39. Gittes GK. Developmental biology of the pancreas: a
heparin-functionalized PEG hydrogels on three- comprehensive review. Dev Biol. 2009;326:4–35.
dimensional human mesenchymal stem cell osteo- 40. Kim SK, Hebrok M. Intercellular signals regulat-
genic differentiation. Biomaterials. 2007;28:66–77. ing pancreas development and function. Genes Dev.
24. Weber LM, He J, Bradley B, Haskins K, Anseth 2001;15:111–27.
KS. PEG-based hydrogels as an in vitro encapsula- 41. Mfopou JK, Chen B, Sui LN, Sermon K, Bouwens
tion platform for testing controlled beta-cell micro- L. Recent advances and prospects in the differentia-
environments. Acta Biomater. 2006;2:1–8. tion of pancreatic cells from human embryonic stem
25. Lin CC, Raza A, Shih H. PEG hydrogels formed by cells. Diabetes. 2010;59:2094–101.
thiol-ene photo-click chemistry and their effect on 42. Patel VN, Rebustini IT, Hoffman MP. Salivary
the formation and recovery of insulin-secreting cell gland branching morphogenesis. Differentiation.
spheroids. Biomaterials. 2011;32:9685–95. 2006;74:349–64.
43. Dalvi MP, Umrani MR, Joglekar MV, Hardikar transplantation to control type 1 diabetes. PLoS One.
AA. Human pancreatic islet progenitor cells dem- 2008;3:e1451.
onstrate phenotypic plasticity in vitro. J Biosci. 57. Gabr MM, Sobh MM, Zakaria MM, Refaie AF,
2009;34:523–8. Ghoneim MA. Transplantation of insulin-produc-
44. Hardikar AA, Marcus-Samuels B, Geras-Raaka E, ing clusters derived from adult bone marrow stem
Raaka BM, Gershengorn MC. Human pancreatic cells to treat diabetes in rats. Exp Clin Transplant.
precursor cells secrete FGF2 to stimulate clustering 2008;6:236–43.
into hormone-expressing islet-like cell aggregates. 58. Gao F, Wu DQ, Hu YH, Jin GX. Extracellular matrix
Proc Natl Acad Sci U S A. 2003;100:7117–22. gel is necessary for in vitro cultivation of insulin
45. Si-Tayeb K, Noto FK, Nagaoka M, Li JX, Battle MA, producing cells from human umbilical cord blood
et al. Highly efficient generation of human hepato- derived mesenchymal stem cells. Chin Med J (Engl).
cyte-like cells from induced pluripotent stem cells 2008;121:811–8.
(vol 51, pg 297, 2010). Hepatology. 2010;51:1094. 59. Otonkoski T, Beattie GM, Mally MI, Ricordi C,
46. Cras-Meneur C, Elghazi L, Czernichow P, Hayek A. Nicotinamide is a potent inducer of endo-
Scharfmann R. Epidermal growth factor increases crine differentiation in cultured human fetal pancre-
undifferentiated pancreatic embryonic cells atic cells. J Clin Invest. 1993;92:1459–66.
in vitro – a balance between proliferation and dif- 60. Gao F, Wu DQ, Hu YH, Jin GX, Li GD, et al. In vitro
ferentiation. Diabetes. 2001;50:1571–9. cultivation of islet-like cell clusters from human
47. Kitade M, Factor VM, Andersen JB, Tomokuni A, umbilical cord blood-derived mesenchymal stem
Kaji K, et al. Specific fate decisions in adult hepatic cells. Transl Res. 2008;151:293–302.
progenitor cells driven by MET and EGFR signal- 61. Aguayo-Mazzucato C, Koh A, El Khattabi I, Li WC,
ing. Genes Dev. 2013;27:1706–17. Toschi E, et al. Mafa expression enhances glucose-
48. Kashimata M, Sayeed S, Ka A, Onetti-Muda A, responsive insulin secretion in neonatal rat beta
Sakagami H, et al. The ERK-1/2 signaling pathway cells. Diabetologia. 2011;54:583–93.
is involved in the stimulation of branching morpho- 62. Timper K, Seboek D, Eberhardt M, Linscheid P,
genesis of fetal mouse submandibular glands by Christ-Crain M, et al. Human adipose tissue-derived
EGF. Dev Biol. 2000;220:183–96. mesenchymal stem cells differentiate into insu-
49. Demeterco C, Beattie GM, Dib SA, Lopez AD, lin, somatostatin, and glucagon expressing cells.
Hayek A. A role for activin A and betacel- Biochem Biophys Res Commun. 2006;341:1135–40.
lulin in human fetal pancreatic cell differen- 63. Mashima H, Shibata H, Mine T, Kojima I. Formation
tiation and growth. J Clin Endocrinol Metab. of insulin-producing cells from pancreatic aci-
2000;85:3892–7. nar AR42J cells by hepatocyte growth factor.
50. Chen LB, Jiang XB, Yang L. Differentiation of rat Endocrinology. 1996;137:3969–76.
marrow mesenchymal stem cells into pancreatic islet 64. Otonkoski T, Cirulli V, Beattie GM, Mally MI, Soto
beta-cells. World J Gastroenterol. 2004;10:3016–20. G, et al. A role for hepatocyte growth factor/scatter
51. Segev H, Fishman B, Ziskind A, Shulman M, factor in fetal mesenchyme-induced pancreatic beta-
Itskovitz-Eldor J. Differentiation of human embry- cell growth. Endocrinology. 1996;137:3131–9.
onic stem cells into insulin-producing clusters. Stem 65. Onitsuka I, Tanaka M, Miyajima A. Characterization
Cells. 2004;22:265–74. and functional analyses of hepatic mesothelial cells
52. Tang DQ, Cao LZ, Burkhardt BR, Xia CQ, in mouse liver development. Gastroenterology.
Litherland SA, et al. In vivo and in vitro charac- 2010;138:1525–U1395.
terization of insulin-producing cells obtained from 66. Ishikawa T, Factor VM, Marquardt JU, Raggi C, Seo
murine bone marrow. Diabetes. 2004;53:1721–32. D, et al. Hepatocyte growth factor/c-met signaling is
53. Sun Y, Chen L, Hou XG, Hou WK, Dong JJ, et al. required for stem-cell-mediated liver regeneration in
Differentiation of bone marrow-derived mesenchymal mice. Hepatology. 2012;55:1215–26.
stem cells from diabetic patients into insulin-producing 67. Wang TC, Bonnerweir S, Oates PS, Chulak M,
cells in vitro. Chin Med J (Engl). 2007;120:771–6. Simon B, et al. Pancreatic gastrin stimulates islet
54. Sun B, Roh KH, Lee SR, Lee YS, Kang KS. Induction differentiation of transforming growth-factor
of human umbilical cord blood-derived stem cells alpha-
induced ductular precursor cells. J Clin
with embryonic stem cell phenotypes into insulin Investig. 1993;92:1349–56.
producing islet-like structure. Biochem Biophys Res 68. Wang RN, Rehfeld JF, Nielsen FC, Kloppel
Commun. 2007;354:919–23. G. Expression of gastrin and transforming growth
55. Wu XH, Liu CP, Xu KF, Mao XD, Zhu J, et al. factor-alpha during duct to islet cell differentiation in
Reversal of hyperglycemia in diabetic rats by por- the pancreas of duct-ligated adult rats. Diabetologia.
tal vein transplantation of islet-like cells generated 1997;40:887–93.
from bone marrow mesenchymal stem cells. World 69. Halban PA, Powers SL, George KL, Bonnerweir
J Gastroenterol. 2007;13:3342–9. S. Spontaneous reassociation of dispersed adult-
56. Chao KC, Chao KF, Fu YS, Liu SH. Islet-like rat pancreatic-islet cells into aggregates with
clusters derived from mesenchymal stem cells in 3-dimensional architecture typical of native islets.
Wharton’s Jelly of the human umbilical cord for Diabetes. 1987;36:783–90.
70. Kamiya A, Kinoshita T, Ito Y, Matsui T, Morikawa factor hepatocyte nuclear factor-6/Onecut-1 con-
Y, et al. Fetal liver development requires a paracrine trols the expression of its paralog Onecut-3 in
action of oncostatin M through the gp130 signal developing mouse endoderm. J Biol Chem.
transducer. EMBO J. 1999;18:2127–36. 2004;279:51298–304.
71. Kamiya A, Kojima N, Kinoshita T, Sakai Y, 86. Du A, Hunter CS, Murray J, Noble D, Cai CL, et al.
Miyaijma A. Maturation of fetal hepatocytes Islet-1 is required for the maturation, proliferation,
in vitro by extracellular matrices and oncostatin M: and survival of the endocrine pancreas. Diabetes.
induction of tryptophan oxygenase. Hepatology. 2009;58:2059–69.
2002;35:1351–9. 87. Pin CL, Rukstalis JM, Johnson C, Konieczny SF. The
72. Matsui T, Kinoshita T, Hirano T, Yokota T, Miyajima bHLH transcription factor Mist1 is required to main-
A. STAT3 down-regulates the expression of tain exocrine pancreas cell organization and acinar
cyclin D during liver development. J Biol Chem. cell identity. J Cell Biochem. 2001;155:519–30.
2002;277:36167–73. 88. Imai Y, Patel HR, Hawkins EJ, Doliba NM,
73. Banas A, Yamamoto Y, Teratani T, Ochiya T. Stem Matschinsky FM, et al. Insulin secretion is increased
cell plasticity: Learning from hepatogenic differen- in pancreatic islets of neuropeptide Y-deficient mice.
tiation strategies. Dev Dyn. 2007;236:3228–41. Endocrinology. 2007;148:5716–23.
74. Lee OK, Kuo TK, Chen WM, Lee KD, Hsieh SL, et al. 89. Gu C, Stein GH, Pan N, Goebbels S, Hornberg H,
Isolation of multipotent mesenchymal stem cells from et al. Pancreatic beta cells require NeuroD to achieve
umbilical cord blood. Blood. 2004;103:1669–75. and maintain functional maturity. Cell Metab.
75. Arany S, Catalan MA, Roztocil E, Ovitt CE. Ascl3 2010;11:298–310.
knockout and cell ablation models reveal complexity 90. Magenheim J, Klein AM, Stanger BZ, Ashery-
of salivary gland maintenance and regeneration. Dev Padan R, Sosa-Pineda B, et al. Ngn3(+) endocrine
Biol. 2011;353:186–93. progenitor cells control the fate and morphogen-
76. Courtois G, Morgan JG, Campbell LA, Fourel G, esis of pancreatic ductal epithelium. Dev Biol.
Crabtree GR. Interaction of a liver-specific nuclear 2011;359:26–36.
factor with the fibrinogen and alpha-1-antitrypsin 91. Papizan JB, Singer RA, Tschen SI, Dhawan S, Friel
promoters. Science. 1987;238:688–92. JM, et al. Nkx2.2 repressor complex regulates islet
77. Cereghini S, Blumenfeld M, Yaniv M. A liver- beta-cell specification and prevents beta-to-alpha-
specific factor essential for albumin transcription cell reprogramming. Genes Dev. 2011;25:2291–305.
differs between differentiated and dedifferentiated 92. Taylor BL, Liu FF, Sander M. Nkx6.1 is essential for
rat hepatoma-cells. Genes Dev. 1988;2:957–74. maintaining the functional state of pancreatic beta
78. Ishiyama T, Kano J, Minami Y, Iijima T, Morishita Y, cells. Cell Rep. 2013;4:1262–75.
et al. Expression of HNFs and C/EBP alpha is cor- 93. Flanagan SE, De Franco E, Lango Allen H, Zerah
related with immunocytochemical differentiation of M, Abdul-Rasoul MM, et al. Analysis of transcrip-
cell lines derived from human hepatocellular carci- tion factors key for mouse pancreatic develop-
nomas, hepatoblastomas and immortalized hepato- ment establishes NKX2-2 and MNX1 mutations
cytes. Cancer Sci. 2003;94:757–63. as causes of neonatal diabetes in man. Cell Metab.
79. Gao N, LeLay J, Vatamaniuk MZ, Rieck S, Friedman 2014;19:146–54.
JR, et al. Dynamic regulation of Pdx1 enhancers by 94. Bonnefond A, Vaillant E, Philippe J, Skrobek B,
Foxa1 and Foxa2 is essential for pancreas develop- Lobbens S, et al. Transcription factor gene MNX1
ment. Genes Dev. 2008;22:3435–48. is a novel cause of permanent neonatal diabe-
80. Xu CH, Lu XW, Chen EZ, He ZY, Uyunbilig B, tes in a consanguineous family. Diabetes Metab.
et al. Genome-wide roles of Foxa2 in directing liver 2013;39:276–80.
specification. J Mol Cell Biol. 2012;4:420–2. 95. Hart AW, Mella S, Mendrychowski J, van Heyningen
81. Gao N, White P, Doliba N, Golson ML, Matschinsky V, Kleinjan DA. The developmental regulator Pax6 is
FM, et al. Foxa2 controls vesicle docking and essential for maintenance of islet cell function in the
insulin secretion in mature beta cells. Cell Metab. adult mouse pancreas. PLoS One. 2013;8:e54173.
2007;6:267–79. 96. Jennings RE, Berry AA, Kirkwood-Wilson R,
82. Carrasco M, Delgado I, Soria B, Martin F, Rojas Roberts NA, Hearn T, et al. Development of the
A. GATA4 and GATA6 control mouse pancreas human pancreas from foregut to endocrine commit-
organogenesis. J Clin Invest. 2012;122:3504–15. ment. Diabetes. 2013;62:3514–22.
83. Haumaitre C, Barbacci E, Jenny M, Ott MO, 97. Gu G, Dubauskaite J, Melton DA. Direct evidence
Gradwohl G, et al. Lack of TCF2/vHNF1 in mice for the pancreatic lineage: NGN3+ cells are islet
leads to pancreas agenesis. Proc Natl Acad Sci U S progenitors and are distinct from duct progenitors.
A. 2005;102:1490–5. Development. 2002;129:2447–57.
84. Watt AJ, Garrison WD, Duncan SA. HNF4: a central 98. Brissova M, Blaha M, Spear C, Nicholson W,
regulator of hepatocyte differentiation and function. Radhika A, et al. Reduced PDX-1 expression impairs
Hepatology. 2003;37:1249–53. islet response to insulin resistance and worsens glu-
85. Pierreux CE, Vanhorenbeeck V, Jacquemin P, cose homeostasis. Am J Physiol Endocrinol Metab.
Lemaigre FP, Rousseau GG. The transcription 2005;288:E707–14.
99. Yoshitomi H, Zaret KS. Endothelial cell inter- gland cells differentiated from mouse early ES cells
actions initiate dorsal pancreas development by in vitro. Hum Cell. 2013;26:80–90.
selectively inducing the transcription factor Ptf1a. 115. Zhou Q, Brown J, Kanarek A, Rajagopal J, Melton
Development. 2004;131:807–17. DA. In vivo reprogramming of adult pancreatic exo-
100. Weedon MN, Cebola I, Patch AM, Flanagan SE, crine cells to beta-cells. Nature. 2008;455:627–U630.
De Franco E, et al. Recessive mutations in a distal 116. Lima MJ, Muir KR, Docherty HM, Drummond R,
PTF1A enhancer cause isolated pancreatic agenesis. McGowan NWA, et al. Suppression of epithelial-
Nat Genet. 2014;46:61–4. to-mesenchymal transitioning enhances ex vivo
101. Kawaguchi Y. Sox9 and programming of liver reprogramming of human exocrine pancreatic tissue
and pancreatic progenitors. J Clin Invest. toward functional insulin-producing beta-like cells.
2013;123:1881–6. Diabetes. 2013;62:2821–33.
102. Mead TJ, Wang Q, Bhattaram P, Dy P, Afelik S, et al. 117. Li L, Yi ZH, Seno M, Kojima I. Activin A and
A far-upstream (−70 kb) enhancer mediates Sox9 betacellulin – effect on regeneration of pancreatic
auto-regulation in somatic tissues during develop- beta-cells in neonatal streptozotocin-treated rats.
ment and adult regeneration. Nucleic Acids Res. Diabetes. 2004;53:608–15.
2013;41:4459–69. 118. Huang PY, He ZY, Ji SY, Sun HW, Xiang D, et al.
103. Kanai-Azuma M, Kanai Y, Gad JM, Tajima Y, Taya C, Induction of functional hepatocyte-like cells from
et al. Depletion of definitive gut endoderm in Sox17- mouse fibroblasts by defined factors. Nature.
null mutant mice. Development. 2002;129:2367–79. 2011;475:386–U142.
104. Tran SD, Liu YN, Xia DS, Maria OM, Khalili S, et al. 119. Yu B, He ZY, You P, Han QW, Xiang D, et al.
Paracrine effects of bone marrow soup restore organ Reprogramming fibroblasts into bipotential hepatic
function, regeneration, and repair in salivary glands stem cells by defined factors. Cell Stem Cell.
damaged by irradiation. PLoS One. 2013;8:e61632. 2013;13:328–40.
105. Caplan AI, Dennis JE. Mesenchymal stem cells as tro- 120. Ahlgren U, Pfaff SL, Jessell TM, Edlund T, Edlund
phic mediators. J Cell Biochem. 2006;98:1076–84. H. Independent requirement for ISL1 in formation
106. Fang D, Hu S, Liu Y, Quan VH, Seuntjens J, et al. of pancreatic mesenchyme and islet cells. Nature.
Identification of the active components in Bone 1997;385:257–60.
Marrow Soup: a mitigator against irradiation-injury 121. Edlund H. Pancreatic organogenesis – developmen-
to salivary glands. Sci Rep. 2015;5:16017. tal mechanisms and implications for therapy. Nat
107. Entesarian M, Dahlqvist J, Shashi V, Stanley CS, Rev Genet. 2002;3:524–32.
Falahat B, et al. FGF10 missense mutations in apla- 122. Matsuoka TA, Kaneto H, Miyatsuka T, Yamamoto T,
sia of lacrimal and salivary glands (ALSG). Eur Yamamoto K, et al. Regulation of MafA expression
J Hum Genet. 2007;15:379–82. in pancreatic beta-cells in db/db mice with diabetes.
108. Scheckenbach K, Balz V, Wagenmann M, Hoffmann Diabetes. 2010;59:1709–20.
TK. An intronic alteration of the fibroblast growth 123. Harmon JS, Bogdani M, Parazzoli SD, Mak SS,
factor 10 gene causing ALSG-(aplasia of lacrimal Oseid EA, et al. Beta-Cell-specific overexpression of
and salivary glands) syndrome. BMC Med Genet. glutathione peroxidase preserves intranuclear MafA
2008;9:114. and reverses diabetes in db/db mice. Endocrinology.
109. Lombaert IMA, Brunsting JF, Wierenga PK, 2009;150:4855–62.
Kampinga HH, De Haan G, et al. Keratinocyte 124. Butler AE, Robertson RP, Hernandez R, Matveyenko
growth factor prevents radiation damage to salivary AV, Gurlo T, et al. Beta cell nuclear musculoapo-
glands by expansion of the stem/progenitor pool. neurotic fibrosarcoma oncogene family A (MafA)
Stem Cells. 2008;26:2595–601. is deficient in type 2 diabetes. Diabetologia.
110. Panman L, Andersson E, Alekseenko Z, Hedlund E, 2012;55:2985–8.
Kee N, et al. Transcription factor-induced lineage 125. Guo S, Dai C, Guo M, Taylor B, Harmon JS, et al.
selection of stem-cell-derived neural progenitor Inactivation of specific beta cell transcription factors
cells. Cell Stem Cell. 2011;8:663–75. in type 2 diabetes. J Clin Invest. 2013;123:3305–16.
111. Jones S. An overview of the basic helix-loop-helix 126. Talchai C, Xuan S, Lin HV, Sussel L, Accili D.
proteins. Genome Biol. 2004;5:226. Pancreatic beta cell dedifferentiation as a mechanism
112. Fairman R, Beran-Steed RK, Anthony-Cahill SJ, of diabetic beta cell failure. Cell. 2012;150:1223–34.
Lear JD, Stafford 3rd WF, et al. Multiple oligo- 127. Ghaye AP, Bergemann D, Tarifeno-Saldivia E,
meric states regulate the DNA binding of helix- Flasse LC, Von Berg V, et al. Progenitor potential
loop-helix peptides. Proc Natl Acad Sci U S A. of nkx6.1-expressing cells throughout zebrafish
1993;90:10429–33. life and during beta cell regeneration. BMC Biol.
113. Shih HP, Seymour PA, Patel NA, Xie R, Wang A, et al. 2015;13:70.
A gene regulatory network cooperatively controlled 128. Chan JY, Luzuriaga J, Bensellam M, Biden TJ,
by Pdx1 and Sox9 governs lineage allocation of fore- Laybutt DR. Failure of the adaptive unfolded protein
gut progenitor cells. Cell Rep. 2015;13:326–36. response in islets of obese mice is linked with abnor-
114. Kawakami M, Ishikawa H, Tachibana T, Tanaka malities in beta-cell gene expression and progression
A, Mataga I. Functional transplantation of salivary to diabetes. Diabetes. 2013;62:1557–68.
129. Li H, Arber S, Jessell TM, Edlund H. Selective agen- 143. Fausto N. Liver regeneration. J Hepatol.
esis of the dorsal pancreas in mice lacking homeo- 2000;32:19–31.
box gene Hlxb9. Nat Genet. 1999;23:67–70. 144. Koniaris LG, McKillop IH, Schwartz SI,
130. Conrad E, Stein R, Hunter CS. Revealing tran-
Zimmers TA. Liver regeneration. J Am Coll Surg.
scription factors during human pancreatic beta 2003;197:634–59.
cell development. Trends Endocrinol Metab. 145. Crissey MAS, Leu JI, De Angelis RA, Greenbaum LE,
2014;25:407–14. Scearce LM, et al. Liver-specific and proliferation-
131. Hamasaki A, Yamada Y, Kurose T, Ban N, Nagashima induced deoxyribonuclease I hypersensitive sites in
K, et al. Adult pancreatic islets require differential the mouse insulin-like growth factor binding pro-
pax6 gene dosage. Biochem Biophys Res Commun. tein-1 gene. Hepatology. 1999;30:1187–97.
2007;353:40–6. 146. Tomizawa M, Garfield S, Xanthopoulos
132. Wen JH, Chen YY, Song SJ, Ding J, Gao Y, et al. KG. Hepatocytes deficient in CCAAT enhancer
Paired box 6 (PAX6) regulates glucose metabo- binding protein alpha (C/EBP alpha) exhibit both
lism via proinsulin processing mediated by pro- hepatocyte and biliary epithelial cell character.
hormone convertase 1/3 (PC1/3). Diabetologia. Hepatology. 1998;28:294a.
2009;52:504–13. 147. Yamasaki H, Sada A, Iwata T, Niwa T, Tomizawa
133. Ding J, Gao Y, Zhao J, Yan H, Guo SY, et al. Pax6 M, et al. Suppression of C/EBP alpha expression
haploinsufficiency causes abnormal metabolic in periportal hepatoblasts may stimulate biliary cell
homeostasis by down-regulating glucagon-like pep- differentiation through increased Hnf6 and Hnf1b
tide 1 in mice. Endocrinology. 2009;150:2136–44. expression. Development. 2006;133:4233–43.
134. Habener JF, Kemp DM, Thomas MK. Minireview: 148. Antoniou A, Raynaud P, Cordi S, Zong YW, Tronche
transcriptional regulation in pancreatic development. F, et al. Intrahepatic bile ducts develop accord-
Endocrinology. 2005;146:1025–34. ing to a new mode of tubulogenesis regulated by
135. Fujimoto K, Polonsky KS. Pdx1 and other factors the transcription factor SOX9. Gastroenterology.
that regulate pancreatic beta-cell survival. Diabetes 2009;136:2325–33.
Obes Metab. 2009;11:30–7. 149. Bullard T, Koek L, Roztocil E, Kingsley PD, Mirels
136. Spence JR, Lange AW, Lin SCJ, Kaestner KH, Lowy L, et al. Ascl3 expression marks a progenitor popula-
AM, et al. Sox17 regulates organ lineage segregation of both acinar and ductal cells in mouse salivary
tion of ventral foregut progenitor cells. Dev Cell. glands. Dev Biol. 2008;320:72–8.
2009;17:62–74. 150. Yoshida S, Ohbo K, Takakura A, Takebayashi H,
137. Kamiya A, Inoue Y, Gonzalez FJ. Role of the hepa- Okada T, et al. Sgn1, a basic helix-loop-helix tran-
tocyte nuclear factor 4 alpha in control of the preg- scription factor delineates the salivary gland duct
nane X receptor during fetal liver development. cell lineage in mice. Dev Biol. 2001;240:517–30.
Hepatology. 2003;37:1375–84. 151. Aure MH, Konieczny SF, Ovitt CE. Salivary gland
138. Li JX, Ning G, Duncan SA. Mammalian hepato- homeostasis is maintained through acinar cell self-
cyte differentiation requires the transcription factor duplication. Dev Cell. 2015;33:231–7.
HNF-4 alpha. Genes Dev. 2000;14:464–74. 152. Gagliardi A, Mullin NP, Tan ZY, Colby D, Kousa
139. Jin JL, Iakova P, Jiang YJ, Lewis K, Sullivan E, et al. AI, et al. A direct physical interaction between
Transcriptional and translational regulation of C/ Nanog and Sox2 regulates embryonic stem cell self-
EBP beta-HDAC1 protein complexes controls differ- renewal. EMBO J. 2013;32:2231–47.
ent levels of p53, SIRT1, and PGC1 alpha proteins at 153. Knox SM, Lombaert IM, Reed X, Vitale-Cross
the early and late stages of liver cancer. J Biol Chem. L, Gutkind JS, et al. Parasympathetic innervation
2013;288:14451–62. maintains epithelial progenitor cells during salivary
140. Schrem H, Klempnauer J, Borlak J. Liver-enriched organogenesis. Science. 2010;329:1645–7.
transcription factors in liver function and develop- 154. Pereira L, Yi F, Merrill BJ. Repression of Nanog
ment. Part II: the C/EBPs and D site-binding pro- gene transcription by Tcf3 limits embryonic
tein in cell cycle control, carcinogenesis, circadian stem cell self-renewal. Mol Cell Biol. 2006;
gene regulation, liver regeneration, apoptosis, and 26:7479–91.
liver-
specific gene regulation. Pharmacol Rev. 155. Wray J, Kalkan T, Gomez-Lopez S, Eckardt D,
2004;56:291–330. Cook A, et al. Inhibition of glycogen synthase
141. Flodby P, Barlow C, Kylefjord H, Ahrlund-Richter kinase-3 alleviates Tcf3 repression of the pluri-
L, Xanthopoulos KG. Increased hepatic cell pro- potency network and increases embryonic stem
liferation and lung abnormalities in mice deficient cell resistance to differentiation. Nat Cell Biol.
in CCAAT/enhancer binding protein alpha. J Biol 2011;13:838–45.
Chem. 1996;271:24753–60. 156. Kelly KF, Ng DY, Jayakumaran G, Wood GA, Koide
142. Nakamura T, Ueno T, Sakamoto M, Sakata R, H, et al. Beta-catenin enhances Oct-4 activity and
Torimura T, et al. Suppression of transforming reinforces pluripotency through a TCF-independent
growth factor-beta results in upregulation of tran- mechanism. Cell Stem Cell. 2011;8:214–27.
scription of regeneration factors after chronic liver 157. Yi F, Pereira L, Hoffman JA, Shy BR, Yuen CM,
injury. J Hepatol. 2004;41:974–82. et al. Opposing effects of Tcf3 and Tcf1 control Wnt
stimulation of embryonic stem cell self-renewal. Nat cells exhibiting regulated exocytosis. Anat Rec.
Cell Biol. 2011;13:762–U383. 2000;259:157–67.
158. Atlasi Y, Noori R, Gaspar C, Franken P, Sacchetti 174. Direnzo D, Hess DA, Damsz B, Hallett JE,
A, et al. Wnt signaling regulates the lineage dif- Marshall B, et al. Induced Mist1 expression pro-
ferentiation potential of mouse embryonic stem motes remodeling of mouse pancreatic acinar cells.
cells through Tcf3 down-regulation. PLoS Genet. Gastroenterology. 2012;143:469–80.
2013;9:e1003424. 175. Rukstalis JM, Kowalik A, Zhu LQ, Lidington D, Pin
159. Wu CI, Hoffman JA, Shy BR, Ford EM, Fuchs E, CL, et al. Exocrine specific expression of Connexin32
et al. Function of Wnt/beta-catenin in counteracting is dependent on the basic helix-loop-helix transcrip-
Tcf3 repression through the Tcf3-beta-catenin inter- tion factor Mist1. J Cell Sci. 2003;116:3315–25.
action. Development. 2012;139:2118–29. 176.
Pan FC, Wright C. Pancreas organogenesis:
160. Marmorstein LY, Kinev AV, Chan GKT, Bochar DA, from bud to plexus to gland. Dev Dyn. 2011;240:
Beniya H, et al. A human BRCA2 complex contain- 530–65.
ing a structural DNA binding component influences 177. McDonald E, Li JM, Krishnamurthy M, Fellows GF,
cell cycle progression. Cell. 2001;104:247–57. Goodyer CG, et al. SOX9 regulates endocrine cell
161. Lin JW, Biankin AV, Horb ME, Ghosh B, Prasad NB, differentiation during human fetal pancreas develop-
et al. Differential requirement for ptf1a in endocrine ment. Int J Biochem Cell Biol. 2012;44:72–83.
and exocrine lineages of developing zebrafish pan- 178. Selman K, Kafatos FC. Transdifferentiation in the
creas. Dev Biol. 2004;274:491–503. labial gland of silk moths: is DNA required for cel-
162. Cano DA, Soria B, Martin F, Rojas A. Transcriptional lular metamorphosis? Cell Differ. 1974;3:81–94.
control of mammalian pancreas organogenesis. Cell 179. Eguchi G, Okada TS. Differentiation of lens tissue
Mol Life Sci. 2014;71:2383–402. from the progeny of chick retinal pigment cells cul-
163. Garg V, Kathiriya IS, Barnes R, Schluterman MK, tured in vitro: a demonstration of a switch of cell
King IN, et al. GATA4 mutations cause human con- types in clonal cell culture. Proc Natl Acad Sci U S
genital heart defects and reveal an interaction with A. 1973;70:1495–9.
TBX5. Nature. 2003;424:443–7. 180. Araki M, Okada TS. Differentiation of lens and pig-
164. Watt AJ, Zhao R, Li JX, Duncan SA. Development ment cells in cultures of neural retinal cells of early
of the mammalian liver and ventral pancreas is chick embryos. Dev Biol. 1977;60:278–86.
dependent on GATA4. BMC Dev Biol. 2007;7:37. 181. Bronnerfraser M. Transdifferentiation – flex-
165. Lee CS, Sund NJ, Vatamaniuk MZ, Matschinsky ibility in cell-differentiation – Okada, Ts. Science.
FM, Stoffers DA, et al. Foxa2 controls Pdx1 gene 1992;256:1054–5.
expression in pancreatic beta-cells in vivo. Diabetes. 182. Eberhard D, Tosh D. Transdifferentiation and meta-
2002;51:2546–51. plasia as a paradigm for understanding development
166. Lee CS, Sund NJ, Behr R, Herrera PL, Kaestner and disease. Cell Mol Life Sci. 2008;65:33–40.
KH. Foxa2 is required for the differentiation of pan- 183. Gruh IMU. Transdifferentiation of stem cells: a criti-
creatic alpha-cells. Dev Biol. 2005;278:484–95. cal view. In: Martin U, editor. Engineering of stem
167. Zaret KS, Carroll JS. Pioneer transcription factors: cells. Heidelberg: Springer; 2009. p. 73–106.
establishing competence for gene expression. Genes 184. Burke ZD, Tosh D. Therapeutic potential of transdif-
Dev. 2011;25:2227–41. ferentiated cells. Clin Sci (Lond). 2005;108:309–21.
168. Le Lay J, Kaestner KH. The fox genes in the liver: 185. Torper O, Pfisterer U, Wolf DA, Pereira M, Lau
from organogenesis to functional integration. S, et al. Generation of induced neurons via direct
Physiol Rev. 2010;90:1–22. conversion in vivo. Proc Natl Acad Sci U S A.
169. Jeon J, Correa-Medina M, Ricordi C, Edlund H, 2013;110:7038–43.
Diez JA. Endocrine cell clustering during human 186. Merrell AJ, Stanger BZ. Adult cell plasticity in vivo:
pancreas development. J Histochem Cytochem. de-differentiation and transdifferentiation are back
2009;57:811–24. in style. Nat Rev Mol Cell Biol. 2016;17:413–145.
170. Rubio-Cabezas O, Jensen JN, Hodgson MI, Codner 187. Burke ZD, Thowfeequ S, Peran M, Tosh D. Stem
E, Ellard S, et al. Permanent neonatal diabetes and cells in the adult pancreas and liver. Biochem
enteric anendocrinosis associated with biallelic muta- J. 2007;404:169–78.
tions in NEUROG3. Diabetes. 2011;60:1349–53. 188. Kume S. Stem-cell-based approaches for regenera-
171. Gradwohl G, Dierich A, LeMeur M, Guillemot tive medicine. Dev Growth Differ. 2005;47:393–402.
F. Neurogenin3 is required for the development of 189. Dabeva MD, Hurston E, Shafritz DA. Transcription
the four endocrine cell lineages of the pancreas. Proc factor and liver-specific mRNA expression in fac-
Natl Acad Sci U S A. 2000;97:1607–11. ultative epithelial progenitor cells of liver and pan-
172. Johnson CL, Kowalik AS, Rajakumar N, Pin creas. Am J Pathol. 1995;147:1633–48.
CL. Mist1 is necessary for the establishment of 190. McLin VA, Zorn AM. Organogenesis: making pan-
granule organization in serous exocrine cells of the creas from liver. Curr Biol. 2003;13:R96–8.
gastrointestinal tract. Mech Dev. 2004;121:261–72. 191. Horb ME, Shen CN, Tosh D, Slack JMW.
173. Pin CL, Bonvissuto AC, Konieczny SF. Mist1 Experimental conversion of liver to pancreas. Curr
expression is a common link among serous exocrine Biol. 2003;13:105–15.
192. Kojima H, Nakamura T, Fujita Y, Kishi A, Fujimiya 206. Rock JR, Onaitis MW, Rawlins EL, Lu Y, Clark CP,
M, et al. Combined expression of pancreatic duo- et al. Basal cells as stem cells of the mouse trachea
denal homeobox 1 and islet factor 1 induces imma- and human airway epithelium. Proc Natl Acad Sci U
ture enterocytes to produce insulin. Diabetes. S A. 2009;106:12771–5.
2002;51:1398–408. 207. Arnold K, Sarkar A, Yram MA, Polo JM, Bronson
193. von Mach MA, Hengstler JG, Brulport M, Eberhardt R, et al. Sox2(+) adult stem and progenitor cells are
M, Schormann W, et al. In vitro cultured islet-derived important for tissue regeneration and survival of
progenitor cells of human origin express human mice. Cell Stem Cell. 2011;9:317–29.
albumin in severe combined immunodeficiency 208. Nelson J, Manzella K, Baker OJ. Current cell models
mouse liver in vivo. Stem Cells. 2004;22:1134–41. for bioengineering a salivary gland: a mini-review of
194. Lardon J, De Breuck S, Rooman I, Van Lommel emerging technologies. Oral Dis. 2013;19:236–44.
L, Kruhoffer M, et al. Plasticity in the adult rat 209. Maimets M, Rocchi C, Bron R, Pringle S, Kuipers J,
pancreas: Transdifferentiation of exocrine to et al. Long-term in vitro expansion of salivary gland
hepatocyte-like cells in primary culture. Hepatology. stem cells driven by Wnt signals. Stem Cells Rep.
2004;39:1499–507. 2016;6:150–62.
195. Kawaguchi Y, Cooper B, Gannon M, Ray M, 210. Yaniv A, Neumann Y, David R, Stiubea-Cohen R,
MacDonald RJ, et al. The role of the transcriptional Orbach Y, et al. Establishment of immortal multipo-
regulator Ptf1a in converting intestinal to pancreatic tent rat salivary progenitor cell line toward salivary
progenitors. Nat Genet. 2002;32:128–34. gland regeneration. Tissue Eng Part C Methods.
196. Ferber S, Halkin A, Cohen H, Ber I, Einav Y, et al. 2011;17:69–78.
Pancreatic and duodenal homeobox gene 1 induces 211. Sutton, M.T. & Bonfield, T.L. Stem cells: innova-
expression of insulin genes in liver and ameliorates tions in clinical applications. Stem Cells Int 2014,
streptozotocin-induced hyperglycemia. Nat Med. 516278 (2014).
2000;6:568–72. 212. Shamoon H, Duffy H, Fleischer N, Engel S,
197. Yi F, Liu GH, Izpisua Belmonte JC. Rejuvenating Saenger P, et al. The effect of intensive treatment
liver and pancreas through cell transdifferentiation. of diabetes on the development and progression
Cell Res. 2012;22:616–9. of long-term complications in insulin-dependent
198. Kaneto H, Nakatani Y, Miyatsuka T, Matsuoka T, diabetes-mellitus. N Engl J Med. 1993;329:977–86.
Matsuhisa M, et al. PDX-1/VP16 fusion protein, 213. Bouwens L, Houbracken I, Mfopou JK. The use of
together with NeuroD or Ngn3, markedly induces stem cells for pancreatic regeneration in diabetes
insulin gene transcription and ameliorates glucose mellitus. Nat Rev Endocrinol. 2013;9:598–606.
tolerance. Diabetes. 2005;54:1009–22. 214. Gennero L, Roos MA, Sperber K, Denysenko T,
199. Hisatomi Y, Okumura K, Nakamura K, Matsumoto Bernabei P, et al. Pluripotent plasticity of stem cells and
S, Satoh A, et al. Flow cytometric isolation of endo- liver repopulation. Cell Biochem Funct. 2010;28:178–89.
dermal progenitors from mouse salivary gland dif- 215. Miyajima A, Tanaka M, Itoh T. Stem/progenitor cells
ferentiate into hepatic and pancreatic lineages. in liver development, homeostasis, regeneration, and
Hepatology. 2004;39:667–75. reprogramming. Cell Stem Cell. 2014;14:561–74.
200. Rothova M, Thompson H, Lickert H, Tucker 216. Schwartz RE, Reyes M, Koodie L, Jiang YH, Blackstad
AS. Lineage tracing of the endoderm during oral M, et al. Multipotent adult progenitor cells from bone
development. Dev Dyn. 2012;241:1183–91. marrow differentiate into functional hepatocyte-like
201. Okumura K, Nakamura K, Hisatomi Y, Nagano
cells. J Clin Investig. 2002;109:1291–302.
K, Tanaka Y, et al. Salivary gland progenitor 217. Reyes M, Lund T, Lenvik T, Aguiar D, Koodie L,
cells induced by duct ligation differentiate into et al. Purification and ex vivo expansion of post-
hepatic and pancreatic lineages. Hepatology. natal human marrow mesodermal progenitor cells
2003;38:104–13. (Retracted article. See vol. 113, pg. 2370, 2009).
202. Denny PC, Denny PA. Dynamics of parenchymal Blood. 2001;98:2615–25.
cell division, differentiation, and apoptosis in the 218. Lu LL, Liu YJ, Yang SG, Zhao QJ, Wang X, et al.
young adult female mouse submandibular gland. Isolation and characterization of human umbilical
Anat Rec. 1999;254:408–17. cord mesenchymal stem cells with hematopoiesis-
203. Konings AW, Coppes RP, Vissink A. On the mecha- supportive function and other potentials.
nism of salivary gland radiosensitivity. Int J Radiat Haematologica. 2006;91:1017–26.
Oncol Biol Phys. 2005;62:1187–94. 219. Banas A, Teratani T, Yamamoto Y, Tokuhara M,
204. Patel VN, Hoffman MP. Salivary gland develop- Takeshita F, et al. Adipose tissue-derived mesenchy-
ment: a template for regeneration. Semin Cell Dev mal stem cells as a source of human hepatocytes.
Biol. 2014;25:52–60. Hepatology. 2007;46:219–28.
205. Nanduri LS, Maimets M, Pringle SA, van der Zwaag 220. Talens-Visconti R, Bonora A, Jover R, Mirabet V,
M, van Os RP, et al. Regeneration of irradiated sali- Carbonell F, et al. Human mesenchymal stem cells
vary glands with stem cell marker expressing cells. from adipose tissue: Differentiation into hepatic lin-
Radiother Oncol. 2011;99:367–72. eage. Toxicol In Vitro. 2007;21:324–9.
221. Planat-Benard V, Silvestre JS, Cousin B, Andre in mice with Sjogren’s-like disease. PLoS One.
M, Nibbelink M, et al. Plasticity of human adi- 2012;7:e38615.
pose lineage cells toward endothelial cells – physi- 231. Coppes RP, Zeilstra LJW, Kampinga HH, Konings
ological and therapeutic perspectives. Circulation. AWT. Early to late sparing of radiation damage to
2004;109:656–63. the parotid gland by adrenergic and muscarinic
222. Seo MJ, Suh SY, Bae YC, Jung JS. Differentiation receptor agonists. Br J Cancer. 2001;85:1055–63.
of human adipose stromal cells into hepatic lineage 232. Konings AWT, Faber H, Cotteleer F, Vissink A,
in vitro and in vivo. Biochem Biophys Res Commun. Coppes RP. Secondary radiation damage as the main
2005;328:258–64. cause for unexpected volume effects: a histopatho-
223. Lange C, Bassler P, Lioznov MV, Bruns H, Kluth logic study of the parotid gland. Int J Radiat Oncol
D, et al. Liver-specific gene expression in mesen- Biol Phys. 2006;64:98–105.
chymal stem cells is induced by liver cells. World 233. Jeong J, Baek H, Kim YJ, Choi Y, Lee H, et al.
J Gastroenterol. 2005;11:4497–504. Human salivary gland stem cells ameliorate hyposal-
224. Rai RM, Yang SQ, McClain C, Karp CL, Klein AS, ivation of radiation-damaged rat salivary glands.
et al. Kupffer cell depletion by gadolinium chloride Exp Mol Med. 2013;45:e58.
enhances liver regeneration after partial hepatec- 234. Lombaert IMA, Brunsting JF, Wierenga PK, Faber
tomy in rats. Am J Physiol. 1996;270:G909–18. H, Stokman MA, et al. Rescue of salivary gland
225. Sato Y, Araki H, Kato J, Nakamura K, Kawano function after stem cell transplantation in irradiated
Y, et al. Human mesenchymal stem cells xeno- glands. PLoS One. 2008;3:e2063.
grafted directly to rat liver are differentiated into 235. Kojima H, Fujimiya M, Matsumura K, Younan P,
human hepatocytes without fusion. Blood. 2005; Imaeda H, et al. NeuroD-betacellulin gene therapy
106:756–63. induces islet neogenesis in the liver and reverses dia-
226. Chamberlain J, Yamagami T, Colletti E, Theise ND, betes in mice. Nat Med. 2003;9:596–603.
Desai J, et al. Efficient generation of human hepato- 236. Nakatsuji N, Nakajima F, Tokunaga K. HLA-
cytes by the intrahepatic delivery of clonal human haplotype banking and iPS cells. Nat Biotechnol.
mesenchymal stem cells in fetal sheep. Hepatology. 2008;26:739–40.
2007;46:1935–45. 237. Lanza RP, Hayes JL, Chick WL. Encapsulated cell
227. Meier RPH, Muller YD, Morel P, Gonelle-Gispert technology. Nat Biotechnol. 1996;14:1107–11.
C, Buhler LH. Transplantation of mesenchymal 238. Dean SK, Yulyana Y, Williams G, Sidhu KS, Tuch
stem cells for the treatment of liver diseases, is BE. Differentiation of encapsulated embryonic
there enough evidence? Stem Cell Res. 2013; stem cells after transplantation. Transplantation.
11:1348–64. 2006;82:1175–84.
228. Bae SH. Clinical application of stem cells in liver 239. Selmani Z, Naji A, Gaiffe E, Obert L, Tiberghien P,
diseases. Korean J Hepatol. 2008;14:309–17. et al. HLA-G is a crucial immunosuppressive mol-
229. Russo FP, Alison MR, Bigger BW, Amofah E, Florou ecule secreted by adult human mesenchymal stem
A, et al. The bone marrow functionally contributes to cells. Transplantation. 2009;87:S62–6.
liver fibrosis. Gastroenterology. 2006;130:1807–21. 240. Jacobs SA, Roobrouck VD, Verfaillie CM, Van Gool
230. Khalili S, Liu YN, Kornete M, Roescher N, Kodama SW. Immunological characteristics of human mes-
S, et al. Mesenchymal stromal cells improve sali- enchymal stem cells and multipotent adult progeni-
vary function and reduce lymphocytic infiltrates tor cells. Immunol Cell Biol. 2013;91:32–9.
Part III
Bioengineering of Salivary Glands
Current Cell Models
for Bioengineering Salivary Glands 7
Olga J. Baker
Abstract
Saliva is critical for sustaining oral health. Patients with salivary gland
hypofunction (symptomatically, xerostomia) have difficulties in basic oral
functions such as chewing, tasting, and swallowing foods. Additionally,
they suffer from caries, periodontal disease, and a variety of microbial
infections. Salivary flow has been significantly improved through the use
of gene therapies and secretory agonists, thereby mitigating dry mouth
symptoms. However, scientific advancements in the area of clinically
applied implants are needed to more fully restore compromised salivary
gland function. In this book chapter, we will evaluate the advantages and
limitations of commonly used salivary cell lines. Furthermore, we will
summarize ongoing studies on current salivary cell isolation methods and
cell culture techniques using different biomaterials. Then, we will describe
the use of stem, progenitor, and acinar cells as candidates for salivary
gland regeneration and bioengineering studies. Finally, we will highlight
the use of scaffolds for growing salivary glands in vitro. Together, these
studies represent the state of the art and trends in the emerging field of sali-
vary gland bioengineering.
7.1 Introduction Hyposalivation is associated with the follow-

ing conditions: (a) Sjögren’s syndrome (SS), an
Hyposalivation contributes to tooth decay, peri- autoimmune disease with a prevalence between
odontitis, and microbial infections. Additionally, 0.1 and 3 % in the United States [48]; (b) radio-
it impairs activities of daily living such as therapy for head and neck cancer, accounting for
speaking, chewing, and swallowing [22].
up to 3 % of all cancers in the United States [35];
(c) side effects of commonly used medications
[3, 76]; and (d) developmental disorders affect-
O.J. Baker, DDS, PhD ing ectodermal tissues and organs [59].
School of Dentistry, The University of Utah,
383 Colorow Dr Room 289A, Salt Lake City,
Common treatments for hyposalivation
UT 84108-1201, USA include the use of secretory agonists and saliva
e-mail: [email protected] substitutes, both of which provide only t emporary

134 O.J. Baker
relief [75]. Experimental treatments such as use c ervical cancer cells [26]. However, many of the
of adenoviral [9] and ultrasound-assisted trans- current immortalized cell lines are intentionally
fection of aquaporin-1 [77] or stem cells [54] created by introduction of a virus that greatly
offer the possibility of significant improvement. extends the number of viable cell replications and
Alternatively, salivary gland transplantation allows for the possibility that a mutation might
remains an attractive option if other treatments occur that could result in a truly immortalized
using a native salivary gland prove ineffective, cell. For instance, overexpression of the large T
unreliable, or fail to maintain treatment gains. antigen of the simian virus 40 (SV40) represses
However, transplantation of a naturally occurring the retinoblastoma (Rb) and p53 genes (both of
salivary gland is problematic because finding which are critical controllers of the cell cycle)
donors, as with any organ donation, is a complex [62]. Other viral genes include those from the
and unpredictable process [78] and the chances human papilloma virus family, which also target
of rejection by the host are high [17]. Rb and p53 [8]. Another method to induce cell
Despite the difficulties involved in salivary immortalization involves the use of the gene
gland implantation, alternative approaches pro- telomerase (hTERT), which extends the DNA
vide new possibilities for making this treatment sequence of telomeres and allows for infinite cell
viable. In this chapter, current trends in the divisions [44]. The use of cell lines to bioengi-
emerging field of salivary gland bioengineering neer a salivary gland would appear to offer an
involving cell lines, primary cells, and stem/pro- attractive alternative to development and implan-
genitor cells will be summarized. Note that par- tation of an artificial salivary gland. However, no
ticular variations on some strategies are too currently available cell line fully recapitulates the
numerous to cover individually; however, no sig- morphological and functional features of the
nificant strategy has been omitted, and multiple native salivary acinar cells; moreover, some are
examples within each category are provided. tumorigenic [62]. Consequently, we believe cell
lines are currently not suitable for bioengineering
and/or implantation and that their use should be
7.2 ommonly Used Salivary Cell
C limited to understanding basic physiological
Lines mechanisms in native glands. Below we evaluate
the advantages and limitations of commonly used
Normal cells are unable to replicate beyond sev- salivary cell lines (see Table 7.1).
eral rounds of proliferation due to progressive
telomere shortening with each successive round.
Specifically, cellular senescence (i.e., naturally 7.2.1 HSY
occurring loss of a cell’s ability to divide and
grow) results when the telomeres reach a criti- Human parotid gland (HSY) is an epithelial cell
cally reduced length and DNA is damaged [14]. line derived from the acinar-intercalated duct
An immortalized cell line is a population of cells region of the human parotid gland [81]. These
from a multicellular organism which would nor- cells are cuboidal in shape, have papillary infold-
mally not proliferate indefinitely but, due to ings as well as microvilli on their free border, and
mutation, has evaded normal cellular senescence exhibit low levels of secretory granules. HSY cells
and undergoes continuous division [49]. maintain polarized monolayer organization [81],
Immortalized cells are different from stem cells, which is critical for engineering a gland capable of
which can also divide indefinitely but form a nor- fluid secretion. They respond to muscarinic and
mal part of the development of a multicellular β-adrenergic autonomic agonists to increase intra-
organism (i.e., are not attributable to mutation or cellular free calcium concentration ([Ca2+]i) and
intentional in vitro modification). Immortalized cyclic AMP ([cAMP]i), respectively [61], features
cell lines can be derived from tumors, such as that are essential for saliva secretion in vivo [73]. It
HeLa cells that were obtained from human has been proposed that salivary intercalated ducts
7 Current Cell Models for Bioengineering Salivary Glands 135
Table 7.1 Commonly used immortalized salivary cell lines

Response to
Amylase secretory
Cell line Source Cell polarity expression agonist Reference
HSY Human parotid + + + Yanagawa et al. [81]
adenocarcinoma
HSG Human submandibular + + + Shirasuna et al. [69]
gland
SMIE Rat submandibular gland + – + He et al. [28, 29]
RSMT-A5 Rat submandibular gland – – – Brown [12]
SMG-C6 Rat submandibular gland + Unknown Unknown Quissell et al. [65]
SMG-C10 Rat submandibular gland + Unknown Unknown Quissell et al. [65]
Par-C10 Rat parotid gland + – + Quissell et al. [66]
Par-C5 Rat parotid gland + – – Quissell et al. [66]
GManSV tdTomato mouse Unknown Unknown Unknown Furukawa et al. [24]
submandibular gland
Modified from Nelson et al. [56]
function as the reservoir for progenitor cells in HSG cells have been used as an in vitro model
salivary glands [55]. More recently, FGF10- for salivary gland secretion, morphology, and
induced ERK1/2 phosphorylation in HSY cells regeneration [31, 37, 38, 45, 70]. Moreover, sev-
was noted, indicating that they respond to growth eral features indicate that they are a potential
factors linked to salivary gland morphogenesis source for developing an artificial salivary gland.
[80]. Given that HSY cells have similar morpho- Specifically, HSG cells differentiate into acinar
logical features to those of intercalated duct cells, structures and express amylase when cultured on
it would be interesting to study whether non-trans- Matrigel [31]. Additionally, they have an innate
fected HSY cells may behave like progenitor stem capacity to increase [Ca2+]i in response to musca-
cells when transplanted in vivo. If this were the rinic and purinergic agonists [52]. Furthermore,
case, they could be used for differentiation studies HSG cells can be modulated by regulators of
(with the aim of developing an artificial salivary apoptosis, thereby providing researchers with a
gland), though we believe such applications have shutoff mechanism for tumorigenic cells in vivo
not yet been explored. [23]. Finally, they are regulated by growth factor
receptors (e.g., EGFR) that may be used to pro-
mote repair or regeneration of salivary tissue [67].
7.2.2 HSG Despite the many advantages of using HSG
for salivary gland bioengineering, there are sig-
Human submandibular gland (HSG) is a neoplas- nificant drawbacks as well. Specifically, HSG
tic intercalated duct cell line derived from an irra- cells grown on plastic appear unable to form TJs.
diated human submandibular gland (SMG) [69]. Interestingly, they have been noted to express TJs
Histologically, HSG cells can be presented either and aquaporins when grown on permeable
in cuboidal or conical shape and have easily vis- supports coated with Matrigel [45]. However,

ible desmosomes with sporadic tight junction future studies will be necessary to understand the
(TJ) complexes [69]. They appear capable of barrier properties of this model (i.e., character-
fluid and protein secretion due to the presence of ization of TJ morphology and in-depth analysis
intercellular connections, microvilli, and protein of monolayer permeability). Finally, a challenge
synthesis machinery. presented by the use of HSG is the frequent
136 O.J. Baker
c ontamination of this cell line with HeLa cells genome into rat SMG acinar cells [65]. Only two of
[18], which may be overcome through proper the formed clones, termed SMG-C6 and SMG-C10,
control protocols, as discussed below. were found to be both well differentiated and of epi-
thelial origin [65]. Structurally, these cell lines
polarize due to their ability of form TJs and desmo-
7.2.3 SMIE somes [65]. Additionally, secretory features (i.e.,
domes, granules, and canaliculi) are observed
Rat submandibular gland acinar epithelial (SMIE) within them [65]. Functionally, SMG-C6 respond
is an immortalized epithelial cell line derived from to muscarinic and purinergic agonists by increasing
rat SMG [29]. This cell line was originally named [Ca2+]i [42]. Furthermore, both SMG-C6 and SMG-
RSMG and later renamed to SMIE because of the C10 respond to β-adrenergic agonists by increasing
adenovirus (12S E1A gene product) used to [cAMP]i [42]. Of the two cell lines, SMG-C6 seems
immortalize the cells [28]. SMIE cells were estab- to be better differentiated than SMG-C10 due to a
lished to study polarized functions in salivary epi- greater quantity of secretory cellular structures and
thelium due to their ability to form TJs when a more stable [Ca2+]i release [65]. Moreover, SMG-
grown on permeable supports [28, 51]. Structurally, C6 and SMG-C10 lines develop a high transepithe-
these cells resemble salivary glandular epithelium lial resistance when grown on collagen-coated
with immature lumens [28]. SMIE cells display polycarbonate filters [16].
selective barrier function and fluid transport [28, In addition to the signaling processes detailed
51] and have also been shown to secrete luciferase, above, additional functions for SMG-C6 and SMG-
when transfected with a pGL3-EGFSP construct C10 have been observed. Specifically, both cell
[1]. These studies indicate that SMIE cells can be lines are excellent models to study sodium channels
used to modulate fluid and protein secretion for and expression of the epithelial sodium channel
future bioengineering applications. protein (ENaC), given their ability to modulate
sodium transport when grown in a culture medium
lacking glucocorticoids or mineralocorticoids [74].
7.2.4 RSMT-A5 Studies using SMG-C10 cells also indicated that the
vanilloid receptor 4 (TRPV4) was functionally con-
Rat submandibular duct epithelial (RSMT-A5) also nected to aquaporin-5 volume [6]. More recently, a
known as (A5) cell line was derived through trans- molecule involved in the regulation of energy
formation of rat SMG cells by way of treatment with metabolism and inflammatory responses (adiponec-
3-methylcholanthrene [12]. This cell line displays a tin) was found to promote fluid secretion in SMG-
ductal epithelium phenotype and expresses a high C6 cells [21]. The above studies indicate SMG-C6
density of α1-andrenergic receptors with metabolic and SMG-C10 are potential candidates for under-
behavior similar to smooth muscle cells [30]. Recent standing regulation of cell volume and secretory
studies demonstrated that A5 cells were able to function in salivary gland bioengineering. Finally,
uptake nanoparticles [27]. Taken together, these studies using SMG-C6 cells demonstrated that
results indicate that A5 cells are useful for receptor apoptosis could be modulated through a Fas-
characterization and signaling studies; however, they mediated pathway [2], indicating their potential for
might not be suitable for the study of protein secre- in vivo transplantation studies without the risk of
tion due to transfection difficulties [1]. uncontrolled cellular growth.
7.2.5 SMG-C6 and SMG-C10 7.2.6 Par-C10 and Par-C5
Rat submandibular gland epithelial (SMG-C) cell Following development of the SMG-C6 and
lines were isolated through transfection of a SMG-C10 cell lines, another study was done to
replication-
defective simian virus 40 (SV40) isolate cells from a rat parotid gland [66]. Similar
to the SMG-C cell line development, parotid sali- 7.2.7 Fluorescent Salivary Cell Lines
vary cells were transfected with an origin-
defective SV40 plasmid [66]. Morphology and Recent studies generated a mesenchymal stem
receptor-mediated [Ca2+]i responses were used as cell line from transgenic mice overexpressing the
a screening technique to monitor cell differentia- red fluorescent protein tdTomato (tdTomato
tion [66]. The rat parotid (Par-C5 and Par-C10) mice). Specifically, they immortalized subman-
cell lines were found to exhibit a significant ele- dibular gland-derived stem cells with the SV40
vation of [Ca2+]i in response to cholinergic, mus- large T antigen mouse submandibular (GManSV)
carinic, and α1-adrenergic agonists [41]. These [24]. These cells exhibited high cell migration
cell lines demonstrated an increase of [cAMP]i in rates, a spindle-shaped fibroblastic morphology,
response to α1-adrenergic agonists [66] as well as and expression of mesenchymal stem cell mark-
increases of [Ca2+]i in response to M3R musca- ers. Moreover, they retained multipotent stem
rinic agonists [10]. No functional amylase cell characteristics, as evidenced by their ability
expression has been observed in Par-C10 cells to differentiate into both osteogenic and adipo-
when grown either on plastic or growth factor- genic lineages [24]. Taken together, these results
reduced (GFR) Matrigel, although an interesting indicate that GManSV cells are useful both for
study on the Par-C 3-9 clones reported a 16-fold imaging and regeneration studies.
increase in amylase content following incubation
with rat serum [83]. However, further studies are
needed to consistently demonstrate improved 7.2.8 Caution When Using Cell Lines
amylase production in these cell lines.
The Par-C10 cell line has been widely studied, We bear a responsibility as researchers to verify the
given its ability to form polarized monolayers, correct use of cell lines and avoid misidentification
which makes it a great model for studying barrier by regularly checking to see that they correspond to
function and ion transport in salivary epithelium their original sources. To that end, short tandem
[72]. Specifically, ion secretion in Par-C10 cells repeat profiling (a method used to compare specific
has been well characterized, thereby establishing loci on DNA from two or more samples) offers an
that it is regulated by basolateral α1-adrenergic excellent solution. Additionally, clear guidelines for
and muscarinic cholinergic receptors as well as authentication testing, documentation of cell line
apical P2Y2 receptors [72]. Furthermore, Par- provenance, and ongoing validation will help ensure
C10 cells express Na+/H+ exchangers, Na+-HCO3- that human cell lines are effective and representa-
cotransporters, and anion exchange proteins on tive models for biomedical research [15].
their basolateral surfaces [20]. These proteins,
which regulate transepithelial transport, are sen-
sitive to changes in both [Ca2+]i and [cAMP]i 7.3 Use of Primary Salivary Cells
[20]. Par-C10 single cells also are capable of
forming salivary spheres when grown on Primary cells are taken directly from living tissue
Matrigel. Under these conditions, Par-C10 (e.g., during biopsies) and established for growth
acinar-like spheres expressed TJs, aquaporins, in vitro. These cells have undergone very few pop-
ion transporters, and muscarinic receptor 3 [7]. ulation doublings and therefore represent well the
These features make Par-C10 acinar-like spheres main functional components of the tissue from
an intriguing model to characterize cell volume which they were derived. As such, the primary
regulation and ion secretion in salivary epithe- cells represent and offer a good option for creating
lium. Moreover, recent studies demonstrated that an artificial salivary gland because they closely
recombinant adenovirus vectors can modify Par- resemble native tissue (see Table 7.2).
C10 cells [11], thereby making them useful not Primary salivary cells for in vitro studies are
only for bioengineering purposes but also as a obtained through cell isolation of salivary glands
gene therapy model. (SMG, parotid, sublingual, and minor salivary
138 O.J. Baker
Table 7.2 Commonly used human and mouse primary salivary cells
Amylase Response
Source Cell expression to secretory agonist References
Human submandibular gland + Unknown + Tran et al. [71], Pradhan-Bhatt et al. [63, 64]
Human minor salivary glands + Unknown + Jang et al. [33]
Human parotid gland + + Unknown Joraku et al. [34]
Mouse parotid gland + + + McCall et al. [50]
Mouse submandibular gland + + + Leigh et al. [40], Maruyama et al. [47]
c
a
d e
Fig. 7.1 Salivary cell isolation and tissue culture. (a) at 150 × g and supernatant is removed. Cells are resus-
Human and (b) mouse submandibular glands are homog- pended and plated on (d) eight-well chambers mounted on
enized in a solution containing dispersion enzymes (e.g., a cover glass and filled with various scaffolds (e.g., lam-
hyaluronidase and collagenase) using a (c) tissue dissocia- inin-111 or hydrogels) within which salivary cell clusters
tor. Following dissociation, cells are centrifuged for 5 min organize into (e) salivary spheres with hollow lumens
glands) using enzymes and mechanical dissociation, these conditions form cell monolayers with key
by which a cell dispersion of predominantly acinar features indicative of a functional gland (e.g., tight
and ductal cells is obtained and plated on multiple junctions, microvilli, and secretory granules) as
substrates (e.g., permeable supports or Matrigel) well as barrier function regulation [71]. More
(see Fig. 7.1). The limitations of this method include recently, human minor salivary glands isolated
the presence of multiple cell types (i.e., a cell disper- with an explant technique were grown on colla-
sion contains acinar, ductal, progenitor, stem, and gen-coated permeable supports, and protein mark-
myoepithelial cells), slow growth, dedifferentiation, ers for progenitor and acinar cells were expressed.
and a finite life span [40, 46, 47, 54]. However, Importantly, as the calcium concentration
ongoing studies aim to improve the quality of pri- increased within the growth medium, these cells
mary cells by optimizing the cell isolation methods acquired a polarized acinar-like phenotype (i.e.,
and using biomaterials that better mimic the extra- increased expression of α-amylase) and demon-
cellular environment in which cells must be grown. strated intact secretion and barrier function [33].
The results shown above indicate that human sali-
vary cells grown as monolayers mimic acinar
7.3.1 Salivary Cell Monolayers functioning and are useful for studying fluid and
electrolyte secretion; however, further study is
A model for secretion studies involves the growth needed to develop bioengineering applications
of human SMG on permeable supports, also (e.g., providing for formation into a branching pat-
known as transwells. SMG cells grown under tern, as noted in functional salivary glands).
7.3.2 Salivary Spheres 7.4 alivary Gland-Derived Stem

S
and Progenitor Cells
Single human parotid cells can be grown on col-
lagen and Matrigel to form salivary spheres with Cells capable of growing more specialized cells
hollow lumens. Under these conditions, cells (e.g., stem cells and progenitor cells) offer sig-
exhibit markers of acinar differentiation, includ- nificant possibilities for salivary gland treatment
ing α-amylase, aquaporin-5, and apical TJ but present serious challenges as well. Stem
expression [34]. Human SMG cells also form cells are undifferentiated but can develop into
salivary spheres when grown on hyaluronic acid specialized cells and reproduce indefinitely to
hydrogels. Under these conditions, cells express produce more cells with the same properties
TJ proteins and α-amylase [63] and respond to [32]. Similarly, progenitor cells are early
neurotransmitters as well [64]. descendants of stem cells that can also differen-
Murine salivary cells form spheres with hol- tiate to form one or more kinds of cells; how-
low lumens when grown on Matrigel or fibrin ever, they cannot divide and reproduce
hydrogels. Previous studies showed that fibrin indefinitely and are more limited in the kinds of
hydrogels polymerized with growth factors cells they can form [79]. Previous studies indi-
(i.e., EGF and IGF-1) induced salivary gland cated that intercalated ducts of salivary glands
differentiation, as indicated by increased levels are enriched with both stem and progenitor cells
of α-amylase expression and response to sali- capable of differentiating and replacing dam-
vary secretory agonists [50]. Further studies aged tissues, thereby contributing to the mainte-
using both parotid and SMG cells demonstrated nance of acinar cells. Additionally, recent
that SMG cell clusters formed more organized studies indicated that the primary mechanism
and larger structures than were formed by for maintaining salivary glands is through dupli-
parotid gland cell clusters. However, both SMG cation of acinar cells [5]. These results indicate
and parotid gland cell clusters maintained that all of these cells (i.e., stem, progenitor, and
α-amylase expression, presence of secretory acinar cells) are candidates for the study of sali-
granules, TJs, and agonist-induced secretory vary gland regeneration and, as such, may have
responses over time [40]. These results indicate bioengineering applications.
that mouse SMG cell clusters are more promis-
ing for the development of a bioengineered sali-
vary gland than parotid gland cell clusters, as 7.4.1 M
arkers of Stem
they form more organized and functional and Progenitor Cells
spheres. Moreover, we recently demonstrated
that conditioned medium (from mesenchymal As indicated above, salivary glands contain pro-
stem cells) enhanced cell organization and genitor cells, and multiple markers have been
multi-lumen formation [47]. These studies indi- used to characterize the various progenitor cell
cate that soluble signals secreted by these populations within them. One such marker is
human mesenchymal stem cells promote for- Ascl3, a transcription factor essential for tissue
mation of a glandular shape. The ability of such development and differentiation. Specifically,
mammalian salivary cells to form salivary transplanted Ascl3-expressing cells were shown
spheres has been useful for understanding cell to induce differentiation and tissue repair in sali-
assembly mechanisms, secretory function, and vary glands and actively promote regeneration
cell behavior in culture. However, further stud- [68]. Moreover, Ascl3 was found to be expressed
ies are needed to better characterize the various in ductal cells and demilune caps of SMG [13].
structural (i.e., acinar, ductal, or myoepithelial) These studies indicate Ascl3-expressing cells
and functional (i.e., serous and mucous) cell contribute to mature salivary gland tissue mainte-
types present in these salivary spheres during nance and are likely useful for salivary gland bio-
time in culture. engineering studies.
140 O.J. Baker
K5 (a structural protein that forms cytoskele- 7.4.2 S

tem and Progenitor Cells
ton intermediate filaments) shows some potential In Vivo
as a marker for salivary gland progenitor cells.
Specifically, cells expressing K5 (also known as Bone marrow-derived mesenchymal stem cells
K5+ cells) have been found in ducts of adult sali- (BM-MSC) can be used for stem cell usage.
vary glands, which contain progenitor cells [19]. Recent studies cocultured BM-MSC with pri-
Likewise, K5+ cells give rise to acinar and ductal mary mouse SMG cells, which led to transdiffer-
cells shortly after birth [36]. However, in order to entiation of BM-MSC into a salivary-like
isolate progenitor cells, co-localization of K5 phenotype, as indicated by the expression of
with other cell markers is necessary [43], indicat- α-amylase, muscarinic type 3 receptor, aquapo-
ing that K5 expression alone may not be a satis- rin-5, and cytokeratin 19. These studies success-
factory marker for salivary progenitor cells. fully identified proteins involved in the process of
Another marker for progenitor cells is Sox2, BM-MSC differentiating into salivary gland epi-
a member of the Sox family of transcription fac- thelial cells; however, further analyses are neces-
tors which has been shown to play key roles in sary to determine the function of these factors in
many stages of mammalian development [39]. mesenchymal stem cell reprogramming [60].
Recent studies showed that an absence of Sox2+ The use of embryonic salivary cells is an alter-
cells in mice resulted in compromised epithelial native option for restoring salivary glands in vivo,
integrity (including salivary glands) and death as they are a good source of progenitor and stem
[4]. These results indicate that Sox2+ progenitor cells. Specifically, previous studies showed that
cells are important for tissue regeneration and mouse embryonic salivary cells (i.e., submandib-
survival of mice. ular, sublingual, and parotid glands) grown in an
Integrins such as α6β1 have been found to organ culture can be transplanted in vivo.
be markers for salivary progenitor cells in However, significant issues appear to limit the
rats [58], and cells expressing them were used utility of this strategy, including a diminished
to establish an immortalized cell line of rat gland size and a brief period of survival for ani-
salivary progenitor cells [82]. This cell line mal subjects following implantation [57].
is capable of differentiating into both acinar- Finally, several studies have shown that stem
and ductal-like structures and has the ability cells derived from postnatal salivary cells can be
to be modulated when grown on Matrigel- used to restore damaged salivary glands in vivo
based scaffolds; however, cells grown under [54, 55]. Unfortunately, as mentioned above,
these conditions display uncontrolled growth. salivary gland specimens express low numbers
Consequently, further studies are needed to of stem and progenitor cells (e.g., c-Kit+ cells),
determine whether the acinar- and ductal-like thereby limiting their utility for clinical pur-
structures generated from this cell line respond poses. In order to generate enough stem/progeni-
to salivary secretory agonists for purposes of tor cells for transplantation, recent studies using
growth suppression. mouse salivary glands expanded the number of
Ductal cells in salivary glands also have been stem cells ex vivo. Specifically, salivary gland
shown to express several stem cell markers (i.e., sphere-derived single cells were differentiated
CD24, CD49f, CD133, and c-Kit+) [55]. in vitro into distinct lobular or ductal/lobular
Particularly, c-Kit was definitively established as organoids containing multiple salivary gland cell
a stem cell marker and therefore gained the high- lineages [53]. This study indicates that func-
est priority; however, flow cytometric analysis of tional salivary gland stem cells can now be puri-
cells obtained from SMG indicated that only fied and expanded ex vivo from single salivary
0.058 % of salivary cells expressed c-Kit [55]. As cells; however, follow-up studies using human
such, c-Kit appears not to be an ideal marker for cells will be needed to translate these findings
stem cell isolation in salivary glands. for clinical work.
a b
Fig. 7.2 Decellularization of a submandibular gland. (a) fold. (c) Primary cells are isolated in the second gland and
Two submandibular glands from a single rat are dissected. (d) combined with the extracellular matrix, then cultured
(b) Cells from the first gland are completely removed, and for 14 days, to generate a salivary structure in vitro
the remaining extracellular matrix is kept for use as a scaf-
7.5 Scaffolds References
Salivary cells can be grown on various sub- 1. Aframian DJ, Amit D, David R, Shai E, Deutsch D,
strates (e.g., permeable supports, collagen, Honigman A, Panet A, Palmon A. Reengineering sali-
vary gland cells to enhance protein secretion for use
Matrigel, hydrogels, etc.), as detailed above. in developing artificial salivary gland device. Tissue
Recently, a study showed that rat SMG can be Eng. 2007;13:995–1001.
decellularized by detergent immersion, thereby 2. Aiba-Masago S, Masago R, Vela-Roch N, Talal N,
removing all cells from the tissue and leaving Dang H. Fas-mediated apoptosis in a rat acinar cell
line is dependent on caspase-1 activity. Cell Signal.
only a scaffold composed of extracellular 2001;13:617–24.
matrix proteins (Fig. 7.2). Using this scaffold 3. Aliko A, Wolff A, Dawes C, Aframian D, Proctor
as a support, primary SMG cells were reseeded G, Ekstrom J, Narayana N, Villa A, Sia YW, Joshi
and cultured in vitro. Results from this study RK, Mcgowan R, Beier Jensen S, Kerr AR, Lynge
Pedersen AM, Vissink A. World Workshop on Oral
demonstrated that recellularized structures Medicine VI: clinical implications of medication-
express salivary differentiation markers in vitro induced salivary gland dysfunction. Oral Surg Oral
[25], thereby offering a promising option for Med Oral Pathol Oral Radiol. 2015;120:185–206.
salivary gland bioengineering if secretory func- 4. Arnold K, Sarkar A, Yram MA, Polo JM, Bronson
R, Sengupta S, Seandel M, Geijsen N, Hochedlinger
tion can be demonstrated in future studies. K. Sox2(+) adult stem and progenitor cells are impor-
tant for tissue regeneration and survival of mice. Cell
Conclusion Stem Cell. 2011;9:317–29.
Construction of an artificial salivary gland is 5. Aure MH, Konieczny SF, Ovitt CE. Salivary gland
homeostasis is maintained through acinar cell self-
an attractive alternative to repair or regenera- duplication. Dev Cell. 2015;33:231–7.
tion of native glands in patients with hyposali- 6. Aure MH, Roed A, Galtung HK. Intracellular
vation. To that end, issues related to the cells Ca2+ responses and cell volume regulation upon
to be grown (i.e., longevity, differentiation, cholinergic and purinergic stimulation in an
immortalized salivary cell line. Eur J Oral Sci.
function) and the environment in which they 2010;118:237–44.
are to be cultured (i.e., ability to produce dif- 7. Baker OJ, Schulz DJ, Camden JM, Liao Z, Peterson
ferentiated structures, biocompatibility, and TS, Seye CI, Petris MJ, Weisman GA. Rat parotid
limitation of tumorigenicity) must be resolved. gland cell differentiation in three-dimensional culture.
Tissue Eng Part C Methods. 2010;16(5):1135–44.
8. Band V, Zajchowski D, Kulesa V, Sager R. Human papil-
Acknowledgments This work was supported by the loma virus DNAs immortalize normal human mammary
NIH-NIDCR grants R01DE022971, R01DE021697, epithelial cells and reduce their growth factor require-
R01DE021697S1, and R01DE021697S2. ments. Proc Natl Acad Sci U S A. 1990;87:463–7.
142 O.J. Baker
9. Baum BJ, Alevizos I, Zheng C, Cotrim AP, Liu S, 22.

Epstein JB, Beier Jensen S. Management of
Mccullagh L, Goldsmith CM, Burbelo PD, Citrin DE, Hyposalivation and Xerostomia: Criteria for
Mitchell JB, Nottingham LK, Rudy SF, Van Waes C, Treatment Strategies. Compend Contin Educ Dent.
Whatley MA, Brahim JS, Chiorini JA, Danielides S, 2015;36:600–3.
Turner RJ, Patronas NJ, Chen CC, Nikolov NP, Illei 23. Fukuda M, Tanaka S, Suzuki S, Kusama K, Kaneko T,
GG. Early responses to adenoviral-mediated trans- Sakashita H. Cimetidine induces apoptosis of human
fer of the aquaporin-1 cDNA for radiation-induced salivary gland tumor cells. Oncol Rep. 2007;17:673–8.
salivary hypofunction. Proc Natl Acad Sci U S A. 24. Furukawa S, Kuwajima Y, Chosa N, Satoh K, Ohtsuka
2012;109:19403–7. M, Miura H, Kimura M, Inoko H, Ishisaki A, Fujimura
10. Bockman CS, Bradley ME, Dang HK, Zeng W,
A, Miura H. Establishment of immortalized mesen-
Scofield MA, Dowd FJ. Molecular and pharmaco- chymal stem cells derived from the submandibular
logical characterization of muscarinic receptor sub- glands of tdTomato transgenic mice. Exp Ther Med.
types in a rat parotid gland cell line: comparison 2015;10:1380–6.
with native parotid gland. J Pharmacol Exp Ther. 25. Gao Z, Wu T, Xu J, Liu G, Xie Y, Zhang C, Wang
2001;297:718–26. J, Wang S. Generation of Bioartificial Salivary Gland
11. Bori E, Racz G, Burghardt B, Demeter I, Hegyesi Using Whole-Organ Decellularized Bioscaffold.
O, Varga G, Foldes A. ParC-10 cells for model- Cells Tissues Organs. 2014;200:171–80.
ling parotid gland tissue reorganization. Fogorv Sz. 26. Gey GO, Coffman WD, Kubicek MT. Tissue cul-
2014;107:99–105. ture studies of the proliferative capacity of cervi-
12. Brown AM. In vitro transformation of subman-
cal carcinoma and normal epithelium. Cancer Res.
dibular gland epithelial cells and fibroblasts of 1952;12:264–5.
adult rats by methylcholanthrene. Cancer Res. 27. Goldman EB, Zak A, Tenne R, Kartvelishvily E,

1973;33:2779–89. Levin-Zaidman S, Neumann Y, Stiubea-Cohen R,
13. Bullard T, Koek L, Roztocil E, Kingsley PD, Mirels L, Palmon A, Hovav AH, Aframian DJ. Biocompatibility
Ovitt CE. Ascl3 expression marks a progenitor popu- of tungsten disulfide inorganic nanotubes and
lation of both acinar and ductal cells in mouse salivary fullerene-like nanoparticles with salivary gland cells.
glands. Dev Biol. 2008;320:72–8. Tissue Eng Part A. 2015;21:1013–23.
14. Campisi J, d’Adda di Fagagna F. Cellular senescence: 28. He X, Kuijpers GA, Goping G, Kulakusky JA, Zheng
when bad things happen to good cells. Nat Rev Mol C, Delporte C, Tse CM, Redman RS, Donowitz M,
Cell Biol. 2007;8:729–40. Pollard HB, Baum BJ. A polarized salivary cell mono-
15. Capes-Davis A, Reid YA, Kline MC, Storts DR,
layer useful for studying transepithelial fluid move-
Strauss E, Dirks WG, Drexler HG, Macleod RA, ment in vitro. Pflugers Arch. 1998;435:375–81.
Sykes G, Kohara A, Nakamura Y, Elmore E, Nims 29. He XJ, Frank DP, Tabak LA. Establishment and char-
RW, Alston-Roberts C, Barallon R, Los GV, Nardone acterization of 12S adenoviral E1A immortalized
RM, Price PJ, Steuer A, Thomson J, Masters JR, rat submandibular gland epithelial cells. Biochem
Kerrigan L. Match criteria for human cell line authen- Biophys Res Commun. 1990;170:336–43.
tication: where do we draw the line? Int J Cancer. 30. He XJ, Wu XZ, Brown AM, Wellner RB, Baum

2013;132:2510–9. BJ. Characteristics of alpha 1-adrenergic receptors
16. Castro R, Barlow-Walden L, Woodson T, Kerecman in a rat salivary cell line, RSMT-A5. Gen Pharmacol.
JD, Zhang GH, Martinez JR. Ion transport in an 1989;20:175–81.
immortalized rat submandibular cell line SMG-C6. 31. Hoffman MP, Kibbey MC, Letterio JJ, Kleinman

Proc Soc Exp Biol Med. 2000;225:39–48. HK. Role of laminin-1 and TGF-beta 3 in acinar dif-
17. Chong AS, Alegre ML. The impact of infection and ferentiation of a human submandibular gland cell line
tissue damage in solid-organ transplantation. Nat Rev (HSG). J Cell Sci. 1996;109(Pt 8):2013–21.
Immunol. 2012;12:459–71. 32. Hsu YC, Fuchs E. A family business: stem cell prog-
18. Committee I. C. L. A. Database of cross-contaminated eny join the niche to regulate homeostasis. Nat Rev
or misidentified cell lines [Online]. 2014. Available: Mol Cell Biol. 2012;13:103–14.
http://iclac.org/databases/cross-contaminations/2014. 33. Jang SI, Ong HL, Gallo A, Liu X, Illei G, Alevizos
19. Coppes RP, Stokman MA. Stem cells and the repair I. Establishment of functional acinar-like cul-
of radiation-induced salivary gland damage. Oral Dis. tures from human salivary glands. J Dent Res.
2011;17:143–53. 2015;94:304–11.
20. Demeter I, Szucs A, Hegyesi O, Foldes A, Racz GZ, 34. Joraku A, Sullivan CA, Yoo J, Atala A. In-vitro recon-
Burghardt B, Steward MC, Varga G. Vectorial bicar- stitution of three-dimensional human salivary gland
bonate transport by Par-C10 salivary cells. J Physiol tissue structures. Differentiation. 2007;75:318–24.
Pharmacol. 2009;60(Suppl 7):197–204. 35. Knecht R. Current treatment concepts in head and
21. Ding C, Li L, Su YC, Xiang RL, Cong X, Yu HK, neck cancer: Highlights of the 2015 Annual Meeting
Li SL, Wu LL, Yu GY. Adiponectin increases secre- of the American Society for Clinical Oncology. HNO.
tion of rat submandibular gland via adiponectin 2015;63:605.
receptors-mediated AMPK signaling. PLoS One. 36. Knox SM, Lombaert IM, Reed X, Vitale-Cross L,
2013;8:e63878. Gutkind JS, Hoffman MP. Parasympathetic i nnervation
maintains epithelial progenitor cells during salivary sion of claudin-4 in tight junctions of submandibular
organogenesis. Science. 2010;329:1645–7. gland cells. Cell Tissue Res. 2008;334:255–64.
37. Kurokawa R, Kyakumoto S, OTA M. Autocrine
52. Nagy K, Szlavik V, Racz G, Ovari G, Vag J, Varga
growth factor in defined serum-free medium of human G. Human submandibular gland (HSG) cell line as
salivary gland adenocarcinoma cell line HSG. Cancer a model for studying salivary gland Ca2+ signalling
Res. 1989;49:5136–42. mechanisms. Acta Physiol Hung. 2007;94:301–13.
38. Lam K, Zhang L, Bewick M, Lafrenie RM. HSG 53. Nanduri LS, Baanstra M, Faber H, Rocchi C, Zwart
cells differentiated by culture on extracellular matrix E, de Haan G, Van OS R, Coppes RP. Purification and
involves induction of S-adenosylmethione decarbox- ex vivo expansion of fully functional salivary gland
ylase and ornithine decarboxylase. J Cell Physiol. stem cells. Stem Cell Rep. 2014;3:957–64.
2005;203:353–61. 54. Nanduri LS, Lombaert I, van der Zwaag M, Faber
39. Lefebvre V, Dumitriu B, Penzo-Mendez A, Han Y, H, Brunsting JF, van OS RP, Coppes RP. Salisphere
Pallavi B. Control of cell fate and differentiation by Sry- derived c-Kit< sup>+ cell transplantation restores
related high-mobility-group box (Sox) transcription fac- tissue homeostasis in irradiated salivary gland.
tors. Int J Biochem Cell Biol. 2007;39:2195–214. Radiother Oncol. 2013a;108:458–63.
40. Leigh NJ, Nelson JW, Mellas RE, Mccall AD, Baker 55. Nanduri LS, Maimets M, Pringle SA, van der Zwaag
OJ. Three-dimensional cultures of mouse submandib- M, van OS RP, Coppes RP. Regeneration of irradiated
ular and parotid glands: a comparative study. J Tissue salivary glands with stem cell marker expressing cells.
Eng Regen Med. 2014. http://onlinelibrary.wiley. Radiother Oncol. 2011;99:367–72.
com/doi/10.1002/term.1952/abstract;jsessionid=08D 56. Nelson J, Manzella K, Baker OJ. Current cell models
6882890A70711C3E23CA9FAB89BD1.f04t02 for bioengineering a salivary gland: a mini-review of
41. Liu X, Mork AC, Sun X, Castro R, Martinez JR, emerging technologies. Oral Dis. 2013;19:236–44.
Zhang GH. Regulation of Ca(2+) signals in a parotid 57. Ogawa M, Oshima M, Imamura A, Sekine Y, Ishida K,
cell line Par-C5. Arch Oral Biol. 2001;46:1141–9. Yamashita K, Nakajima K, Hirayama M, Tachikawa
42. Liu XB, Sun X, Mork AC, Dodds MW, Martinez JR, T, Tsuji T. Functional salivary gland regeneration by
Zhang GH. Characterization of the calcium signaling transplantation of a bioengineered organ germ. Nat
system in the submandibular cell line SMG-C6. Proc Commun. 2013;4:1–10.
Soc Exp Biol Med. 2000;225:211–20. 58. Okumura K, Nakamura K, Hisatomi Y, Nagano

43. Lombaert IM, Knox SM, Hoffman MP. Salivary
K, Tanaka Y, Terada K, Sugiyama T, Umeyama K,
gland progenitor cell biology provides a rationale Matsumoto K, Yamamoto T, Endo F. Salivary gland
for therapeutic salivary gland regeneration. Oral Dis. progenitor cells induced by duct ligation differenti-
2011;17:445–9. ate into hepatic and pancreatic lineages. Hepatology.
44. Lu MH, Liao ZL, Zhao XY, Fan YH, Lin XL, Fang 2003;38:104–13.
DC, Guo H, Yang SM. hTERT-based therapy: a uni- 59. Pagnan NA, Visinoni AF. Update on ectodermal dys-
versal anticancer approach (Review). Oncol Rep. plasias clinical classification. Am J Med Genet A.
2012;28:1945–52. 2014;164A:2415–23.
45. Maria OM, Maria O, Liu Y, Komarova SV, Tran
60.
Park YJ, Koh J, Gauna AE, Chen S, Cha
SD. Matrigel improves functional properties of human S. Identification of regulatory factors for mesenchy-
submandibular salivary gland cell line. Int J Biochem mal stem cell-derived salivary epithelial cells in a co-
Cell Biol. 2011a;43:622–31. culture system. PLoS One. 2014;9:e112158.
46.
Maria OM, Zeitouni A, Gologan O, Tran 61. Patton LL, Pollack S, Wellner RB. Responsiveness
SD. Matrigel improves functional properties of pri- of a human parotid epithelial cell line (HSY) to auto-
mary human salivary gland cells. Tissue Eng Part A. nomic stimulation: muscarinic control of K+ trans-
2011b;17:1229–38. port. In Vitro Cell Dev Biol. 1991;27A:779–85.
47. Maruyama CL, Leigh NJ, Nelson JW, Mccall AD, 62. Pipas JM. SV40: cell transformation and tumorigen-
Mellas RE, Lei P, Andreadis ST, Baker OJ. Stem esis. Virology. 2009;384:294–303.
cell-soluble signals enhance multilumen formation in 63. Pradhan S, Liu C, Zhang C, Jia X, Farach-Carson
SMG cell clusters. J Dent Res. 2015;94:1610–7. MC, Witt RL. Lumen formation in three-dimensional
48. Maslinska M, Przygodzka M, Kwiatkowska B,
cultures of salivary acinar cells. Otolaryngol Head
Sikorska-Siudek K. Sjogren’s syndrome: still not fully Neck Surg. 2010;142:191–5.
understood disease. Rheumatol Int. 2015;35:233–41. 64. Pradhan-Bhatt S, Harrington DA, Duncan RL, Jia

49. Masters JR. Human cancer cell lines: fact and fantasy. X, Witt RL, Farach-Carson MC. Implantable three-
Nat Rev Mol Cell Biol. 2000;1:233–6. dimensional salivary spheroid assemblies demonstrate
50. Mccall AD, Nelson JW, Leigh NJ, Duffey ME,
fluid and protein secretory responses to neurotrans-
Lei P, Andreadis ST, Baker OJ. Growth factors mitters. Tissue Eng Part A. 2013;19:1610–20.
polymerized within fibrin hydrogel promote amy- 65. Quissell DO, Barzen KA, Gruenert DC, Redman

lase production in parotid cells. Tissue Eng Part A. RS, Camden JM, Turner JT. Development and char-
2013;19(19–20):2215–25. acterization of SV40 immortalized rat submandibu-
51.
Michikawa H, Fujita-Yoshigaki J, Sugiya lar acinar cell lines. In Vitro Cell Dev Biol Anim.
H. Enhancement of barrier function by overexpres- 1997;33:164–73.
144 O.J. Baker
66. Quissell DO, Barzen KA, Redman RS, Camden JM, 76. Villa A, Wolff A, Aframian D, Vissink A, Ekstrom
Turner JT. Development and characterization of SV40 J, Proctor G, Mcgowan R, Narayana N, Aliko
immortalized rat parotid acinar cell lines. In Vitro Cell A, Sia YW, Joshi RK, Jensen SB, Kerr AR, Dawes C,
Dev Biol Anim. 1998;34:58–67. Pedersen AM. World Workshop on Oral Medicine VI:
67. Ratchford AM, Baker OJ, Camden JM, Rikka S, Petris a systematic review of medication-induced salivary
MJ, Seye CI, Erb L, Weisman GA. P2Y2 nucleotide gland dysfunction: prevalence, diagnosis, and treat-
receptors mediate metalloprotease-dependent phos- ment. Clin Oral Investig. 2015b;19:1563–80.
phorylation of epidermal growth factor receptor and 77. Wang Z, Zourelias L, Wu C, Edwards PC, Trombetta
ErbB3 in human salivary gland cells. J Biol Chem. M, Passineau MJ. Ultrasound-assisted non-viral
2010;285:7545–55. gene transfer of AQP1 to the irradiated minipig
68. Rugel-Stahl A, Elliott ME, Ovitt CE. Ascl3 marks parotid gland restores fluid secretion. Gene Ther.
adult progenitor cells of the mouse salivary gland. 2015;22(9):739–49.
Stem Cell Res. 2012;8:379–87. 78. Watson CJ, Dark JH. Organ transplantation: histori-
69. Shirasuna K, Sato M, Miyazaki T. A neoplastic epi- cal perspective and current practice. Br J Anaesth.
thelial duct cell line established from an irradiated 2012;108(Suppl 1):i29–42.
human salivary gland. Cancer. 1981;48:745–52. 79. Weissman IL, Anderson DJ, Gage F. Stem and pro-
70. Soltoff SP, Hedden L. Isoproterenol and cAMP block genitor cells: origins, phenotypes, lineage commit-
ERK phosphorylation and enhance [Ca2+]i increases ments, and transdifferentiations. Annu Rev Cell Dev
and oxygen consumption by muscarinic receptor Biol. 2001;17:387–403.
stimulation in rat parotid and submandibular acinar 80. Yamada A, Futagi M, Fukumoto E, Saito K, Yoshizaki K,
cells. J Biol Chem. 2010;285:13337–48. Ishikawa M, Arakaki M, Hino R, Sugawara Y, Ishikawa
71. Tran SD, Wang J, Bandyopadhyay BC, Redman RS, M, Naruse M, Miyazaki K, Nakamura T, Fukumoto
Dutra A, Pak E, Swaim WD, Gerstenhaber JA, Bryant S. Connexin 43 is necessary for salivary gland branch-
JM, Zheng C, Goldsmith CM, Kok MR, Wellner RB, ing morphogenesis and FGF10-induced ERK1/2 phos-
Baum BJ. Primary culture of polarized human sali- phorylation. J Biol Chem. 2016;291:904–12.
vary epithelial cells for use in developing an artificial 81. Yanagawa T, Hayashi Y, Nagamine S, Yoshida H,

salivary gland. Tissue Eng. 2005;11:172–81. Yura Y, Sato M. Generation of cells with phenotypes
72. Turner JT, Redman RS, Camden JM, Landon LA, of both intercalated duct-type and myoepithelial cells
Quissell DO. A rat parotid gland cell line, Par-C10, in human parotid gland adenocarcinoma clonal cells
exhibits neurotransmitter-regulated transepithelial grown in athymic nude mice. Virchows Arch B Cell
anion secretion. Am J Physiol. 1998;275:C367–74. Pathol Incl Mol Pathol. 1986;51:187–95.
73. Turner RJ, Sugiya H. Understanding salivary fluid 82. Yaniv A, Neumann Y, David R, Stiubea-Cohen R, Orbach
and protein secretion. Oral Dis. 2002;8:3–11. Y, Lang S, Rotter N, DVIR-Ginzberg M, Aframian DJ,
74. Vasquez MM, Mustafa SB, Choudary A, Seidner
Palmon A. Establishment of immortal multipotent rat
SR, Castro R. Regulation of epithelial Na+ channel salivary progenitor cell line toward salivary gland regen-
(ENaC) in the salivary cell line SMG-C6. Exp Biol eration. Tissue Eng Part C Methods. 2011;17:69–78.
Med (Maywood). 2009;234:522–31. 83. Zhu Y, Aletta JM, Wen J, Zhang X, Higgins D,

75. Villa A, Connell CL, Abati S. Diagnosis and manage- Rubin RP. Rat serum induces a differentiated phe-
ment of xerostomia and hyposalivation. Ther Clin notype in a rat parotid acinar cell line. Am J Physiol.
Risk Manag. 2015a;11:45–51. 1998;275:G259–68.
Matrix Biology of the Salivary
Gland: A Guide for Tissue 8
Engineering
Mariane Martinez, Danielle Wu,
Mary C. Farach-Carson, and Daniel A. Harrington
Abstract
Salivary glands produce saliva needed to carry out daily functions such as
initiation of food digestion, lubrication of the oral cavity, and prevention
of oral diseases. Xerostomia, or dry mouth due to hyposalivation, can
occur in individuals with Sjögren’s syndrome or in patients who receive
radiation therapy to treat head and neck cancer. This chapter will focus on
the bioengineering approaches in salivary gland regeneration that seek to
restore salivary function in patients suffering from xerostomia. A brief
description of salivary gland function, structure, and development will be
provided first as this information is vital to inform any salivary gland tis-
sue engineering efforts. Additionally, examples of salivary gland-derived
stem/progenitor cells that are used in various salivary gland regeneration
models will be introduced along with a brief description of each utility as
source material for tissue engineering. Lastly, we will review matrices for
three-dimensional cell culture, including decellularized native extracellu-
lar matrix scaffolds, Matrigel®, and scaffolds containing biologically
derived natural polymers, polysaccharides, and biologically active protein
fragments. Together, these elements will provide a current view of the
state-of-the-art clinical approaches to relieve xerostomia using three-
dimensional culture techniques.
8.1 otivation for Engineering

M
M. Martinez • D. Wu, PhD
a Neo-Salivary Gland
D.A. Harrington, PhD (*)
Department of BioSciences, Rice University, Over 50,000 Americans suffer from head and
Houston, TX, USA neck cancer annually [1]. The majority of those
e-mail: [email protected];
cancer patients with locally invasive forms of the
[email protected]; [email protected]
disease will receive a standard treatment that
M.C. Farach-Carson, PhD
includes radiation therapy (RT). Despite the
Departments of BioSciences and Bioengineering,
Rice University, Houston, TX, USA broad availability and use of intensity-modulated
e-mail: [email protected] radiotherapy (IMRT) as an RT delivery method,

146 M. Martinez et al.
and its high spatial precision, RT still too often salivary glands lining the oral cavity [7]. All of
leads to drastic adverse morphological changes these salivary glands are responsible for produc-
in salivary gland structure, including ductal meta- ing the salivary components that initiate diges-
plasia, periductal fibrosis, and necrotic loss of tion of food, lubricate the oral cavity, facilitate
salivary acinar cells [2, 3]. Consequently, ~64 % swallowing, and maintain the dental flora to
of long-term head and neck cancer survivors who avoid dental diseases [8].
were treated with conventional two-dimensional Functional salivary glands have a complex
(2D) RT suffer from moderate to severe cellular organization that allows saliva to be pro-
RT-induced xerostomia, or dry mouth, which duced constitutively and in greater amounts upon
greatly reduces their quality of life due to demand. A tightly structured and extensively
hyposalivation [4]. Current treatments for xero- routed network ensures that saliva is unidirec-
stomia include cholinergic sialagogues, artificial tionally secreted into the oral cavity. Proximal
saliva substitutes, strict oral hygiene, and other functional units in salivary glands consist of aci-
largely palliative treatments. However, these nar cells (serous and/or mucinous cells) that pro-
strategies are only temporary, and many produce duce proteins and fluid transported in saliva. The
several undesirable side effects including, but not spherically organized pyramidal acinar cells are
limited to, nausea, vomiting, and/or excessive highly interconnected to each other and to the
sweating [5]. Hence, they are often abandoned by surrounding myoepithelial cells via cell-cell
patients, as a majority of patients fail to return to adhesions (see Sect. 8.2.2.1). Stellate-shaped
their dental follow-up appointments after com- myoepithelial cells are postulated to be responsi-
pletion of radiation therapy [6]. ble for contracting the acinus and forcing saliva
For these reasons, a tissue-engineered salivary out into an elaborate system of ducts leading into
gland, created with patient-derived cells and sur- the oral cavity [9–11]. Functional acini secrete
gically grafted into the parotid of the xerostomic salivary contents into their lumens that merge
patient after successful cancer therapy, offers an into a hierarchical transport system composed of
alternative treatment for this syndrome. intercalated, striated, and excretory ductal
Inspiration for designing a salivary neotissue is regions, which both transport and continue to
found in the organization and composition of the modify the composition of saliva as it exits the
native tissue. This chapter considers the cellular gland [12].
and extracellular components of salivary glands Polarized acinar cells comprise the acinar units
and their potential for implementation in the cre- that selectively secrete salivary components into
ation of an engineered tissue replacement. the lumens of the glands. Polarized, or asymmet-
rically organized, acinar cells are interconnected
on their lateral surfaces by cell-cell junctions
8.2 alivary Gland Structure
S including adherens junctions, tight junctions, gap
and Development junctions, and desmosomes and are anchored on
their basal surface to the basement membrane
8.2.1 S
alivary Gland Structure through integrin-mediated binding [13–20].
and Function Acinar cells also interact with myoepithelial cells
on their basal sides. Organelle positioning is dis-
The salivary system in humans consists of three tinct in polarized cells; in salivary acinar cells, the
major bilateral salivary glands: the parotid, sub- nucleus is positioned at the base of the cell, near
mandibular, and sublingual glands. Parotid the basement membrane and far away from the
glands are the largest and are located inferior and lumen. The smooth and extensive rough endo-
anterior to the ear. Submandibular and sublingual plasmic reticulum (ER) and the Golgi apparatus
glands are located inferior to the tongue and floor that synthesize and package salivary contents for
of the oral cavity. Additionally, there are numer- transport are positioned between the nucleus and
ous minor serous fluid and mucous-producing the secretory lumen. The many secretory vesicles
8 Matrix Biology of the Salivary Gland: A Guide for Tissue Engineering 147
a b c
Fig. 8.1 Basement membrane proteins assist in the orga- laminin (red), and (c) perlecan/HSPG2 (green) that sur-
nization and function of acinar and ductal compartments round acini and intercalated ducts. Scale bars are 20 μm,
in human parotid glands. Immunohistochemistry is used and nuclei (blue) are counterstained with 4’,6-diamidino-
to identify critical extracellular matrix proteins in human 2-phenylindole (DAPI)
salivary gland tissue including (a) collagen IV (green), (b)
that release salivary contents into the lumen are in about 7 mL/min during a meal when they are
close proximity to the apical membrane [21–23]. stimulated by the surrounding autonomic nerves
This polarized organization and unique nature of [32, 33]. Cholinergic neurotransmitters released
secretory vesicles have been described as key fac- by the surrounding parasympathetic innervation
tors for proper secretion, which can be lost in the stimulate the glands to secrete salivary fluid,
selective destruction of the salivary acinar units while adrenergic neurotransmitters released by
during radiation [24–26]. the surrounding sympathetic innervation stimu-
A well-organized basement membrane is cru- late the glands to secrete salivary proteins (see
cial for maintaining cell polarity. The basement 8.2.2). Additionally, autonomic innervation is
membrane is a specialized network of extracel- required for proper development of the salivary
lular matrix (ECM) proteins mainly composed of glands during embryogenesis and in postnatal
collagen IV, laminin, perlecan/HSPG2 (Fig. 8.1), maturation, as discussed in the following section.
and entactin/nidogen [27, 28]. Each of the acini
and ductal cells in the salivary glands is separated
from underlying connective tissue by a mesh- 8.2.2 Salivary Gland Development
work of laminin heterotrimers and collagen IV,
which are spot-welded together by entactin [29]. Much of what is known of salivary gland develop-
The ~800-million-year-old heparan sulfate pro- ment was discovered through the study of devel-
teoglycan 2, perlecan/HSPG2, plays a unique opmental processes in a mouse submandibular
role in the border functions of the basement gland model and is summarized briefly below
membrane and as a growth factor supply depot [34–36]. Salivary gland development consists of
important for wound healing [30, 31]. Additional five distinct stages: prebud, initial bud, pseudo-
functions of the basement membrane include glandular, canalicular, and terminal bud. During
promoting salivary cell organization, providing the prebud stage at embryonic day 11.5 (E11.5),
mechanical support to the gland, providing guid- the ectodermal epithelium adjacent to the tongue
ance cues during salivary development and hence thickens, marking the location of the base of the
regeneration, and storing heparin-binding (HB) salivary gland. The thickened epithelium contin-
factors that can be actively released in case of ues to proliferate and grow, forming an initial bud
immediate need. that extends into the neighboring neural crest mes-
Salivary glands produce and secrete saliva enchyme (E12.5). Simultaneously, the base of the
constitutively at a relatively constant rate of bud invaginates to create the duct of the devel-
0.3 mL/min or on demand at higher levels of oping salivary bud. Branching morphogenesis of
the salivary gland commences during the pseu- proper branching morphogenesis and differentia-
doglandular stage (E13.5), yielding a multilobed tion of the salivary gland.
structure with multiple salivary buds connected Studies of embryonic salivary gland develop-
to the base of the gland via epithelial branches. ment have identified important factors needed for
As branching morphogenesis continues during successful development of a functional salivary
the canalicular stage (E15.5), these epithelial gland: growth factors (GFs) (i.e., EGF and fibro-
branches begin to develop a lumen via apoptosis blast growth factors (FGFs)), innervation, vascu-
of inner epithelial cells, migration of outer epi- larization, and epithelial and mesenchymal
thelial cells, and proliferation of epithelial cells cell-derived ECM (discussed further in the fol-
at the tips of the branches [36, 37]. Salivary gland lowing section). To advance the field of salivary
differentiation begins and the majority of ductal tissue replacement, these critical factors should
lumens are formed by the terminal bud stage be strategically incorporated into the design of
(E17.5), but the remaining undifferentiated cells the bioengineered organ, regardless of whether or
and unformed lumens will be differentiated and not they are integrated in vitro into the scaffold
formed postnatally, respectively. for organ culture or added in vivo after implanta-
Aside from being instrumental for the physi- tion of the scaffold. Moreover, to fully recapitu-
ology of the mature gland, nerves also are essen- late the events in development, the addition of
tial during salivary gland development. Salivary key morphogenetic factors needs to be tempo-
glands are innervated by postganglionic parasym- rally and spatially controlled, as they would be
pathetic and sympathetic nerve fibers that stimu- during native gland development. For example,
late fluid-rich and protein-rich secretions [12, 32, the introduction of certain ECM components like
38]. Parasympathetic postganglionic nerve fibers fibronectin, which is critical for cleft formation
originate from the submandibular ganglion for needed during branching morphogenesis, and
the SMG, while they originate from the otic gan- FGFs needed for bud elongation and cleft forma-
glion for the parotid gland. Sympathetic nerve tion should only be introduced during the onset
fibers innervating all glands originate from the of branching and not during initial clustering and
superior cervical ganglion, using the external assembly. The ideal salivary gland tissue engi-
carotid plexus as a guide to branch off to the sali- neering scaffold would support these temporo-
vary glands [32, 38, 39]. Sympathetic denerva- spatially distinguished functions and promote
tion in neonatal rats led to a reduction in acinar stable long-term ingrowth of the host’s native
cell size and in the number of granules produced, vascularization and innervation as the neogland
suggesting that proper sympathetic innervation is develops.
important for acinar cell maturation [40]. Studies
performed with fetal mice showed that the para- 8.2.2.1 Importance of the ECM
sympathetic ganglia from the submandibular During Salivary Gland
ganglion grow in parallel and are directed by the Development
developing SMG epithelium [41, 42]. Lastly, dis- Successful salivary gland development depends
ruption of parasympathetic innervation in mouse on the temporospatial signaling provided by the
embryonic SMG explant cultures reduced the development and maturation of the underlying
number of cytokeratin 5 (K5) expressing pro- basement membrane that separates the basal sur-
genitor cells, as well as a reduction in branching face of the glandular acinar and myoepithelial
morphogenesis of the organ [43]. The epithelial cells from the underlying connective tissue. The
progenitor population needed for organogenesis basement membrane is dynamic and is actively
was rescued by stimulation of the muscarinic M1 remodeled throughout the development,
receptor (M1), via acetylcholine treatment, and homeostasis, and wound healing. Under normal
was dependent on epithelial growth factor (EGF) conditions, the basement membrane maintains
receptor signaling. Together, these studies indi- tissue homeostasis of the salivary organ and pro-
cate that innervation contributes significantly to vides direction and orientation for secretion.
Proteases, such as matrix metalloproteinases enchyme. Unexpectedly, fibronectin expression

(MMPs), partially cleave the basement mem- was observed in salivary epithelial cells, with
brane to loosen the matrix and allow for acinar increased expression of fibronectin localized to
growth. Moreover, in addition to providing the the epithelial cells closest to where the cleft will
extra space needed for tissue expansion, turnover form. This fibronectin deposition was accompa-
releases a number of locally sequestered growth nied by a decrease in E-cadherin expression [46].
factors needed to further stimulate gland devel- To test fibronectin’s role in branching mor-
opment [37, 44, 45]. One study using mouse phogenesis, the protein was targeted with specific
embryonic SMG explants showed the dynamics blocking polyclonal or monoclonal antibodies
of the basement membrane during branching that could reduce its interactions with epithelial
morphogenesis [44]. Salivary epithelial cells at cells in the developing mouse SMG [46].
the tip of the buds digested the basement mem- Branching morphogenesis was partially inhibited
brane through the activation of MMPs, creating by anti-fibronectin antibody treatment, shown by
perforations through which they could migrate. a dose-dependent reduction in bud formation. A
The basement membrane is seen to move inward similar decrease in branching was observed in
toward the clefts, forming the ducts that appear early salivary buds treated with siRNA against
during development. Furthermore, interactions fibronectin and was rescued by exogenous fibro-
between the developing salivary epithelium and nectin in a concentration-dependent manner if it
the basement membrane are vital for the growth was added to the organ culture. Treatment with
of an initial salivary epithelial bud, branching antibodies targeting integrins α5 and β1 had the
morphogenesis of that bud, and differentiation greatest negative effect on branching morpho-
and polarization of the salivary parenchymal genesis, suggesting an important role for the
cells including acinar and ductal cells. fibronectin receptor α5β1 in fibronectin-mediated
The ECM is crucial for the formation of a branching morphogenesis.
multilobed secretory organ such as the salivary Interstitial collagens also are important to ini-
gland through branching morphogenesis, and in tiate branching morphogenesis during the early
the case of salivary acinus, the ECM is secreted development of salivary glands [47, 53, 54].
both by the epithelial cells, particularly the myo- Using radiolabeled amino acids, soluble tropo-
epithelial cells, and the neighboring mesenchy- collagen was shown to be secreted by mesenchy-
mal cells [46–48]. Studies in which interactions mal cells and selectively polymerized into
between the salivary epithelial cells and the sur- collagen fibrils by salivary epithelium from
rounding basement membrane proteins were embryonic SMG rudiments [47]. Studies in
decreased inhibited branching morphogenesis, which collagen synthesis was inhibited, or the
demonstrating the need for a well-organized protein was degraded with collagenases, reduced
basement membrane to guide gland develop- the extent of branching morphogenesis of the
ment [46, 49, 50]. Fibronectin, a glycoprotein in SMG in the ex vivo embryonic model [54–58].
the ECM that is arranged into fibrils, is present Inhibition of collagen synthesis and secretion by
early on during gland development and estab- treatment with L-azetidine-2-carboxylic acid
lishes cleft formation during branching mor- (LACA) or α,α’-dipyridyl drastically reduced
phogenesis [46, 51, 52]. Fibronectin expression branching of the embryonic developing SMGs
in developing mouse SMGs was assessed via [57]. However, inhibition of collagen cross-
laser microdissection and RT-PCR of the sali- linking by treatment with β-aminopropionitrile
vary tissue, allowing the tissue to be divided into had no effect on the morphogenetic activity of the
cell populations of cleft-derived epithelial cells, SMGs treated. Collagenase treatment of
bud-derived epithelial cells, cleft-neighboring developing salivary glands, whether for a short or
mesenchyme, and bud-neighboring mesenchyme long period, inhibited branching morphogenesis
[46]. Fibronectin expression was high in both of the gland [58]. Expression of collagens I, III,
the cleft-neighboring and bud-neighboring mes- IV, and V was analyzed by immunohistochemical
techniques [51, 53]. Collagens I, III, IV, and V Antibodies against the E3 domain of the lam-
were found to be present throughout the salivary inin α1 chain inhibited branching morphogen-
mesenchyme; however, collagen III accumulated esis and led to the formation of a discontinuous
distinctly at the indented sites and clefts of the basement membrane [49]. Targeting integrin α6
salivary bud. Although collagen IV was present with antibodies also disturbed branching mor-
in the mesenchyme, it was localized in higher phogenesis of the embryonic SMG, but did not
concentrations in the surrounding basement disrupt the formation of a continuous basement
membrane layer [51]. Collagenase derived from membrane. Furthermore, this study suggests
bovine dental pulp that specifically cleaves col- that the E3 domain from laminin α1 chain is
lagen I and III was shown to inhibit branching important for the formation of a well-organized,
morphogenesis of the developing salivary gland, continuous basement membrane. Kadoya et al.
and this effect was reversible with the addition of tested the activity of various sequences from
a collagenase inhibitor (bovine serum albumin or laminin α1 and α2 chains, and a specific sequence
fetal calf serum) [56]. (RKRLQVQLSIRT) in laminin α1 LG4 domain
Laminin is an abundant component of the base- was shown to be important for salivary gland
ment membrane that also plays an important role development [66]. Embryonic mouse SMG
in salivary gland development. The heterotrimer is glands failed to undergo branching morphogen-
made up of one of each subunit chain: α (1–5), β esis or form a continuous basement membrane
(1–3), and γ (1–3), which co-assembles into a when treated with AG-73 peptide. Furthermore,
polymeric network lying directly under the epithe- the homologous sequence to AG-73, MG-73, in
lium and/or endothelium [27]. The α chain of lam- laminin α2 chain had no effect on branching mor-
inin uniquely contributes five LG domains at the phogenesis of the developing gland, suggesting
C-terminal end, which make up the globular, or G, that laminin α2 may be important for terminal bud
domain that has many cell-binding sites enabling differentiation and not for branching morphogen-
cell-laminin adhesion [59]. Cellular receptors for esis. Targeting laminin α5 activity in the devel-
laminin mediate the biological responses by the oping SMG with A5G77f peptide derived from
salivary epithelium, and those include integrins α3, globular domain LG4 inhibited branching mor-
α6, β1, and α-dystroglycan [37, 50, 60–63]. phogenesis, but did not inhibit basement mem-
Laminin glycoproteins mainly are localized to the brane formation [65]. The salivary rudiments
basement membrane surrounding the developing treated with A5G77f contained terminal epithe-
mouse SMG, but expression of laminin chain iso- lial buds with cleft formation, but they failed to
forms varies during development and continues to form elongated ducts, suggesting that laminin α5
be expressed up to the time of maturation of the cell adhesion is important for ductal elongation.
gland [37, 51, 62, 64]. Laminins α1 and α5 are Perlecan/HSPG2 is a heparan sulfate proteo-
expressed throughout the basement membrane of glycan present in the basement membrane sur-
developing mouse SMGs during branching mor- rounding various epithelia, including the salivary
phogenesis and are localized to the basement gland epithelium. Successful salivary gland
membrane surrounding the ducts in the branched growth and development relies on the presence of
organ; specifically, α1 is restricted to the only the perlecan because it serves as a depot for the broad
excretory ducts [49, 65]. Laminin α2 was shown to category of heparin-binding (HB) GFs that
be expressed at lower levels early during develop- enable cell proliferation and differentiation [30,
ment at E13 and at higher levels in the basement 37, 68, 69]. When properly synthesized by the
membrane surrounding matured acini [66]. cell, perlecan is modified within domain I and V
Laminin α3 is present in the early stages of devel- with long glycosaminoglycan chains, primarily
opment (mouse) around the stalk of the bud and heparan sulfate and some chondroitin sulfate.
growing ducts and in later stages in the basement The presence of heparan sulfate enables its
membrane surrounding the excretory duct and growth factor binding and delivery function [30,
myoepithelial cells [67]. 70–72]. Important GFs for salivary gland
d evelopment delivered by perlecan’s heparan sul- composed of cytoplasmic scaffolding proteins

fate chains include FGF-1, FGF-2, FGF-9, FGF- zonula occludins, and transmembrane proteins
10, HB-EGF, vascular endothelial growth factor claudins, occludins, and junctional adhesion mol-
(VEGF), and platelet-derived growth factor ecules [14, 78]. Tight junctions are responsible
(PDGF) for vascularization [30]. Furthermore, for forming an epithelial barrier to prevent diffu-
these growth factors, often in the presence of sion of solutes via the paracellular space, allow-
heparan sulfate, stimulate FGFRs, and this acti- ing for the formation of a transepithelial ion
vation in turn triggers the biological processes gradient [78]. Additionally, tight junctions pre-
that occur during salivary gland development. vent the fluid movement of integral membrane
Thus, the contribution of perlecan as a key coproteins between the apical surface and the baso-
receptor providing heparan sulfate for growth lateral membrane [79]. Adherens junctions are
factor presentation remains very important [69, composed of transmembrane protein cadherins
73–76]. Through the use of the ex vivo embry- and cytoplasmic scaffolding proteins including
onic mouse SMG model, FGF-7 and FGF- 10 catenins [80]. In polarized cells, catenins associ-
have been shown to play a positive role in stimu- ate with cadherins to form the complex that links
lating salivary epithelial bud growth. Embryonic with F-actin, and these F-actin-linked complexes
mouse SMGs treated with FGF-7 and FGF-10 are directly linked by cadherins between the cells
had larger buds when compared to the control in the epithelial sheet, forming a continuous actin
SMGs, while those treated with FGF-2 had a belt that strengthens the epithelium [81].
decrease in the number of buds [73]. Additionally, Desmosomes are another form of cell-adhesions
FGF-7 and FGF-10 have differential effects on that use cadherins to interconnect neighboring
ductal formation within a heparan sulfate-rich epithelial cells [82]. Desmosomal cadherins des-
ECM, driving ductal branching or elongation, moglein and desmocollin are the transmembrane
respectively [69]. proteins, and their N-terminal domains interact
Aside from its role in branching morphogen- with each other in the intercellular space to make
esis, the ECM also is critical for the establish- the cell-cell adhesions that provide mechanical
ment and stabilization of polarity of the salivary strength to the epithelial sheet [83]. Cytoplasmic
parenchymal cells. Salivary epithelial cells asym- proteins plakoglobin and plakophilin are part of
metrically organize their cytoplasmic and the scaffolding complex that connects the cadher-
membrane-bound proteins during cytodifferenti- ins to the intermediate filament-binding protein
ation, leading to a polarized epithelial cell with desmoplakin, forming cell-cell adhesions linked
an apical membrane facing the lumen separated to the cytoskeletal intermediate filaments [84].
from the basolateral membrane. Polarization of Lastly, polarized epithelial cells are physically
epithelial cells involves both cell-cell and cell- and chemically interconnected by gap junctions
ECM adhesion, but how exactly each one aids in that are composed of connexin (Cx) proteins that
establishment of polarity is still unclear. This form cell-cell channels that allow the transport of
asymmetric organization, or polarity, of salivary ions and small molecules from the cytoplasm of
acinar and ductal cells is required for the proper one cell to another adjacent cell [20]. Connexin
function of the salivary gland, i.e., the directional (Cx) proteins mediate coordinated intercellular
secretion of salivary proteins and fluid. It is signaling during secretion [19], and they also
essential that these polarized functions be recre- play a role in the proper development of the sali-
ated in any engineered gland designed to restore vary gland [85].
directional secretion of saliva. Asymmetrically organized salivary epithelial
Polarized epithelial cells are physically inter- cells also are connected to the underlying base-
connected by cell-cell adhesions including tight ment membrane via integrin-mediated binding in
junctions, adherens junctions, desmosomes, and hemidesmosomes and focal adhesions, which
gap junctions, all of which aid in establishing and also play a role in establishing and maintaining
maintaining cell polarity [77]. Tight junctions are epithelial cell polarity. Focal adhesions and
hemidesmosomes are composed of transmem- other. The PAR complex binds to tight junctions
brane integrin receptors which bind to the ECM via Par3, creating an apical-basal border in polar-
outside of the cell and are linked to a scaffold ized epithelial cells [14, 77]. Additionally, Par6 in
complex that is associated with the actin and ker- this polarity complex associates with RhoGTPase
atin intermediate filament cytoskeleton, respec- Cdc42, and together this complex enables actin
tively [86, 87]. Some of the proteins that make up cytoskeletal reorganization and organelle translo-
the intracellular scaffold complex for focal adhe- cation [95, 96]. The Scribble complex contains
sions include talin, vinculin, focal adhesion the conserved proteins scribble, Discs large
kinase, and α-actinin. The intracellular scaffold (Dlg), and lethal giant larvae (Lgl), altogether
of hemidesmosomes is composed of an inner making a polarity complex that defines the baso-
plaque that connects the adhesion site to the lateral membrane of a polarized cell [77, 97]. The
intermediate filaments and an outer plaque that presence of noncoding mir200c, a key regulator
lies parallel to the cell membrane [88]. Formation of polarity in epithelial cells, in the developing
of these integrin-mediated adhesion sites along submandibular end bud lends further support to
with cell-cell adhesion sites activates members of the notion that the salivary gland uses many of
the Rho family small guanosine triphosphatases the same mechanisms to establish polarity as do
(RhoGTPases), such as Rac1, Cdc42, and RhoA, other polarized epithelia [98]. It remains unclear
which mediate the reorganization of the cytoskel- whether the secretory acinar cells of the salivary
etal filaments [77, 89, 90]. This cytoskeletal reor- gland polarize precisely in the same way as do
ganization is vital for the asymmetrical the ductal cells, but it is likely many of the same
positioning of organelles and polarity complexes proteins and complexes are involved.
within the polarized epithelial cells [13, 77]. Together, these cell adhesion and polarity
Because of the importance of cell-matrix adhe- complexes regulate the asymmetric cell organiza-
sions on cell polarity, a scaffold for three- tion needed for proper function of the variety of
dimensional (3D) organ culture should provide epithelial cells found in the salivary gland.
integrin-binding sites to the cells to activate the However, characterization of these protein com-
players needed for cytoskeletal and organelle plexes in native salivary epithelial tissue remains
reorganization in vitro. to be investigated, particularly with respect to the
Polarity complexes, primarily characterized to differences in the acinar and ductal regions.
date in ductal-like cells, are crucial to the estab- When the exact nature of the cell adhesion and
lishment and maintenance of epithelial cell polar- polarity complexes present in the various salivary
ity, and they include the Crumbs (CRB) complex, regions is determined, then these polarity com-
the partitioning-defective (PAR) complex, and plex proteins can serve as markers for correct
the Scribble complex [91].The CRB complex salivary epithelial assembly and polarization in a
consists of the transmembrane homologous CRB 3D in vitro organ culture model suitable for tis-
proteins, and its cytoplasmic domain is associ- sue repair.
ated with protein associated with lin seven
(PALS-1) [77, 91, 92]. PALS-1 links with PALS-
1-associated tight junction protein (PATJ) via one 8.3 Salivary Gland-Derived
of its L27 domains [77]. PATJ then connects the Stem/Progenitor Cells Used
CRBs complex to tight junctions by interaction for Regeneration Models
with its PDZ domains, localizing the polarity
complex to the apicolateral side of the mem- Engineering a functional organ in vitro requires
brane. The PAR complex is made up of PDZ- all of the cell types, and their distinct functions,
domain- containing scaffold proteins Par3 and present in that native organ. Those various cell
Par6 and serine/threonine kinase atypical protein types can be isolated and maintained in various
kinase C (aPKC) [14, 77, 93, 94]. Par3 and Par6 differentiated states, or stem/progenitor cells can
both have the ability to bind to aPKC and to each be induced to differentiate into the various cell
types needed. Much work has been performed to salisphere culture, but the expression of c-Kit
locate, isolate, and characterize salivary stem/ increased by tenfold (0.65 %). Salivary restora-
progenitor cells for tissue engineering purposes. tion (measured by salivary flow rate) in the
This section will focus on some of the efforts per- mouse-irradiated model was achieved most effi-
formed by multiple research groups to isolate and ciently with c-Kit+ cells, as it only required an
characterize salivary stem/progenitor cells and injection of 300 c-Kit+ cells, while the rest of the
determine their utility in gland regeneration. stem cell populations required 10,000–134,000
cells to achieve salivary restoration.
A subsequent study assessed the ability of a
8.3.1 Salisphere-Derived c-Kit+ stem cell population positive for c-Kit, CD24,
Cells for Salivary Gland and/or CD49f to restore salivary function in irra-
Regeneration diated salivary glands [102]. Although all of the
stem cell populations (c-Kit+, c-Kit+/CD24+,
A salisphere cell culture method was developed c-Kit+/CD49f+, and c-Kit+/CD24+/CD49f+)
to culture salivary stem cells that can be used for restored some percentage of salivary flow in irra-
salivary gland function restoration [99]. The diated salivary glands, the stem cell population
method entails mechanically and enzymatically expressing c-Kit, CD24, and CD49f restored sali-
dissociating salivary tissue and culturing the iso- vary flow rate in irradiated salivary glands to an
lated cells on nonadherent plates to drive cell average of >50 % of the salivary flow rate of the
aggregation into salispheres. The salisphere cell untreated control. Restoration of salivary gland
culture technique allowed for fluorescence- tissue morphology and cell phenotype by trans-
activated cell sorting (FACS)-mediated isolation plantation of c-Kit+/CD24+/CD49f+ cells was
of c-Kit+/Sca-1+/Musashi-1+ cells that could shown by a reduction in fibrosis, reduction in
differentiate into acinar and ductal cell types oversized blood vessels, and increased expres-
[99]. Salivary c-Kit+ stem cells transplanted into sion of differentiated ductal cell markers and
irradiated mouse salivary beds restored salivary ductal stem markers. Lastly, another lab showed
tissue morphology and function after 90 days that CD49f and Thy-1 expressing salivary pro-
compared to the irradiated-only control. To verify genitor cells isolated from adult human salivary
that the same population of stem cells was pres- glands could differentiate into pancreatic cell
ent in human salivary glands, salivary c-Kit+ phenotypes when cultured as spheres, showing
stem cells were isolated from non-tumorigenic their potential to transdifferentiate into other cell
parotid and SMG tissue and were cultured via the types [103].
salisphere method [100]. Human-derived c-Kit+
cells also had the ability to self-renew and dif-
ferentiate when cultured in vitro in Matrigel®. 8.3.2 Isolation of Adult Stem Cells
A study by Nanduri et al. further characterized from Human Salivary Tissue
stem cell marker expression of murine cells
derived from the salisphere culture system [101]. Researchers also have reported other methods for
Salivary stem cell populations expressing c-Kit, the isolation and maintenance of salivary-derived
CD133, CD49f, and CD24 were isolated, and cells with stem/progenitor markers [104, 105].
stem cell populations expressing each marker Our own group has isolated primary human sali-
were quantified using FACS. The expression vary stem/progenitor cells (hS/PCs) from healthy
level of CD49f and CD24 was high in the parotid and SMG tissue that can organize into
salisphere-derived stem cells at day 0, with posi- salivary structures and that display self-renewal
tive expression in more than 50 % of cells, while and extended proliferation capabilities [106–
the expression level of CD133 (6 %) and c-Kit 109]. Specifically, hS/PCs cultured in 3D hyal-
(0.058 %) was low. Expression levels of CD133, uronic acid-based hydrogels retain self-renewing
CD49f, and CD24 remained constant at day 3 of properties that allow them to continue to prolifer-
ate for over 48 days (further discussed in poly- MSCs derived from human salivary tissue, but
saccharide Sect. 8.4.3.2) [108]. These hS/PCs this approach has many challenges to overcome
express c-Kit, K5, and K14 and can be differenti- before such use can be achieved.
ated to display salivary and ductal phenotypes
(data in press and [108]).
Rotter et al. reported a population of stem 8.4 D Scaffolds Used in Salivary
3
cells isolated from adult human parotid tissue Gland Regeneration Models
using enzymatic digestion methods [104].
Mesenchymal stem cell (MSC) marker expres- Tissue engineering of a functional organ requires
sion was assessed with flow cytometry analysis a 3D biomimetic scaffold that allows cells to
and showed that these isolated cells expressed organize into their native structure. There have
CD29, CD44, CD73, and CD90. Analysis of been a myriad of scaffolds generated and utilized
these cells for hematopoietic stem cell marker for 3D cell culture for tissue engineering pur-
expression showed that these cells are negative poses. Specifically, studies using 3D scaffolds for
for CD34, CD45, and CD133, indicating that salivary cell culture have been extremely infor-
they are not of hematopoietic origin. Salivary mative for future salivary gland tissue engineer-
gland-derived MSCs were cultured in various ing efforts; these studies will be reviewed in this
tissue-specific differentiation media to demon- section and are outlined in Table 8.1.
strate their ability to differentiate into adipogenic,
osteogenic, or chondrogenic cell phenotypes.
Importantly, these authors do not suggest that 8.4.1 D
ecellularized Native ECM
these cells demonstrate any salivary-specific Scaffolds for Regenerative
function or ability to differentiate into acinar, Purposes
ductal, or myoepithelial phenotypes.
Another lab also reported the isolation of Organ donor shortage is a significant problem in
MSCs from adult human salivary tissue that the United States. There are currently over
could be used for salivary regenerative purposes 123,000 people on the waitlist to receive an
[110]. In this case, human SMGs were processed organ, and more than 6500 people will die wait-
enzymatically and isolated cells were cultured on ing each year [118]. One solution to the shortage
collagen I-coated tissue culture flasks. Human of donated organs is to utilize allogeneic and
salivary gland stem cells (hSGSCs) were cultured xenogeneic organs to create ECM scaffolds for
for longer periods of up to 5 weeks, and then organ regeneration. Decellularization of organs
assessed for mesenchymal and hematopoietic via physical, chemical, and/or enzymatic meth-
stem cell marker expression via FACS. Isolated ods leaves behind a natural scaffold composed of
hSGSCs were positive for MSC markers CD44, the ECM proteins and glycosaminoglycans
CD49f, CD90, and CD105, but not for hemato- (reviewed in [119, 120]). Published decellular-
poietic stem cell markers CD34 and CD45. These ization methods vary greatly depending on the
MSCs could be differentiated into adipogenic, chemical agents, physical forces, organ type, and
osteogenic, and chondrogenic cell types. organ species used (Table 8.2) [113, 119, 120,
Sequential treatments with differentiation media 140, 151–156]. In general, mechanical/physical
(containing EGF/hepatocyte growth factor (HGF) methods, chemical methods, and/or enzymatic
or solely EGF) resulted in amylase expression in methods can be applied when decellularizing
a small fraction of these cells. Transplantation of organs, and these methods can be used in combi-
hSGSCs into radiation-damaged rat salivary beds nation for more efficient decellularization
partially restored salivary flow and normalized depending on tissue type (refer to [119, 120] for
tissue morphology and reduced the number of a more detailed review of techniques).
cells undergoing apoptosis. This study suggests Efforts to decellularize various organs includ-
some salivary gland regenerative potential of ing the heart, lungs, kidneys, and trachea, among
Table 8.1 Scaffolds used for tissue engineering purposes

Type of matrix Examples
Decellularized ECM scaffold Decellularized mouse, rat, porcine, and human lungs [111, 112]
Decellularized and recellularized rat submandibular gland [113]
Matrigel®/GFR Matrigel® Matrigel® used for culturing primary salivary gland cells [114]
Growth factor-reduced Matrigel® [115]
Biologically derived polymers/ Proteins:
natural polymers Collagen
Laminin
Fibrin [115]
Polysaccharides:
Hyaluronic acid
Agarose
Dextran
Chitosan
Protein/polysaccharide hybrid polymers:
Collagen/HA
Laminin/cellulose
Gelatin/chitosan
Fibrin/alginate
Fibrin/agarose
Protein fragments/peptides Laminin peptides integrated into agarose gels [116]
Collagen-derived degradable peptides (PQ) [117]
Collagen I
Collagen IV
Perlecan domain IV peptide [106]
Fibronectin-derived RGD peptide
others, in the past decade have paved the way for perfused with blood and cultured under physi-
future advancements in decellularization of other ological conditions with the use of a bioreactor,
organs [113, 136, 140, 151–156]. An example of and were orthotopically transplanted into a lung
a highly branched organ of high relevance to sali- resection rat model. Regenerated rat lungs were
vary glands, with respect to tissue structure and connected to the recipient’s pulmonary artery and
organization, is the lung. Like salivary gland vein to restore blood flow and were ventilated via
development, lung development depends on the recipient’s airway. Although the regenerated
branching morphogenesis of the primary buds lung functionally provided gas exchange in vivo
formed to develop the fully branched, differenti- for up to 6 h before respiratory failure, there
ated organ [157–159]. were observable complications including pul-
Ott et al. developed a low-concentration monary secretions and interstitial edema in the
sodium dodecyl sulfate (SDS)-based perfu- graft. Additionally, the lung was not completely
sion decellularization protocol for decellu- regenerated as there were areas of the lung tis-
larizing rat lungs to yield ECM scaffolds for sue lacking surfactants, containing squamous
lung regeneration purposes [141]. Perfusion epithelium instead of mature secretory cells,
decellularization of rat lungs created acellular and possible type II cell hyperplasia. A follow-
ECM scaffolds that maintained the native tis- up study showed that integrating improved graft
sue architecture including vasculature, airways, preservation and postoperative weaning proto-

and alveoli. Decellularized lung ECM scaffolds cols allowed for enhanced in vivo gas exchange
were recellularized with human umbilical cord in the rat lung resection model [111]. However,
endothelial cells (HUVECs) and fetal rat lung the recellularized scaffolds again failed to be
epithelial cells to restore vasculature and the completely regenerated and led to fibrosis sur-
lung parenchyma. Recellularized rat lungs were rounding the graft and infection.
Table 8.2 Decellularization methods

Method Mechanism of action Examples; references
Physical methods Freeze/thaw cycles Disrupts cell membrane through [121–126]
the formation of ice crystals
Sonication Disrupts cell membrane through [122, 127–129]
the use of applied sound energy
Orbital shaking conditions Typically used to enhance the [130–132]
removal of cell particles from
ECM
Pressure Pressure applied to tissue can [133–135]
aid in bursting cells
Chemical methods Nonionic detergents (Triton Permeabilizes the cell [124, 128, 133,
X-100) membrane without denaturing 136–139]
proteins
Ionic detergents (sodium Permeabilizes the cell [111–113, 133,
dodecyl sulfate, sodium membrane while also denaturing 136–142]
deoxycholate, Triton-X-200) proteins
Zwitterionic detergents Permeabilizes the cell [112, 139, 142, 143]
(CHAPS, sulfobetaines) membrane with little or no
denaturing of proteins
Hypotonic and hypertonic Alternating between low [125, 144–146]
solutions concentrations and high
concentrations of NaCl solution
to disrupt cell membranes
Alcohols (isopropanol, Disrupts cell membranes but [147–149]
ethanol, methanol, glycerol) also fixes tissue; the remaining
ECM is partially cross-linked
Acids/bases (peracetic acid, Disinfects while also removing [112, 140, 150]
acetic acid, sodium nucleic acids and hydrolyzing
hydroxide) the ECM proteins, especially
collagen
Damages ECM as it degrades it
to a higher extent
Chelating agents (EDTA, Takes up ions necessary for [125, 126, 149]
EGTA) cell-cell and cell-ECM binding
Enzymatic methods Enzymes targeting proteins Typically used in combination [125, 126, 149]
(trypsin, collagenase, lipase) with other methods as it
removes cellular proteins and
ECM proteins, depending on the
enzyme type
Enzymes targeting Typically used in combination [125, 126, 132, 146]
nucleotides (nucleases) with other decellularization
methods because it degrades
nucleic acids left behind from
cell lysis
Other lung decellularization examples d ecellularized with Triton X/SDC-based decel-

contain the use of Triton X/sodium deoxy- lularization method and have been recellular-
cholate (SDC)-based and 3-[(3-cholamidopro- ized with different cell types [137, 138]. Mouse
pyl)dimethylammonio]-1-propanesulfonate lungs decellularized with the detergent solution
(CHAPS)-based detergents [137, 138, 140, 143, via the trachea, and the right ventricle gener-
153]. Mouse lungs have been successfully ated lung ECM scaffolds that maintained the
tissue architecture and ECM integrity including dimensional tissue architecture [140]. Acellular
an intact vasculature network and ECM proteins ECM scaffolds decellularized by this method had
collagen I, collagen IV, laminin, and fibronectin, reduced DNA content compared to native tissue,
respectively [138]. However, scaffolds decellu- maintained expression of collagen I, elastin, and
larized by this method contained higher amounts GAGs, and had comparable biomechanical
of remaining contaminating cell cytoskeletal pro- integrity.
teins. Lung scaffolds were recellularized with CHAPS-based detergent solution has been
bone marrow-derived mouse MSCs, and although used in the successful decellularization of rat
cell attachment was successful, MSCs failed to lungs [112, 143]. Using this method, Petersen
differentiate into differentiated lung epithelial et al. generated acellular lung ECM scaffolds that
cells. Price et al. integrated the use of a venti- maintained the three-dimensional branched tis-
lator to their decellularization method to apply sue architecture and could be recellularized with
physiological levels of mechanical stretch [137]. lung epithelial and rat lung microvascular endo-
Decellularized ECM scaffolds in this study also thelial cells [143]. DNA content was reduced by
retained their native tissue architecture including 99 %, and cellular proteins’ major histocompati-
airways, alveoli, and blood vessel network. The bility complex (MHC) I, MHC II, and β-actin
remaining ECM contained normal levels of col- were removed from acellular scaffolds generated
lagens, but lower levels of elastin, laminin, and with this CHAPs-based decellularization method.
GAGs. These scaffolds were reseeded with fetal ECM proteins’ collagen, laminin, and elastin, at
lung cells that were localized to alveolar areas lower levels, were preserved in the acellular scaf-
and expressed cytokeratin 18 and pro-Sp-C, a folds, while most of the GAGs were depleted.
pulmonary-associated surfactant protein. However, in a study by the Ott Lab that used
Other studies using the Triton X/SDC-based SDS, Triton X/SDC, and CHAPs-based decellu-
decellularization method translated this technol- larization methods on rat lungs, the CHAPS-
ogy to clinically relevant, human-sized lungs based method led to acellular scaffolds with
[140, 153]. Price et al. followed up their previous higher levels of DNA content and cytoplasmic
lung decellularization work by developing an proteins and lower levels of collagen and laminin
automated system for organ decellularization via peptides [112]. Based on this study, this group
perfusion that allows for the control of airway chose the SDS-based detergent decellularization
and vascular perfusion pressures [153]. Porcine method to decellularize clinically relevant
lungs were decellularized via perfusion with the porcine and human lungs. Acellular porcine and
following solutions in sequential order: deion- lung ECM scaffolds had a reduced DNA and
ized water (2 h), Triton X (5 h), sodium deoxy- cytoplasmic protein content, with preservation of
cholate (5 h), NaCl (5 h), DNase (2 h), and PBS ECM components’ elastin, collagen IV, fibronec-
(5 h). This automated method reduced decellular- tin, laminin, and GAGs. Additionally, the human
ization time from days to a day and made decel- ECM scaffolds were biocompatible with small
lularization of the lung more consistent when airway epithelial cells (SAECs), pulmonary alve-
compared to a manual decellularization. olar epithelial cells (PAECs), and human umbili-
Decellularization of the lung with this automated cal vein endothelial cells (HUVECs), shown by
system yielded decellularized lung ECM that cell adhesion onto the scaffold and cell viability
maintained the structural integrity of the tissue, after 5 days of culture. The whole human lung
removed α-galactose, reduced levels of DNA, ECM scaffold, and not just tissue slices, also was
and retained expression of collagen I, collagen successfully seeded with human PAECs that
IV, elastin, fibronectin, vitronectin, and laminin. were distributed via the airways by gravity and
Weymann et al. similarly used Triton X/SDC were cultured under constant pressure for 4 days
detergent-based perfusion decellularization with no visible tissue damage.
methods to successfully generate an acellular Another study also used the three differ-
lung ECM scaffold that retained the three- ent published detergent-based decellularization
methods (SDS [111, 141], Triton X/SDC [137, detergent solution. Another option is to perfuse
138, 153], and CHAPS [143]) on mouse lungs to the detergent solution through the salivary ductal
compare the effectiveness of each decellulariza- system, which also is quite small and delicate
tion method and the ability to recellularize each when attempting to cannulate, making it techni-
scaffold [139]. ECM scaffolds processed by the cally difficult to connect to an external perfusion
three different protocols were assessed for pres- system.
ervation of the overall tissue architecture by his- One attempt to translate this SDS-based
tological methods. The remaining ECM scaffolds method to the salivary gland was reported by Gao
from SDS and Triton X/SDC better maintained et al. [113]. Rat SMGs were successfully col-
the native lung tissue architecture when compared lected and decellularized via a previously pub-
to ECM scaffolds from CHAPS. Decellularized lished method that employed detergent solutions
ECM from all three methods was analyzed for containing SDS for 32 h, Triton X-100 for 2 h,
protein retention by performing immunohisto- and DNase I for 1 h [119, 155]. However, in this
chemistry, mass spectrometry, and western blot- study, perfusion decellularization techniques were
ting. Lung ECM decellularized with SDS and unsuccessful because of the difficult anatomy.
CHAPS had higher retention of collagen I, colla- After the decellularization process was com-
gen IV, and fibronectin when compared to Triton pleted, the remaining ECM was assessed through
X/SDC. Tissue decellularized with SDS and histology, immunohistochemistry, scanning elec-
Triton X/SDC had higher retention of laminin tron microscopy (SEM), and DNA and protein
compared to CHAPS. Decellularization meth- quantification analysis, showing that the ECM
ods using Triton X/SDC and CHAPS were more retained its natural structure and comparable lev-
effective at removing nuclear proteins, but less els of ECM proteins that include collagen I, col-
effective at removing cytoplasmic proteins when lagen IV, laminin, and fibronectin. Decellularized
compared to the method using SDS. Proteolytic ECM scaffolds were recellularized with primary
activity was measured in each decellularized rat SMG cells injected into the main duct of the
tissue treated with the various detergent-based gland and were maintained in suspension in a
methods, and methods using SDS and CHAPS rotary cell culture system. Recellularized scaf-
had the lowest proteolytic activity after 24 h. folds were assessed by histology and immunohis-
Remaining ECM from all three detergent-based tochemistry, which showed that SMG cells
methods were recellularized using bone marrow- remained in the scaffold after injection and
derived mouse mesenchymal stem cells and C10 expressed cell-cell adhesion markers (E-cadherin
mouse lung epithelial cells. Notably, there were and occludin), aquaporin 5 (AQP-5), and
no observable differences in cell attachment, α-amylase. This study was useful in showing the
proliferation, and apoptosis between the varying possibilities of using decellularization and recel-
detergent-based methods. lularization techniques to salivary gland regenera-
Lessons learned from decellularization of the tion purposes. However, the salivary glands of
lung studies are highly relevant and applicable to humans are much larger organs than those of rats.
any attempts to successfully decellularize the An animal source such as the pig that retains a
salivary glands. Based on the successful genera- similar facial anatomy to that of humans would be
tion of an acellular lung ECM scaffold that has more ideal for generating an organ scaffold that
the native three-dimensional tissue architecture could restore salivary function in humans.
preserved with the use of SDS-based decellular-
ization method, this method seems to be the most
useful in an organ with similar tissue architecture 8.4.2 Matrigel®/GFR Matrigel®
to the lung. However, decellularizing the salivary
gland poses challenges when compared to the Matrigel®, a basement membrane extract derived
lungs. The vasculature network surrounding the from Engelbreth-Holm-Swarm (EHS) mouse
salivary gland is too small to use for perfusion of sarcoma cells, is commonly used for 3D cell
culture because it contains growth factors and components on acinar development and cytodif-
basement membrane proteins found in the ECM, ferentiation, showing that laminin-1 and trans-
which greatly enhance the growth and viability forming growth factor-β3 are important for acinar
of cells that may not grow well in GF-depleted cell differentiation [164]. In another study, HSG
matrices. Matrigel® often is used for cell culture cells cultured on Matrigel® or laminin-1 and
to assess the biological activity of those cells in treated with transforming growth factor-α had
a growth factor and basement membrane replete a synergistic increase in α-amylase expression,
matrix. However, commercial Matrigel® typi- which was mediated through protein kinase C
cally is not considered a controlled and fully (PKC) and ERK1/2 [165]. A study performed by
characterized matrix as it contains various Maria et al. showed that 3D spheroids formed on
growth factors and basement membrane pro- Matrigel® expressed α-amylase, tight junction
teins, and their concentrations vary from batch proteins, AQP-5, CD44, and CD166 [114]. HSG
to batch. Although Matrigel® is a useful matrix cells grown in Matrigel® formed acini-like struc-
for studying the biological activity of cells, it is tures that displayed reduced cell proliferation
not ideal for regeneration or tissue engineering and increased in cell apoptosis [166]. Although
purposes as it is not biocompatible with humans decreased cell proliferation tends to accompany
because it is rodent derived. Despite this, the cell differentiation, further characterization of
results of studies using Matrigel® for salivary these salivary structures for acinar phenotypic
cell culture are insightful because they can pro- markers is needed to fully show that they are
vide a reference scaffold for comparison when differentiated structures. Additionally, the HSG
trying to generate and assess the usefulness of a cell line is listed on the International Cell Line
customized, biocompatible scaffold with ECM Authentication Committee’s (ICLAC) Database
proteins/peptides. of Cross-Contaminated or Misidentified Cell
Rat parotid gland cells (Par-C10) are salivary Lines (version 7.1, 2013-08-22) as likely to be
acinar cells isolated from rat parotid glands and contaminated by HeLa cells [167]. This is addi-
transformed with simian virus 40 [160]. Baker tionally reported by the European Collection
et al. have cultured Par-C10s on Matrigel® and of Cell Cultures (ECACC) and the Japanese
demonstrated their arrangement into acinar- Collection of Research Bioresources (JCRB)
like spheres that expressed tight junction pro- Cell Bank. For these reasons, HSG cell lines
teins, M3 receptor, and AQP-3, AQP-5 and are no longer a reliable model for salivary cell
were responsive to carbachol treatment [161]. function.
Other studies have used the human submandibu- Although cell lines are convenient to use
lar gland (HSG) cell line in combination with because of their hardiness and extended passag-
Matrigel®/basement membrane extract to study ing capability, they are not always representative
the in vitro cell organization and phenotype of of the cell phenotype/activity from the native tis-
salivary cells. Human submandibular gland sue. Additionally, immortalized cell lines are not
(HSG) cells, a cell line derived from interca- useful for regenerative or tissue engineering pur-
lated ducts of irradiated SMGs (now known to poses as they can lead to tumor formation as a
be contaminated with HeLa, see below), have result of their uncontrolled proliferation subse-
been shown to form a 3D reticular network quent to immortalization. Primary human sali-
resembling acini-like structures connected to vary cells grown on top of Matrigel® (“2.5D” cell
duct-like structures, with the HSG cells taking culture) formed acini-like spheroids that
on a differentiated acinar cell phenotype, i.e., expressed tight junction proteins, AQP-5,
well-developed Golgi apparatus, microvilli-like α-amylase, CD44, and CD166 [168]. In another
projections from the apical surface, the presence study, primary human SMG cells were isolated
of granules, amylase production, and a decrease from fresh salivary tissue that was processed with
in cell division [162, 163]. Hoffman et al. fur- a dissociation buffer to yield single cells that
ther investigated the role of basement membrane could be passaged on tissue culture plates [169].
These primary human SMG cells cultured on 8.4.3.1 Salivary Cell Culture Utilizing
basement membrane extract (BME) formed aci- Whole Proteins in the Matrix
notubular structures, which is believed to be Collagen I gels have been used frequently as a
differentiated because of decreased proliferation matrix for culture of both epithelial and mesen-
and increased expression of α-amylase, occludin, chymal cells. Collagen I is a heterotrimer com-
claudin-1, and claudin-3 [144]. Our group posed of two α1 chains and one α2 chain that
showed that primary human salivary stem/pro- organize into triple-helical fibers [171]. This
genitor cells (hS/PCs) cultured on Matrigel® self- fibrous protein polymer is the most abundant pro-
assembled into 3D acini-like structures [106]. tein in connective tissue, providing multiple
Salivary cells cultured on Matrigel® expressed integrin-binding sites to promote cell migration,
E-cadherin, AQP-5, K19, and α-amylase, show- phospho-FAK activation, phosphoinositol-3-
ing that these acini-like structures are at least par- kinase (PI3-kinase), and mitogen-activated pro-
tially differentiated. Additionally, the cells grown tein kinase (MAPK) cascades [27, 172].
on Matrigel® formed stress fibers and had local- Mouse-derived SMG salivary gland cells
ized phospho-FAK expression at focal adhesion grown on rat tail-derived collagen I gels formed
sites. The components present in Matrigel® that ductal-like structures that retained expression of
stimulate salivary-derived hS/PCs, or other sali- EGF when cultured with media containing tes-
vary stem/progenitor cells, to differentiate into tosterone, triiodothyronine, and hydrocortisone
the multiple salivary epithelial cell types include [173]. Rat SMG-derived salivary (RSMG-1) cells
laminin, perlecan/HSPG2, and collagen cultured under serum-free conditions in collagen
IV. Integration of these motifs into a scaffold I gels underwent branching morphogenesis and
used for tissue engineering can thus provide a formed branched structures when treated with
defined, scalable, and bioactive support for sali- HGF [174]. In another study, human primary
vary gland replacement. parotid gland cells were isolated using tissue
explant culture methods and used at early
passages [175]. These primary parotid gland cells
8.4.3 Biologically Derived organized into acini-like and ductal-like struc-
Polymers/Natural Polymers tures when grown on collagen I/GFR Matrigel®
gels [175]. These salivary structures expressed
As previously mentioned, Matrigel® contains α-amylase, AQP-5, and tight junction proteins
many components from the ECM, and, thus, it is occludin, claudin-1, and ZO-1. Burford-Mason
not easy to determine the connection between et al. reported the use of collagen I gels for cul-
biological responses of cells grown on Matrigel® turing rat SMG glands as a model for the patho-
and a specific protein component(s). Growing biology of salivary glands [176]. In this study, rat
cells on matrices with limited numbers of pre- SMG organoids cultured in collagen retained
cisely defined components helps to identify a their secretory activity, and each salivary cell
connection between the activity of cells and an type (acinar, ductal, and myoepithelial) retained
individual ECM component. Matrices can be their phenotypic marker expression for some
made up of biologically derived whole proteins, time in culture.
polysaccharides, peptides, or combinations of Although collagen I provides integrin-binding
these. As the size of some ECM proteins are very sites for cell attachment and activation of signal-
large, it is difficult to integrate multiple purified ing cascades, it is not the most appropriate ECM
proteins into one scaffold. For this reason, recent component to use for salivary gland regeneration
work has focused on mining small peptide as it is most commonly associated with connec-
sequences or recombinant protein subdomains tive tissue, wound healing, scarring, and fibrosis.
with specific biological activity that can be easily A scaffold with components from the basement
integrated into tissue engineering scaffolds or membrane, which directly surrounds the salivary
hydrogels [170]. epithelial cells, should be more effective for pro-
viding a scaffold that recreates the native envi- is a linear polymer of 1,4-β-linked D-glucose
ronment and signals and enables regeneration or units commonly found in the cell wall of plant
tissue engineering based reconstruction of the cells and produced by some microbial organisms;
salivary gland. its derivatives have been used for microbial and
Fibrin hydrogels provide another example of mammalian cell culture [179–181]. Agarose and
how intact proteins are used for salivary gland alginate also are polysaccharides, derived from
cell culture. Fibrin, a nonglobular protein derived natural biological sources, generally cell walls of
from fibrinogen, polymerizes into non-soluble algae. These materials are readily commercially
fibers that play a role in blood clot formation available, and both are frequently used in the
[177]. Fibrin polymers also can be cross-linked laboratory for a variety of biological applica-
to form hydrogels for 3D cell culture. Mouse tions. Both are largely considered as nonadhe-
parotid gland-derived salivary cells cultured on sive, “blank slates” when used with many
fibrin-based hydrogel polymerized with growth mammalian cell types, as neither contains the
factors (EGF and insulin-like growth factor-1) common adhesive proteins (e.g. fibronectin, vit-
and combined with growth factor-reduced (GFR) ronectin) nor other adhesion motifs found in a
Matrigel® formed 3D salivary spheroids [115]. complete ECM. Yet, these materials preserve
Laminin is a major component of the base- much of the mechanical properties (viscoelastic-
ment membrane that directs salivary gland devel- ity and low elastic modulus) found in native
opment. Specifically, Cantara et al. developed ECM. Agarose powders can be dissolved with
poly(lactic-co-glycolic acid) nanofiber scaf- heat and cooled to form aqueous hydrogels, while
folds that were functionalized with laminin-111 alginates are readily soluble and can be triggered
[178]. Two salivary cell lines were used in this to gel with ionic gradients (Ca2+ is most com-
study: (1) immortalized mouse ductal SMG epi- mon). Cells seeded onto these materials, or
thelial cells (SIMS) and (2) immortalized rat encapsulated within them, tend to aggregate into
SMG acinar epithelial cells (SMGC10). Both multicellular spheroids for survival. The
SIMS and SMGC10 cells grown on laminin- nonadhesive nature of these polysaccharide

111-functionalized nanofibers showed enhanced hydrogels preferentially supports cell-cell inter-
formation of tight junctions, shown by immu- actions, because there are no cell-ECM interac-
nofluorescence for ZO-1 and occludins at the tions available from these substrates.
apical-basal membrane border. Moreover, lam- HA-based systems have seen a significant
inin-111-modified nanofibers increased cell pro- increase in interest over recent years, as research-
liferation, without reducing cell viability. This ers have developed the tools to purify and func-
study showed that a matrix for salivary gland tionalize the base polymer. HA is a linear
tissue engineering could be functionalized with polysaccharide composed of the repeating disac-
basement membrane components, especially charide unit β-1,4-D glucuronic acid-β-1,3-N-
laminin-111, to promote the polarization and acetyl-D-glucosamine [182, 183]. HA is a
differentiation of salivary epithelial cells. In par- ubiquitously expressed glycosaminoglycan that
ticular, this study utilized the whole laminin-111 is found throughout the body as an essential com-
isoform in the scaffold for cell culture, but pep- ponent of ECM. Unlike many other hydrogel sys-
tides with the biologically active sequences also tems, HA is inherently bioactive, as many cells
can be synthesized and incorporated into a vari- have HA receptors for interaction with the base
ety of cell culture matrices. material. Moreover, cells express HA receptors
CD44 and hyaluronan-mediated motility recep-
8.4.3.2 Use of Polysaccharides in Tissue tor (RHAMM), allowing them to interact with
Engineering Scaffolds the surrounding HA polymers and activate cellu-
Polysaccharides, including cellulose, alginate, lar signaling processes [184, 185]. Carboxylic
agarose, and hyaluronic acid (HA), often are used acids along each saccharide unit offer preferred
to generate scaffolds for 3D cell culture. Cellulose reactive sites for selectively functionalizing the
polymer. Many orthogonal chemistries exist for neled or undermined wounds, or over superficial
initiating cross-linking and hydrogel network wounds. HyStem’s Renevia™ product is simi-
formation; however these must be cytocompati- larly based on the HyStem platform and is under
ble, especially for cells embedded within the pre- evaluation in clinical trials in Spain as an inject-
gelled matrix [186]. HA is ideal for tissue able hydrogel and cell carrier to treat facial wast-
engineering purposes as it is nonimmunogenic, ing in HIV+ patients. Renevia™ has passed Phase
biocompatible and biodegradable and can be I clinical trials for safety in humans as an inject-
chemically functionalized with biologically able in the retro-auricular area. Other licensees of
active peptides from the ECM [183, 186]. Also, this same technology target veterinary wound
as HA is negatively charged and highly polar, repair applications, and other human wound
therefore, hydrogels composed of HA are very repair or topical applications (e.g., BakerDVM’s
hydrophilic, hydrated gels with variable swelling Remend® corneal repair, wound repair, and eye
ratios [187]. lubrication drops). The progression of these and
The Prestwich Laboratory has published other biomaterials through FDA approval is a
extensively on the preparation and application of critical step in their eventual use in tissue engi-
various HA systems, and their work has resulted neering applications.
in commercially available preparations as the
HyStem® product line. Our lab has used the com-
mercially available HyStem® hydrogel composed 8.4.4 Protein Fragments/
of thiol-modified hyaluronic acid cross- linked Biologically Active Peptides
with polyethylene glycol diacyrlate (PEGDA) for
culture of salivary cells. Primary human salivary Scaffolds for tissue engineering purposes should
stem/progenitor-like cells (hS/PCs) cultured on be strategically designed to have bioactive compo-
2.5D or in 3D HA-based HyStem® hydrogels nents to induce the cellular processes needed for
formed acini-like structures that expressed cell- the development, or regeneration, of the desired
cell adhesion markers (β-catenin, E-cadherin, specific organ. Peptides derived from ECM com-
ZO-1), cholinergic M3-muscarinic receptors, ponents can be integrated into scaffolds to pro-
and β-adrenergic receptors [107, 108]. Salivary mote cell-ECM adhesion and cell-cell adhesion in
structures formed were proliferative up to cell culture. Additionally, peptides can be inte-
48 days in culture and were responsive to isopro- grated into hydrogels to aid in polymerization, as
terenol and norepinephrine stimulation [108]. in the case of step-growth thiol-ene polymeriza-
HA hydrogels seeded with hS/PCs were tions of poly(ethylene glycol) hydrogels [188].
implanted in a parotid gland three-fourths resec- There are several available laminin-derived
tion model, allowing the hydrogels to be directly peptides including IKVAV, AG10, AG32, and
in contact with the salivary bed [109]. Implanted AG73/MG73 from the α-chain, and YIGSR from
acini-like spheroids expressed HA receptors the β chain [189]. HSG formed spheroids that
CD168/RHAMM and CD44, also a progenitor resemble acini when cultured on scaffolds con-
marker, and retained progenitor marker CD44 taining AG73 (RKRLQVQLSIRT) or the homol-
expression after at least 1 week of being ogous MG73 (KNRLTIELEVRT); however,
implanted in the rat salivary bed [109]. these spheroids lacked polarity and lumens [116,
HyStem® materials are sold by BioTime and 162, 190]. Branching morphogenesis was inhib-
have received FDA 510(k) approval for market- ited in embryonic SMGs cultured on a combina-
ing in use as a wound matrix (K134037 for tion of Matrigel®, laminin, and nidogen when
BioTime’s Premvia™ product). It has been treated with soluble AG73 peptide [189].
assessed for ISO 10993 biocompatibility tests, Therefore, epithelial cell interaction with AG73
EN 13276-1 tests for primary wound dressings, peptide sequence is vital for branching, but it is
and human cytocompatibility tests. Premvia™ is not sufficient to promote acinar polarity and
approved for syringe-based delivery within tun- differentiation.
Basement
a Organ b Salivary acinus c d Peptide
membrane
HA-SH
Ac-PEG=RGD
MMP-labile crosslinker
Fig. 8.2 Illustration of multiple organ levels for consider- surrounds each acinar structure. (c) Within that BM, a
ation in salivary gland engineering. (a) Examination at the meshwork of protein and glycosaminoglycan components
organ level (bright field DIC image, scale bar is 20 μm) of arranges to direct cell polarization. (d) On the molecular
human parotid gland demonstrates organization into acini level, key elements of this supportive BM may be repro-
and ductal systems. (b) Closer study of the salivary acinus duced by employing modular peptides with controlled
with an antibody against collagen IV (green) and nuclear spatial and/or temporal positioning
stain (blue) identifies the basement membrane (BM) that
Our lab previously discovered a novel peptide MMP1 diminishes cleft formation [56, 192]. Our
(TWSKVGGHLRPGIVQSG) from domain IV lab has integrated these peptides into HA-based
of perlecan/HSPG2 that promotes cell adhesion, scaffolds for prostate xenograft-derived cancer
spreading, and FAK activation of various cell cells and is currently working on translating this
types [191]. Salivary hS/PCs cultured on per- scaffold for 3D culture of hS/PCs [193].
lecan domain IV peptide formed acini-like
spheroids that secreted α-amylase over 6 days at
comparable levels to structures formed on 8.5 Future Directions
Matrigel® [106]. Additionally, salivary spheroids
cultured on perlecan domain IV peptide Much has been learned about the important role
expressed tight junction protein E-cadherin, that ECM plays in salivary gland development,
water channel protein AQP-5, as well as acti- and thus, many efforts have focused on develop-
vated FAK. Other peptides can be integrated into ing biologically active scaffolds with incorpo-
three-dimensional cell culture for tissue engi- rated ECM components. An ideal scaffold for
neering to promote cell adhesion and motility, salivary gland engineering would contain cell
and these include fibronectin-derived RGD pep- adhesion sites and degradable cross-linkers that
tide and MMP-sensitive, collagen I-derived PQ salivary epithelial cells could selectively cleave
peptide commonly used as a cleavable cross- to make space for growth. Cell adhesion
linker (Fig. 8.2). Fibronectin is one of the ECM sequences from laminin, fibronectin, and colla-
proteins that enables cleft formation during gen are available for use in tissue engineering
branching morphogenesis, thus, integrating an scaffolds, but should be used in a manner that is
integrin-binding site from fibronectin may per- relevant to the development of the organ. For
mit clefting of the hS/PC spheroids in our example, cell adhesion sites that are vital for
HA-based hydrogel model. Additionally, colla- branching morphogenesis should be integrated
gen I is expressed and localized throughout the early on in the scaffold to promote branching of
surrounding salivary mesenchyme. MMPs salivary acini-like spheroids. It should be noted
expressed by the salivary epithelium during that salivary epithelial cells, in the case of hS/PCs
development play a role in modulating cleft for- grown in HA-based hydrogels, secrete their own
mation, as inhibiting MMP1 with TIMPs leads ECM after a few days in culture, although it is
to increased cleft formation in embryonic sali- unclear if it is well organized [106]. Therefore, it
vary glands, while treatment with exogenous is possible that a scaffold for successful salivary
gland engineering may need only the initial ECM much work remaining to be done to design a scaf-
components to trigger branching morphogenesis, fold with the proper spatial-temporal cues for com-
and the ECM components needed for terminal plete salivary gland functional restoration. The
differentiation may be expressed, secreted, and successful generation of a salivary gland in vitro will
organized later by salivary cells in culture. not only drastically improve head and neck cancer
A decellularized native ECM scaffold from patients’ quality of life, but it would be informative
the salivary gland also would be ideal to generate for tissue engineering of other secretory organs.
a salivary gland in vitro. The acellular scaffold
would contain the 3D tissue architecture present Acknowledgments This work was supported by NIH
in the native salivary gland, including a vascular R01DE022969 (MCFC, DAH), NIH F32DE024697
(DW), NSF Graduate Research Fellowship Program
network, branched lobules, and a ductal system. (GRFP) DGE-1450681 (MM), and private philanthropic
Clinically relevant acellular scaffolds of human donations. The authors would like to thank Dr. Robert
size, whether porcine or human derived, would L. Witt, Dr. Xinqiao Jia, Dr. Swati Pradhan-Bhatt, as well
be best to generate optimal decellularization and as all of the salivary team members for many helpful
discussions.
recellularization methods. However, such scaf-
folds would need to be efficiently decellularized,
leaving behind little to no antigens from the
donor that would cause any immunoreaction References
from the host. Moreover, the decellularized scaf-
fold also must be effectively sterilized to avoid 1. American Cancer Society. Cancer Facts & Figures.
causing any infection at the graft site. Regardless 2016.
2. Sullivan CA, Haddad RI, Tishler RB, et al.
of whether the scaffold is a hydrogel with ECM Chemoradiation-induced cell loss in human
peptides or a native acellular ECM, the mechani- submandibular glands. Laryngoscope. 2005;115:958–
cal properties must be equal or similar to the 64. doi:10.1097/01.MLG.0000163340.90211.87.
mechanical properties of the ECM of the natural 3. AlDuhaiby EZ, Breen S, Bissonnette J-P, et al.
A national survey of the availability of intensity-
gland. modulated radiation therapy and stereotactic
The cell type used to engineer a salivary gland radiosurgery in Canada. Radiat Oncol. 2012;7:18.
in vitro for implantation should be chosen wisely. doi:10.1186/1748-717X-7-18.
Development of a salivary gland either will 4. Wijers OB, Levendag PC, Braaksma MMJ, et al.
Patients with head and neck cancer cured by radia-
require all of the various differentiated salivary tion therapy: a survey of the dry mouth syndrome in
cell types (acinar, ductal, and myoepithelial) or a long-term survivors. Head Neck. 2002;24:737–47.
salivary stem/progenitor cell that can differenti- doi:10.1002/hed.10129.
ate into all of the cell types upon receiving the 5. Dirix P, Nuyts S, Van den Bogaert W. Radiation-
induced xerostomia in patients with head and neck
proper ECM and growth factor/morphogenic cancer: a literature review. Cancer. 2006;107:2525–
cues. In addition, salivary cells for implantation 34. doi:10.1002/cncr.22302.
into a human host must be human derived and 6. Toljanic JA, Heshmati RH, Bedard J-F. Dental fol-
ideally derived from the patient to receive the low-up compliance in a population of irradiated head
and neck cancer patients. Oral Surg Oral Med Oral
graft to avoid host rejection, i.e., an autograft Pathol Oral Radiol Endod. 2002;93:35–8. doi:http://
model. Our lab routinely uses isolated healthy, dx.doi.org/10.1067/moe.2002.116599.
non-tumorigenic salivary stem/progenitor cells 7. Ellis H. Anatomy of the salivary glands. Surg.
from the head and neck cancer patient obtained 2012;30:569–72. doi:10.1016/j.mpsur.2012.09.008.
8. Marsh PD, Do T, Beighton D, DA D. Influence of
before radiation therapy. The plan is to use these saliva on the oral microbiota. Periodontol 2000.
cells to grow the salivary gland implant during 2016;70:80–92. doi:10.1111/prd.12098.
the time that the patient receives radiotherapy, 9. Emmelin N, Garrett JR, Ohlin P. Neural con-
such that it will be ready to transplant back into trol of salivary myoepithelial cells. J Physiol.
1968;196:381–396.1.
the patient after therapy has been completed. 10. Emmelin N, Gjörstrup P. On the function of myoepi-
Although there have been advances in research thelial cells in salivary glands. J Physiol. 1973;230:
aimed toward generating a salivary gland, there is 185–98.
11. Chitturi RT, Veeravarmal V, Nirmal RM, Reddy BVR. from animal models to therapies. J Dent Res.
Myoepithelial cells (MEC) of the salivary glands 2009;88:894–903. doi:10.1177/0022034509343143.
in health and tumours. J Clin Diagn Res. 2015;9: 27. LeBleu VS, Macdonald B, Kalluri R. Structure
ZE14–8. doi:10.7860/JCDR/2015/11372.5707. and function of basement membranes. Exp
12. Diaz Silva G. Physiology of the salivary glands. Biol Med (Maywood). 2007;232:1121–9.
Odontol Chil. 1987;35:130–6. doi:10.3181/0703-MR-72.
13. Bornens M. Organelle positioning and cell polarity. 28. Yurchenco PD, Amenta PS, Patton BL. Basement
Nat Rev Mol Cell Biol. 2008;9:874–86. doi:10.1038/ membrane assembly, stability and activities
nrm2524. observed through a developmental lens. Matrix Biol.
14. Shin K, Fogg VC, Margolis B. Tight junctions and 2004;22:521–38. doi:10.1016/j.matbio.2003.10.006.
cell polarity. Annu Rev Cell Dev Biol. 2006;22:207– 29. Skalova A, Leivo I. Basement membrane pro-
35. doi:10.1146/annurev.cellbio.22.010305.104219. teins in salivary gland tumours. Virchows Arch A.
15. Dbouk HA, Mroue RM, El-Sabban ME, Talhouk RS. 1992;420:425–31. doi:10.1007/BF01600514.
Connexins: a myriad of functions extending beyond 30. Farach-Carson MC, Warren CR, Harrington DA,
assembly of gap junction channels. Cell Commun Carson DD. Border patrol: insights into the unique
Signal. 2009;7:1–17. doi:10.1186/1478-811X-7-4. role of perlecan/heparan sulfate proteoglycan 2 at
16. Shimono M, Muramatsu T, Hashimoto S, Inoue cell and tissue borders. Matrix Biol. 2014;34:64–79.
T. Connexin expression in the salivary glands. doi:10.1016/j.matbio.2013.08.004.
Eur J Morphol. 1996;34:197–202. doi:10.1076/ 31. Warren CR, Kassir E, Spurlin J, et al. Evolution of the
ejom.34.3.197.13027. perlecan/HSPG2 gene and its activation in regenerating
17. Lourenço SV, Lima DMC, Uyekita SH, et al. nematostella vectensis. PLoS One. 2015;10:e0124578.
Expression of beta-1 integrin in human developing doi:10.1371/journal.pone.0124578.
salivary glands and its parallel relation with matura- 32. Proctor GB, Carpenter GH. Regulation of salivary
tion markers: in situ hybridisation and immunofluo- gland function by autonomic nerves. Auton Neurosci.
rescence study. Arch Oral Biol. 2007;52:1064–71. 2007;133:3–18. doi:10.1016/j.autneu.2006.10.006.
doi:10.1016/j.archoralbio.2007.05.002. 33. Humphrey SP, Williamson RT. A review of
18. Lourenço SV, Nico MMS, Kapas S. Integrin expres- saliva: normal composition, flow, and function.
sion in developing human salivary glands. Oral J Prosthet Dent. 2001;85:162–9. doi:10.1067/
Surg Oral Med Oral Pathol Oral Radiol Endod. mpr.2001.113778.
2005;100:193. doi:10.1016/j.tripleo.2005.05.036. 34. Borghese E. The development in vitro of the sub-
19. Segawa A, Takemura H, Yamashina S. Calcium sig- mandibular gland and sublingual glands of the mus
naling in tissue: diversity and domain-specific inte- musculus. J Anat. 1950;84:287–302.
gration of individual cell response in salivary glands. 35. Borghese E. Explanation experiments on the influ-
J Cell Sci. 2002;115:1869–76. ence of connective tissue capsule on the develop-
20. Meda P, Pepper MS, Traub O, et al. Differential ment of the epithelial part of the submandibular
expression of gap junction connexins in endocrine and gland of Mus musculus. J Anat. 1950;84:303–18.
exocrine glands. Endocrinology. 1993;133:2371–8. 36. Tucker AS. Salivary gland development. Semin
doi:10.1210/endo.133.5.8404689. Cell Dev Biol. 2007;18:237–44. doi:10.1016/j.
21. Castle JD. Protein secretion by rat parotid acinar semcdb.2007.01.006.
cells: pathways and regulation. Ann N Y Acad Sci. 37. Patel VN, Rebustini IT, Hoffman MP. Salivary gland
1998;842:115–24. doi:10.1111/j.1749-6632.1998. branchingmorphogenesis.Differentiation.2006;74:349–
tb09639.x. 64. doi:10.1111/j.1432-0436.2006.00088.x.
22. Gorr S-U, Venkatesh SG, Darling DS. Parotid secre- 38. Ferreira JN, Hoffman MP. Interactions between
tory granules: crossroads of secretory pathways and developing nerves and salivary glands.
protein storage. J Dent Res. 2005;84:500–9. Organogenesis. 2013;9:199–205.
23. Ishikawa Y, Cho G, Yuan Z, et al. Water chan- 39. Werner Kahle M, Michael Frotscher M. Color atlas
nels and zymogen granules in salivary glands. of human anatomy. Volume 3, nervous system and
J Pharmacol Sci. 2006;100:495–512. doi:10.1254/ sensory organs. Stuttgart/New York: Thieme; 2003.
jphs.CRJ06007X. 40. Henriksson R, Carlsöö B, Danielsson Å, et al.
24. Nagler R, Marmary Y, Fox PC, et al. Irradiation- Developmental influences of the sympathetic nervous
induced damage to the salivary glands: the role system on rat parotid gland. J Neurol Sci. 1985;71:183–
of redox-active iron and copper. Radiat Res. 91. doi:10.1016/0022-510X(85)90058-9.
1997;147:468–76. doi:10.2307/3579504. 41. Coughlin MD. Early development of para-
25. Nagler RM, Laufer D. Protection against sympathetic nerves in the mouse subman-
irradiation-induced damage to salivary glands dibular gland. Dev Biol. 1975;43:123–39.
by adrenergic agonist administration. Int doi:10.1016/0012-1606(75)90136-0.
J Radiat Oncol. 1998;40:477–81. doi:10.1016/ 42. Coughlin MD. Target organ stimulation of para-
S0360-3016(97)00574-9. sympathetic nerve growth in the developing mouse
26. Grundmann O, Mitchell GC, Limesand submandibular gland. Dev Biol. 1975;43:140–58.
KH. Sensitivity of salivary glands to radiation: doi:10.1016/0012-1606(75)90137-2.
43. Knox SM, Lombaert IM, Reed X, et al. 57. Spooner BS, Faubion JM. Collagen involvement
Parasympathetic innervation maintains epithelial proin branching morphogenesis of embryonic lung
genitor cells during salivary organogenesis. Science. and salivary gland. Dev Biol. 1980;77:84–102.
2010;329:1645–7. doi:10.1126/science.1192046. doi:10.1016/0012-1606(80)90458-3.
44. Harunaga JS, Doyle AD, Yamada KM. Local 58. Grobstein C, Cohen J. Collagenase: effect on the
and global dynamics of the basement mem- morphogenesis of embryonic salivary epithelium
brane during branching morphogenesis require in vitro. Science (80- ). 1965;150:626–8.
protease activity and actomyosin contractil- 59. Durbeej M. Laminins. Cell Tissue Res.
ity. Dev Biol. 2014;394:197–205. doi:10.1016/j. 2010;339:259–68. doi:10.1007/s00441-009-0838-2.
ydbio.2014.08.014. 60. Kadoya Y, Yamashina S. Distribution of alpha 6 inte-
45. Koyama N, Hayashi T, Kashimata M. Regulation grin subunit in developing mouse submandibular
of branching morphogenesis in fetal mouse sub- gland. J Histochem Cytochem. 1993;41:1707–14.
mandibular gland by signaling pathways activated 61. Belkin AM, Stepp MA. Integrins as receptors for
by growth factors and α6 integrin. J Oral Biosci. laminins. Microsc Res Tech. 2000;51:280–301.
2011;53:298–303. doi:http://dx.doi.org/10.1016/ doi:10.1002/1097-0029(20001101)51:3<280::AID-
S1349-0079(11)80022-8. JEMT7>3.0.CO;2-O.
46. Sakai T, Larsen M, Yamada KM. Fibronectin 62. KadoyaY,Yamashina S. Salivary gland morphogenesis
requirement in branching morphogenesis. Nature. and basement membranes. Anat Sci Int. 2005;80:71–
2003;423:876–81. 9. doi:10.1111/j.1447-073x.2005.00102.x.
47. Kallman F, Grobstein C. Source of collagen at 63. Miner JH, Yurchenco PD. Laminin functions
epitheliomesenchymal interfaces during induc- in tissue morphogenesis. Annu Rev Cell Dev
tive interaction. Dev Biol. 1965;11:169–83. Biol. 2004;20:255–84. doi:10.1146/annurev.
doi:10.1016/0012-1606(65)90055-2. cellbio.20.010403.094555.
48. Bernfield MR. Collagen synthesis during epithelio- 64. Ekblom P, Lonai P, Talts JF. Expression and biologi-
mesenchymal interactions. Dev Biol. 1970;22:213– cal role of laminin-1. Matrix Biol. 2003;22:35–47.
31. doi:10.1016/0012-1606(70)90151-X. doi:10.1016/S0945-053X(03)00015-5.
49. Kadoya Y, Kadoya K, Durbeej M, et al. Antibodies 65. Kadoya Y, Mochizuki M, Nomizu M, et al. Role for
against domain E3 of laminin-1 and integrin alpha laminin-α5 chain LG4 module in epithelial branch-
6 subunit perturb branching epithelial morpho- ing morphogenesis. Dev Biol. 2003;263:153–64.
genesis of submandibular gland, but by different doi:10.1016/S0012-1606(03)00446-9.
modes. J Cell Biol. 1995;129:521–34. doi:10.1083/ 66. Kadoya Y, Nomizu M, Sorokin LM, et al. Laminin
jcb.129.2.521. alpha1 chain G domain peptide, RKRLQVQLSIRT,
50. Durbeej M, Talts JF, Henry MD, et al. Dystroglycan inhibits epithelial branching morphogenesis of
binding to laminin alpha1LG4 module influences cultured embryonic mouse submandibular gland.
epithelial morphogenesis of salivary gland and Dev Dyn. 1998;212:394–402. doi:10.1002/
lung in vitro. Differentiation. 2001;69:121–34. (SICI)1097-0177(199807)212:3<394::AID-
doi:10.1046/j.1432-0436.2001.690206.x. AJA7>3.0.CO;2-C.
51. Hardman P, Spooner BS. Localization of extra- 67. Kadoya Y, Yamashina S. Localization of laminin-5,
cellular matrix components in developing mouse HD1/plectin, and BP230 in the submandibular
salivary glands by confocal microscopy. Anat Rec. glands of developing and adult mice. Histochem Cell
1992;234:452–9. doi:10.1002/ar.1092340315. Biol. 1999;112:417–25.
52. Kumagai M, Sato I. Immunolocalization of fibronec- 68. Patel VN, Knox SM, Likar KM, et al. Heparanase
tin and collagen types I and III in human fetal parotid cleavage of perlecan heparan sulfate modulates
and submandibular glands. Cells Tissues Organs. FGF10 activity during ex vivo submandibular
2003;173:184–90. gland branching morphogenesis. Development.
53. Nakanishi Y, Nogawa H, Hashimoto Y, et al. 2007;134:4177–86. doi:10.1242/dev.011171.
Accumulation of collagen III at the cleft points 69. Patel VN, Likar KM, Zisman-Rozen S, et al. Specific
of developing mouse submandibular epithelium. heparan sulfate structures modulate FGF10-mediated
Development. 1988;104:51–9. submandibular gland epithelial morphogenesis and
54. Hayakawa T, Kishi J, Nakanishi Y. Salivary gland differentiation. J Biol Chem. 2008;283:9308–17.
morphogenesis: possible involvement of collage- doi:10.1074/jbc.M709995200.
nase. Matrix Suppl. 1992;1:344–51. 70. Knox SM, Whitelock JM. Perlecan: how does one
55. Spooner BS, Thompson-Pletscher HA, Stokes B, molecule do so many things? Cell Mol Life Sci.
Bassett KE. Extracellular matrix involvement in 2006;63:2435–45. doi:10.1007/s00018-006-6162-z.
epithelial branching morphogenesis. Dev Biol. 71. Farach-Carson MC, Carson DD. Perlecan--a mul-
1986;3:225–60. tifunctional extracellular proteoglycan scaffold.
56. Fukuda Y, Masuda Y, Kishi J, et al. The role of inter- Glycobiology. 2007;17:897–905. doi:10.1093/
stitial collagens in cleft formation of mouse embry- glycob/cwm043.
onic submandibular gland during initial branching. 72. Hopf M, Göhring W, Kohfeldt E, et al. Recombinant
Development. 1988;103:259–67. domain IV of perlecan binds to nidogens,
laminin-nidogen complex, fibronectin, fibulin-2 and ior. Biochim Biophys Acta. 2004;1692:103–19.
heparin. Eur J Biochem. 1999;259:917–25. doi:10.1016/j.bbamcr.2004.04.007.
73. Hoffman MP, Kidder BL, Steinberg ZL, et al. Gene 87. Green K, Jones J. Desmosomes and hemidesmo-
expression profiles of mouse submandibular gland somes: structure and function of molecular compo-
development: FGFR1 regulates branching morpho- nents. FASEB J. 1996;10:871–81.
genesis in vitro through BMP- and FGF-dependent 88. Litjens SHM, Sonnenberg A. Hemidesmosomes. In:
mechanisms. Development. 2002;129:5767–78. Encyclopedic reference genomics and proteomics
74. Ohuchi H, Hori Y, Yamasaki M, et al. FGF10 acts molecular medicine. Berlin/Heidelberg: Springer
as a major ligand for FGF receptor 2 IIIb in mouse Berlin Heidelberg; 2006. p. 754–8.
multi-organ development. Biochem Biophys 89. Yu W, Datta A, Leroy P, et al. Beta1-integrin orients
Res Commun. 2000;277:643–9. doi:10.1006/ epithelial polarity via Rac1 and laminin. Mol Biol Cell.
bbrc.2000.3721. 2005;16:433–45. doi:10.1091/mbc.E04-05-0435.
75. De Moerlooze L, Spencer-Dene B, Revest JM, et al. 90. Liu KD, Datta A, Yu W, et al. Rac1 is required
An important role for the IIIb isoform of fibroblast for reorientation of polarity and lumen forma-
growth factor receptor 2 (FGFR2) in mesenchymal- tion through a PI 3-kinase-dependent pathway.
epithelial signalling during mouse organogenesis. Am J Physiol Renal Physiol. 2007;293:F1633–40.
Development. 2000;127:483–92. doi:10.1152/ajprenal.00053.2007.
76. Steinberg Z, Myers C, Heim VM, et al. FGFR2b 91. Assémat E, Bazellières E, Pallesi-Pocachard E,
signaling regulates ex vivo submandibular gland et al. Polarity complex proteins. Biochim Biophys
epithelial cell proliferation and branching mor- Acta Biomembr. 2008;1778:614–30. doi:10.1016/j.
phogenesis. Development. 2005;132:1223–34. bbamem.2007.08.029.
doi:10.1242/dev.01690. 92. Bulgakova NA, Knust E. The Crumbs complex: from
77. Roignot J, Peng X, Mostov K. Polarity in mamma- epithelial-cell polarity to retinal degeneration. J Cell
lian epithelial morphogenesis. Cold Spring Harb Sci. 2009;122:2587–96. doi:10.1242/jcs.023648.
Perspect Biol. 2013;5(2) doi:10.1101/cshperspect. 93. Goldstein B, Macara IG. The PAR proteins: funda-
a013789. mental players in animal cell polarization. Dev Cell.
78. Baker OJ. Tight junctions in salivary epithe- 2007;13:609–22. doi:10.1016/j.devcel.2007.10.007.
lium. J Biomed Biotechnol. 2010;2010:278948. 94. Chen J, Zhang M. The Par3/Par6/aPKC com-
doi:10.1155/2010/278948. plex and epithelial cell polarity. Exp Cell Res.
79. Trimble WS, Grinstein S. Barriers to the free dif- 2013;319:1357–64. doi:http://dx.doi.org/10.1016/j.
fusion of proteins and lipids in the plasma mem- yexcr.2013.03.021.
brane. J Cell Biol. 2015;208:259–71. doi:10.1083/ 95. Iden S, Collard JG. Crosstalk between small
jcb.201410071. GTPases and polarity proteins in cell polarization.
80. Hartsock A, Nelson WJ. Adherens and tight junc- Nat Rev Mol Cell Biol. 2008;9:846–59. doi:10.1038/
tions: structure, function and connections to nrm2521.
the actin cytoskeleton. Biochim Biophys Acta 96. Joberty G, Petersen C, Gao L, Macara IG. The cell-
Biomembr. 2008;1778:660–9. doi:10.1016/j. polarity protein Par6 links Par3 and atypical protein
bbamem.2007.07.012. kinase C to Cdc42. Nat Cell Biol. 2000;2:531–9.
81. Harris TJ, Tepass U. Adherens junctions: from mol- doi:10.1038/35019573.
ecules to morphogenesis. Nat Rev Mol Cell Biol. 97. Su W-H, Mruk DD, Wong EWP, et al. Polarity pro-
2010; 11:502–14. doi: nrm2927 [pii]\r10.1038/ tein complex Scribble/Lgl/Dlg and epithelial cell
nrm2927. barriers. Adv Exp Med Biol. 2012;763:149–70.
82. Nekrasova O, Green KJ. Desmosome assembly 98. Rebustini IT, Hayashi T, Reynolds AD, et al. miR-
and dynamics. Trends Cell Biol. 2013;23:537–46. 200c regulates FGFR-dependent epithelial prolifera-
doi:10.1016/j.tcb.2013.06.004. tion via Vldlr during submandibular gland branching
83. Garrod D, Chidgey M. Desmosome structure, morphogenesis. Development. 2012;139:191–202.
composition and function. Biochim Biophys Acta doi:10.1242/dev.070151.
Biomembr. 2008;1778:572–87. doi:http://dx.doi. 99. Lombaert IMA, Brunsting JF, Wierenga PK, et al.
org/10.1016/j.bbamem.2007.07.014. Rescue of salivary gland function after stem cell
84. Delva E, Tucker DK, Kowalczyk AP. The des- transplantation in irradiated glands. PLoS One.
mosome. Cold Spring Harb Perspect Biol. 2008;3:e2063. doi:10.1371/journal.pone.0002063.
2009;1(2):a002543. doi:10.1101/cshperspect. 100. Feng J, van der Zwaag M, Stokman MA, et al.
a002543. Isolation and characterization of human salivary
85. Yamada A, Futagi M, Fukumoto E, et al. Connexin gland cells for stem cell transplantation to reduce
43 is necessary for salivary gland branching mor- radiation-induced hyposalivation. Radiother Oncol.
phogenesis and FGF10-induced ERK1/2 phosphory- 2009;92:466–71.
lation. J Biol Chem. 2016;291:904–12. doi:10.1074/ 101. Nanduri LSY, Maimets M, Pringle SA, et al.
jbc.M115.674663. Regeneration of irradiated salivary glands with
86. Wozniak MA, Modzelewska K, Kwong L, Keely stem cell marker expressing cells. Radiother Oncol.
PJ. Focal adhesion regulation of cell behav- 2011;99:367–72. doi:10.1016/j.radonc.2011.05.085.
102. Nanduri LSY, Lombaert IM, der Zwaag M v, 115. McCall AD, Nelson JW, Leigh NJ, et al. Growth
et al. Salisphere derived c-Kit(+) cell transplanta- factors polymerized within fibrin hydrogel promote
tion restores tissue homeostasis in irradiated sali- amylase production in parotid cells. Tissue Eng
vary gland. Radiother Oncol. 2013;108:458–63. Part A. 2013;19(19–20):2215–25. doi:10.1089/ten.
doi:10.1016/j.radonc.2013.05.020. tea.2012.0674.
103. Sato A, Okumura K, Matsumoto S, et al. Isolation, 116. Yamada Y, Hozumi K, Aso A, et al. Laminin active
tissue localization, and cellular characterization peptide/agarose matrices as multifunctional bio-
of progenitors derived from adult human salivary materials for tissue engineering. Biomaterials.
glands. Cloning Stem Cells. 2007;9:191–205. 2012;33:4118–25. doi:10.1016/j.biomaterials.2012.
doi:10.1089/clo.2006.0054. 02.044.
104. Rotter N, Oder J, Schlenke P, et al. Isolation and 117. Miller JS, Shen CJ, Legant WR, et al. Bioactive
characterization of adult stem cells from human hydrogels made from step-growth derived PEG-
salivary glands. Stem Cells Dev. 2008;17:509–18. peptide macromers. Biomaterials. 2010;31:3736–43.
doi:10.1089/scd.2007.0180. doi:10.1016/j.biomaterials.2010.01.058.Bioactive.
105. Wang Y, Shnyra A, Africa C, et al. Activation of 118. American Transplant Foundation. About trans-
the extrinsic apoptotic pathway by TNF-alpha in plant: facts and myths. 2016. http://www.ameri-
human salivary gland (HSG) cells in vitro, suggests cantransplantfoundation.org/about-transplant/
a role for the TNF receptor (TNF-R) and intercel- facts-and-myths/.
lular adhesion molecule-1 (ICAM-1) in Sjögren’s 119. Gilbert TW, Sellaro TL, Badylak SF. Decellularization
syndrome- associated autoimmune sialadenitis. of tissues and organs. Biomaterials. 2006;27:3675–
Arch Oral Biol. 2009;54:986–96. doi:10.1016/j. 83. doi:10.1016/j.biomaterials.2006.02.014.
archoralbio.2009.07.011. 120. Crapo PM, Gilbert TW, Badylak SF. An overview of
106. Pradhan S, Zhang C, Jia X, et al. Perlecan domain tissue and whole organ decellularization processes.
IV peptide stimulates salivary gland cell assem- Biomaterials. 2011;32:3233–43. doi:10.1016/j.
bly in vitro. Tissue Eng Part A. 2009;15:3309–20. biomaterials.2011.01.057.
doi:10.1089/ten.TEA.2008.0669. 121. Cortiella J, Niles J, Cantu A, et al. Influence of acel-
107. Pradhan S, Liu C, Zhang C, et al. Lumen formation lular natural lung matrix on murine embryonic stem
in three-dimensional cultures of salivary acinar cells. cell differentiation and tissue formation. Tissue
Otolaryngol Head Neck Surg. 2010;142:191–5. Eng Part A. 2010;16:2565–80. doi:10.1089/ten.
doi:10.1016/j.otohns.2009.10.039. tea.2009.0730.
108. Pradhan-Bhatt S, Harrington DA, Duncan RL, et al. 122. Hung S-H, Su C-H, Lee F-P, Tseng H. Larynx decel-
Implantable three-dimensional salivary spheroid lularization: combining freeze-drying and sonication
assemblies demonstrate fluid and protein secretory as an effective method. J Voice. 2013;27:289–94.
responses to neurotransmitters. Tissue Eng Part A. doi:http://dx.doi.org/10.1016/j.jvoice.2013.01.018.
2013;19:1610–20. doi:10.1089/ten.tea.2012.0301. 123. Burk J, Erbe I, Berner D, et al. Freeze-thaw cycles
109. Pradhan-Bhatt S, Harrington DA, Duncan RL, et al. enhance decellularization of large tendons. Tissue
A novel in vivo model for evaluating functional Eng Part C Methods. 2014;20:276–84. doi:10.1089/
restoration of a tissue engineered salivary gland. ten.TEC.2012.0760.
Laryngoscope. 2014;124:456–61. doi:10.1002/ 124. Gardin C, Ricci S, Ferroni L, et al. Decellularization
lary.24297. and delipidation protocols of bovine bone and peri-
110. Jeong J, Baek H, Kim Y-J, et al. Human salivary cardium for bone grafting and guided bone regenera-
gland stem cells ameliorate hyposalivation of tion procedures. PLoS One. 2015;10(7):e0132344.
radiation-damaged rat salivary glands. Exp Mol doi:10.1371/journal.pone.0132344.
Med. 2013;45:e58. doi:10.1038/emm.2013.121. 125. Xu H, Xu B, Yang Q, et al. Comparison of decel-
111. Song JJ, Kim SS, Liu Z, et al. Enhanced in vivo func- lularization protocols for preparing a decellu-
tion of bioartificial lungs in rats. Ann Thorac Surg. larized porcine annulus fibrosus scaffold. PLoS
2011;92:998–1006. doi:http://dx.doi.org/10.1016/j. One. 2014;9(1):e86723. doi:10.1371/journal.
athoracsur.2011.05.018. pone.0086723.
112. Gilpin SE, Guyette JP, Gonzalez G, et al. Perfusion 126. Porzionato A, Sfriso MM, Macchi V, et al.
decellularization of human and porcine lungs: Decellularized omentum as novel biologic scaffold
bringing the matrix to clinical scale. J Heart Lung for reconstructive surgery and regenerative medi-
Transplant. 2014;33:298–308. doi:http://dx.doi. cine. Eur J Histochem. 2013;57:24–30. doi:10.4081/
org/10.1016/j.healun.2013.10.030. ejh.2013.e4.
113. Gao Z, Wu T, Xu J, et al. Generation of bioartifi- 127. Azhim A, Syazwani N, Morimoto Y, et al. The use
cial salivary gland using whole-organ decellularized of sonication treatment to decellularize aortic tissues
bioscaffold. Cells Tissues Organs. 2014;200:171–80. for preparation of bioscaffolds. J Biomater Appl.
114. Maria OM, Maria O, Liu Y, et al. Matrigel improves 2014;29:130–41. doi:10.1177/0885328213517579.
functional properties of human submandibular 128. Oliveira AC, Garzón I, Ionescu AM, et al.
salivary gland cell line. Int J Biochem Cell Biol. Evaluation of small intestine grafts decellulariza-
2011;43:622–31. doi:10.1016/j.biocel.2011.01.001. tion methods for corneal tissue engineering. PLoS
One. 2013;8(6):e66538. doi:10.1371/journal. 141. Ott HC, Clippinger B, Conrad C, et al. Regeneration
pone.0066538. and orthotopic transplantation of a bioartificial lung.
129. Sarig U, Au-Yeung GCT, Wang Y, et al. Thick acel- Nat Med. 2010;16:927–33.
lular heart extracellular matrix with inherent vas- 142. Hudson TW, Liu SY, Schmidt CE. Engineering
culature: a potential platform for myocardial tissue an improved acellular nerve graft via optimized
regeneration. Tissue Eng Part A. 2012;18:2125–37. chemical processing. Tissue Eng. 2004;10:1346–58.
doi:10.1089/ten.TEA.2011.0586. doi:10.1089/ten.2004.10.1346.
130. Barnes CA, Brison J, Michel R, et al. The surface 143. Petersen TH, Calle EA, Zhao L, et al. Tissue-
molecular functionality of decellularized extra- engineered lungs for in vivo implantation. Science.
cellular matrices. Biomaterials. 2011;32:137–43. 2010;329:538–41. doi:10.1126/science.1189345.
doi:10.1016/j.biomaterials.2010.09.007. 144. Elder BD, Eleswarapu SV, Athanasiou
131. Nara S, Chameettachal S, Midha S, et al. Preservation KA. Extraction techniques for the decellularization
of biomacromolecular composition and ultrastruc- of tissue engineered articular cartilage constructs.
ture of a decellularized cornea using a perfusion Biomaterials. 2009;30:3749–56. doi:10.1016/j.
bioreactor. RSC Adv. 2016;6:2225–40. doi:10.1039/ biomaterials.2009.03.050.
C5RA20745B. 145. Meyer SR, Nagendran J, Desai LS, et al.
132. Baiguera S, Del Gaudio C, Kuevda E, et al. Dynamic Decellularization reduces the immune response
decellularization and cross-linking of rat tracheal to aortic valve allografts in the rat. J Thorac
matrix. Biomaterials. 2014;35:6344–50. doi:http:// Cardiovasc Surg. 2005;130:469–76. doi:10.1016/j.
dx.doi.org/10.1016/j.biomaterials.2014.04.070. jtcvs.2005.03.021.
133. Struecker B, Hillebrandt KH, Voitl R, et al. Porcine 146. Choi YC, Choi JS, Kim BS, et al. Decellularized
liver decellularization under oscillating pressure extracellular matrix derived from porcine adipose
conditions: a technical refinement to improve the tissue as a xenogeneic biomaterial for tissue engi-
homogeneity of the decellularization process. Tissue neering. Tissue Eng Part C Methods. 2012;18:866–
Eng Part C Methods. 2015;21:303–13. doi:10.1089/ 76. doi:10.1089/ten.tec.2012.0009.
ten.tec.2014.0321. 147. Lumpkins SB, Pierre N, McFetridge PS. A mechani-
134. Faulk DM, Wildemann JD, Badylak SF. cal evaluation of three decellularization methods
Decellularization and cell seeding of whole liver in the design of a xenogeneic scaffold for tissue
biologic scaffolds composed of extracellular matrix. engineering the temporomandibular joint disc. Acta
J Clin Exp Hepatol. 2015;5:69–80. doi:http://dx.doi. Biomater. 2008;4:808–16. doi:http://dx.doi.
org/10.1016/j.jceh.2014.03.043. org/10.1016/j.actbio.2008.01.016.
135. Momtahan N, Poornejad N, Struk JA, et al.Automation 148. Gratzer PF, Harrison RD, Woods T. Matrix alteration
of pressure control improves whole porcine heart and not residual sodium dodecyl sulfate cytotoxicity
decellularization. Tissue Eng Part C Methods. affects the cellular repopulation of a decellularized
2015;21(11):1148–61. 00:150714090832003 matrix. Tissue Eng. 2006;12:2975–83. doi:10.1089/
doi:10.1089/ten.tec.2014.0709. ten.2006.12.ft-234.
136. Sullivan DC, Mirmalek-Sani S-H, Deegan DB, 149. Sawkins MJ, Bowen W, Dhadda P, et al. Hydrogels
et al. Decellularization methods of porcine kid- derived from demineralized and decellularized bone
neys for whole organ engineering using a high- extracellular matrix. Acta Biomater. 2013;9:7865–
throughput system. Biomaterials. 2012;33:7756–64. 73. doi:10.1016/j.actbio.2013.04.029.
doi:10.1016/j.biomaterials.2012.07.023. 150. Saghizadeh M, Winkler MA, Kramerov AA, et al.
137. Price AP, England KA, Matson AM, et al. A simple alkaline method for decellularizing
Development of a decellularized lung bioreactor sys- human amniotic membrane for cell culture. PLoS
tem for bioengineering the lung: the matrix reloaded. One. 2013;8(11):e79632. doi:10.1371/journal.
Tissue Eng Part A. 2010;16:2581–91. doi:10.1089/ pone.0079632.
ten.tea.2009.0659. 151. Ott HC, Matthiesen TS, Goh S-K, et al. Perfusion-
138. Daly AB, Wallis JM, Borg ZD, et al. Initial binding decellularized matrix: using nature’s plat-
and recellularization of decellularized mouse lung form to engineer a bioartificial heart. Nat Med.
scaffolds with bone marrow-derived mesenchymal 2008;14:213–21.
stromal cells. Tissue Eng Part A. 2011;18:1–16. 152. Robertson MJ, Dries-Devlin JL, Kren SM, et al.
doi:10.1089/ten.tea.2011.0301. Optimizing recellularization of whole decel-
139. Wallis JM, Borg ZD, Daly AB, et al. Comparative lularized heart extracellular matrix. PLoS One.
assessment of detergent-based protocols for mouse 2014;9:e90406. doi:10.1371/journal.pone.0090406.
lung de-cellularization and re-cellularization. Tissue 153. Price AP, Godin LM, Domek A, et al. Automated
Eng Part C Methods. 2012;18:420–32. doi:10.1089/ decellularization of intact, human-sized lungs for
ten.tec.2011.0567. tissue engineering. Tissue Eng Part C Methods.
140. Weymann A, Patil NP, Sabashnikov A, et al. 2015;21:94–103. doi:10.1089/ten.tec.2013.0756.
Perfusion-decellularization of porcine lung and 154. Moroni F, Mirabella T. Decellularized matrices for
trachea for respiratory bioengineering. Artif Organs. cardiovascular tissue engineering. Am J Stem Cells.
2015;39:1024–32. doi:10.1111/aor.12481. 2014;3:1–20.
155. Traphagen SB, Fourligas N, Xylas JF, et al. 169. Szlávik V, Szabó B, Vicsek T, et al. Differentiation
Characterization of natural, decellularized and of primary human submandibular gland cells
reseeded porcine tooth bud matrices. Biomaterials. cultured on basement membrane extract. Tissue
2012;33:5287–96. doi:10.1016/j.biomaterials.2012. Eng Part A. 2008;14:1915–26. doi:10.1089/ten.
04.010. tea.2007.0208.
156. Macchiarini P, Jungebluth P, Go T, et al. Clinical 170. Pradhan S, Farach-Carson MC. Mining the extra-
transplantation of a tissue-engineered air- cellular matrix for tissue engineering applica-
way. Lancet. 2008;372:2023–30. doi:10.1016/ tions. Regen Med. 2010;5:961–70. doi:10.2217/
S0140-6736(08)61598-6. rme.10.61.
157. Chuang P-T, McMahon AP. Branching morpho- 171. Mouw JK, Ou G, Weaver VM. Extracellular matrix
genesis of the lung: new molecular insights into assembly: a multiscale deconstruction. Nat Rev Mol
an old problem. Trends Cell Biol. 2016;13:86–91. Cell Biol. 2014;15:771–85.
doi:10.1016/S0962-8924(02)00031-4. 172. Giancotti FG, Ruoslahti E. Integrin signaling.
158. Andrew DJ, Ewald AJ. Morphogenesis of epithe- Science (80- ). 1999;285:1028–33.
lial tubes: insights into tube formation, elonga- 173. Durban EM. Mouse submandibular salivary epithe-
tion, and elaboration. Dev Biol. 2010;341:34–55. lial cell growth and differentiation in long-term cul-
doi:10.1016/j.ydbio.2009.09.024. ture: influence of the extracellular matrix. In Vitro
159. Swarr DT, Morrisey EE. Lung endoderm mor- Cell Dev Biol. 1990;26:33–43.
phogenesis: gasping for form and function. Annu 174. Furue M, Okamoto T, Hayashi H, et al. Effects of
Rev Cell Dev Biol. 2015;31:553–73. doi:10.1146/ hepatocyte growth factor (HGF) and activin A on the
annurev-cellbio-100814-125249. morphogenesis of rat submandibular gland-derived
160. Quissell DO, Barzen KA, Redman RS, et al. epithelial cells in serum-free collagen gel culture. In
Development and characterization of SV40 immor- Vitro Cell Dev Biol Anim. 1999;35:131–5.
talized rat parotid acinar cell lines. Vitr Cell Dev Biol. 175. Joraku A, Sullivan CA, Yoo J, Atala A. In-vitro recon-
1998;34:58–67. doi:10.1007/s11626-998-0054-5. stitution of three-dimensional human salivary gland
161. Baker OJ, Schulz DJ, Camden JM, et al. Rat parotid tissue structures. Differentiation. 2007;75:318–24.
gland cell differentiation in three-dimensional cul- doi:10.1111/j.1432-0436.2006.00138.x.
ture. Tissue Eng Part C Methods. 2010;16:1135–44. 176. Burford-Mason AP, Dardick I, Mackay A. Collagen
doi:10.1089/ten.tec.2009.0438. gel cultures of normal salivary gland: conditions for
162. Shirasuna K, Sato M, Miyazaki T. A neoplastic epi- continued proliferation and maintenance of major cell
thelial duct cell line established from an irradiated phenotypes in vitro. Laryngoscope. 1994;104:335–
human salivary gland. Cancer. 1981;48:745–52. 40. doi:10.1288/00005537-199403000-00016.
doi:10.1002/1097-0142(19810801)48:3<745::AID- 177. Weisel JW. Fibrinogen and fibrin. Adv
CNCR2820480314>3.0.CO;2-7. Protein Chem. 2005;70:247–99. doi:10.1016/
163. Royce LS, Kibbey MC, Mertz P, et al. Human neo- S0065-3233(05)70008-5.
plastic submandibular intercalated duct cells express 178. Cantara SI, Soscia DA, Sequeira SJ, et al. Selective
an acinar phenotype when cultured on a basement functionalization of nanofiber scaffolds to regulate
membrane matrix. Differentiation. 1993;52:247–55. salivary gland epithelial cell proliferation and polar-
doi:10.1111/j.1432-0436.1993.tb00637.x. ity. Biomaterials. 2012;33:8372–82. doi:10.1016/j.
164. Hoffman MP, Kibbey MC, Letterio JJ, Kleinman biomaterials.2012.08.021.
HK. Role of laminin-1 and TGF-beta 3 in acinar 179. Yin N, Santos TMA, Auer GK, et al. Bacterial cel-
differentiation of a human submandibular gland cell lulose as a substrate for microbial cell culture.
line (HSG). J Cell Sci. 1996;109(Pt 8):2013–21. Appl Environ Microbiol. 2014;80(6):1926–32.
165. Jung DW, Hecht D, Ho SW, et al. PKC and ERK1/2 doi:10.1128/AEM.03452-13.
regulate amylase promoter activity during dif- 180. Modulevsky DJ, Lefebvre C, Haase K, et al. Apple
ferentiation of a salivary gland cell line. J Cell derived cellulose scaffolds for 3D mammalian cell
Physiol. 2000;185:215–25. doi:10.1002/1097-4652 culture. PLoS One. 2014;9:e97835. doi:10.1371/
(200011)185:2<215::AID-JCP6>3.0.CO;2-L. journal.pone.0097835.
166. Szlávik V, Vág J, Markó K, et al. Matrigel-induced 181. Bhattacharya M, Malinen MM, Lauren P, et al.
acinar differentiation is followed by apoptosis in Nanofibrillar cellulose hydrogel promotes three-
HSG cells. J Cell Biochem. 2008;103:284–95. dimensional liver cell culture. J Control Release.
doi:10.1002/jcb.21404. 2012;164:291–8. doi:http://dx.doi.org/10.1016/j.
167. Capes-Davis A, Theodosopoulos G, Atkin I, et al. jconrel.2012.06.039
Check your cultures! A list of cross-contaminated or 182. Hardingham T. Chemistry and biology of hyaluro-
misidentified cell lines. Int J Cancer. 2010;127:1–8. nan. Chem Biol Hyaluronan. 2004. doi:10.1016/
doi:10.1002/ijc.25242. B978-008044382-9/50032-7.
168. Maria OM, Zeitouni A, Gologan O, Tran 183. Xu X, Jha AK, Harrington DA, et al. Hyaluronic
SD. Matrigel improves functional properties of pri- acid-based hydrogels: from a natural polysaccharide
mary human salivary gland cells. Tissue Eng Part A. to complex networks. Soft Matter. 2012;8:3280–94.
2011;17:1229–38. doi:10.1039/C2SM06463D.
184. Turley EA, Noble PW, Bourguignon LYW. Signaling 190. Hoffman MP, Nomizu M, Roque E, et al. Laminin-1
properties of hyaluronan receptors. J Biol Chem. and laminin-2 G-domain synthetic peptides bind
2002;277:4589–92. doi:10.1074/jbc.R100038200. syndecan-1 and are involved in acinar forma-
185. Peach RJ, Hollenbaugh D, Stamenkovic I, Aruffo tion of a human submandibular gland cell line.
A. Identification of hyaluronic acid binding sites J Biol Chem. 1998;273:28633–41. doi:10.1074/
in the extracellular domain of CD44. J Cell Biol. jbc.273.44.28633.
1993;122:257–64. 191. Farach-Carson MC, Brown AJ, Lynam M, et al.
186. Burdick JA, Prestwich GD. Hyaluronic acid hydro- A novel peptide sequence in perlecan domain IV
gels for biomedical applications. Adv Mater. supports cell adhesion, spreading and FAK activa-
2011;23:41–56. doi:10.1002/adma.201003963. tion. Matrix Biol. 2008;27:150–60. doi:10.1016/j.
187. Lee JY, Spicer AP. Hyaluronan: a multifunctional, matbio.2007.09.007.
megaDalton, stealth molecule. Curr Opin Cell Biol. 192. Nakanishi Y, Sugiura F, Kishi JI, Hayakawa
2000;12:581–6. T. Collagenase inhibitor stimulates cleft for-
188. Shubin AD, Felong TJ, Graunke D, et al. mation during early morphogenesis of mouse
Development of poly(ethylene glycol) hydrogels salivary gland. Dev Biol. 1986;113:201–6.
for salivary gland tissue engineering applications. doi:10.1016/0012-1606(86)90122-3.
Tissue Eng Part A. 2015;21:1733–51. doi:10.1089/ 193. Fong ELS, Wan X, Yang J, et al. A 3D in vitro
ten.tea.2014.0674. model of patient-derived prostate cancer xenograft
189. Hosokawa Y, Takahashi Y, Kadoya Y, et al. Significant for controlled interrogation of in vivo tumor-stro-
role of laminin-1 in branching morphogenesis of mal interactions. Biomaterials. 2016;77:164–72.
mouse salivary epithelium cultured in basement doi:10.1016/j.biomaterials.2015.10.059.
membrane matrix. Dev Growth Differ. 1999;41:207–
16. doi:10.1046/j.1440-169x.1999.00419.x.
3D Printing Technology
in Craniofacial Surgery 9
and Salivary Gland Regeneration
Jong Woo Choi, Namkug Kim,
and Chang Mo Hwang
Abstract
Patient-specific three-dimensional (3D)-printed phantoms and surgical
guides are being utilized more often nowadays to assist diagnosis and
treatment planning for surgery, which are tailored to individual’s unique
needs. 3D printing surgical guides made of temporary materials can be
fabricated to fit the surface of the hard or soft tissue organs by 3D model-
ing of the surgical interface. To date, the value of 3D printing for surgical
planning as a guidance tool has been proven in various hard tissue surgical
applications, such as craniofacial and maxillofacial surgery, spine surgery,
cardiovascular surgery, neurosurgery, pelvic surgery, and visceral surgery.
Craniofacial plastic surgery is one of the medical fields that pioneered the
use of the 3D printing concept. Rapid prototyping technology was intro-
duced to medicine in the 1990s via CAD-CAM (computer-aided design,
computer-aided manufacturing). The medical models or bio-models based
on the 3D printing technique represent 1:1 scale reproductions of the
human anatomical region of interest that can be obtained via 3D medical
imaging. The procedure for the fabrication of medical models comprises
multiple steps: (1) acquisition of high-quality volumetric 3D image data of
the anatomical structure to be modeled, (2) 3D image processing to extract
the region of interest from the surrounding tissues, (3) mathematical sur-
face modeling of the anatomic surfaces, (4) formatting of data for rapid
prototyping, (5) model building, and (6) quality assurance of the model
and its dimensional accuracy. Furthermore, tissue engineers also experi-
ence the advent of a new 3D printing era. The tissue engineering triad
comprises cells, scaffolds, and growth factors. Recently, 3D technology
has become sufficiently evolved to enable printing of living cells. Although
J.W. Choi, MD, PhD, MMM (*) • N. Kim, PhD

C.M. Hwang, PhD
Department of Plastic and Reconstructive Surgery,
Department of Bioengineering, Ulsan University,
College of Medicine, Seoul, South Korea
e-mail: [email protected]; namkugkim@gmail.
com; [email protected]

174 J.W. Choi et al.
many challenging issues remain to be resolved for such complex struc-

tures, heart, kidney, and skin regenerations are being investigated using 3D
bioprinting technology. A potential candidate for a clinical success resides
in the regeneration of the major salivary glands, which consist of various
cells encapsulated by a connective tissue membrane.
9.1 Introduction surgery of complex renal cell carcinomas in

advance of actual surgery using 3D printed tactile
Three-dimensional (3D) printing is a rapidly prototype models that include the vessels and
developing technology that is applied worldwide parenchyma of the kidney. Neurosurgeons utilize
in various fields, such as business, fashion, similar approaches for neurosurgery of brain can-
mechanical engineering, and medicine [1, 2]. 3D cer. These types of efforts allow surgeons in vari-
printing technology has already been used in the ous specialties to perform advanced analyses of
mock formation of various products including the patient’s specific status. In addition, tactile
cellular phones [2], and the extent of its applica- models with a real intraoperative 1:1 scale refer-
tions has greatly expanded in medicine as this ence can be very useful for preoperative consul-
technology has evolved in recent years. A major tations with patients [3–5].
advantage is that it can generate a unique product Craniofacial plastic surgery is one of the medi-
in a short period of time, which is suitable for cal fields that pioneered the use of the 3D printing
individualized medicine where each patient concept. Rapid prototype (RP) technology was
requires a specific treatment, tailored to a thera- introduced to medicine in the 1990s via computer-
peutic approach. As opposed to the products gen- aided design and computer-aided manufacturing
erated by 3D printing, most industrial products (CAD-CAM). The medical models or bio-models
are mass produced, and every unit has the same based on the 3D printing technique represent 1:1
dimensions. As one would expect, in the field of scale reproductions of the human anatomical
medicine, patients are all different in terms of any region of interest that can be obtained via 3D med-
shapes and sizes that require surgical attention. ical imaging [5]. The procedure for the fabrication
The 3D printing technique supports the contem- of medical models comprises multiple steps: (1)
porary aim of implementing personalized medi- acquisition of high-quality volumetric 3D image
cine by providing a patient-specific product in a data of the anatomical structure to be modeled, (2)
short period of time at reasonable prices [3]. 3D image processing to extract the region of inter-
Therefore, the clinical applications of the 3D est from the surrounding tissues, (3) mathematical
printing technology are expanding more rapidly surface modeling of the anatomic surfaces, (4) for-
in recent years. The affordability and conve- matting of data for rapid prototyping, (5) model
nience of this technology have spurred its adop- building, and (6) quality assurance of the model
tion in a variety of medical fields. This and its dimensional accuracy [3, 6].
revolutionary technique may ultimately allow the For instance, because patients requiring cranio-
printing of tissue and organ structures to replace facial surgery tend to have very specific malforma-
damaged or missing body parts. Although out- tions or deformities, mostly in the bone, a 3D
comes and efficacy of 3D printing require more printing prototype model can greatly assist with
scientific research, it is clear that 3D printing preoperative evaluation and intraoperative proce-
technology is unique and has invaluable innova- dures. Medical modeling in craniofacial surgery
tions in medicine. For example, pediatric cardiac based on 3D printing has mainly been developed
surgeons use 3D printing-based tactile models over the last 15 years. It can incorporate (1) aiding
for analyzing and visualizing complex congenital in the production of surgical implants, (2)
heart diseases. Urologic surgeons simulate improving surgical planning, (3) acting as an

9 3D Printing Technology in Craniofacial Surgery and Salivary Gland Regeneration 175
o rientation aid during surgery, (4) enhancing diag- the techniques includes stereolithography (SL),
nostic quality, (5) assisting preoperative simulation, selective laser sintering (SLS), 3D printing
(6) obtaining a patient’s consent prior to surgery, (3D printer-based SLS: 3DP), fused deposition
and (7) preparing a template for resection for sur- modeling (FDM), direct metal laser sinter-
geons as well as providing an educational tool for ing (DMLS), laminated object manufacturing
medical students and residents [3, 5, 7]. (LOM), and electron beam melting (EBM). The
Meanwhile, tissue engineers also experience materials used for the 3D printing technology
the advent of a new 3D printing era. The tissue include thermoplastic, metal powder, ceramic
engineering triad comprises cells, scaffolds, and powder, eutectic metals, alloy metal, photo-
growth factors. Recently, 3D technology has polymer, paper, foil, plastic film, and titanium
become sufficiently evolved to enable printing of alloys. The 3D technology can be classified by
living cells. Although many obstacles need to be the aimed deposition process. PolyJet model-
overcome, 3D bioprinting provides bioengineers ing and 3D plotting technology are based on
with a new modality such as 3D cell culture on drop-on-drop deposition. 3D printing is based
scaffolds that might be superior to conventional on drop-on-powder deposition. Fused deposition
cell culture systems. Bioprinting is an emerging modeling is based on continuous deposition. The
technology that is expected to eventually regener- most frequently used representative methods are
ate biological tissues and even solid organs. As a reviewed and summarized in Table 9.1 [3].
combination of techniques, a nonliving scaffold
could be constructed using 3D technology, while
bioprinting simultaneously adds a living tissue [1, 9.2.1 L
iquid-Based 3D Printing
8–12]. More specifically, tissue-compatible scaf- Technology
folds are generated with bioprinting, and living
cells are incorporated into them, along with vari- 9.2.1.1 Stereolithography (SL or SLA)
ous growth factors, depending on the application. Stereolithography (SL) has been the most widely
Heart, kidney, and skin regenerations are being used 3D printing technique for craniofacial sur-
investigated using 3D bioprinting technology, gery since it was first applied for grafting a skull
although many challenging issues remain to be defect in 1994 [13]. The SL RP system consists
resolved for such complex structures. The regen- of a bath of photosensitive resin, a model-building
eration of the major salivary glands, which consist platform, and an ultraviolet (UV) laser for curing
of various cells encapsulated by connective tissue the resin. A mirror is used to guide the laser focus
membrane, certainly requires further investigation onto the surface of the resin; the resin becomes
and attention for a clinical success of 3D bioprint- cured when exposed to the UV radiation. The
ing. In this book chapter, the current status of 3D mirror is computer controlled and is guided to
printing technology and its clinical applications in cure the resin on a slice-by-slice basis. These
craniofacial surgery are reviewed. A potential slice data are fed into the RP machine that directs
application of 3D bioprinting for salivary gland the exposure path of the UV laser onto the sur-
regeneration is discussed at the end of the chapter. face of the resin. The layers are cured sequen-
tially and bind together to form a solid object,
beginning from the bottom of the model and
9.2 eview of Current 3D
R building upward. Each new layer of resin is
Printing in Craniofacial wiped across the surface of the previous layer
Surgery (Reproduced using a wiper blade before being exposed and
from Ref. [28]) cured. The model is then removed from the bath
and cured for an additional period of time in a
3D printing technology can be categorized UV cabinet. [14].
by the techniques, the materials, or the aimed Generally, SL is considered to provide the
deposition process. The classification based on greatest accuracy and best surface finish of any
Table 9.1 A comparison of current 3D printing technologies

3D printing technology Materials Aimed deposition process
Liquid base SL (Stereolithography) Photopolymer
Polyjet or Multijet Printing ABS, Acryl Drop-on-drop
deposition
Powder base SLS (Selective Laser Sintering) Thermoplastics
Metal powder
3DP (3D printing) Plastic powder Drop-on-powder
deposition
DMLS (Direct Metal Laser Sintering) Alloy metal
Ceramic powder
EBM (Electron beam melting) Titanium alloys
Solid base FDM (Fused Deposition Modeling) Thermoplastics Continuous deposition
Eutectic metals
ABS
LOM (Laminated object manufacturing) Paper
Foil
Plastic film
Reprinted from ref [18]. ABS, acrylonitrile butadiene styrene
RP technology. The model material is robust, 9.2.1.2 PolyJet Modeling

slightly brittle, and relatively light [15]. SL PolyJet modeling is performed by jetting state-of-
accuracy is 1.2 mm (range, 0–4.8 mm) for skull the-art photopolymer materials in ultrathin layers
base measures, 1.6 mm (range, 0–5.8 mm) for (16 μm) onto a build tray layer by layer until the
midface measures, 1.9 mm (range, 0–7.9 mm) model is completed. Each photopolymer layer is
for maxilla measures, and 1.5 mm (range, cured by UV light immediately after it is jetted,
0–5.7 mm) for orbital measures. The mean dif- producing fully cured models that can be handled
ferences in defect dimensions are 1.9 mm (range, and used immediately without post-curing. The
0.1–5.7 mm) for unilateral maxillectomy, gel-like support material used, which is specially
0.8 mm (range, 0.2–1.5 mm) for bilateral maxil- designed to support complicated geometries, is
lectomy, and 2.5 mm (range, 0.2–7.0 mm) for easily removed by hand and water jetting [14]. At
orbitomaxillectomy defects [16]. Midface SL present, this technique is too time-consuming and
models may be more prone to error than those of expensive to be used in craniofacial surgery clini-
other craniofacial regions because of the pres- cal applications. Ibrahim et al. reported a dimen-
ence of thin walls and small projections. Choi sional error of 2.14 % in reproducing a dry
et al. [29] found that the absolute mean deviation mandible when using this technique [18].
between an original dry skull and an SL RP
model over 16 linear measurements was 0.62 ±
0.5 mm (0.56 ± 0.39 %) [15, 17]. The accuracy 9.2.2 P
owder-Based 3D Printing
of computed tomography (CT) and SL models Technology
was compared. The accuracy for SL models
expressed as the arithmetic mean of the relative 9.2.2.1 Selective Laser Sintering (SLS)
deviations ranged from 0.8 to 5.4 %, with an The selective laser sintering (SLS) technique uses
overall mean deviation of 2.2 %. The mean devi- a CO2 laser beam to selectively fabricate models in
ations of the investigated anatomical structures consecutive layers. First, the laser beam scans over
ranged from 0.8 to 3.2 mm. An overall mean a thin layer of powder previously deposited on the
deviation (comprising all structures) of 2.5 mm build tray and leveled with a roller. The laser heats
was found. the powder particles, fusing them to form a solid
layer, and then moves along the x- and y-axes to associated with low waste, and accurate (± 0.1 mm
design the structures according to the CAD data. in the Z plane, ± 0.2 mm in the X and Y planes),
After the first layer fuses, the build tray moves and they can make hard, soft, or flexible models.
downward, and a new layer of powder is deposited These printers can also be used to identify differ-
and sintered, and the process is repeated until the ent types of body tissue depending on the pre-
object is completed. The prototype surface is fin- defined threshold setting selected. Silva et al.
ished by sandblasting [14]. The SLS prototype is reported a mean dimensional error of 2.67 % in
opaque, and its surface is abrasive and porous. prototypes produced using 3D printing technolo-
Prototype fabrication time is 15 h. The accuracy of gies in comparison with a dry human skull [19].
the SLS model is relatively high, with maximum
standard errors of 0.1–0.6 mm. This accuracy
depends on the thickness of the CT scans used, 9.2.3 S
olid-Based 3D Printing
which should be as thin as possible (1–2 mm is a Technology
good compromise for a skull study). Because of
the high cost of the materials, several parts are fab- 9.2.3.1 Fused Deposition Modeling
ricated simultaneously. The long fabrication time Fused deposition modeling (FDM) uses a similar
for the SLS technique (16 h) is close to the time principle to SL in that it builds models on a layer-
required for fabrication with the SL system [19]. by-layer basis. The main difference is that the lay-
ers are deposited as a thermoplastic that is extruded
9.2.2.2 3D Printer-Based SLS from a fine nozzle. A commonly used material for
(3D Printing) this procedure is acrylonitrile butadiene styrene
The 3D printing system uses a print head to selec- (ABS). The 3D model is constructed by extruding
tively disperse a binder onto powder layers. This the heated thermoplastic material onto a foam sur-
technology has a lower cost than similar tech- face along a path indicated by the model data. Once
niques. First, a thin layer of powder is spread over a layer has been deposited, the nozzle is raised
a tray using a roller similar to that used in the SLS between 0.278 and 0.356 mm, and the next layer is
system. The print head scans the powder tray and deposited on top of the previous layer. This process
delivers a continuous jet of a solution that binds is repeated until the model is completed [14]. As
the powder particles as it touches them. No sup- with SL, support structures are required for FDM
port structures are required while the prototype is models because time is needed for the thermoplas-
being fabricated because the surrounding powder tic to harden and the layers to bond together [20].
supports the unconnected parts. When the process
is complete, the surrounding powder is aspirated.
In the finishing process, the prototype surfaces are 9.3 Patient-Specific Modeling
infiltrated with a cyanoacrylate-based material to and Its Clinical Application
harden the structure [19]. The printing technique Using 3D Printing
enables the formation of complex geometrical Technology
structures, such as hanging partitions inside the
cavities, without artificial support structures [14]. 9.3.1 Patient-Specific Modeling
After the CT scan, the rendering of the DICOM from Medical Images
data and transformation into STL data files take a and Computer-Aided Design
maximum of 30 min, and the printing and infiltra-
tion process takes approximately 4–6 h. Simpler As depicted in Fig. 9.1, after patient scanning with
models can be purchased for as little as $300–$400 CT and/or MRI, the DICOM data can be trans-
[19]. The 3D printers used in this process are ferred and processed into STL data files or other
relatively inexpensive ($2500–$3000), have fast 3D file formats by using segmentation, surface
build times (4 h for a full skull), and are easy to extraction, and 3D model post-processing. Less
maintain. Additionally, 3D printers are cost-effective, than a 1-mm CT slice thickness and voxel with iso-
Process of 3D printer
Materials 3D printer
Conversion to 3D files
In the finishing process, the

3D model design prototype surfaces are infiltrated
on the computer with a cyanoacrylate-based
material to be harden the structure
a print head to selectively

disperse a binder onto The print head scans the
powder layers Platform moved according to powder tray and delivers a
the movement of nozzle continuous jet of a solution that
binds the powder particles
Fig. 9.1 The overall process of 3D printing in craniofa- printer. Rapid prototyping (RP) uses layer-by-layer ste-
cial surgery. After the patient is scanned via CT, DICOM reolithographic accumulation. The RP model is then fab-
files should be exported. Less than a 1-mm CT slice ricated on plaster via jetting of a material that consists of
thickness is recommended. DICOM data are imported plaster (<90 %), vinyl polymer (<20 %), and carbohy-
and converted to stereolithography (STL) files. Rendering drate (<10 %). The printing and infiltration process takes
of CT scan DICOM data into STL data files takes about about 4–6 h. Finally, unsintered sections are removed
30 min. Converted 3D files are uploaded into the 3D (Reprinted from Ref. [18])
cubic spacing are recommended. The time required surgical planning. Based on this planning, surgical
mainly rests on the clinical application. In particu- guides are designed by computer-aided design
lar, segmentation is a critical procedure for improv- (CAD) software. After the generation of a 3D
ing the overall accuracy and needs considerable model, the most suitable 3D printer for their appli-
time. No satisfactory fully automated medical cations is selected among various kinds of 3DP
image segmentation algorithms have been estab- techniques. The 3D model file is uploaded into the
lished. Therefore, manual or semiautomated seg- 3D printer. The 3D printer uses layer-by-layer STL
mentation algorithms have generally been used, accumulation to fabricate the 3D physical model.
which have enhanced the importance of operator
experience. After segmentation, a surface model
should be produced by a marching cube [21, 22] or 9.3.2 A
pplications for Personalized
other 3D contour extraction algorithms [23]. For Treatment
medical visualization, these kinds of shaded sur-
face display techniques are well established. 9.3.2.1 Surgical Planning and Guidance
However, this 3D model by itself is not good Tools
enough for 3DP, due to, for example, too many Patient-specific 3D printed phantoms and surgi-
mesh units and incomplete topological soundness. cal guides are being used more often to aid diag-
Therefore, topological correction [24], decimation nosis and treatment planning for surgery, which
[25], Laplacian smoothing [26], and local smooth- allow individual customization. 3D printing sur-
ing [27] are required to make a 3D model for 3DP. In gical guides made of temporary materials can be
addition, virtual simulation, including determination fabricated to fit the surface of the hard or soft tis-
of the entry point and direction of the screw and sue organs by 3D modeling of the surgical inter-
surgical line, is accomplished for patient-specific face. To date, the value of 3D printing for surgical
planning as a guidance tool has been proven in rubberlike and transparent material (Fig. 9.1). The
various hard tissue surgical applications, such as 3D printed phantom provided invaluable informa-
craniofacial and maxillofacial surgery [28–33], tion on the LV geometry. It is known that the softest
spine surgery [34], cardiovascular surgery [35, 3D printable materials cannot be directly used as
36], neurosurgery [37, 38], pelvic surgery [39, surgical simulators because they are still too hard
40], and visceral surgery [41]. for scalpel incision and suturing. Therefore, addi-
Recent advances in 3D printable materials have tional post-processing using gelatin or silicone
increased the level of realism of the 3D phantoms molding techniques or a novel 3D printing system
used for surgical planning. Improved diversity due that can directly jet a variety of silicone materials
to better transparency, color, and softness facilitates needs to be developed.
better understanding of complex 3D anatomical
structures and guidance functions for soft tissues 9.3.2.2 Implantable Devices
[42]. Yang et al. [43] used a full-colored and flexi- 3D printing techniques are also used in implant
ble 3D printed phantom as a preplanning simulator design to make patient-specific prosthetics, out-
for extended septal myectomy. From the cardiac side the standard range of ready-made commer-
CT data, a myocardial 3D model was made by in- cial implants (Fig. 9.2). In addition, this approach
house software (A-view Cardiac; Asan Medical has improved surgical performance by enabling
Center, Seoul, Republic of Korea). Using a 3D the creation of patient-specific anatomy-based
printer (Connex3 Objet500; Stratasys Corporation, implants. For hard tissue structures, metal
Rehovot, Israel), the left ventricular (LV) myocar- implants have, in particular, been successfully
dium, papillary muscle, and intraventricular m uscle used in various applications [44, 45], which were
band (including the accessory papillary muscle) mostly FDA cleared, such as mandible [33] and
were fabricated with differently colored materials, dental [46] restoration and hip [47], femoral [48],
whose flexibility could be controlled by adding a and hemi-knee joint reconstruction [44, 45]. In
Fig. 9.2 Large cranial defect reconstructed with 3D bone defect (Modified from Ref. [18]). Bottom panel:
printed titanium implant. Top panels: The contralateral Computer-simulated skull defect images before (left) and
normal cranium was mirrored and the 3D printed titanium after (right) titanium implant was inserted
implant was inserted for the correction of the calvarial
addition, the biocompatible ceramic hydroxyapa- regurgitation using a braided stent in an animal
tite [49] and the biodegradable polymer polycap- study. From the cardiac CT data, the 3D recon-
rolactone [50] have been used in 3D printing-based structed model of the right-sided cardiac cavities
applications to substitute hard tissues with cus- of a pig was obtained (OsiriX® Imaging Software;
tomized implants. Pixmeo, Switzerland). A solid Alumide® mold
Beyond the hard tissue applications, custom- was manufactured using a 3D printing system, and
ized implants created using 3D printing have then a personalized compressible nitinol stent was
recently been used in the interventional field. subsequently produced and fitted onto the 3D
Amerini et al. [51] revealed the feasibility of a per- printing mold (Fig. 9.3). This customized stent
sonalized interventional treatment for tricuspid was almost completely fitted onto the right atrium,
a b
c d e
f g
and an additional tubular stent component contain- In this book chapter, only clinical applica-
ing a tissue valve prosthesis was established. In the tions with previously developed 3D printing
feasibility study performed in animals, they found technologies were discussed. However, other

that the 3D printing-based stent could stabilize the approaches for personalized implants have been
biological valve prostheses by force transmission proposed, including bioprinting of tissues and
from the annulus to the atrial wall and the adjacent organs [52–54] and the organ-on-a-chip tech-
vena cava. nique [55, 56].
a b c d h
e f g
Fig. 9.4 In silico tridimensional reconstruction of the developed stent in an introducer. (f, g) Two different types
right-sided cardiac cavities of a female pig. (a) CT-based of the prototype equipped with a self-expanding biopros-
primary 3D reconstruction. (b) 3D reconstructed model of thetic valve. (h, j) Study results [51] showing implantation
main structural parts. (c) 3D printed phantom mold of the of the developed stent. Postmortem autopsy (h, i) and CT
main structural parts with alumide material. (d) A personal- fluoroscopy (j) both revealed accurate positioning of the
ized stent with nitinol material. (e) An equipped state of the valve prostheses (Reproduced from Ref. [51])
Fig. 9.3 A cardiac three-chamber CT image and 3D the myocardium showing the geometric relationship
printing of the heart. (a) CT imaging demonstrating a among the hypertrophied septum (asterisks), papillary
hypertrophied interventricular septum (asterisks), poste- muscle (A anterior, P posterior), and intraventricular
rior papillary muscle (P), and intraventricular muscle muscle band (asterisks). (g) Intraoperative photography
band or accessory papillary muscles (arrowhead). (b) A via the apical approach shows the limited visual field of
bull’s-eye map generated by using the end-diastolic the LV cavity. The base of the anterior papillary muscle is
phase of the CT imaging shows the extent of the hyper- exposed after excision of the muscle band (not shown)
trophied myocardium (red area, >15 mm in thickness). near the anterior papillary muscle. LV left ventricle
(c) 3D reconstructed model. (d–f) 3D printed phantom of (Reprinted from Ref. [43])
a1 air a2
adapter connected spring
piston
to air supply
Pneumatic micro-extrusion
air valve-based air

syringe nozzle
piston barrel
air
(optional)
bioink
tube bioink
extruded Micro-needle bioink extruded
filament filament Micro-needle
bioink
container
scaffold scaffold
b1 b2
stepper
motor
Mechanical micro-extrusion
bioink connector
stepper
feeder
motor
connector lead- rotating

piston screw screw gear
bioink bioink
extruded micro-needle extruded

micro-needle
filament filament
scaffold scaffold
actuator
c pressure
bioink
adapter connected
to actuator pressure Ferro-magnetic
Solenoid micro-extrusion
plunger
actuating
syringe coil
barrel
nozzle N N
Ferro-magnetic S S
ring magnet closed
(no flow) open
extruded
(flow)
filament
scaffold
Fig. 9.5 Extrusion-based bioprinting systems: (a) pneu- ing piston (B1) and screw-driven (B2) and (c) solenoid
matic micro-extrusion including valve-free (A1) and valve micro-extrusion (Reprinted from Ref. [71])
based (A2) and (b) motor-driven micro-extrusion includ-
Fig. 9.6 Extrusion bioprinting with tissue originated bioinks. Biodegradable synthetic polymer scaffold is coextruded
side by side to hydrogel bioink to maintain 3D architecture of printed objects (Reprinted from Ref. [70])
Fig. 9.7 Salivary gland regeneration with microsphere 3D bioprinting (Reproduced from Refs. [82, 88])
9.4 D Bioprinting and Salivary

3 tissue growth and organization. Similarly, key
Gland Regeneration elements of bioprinting consist of cells, bio-
printer, bioink, and the bioreactor system. The
9.4.1 3D Bioprinting Considerations choice of cells for tissue reconstruction depends
on the types of cells in the target tissues and
An increasing number of publications for 3D organs. For example, vascular endothelial cells
bioprinting report significant progress and suc- and smooth muscle cells would be appropriate
cesses in vitro and in vivo. The three major for blood vessel printing and fibroblasts for
components of tissue engineering include cells, connective tissues. Stem cells are frequently
scaffolds, and biological factors that facilitate considered as a potential source of cells as well.
The selection of cell types has been widely The first consideration of materials as a bio-
investigated in the tissue engineering literature ink is printability. An appropriate shape holding
[57–61]. Therefore, it will not be explained fur- mechanism is necessary to maintain 3D configu-
ther here. Bioreactors can be employed for the ration of printed objects. The transition of bio-
maturation of printed tissue constructs into inks from liquid to solid (semisolid) should be
functional tissue units and organs [62–64]. shorter than significant shape change. The stabil-
Generally, bioprinted three-dimensional tissue ity of bioprinted constructs mainly depends on
constructs are formed layer by layer by printing the viscosity of the bioinks after printing.
bioinks that contain living cells. Current 3D Liquid phase bioinks out of the printer nozzle
bioprinting research mainly focuses on the are subject to surface tension and gravitational
printing device and material compositions. The force. These external forces affect shape change
two most widely employed bioprinting of printed bioinks until possessing high enough
mechanisms would be the inkjet printing and viscosity.
extrusion printing [65]. From the viscosity point of view, materials
with short cross-linking time can be a first con-
9.4.1.1 Inkjet Printing sideration as bioink candidates. Bioink materials
Inkjet printing has a mechanism which is very modified to have a short gelling time are widely
similar to conventional office inkjet printers, with used in 3D bioprinting. Hydrogels with short
serial deposition of cell-containing bioink drop- cross-linking time are widely employed as bio-
lets. Piezoelectric actuators, heat-assisted bubble inks because these materials have dimensional
jet actuators, and pneumatic pressurization with stability in a relatively short time after printing
solenoid valves are examples of inkjet printing [65, 66, 69, 74, 75].
techniques used to generate droplets [66–68]. Another approach for high viscosity is
The electronic control system of inkjet bioprint- employing thixotropic materials to improve the
ers enables relatively precise cell positioning, stability of printed 3D constructs during cross-
which can be used for drug testing or small-scale linking. Thixotropic materials are usually semi-
tissue unit fabrication. solid and have a shear thinning property
(thixotropic means “shear thinning”). During
9.4.1.2 Extrusion Printing bioink printing through the printer nozzle, shear
Extrusion-based printing is the most widely used forces induce a lowering of the viscosity, and bio-
bioprinting system [57, 69, 70]. Cell-containing inks have low flow resistance, with minimal
material (bioink) is extruded from a reservoir to harmful effect to the suspended cells. After exit-
the printing bed through printer nozzles as shown ing the nozzle, the thixotropic materials regain
in Fig. 9.5. The driving mechanism of extrusion their high viscosity, and shape changes are mini-
can be pneumatic pressure or motor-driven mized [76]. This prevents the collapse of printed
syringe plunger movement [70, 71]. 3D constructs.
One additional advantage of thixotropic bio-
9.4.1.3 Bioinks ink is the absence of cell sedimentation in the
Cell behaviors including adhesion, migration, reservoir during the printing process. As the spe-
proliferation, differentiation, and tissue f ormation cific gravity of a cell is slightly higher than
are influenced by the extracellular microenviron- water, suspended cells tend to localize on the
ments, both in vivo and in vitro. After printing, bottom of a reservoir. This effect is significant
cells are encapsulated in the bioink, and cell when cells are suspended in a low viscosity liq-
behavior is mainly affected by the biophysico- uid. In thixotropic bioinks, suspended cells may
chemical properties of the bioink, such as stiff- show no or negligible displacement. As printing
ness, molecular structure, cytokines or growth time is proportionally increased with the volume
factors, degradability, and permeability [71–73]. of an object, inhomogeneous cell distribution
would be a significant defect in human-sized Photo-cross-linking polymers are also being

organs made with non-thixotropic aqueous bio- widely investigated as bioinks. Hydrogel
inks. While extruding with thixotropic bioinks, precursors, including methacrylated gelatin

care must be taken to keep the proper shear stress (GelMA), star poly(ethylene glycol-co-lactide)-
range to avoid lowering suspended cell viability. acrylate (SPELA), poly(ethylene glycol)
In side-by-side polymer printing the nonporous dimethacrylate (PEGDMA), and polyethylene
structure of each layer can be employed as a sup- glycol) diacrylate (PEGDA), can be cross-
porting structure to improve dimensional stabil- linked using UV light [73, 79, 80]. A brief sum-
ity of constructs during 3D bioprinting as shown mary of bioinks currently used are listed in
in Fig. 9.6 [57, 70, 77, 78]. Table 9.2. Bioink materials that support cell
Hydrogels can provide cells with a minimum viability and proliferation and have short cross-
damage environment during the bioprinting pro- linking time are still needed to be developed for
cess. Hydrogels are widely used as a bioink mate- employing 3D bioprinting process for tissue
rial with a cell compatible pH and appropriate regeneration. Bioprinters should have appropri-
osmolarity. Examples of biomaterials with natural ate design compatible to bioink’s cross-linking
origins are alginate, fibrin, gelatin, hyaluronic mechanism. Dual or multiple mixing nozzle
acid, and collagen, and synthetic biomaterials are configuration is required for mixing precursor
polyethylene glycol and Pluronic® F-127 [70, solutions. Cooling or heating temperature con-
72–74, 78]. Mixtures of these materials are also trol should be considered for temperature-
used with optimized printability, low cell damage, induced cross-linking materials [80, 81].
and higher 3D printed construct stability.
The cross-linking mechanism depends on
hydrogels’ intrinsic characteristics. Alginate 9.4.2 S
alivary Gland Regeneration
has ionically cross-linking, and simple contact by 3D Bioprinting
of alginate solution with divalent cationic solu-
tions, such as calcium, barium, and strontium, The ultimate goal of 3D bioprinting is to provide
can generate cross-linked hydrogel. Due to its vascularized functional living organs, which can
low cost and simple cross-linking process, algi- be applied to the replacement of missing or dis-
nate is often employed as an initial test material abled tissues and organs. Observations and lessons
for various bioprinters. Collagen and decellular- from developmental biology can provide funda-
ized extracellular matrix (dECM) have pH- and mental and practical ideas for tissue engineering
temperature- dependent cross-linking manner approaches. Specific tissues or organs at different
[57, 70, 74, 77]. Under the physiologic pH condi- stages of development will have varying structural
tion and temperature (pH 7.4 and 37 °C, respec- requirements. The essential morphogenetic steps
tively), these materials cross-link to form stable and events of organogenesis during developmental
hydrogel matrix. Further, with high cytocompat- stage can provide insights for salivary gland regen-
ibility, cells in collagen and dECM show high eration through 3D bioprinting [82].
tissue formation superior to alginate. However, Salivary glands consist of saliva-secreting
relatively long cross-linking time (~30 min under acinar cells and various other types of cells.
37 °C) hampers widespread use of these materials Tissue engineering of salivary glands was tried
as bioink [57, 75]. Fibrin has enzyme-activated with several different approaches with hydrogel
cross-linking mechanism. By mixing fibrinogen material for tissue regeneration [83–85]. Tissue
solution with thrombin solution, a stable fibrin spheroids, which have been used as an in vitro
hydrogel forms. Fibrinogen is a blood coagu- 3D model system in biomedical and tumor
lation protein and has high cytocompatibility research for several decades, may be a useful
but still has relatively longer cross-linking time candidate in salivary gland regeneration with 3D
(0.5~10 min) than alginate (0.5~5 s). bioprinting technology (Fig. 9.7) [77, 82].
Table 9.2 Hydrogels used in extrusion-based bioprinting
186
Cross-linking
mechanism in
extrusion Solidification Extrusion
Hydrogel type Bioink bioprinting reversibility bioprinting system Advantages Disadvantages
Alginate Aggregates, proteins, Ionic − Pneumatic Biocompatibility, good Low cell adhesion and
encapsulated cells micro-extrusion extrudability and spreading without
(skeletal myoblasts, and bioplotter bioprintability, fast modification of
BMSC, SMC, MSC, gelation, good stability and hydrogel
ASC, CPC, integrality of printed
chondrocytes, construct, medium
cardiomyocytes) elasticity, low cost,
nonimmunogenic
Collagen type I Encapsulated cells pH mediated or − Pneumatic Cell adherent, promote Poor mechanical
(bovine aortic endothelial thermal micro-extrusion proliferation, signal properties, slow
cells, keratinocytes, transducer, good extrusion gelation, unstable
fibroblasts, rat neural and bioprinting abilities,
cells, MSC, AFS) nonimmunogenic
Gelatin Encapsulated cells Thermal + Mechanical and Cell adherent, Unstable, fragile, weak
(HepG2, hepatocytes, pneumatic biocompatible, mechanical properties
fibroblasts, SMC) micro-extrusion nonimmunogenic at physiological
temperature and low
abilities to extrude and
print without
modification
PEG Encapsulated cells (bone Ionic, physical, − Pneumatic Support cell viability, Low proliferation rate,
marrow stem cells or or covalent micro-extrusion biocompatible, low cell adhesion,
porcine aortic valve agents nonimmunogenic, widely weak mechanical
interstitial cells) used in tissue engineering properties and stability
when modified without modification
Fibrin Acellular scaffolds or Enzymatic − Pneumatic Promote angiogenesis Difficult to control
encapsulated cells (AFS, micro-extrusion (causes inflammatory geometry, low
HUVEC) response), fast gelation, mechanical properties,
good integrality, medium limited EBB
elasticity printability
J.W. Choi et al.
Matrigel Encapsulated cells Thermal − Pneumatic Promote cell differentiation Slow gelation, which
(HepG2, BMSCs, gMSC, micro-extrusion and vascularization of affects mechanical
gEPC) construct, support cell stability, requires
viability, good cooling system for
bioprintability, highly extrusion bioprinting,
suitable particularly for expensive
cardiac tissue engineering
Agarose Encapsulated cells Thermal + Pneumatic and High mechanical Low cell adhesion,
(BMSCs osteosarcoma mechanical properties, stable, resistant fragile, require heating
cells, MSC) micro-extrusion for protein adsorption, low system for extrusion
cost, good integrality, bioprinting
nonimmunogenic
Chitosan Acellular scaffolds, Ionic or covalent − Pneumatic Antibacterial and Weak mechanical and
encapsulated cells agents micro-extrusion antifungal, medium stability properties
(cartilage progenitor printability, without modification,
cells, MSC, CPC) nonimmunogenic slow gelation rate
Pluronic® F-127 Encapsulated cells Thermal + Pneumatic and High printability, good Poor mechanical and
(human primary mechanical bioprintability, structural properties,
fibroblasts, BMSC, micro-extrusion nonimmunogenic slow gelation, rapid
HepG2) degradation, require
heating system for
extrusion bioprinting
Hyaluronic acid Encapsulated cells Ionic, covalent − Pneumatic and Promote proliferation and Rapid degradation,
(chondrocytes, HepG2, agents mechanical angiogenesis, fast gelation, poor mechanical
C3A, fibroblasts) micro-extrusion good bioprintability, properties and low
nonimmunogenic stability without
modification
9 3D Printing Technology in Craniofacial Surgery and Salivary Gland Regeneration
Methylcellulose Encapsulated Thermal, pH + Mechanical micro- High printability, Low bioprintability,

chondrocytes mediated extrusion nonimmunogenic sensitive on common
cell culture media,
unstable
Reprinted from Ref. [71]
187
Bioprinting, or robotic additive biomanufac 2. Poukens J, Haex J, Riediger D. The use of rapid pro-
totyping in the preoperative planning of distraction
turing, could be implemented by a precise layer-
osteogenesis of the cranio-maxillofacial skeleton.
by-layer placement of self-assembled tissue Comput Aided Surg. 2003;8(3):146–54.
spheroids in advanced hydrogels. The rapid pro- 3. Wagner JD, Baack B, Brown GA, Kelly J. Rapid
cess of tissue spheroids to self-assemble and to 3-dimensional prototyping for surgical repair of max-
illofacial fractures: a technical note. J Oral Maxillofac
form mature tissue in a relatively short time
Surg. 2004;62(7):898–901.
scale may provide the versatility needed for suc- 4. Faber J, Berto PM, Quaresma M. Rapid prototyping
cessful 3D bioprinting. Advancement of the tis- as a tool for diagnosis and treatment planning for
sue spheroids-based approach demands the maxillary canine impaction. Am J Orthod Dentofacial
Orthop. 2006;129(4):583–9.
synthesis of sophisticated soft biomaterials and
5. Mavili ME, Canter HI, Saglam-Aydinatay B, Kamaci
extracellular matrices, such as bio-processible S, Kocadereli I. Use of three-dimensional medical
and biomimetic stimuli- sensitive functional modeling methods for precise planning of orthogna-
hydrogels as bioink materials [71]. thic surgery. J Craniofac Surg. 2007;18(4):740–7.
6. D'Urso PS, Effeney DJ, Earwaker WJ, Barker TM,
Salivary gland regeneration is also possible
Redmond MJ, Thompson RG, et al. Custom cranio-
using 3D bioprinting with cells and hydrogels. plasty using stereolithography and acrylic. Br J Plast
Cells in the duct close to the acini are believed to Surg. 2000;53(3):200–4.
provide all the cell types required for the formation 7. Paiva WS, Amorim R, Bezerra DAF, Masini M.
Aplication of the stereolithography technique in complex
of acini and ducts. In vitro cultured salivary cells
spine surgery. Arq Neuropsiquiatr. 2007;65(2B):443–5.
could be assembled into three-dimensional acinar 8. Armillotta A, Bonhoeffer P, Dubini G, Ferragina S,
and ductal structures in the presence of collagen Migliavacca F, Sala G, et al. Use of rapid prototyping
and Matrigel® [86]. Bioprinting of three-dimen- models in the planning of percutaneous pulmonary
valved stent implantation. Proceedings of the
sional salivary gland structures may be guided by
Institution of Mechanical Engineers, Part H. J Eng
present experience with 3D bioprinting of vascular Med. 2007;221(4):407–16.
branch formation [72, 87]. Advancement in 3D 9. Kim MS, Hansgen AR, Wink O, Quaife RA, Carroll
bioprinting technology, in combination with a fun- JD. Rapid prototyping a new tool in understanding
and treating structural heart disease. Circulation.
damental understanding of the molecular mecha-
2008;117(18):2388–94.
nisms of development, provides a novel strategy 10. Wurm G, Tomancok B, Pogady P, Holl K, Trenkler
for salivary gland regeneration. J. Cerebrovascular stereolithographic biomodeling for
aneurysm surgery: technical note. J Neurosurg.
2004;100(1):139–45.
Conclusions
11. Giesel FL, Hart AR, Hahn HK, Wignall E, Rengier F,
3D printing technology enables more effec- Talanow R, et al. 3D reconstructions of the cerebral
tive patient consultations, increases diagnos- ventricles and volume quantification in children with
tic quality, improves surgical planning, acts brain malformations. Acad Radiol. 2009;16(5):610–7.
12. Guarino J, Tennyson S, McCain G, Bond L, Shea K,
as an orientation aid during surgery, and pro-
King H. Rapid prototyping technology for surgeries
vides a template for surgical resection. In of the pediatric spine and pelvis: benefits analysis.
addition, as bioprinting technology further J Pediatr Orthop. 2007;27(8):955–60.
evolves, tissues or organs might one day be 13. Hurson C, Tansey A, O’Donnchadha B, Nicholson P,
Rice J, McElwain J. Rapid prototyping in the assess-
made with patient-specific shapes and dimen-
ment, classification and preoperative planning of ace-
sions, thus substantializing the goal of indi- tabular fractures. Injury. 2007;38(10):1158–62.
vidualized medicine. 14. Hiramatsu H, Yamaguchi H, Nimi S, Ono H. Rapid
prototyping of the larynx for laryngeal frame work
surgery. Nihon Jibiinkoka Gakkai Kaiho. 2004;
107(10):949–55.
References 15. Winder J, Bibb R. Medical rapid prototyping tech-
nologies: state of the art and current limitations for
1. Müller A, Krishnan KG, Uhl E, Mast G. The applica- application in oral and maxillofacial surgery. J Oral
tion of rapid prototyping techniques in cranial recon- Maxillofacial Surgery: official journal of the American
struction and preoperative planning in neurosurgery. Association of Oral and Maxillofacial Surgeons.
J Craniofac Surg. 2003;14(6):899–914. 2005;63(7):1006–15.
16. Fullerton JN, Frodsham GC, Day RM. 3D printing for based on stereolithography in computer assisted sur-
the many, not the few. Nat Biotechnol. 2014;32(11): gery. Med Phys. 2006;33(9):3408–17.
1086–7. 32. Ibrahim D, Broilo TL, Heitz C, de Oliveira MG, de
17. Hoy MB. 3D printing: making things at the library. Oliveira HW, Nobre SM, et al. Dimensional error of
Med Ref Serv Q. 2013;32(1):94–9. selective laser sintering, three-dimensional printing
18. Choi JW, Kim N. Clinical application of three-
and PolyJet models in the reproduction of mandibular
dimensional printing technology in craniofacial plas- anatomy. J Cranio Maxillo Facial Surg: official publi-
tic surgery. Arch Plast Surg. 2015;42(3):267–77. cation of the European Association for Cranio-
19. Ventola CL. Medical Applications for 3D Printing: Maxillo-Facial Surgery. 2009;37(3):167–73.
Current and Projected Uses. P T: a peer-reviewed 33. Silva DN, de Oliveira MG, Meurer E, Meurer MI, da
journal for formulary management. 2014;39(10): Silva JVL, Santa-Barbara A. Dimensional error in
704–11. selective laser sintering and 3D-printing of models for
20. Olszewski R, Tranduy K, Reychler H. Innovative pro- craniomaxillary anatomy reconstruction. J Cranio
cedure for computer-assisted genioplasty: three- Maxillo Facial Surg: official publication of the
dimensional cephalometry, rapid-prototyping model European Association for Cranio-Maxillo-Facial
and surgical splint. Int J Oral Maxillofac Surg. Surgery. 2008;36(8):443–9.
2010;39(7):721–4. 34. Ohtani T, Kusumoto N, Wakabayashi K, Yamada S,
21. Winder J, McRitchie I, McKnight W, Cooke S. Virtual Nakamura T, Kumazawa Y, et al. Application of haptic
surgical planning and CAD/CAM in the treatment of device to implant dentistry--accuracy verification of
cranial defects. Stud Health Technol Inform. drilling into a pig bone. Dent Mater J. 2009;28(1):
2005;111:599–601. 75–81.
22. Dankowski R, Baszko A, Sutherland M, Firek L,
35. WE L, HE C, editors. Marching cubes: a high resolu-
Kalmucki P, Wroblewska K, et al. 3D heart model tion 3D surface construction algorithm. ACM sig-
printing for preparation of percutaneous structural graph computer graphics. New York: ACM; 1987.
interventions: description of the technology and case 36. Tiede U, Hoehne KH, Bomans M, Pommert A,

report. Kardiol Pol. 2014;72(6):546–51. Riemer M, Wiebecke G. Investigation of medical
23. Rengier F, Mehndiratta A, von Tengg-Kobligk H,
3D-rendering algorithms. IEEE Computer Graphics
Zechmann CM, Unterhinninghofen R, Kauczor HU, and Applications. 1990;10(2):41–53.
et al. 3D printing based on imaging data: review of 37. Yushkevich PA, Piven J, Hazlett HC, Smith RG, Ho S,
medical applications. Int J Comput Assist Radiol Gee JC, et al. User-guided 3D active contour segmentation
Surg. 2010;5(4):335–41. of anatomical structures: significantly improved efficiency
24. McGowan J. 3D printing technology speeds develop- and reliability. Neuroimage. 2006;31(3):1116–28.
ment. Health Estate. 2013;67(9):100–2. 38. Shattuck DW, Leahy RM. BrainSuite: an automated
25. Lee H, Fang NX. Micro 3D printing using a digital cortical surface identification tool. Med Image Anal.
projector and its application in the study of soft 2002;6(2):129–42.
materials mechanics. J Vis Exp: JoVE. 2012;69: 39. Schroeder WJ, Zarge JA, Lorensen WE, editors.

e4457. Decimation of triangle meshes. ACM Siggraph
26. Mironov V, Boland T, Trusk T, Forgacs G, Markwald Computer Graphics. New York: ACM; 1992.
RR. Organ printing: computer-aided jet-based 3D tis- 40. Field DA. Laplacian smoothing and Delaunay trian-
sue engineering. Trends Biotechnol. 2003;21(4): gulations. Comm Appl Numer Methods. 1988;4(6):
157–61. 709–12.
27. Mankovich NJ, Samson D, Pratt W, Lew D, Beumer 41. Hinton E, Campbell J. Local and global smoothing of
3rd J. Surgical planning using three-dimensional discontinuous finite element functions using a least
imaging and computer modeling. Otolaryngol Clin squares method. Int J Numer Methods Eng. 1974;8(3):
North Am. 1994;27(5):875–89. 461–80.
28. Olszewski Raphael RH. Clinical applications of rapid 42. Mahmood F, Owais K, Taylor C, Montealegre-Gallegos
prototyping models in cranio-maxillofacial surgery. M, Manning W, Matyal R, et al. Three-dimensional
In: Hoque M, editor. Advanced applications of rapid printing of mitral valve using echocardiographic data.
prototyping technology in modern engineering. J Am Coll Cardiol Img. 2015;8(2):227–9.
InTech; 2011. p. 173–206. 43. Yang DH, Kang J-W, Kim N, Song J-K, Lee J-W, Lim
29. Choi JY, Choi JH, Kim NK, Kim Y, Lee JK, Kim MK, T-H. Myocardial 3-dimensional printing for septal
et al. Analysis of errors in medical rapid prototyping myectomy guidance in a patient with obstructive
models. Int J Oral Maxillofac Surg. 2002;31(1):23–32. hypertrophic cardiomyopathy. Circulation. 2015;
30. Chang PS, Parker TH, Patrick Jr CW, Miller MJ. The 132(4):300–1.
accuracy of stereolithography in planning craniofacial 44. Jiankang H, Dichen L, Bingheng L, Zhen W, Tao
bone replacement. J Craniofac Surg. 2003;14(2): Z. Custom fabrication of composite tibial hemi-knee
164–70. joint combining CAD/CAE/CAM techniques.
31. Schicho K, Figl M, Seemann R, Ewers R, Lambrecht Proceedings of the Institution of Mechanical
JT, Wagner A, et al. Accuracy of treatment planning Engineers, Part H. J Eng Med. 2006;220(8):823–30.
45. Wang Z, Teng Y, Li D. Fabrication of custom-made 60. Moon KH, Ko IK, Yoo JJ, Atala A. Kidney diseases
artificial semi-knee joint based on rapid prototyping and tissue engineering. Methods. 2016;99:112–9.
technique: computer-assisted design and manufactur- 61. Ohashi K, Okano T. Functional tissue engineering of
ing. Zhongguo xiu fu chong jian wai ke za zhi= the liver and islets. Anat Rec. 2014;297(1):73–82.
Zhongguo xiufu chongjian waike zazhi= Chinese 62. Scarritt ME, Pashos NC, Bunnell BA. A review of cel-
journal of reparative and reconstructive surgery. lularization strategies for tissue engineering of whole
2004;18(5):347–51. organs. Front Bioeng Biotechnol. 2015;3:43.
46. Lee M-Y, Chang C-C, Ku Y. New layer-based imag- 63. Price AP, Godin LM, Domek A, Cotter T, D'Cunha J,
ing and rapid prototyping techniques for computer- Taylor DA, et al. Automated decellularization of
aided design and manufacture of custom dental intact, human-sized lungs for tissue engineering.
restoration. J Med Eng Technol. 2008;32(1):83–90. Tissue Eng Part C Methods. 2015;21(1):94–103.
47. Dai K-R, Yan M-N, Zhu Z-A, Sun Y-H. Computer- 64. Bueno EM, Diaz-Siso JR, Sisk GC, Chandawarkar A,
aided custom-made hemipelvic prosthesis used in Kiwanuka H, Lamparello B, et al. Vascularized com-
extensive pelvic lesions. J Arthroplasty. 2007;22(7): posite allotransplantation and tissue engineering.
981–6. J Craniofac Surg. 2013;24(1):256–63.
48. Harrysson OL, Hosni YA, Nayfeh JF. Custom-
65. Sears NA, Dhavalikar PS, Seshadri D, Cosgriff-

designed orthopedic implants evaluated using finite Hernandez E. A review of 3D printing of tissue engi-
element analysis of patient-specific computed tomog- neering constructs. Tissue Eng Part B Rev. 2016;
raphy data: femoral-component case study. BMC 22(4):298–310.
Musculoskelet Disord. 2007;8(1):91. 66. Rodriguez-Devora JI, Zhang B, Reyna D, Shi ZD, Xu
49. Stevens B, Yang Y, Mohandas A, Stucker B, Nguyen T. High throughput miniature drug-screening plat-
KT. A review of materials, fabrication methods, and form using bioprinting technology. Biofabrication.
strategies used to enhance bone regeneration in engi- 2012;4(3):035001.
neered bone tissues. J Biomed Mater Res B Appl 67. Hendriks J, Willem Visser C, Henke S, Leijten J, Saris
Biomater. 2008;85(2):573–82. DB, Sun C, et al. Optimizing cell viability in droplet-
50. Peltola SM, Melchels FP, Grijpma DW, Kellomäki based cell deposition. Sci Rep. 2015;5:11304.
M. A review of rapid prototyping techniques for tissue 68. Gao G, Cui X. Three-dimensional bioprinting in tis-
engineering purposes. Ann Med. 2008;40(4):268–80. sue engineering and regenerative medicine.
51. Amerini A, Hatam N, Malasa M, Pott D, Tewarie L, Biotechnol Lett. 2016;38(2):203–11.
Isfort P, et al. A personalized approach to interven- 69. Wu Z, Su X, Xu Y, Kong B, Sun W, Mi S. Bioprinting
tional treatment of tricuspid regurgitation: experiences three-dimensional cell-laden tissue constructs with
from an acute animal study. Interact Cardiovasc controllable degradation. Sci Rep. 2016;6:24474.
Thorac Surg. 2014;19(3):414–8. 70. Pati F, Jang J, Ha DH, Won Kim S, Rhie JW, Shim JH,
52. Griffith LG, Naughton G. Tissue engineering--current et al. Printing three-dimensional tissue analogues with
challenges and expanding opportunities. Science decellularized extracellular matrix bioink. Nat
(New York, NY). 2002;295(5557):1009–14. Commun. 2014;5:3935.
53. Melchels FP, Domingos MA, Klein TJ, Malda J, Bartolo 71. Ozbolat IT, Hospodiuk M. Current advances and

PJ, Hutmacher DW. Additive manufacturing of tissues future perspectives in extrusion-based bioprinting.
and organs. Prog Polym Sci. 2012;37(8):1079–104. Biomaterials. 2016;76:321–43.
54. Murphy SV, Atala A. 3D bioprinting of tissues and 72. Duan B. State-of-the-art review of 3D bioprinting for
organs. Nat Biotechnol. 2014;32(8):773–85. cardiovascular tissue engineering. Ann Biomed Eng.
55. Huh D, Matthews BD, Mammoto A, Montoya-Zavala 2016, in press.
M, Hsin HY, Ingber DE. Reconstituting organ-level 73. Skardal A, Devarasetty M, Kang HW, Mead I, Bishop
lung functions on a chip. Science (New York, NY). C, Shupe T, et al. A hydrogel bioink toolkit for mim-
2010;328(5986):1662–8. icking native tissue biochemical and mechanical
56. Huh D, Hamilton GA, Ingber DE. From 3D cell cul- properties in bioprinted tissue constructs. Acta
ture to organs-on-chips. Trends Cell Biol. Biomater. 2015;25:24–34.
2011;21(12):745–54. 74. Stanton MM, Samitier J, Sanchez S. Bioprinting of
57. Shim JH, Jang KM, Hahn SK, Park JY, Jung H, Oh K, 3D hydrogels. Lab Chip. 2015;15(15):3111–5.
et al. Three-dimensional bioprinting of multilayered 75. Chua CK, Yeong WY. Materials for bioprinting. In:
constructs containing human mesenchymal stromal Bioprinting: principles and applications. Singapore:
cells for osteochondral tissue regeneration in the rab- World Scientific Publishing Co. Pte. Ltd.; 2014.
bit knee joint. Biofabrication. 2016;8(1):014102. p. 117–64.
58. Lewis PL, Shah RN. 3D printing for liver tissue engi- 76. Ozbolat IT, Yu Y. Bioprinting toward organ fabrica-
neering: current Approaches and future challenges. tion: challenges and future trends. IEEE Trans Biomed
Curr Transpl Reports. 2016;3(1):100–8. Eng. 2013;60(3):691–9.
59. Maeshima A, Takahashi S, Nakasatomi M, Nojima 77. Murphy SV, Atala A. 3D bioprinting of tissues and
Y. Diverse cell populations involved in regeneration organs. Nat Biotechnol. 2014;32(8):773–85.
of renal tubular epithelium following acute kidney 78. Kang H-W, Lee SJ, Ko IK, Kengla C, Yoo JJ, Atala
injury. Stem Cells Int. 2015;2015:964849. A. A 3D bioprinting system to produce human-scale
tissue constructs with structural integrity. Nat 84. Chan YH, Huang TW, Young TH, Lou PJ. Human
Biotechnol. 2016;34(3):312–9. salivary gland acinar cells spontaneously form three-
79. Luiz EB, Juliana CC, Vijayan M, Ana LC, Nupura dimensional structures and change the protein
SB, Wesleyan AA, et al. Direct-write bioprinting of expression patterns. J Cell Physiol. 2011;226(11):
cell-
laden methacrylated gelatin hydrogels. 3076–85.
Biofabrication. 2014;6(2):024105. 85. Joraku A, Sullivan CA, Yoo J, Atala A. In-vitro recon-
80. Skardal A, Zhang J, Prestwich GD. Bioprinting
stitution of three-dimensional human salivary gland
vessel-like constructs using hyaluronan hydrogels tissue structures. Differentiation. 2007;75(4):318–24.
crosslinked with tetrahedral polyethylene glycol tet- 86. Maria OM, Maria O, Liu Y, Komarova SV, Tran

racrylates. Biomaterials. 2010;31(24):6173–81. SD. Matrigel improves functional properties of human
81. Kolesky DB, Homan KA, Skylar-Scott MA, Lewis JA. submandibular salivary gland cell line. Int J Biochem
Three-dimensional bioprinting of thick vascularized tis- Cell Biol. 2011;43(4):622–31.
sues. Proc Natl Acad Sci U S A. 2016;113(12):3179–84. 87. Mandrycky C, Wang Z, Kim K, Kim DH. 3D bio-
82. Holmberg KV, Hoffman MP. Anatomy, biogenesis
printing for engineering complex tissues. Biotechnol
and regeneration of salivary glands. Monogr Oral Sci. Adv. 2016;34(4):422–34.
2014;24:1–13. 88. Jeong GS, Song JH, Kang AR, Jun Y, Kim JH, Chang
83. Jakab K, Norotte C, Marga F, Murphy K, Vunjak- JY, et al. Surface tension-mediated, concave-
Novakovic G, Forgacs G. Tissue engineering by self- microwell arrays for large-scale, simultaneous pro-
assembly and bio-printing of living cells. duction of homogeneously sized embryoid bodies.
Biofabrication. 2010;2(2):022001. Adv Healthc Mater. 2013;2(1):119–25.
Functional Salivary Gland
Regeneration by Organ 10
Replacement Therapy
Miho Ogawa and Takashi Tsuji
Abstract
The salivary glands are exocrine organs that secrete saliva to maintain oral
health and homeostasis. Dysfunctional salivary glands exhibit symptoms
of dry mouth, including dental caries and dysfunction in speech and swal-
lowing. Current clinical therapies for dry mouth disease include artificial
saliva substitutes or parasympathetic stimulants, but these are transient
and palliative approaches. To achieve the functional recovery of dysfunc-
tional salivary glands, salivary gland tissue stem cells are thought to be
candidate cell sources for salivary gland tissue repair therapies. In addi-
tion, whole salivary gland replacement therapy is expected to be a novel
therapy resulting in the regeneration of fully functional salivary glands.
The salivary glands arise from their organ germs, which are induced by
epithelial-mesenchymal interactions. Recently, we developed a novel bio-
engineering method, i.e., the organ germ method, which can regenerate
the ectodermal organs, including the teeth, hair, lacrimal glands, and sali-
vary glands. The bioengineered salivary glands successfully secrete saliva
into the oral cavity and can also improve the symptoms of dry mouth, such
as bacterial infection and swallowing dysfunction. In this review, we sum-
marize recent findings and bioengineering methods for salivary gland
regeneration therapy.
10.1 Introduction
Exocrine glands, such as the sweat glands, lacri-

mal glands, and salivary glands, produce secre-
tory fluids such as sweat, tears, and saliva. These
M. Ogawa, PhD • T. Tsuji, PhD (*)
secretory fluids have important roles in maintain-
Organ Technologies Inc., Tokyo 101-0048, Japan
ing health and homeostasis. For example, saliva
RIKEN Center for Developmental Biology,
is secreted into the oral cavity and functions dur-
Kobe, Hyogo 650-0047, Japan
e-mail: [email protected]; ing chewing, digestion, cleaning, and swallowing
[email protected] [1, 2]. The salivary glands arise from the salivary

194 M. Ogawa and T. Tsuji
gland organ germ, which is generated by the chapter, we discuss the novel findings and bioen-
interaction of epithelial-mesenchymal stem cells gineering methods used in salivary gland regen-
during embryonic development [3, 4]. Salivary eration and the feasibility of these methods for
glands consist of three major glands, including future organ replacement regenerative therapy.
the submandibular gland (SMG), sublingual
gland (SLG), and parotid gland (PG), and many
minor glands. Overall, 95 % of the saliva secreted 10.2 D
evelopment of Salivary
per day is secreted by the SMG, SLG, and PG, Glands
and 5 % is secreted by the minor salivary glands. During Embryogenesis
The SMG and PG secrete serous saliva that con-
tributes mainly to the digestion of food. The SLG The salivary glands are generated from the
secretes mucous saliva, which protects the oral organ germ, which is produced by epithelial and
cavity from drying. Therefore, salivary gland mesenchymal stem cell interactions during early
dysfunction induces xerostomia and has an embryonic development. The SMG, SLG, and
adverse effect on bodily health [5, 6]. PG are generated through similar morphoge-
Xerostomia induces some clinical problems, netic events but differ in the timing and position
including dental decay, bacterial infection, masti- at which generation begins [2–4, 19–22]. The
cation and swallowing dysfunction, and a general development of the SMG is produced by the
reduction in quality of life [5–7]. Xerostomia invagination of the oral epithelium into the mes-
develops due to autoimmune diseases, such as enchymal region derived from the base of the
Sjögren’s syndrome, aging, and radiation therapy tongue on embryonic day (ED) 11 (prebud)
for head and neck cancer. Current therapies for (Fig. 10.1). The invaginated epithelial tissue
xerostomia rely on the use of artificial saliva sub- proliferates to form an epithelial stalk and a ter-
stitutes or parasympathetic stimulants to promote minal bud at the tip (initial bud). The epithelial
saliva secretion and to prevent dry mouth [8, 9]. stalk differentiates into the ducts, which are
However, these therapies only provide temporary called the intercalated, striated, and excretory
effects and cannot result in the recovery of sali- ducts depending on their position relative to the
vary gland dysfunction, which is why the devel- side of the opening. The terminal bud forms the
opment of novel therapies for restoration of branched structure by forming a cleft and by
salivary gland function is necessary [10]. repeating the elongation and branching process
Regenerative therapies utilizing stem cell during ED 12.5–13.5 (pseudoglandular) [23–
transplantation have been conducted in various 25]. From ED 15.0, the terminal bulbs differen-
organs, including salivary glands [11, 12]. In tiate into the acinar cells and begin the synthesis
addition, salivary gland regeneration therapy of secretory proteins, which differ depending on
involving gene modification and tissue engineer- the type of salivary gland [26]. The SMG and
ing may eventually be used to restore damaged PG secrete serous saliva, which contains a large
tissue and recover the flow of saliva [13]. Similar amount of digestive enzymes such as α-amylase,
organ replacement therapy approaches for ecto- which degrades starches and aids in digestion.
dermal organs such as the teeth and hair follicles, The SLG secretes mucous saliva, which con-
which can be achieved by transplantation of bio- tains rich mucin protein to protect the mouth
engineered organ germs that have been reconsti- against dryness. The epithelial cells also differ-
tuted using organ germ methods, have been entiate into myoepithelial cells, and adult epi-
reported [14–16]. Recently, we induced the thelial tissue stem cells are maintained in the
regeneration of salivary glands and lacrimal excretory duct to contribute to the repair of
glands using this method [17, 18]. In this book injured tissue [27–29].
10 Functional Salivary Gland Regeneration by Organ Replacement Therapy 195
ED11 ED12 ED12.5 ED13 ED14-15 Adult

Prebud Early initial Late initial Pseudo- Canalicular salivary
bud bud glanduar stage gland
Fig. 10.1 Schematic representation of the developmental chymal tissue and forms a certain morphology according to
stages of the salivary gland. The salivary glands, including the development of each organ. The salivary gland epithelial
SMG, SLG, and PG, are produced from organ germs induced tissue is formed by the epithelial stalk and terminal bulb,
by the interaction of the epithelial tissue and the mesenchy- which form the duct and acinar cells. The acinar cells mature
mal tissue. The epithelial tissue invaginates into the mesen- and begin to synthesize and secrete secretory proteins
10.3 D
iseases and Treatments 15,000–20,000 [30]. Of all SS patients, approxi-
of Salivary Glands mately 70 % are positive for the SS antibody SSA
(anti-Ro), and 40 % are positive for the SS anti-
The various types of salivary gland diseases body SSB (anti-La) [31–33]. However, these
include salivary gland-specific diseases, such as antibodies are not common to all patients, and the
salivary tumors, obstructive disorders, and infec- details of the pathogenic mechanism are not
tions, as well as the symptoms of systemic dis- clear. Current therapies for dry mouth syndrome
eases, such as Sjögren’s syndrome (SS), include symptomatic treatments such as the
lymphoma, and metabolic diseases [2]. Salivary administration of artificial saliva and sialogogues
dysfunction resulting from the atrophy of acinar to enhance moisture retention in the oral cavity
cells and saliva reduction leads to xerostomia [9]. In addition, parasympathomimetic drugs
(dry mouth syndrome). In Europe, approximately such as pilocarpine and cevimeline have been
20 % of the population is thought to suffer from used to stimulate the muscarinic M3 receptor and
dry mouth syndrome, and this disease has been induce salivary flow [32].
estimated to occur in approximately 800 million
people in Japan [30]. The treatment of head and
neck cancer, including salivary tumors, has been 10.4 S
alivary Gland Regeneration
performed using radiation therapy. However, as Using Stem Cells and Gene
the salivary glands are more sensitive to radia- Therapy
tion, this treatment can cause atrophy of the aci-
nar cells. Another condition that affects the 10.4.1 Tissue Regeneration Using
salivary glands is SS, an autoimmune disease that Adult Tissue Stem Cells
occurs frequently in middle-aged and elderly
women. It also affects the salivary glands as well Transplantation of adult tissue stem cells has
as other glands such as the lacrimal glands, become a recognized method for regenerative
resulting in dry eyes. The annual number of SS therapy to restore damaged tissues and organs in
patients has been reported to be approximately diverse diseases [11, 34]. Regarding salivary
gland regeneration, tissue stem/progenitor cell fer and circulated throughout the body by the
studies have reported that tissue stem cells have bloodstream [49–53]. Stem cell transplantation
the capacity for tissue repair. The c-kit- and sca- therapy and gene therapy are expected to be a
1-positive tissue stem cells are localized to the new treatment strategy for salivary gland disor-
intercalated duct of adult salivary glands, where ders and other diseases.
these cells can induce the acinar and duct cells
[35–38]. Furthermore, these stem cells are plu-
ripotent and can differentiate into the liver or 10.5 Functional Regeneration
pancreas’ tissues [39, 40]. The c-kit-positive sali- of a Bioengineered Salivary
vary gland stem cells can be cultured while main- Gland
taining the tissue repair capacity in vitro [12,
41–43]. A transplant of these cells can recover The current research for regenerating three-
the decreased salivation resulting from dimensional organs mimics organogenesis in
irradiation-induced atrophy of the acinar cells. In the developing embryo. In the salivary gland
addition, it has been reported that the bone regeneration field, epithelial cell aggregates are
marrow- derived mesenchymal stem cells have used to elucidate the mechanism of regeneration
the ability to promote the regenerative capacity and branching morphogenesis in vitro [54]. In
of the salivary gland stem cells that remain in the addition, the aggregate mix of epithelial and
damaged salivary glands after irradiation [44]. mesenchymal stem cells has been reported to
Stem cell transplantation is expected to serve as increase the number of branches and rate of
an effective means of achieving salivary gland branch formation [54].
regeneration. Recently, we demonstrated the possibility of
full functional regeneration of the ectodermal
organs, including the teeth, hair follicles, lacri-
10.4.2 Gene Therapy for Salivary mal glands, and salivary glands, using “organ
Gland Regeneration germ methods” that involved epithelial and mes-
enchymal stem cell manipulation techniques to
Because the salivary glands are located close to induce the formation of an organ germ (Fig.
the body surface, regeneration of damaged sali- 10.2a) [14–18]. Using this method, it is possible
vary glands via gene therapy has been studied. to control the size, number, morphology, and
The salivary glands open via the duct into the oral invagination direction of the regenerated organ
cavity, and thus methods of direct gene transfec- [16, 55]. For successful salivary gland replace-
tion into the salivary glands via the duct have ment therapy, it is important that the invagination
been reported. After the transfection of the water direction is controlled in such a way that the
channel aquaporin-1 (AQP1) gene using adeno- invaginated tissue connects to the ducts to secrete
virus or adeno-associated virus, the saliva secre- saliva into the oral cavity.
tion of irradiated salivary glands was significantly
recovered [45, 46]. However, salivary glands are
known to function as exocrine glands that secrete 10.5.1 Development
saliva in the oral cavity and as endocrine glands of a Bioengineered Salivary
that secrete substances into the bloodstream. Gland
Gene therapy using the salivary glands has also
been performed as a treatment for other diseases, To reconstruct the bioengineered salivary gland,
including SS and other genetic diseases [47, 48]. the germs including the SMG, SLG, and PG were
It has been reported that some materials, such as isolated from mice at embryonic day (ED) 13.5–
IL-17 receptor antibodies, growth hormones, 14.5. The bioengineered SMG germ showed
parathyroid hormones, and erythropoietin, can be epithelial-mesenchymal interactions and epithelial
expressed in adult salivary glands via gene trans- bud formation in organ culture (Fig. 10.2b) [17].
a Ectodermal organs
Epitherial
stem cell
Bio-engineered Tooth Regeneration:
organ germ PNAS, 2009
PLos ONE, 2011
Hair Follicle
Regeneratiion:
Mesenchymal Nature Commun.,
stem cell 2012
Sci. Rep., 2012
Organ Germ Method: Salivary and

Nat. Methods, 2007 lacrimal glands
regeneration
Nature Commun., 2013
b
0h 24 h 72 h
Fig. 10.2 Regeneration of salivary gland germs using contrast images of the bioengineered submandibular
organ germ methods. (a) Ectodermal organs including the gland germ at 0, 24, and 72 h of in vitro culture. The bio-
teeth, hair follicle, and secretory glands can be regener- engineered submandibular gland showed the interaction
ated in vivo by transplanting bioengineered organ germs between the epithelial and mesenchymal cells (24 h) and
that are reconstituted by organ germ methods. (b) Phase- invagination of the epithelial tissue (72 h)
The regenerated SLG and PG germs also showed 10.5.2 Secretion of a Bioengineered
patterns that were similar to that of the SMG germ Saliva
and structurally correct based on the natural sali-
vary gland germ. The correct transplantation of the About 1–1.5 L of saliva is secreted per day from
bioengineered salivary gland is important to the salivary glands. This secretion is induced by
achieving the secretion of saliva into the oral cav- eating, heat, and painful stimulation to the oral
ity. A bioengineered salivary gland germ was cavity. These stimulations are transmitted via the
engrafted into the PG duct of the model mice with afferent and efferent neural networks from the
salivary gland defects using an intraepithelial tis- oral cavity to the salivary glands (Fig. 10.4a) [56–
sue-connecting plastic method. In these mice, the 60]. Moreover, because secreted saliva also plays
SMG, SLG, and PG were excised. After 30 days, an important role in taste perception, the hypose-
the growth of the bioengineered salivary gland and cretion of saliva has been known to cause taste
its connection to the PG duct was successfully disorders [61–63]. In the medical field, the secre-
achieved (Fig. 10.3a) [17]. The bioengineered sali- tion of saliva from the salivary gland has been
vary gland structures, including the localization of analyzed using five tastes, including sour (citrate),
myoepithelial cells, the water channel aquaporin 5 bitter (quinine hydrochloride), salty (NaCl), sweet
(AQP5), and neuronal connections, were similar to (sucrose), and umami (glutamate) [63, 64].
those of a natural tissue (Fig. 10.3b) [17]. Compared to the control substance, citrate
Bioengineered
Oral side SMG
Palotid
duct
AQP5/E-cadhelin/ Calponin/NF-H/
b
HE staining PAS staining nuclei nuclei
submandibular gland
Bioengineered
sublingual gland
Bioengineered
Fig. 10.3 In vivo transplantation of a bioengineered sali- columns). Images of HE staining (left) and periodic acid
vary gland. (a) Photographs of the bioengineered subman- and Schiff (PAS) staining (second from the left). The bio-
dibular gland at day 30 after transplantation (left). The engineered SLG showed a strongly positive PAS staining.
bioengineered submandibular gland duct connected with Immunohistochemical images of calponin (red),
natural PG duct (right). (b) Histological analysis of the E-cadherin (green, third from the left), and NF-H (green,
bioengineered SMG (upper columns) and the SLG (lower right) are shown (Modified from Ref. Ogawa et al. [17])
s timulation induced significant quantities of saliva demonstrate that saliva secretion by the bioengi-
secretion from both the natural and bioengineered neered salivary gland may be controlled through
salivary gland (Fig. 10.4b) [17]. Saliva secretion the afferent-efferent neural network.
was induced in response to all tastes in addition to
the sour stimulus. The secretion amount depended
on the type of stimulus and exhibited the follow- 10.5.3 Functional Restoration
ing order: sour > bitter > umami > salty = sweet of Swallowing Dysfunction
[64]. In addition, the secreted bioengineered Using a Bioengineered
saliva contained the amylase protein, which has Salivary Gland
starch-degrading activity [17]. Salivation was
measured about 1–3 months after transplantation Oral health and homeostasis are maintained by
and followed up for 6 months. These findings saliva and saliva proteins, such as amylase,
0.8
a Salivatory b
Solitary
nucleus
nucleus
(ml/mg of salivaly gland weight, 5 min)

0.6
Cerebrum
Saliva flow
0.4
Salivary
gland 0.2
Gustatory
stimulation
by citrate
Saliva 0 1
secretion Natural Bio-
engineered
Fig. 10.4 Saliva secretion induced by gustatory stimula- from natural SMG (light bar) and bioengineered SMG
tion. (a) Schematic representation of saliva secretion (dark bar) after gustatory stimulation by citrate. The
induced by gustatory stimulation via the central nervous amount of secreted saliva exhibited no significant differ-
system. (b) Assessment of the amount of saliva secretion ence (b reprinted from Ref. Ogawa et al. [17])
lysozyme, IgA, lactoferrin, myeloperoxidase, survived, and their body weight increased within
NGF, EGF, and parotin. Therefore, the hypose- 4 days after transplantation [17]. These results
cretion of saliva causes various problems, includ- indicated that the bioengineered salivary gland
ing dental caries, bacterial infection, sleep can improve the swallowing function associated
disorders, and swallowing dysfunction [65, 66]. with the maintenance of oral health.
The bioengineered salivary gland-engrafted
mouse had fewer bacteria compared with the sali-
vary gland defect model mouse. Among the sali- 10.6 F
uture Directions of Salivary
vary gland functions, the swallowing function is Gland Regeneration
critical for reducing the risk of aspiration. The
saliva promotes the formation of a bolus of food Organ regenerative technology has advanced sig-
or water and triggers the swallowing reflex. nificantly. To achieve future clinical applications
Therefore, salivary gland dysfunction can cause of salivary gland replacement therapy, it is impor-
chronic lung disease and can affect the survival, tant to identify suitable cell sources. The ideal cell
quality of life, and overall health of an individual source is the patient’s own cells because there is no
[67]. In the salivary gland defect model mouse, immunological rejection. Recent stem cell biology
the body weight was abnormally decreased, and studies have revealed the presence of adult tissue
all mice died within 5 days, despite having free stem cells in the salivary gland. These adult tissue-
access to food and water (Fig. 10.5) [17]. Dry derived stem cells, which include c-kit- and sca-
mouth patients often drink high-viscosity water 1-positive cells, can repair the acinar cells injured
because they cannot swallow water. Similarly, by radiation and can partially recover the total
the salivary gland defect model mouse exhibited amount of secreted saliva [12, 41–44]. The sali-
a recovery of body weight and an increased sur- vary gland of adult stem cells would be valuable
vival rate when drinking high-viscosity water; cell sources for achieving salivary gland tissue
this result in the model mouse raised the possibil- regeneration via stem cell transplantation therapy.
ity that dysphagia may occur. In contrast, all of In contrast, pluripotent stem cells and induced plu-
the bioengineered salivary gland-engrafted mice ripotent stem cells are also potential cell sources
Fig. 10.5 Analysis of 120

body weight.
Measurement of body
weight (left graphs)
every 0.5 days after 110
transplantation in
normal mice (gray
dots); salivary gland
Body weight (%)

defect model mice 100
(black dots), salivary
gland-engrafted mice
(red dots), and salivary
gland defect model 90
mice were given
high-viscosity water
(green dots). All
salivary gland defect 80
model mice died within
5 days (✝) after
removing all of the
major salivary glands. 70
Salivary gland-
engrafted mice
recovered the body
weight (Reprinted from 60
Ref. Ogawa et al. [17]) 0 1 2 3 4 5 6 7
Date
for salivary gland regeneration because these cells Current whole organ regenerative therapy has
can differentiate into all types of cells, including the potential as a future therapeutic technology
endodermal, ectodermal, and mesodermal cells for several diseases. Salivary gland regenerative
[68–70]. It has been reported that some organs, therapy is regarded as a model for future secre-
such as the optic cup and pituitary gland, can be tory organ replacement therapies that will sub-
derived from ES cells or iPS cells. In the future, stantially contribute to achieving an understanding
the method of salivary gland regeneration using related to organ regeneration technology.
iPS cells is expected to be established [71–73].
Another important direction for future salivary Acknowledgments This work was partially supported by
gland regeneration therapy is to establish the mech- a Grant-in-Aid for KIBAN (A) from the Ministry of
anisms by which autoimmune diseases such as SS Education, Culture, Sports, Science and Technology (no.
cause xerostomia [31–33]. In autoimmune dis- 25242041). This work was also partially supported by
Organ Technologies, Inc.
eases, salivary gland damage, such as the atrophy
of acinar cells, is caused by self-antigens. Therefore,
even if the bioengineered salivary gland is trans- Conflict of Interest This work was partially
planted and can temporarily recover the saliva funded by Organ Technologies Inc. M.O. is a
secretion, there is the possibility that the bioengi- researcher and T.T. is a director at Organ
neered acinar cells will again undergo atrophy. To Technologies Inc. This work was performed
achieve future clinical applications of salivary under an Invention Agreement between Tokyo
gland replacement therapy, genetic modifications University of Science, RIKEN and Organ
of patient-derived stem cells will be necessary to Technologies Inc.
decrease the expression of autoantigens.
References transplantation of a bioengineered organ germ. Nat

Commun. 2013;4:2498.
1. Edgar M, Dawes C, Mullane OD. Saliva and oral 18. Hirayama M, Ogawa M, Oshima M, Sekine Y, Ishida
health. 3rd ed. London: British Dental Association; K, Yamashita K, Ikeda K, Shimmura S, Kawakita T,
2004. Tsubota K, Tsuji T. Functional lacrimal gland regen-
2. Tucker AS, Miletich I. Salivary glands; development, eration by transplantation of a bioengineered organ
adaptations, and disease. London: Karger; 2010. germ. Nat Commun. 2013;4:2497.
3. Avery JK. Oral development and histology. New York: 19. Jiménez-Rojo L, Granchi Z, Graf D, Mitsiadis

Thieme Press; 2002. p. 292–330. TA. Stem cell fate determination during development
4. Tucker AS. Salivary gland development. Semin Cell and regeneration of ectodermal organs. Front Physiol.
Dev Biol. 2007;18:237–44. 2012;3:107.
5. Vissink A, Burlage FR, Spijkervet FK, Jansma J, 20. Pispa J, Thesleff I. Mechanisms of ectodermal organ-
Coppes RP. Prevention and treatment of the conse- ogenesis. Dev Biol. 2003;262(2):195–205.
quences of head and neck radiotherapy. Crit Rev Oral 21. Jaskoll T, Melnick M. Embryonic salivary gland

Biol Med. 2003;14:213–25. branching morphogenesis. Austin (TX); Madame
6. Vissink A, Jansma J, Spijkervet FK, Burlage FR, Curie Bioscience Database [Internet]; 2004.
Coppes RP. Oral sequelae of head and neck radiother- 22. Knosp WM, Knox SM, Hoffman MP. Salivary gland
apy. Crit Rev Oral Biol Med. 2003;14:199–212. organogenesis. Wiley Interdiscip Rev Dev Biol.
7. Atkinson JC, Grisius M, Massey W. Salivary hypo- 2012;1(1):69–82.
function and xerostomia: diagnosis and treatment. 23. Sakai T. Epithelial branching morphogenesis of sali-
Dent Clin N Am. 2005;49:309–26. vary gland: exploration of new functional regulators.
8. Ship JA, Pillemer SR, Baum BJ. Xerostomia and the J Med Investig. 2009;56(Suppl):234–8.
geriatric patient. J Am Geriatr Soc. 2002;50:535–43. 24. Hsu JC, Yamada KM. Salivary gland branching mor-
9. Fox PC. Salivary enhancement therapies. Caries Res. phogenesis – recent progress and future opportunities.
2004;38:241–6. Int J Oral Sci. 2010;2(3):117–26.
10. Kagami H, Wang S, Hai B. Restoring the function of 25. Harunaga J, Hsu JC, Yamada KM. Dynamics of sali-
salivary glands. Oral Dis. 2008;14:15–24. vary gland morphogenesis. J Dent Res. 2011;90(9):
11. Körbling M, Estrov Z. Adult stem cells for tissue 1070–7.
repair – a new therapeutic concept? N Engl J Med. 26. Denny PC, Denny PA. Dynamics of parenchymal cell
2003;349:570–82. division, differentiation, and apoptosis in the young
12. Lombaert IM, Brunsting JF, Wierenga PK, Faber H, adult female mouse submandibular gland. Anat Rec.
Stokman MA, Kok T, Visser WH, Kampinga HH, de 1999;254:408–17.
Haan G, Coppes RP. Rescue of salivary gland func- 27. Man YG, Ball WD, Marchetti L, Hand AR.

tion after stem cell transplantation in irradiated Contributions of intercalated duct cells to the normal
glands. PLoS ONE. 2008;3:e2063. parenchyma of submandibular glands of adult rats.
13. O’Connell AC, Baccaglini L, Fox PC, O’Connell BC, Anat Rec. 2001;263(2):202–14.
Kenshalo D, Oweisy H, Hoque AT, Sun D, Herscher LL, 28. Ihrler S, Zietz C, Sendelhofert A, Lang S, Blasenbreu-
Braddon VR, Delporte C, Baum BJ. Safety and efficacy Vogt S, Löhrs U. A morphogenetic concept of salivary
of adenovirus-mediated transfer of the human aquapo- duct regeneration and metaplasia. Virchows Arch.
rin-1 cDNA to irradiated parotid glands of non-human 2002;440(5):519–26.
primates. Cancer Gene Ther. 1999;6(6):505–13. 29. Lombaert IM, MP H. Stem cells in salivary gland
14. Nakao K, Morita R, Saji Y, Ishida K, Tomita Y, Ogawa development and regeneration. In: Stem cells in cra-
M, Saitoh M, Tomooka Y, Tsuji T. The development niofacial development and regeneration. Hoboken:
of a bioengineered organ germ method. Nat Methods. Wiley-Blackwell; 2013. p. 271–84.
2007;4(3):227–30. 30. Hayashi Y, Arakaki R, Ishimaru N. Salivary gland and
15. Ikeda E, Morita R, Nakao K, Ishida K, Nakamura T, autoimmunity. J Med Investig. 2009;56:185–91.
Takano-Yamamoto T, Ogawa M, Mizuno M, Kasugai 31. Fox RI, Stern M, Michelson P. Update in Sjögren syn-
S, Tsuji T. Fully functional bioengineered tooth drome. Curr Opin Rheumatol. 2000;12(5):391–8.
replacement as an organ replacement therapy. Proc 32. Nakamura T, Matsui M, Uchida K, Futatsugi A,

Natl Acad Sci U S A. 2009;106(32):13475–80. Kusakawa S, Matsumoto N, Nakamura K, Manabe T,
16. Toyoshima KE, Asakawa K, Ishibashi N, Toki H, Taketo MM, Mikoshiba K. M3 muscarinic acetylcho-
Ogawa M, Hasegawa T, Irié T, Tachikawa T, Sato A, line receptor plays a critical role in parasympathetic
Takeda A, Tsuji T. Fully functional hair follicle regen- control of salivation in mice. J Physiol. 2004;558:
eration through the rearrangement of stem cells and 561–75.
their niches. Nat Commun. 2012;3:784. 33. Copelan EA. Hematopoietic stem-cell transplanta-

17. Ogawa M, Oshima M, Imamura A, Sekine Y, Ishida tion. N Engl J Med. 2006;354:1813–26.
K, Yamashita K, Nakajima K, Hirayama M, Tachikawa 34. Segers VFM, Lee RT. Stem-cell therapy for cardiac
T, Tsuji T. Functional salivary gland regeneration by disease. Nature. 2008;451:937–42.
35. Rotter N, Oder J, Schlenke P. Isolation and character- the C57BL/6.NOD-Aec1Aec2 mouse model of
ization of adult stem cells from human salivary Sjögren’s syndrome. Arthritis Res Ther. 2012;14(1):
glands. Stem Cells Dev. 2008;17(3):509–18. R40.
36. Sugito T, Kagami H, Hata K, Nishiguchi H, Ueda 48. Wang J, Wang F, Xu J, Ding S, Guo Y. Double-strand
M. Transplantation of cultured salivary gland cells adeno-associated virus-mediated exendin-4 expression
into an atrophic salivary gland. Cell Transplant. in salivary glands is efficient in a diabetic rat model.
2004;13(6):691–9. Diabetes Res Clin Pract. 2014;103(3):466–73.
37. Kishi T, Takao T, Fujita K, Taniguchi H. Clonal prolif- 49. Voutetakis A, Bossis I, Kok MR, Zhang W, Wang J,
eration of multipotent stem/progenitor cells in the Cotrim AP, Zheng C, Chiorini JA, Nieman LK, Baum
neonatal and adult salivary glands. Biochem Biophys BJ. Salivary glands as a potential gene transfer target
Res Commun. 2006;340(2):544–52. for gene therapeutics of some monogenetic endocrine
38. Takahashi S, Schoch E, Walker NI. Origin of acinar disorders. J Endocrinol. 2005;185(3):363–72.
cell regeneration after atrophy of the rat parotid 50. Samuni Y, Zheng C, Cawley NX, Cotrim AP, Loh YP,
induced by duct obstruction. Int J Exp Pathol. Baum BJ. Sorting of growth hormone-erythropoietin
1998;79:293–301. fusion proteins in rat salivary glands. Biochem
39. Hisatomi Y, Okumura K, Nakamura K, Matsumoto S, Biophys Res Commun. 2008;373(1):136–9.
Satoh A, Nagano K, Yamamoto T, Endo F. Flow cyto- 51. Samuni Y, Cawley NX, Zheng C, Cotrim AP, Loh YP,
metric isolation of endodermal progenitors from Baum BJ. Sorting behavior of a transgenic
mouse salivary gland differentiate into hepatic and erythropoietin- growth hormone fusion protein in
pancreatic lineages. Hepatology. 2004;39(3):667–75. murine salivary glands. Hum Gene Ther. 2008;19(3):
40. Okumura K, Nakamura K, Hisatomi Y, Nagano K, 279–86.
Tanaka Y, Terada K, Sugiyama T, Umeyama K, 52. Voutetakis A, Zheng C, Metzger M, Cotrim AP,

Matsumoto K, Yamamoto T, Endo F. Salivary gland Donahue RE, Dunbar CE, Baum BJ. Sorting of trans-
progenitor cells induced by duct ligation differentiate genic secretory proteins in rhesus macaque parotid
into hepatic and pancreatic lineages. Hepatology. glands after adenovirus-mediated gene transfer. Hum
2003;38(1):104–13. Gene Ther. 2008;19(12):1401–5.
41. Feng J, Van der Zwaag M, Stokman MA, Van Os R, 53. Nguyen CQ, Yin H, Lee BH, Chiorini JA, Peck

Coppes RP. Isolation and characterization of human AB. IL17: potential therapeutic target in Sjögren’s
salivary gland cells for stem cell transplantation to syndrome using adenovirus-mediated gene transfer.
reduce radiation-induced hyposalivation. Radiother Lab Investig. 2011;91(1):54–62.
Oncol. 2009;92:466–71. 54. Wei C, Larsen M, Hoffman MP, Yamada KM. Self-
42. Okumura K, Shinohara M, Endo F. Capability of tis- organization and branching morphogenesis of pri-
sue stem cells to organize into salivary rudiments. mary salivary epithelial cells. Tissue Eng. 2007;13(4):
Stem Cells Int. 2012;2012:502136. 721–35.
43. Nanduri LS, Maimets M, Pringle SA, van der Zwaag 55. Ishida K, Murofushi M, Nakao K, Morita R, Ogawa
M, van Os RP, Coppes RP. Regeneration of irradiated M, Tsuji T. The regulation of tooth morphogenesis is
salivary glands with stem cell marker expressing cells. associated with epithelial cell proliferation and the
Radiother Oncol. 2011;99(3):367–72. expression of Sonic hedgehog through epithelial-
44. Sumita Y et al. Bone marrow-derived cells rescue mesenchymal interactions. Biochem Biophys Res
salivary gland function in mice with head and neck Commun. 2011;405(3):455–61.
irradiation. Int J Biochem Cell Biol. 2011;43:80–7. 56. Ekström J, Khosravani N, Castagnola M, Messana
45. Shan Z, Li J, Zheng C, Liu X, Fan Z, Zhang C, I. Saliva and the control of its secretion. Berlin:
Goldsmith CM, Wellner RB, Baum BJ, Wang Springer; 2012.
S. Increased fluid secretion after adenoviral-mediated 57. Matsuo R, Yamamoto T, Yoshitaka K, Morimoto T.
transfer of the human aquaporin-1 cDNA to irradiated Neural substrates for reflex salivation induced by
miniature pig parotid glands. Mol Ther. 2005;11(3): taste, mechanical, and thermal stimulation of the oral
444–51. region in decerebrate rats. Jpn J Physiol.
46. Baum BJ, Alevizos I, Zheng C, Cotrim AP, Liu S, 1989;39:349–57.
McCullagh L, Goldsmith CM, Burbelo PD, Citrin 58. Matsuo R, Garrett JR, Proctor GB, Carpenter

DE, Mitchell JB, Nottingham LK, Rudy SF, Van Waes GH. Reflex secretion of proteins into submandibular
C, Whatley MA, Brahim JS, Chiorini JA, Danielides saliva in conscious rats, before and after preganglionic
S, Turner RJ, Patronas NJ, Chen CC, Nikolov NP, Illei sympathectomy. J Physiol. 2000;527:175–84.
GG. Early responses to adenoviral-mediated transfer 59. Proctor GB, Guy HC. Regulation of salivary gland
of the aquaporin-1 cDNA for radiation-induced sali- function by autonomic nerves. Auton Neurosci.
vary hypofunction. Proc Natl Acad Sci U S A. 2007;133:3–18.
2012;109(47):19403–7. 60. Turner RJ, Sugiya H. Understanding salivary fluid
47. Yin H, Nguyen CQ, Samuni Y, Uede T, Peck AB, and protein secretion. Oral Dis. 2002;8(1):3–11.
Chiorini JA. Local delivery of AAV2-CTLA4IgG 61. Matsuo R. Role of saliva in the maintenance of taste
decreases sialadenitis and improves gland function in sensitivity. Crit Rev Oral Biol Med. 2000;11:216–29.
62. Froehlich DA, Pangborn RM, Whitaker JR. The effect 68. Wu SM, Hochedlinger K. Harnessing the potential of
of oral stimulation on human parotid salivary flow induced pluripotent stem cells for regenerative medi-
rate and alpha-amylase secretion. Physiol Behav. cine. Nat Cell Biol. 2011;13(5):497–505.
1987;41(3):209–17. 69. Cohen DE, Melton D. Turning straw into gold: direct-
63. Sasano T, Satoh-Kuriwada S, Shoji N, Sekine-
ing cell fate for regenerative medicine. Nat Rev
Hayakawa Y, Kawai M, Uneyama H. Application of Genet. 2011;12(4):243–52.
umami taste stimulation to remedy hypogeusia based 70. Yan X, Qin H, Qu C, Tuan RS, Shi S, Huang GT. iPS
on reflex salivation. Biol Pharm Bull. 2010;33(11): cells reprogrammed from human mesenchymal-like
1791–5. stem/progenitor cells of dental tissue origin. Stem
64. Ogawa M, Yamashita K, Niikura M, Nakajima K, Cells Dev. 2010;19(4):469–80.
Toyoshima KE, Oshima M, Tsuji T. Saliva secretion 71. Eiraku M, Takata N, Ishibashi H, Kawada M,

in engrafted mouse bioengineered salivary glands Sakakura E, Okuda S, Sekiguchi K, Adachi T, Sasai
using taste stimulation. J Prosthodont Res. 2014;58(1): Y. Self-organizing optic-cup morphogenesis in three-
17–25. dimensional culture. Nature. 2011;472(7341):51–6.
65. Lamy E, Graca G, Costa GD, Franco C, Silva FC, 72. Suga H, Kadoshima T, Minaguchi M, Ohgushi M,
Baptista ES, et al. Changes in mouse whole saliva Soen M, Nakano T, Takata N, Wataya T, Muguruma
soluble proteome induced by tannin-enriched diet. K, Miyoshi H, Yonemura S, Oiso Y, Sasai Y. Self-
Proteome Sci. 2010;8:65. formation of functional adenohypophysis in three-
66. Cohen S. Isolation of a mouse submaxillary gland dimensional culture. Nature. 2011;480(7375):57–62.
protein accelerating incisor eruption and eyelid 73. Ozone C, Suga H, Eiraku M, Kadoshima T, Yonemura
opening in the new-born animal. J Biol Chem. S, Takata N, Oiso Y, Tsuji T, Sasai Y. Functional anterior
1962;237:1555–62. pituitary generated in self-organizing culture of human
67. Sreebny LM, Schwartz SS. A reference guide to drugs embryonic stem cells. Nat Commun. 2016;7:10351.
and dry mouth – 2nd edition. Gerodontology.
1997;14(1):33–47.
Part IV
Therapeutic Considerations for Restoration
of Salivary Function
Regulation of Salivary Secretion
11
Guy Carpenter and Polliane Carvalho
Abstract
The production of saliva in conscious humans is under the control of the
higher centers of the brain which is upregulated by an autonomic reflex in
response to taste, chewing, and smell. The higher centers of the brain
maintain the resting rate of salivary secretion which in the healthiest sub-
jects is sufficient to maintain oral health and perform the functions of the
mouth such as speaking, swallowing, and preventing the overgrowth of
microbial colonies on oral tissues. When asleep, the same higher centers
reduce their neural output leading to very low salivary flow, which pre-
vents choking or aspiration of saliva into the lungs leading to pneumonia.
An understanding of this complex control of salivary secretion is particu-
larly important for the regeneration of the salivary glands and their func-
tions. Stem cell treatments to replace the salivary tissue are an impressive
first step, but the new tissue needs to be under neural control. Inappropriate
salivary secretion can be just as much a problem as insufficient salivary
flow as demonstrated by drooling in stroke patients and patients on certain
antipsychotic medications, who “drown” in their own saliva at night.
11.1 Introduction by the three pairs of major glands (parotid, sub-

mandibular, and sublingual) and the hundreds of
Salivary secretion is an autonomic reflex acti- minor glands. The resting salivary rate is influ-
vated mainly by taste, chewing, and smell for enced mostly by the higher centers of the brain
some salivary glands [10, 17, 36]. In conscious (such as the hypothalamus and amygdala), which
humans, there is a resting rate (approximately increase their neural input to the salivary centers
0.5 ml/min) of salivary secretion into the mouth located in the brainstem during the day but
decrease at night and during times of anxiety
leading to a dry mouth at those times. Upon stim-
ulation by taste, smell, or chewing, salivary
G. Carpenter, PhD (*) • P. Carvalho, MSc, PhD
secretion is greatly upregulated two to three times
Mucosal and Salivary Biology Division,
King’s College London Dental Institute, London, UK greater than the resting rate. Thus when we put
e-mail: [email protected] food into the mouth, nerves associated with taste

208 G. Carpenter and P. Carvalho
Fig. 11.1 Sympathetic nerve innervation of acinar and empty-looking fat cells. The empty appearance of the fat
ductal cells in the submandibular glands of humans and cells is caused by the tissue fixation process. (b) Tyrosine
mice. (a) Tyrosine hydroxylase (brown) staining (1:100 hydroxylase (1:100) staining of mouse submandibular
dilution) of human submandibular glands lightly counter- gland also demonstrates plentiful sympathetic nerves
stained with hematoxylin (blue) shows dense sympathetic around acini and ducts. However, there are no fat cells in
innervation of the acinar and ductal cells as well as the mouse salivary glands
buds, bare nerve endings in the mucosa, and nucleus [28]. Sympathetic and parasympathetic
olfactory receptors in the nose and mechanore- efferents are then sent to each of the glands to
ceptors (Ruffini endings) in the mouth are all increase salivary secretion [38] (Fig. 11.1a, b).
stimulated and send signals via autonomic affer- Salivary flow is controlled separately from
ents back to the salivary centers in solitary tract salivary protein secretion, which itself is
11 Regulation of Salivary Secretion 209
d ifferently regulated depending on the secretory fat detection or digestion within food. However,
cell type, granule or vesicular secretion route, it is possible these lipases maintain the environ-
and even the protein itself [30]. Anaesthetized ment within the crypts to maintain taste bud
animals with isolated nerve preparations allowed acuity [29, 45].
researchers to show that parasympathetic stimu- Much progress has been made in characteriz-
lation per se caused a high flow with a low pro- ing the different channels responsible for the
tein concentration-type saliva, whereas detection of the basic tastes by taste bud cells
sympathetic stimulation evoked a high protein [9]. Salt taste is transmitted by sodium and pos-
with low-flow saliva [13, 38]. In conscious ani- sibly potassium channels located on the apical
mals, however, it was found that most salivary surface of taste bud cells that signal to afferent
secretion is composed of parasympathetic stimu- nerves via ATP molecules, whereas sour taste
lation with smaller amounts of sympathetic stim- (which is composed of protons) is detected by a
ulation overlaid [7, 28]. In contrast to the rest of separate channel. Receptors for bitter tastes and
the body, the autonomic nerves within the sali- glutamate have also been determined [3, 19].
vary glands work in harmony rather than antago- Now that specific receptors have been cloned,
nistically. Similar experiments in humans using more studies are examining the confounding fac-
adrenergic and cholinergic blocking drugs tors of taste receptors, such as age [31] and
revealed that a similar situation occurs [2]. obesity.
In addition to taste, the other major stimulus for
increased salivary secretion is chewing.
Mechanoreceptors in the gingival pocket sur-
11.2 S
alivary Secretion by Taste, rounding each tooth are the main receptor for tooth
Chewing, and Smell movement related to chewing. Several studies
have shown that increases in chewing activity lead
Taste buds are mostly located at the back of the to increased salivary secretion [1, 17] although,
tongue within the foliate and circumvallate interestingly, empty chewing (i.e., clenching teeth)
papilla. Some taste buds occur at the front of the does not lead to salivary secretion. Under normal
tongue associated with (but not always within) eating conditions, taste and chewing afferent nerve
the fungiform papillae, which are the red dots signals are combined to cause, at best, an additive
readily apparent on the tongue. The taste maps effect on salivary secretion.
of the tongue depicting sweet tastes at the front Additionally, smells can also stimulate sali-
of the tongue and salt at the sides, etc. often vary secretion. Olfactory stimuli have been
reproduced in textbooks are now largely dis- shown to stimulate submandibular/sublingual
counted. There is abundant evidence to show secretion but not parotid glands [24, 25]. When
most areas of the tongue are able to detect most food is consumed, aerosols are released from
tastes. There is considerable variation in the the food, probably aided by mixing with saliva,
number of taste buds between people, which has which travel via the retronasal route to the
some correlation with super-taster status – a olfactory neuroepithelium in the nose, and con-
heightened ability to detect and discriminate tribute flavor signals to the basic tastes detected
tastes [16]. At both circumvallate and foliate by the tongue. Indeed, much of the taste of food
papilla, the taste buds located within crypts are comes from the olfactory input rather than the
bathed in a secretion from serous minor salivary taste or chewing that occurs in the mouth.
glands. These glands (von Ebner’s) have some Olfactory stimulation of salivary glands is prob-
interesting proteins that have been suggested to ably of least importance to salivary secretion
be involved in fat detection, such as lingual stimulated by food in the mouth but does appear
lipase and lipocalin [22]. However, the output to contribute to the mouthwatering phenome-
from these glands is so small that it would be non. This is the subjective feeling of excessive
highly unlikely that they play a significant role in saliva in the mouth often associated with the
thought of food. However, few scientists have 11.4 Neural Connections

been able to show a thought or sight evoked to the Different Glands
secretion of saliva. It would appear that in some
situations, smells are apparent which could lead Salivary glands are unique in utilizing parasym-
to some secretion and the mouthwatering pathetic and sympathetic innervation in an addi-
response. However, in many situations where a tive/synergistic manner rather than the more
mouthwatering response occurs in the absence usual antagonist setup found for the regulation of
of food-related smells, it could also be due to blood flow and other functions in the body. Taste,
facial muscles squeezing on turgid salivary mechanical, or smell signals generate afferent
ducts to cause transient flows sufficient to be signals in fibers of the facial (CNVII), glossopha-
detected as mouthwatering [18]. ryngeal (CNIX), and trigeminal (CNV) nerves.
The nucleus of the solitary tract is innervated by
the CNVII and CNIX and sends interneurons to
the salivary centers. Interneurons also supply the
11.3 M
echanism of Salivary primary sympathetic salivary centers which are
Secretion located in the upper thoracic segments of the spi-
nal cord. Efferent nerve fibers from the salivary
As noted above, the fluid component of saliva is nuclei conduct signals via the chorda lingual
differently regulated to the protein (and to some nerve to the submandibular ganglion and onto the
extent the ionic component) of saliva. Salivary submandibular and sublingual glands. The
glands are composed of polarized epithelial parotid gland is supplied by efferent fibers in the
cells and have two main forms – the acini and glossopharyngeal (tympanic branch) nerve to the
the ducts. Often described as resembling a otic ganglion and postganglionic fibers in the
bunch of grapes, the acini are the site of fluid auriculotemporal nerve. There also appears to be
formation, while the ducts modify the saliva a contribution to the parotid gland efferent supply
and convey it to the mouth. Fluid is mobilized from the facial nerve. Minor salivary glands are
by creating an ionic gradient across the acinar supplied by parasympathetic nerve fibers in the
cells (primary saliva) which then is modified by buccal branch of the mandibular nerve, the lin-
the ducts [43]. The osmotic gradient is created gual nerve, and the palatine nerve.
by the selective secretion of chloride ions The salivary reflex is affected by the higher
through the apical membranes of polarized centers of the brain and shows circadian-like
acini. Thus once the parasympathetic nerves variations particularly in the resting salivary flow.
from the brain have conveyed the signal to This central neural activity appears to contribute
secrete by releasing acetylcholine which binds towards the lower salivary secretion during sleep
to muscarinic receptors on the acinar cells, acti- and zero flow during anesthesia. Suppression of
vation of intracellular calcium signaling elicits impulse traffic from the salivary nuclei to sali-
the opening of chloride channels on the apical vary glands leading to reduced salivation and dry
side of the acini. Sodium ions follow the chlo- mouth is most obviously demonstrated during
ride ions through an electrochemical attraction fear and anxiety. However, these effects are less
so that a higher concentration of sodium chlo- obvious in stimulated flow rates, where the
ride exists in the ductal/apical side of the cell effects of taste and chewing predominate.
compared to the basolateral/interstitial side. Significant advancements in our understand-
This osmotic gradient draws fluid from blood ing of the brain have been made possible by func-
vessels, via the interstitial compartment, toward tional MRI [40]. By the injection of labeled
the apical side and into the ductal system. Water glucose (or other substrates), the active regions of
may pass either around the acini through the the brain can be imaged when stimuli such as
tight junctions between cells or via the aquapo- food or drinks are put in the mouth. Despite some
rin channels within the acini [27]. recent advances in understanding of how tastes
are perceived, relatively little attention has been removal of calcium through the actions of plasma
paid as to how taste affects the salivary nuclei. membrane and ER calcium pumps. Store-
Most people believe that the thought of foods operated calcium entry has been shown to be
activates salivary secretion, the so-called mouth- dependent upon the presence of three proteins,
watering [20]. However, neither Pavlov nor STIM1, Orai1, and TRPC1 channels. Other
Lashley found any evidence to support the pres- receptors (α1-adrenoceptor, substance P neuroki-
ence of a conditional salivary reflex in man. fMRI nin 1 receptor, P2Y receptor, P2X receptors) uti-
studies have demonstrated the considerable dif- lize intracellular calcium signaling mechanisms
ferences between animal and human brains in but may make comparatively minor contributions
response to food [41]. Experiments by the author to salivary fluid secretion under physiological
also suggest that just the thought of food does not conditions.
sustain a stimulated salivary flow and that most
mouthwatering experiences are the result of
smells evoking submandibular/sublingual sali- 11.6 Protein Secretion
vary flow [18]. Using flow meters, it was possible
to detect, particularly when subjects were hun- Protein secretion, following stimulation by sym-
gry, small spikes of salivary flow. It was specu- pathetic and to lesser extent parasympathetic
lated that facial muscles compress the turgid nerves, activates adrenergic receptors on cells
ducts coming from salivary glands to the mouth and via intracellular cyclic AMP signaling causes
to cause small transient “flows” of saliva that can the storage granules to migrate toward the apical
be easily perceived by the subject. membrane, fuse, and then release their secretory
protein cargo into the ductal lumen. While the
storage granule mechanism is the major route by
11.5 Neurotransmitters which proteins enter the ductal lumen, non-
and Receptors storage vesicles also transport other proteins such
as secretory IgA (sIgA). This is the main anti-
Salivary acinar and ductal cells have been well body in saliva since it is actively transported via a
studied for their receptors. Muscarinic receptors membrane receptor (polymeric immunoglobulin
are usually cited as most important for fluid receptor) into saliva, whereas other classes of
secretion, but they also cause a significant degree antibody such as IgG and IgE are unable to bind
of protein secretion, mostly mucin and sIgA [8]. the membrane receptor and so passively diffuse
Using mouse knockout models, M3 appears most into saliva. Differences in the secretion of sIgA
important with smaller contributions from M1 and other proteins highlight differences between
[11, 14] although the in vivo situation is likely to different secretory mechanisms within one cell
be far more complex with inputs from purinergic [7]. However acinar and ductal cells have differ-
[4] and peptidergic neurotransmitters [12]. ent secretory proteins and are regulated by differ-
Acinar cell activation of fluid transport is ent neural impulses [39]. Thus considerable
achieved through increases in intracellular cal- complexity can exist in protein secretion within a
cium concentration and binding of calcium to single salivary gland.
ion-transporting proteins. The acinar cell musca- In humans, during normal conscious reflex
rinic receptors are G-protein-coupled receptors; secretion, this complexity is less apparent since
binding of acetylcholine leads to a G-protein/ secretory inputs are processed centrally and so
phospholipase C-mediated generation of inositol fluid and protein secretion seem to occur together.
triphosphate (IP3) from phosphatidylinositol Thus, from a single gland, such as the parotid,
4,5-bisphosphate. IP3 interacts with IP3 recep- which is the easiest to collect from using a
tors (IP3Rs) on the endoplasmic reticulum caus- Lashley suction cup, a similar range of proteins
ing release of stored calcium. Cytoplasmic are secreted at rest and when stimulated by dif-
calcium levels are tightly controlled by rapid ferent taste stimuli although the relative
p roportions of some proteins (such as sIgA and and has led some clinicians to deliberately dam-
amylase) may vary [37]. A more detailed study age the salivary glands by irradiation to effect a
by mass spectrometer methods has revealed that remedy [21]. Studies on rabbits have demon-
there are some small changes in the composition strated that irradiation has atrophic effects on the
of proteins [33]. The greatest variation in protein salivary glands [46] partially by damaging exist-
secretion is seen in whole-mouth saliva, which is ing cells but also by damaging stem cells leading
the combination of all the salivary glands. At rest, to reduced repopulation during normal cell turn-
submandibular and sublingual glands predomi- over [35]. Currently Botox treatment is limited to
nate; when stimulated by taste, parotid is the terminally ill patients due to the risk of whole
single most dominant contributor to salivary pro- body neurotoxicity and inhibiting the muscles
tein. However, smell or chewing without taste involved in swallowing thus potentially leading
evokes some differences in salivary proteins, to even greater excessive salivation (sialorrhea)/
most noticeably muc7 and statherin [18]. choking.
11.7 S
tudies of Neural Agonists 11.8 Considerations
and Antagonists for Regeneration of Salivary
Glands
α2-Adrenoceptor agonists (e.g., clonidine) and
antagonists (e.g., yohimbine) have been demon- Initial and recent studies by Coppes and col-
strated to act centrally in studies of reflex secre- leagues [32, 35] have demonstrated that stem cell
tion in human subjects and cholinergically evoked therapy of salivary glands involves an initial
secretion in animal models. α2-Adrenoceptor short-term recovery of the already existing sali-
blockade can increase salivary secretion, while vary cells and a longer-term repopulation of the
α2-adrenoceptor agonists inhibit secretion. It glandular stem cells. Short-term effects demon-
appears that adrenergic agonists such as amphet- strated a recovery of salivary flow in response to
amine exert an inhibitory effect on the flow of autonomimetics demonstrating a functional
saliva through the release of noradrenaline from recovery of the acinar cells. These studies have
nerves in the medulla causing activation of inhib- been a vital step in demonstrating the potential of
itory α2-adrenoceptors rather than through a the treatment and opens further lines of inquiry.
peripheral vasoconstrictive effect. These central The treatment with stem cells is though
effects of amphetamine that cause a dry mouth fraught with potential problems, the most serious
contrast with its action in the periphery leading of which is the potential transformation of the
to increased secretion of protein by salivary cells injected stem cells into noncancerous growths
and increased salivary protein concentration. The called teratomas. Even if this risk is extremely
presence of muscarinic receptors on neurons of low, it has to be balanced against the benefits to
the salivary nuclei may also partly explain the patients from increased salivary production.
observed effects on salivary secretion evoked by Injected stem cells appear to help preexisting
intracerebro-ventricular injection of pilocarpine salivary acinar cells recover function as well as
or atropine which were found to, respectively, repopulating the stem cell pool for longer-term
stimulate and inhibit salivation. function. Bone marrow soups [44] or mesenchy-
The recent use of botulinum toxin for intra- mal stem cell therapy has multiple growth factors
muscular paralysis has prompted a number of that probably boost the recovery of acinar cells in
researchers to use this on salivary glands, princi- an atrophic/diseased state. Epithelial stem cells
pally as a treatment for drooling [26]. Drooling extracted from existing salivary glands (labeled
represents the greatest concern of cares of stroke with anti-c-Kit antibodies) are probably required
victims, Parkinson’s, and other degenerative dis- for longer-term repopulation of the stem/progeni-
eases as the drooling constantly wets clothing tor cell pool [35]. In both cases, the endogenous
structure of the gland is used to position these responds in a manner of reflex to stimuli that
cells so that they can contribute to salivary secre- initiate increased salivation.
tion into the mouth. Although some animal studies have shown the
An alternative approach is to bioengineer the ability of salivary gland nerves to regrow into
gland in vitro and then transplant the gland into regenerating glands [6], this has not yet been
the existing ductal (excretory duct) structures shown in humans. Certainly with disease such as
[34]. In most of these studies, an autonomimetic Sjögren’s syndrome, it is known that there can be
has been used to test the function or ability of the a loss of the fine nerve fibers adjacent to areas of
gland to produce saliva. This drug (usually pilo- inflammation [42]. Thus, in any treatment of sali-
carpine) is injected into the body and reaches the vary glands, some account has to be taken of the
salivary glands by the bloodstream. It then stimu- preexisting innervation to determine the likely
lates salivary secretion by directly binding to chances of successful regeneration. It is well
muscarinic receptors on the acinar cells bypass- established that salivary glands require an intact
ing any nerve-acinar cell junction. Few papers innervation to maintain their histological appear-
have considered whether the nerve-acinar cell ance and functional capability.
junction has formed or is even functional.
Any researcher devising therapies for the
recovery of salivary glands has to have an appre- References
ciation of the complex neural control of salivary
glands as detailed above as well as the structural 1. Anderson DJ, Hector MP, Linden RWA. The effects of
architecture. In particular, the diurnal variation unilateral and bilateral chewing, empty clenching and
simulated bruxism, on the masticatory-parotid sali-
in salivary flow with an upregulation of the rest- vary reflex in man. Exp Physiol. 1996;81:305–12.
ing flow rate during periods of eating and chew- 2. Baum BJ, Wellner RB. Receptors in salivary glands.
ing but a downregulation of salivary secretion In: Garrett JR, Ekstrom J, Anderson LC, editors.
during sleep will need careful attention. The Neural mechanisms of salivary secretion. Basel:
Karger; 1999.
problems of a high flow at night have been well 3. Behrens M, Meyerhof W. Bitter taste receptor research
documented by Ekström and colleagues when comes of age: from characterization to modulation of
describing patients on clozapine – an antipsy- TAS2Rs. Semin Cell Dev Biol. 2013;24:215–21.
chotic prescribed to treatment-resistant schizo- 4. Bhattacharya S, Imbery JF, Ampem PT, Giovannucci
DR. Crosstalk between purinergic receptors and
phrenia patients. The night-time sialorrhea canonical signaling pathways in the mouse salivary
causes patients to choke at night with the feeling gland. Cell Calcium. 2015;58:589–97.
of “drowning” frequently reported. Clearly then 5. Burghartz M, Ginzkey C, Hackenberg S, Hagen
we need to carefully regulate the degree of R. Two-stage autotransplantation of the human sub-
mandibular gland: first long-term results.
regeneration of salivary glands whether by stem Laryngoscope. 2016;126:1551–5.
cells or drugs. 6. Carpenter GH, Khosravani N, Ekstrom J, Osailan SM,
In studies of transplanted salivary glands to Paterson KP, Proctor GB. Altered plasticity of the
avoid irradiation fields in treatment of head and parasympathetic innervation in the recovering rat sub-
mandibular gland following extensive atrophy. Exp
neck cancers [5] or treatment of chronic dry Physiol. 2009;94:213–9.
eyes [15, 23], it has become apparent that the 7. Carpenter GH, Proctor GB, Anderson LC, Zhang XS,
transplanted glands can become innervated Garrett JR. Immunoglobulin A secretion into saliva
from nerves attached to local blood vessels [15] during dual sympathetic and parasympathetic nerve
stimulation of rat submandibular glands. Exp Physiol.
leading to some interesting effects whereby 2000;85:281–6.
salivary secretion increased with exercise and 8. Carpenter GH, Proctor GB, Ebersole LE, Garrett
temperature (reflecting increased blood flow). JR. Secretion of IgA by rat parotid and submandibular
The preferred option that is required is the cells in response to autonomimetic stimulation
in vitro. Int Immunopharmacol. 2004;4:1005–14.
regenerated salivary gland to be innervated by 9. Chandrashekar J, Hoon MA, Ryba NJP, Zuker
both parasympathetic and sympathetic nerves CS. The receptors and cells for mammalian taste.
that were originally in the gland so that it Nature. 2006;444:288–94.
10. Chaudhari N, Roper SD. The cell biology of taste 26. Lungren MP, Halula S, Coyne S, Sidell D, Racadio
journal of cell biology. J Cell Biol. 2010;190: JM, Patel MN. Ultrasound-guided botulinum toxin
285–96. type a salivary gland injection in children for refrac-
11. Culp DJ, Luo W, Richardson LA, Watson GE,
tory sialorrhea: 10-year experience at a large tertiary
Latchney LR. Both M(1) and M(3) receptors regulate children’s hospital. Pediatr Neurol. 2016;54:70–5.
exocrine secretion by mucous acini. Am J Phys Cell 27. Ma TH, Song YL, Gillespie A, Carlson EJ, Epstein
Phys. 1996;40:C1963–72. CJ, Verkman AS. Defective secretion of saliva in
12. Del Fiacco M, Quartu M, Ekstrom J, Melis T, Boi M, transgenic mice lacking aquaporin-5 water channels.
Isola M, Loy F, Serra MP. Effect of the neuropeptides J Biol Chem. 1999;274:20071–4.
vasoactive intestinal peptide, peptide histidine methi- 28. Matsuo R. Central connections for salivary innerva-
onine and substance P on human major salivary gland tions and efferent impulse formation. In: Garrett JR,
secretion. Oral Dis. 2015;21:216–23. Ekstrom J, Anderson LC, editors. Neural mechanisms
13. Garrett JR. The Proper Role of Nerves in Salivary of salivary secretion. Basel: Karger; 1999.
Secretion - A Review. J Dent Res. 1987;66:387–97. 29. Matsuo R. Role of saliva in the maintenance of taste
14. Gautam D, Heard TS, Cui YH, Miller G, Bloodworth sensitivity. Crit Rev Oral Biol Med. 2000;11:216–29.
L, Wess J. Cholinergic stimulation of salivary secre- 30. Melvin JE, Yule D, Shuttleworth T, Begenisich T.
tion studied with M-1 and M-3 muscarinic receptor Regulation of fluid and electrolyte secretion in salivary
single- and double-knockout mice. Mol Pharmacol. gland acinar cells. Annu Rev Physiol. 2005;67:445–69.
2004;66:260–7. 31. Mennella JA, Pepino MY, Duke FF, Reed DR. Age
15. Gerling G, Garrett JR, Paterson KP, Sieg JP, Collin modifies the genotype-phenotype relationship for the
RO, Carpenter GH, Hakim SG, Lauer I, Proctor bitter receptor TAS2R38. BMC Genet. 2010;11:60.
GB. Innervation and secretory function of trans- 32. Nanduri LSY, Maimets M, Pringle SA, Van Der

planted human submandibular glands. Transplantation. Zwaag M, Van Os RP, Coppes RP. Regeneration of
2008;85:135–40. irradiated salivary glands with stem cell marker
16.
Hayes JE, Bartoshuk LM, Kidd JR, Duffy expressing cells. Radiother Oncol. 2011;99:367–72.
VB. Supertasting and PROP bitterness depends on 33.
Neyraud E, Sayd T, Morzel M, Dransfield
more than the TAS2R38 gene. Chem Senses. E. Proteomic analysis of human whole and parotid
2008;33:255–65. salivas following stimulation by different tastes.
17. Hector MP, Linden RW. Reflexes of salivary secre- J Proteome Res. 2006;5:2474–80.
tion. In: Garrett JR, Ekstrom J, Anderson LC, editors. 34. Ogawa M, Oshima M, Imamura A, Sekine Y, Ishida
Neural mechanisms of salivary secretion. Basel: K, Yamashita K, Nakajima K, Hirayama M, Tachikawa
Karger; 1999. T, Tsuji T. Functional salivary gland regeneration by
18. Ilangakoon Y, Carpenter GH. Is the mouthwatering transplantation of a bioengineered organ germ. Nat
sensation a true salivary reflex? J Texture Stud. Commun. 2013;4:2498.
2011;42:212–6. 35. Pringle S, Maimets M, Van Der Zwaag M, Stokman
19. Iwatsuki K, Ichikawa R, Uematsu A, Kitamura A, MA, Van Gosliga D, Zwart E, Witjes MJH, De Haan
Uneyama H, Torii K. Detecting sweet and umami G, Van Os R, Coppes RP. Human salivary gland stem
tastes in the gastrointestinal tract. Acta Physiol. cells functionally restore radiation damaged salivary
2012;204:169–77. glands. Stem Cells. 2016;34:640–52.
20. Jenkins GN, Dawes C. Psychic flow of saliva in man. 36. Proctor GB. The physiology of salivary secretion.

Arch Oral Biol. 1966;11:1203–4. Periodontology 2000. 2016;70:11–25.
21. Kasarskis EJ, Hodskins JST, WH C. Unilateral parotid 37. Proctor GB, Carpenter GH. Chewing stimulates

electron beam radiotherapy as palliative treatment for secretion of human salivary secretory immunoglobu-
sialorrhea in amyotrophic lateral sclerosis. J Neurol lin A. J Dent Res. 2001;80:909–13.
Sci. 2011;308:155–7. 38. Proctor GB, Carpenter GH. Regulation of salivary
22. Kawai T, Fushiki T. Importance of lipolysis in oral gland function by autonomic nerves. Auton Neurosci-
cavity for orosensory detection of fat. Am J Phys Basic Clin. 2007;133:3–18.
Regul Integr Comp Phys. 2003;285:R447–54. 39. Proctor GB, Carpenter GH. Salivary secretion: mech-
23. Konitzer J, Gerdan L, Sieg P, Hakim SG, Bruggemann anism and neural regulation. Monogr Oral Sci.
A, Gebhard M, Buchmann I. Salivary gland scintigra- 2014;24:14–29.
phy in patients suffering from severe keratoconjunctivi- 40. Rolls ET. Taste, olfactory and food texture reward
tis sicca Evaluation of the viability of submandibular processing in the brain and obesity. Int J Obes.
glands before and after transfer into the temporal fossa. 2011;35:550–61.
Nuklearmedizin-Nuclear Med. 2015;54:189–95. 41. Small DM. Taste representation in the human insula.
24. Lee VM, Linden RWA. An olfactory submandibular Brain Struct Funct. 2010;214:551–61.
salivary reflex in humans. Exp Physiol. 1992a;77: 42. Sorensen CE, Larsen JO, Reibel J, Lauritzen M,

221–4. Mortensen EL, Osler M, Pedersen AML. Associations
25. Lee VM, Linden RWA. The effect of odors on stimu- between xerostomia, histopathological alterations,
lated parotid salivary flow in humans. Physiol Behav. and autonomic innervation of labial salivary glands in
1992b;52:1121–5. men in late midlife. Exp Gerontol. 2014;57:211–7.
43. Thaysen JH, Thorn NA, Schwartz IL. Excretion of 45. Voigt N, Stein J, Galindo MM, Dunkel A, Raguse JD,
sodium, potassium, chloride and carbon dioxide in Meyerhof W, Hofmann T, Behrens M. The role of
human parotid saliva. Am J Physiol. 1954;178:155–9. lipolysis in human orosensory fat perception. J Lipid
44. Tran SD, Liu YN, Xia DS, Maria OM, Khalili S, Res. 2014;55:870–82.
Wang RWJ, Quan VH, Hu S, Seuntjens J. Paracrine 46. Xu H, Shan XF, Cong X, Yang NY, Wu LL, Yu GY,
effects of bone marrow soup restore organ function, Zhang Y, Cai ZG. Pre- and Post-synaptic Effects of
regeneration, and repair in salivary glands damaged Botulinum Toxin A on Submandibular Glands. J Dent
by irradiation. PLoS ONE. 2013;8:e61632. Res. 2015;94:1454–62.
Salivary Gland Gene Therapy
in Experimental and Clinical Trials 12
Michael Passineau
Abstract
Salivary gland gene therapy presents an opportunity to reprogram the
organ on the molecular level and achieve unprecedented therapeutic
advancements. This chapter will review the basic biology of gene transfer,
with emphasis on those vector systems that have performed well in the
salivary gland in animal models. Various therapeutic applications of sali-
vary gland gene therapy will be discussed, including radiation-induced
xerostomia and Sjögren’s syndrome. The concept of salivary glands as
endogenous bioreactors for systemic gene therapeutics in monogenetic
and acquired diseases will also be reviewed.
A brief history of the field, with regard to animal models, clinical trans-
lational studies, and ultimately a successful phase I/II clinical trial, will be
presented. The merits and limitations of the several animal models of sali-
vary gland gene therapy will be reviewed. The chapter concludes with a
discussion of human salivary gland gene therapy clinical trials, completed
and ongoing, and will point out congruence and discord between preclini-
cal animal studies and clinical trials. Salivary gland gene therapy is now
established as safe and therapeutically effective in humans, and the near
future of this field will be focused on making this technology practical for
outpatient use and broadly disseminating it into the practice of oral and
dental medicine.
12.1 Overview
Gene therapy may be broadly defined as the act

of delivering a genetic sequence to a target cell or
tissue to effect changes in gene expression.
Typically, this involves utilizing a vector contain-
M. Passineau, PhD ing an expression cassette comprised of a pro-
Gene Therapy Program, Allegheny Health Network, moter, open-reading frame for the gene to be
Pittsburgh, PA, USA expressed (referred to as the “transgene”), and a

218 M. Passineau
polyA sequence. The presence of these three has facilitated careful and comparative studies of
minimalist elements results in the expression of vectorology in the salivary gland and enabled
the transgene within the target cell/tissue at levels translation of salivary gland gene therapy from a
proportional to the activity of the promoter in the proof of concept in 1991 [2] to a successful phase
target. While transgene expression is the classic I human gene therapy clinical trial in 2006. This
example of gene therapy, other approaches chapter will review the roughly two decades
include delivery of silencing RNA, microRNA, between those milestone events, as well as specu-
and, increasingly, gene editing. lating on the future of this promising new
Gene therapy is fundamentally different from approach to salivary gland therapeutics, repair,
traditional pharmacology and even biopharma- and regeneration.
cology in that rather than manipulating the exist-
ing cellular machinery, gene therapy allows the
manipulation of the composition of the cellular 12.2 Gene Delivery Technology
machinery itself. Accordingly, gene therapy the-
oretically expands the armamentarium of thera- The fundamental challenge facing the gene thera-
peutic options beyond what is available through pist is, simply put, how to get a gene drug from
conventional pharmacology, and the two the outside of a target cell into the cytoplasm.
approaches could potentially be synergistic. While only 7 nm of cell membrane stands in the
When considering salivary gland disorders, gene way of this objective, that barrier has proven
therapy is particularly attractive due to the rela- exceptionally difficult to traverse, to the extent
tive paucity of conventional treatment options for that the gene therapist will often describe the cell
salivary gland disorders. membrane as “the longest 7 nm in nature.” In
Historically, gene therapy as a broad field has marked contrast to prokaryotic cells, eukaryotic
been very slow to meet its initial promise, and by cells repel foreign nucleic acids, based both upon
far the most important factor limiting the main- passive biophysical principles (DNA is strongly
streaming of gene therapy has been the difficulty hydrophilic, precluding its diffusion through the
of achieving vector delivery to the target cell/tis- cell membrane) and active immunological barri-
sue. As detailed below, cellular entry represents a ers, both extracellular and intracellular. In one
fundamental challenge for gene therapy vectorol- often-repeated statement to Time magazine in
ogy, but so does macroscopic delivery of the vec- 1999 [3], Inder Verma remarked, “There are only
tor to the target tissue. In particular, delivery of three problems in gene therapy: delivery, delivery
gene therapy vectors to internal, solid organs is and delivery.”
inherently challenging, since this requires either Dr. Verma’s statement accurately reflected the
surgical intervention or intravenous administra- driving force behind the remarkable adherence to
tion of a vector capable of efficiently trafficking Gartner’s curve that the gene therapy field has
to the target tissue without unwanted accumula- observed. On the positive side, this singular chal-
tion in off-target tissues. This latter consideration lenge of delivery has not diminished the theoreti-
has proven to be a major drawback of intravenous cal potential of gene therapy over the past two
administration of gene therapy vectors, particu- decades, and indeed this challenge has driven
larly viral vectors. rational and often successful research designed to
The salivary gland has unique attributes that meet it. What is now clear is that the diversity of
make it an intriguing and practical site for gene gene therapy applications precludes generalized
therapy, obviating some of the above concerns. applications of gene therapy techniques. What is
These considerations have been previously elab- needed rather is a gene therapy “toolbox” com-
orated [1] and include the direct accessibility of prised of dozen of vectors, devices, and tech-
the organ through bloodless intraoral cannulation niques, each of which may address only a few or
of the salivary ducts (Wharton’s and Stensen’s). even a single disease state. This principle has
This exceptionally simple and safe accessibility been well demonstrated in salivary gland gene
12 Salivary Gland Gene Therapy in Experimental and Clinical Trials 219
therapy over the past two decades, and this chap- gland gene therapy since 2002. Unfortunately, it
ter will pay particular attention to the refinement is known that AAV does not transduce acinar
of the salivary gland gene therapy “toolkit” into cells [5], and thus the field does not yet have a
its present form. viral vector that is well-suited for any gene ther-
apy application requiring gene transfer to the
acini of the salivary gland. It is this author’s opin-
12.2.1 Viral Vectors ion that an alternate AAV serotype will ultimately
be discovered that can transduce acinar cells
The only biological entity in nature capable of (source: unpublished data from John Chiorini,
efficient transport of nucleic acid across eukary- PhD), but given the variability in viral tropism for
otic cell membranes is the virus. Indeed, the the salivary glands of different species, any AAV
life cycle of the virus depends upon success- candidates for acinar cell gene transfer will need
ful transmembrane transfer of genetic material to be empirically validated in humans, a complex
into the host cell cytoplasm. For this reason, the and challenging task.
early days of gene therapy research were domi-
nated by viral vectors, and as of data from July
2015 (http://www.wiley.com//legacy/wileychi/ 12.2.2 Nonviral Vectors
genmed/clinical/), >70 % of all human gene ther-
apy clinical trials have utilized a viral vector. Nonviral vectors can potentially obviate both of
Viral vectors have two key attributes that the disadvantages of viral vectors in that target
determine their suitability for a given gene ther- cell affinity can be engineered directly into a syn-
apy application, such as gene transfer to the sali- thetic construct (often referred to generally as
vary gland (see Fig. 12.1, left panel): (1) proteins “nanoparticles”; see Fig. 12.1, right panel), and
in the capsid or envelope of the virus mediate cell lack of viral protein antigens can evade host
binding through receptor/ligand interactions, and immune recognition. The downside of nonviral
thus cell binding affinity can vary dramatically vectors is that they lack the viral mechanisms that
from one cell type to another, and (2) endosome mediate escape from the endosome. Endosomal
escape is a key feature of the viral life cycle but escape has proven extraordinarily difficult to
results in deposition of antigenic viral capsid pro- engineer artificially, and without robust endo-
teins on MHC receptors, triggering host response some escape, gene transfer efficiency is low due
to the infected cell. Both of these attributes must to the vector remaining entombed within the
be managed in such a way as to match the viral endosome. There has been one report of an endo-
vector choice to the particular application. For somolytic nanoparticle capable of effective
example, the canonical adenovirus type 5 (Ad5) siRNA to the salivary gland of the rodent [6], but
is strongly immunogenic in the host, but this can questions remain as to the relevance of this tech-
actually be an advantage in gene therapy applica- nology for accomplishing transgene expression.
tions in oncolysis or immunization. In the sali- The general consensus of the gene therapy litera-
vary gland, the most promising applications of ture thus far regarding nanoparticles is that they
gene therapy are for chronic conditions, and the are often capable of delivering siRNA to diverse
anti-Ad5 host response is very undesirable. targets but are far less effective in accomplishing
In the salivary gland, a very helpful study pub- expression of an exogenous transgene.
lished early on surveyed the efficacy of several Ultrasound-assisted gene transfer (UAGT; see
viral vectors for gene transfer to the salivary Fig. 12.1, center panel) was proposed in 2010 [7]
gland [4]. This work demonstrated that the only as a novel method for accomplishing gene transfer
viral vectors capable of robust transduction of the to the salivary gland of animal models. This tech-
salivary gland are adenovirus and adeno- nique circumvents endosomal escape altogether
associated virus (AAV), and these vectors have by relying upon transient disruption of the cell
formed the sole basis of viral-mediated salivary membrane and direct transmembrane transit of a
220 M. Passineau
Fig. 12.1 Basic biology of exogenous gene transfer to a engineered to evade MHC activation, but endosomal
target cell. The cell membrane prohibits the passage of escape has proven difficult to artificially engineer.
genetic material. Ligand/receptor interactions by viral Sonoporation (center) involves the direct transmembrane
vectors (left) or nanoparticles (right) can mediate endocy- transit of the nonviral vector via transient pores in the cell
tosis of the vector, but this leads to a destination in an membrane produced by fluid-phase cavitation. In all
endosome, not the cytoplasm. Viral mechanisms (left) instances, once the genetic material has been deposited in
lead to endosomal disruption, but viral antigens remain the cytoplasm, transgene expression can occur (© 2013
and are presented on MHC receptors. Nanoparticles are Michael J Passineau, PhD)
naked DNA vector. The technique relies upon bio- its own, because many other monogenetic disease
physical phenomenon referred to as “sonoporation” states were pursued from the earliest days of gene
in which the DNA vector is associated with therapy research. Nevertheless, CF is the prototype
~2.5 μm microbubbles comprised of a lipid bilayer gene therapy application and, ironically, one of the
surrounding a perfluoropropane gas core. These most difficult to address, with no effective gene
microbubbles resonate in a 1 MHz acoustic field, therapy treatment yet manifest, despite almost
and if the acoustic field is of sufficient power, the 30 years of research.
microbubbles are violently destroyed, resulting in Salivary gland gene therapy has never
fluid cavitation, which disrupts the cell membrane. employed the gene repair paradigm, chiefly
Sonoporation alone appears to have minimal because of the extremely low impact of monoge-
effects upon cellular homeostasis, and no damage netic diseases of the salivary gland. However, an
to the gland is apparent after UAGT, either by his- intriguing application of salivary gland gene ther-
tological or proteomic analysis [8]. apy has been proposed that attempts to address
systemic monogenetic diseases by repurposing
a portion of the salivary gland into an “endog-
12.3 Gene Repair enous bioreactor” [10, 11] for systemic delivery
of biomolecules deficient in these disease states.
Gene repair refers to the use of gene transfer to cor- A variety of therapeutic biomolecules have been
rect a deleterious genetic polymorphism in the tar- expressed in the salivary glands of various animal
get cell. The gene repair paradigm was the founding models and been shown to circulate systemically,
principle of gene therapy many decades ago, driven with examples including erythropoietin [10, 11],
primarily by the well-characterized gene/disease human growth hormone [10–12], α-galactosidase
relationship that exists between CFTR mutations A [13], and GLP-1 [14].
and cystic fibrosis [9]. It is not fair to credit CF as The clinical translation of the salivary
the genesis of the gene therapy research field all on gland bioreactor paradigm faces two principal
c hallenges. First, the sorting of the transgene 12.4 Genetic Medicine

between the apical (exocrine-directed) and
basal (endocrine- directed) compartments of Genetic medicine as a gene therapy strategy dif-
the acinar cell cannot be predicted and varies fers from gene repair in that the therapeutic effect
between species. A substantial body of care- is achieved not by replacement of a defective gene
ful research, almost all of it carried out at the but by the use of a transgene to manipulate the
NIDCR, has sought to understand and manipu- physiology of the target cell. In this paradigm, the
late these two sorting pathways, but the results context is rarely inherited monogenetic disease
have not yet produced clear-cut principles likely (where a gene repair strategy would presumably
to apply to humans [15–19]. Thus, the animal be more direct) but rather complex acquired dis-
models, particularly rodent models, are lim- ease. As mentioned above, monogenetic inherited
ited in their ability to predict whether a trans- diseases of the salivary gland are very rare, so
gene intended for endocrine circulation after genetic medicine is by far the more important
expression in the salivary glands of humans paradigm to consider in the near-to-intermediate
would indeed traffic as intended. Heretofore, future of salivary gland gene therapy.
no clinical trials have been approved to study The majority of genetic medicine research in
this gene therapy strategy in humans, and until the salivary gland has focused on radiation-
such a trial can empirically address these sort- induced xerostomia, likely due to the consistency
ing issues, the idea remains extremely intrigu- and reasonably direct clinical relevance of irradi-
ing but unrealized. ated animal models. Sjögren’s syndrome (SS)
The second issue that confounds the use of has also attracted the interest of the salivary gland
salivary glands as endogenous bioreactors is gene therapy community, owing to its very high
the imprecision of the relationship [vector dose/ prevalence, but the clinical relevance of the ani-
systemic transgene circulating]. The principles mal models is far more problematic. Animal
of pharmacology as they relate to biopharma- models of salivary gland dysfunction, and their
ceuticals dictate that dosing must be maintained limitations, will be discussed in the following
within a relatively narrow window in order to section. Recently, an intriguing application of
maximize efficacy while avoiding intolerable genetic medicine involves the ectopic synthesis
side effects. With an exogenously delivered of a hormone called PYY, normally produced by
biological agent, this dosing can be tightly con- endocrine cells of the gut epithelium, in the sali-
trolled, but producing the biological agent vary gland. This approach has been shown to
endogenously simply does not allow this level modulate taste [20] and induce satiety [21], with
of precision. For this reason, the concept of potential applications to the treatment of obesity
producing a growth hormone such as HGH, and some forms of anorexia.
endogenously, might not be workable due to the Animal studies utilizing genetic medicine
risk of overdose and the tendency of transgene approaches to radiation-induced xerostomia have
expression to degrade over time. The notable been impressively varied but fall into one of two
exception might be orphan diseases such as categories: (1) protection of the gland from the
lysosomal storage diseases, where even a small predictable radiation insult or (2) functional res-
amount of circulating enzyme can be benefi- toration of a salivary gland where damage is
cial, and increased enzyme is only additive to already manifest. The latter application has here-
the therapeutic benefit. This might also be the tofore been focused exclusively on a single trans-
case in a schema like the expression of GLP-1 in gene, aquaporin-1 (AQP1) that presumably
type 2 diabetes mellitus where the salivary localizes circumferentially in the cell membrane
gland provides a baseline level of the therapeu- of surviving ductal cells (and possibly acinar
tic that could reduce (but not eliminate) the cells, which may survive in small numbers) and
need for exogenous administration of conven- facilitates transmembrane flux [22] of interstitial
tional pharmaceuticals. fluid into the intraductal labyrinth. The first report
222 M. Passineau
of this genetic medicine treatment paradigm uti- molecular etiology of local salivary gland dys-
lizing the archetype adenoviral type 5 vector was function in SS is poorly understood, meaning that
in 1997 [23], and since that time, this therapeutic the molecular targets for a gene therapy strategy
approach has been upscaled from rodents to min- in humans are not at all clear. The advent of
iature swine [24], replicated with an AAV vector UAGT as a nonviral platform for salivary gland
in rodents [25] and later miniswine [26], carried gene therapy may obviate the former concern, but
out with nonviral ultrasound-assisted gene trans- the latter remains unresolved.
fer in miniswine [27], and finally shown both Gene therapy is fundamentally a method for
safety and efficacy in a phase I dose-escalation altering the intracellular programming of the tar-
human clinical trial [28]. The field-wide implica- get cell and as such presents nearly unlimited
tions of this milestone clinical trial are discussed versatility. However – and to extend the metaphor
at the end of this chapter. of software – since the “program” of the pathobi-
Radioprotective gene therapy strategies have ology in SS is not understood, there is no basis
been more diverse but are much less advanced upon which to directly act upon it with gene ther-
down the clinical translational pathway. Examples apy. At the highest level, it might be effective to
include Tousled kinase [29, 30], human keratino- simply utilize transgenes with broad anti-
cyte growth factor [31, 32], vascular endothelial inflammatory activity, based upon what is known
growth factor and/or fibroblast growth factor [33, about the disease at the histological level. Even
34], and heat shock protein 25 [35]. The mecha- starting with this basic premise, a meaningful
nisms by which these various treatments exert animal model must be engaged before a clinical
their therapeutic effect are well understood in translational strategy can approach clinical trials,
some cases, less so in others. All of these biologi- and it is principally animal models that have hin-
cal therapies appear to be safe and could poten- dered the progress of gene therapy for SS.
tially be candidates for human clinical trials. In the main, it is this author’s opinion that an
However, the cost/benefit analysis of using an animal model with direct relevance to translating
adenoviral vector, which is itself inflammatory, SS gene therapy to humans does not yet exist.
must be considered. Alternatives, such as AAV or This is not for lack of effort, as evidenced by a
sonoporation might be considered for clinical tri- recent and very helpful review by Park et al. [36]
als but only after more work is done to optimize that inventories more than a dozen mouse models
the transgene expression dynamics of each thera- of SS. Missing from the Park et al. review are
peutic candidate to maximize protection and several additional animal models where induc-
minimize unintended biological consequences. tion of SS-like disease is itself accomplished by
SS is the most prevalent salivary gland dis- viral gene transfer [37–39]. The phenotypes of
ease, affecting roughly 0.5–3 % of the general these mouse models are variable, with respect to
population, with a 1:9 male to female ratio. Thus, salivary and lacrimal gland manifestations, as
SS presents the greatest single opportunity for well as systemic manifestations and autoantibod-
clinical impact using salivary gland genetic med- ies. Similarly, a number of gene therapy studies
icine strategies. Given the sophisticated state of have been carried out targeting the salivary
salivary gland gene therapy relative to gene ther- glands of these mouse models, in particularly the
apy applications in other organs and tissues, there C57BL/6.NOD-Aec1Aec2 model [40–44]. These
is reason to be optimistic that this hope may be studies do suggest the potential of gene therapy
realized in the coming decade(s). However, there for SS, but their relevance to the human condition
are two principal hurdles to exploiting gene ther- is not clear, and at this point it is difficult to imag-
apy to disrupt and/or reverse SS: (1) since a major ine a successful Investigational New Drug (IND)
element of primary SS is inflammatory, the pros- application for a human gene therapy clinical
pects for using a virus, even AAV, to treat this trial based upon any of this evidence.
disease locally within the salivary gland are So what role can mouse models play in unrav-
doubtful, and (2) despite decades of research, the eling the complex molecular choreography of
human SS and more importantly in providing the (1990). Despite this history of promise and
rationale for a human clinical trial involving gene setbacks, gene therapy has blossomed during the
transfer to the salivary gland in SS? Perhaps one second decade of this century, with several
answer lies in working backward to mouse mod- industry-sponsored approval applications now
els rather than forward. The availability of sali- pending before the FDA and the EMA. Some of
vary gland biopsies from SS patients through the trials and tribulations of the gene therapy
various tissue banking efforts provides a rich field, such as the gradual waning of viral-
resource for genomic and transcriptomic studies mediated RPE65 gene repair in congenital blind-
of the molecular pathobiology of SS in the ness [47], were perhaps predictable. Others,
affected salivary gland, already yielding insights including the persistence of AQP1 expression in
on the human condition [45, 46]. Since the only the human salivary gland following viral-
relevant targets for gene therapy are the human mediated gene therapy [48], were not. As the
ones, it may be best to de-emphasize the impor- field traverses this critical juncture in its history,
tance of SS-like phenotype in mice and rather it is absolutely clear that high-quality, large ani-
focus gene therapy efforts going forward on mal preclinical models of gene therapy are one
mouse models that allow demonstration of clear- key to assuring smooth clinical translation of
cut modulation of genetic targets known to be candidate gene therapies. In this regard, the sali-
relevant to the human condition. vary gland gene therapy field enjoys a distinct
A final potential application of salivary gland- advantage, as discussed below.
based genetic medicine strategies bears mention- One of the major advantages of salivary
ing, although it has not yet been exploited in a gland gene therapy is the accessibility of the
peer-reviewed research manuscript. It is theoreti- organ itself, via intraoral cannulation of the sali-
cally possible to use gene therapy to alter the pro- vary duct (parotid or submandibular).
tein composition of saliva for applications Technically, the cells of the salivary gland are
focused on the oral cavity itself. One can envi- epithelium, and the tight junctions between
sion the introduction of proteins or peptides with these cells, combined with the encapsulation of
specific activity against intractable or opportunis- the organ, make the salivary gland a cutaneous
tic dental pathogens, such as Candida albicans or structure. Delivery of a vector to the salivary
Aggregatibacter actinomycetemcomitans. In this gland via cannulation avoids communication
author’s opinion, this novel methodology for with the systemic circulation and avoids many
chronic oral disease is extremely promising and of the complexities of systemic toxicity that
warrants greater attention. have complicated other applications of gene
therapy. Fundamentally, cannulating the sali-
vary duct of large animals (e.g., miniswine) or
even that of a rodent is very similar to the actual
12.5 Experimental Models situation that would be faced in humans, further
of Salivary Gland Disease increasing the relevance of animal models to
and Gene Therapy clinical translation.
The first models of salivary gland gene ther-
The great promise of gene therapy is that it pres- apy were rodents, as expected, and their low cost,
ents therapeutic opportunities that are simply not as well as the availability of transgenic modifica-
possible with traditional pharmacotherapies. tion (in mice), makes rodents the mainstay of
However, with this new paradigm come addi- research development in this field. The primary
tional risks, some known, and some yet unknown. drawback of these animals is the very small size
Gene therapy is not a new field and has been an and relative fragility of the salivary duct, making
active area of research for more than four decades, the technique extremely challenging from a tech-
with the first successful human gene therapy nical standpoint and also prone to variability. A
intervention now almost three decades in the past second issue that limits rodent models is the
224 M. Passineau
major difference in salivary gland structure and 12.6 C

linical Trials of Gene
function between rodents and humans. In rodents, Therapy
the submandibular gland (SMG) is the largest
gland and plays a much greater role in saliva pro- As of the date of this writing, clinicaltrials.gov
duction than the parotid gland. In humans, the lists only two gene therapy studies involving the
roles of the parotid and submandibular are salivary gland, one completed, and the other
reversed. Thus, caution must be exercised when approved but not yet recruiting. Both trials utilize
interpreting the results of gene therapy interven- the same therapeutic philosophy, expressing a
tions in the SMG of rodents. Nevertheless, with water channel called aquaporin-1 (AQP1) in the
the notable exception of exocrine/endocrine sort- salivary glands of patients whose salivary glands
ing, most gene therapy insights gained in the have been damaged by radiotherapy for head-
SMG of rodents do faithfully upscale to larger and-neck cancer. Since acinar cells are known to
animal models. be destroyed in the context of radiation-induced
Between rodents and humans, a large ani- xerostomia, it is presumed that AQP1 achieves
mal model of gene therapy is highly desirable, expression in the surviving ductal cells, driving
if not essential. Whereas primates are indis- transcellular fluid flux from the interstitial space
pensable as preclinical models in other areas into the intraductal labyrinth.
of research, they present very considerable The first trial, NCT00372320, is usually
challenges, primarily in ethical and cost con- referred to as the “AdAQP1” trial and involved a
siderations. Fortunately, primates have proven phase I dose-escalation paradigm to evaluate the
unnecessary for translation of salivary gland safety of adenoviral gene therapy in the human
gene therapy to clinical trials and may even be salivary gland [28]. As a first-in-man study, the
less informative than other animal models [49]. primary outcome measure was safety, but second-
To a lesser degree, canines also present chal- ary measures included both objective (salivary
lenges as preclinical models, primarily due to flow) and subjective (xerostomia) metrics of thera-
their status as companion animals. Fortunately, peutic efficacy. It is difficult to overstate the
the pig has proven to be nearly ideal as a pre- importance of this successful trial as an inflection
clinical model of salivary gland gene therapy point for the field, as it established proof-of-
and has proven sufficient for clinical transla- principle for the safety and efficacy of salivary
tion of the first-in-man salivary gland gene gland gene therapy, allowing subsequent studies to
therapy [24]. The parotid glands of pigs are focus on practicality of this approach for dissemi-
similar in overall size and location to humans nation to the oral medicine clinical community.
(see Fig. 12.2), and while pigs are highly intel- Two important observations from the AdAQP1
ligent animals, their status as an agricultural trial are worth noting. First, the dose-escalation
product obviates ethical concerns that plague strategy demonstrated that there is an ideal dos-
canine or primate studies. The principal chal- ing range for efficacy, above and below which is
lenge in the use of pigs as models of salivary ineffective and possibly harmful [28]. The sec-
gland gene therapy is the forceful growth ond observation, specifically addressed in a later
kinetics of the domestic farm swine, a trait that report [48], was the duration of therapeutic effect
has been bred into these animals over millen- in the human patients, which was unexpected in
nia but is cumbersome for chronic studies. For that it substantially exceeded the transient effects
reasons of convenience, the miniature pig has seen in preclinical large animal studies [24]. This
been established as the penultimate preclini- issue will need to be considered in future clinical
cal model of salivary gland gene therapy after trials, and while it is premature to draw broad
an impressive body of collaborative work by conclusions, this single observation gives reason
Songlin Wang in Beijing and Bruce Baum in to hope that salivary gland gene therapy in
Bethesda characterized the essential elements humans may produce therapeutic effects that last
of this animal model [50, 51]. for at least several months.
Fig. 12.2 Comparison of parotid gland position and shows (DRR) of a miniswine subject. The right parotid
relative size and humans versus miniswine. Upper is shown in magenta. Note that human and swine
shows digital radiography reconstruction (DRR) of images are not referenced to the same scale. (Upper,
parotid glands from two de-identified patients from the reproduced from [18], lower © 2014 Olivier Gayou,
Allegheny Cancer Center (Pittsburgh, PA). The lower PhD)
The second trial, NCT02446249, builds upon genic adenoviral vector. In this follow-up trial,
the success of the AdAQP1 clinical trial, while AAV2 will be used to express the AQP1 trans-
attempting to improve upon what is presumed to gene, and preclinical studies suggest that the
be its principal weakness, the highly immuno- duration of therapeutic efficacy could be much
226 M. Passineau
longer than that seen with adenovirus [26]. AAV 5. Katano H, Kok MR, Cotrim AP, Yamano S, Schmidt
M, Afione S, Baum BJ, Chiorini JA. Enhanced trans-
has been used safely in other human clinical tri-
duction of mouse salivary glands with aav5-based
als, and this trial holds great promise for long- vectors. Gene Ther. 2006;13:594–601.
lasting palliative therapy in radiation-induced 6. Arany S, Benoit DS, Dewhurst S, Ovitt CE.
xerostomia. Nanoparticle-mediated gene silencing confers radio-
protection to salivary glands in vivo. Mol Ther.
Looking into the future, it is now clear that sali-
2013;21:1182–94.
vary gland gene therapy is a promising new modal- 7. Passineau MJ, Zourelias L, Machen L, Edwards PC,
ity for treating salivary gland dysfunction in Benza RL. Ultrasound-assisted non-viral gene
radiation-induced xerostomia and may soon find transfer to the salivary glands. Gene Ther. 2010;17:
1318–24.
widespread application in this condition [52].
8. Geguchadze R, Wang Z, Zourelias L, Perez-Riveros P,
However, this patient population represents but a Edwards PC, Machen L, Passineau MJ. Proteomic
small niche of patients suffering from xerostomia profiling of salivary gland after nonviral gene transfer
and hyposalivation, with age-related xerostomia mediated by conventional plasmids and minicircles.
Mol Ther Methods Clin Dev. 2014;1:14007.
and SS representing tens of millions in the devel-
9. Riordan JR, Rommens JM, Kerem B, Alon N,
oped world and presumably hundreds of millions Rozmahel R, Grzelczak Z, Zielenski J, Lok S, Plavsic
of individuals worldwide. Since these conditions N, Chou JL, et al. Identification of the cystic fibrosis
are chronic but not lethal, the key to effective ther- gene: cloning and characterization of complementary
DNA. Science. 1989;245:1066–73.
apy will be durable transgene expression, either
10. Voutetakis A, Kok MR, Zheng C, Bossis I, Wang J,
with single treatment or (more likely) with a thera- Cotrim AP, Marracino N, Goldsmith CM, Chiorini JA,
peutic strategy that allows for serial readministra- Loh YP, Nieman LK, Baum BJ. Reengineered sali-
tion. If the latter is required, questions remain as to vary glands are stable endogenous bioreactors for sys-
temic gene therapeutics. Proc Natl Acad Sci U S A.
the approach that might be taken with viral vectors,
2004;101:3053–8.
as even AAV generates host immune response with 11. Voutetakis A, Bossis I, Kok MR, Zhang W, Wang J,
repeated dosing. Nonviral alternatives, such as Cotrim AP, Zheng C, Chiorini JA, Nieman LK, Baum
UAGT or nanoparticles, might also be considered BJ. Salivary glands as a potential gene transfer target
for gene therapeutics of some monogenetic endocrine
but have not yet been evaluated in the salivary
disorders. J Endocrinol. 2005;185:363–72.
glands of humans. With clinical data demonstrating 12. Racz GZ, Zheng C, Goldsmith CM, Baum BJ, Cawley
the safety of salivary gland gene therapy now NX. Toward gene therapy for growth hormone defi-
firmly in hand, it is important that clinical develop- ciency via salivary gland expression of growth hor-
mone. Oral Dis. 2015;21:149–55.
ment of these various modalities be accelerated in
13. Passineau MJ, Fahrenholz T, Machen L, Zourelias L,
order to mainstream salivary gland gene therapy Nega K, Paul R, MacDougall MJ, Mamaeva O, Steet
into the practice of oral and dental medicine. R, Barnes J, Kingston HM, Benza RL. Alpha-
galactosidase a expressed in the salivary glands par-
tially corrects organ biochemical deficits in the fabry
mouse through endocrine trafficking. Hum Gene
References Ther. 2011;22:293–301.
14. Voutetakis A, Cotrim AP, Rowzee A, Zheng C, Rathod
1. Perez P, Rowzee AM, Zheng C, Adriaansen J, Baum T, Yanik T, Loh YP, Baum BJ, Cawley NX. Systemic
BJ. Salivary epithelial cells: An unassuming target delivery of bioactive glucagon-like peptide 1 after
site for gene therapeutics. Int J Biochem Cell Biol. adenoviral-mediated gene transfer in the murine sali-
2010;42:773–7. vary gland. Endocrinology. 2010;151:4566–72.
2. Mastrangeli A, O'Connell B, Aladib W, Fox PC, 15. Voutetakis A, Zheng C, Wang J, Goldsmith CM, Afione
Baum BJ, Crystal RG. Direct in vivo adenovirus- S, Chiorini JA, Wenk ML, Vallant M, Irwin RD, Baum
mediated gene transfer to salivary glands. Am J Phys. BJ. Gender differences in serotype 2 adeno-associated
1994;266:G1146–55. virus biodistribution after administration to rodent sali-
3. Leon J, Park A. Time. 1999;153:68–70, 73. vary glands. Hum Gene Ther. 2007;18:1109–18.
4. Shai E, Falk H, Honigman A, Panet A, Palmon 16. Yan X, Voutetakis A, Zheng C, Hai B, Zhang C, Baum
A. Gene transfer mediated by different viral vectors BJ, Wang S. Sorting of transgenic secretory proteins
following direct cannulation of mouse submandibular in miniature pig parotid glands following adenoviral-
salivary glands. Eur J Oral Sci. 2002;110:254–60. mediated gene transfer. J Gene Med. 2007;9:779–87.
17. Samuni Y, Cawley NX, Zheng C, Cotrim AP, Loh YP, C, Whatley MA, Brahim JS, Chiorini JA, Danielides
Baum BJ. Sorting behavior of a transgenic erythropoietin- S, Turner RJ, Patronas NJ, Chen CC, Nikolov NP, Illei
growth hormone fusion protein in murine salivary GG. Early responses to adenoviral-mediated transfer
glands. Hum Gene Ther. 2008;19:279–86. of the aquaporin-1 cdna for radiation-induced salivary
18. Samuni Y, Zheng C, Cawley NX, Cotrim AP, Loh YP, hypofunction. Proc Natl Acad Sci U S A. 2012;109:
Baum BJ. Sorting of growth hormone-erythropoietin 19403–7.
fusion proteins in rat salivary glands. Biochem 29. Palaniyandi S, Odaka Y, Green W, Abreo F, Caldito
Biophys Res Commun. 2008;373:136–9. G, De Benedetti A, Sunavala-Dossabhoy
19. Voutetakis A, Zheng C, Metzger M, Cotrim AP,
G. Adenoviral delivery of tousled kinase for the pro-
Donahue RE, Dunbar CE, Baum BJ. Sorting of trans- tection of salivary glands against ionizing radiation
genic secretory proteins in rhesus macaque parotid damage. Gene Ther. 2011;18:275–82.
glands after adenovirus-mediated gene transfer. Hum 30. Timiri Shanmugam PS, Dayton RD, Palaniyandi S,
Gene Ther. 2008;19:1401–5. Abreo F, Caldito G, Klein RL, Sunavala-Dossabhoy
20. La Sala MS, Hurtado MD, Brown AR, Bohorquez G. Recombinant aav9-tlk1b administration amelio-
DV, Liddle RA, Herzog H, Zolotukhin S, Dotson rates fractionated radiation-induced xerostomia. Hum
CD. Modulation of taste responsiveness by the Gene Ther. 2013;24:604–12.
satiation hormone peptide yy. FASEB J. 2013;27: 31. Zheng C, Cotrim AP, Sunshine AN, Sugito T, Liu L,
5022–33. Sowers A, Mitchell JB, Baum BJ. Prevention of
21. Acosta A, Hurtado MD, Gorbatyuk O, La Sala M, radiation-induced oral mucositis after adenoviral
Duncan D, Aslanidi G, Campbell-Thompson M, vector-mediated transfer of the keratinocyte growth
Zhang L, Herzog H, Voutetakis A, Baum BJ, factor cdna to mouse submandibular glands. Clin
Zolotukhin S. Salivary pyy: a putative bypass to sati- Cancer Res. 2009;15:4641–8.
ety. PLoS ONE. 2011;6:e26137. 32. Zheng C, Cotrim AP, Rowzee A, Swaim W, Sowers
22. Delporte C, Hoque AT, Kulakusky JA, Braddon VR, A, Mitchell JB, Baum BJ. Prevention of radiation-
Goldsmith CM, Wellner RB, Baum BJ. Relationship induced salivary hypofunction following hkgf gene
between adenovirus-mediated aquaporin 1 expression delivery to murine submandibular glands. Clin Cancer
and fluid movement across epithelial cells. Biochem Res. 2011;17:2842–51.
Biophys Res Commun. 1998;246:584–8. 33. Guo L, Gao R, Xu J, Jin L, Cotrim AP, Yan X, Zheng
23. Delporte C, O'Connell BC, He X, Lancaster HE,
C, Goldsmith CM, Shan Z, Hai B, Zhou J, Zhang C,
O'Connell AC, Agre P, Baum BJ. Increased fluid Baum BJ, Wang S. Adltr2ef1alpha-fgf2-mediated
secretion after adenoviral-mediated transfer of the prevention of fractionated irradiation-induced sali-
aquaporin-1 cdna to irradiated rat salivary glands. vary hypofunction in swine. Gene Ther. 2014;21:
Proc Natl Acad Sci U S A. 1997;94:3268–73. 866–73.
24. Shan Z, Li J, Zheng C, Liu X, Fan Z, Zhang C, 34. Cotrim AP, Sowers A, Mitchell JB, Baum BJ.

Goldsmith CM, Wellner RB, Baum BJ, Wang Prevention of irradiation-induced salivary hypofunc-
S. Increased fluid secretion after adenoviral-mediated tion by microvessel protection in mouse salivary
transfer of the human aquaporin-1 cdna to irradiated glands. Mol Ther. 2007;15:2101–6.
miniature pig parotid glands. Mol Ther. 2005;11: 35. Lee HJ, Lee YJ, Kwon HC, Bae S, Kim SH, Min JJ,
444–51. Cho CK, Lee YS. Radioprotective effect of heat shock
25. Braddon VR, Chiorini JA, Wang S, Kotin RM, Baum protein 25 on submandibular glands of rats. Am
BJ. Adeno-associated virus-mediated transfer of a J Pathol. 2006;169:1601–11.
functional water channel into salivary epithelial cells 36. Park YS, Gauna AE, Cha S. Mouse models of primary
in vitro and in vivo. Hum Gene Ther. sjogren’s syndrome. Curr Pharm Des. 2015;21:
1998;9:2777–85. 2350–64.
26. Gao R, Yan X, Zheng C, Goldsmith CM, Afione S, 37. Nguyen CQ, Yin H, Lee BH, Carcamo WC, Chiorini
Hai B, Xu J, Zhou J, Zhang C, Chiorini JA, Baum BJ, JA, Peck AB. Pathogenic effect of interleukin-17a in
Wang S. Aav2-mediated transfer of the human aqua- induction of sjogren’s syndrome-like disease using
porin-1 cdna restores fluid secretion from irradiated adenovirus-mediated gene transfer. Arthritis Res
miniature pig parotid glands. Gene Ther. 2011;18: Ther. 2010;12:R220.
38–42. 38. Vosters JL, Landek-Salgado MA, Yin H, Swaim WD,
27. Wang Z, Zourelias L, Wu C, Edwards PC, Trombetta Kimura H, Tak PP, Caturegli P, Chiorini
M, Passineau MJ. Ultrasound-assisted nonviral gene JA. Interleukin-12 induces salivary gland dysfunction
transfer of aqp1 to the irradiated minipig parotid in transgenic mice, providing a new model of
gland restores fluid secretion. Gene Ther. 2015;22: sjogren’s syndrome. Arthritis Rheum. 2009;60:
739–49. 3633–41.
28. Baum BJ, Alevizos I, Zheng C, Cotrim AP, Liu S, 39. Vosters JL, Yin H, Roescher N, Kok MR, Tak PP,
McCullagh L, Goldsmith CM, Burbelo PD, Citrin Chiorini JA. Local expression of tumor necrosis
DE, Mitchell JB, Nottingham LK, Rudy SF, Van Waes factor-receptor 1:Immunoglobulin g can induce sali-
228 M. Passineau
vary gland dysfunction in a murine model of sjogren’s do the genes tell us about disease pathogenesis?
syndrome. Arthritis Res Ther. 2009;11:R189. Autoimmun Rev. 2014;13:756–61.
40. Nguyen CQ, Yin H, Lee BH, Chiorini JA, Peck
47. Kaiser J. Gene therapy for blindness may fade with
AB. Il17: potential therapeutic target in sjogren’s syn- time. Science. 2015. http://www.sciencemag.org/
drome using adenovirus-mediated gene transfer. Lab news/2015/05/gene-therapy-blindness-may-fade-time
Investig. 2011;91:54–62. 48. Zheng C, Baum BJ, Liu X, Goldsmith CM, Perez P,
41. Roescher N, Vosters JL, Yin H, Illei GG, Tak PP, Jang SI, Cotrim AP, McCullagh L, Ambudkar IS,
Chiorini JA. Effect of soluble icam-1 on a sjogren’s Alevizos I. Persistence of haqp1 expression in human
syndrome-like phenotype in nod mice is disease stage salivary gland cells following adhaqp1 transduction is
dependent. PLoS ONE. 2011;6:e19962. associated with a lack of methylation of hcmv pro-
42. Lee BH, Carcamo WC, Chiorini JA, Peck AB,
moter. Gene Ther. 2015;22:758–66.
Nguyen CQ. Gene therapy using il-27 ameliorates 49. Voutetakis A, Zheng C, Cotrim AP, Mineshiba F, Afione
sjogren’s syndrome-like autoimmune exocrinopathy. S, Roescher N, Swaim WD, Metzger M, Eckhaus MA,
Arthritis Res Ther. 2012;14:R172. Donahue RE, Dunbar CE, Chiorini JA, Baum BJ. Aav5-
43. Vosters JL, Roescher N, Illei GG, Chiorini JA, Tak mediated gene transfer to the parotid glands of non-
PP. Taci-fc gene therapy improves autoimmune sial- human primates. Gene Ther. 2010;17:50–60.
adenitis but not salivary gland function in non-obese 50. Li J, Zheng C, Zhang X, Liu X, Zhang C, Goldsmith
diabetic mice. Oral Dis. 2012;18:365–74. CM, Baum BJ, Wang S. Developing a convenient
44. Wu C, Wang Z, Zourelias L, Thakker H, Passineau large animal model for gene transfer to salivary glands
MJ. Il-17 sequestration via salivary gland gene ther- in vivo. J Gene Med. 2004;6:55–63.
apy in a mouse model of sjogren’s syndrome sup- 51. Li J, Shan Z, Ou G, Liu X, Zhang C, Baum BJ, Wang
presses disease-associated expression of the putative S. Structural and functional characteristics of irradia-
autoantigen klk1b22. Arthritis Res Ther. 2015;17:198. tion damage to parotid glands in the miniature pig. Int
45. Ice JA, Li H, Adrianto I, Lin PC, Kelly JA,
J Radiat Oncol Biol Phys. 2005;62:1510–6.
Montgomery CG, Lessard CJ, Moser KL. Genetics of 52. Baum BJ. Radiation-induced salivary hypofunction
sjogren’s syndrome in the genome-wide association may become a thing of the past. Oral Dis. 2016;
era. J Autoimmun. 2012;39:57–63. 22:81–4.
46. Burbelo PD, Ambatipudi K, Alevizos I. Genome-

wide association studies in sjogren’s syndrome: What
Surgical Management of Salivary
Gland Disease 13
Varun V. Varadarajan and Peter T. Dziegielewski
Abstract
The study of salivary gland tissue and the surgical management of salivary
gland pathology are fundamental to the practicing otolaryngologist-head
and neck surgeon. Traditional surgical intervention for both neoplastic and
nonneoplastic disease of the salivary glands includes sialadenectomy,
superficial or complete parotidectomy, minor procedures involving the
salivary ducts, and procedural interventions for xerostomia and sialorrhea.
Recent surgical advances of the salivary glands and ducts such as mini-
mally invasive, endoscopic, and robotic techniques have augmented the
surgeon’s armamentarium for managing salivary gland disease. Novel
techniques such as salivary gland transfer are also being pioneered. The
mechanisms of salivary gland function remain an active research topic,
and future applications may include regeneration of functional salivary
gland tissue. This chapter briefly reviews the basic surgical anatomy and
physiology of the major and minor salivary glands and describes tradi-
tional indications for surgical intervention. The recent advances in salivary
gland surgery are described, and the chapter concludes by highlighting
recent discoveries in the field of salivary gland regeneration. The implica-
tions of these advances for the head and neck surgeon and the potential
future of surgical management of salivary gland pathology are discussed.
13.1 I ntroduction and Historical

Perspective
The anatomic study of major salivary glands is

documented as early as the second century AD
V.V. Varadarajan, MD • P.T. Dziegielewski, when Galen described anatomic relationships
MD FRCS(C) (*) of the major salivary glands [1, 2]. Detailed
University of Florida Department of Otolaryngology,
anatomic depictions were not available in the
Gainesville, FL, USA
e-mail: [email protected]; western world until the fifteenth to sixteenth
[email protected] centuries when anatomists including Andreas

230 V.V. Varadarajan and P.T. Dziegielewski
Vesalius, Realdus Columbus, William Harvey, neoplastic and nonneoplastic disease and briefly
Bartholomaeus Eustachius, and others popular- discuss the most commonly described surgical
ized systematic human body dissection. Vesalius approaches to the major salivary glands. The
is the first anatomist to use the term “salivary recent advances in salivary gland surgery such
glands” and attribute their presence to the secre- as sialendoscopy, salivary gland transfer, and
tion of saliva in his writing De humani corporus minimally invasive surgery will then be dis-
fabrica in 1543. A more sophisticated anatomic cussed. The chapter will conclude by highlight-
understanding of the salivary glands was not ing recent discoveries in the field of functional
attained until the seventeenth century. Nicholas salivary gland regeneration and discuss the
Stenson first described the parotid gland duct in implications of these advances for the head and
his writing De glandulis oris et novis earandum neck surgeon.
vasis in 1661. Thomas Wharton is credited with
the discovery of the submandibular gland duct,
which he described in his writing Adenographia 13.1.1 Anatomic and Physiologic
sive glandularum totius corporis descriptio in Principles of Major and Minor
1656. In this writing he also describes what we Salivary Glands
believe to be the sublingual gland and ducts.
Caspar Bartholin further characterized the Salivary glands are accessory digestive glands
anatomy of the sublingual gland and duct sys- and begin their development during the 6th week
tem in 1685 [1]. Anatomists further character- of gestation. Epithelial buds invaginate from oral
ized the structural relationships, histology, and ectoderm into connective tissue mesenchyme.
physiologic function over the ensuing centuries. There are three paired major salivary glands
Traditional surgical interventions were devel- (parotid, submandibular, and sublingual glands)
oped to address a range of pathologies including and presumably 100s of minor salivary glands.
neoplastic, nonneoplastic, obstructive, inflam- The site of invagination defines the location of
matory, infectious, and iatrogenic conditions. the ductal orifice. Tunnels created by ectodermal
The twentieth century allowed further develop- outpouchings proliferate and branch, creating
ment and refinement of salivary gland surgery. tubules and acini that ultimately form the struc-
Janes was the first surgeon to describe a process ture of the salivary glands. The parotid gland
for the intraoperative identification of the facial develops first among the major salivary glands
nerve during parotid surgery in 1940 [3]. The followed by the submandibular and sublingual
development and widespread availability of com- glands. The parotid gland develops around the
puterized tomography and magnetic resonance branches of the facial nerve and is the last to
imaging allowed the medical and surgical com- become encapsulated by connective tissue fascia.
munity to a gain a sophisticated understanding The associated lymphatic vessels develop after
of the salivary gland anatomy. The structure and the submandibular and sublingual glands become
function of the salivary glands continue also to encapsulated but before parotid gland encapsula-
play an important role in regenerative medicine; tion [2–7]. The result of this unique aspect of
functional salivary gland regeneration is an active embryogenesis is the presence of lymphatic
research topic. Researchers aim to replicate and channels and lymph nodes within the parotid
regenerate the complex histologic organization gland.
and function of the human salivary glands in an
attempt to potentially allow salivary gland pres- 13.1.1.1 Parotid Gland and Facial
ervation and regrowth. Nerve Anatomy
This chapter will begin by providing an over- The parotid gland is the largest of the major sali-
view of the anatomic and physiologic principles vary glands and is located between the external
of the salivary glands. We will then review the auditory canal and the mandibular ramus. It is
indications for salivary gland surgery including classically described as wedge shaped and extends
13 Surgical Management of Salivary Gland Disease 231
superficially over the masseter muscle. The parotid vein provides venous drainage into the retro-
tail is a posterior and inferior extension into the mandibular vein. The embryological develop-
neck over the sternocleidomastoid muscle. The ment of the lymphatic tissues prior to parotid
parotid gland fascia encapsulates glandular tissue, gland encapsulation leads to the presence of
blood vessels, and lymphatic tissue. The parotid intraparotid and periparotid lymph nodes and
gland is bordered medially by the parapharyngeal lymphatic channels that drain the forehead,
space and medial pterygoid muscle, laterally by scalp, periorbital regions, auricles, and external
subcutaneous fat and dermis, superiorly by the auditory canals. Intraparotid lymph nodes also
zygomatic arch, inferiorly by the styloid process serve as lymphatic drainage to the posterior
and associated muscles and ligaments, anteriorly aspects of the nasopharynx and soft palate [2–6,
by the mandibular ramus and masseter muscle, 8]. This has clinical implications in head and
and posteriorly by the external auditory canal. The neck malignancy in the abovementioned sites
styloid process, stylohyoid muscle, and digastric with lymph node metastasis that may require
muscle separate the gland from the vessels and parotidectomy despite no primary salivary gland
nerves of the parapharyngeal space. The medial disease.
aspect of the gland, which contacts these struc- Associated nerves are the facial nerve and its
tures, is termed the “deep lobe” of the parotid branches, the auriculotemporal nerve, and the
gland although the gland is technically unilobular. great auricular nerve. The parasympathetic inner-
The facial nerve is considered by many to be the vation to the parotid gland stimulates saliva
dividing structure between the superficial and secretion. Preganglionic parasympathetic fibers
deep lobes of the parotid gland [3–6, 8]. originate from the inferior salivary nucleus and
The parotid gland duct, known as the Stensen’s travel along the glossopharyngeal nerve to the
duct (named after Nicholas Stenson), is 4–6 cm otic ganglion via the lesser superficial petrosal
in length and arises from the anterior aspect of nerve. The auriculotemporal nerve carries sensa-
the parotid gland [1, 5]. The Stensen’s duct trav- tion from the otic ganglion to the parotid gland.
els in the anterior direction lateral to the masseter The auriculotemporal nerve is a branch of the
muscle. The buccal branch of the facial nerve mandibular division of the trigeminal nerve; it
often travels parallel to the duct. The duct ulti- exits the skull base at foramen ovale and travels
mately makes a 90° medial turn (the “masseteric anteriorly and laterally from the skull base and
bend”) to pierce the buccinator muscle and opens infratemporal fossa to the external auditory canal.
into the buccal mucosa at the level of the second Sympathetic stimulation originates from the
maxillary molar tooth. In 21 % of the human superior cervical ganglion; postganglionic fibers
population, accessory parotid tissue is found in travel to the parotid gland via the external carotid
proximity to the duct and ductal orifice [4, 9, 10]. artery [2, 4–6].
The connective tissue fascia that encapsulates The great auricular nerve originates from cer-
the parotid gland is contiguous with the superfi- vical rootlets C2–C3 and is a branch of the cervi-
cial layer of the deep cervical fascia. The fascia cal plexus. This nerve branches from the cervical
sends septations into the parotid tissue. The plexus at Erb’s point and courses superiorly from
parotid gland is separated from the submandibu- the posterior aspect of the sternocleidomastoid to
lar gland by the stylomandibular ligament, which the superficial aspect of the parotid gland. The
is a continuation of the fascia of the posterior great auricular nerve supplies sensation to the
belly of the digastric muscle. The gland has skin overlying the parotid gland, the mastoid and
fibrous attachments to the anterior wall of the mandibular angle, and the inferior and posterior
external auditory canal, mastoid process, and the aspects of the auricle. This nerve may be sacri-
fascia of the sternocleidomastoid [5]. ficed during a parotidectomy [3–6, 8].
The transverse facial artery branch of the The facial nerve is intimately associated with
external carotid artery serves as the arterial sup- the parotid gland tissue, and a discussion of the
ply to the parotid gland, and the transverse facial surgical anatomy of the parotid gland is incom-
plete without describing the course of the facial the submandibular triangle in the neck. The
nerve. The main trunk of the nerve exits the sty- gland is associated with neck level IB lymph
lomastoid foramen and provides branches to the nodes and extends medial and deep to the infe-
posterior belly of the digastric muscle, posterior rior border of the posterior mandibular body.
auricular muscle, and the stylohyoid muscle The gland curves over the posterior border of the
before entering the parotid gland. The nerve mylohyoid muscle, which anatomically divides
enters the gland approximately 1 cm after exiting the gland into two lobes. The superficial lobe is
the temporal bone [4–6]. At this point, the nerve located in the posterolateral sublingual space,
divides into superior temporofacial and inferior while the larger deep lobe is located inferior to
cervicofacial divisions at the pes anserinus; the mylohyoid muscle. Like the parotid gland,
13.3 % of patients have three divisions [11]. The the fibrous encapsulation of the submandibular
five terminal branches of the nerve from superior gland derives from the superficial layer of the
to inferior are the temporal, zygomatic, buccal, deep cervical fascia [2–6, 13]. The submandibu-
marginal mandibular, and cervical branches. lar duct is termed the Wharton’s duct (named
Communicating branches between these terminal after anatomist Thomas Wharton) [1]. The duct
branches are very common, and the terminal extends from the medial aspect of the gland and
branching is variable. Identification of the facial extends anteriorly to open into the oral cavity
nerve is critical in parotidectomy. Several classic lateral to the lingual frenulum. The duct courses
anatomic relationships have been used to localize between the mylohyoid and hyoglossus muscles.
the main trunk of the facial nerve. The “tragal The opening is at the apex or on the walls of the
pointer” is a deep extension of conchal cartilage papilla on the anterior floor of mouth. The duct is
that is an anatomic landmark; the nerve is located approximately 5 cm in length and is between 0.5
1 cm inferior and medial to the tragal pointer [4, and 1.5 mm in diameter [4, 14]. The lingual
9, 12]. The nerve is located posterior and lateral nerve curves around the inferior border of the
to the base of the styloid process. The main trunk duct from a lateral to anteromedial direction to
is also located 6–8 mm deep to the tympanomas- provide sensory innervation to the anterior
toid suture line of the temporal bone exiting the 2/3rds of the tongue. The arterial supply of the
stylomastoid foramen. The nerve is also located submandibular gland is via the glandular branch
superior and deep to the proximal attachment of of the facial artery branch of the external carotid
the posterior belly of the digastric muscle. Facial artery. The facial artery travels deep to the digas-
nerve dissection is discussed elsewhere in this tric and stylohyoid muscle to pass into a groove
text. The mastoid cortex and air cells can be on the posterior and deep surface of the gland.
removed to identify the facial canal if the above The artery courses both anteriorly and superiorly
methods do not allow identification of the nerve. to the superior aspect of the gland until it curves
Anterograde dissection and further skeletoniza- over the facial notch of the mandibular body to
tion of the nerve starting from the main trunk then ascend over the lateral aspect of the man-
allow safe removal of parotid gland tissue. If a dibular body anterior to the masseter muscle.
distal branch is found before the main trunk, ret- Venous drainage is provided by the facial vein
rograde dissection can also be performed to trace which travels superficial to the submandibular
the nerve to the main trunk [3, 4, 6]. Further gland and drains into the common facial vein [2,
description of parotidectomy is described later in 4–6]. Figure 13.1 depicts the anatomic relation-
this chapter. ships of the structures to the submandibular
gland.[4].
13.1.1.2 Submandibular Gland Like the parotid gland, the sympathetic
Anatomy innervation to the submandibular gland is pro-
The submandibular gland is the second largest vided by postganglionic sympathetic fibers origi-
paired major salivary gland and is located within nating from the superior cervical ganglion, which
Lingual nerve Inferior alveolar

nerve
Submandibular
ganglion
Parotid gland
Submandibular
gland
Sublingual gland
Mylohyoid muscle
Submandibular duct
Fig. 13.1 Anatomic relationships of the submandibular gland to its adjacent structures (Reprinted from Ref. 4)
travel along the external carotid artery branches. 13.1.1.3 Sublingual Gland Anatomy
The parasympathetic innervation originates in The sublingual gland is the smallest of the paired
the superior salivatory nucleus, and travels down major salivary glands. This gland is located in the
the facial nerve via the nervus intermedius and sublingual space in between the mylohyoid mus-
ultimately joins the lingual nerve via the chorda cle and the oral cavity floor mucosa. The genio-
tympani nerve to synapse in the submandibular glossus muscle is medial to the gland, while the
ganglion. Postganglionic fibers synapse onto mandible is lateral to the gland. The Wharton’s
glandular cells [2, 4]. duct also travels within this space along with the
The marginal mandibular branch of the facial terminal branches of the lingual and hypoglossal
nerve is closely associated with the submandibu- nerves. The sublingual gland is located laterally
lar gland and is often found coursing anteriorly to these structures. This gland is approximately
within 1–2 cm of the angle of the mandible [3–6]. 3 cm in length and oval in shape and has no
The nerve loops below the mandible and has a fibrous capsule. The sublingual gland may have a
variable course and superior-inferior position to major drainage duct (Bartholin duct) and minor
the inferior border of the mandible. The facial drainage duct but drains into the oral cavity along
vein is deep to this nerve; the vein can be ligated the sublingual fold via 8–20 small ducts termed
and reflected superiorly from the gland dur- the ducts of Rivinus [3–6].
ing submandibular gland surgery to protect the The arterial supply of the sublingual gland is
nerve. This maneuver has been termed the Hayes- mainly from the sublingual branches from the lin-
Martin maneuver after the well-known head and gual branch of the external carotid artery. There
neck surgeon [3, 13]. Several surgical approaches are also branches from the submental branch of
have been described for submandibular gland the facial artery. The lingual and facial veins pro-
surgery including transcervical, submental, retro- vide venous drainage to the sublingual gland. The
auricular, or intraoral approaches [3]. The lateral sympathetic and parasympathetic innervation to
cervical approach is most often described and the gland is similar to the submandibular gland as
allows direct access to the gland; this technique described above. Postganglionic parasympathetic
is again described later in this chapter. nerves originate in the submandibular ganglion.
Fig. 13.2 Basic Acinus

salivary gland unit Intercalated Striated Excretory
(Reprinted from Ref. 8) duct duct duct
Myoepithelial cell
The lingual gland can be surgically approached in face. The acinus expresses saliva into the secre-
a transoral fashion with direct incision into floor tory duct, which consists of intercalated and
of mouth into the sublingual space [3, 5, 6]. striated duct. Myoepithelial cells also surround
the intercalated ducts. The intercalated ducts
13.1.1.4 Minor Salivary Glands consist of cuboidal shaped cells and continue as
There are 100–1000 minor salivary glands that striated ducts, which contain columnar cells with
are distributed throughout the oral cavity, oro- microvilli on their luminal surface. The acinus
pharynx, larynx, tracheobronchial tree, and nasal and these proximal ductal components are
cavity. The arterial supply, venous and lymphatic together considered the secretory end piece and
drainage, and innervation depend on the ana- are organized into lobules [2]. These ducts drain
tomic location of the minor salivary glands. into excretory and collecting ducts, which con-
Postganglionic fibers from the submandibular sist of a bicellular layer (apical flat epithelial
gland innervate the minor salivary glands of the cells and basal columnar cells) and lie outside of
inferior oral cavity and oropharynx. Palatine the lobules. This structural organization varies
nerves supply postganglionic fibers from the between glands. Figure 13.2 depicts the basic
pterygopalatine ganglion to the superior oral cav- salivary gland unit [8].
ity and palate [2, 4]. The ducts of Rivinus are the collecting ducts
of the sublingual gland, while the Stenson and
Wharton’s ducts are the terminal collecting ducts
13.1.2 Salivary Gland Physiology of the parotid and submandibular glands, respec-
tively. The ducts serve as transport conduits
The basic histologic architecture of all salivary while also modifying saliva composition. The
glands consists of a branching duct system that medullary brainstem salivary center is a major
terminates at the salivary acini. The acinus is the central neural control center for salivation; how-
site of production of saliva and is surrounded and ever, there are multiple other stimuli for saliva
supported by myoepithelial cells which contract secretion including taste and olfaction and the
to express saliva into the ducts, myofibroblasts, mechanical act of mastication. Salivation can be
extracellular matrix and stromal cells, immune increased or decreased as a side effect of medi-
cells, vascular endothelial cells, and nerve cells. cations and can be affected by systemic medical
The acinus contains many acinar cells that pro- conditions [2, 4].
duce saliva into the acinar lumen [2, 4, 7]. Acinar The average human produces between 1 and
cells are bipolar, pyramidal shaped cells which 1.5 L of saliva daily. The minimal human sali-
secrete fluid and proteins from their apical sur- vary flow rate is at least 0.1 mL per min when
unstimulated and at least 0.2 mL/min when secretion. Potassium is unaffected by flow rates
stimulated although the average range of salivary [2, 4, 15, 17].
flow rates are 0.3 mL/min when unstimulated to
7 mL/min as the maximum stimulated flow rate
[2, 4, 15, 16]. The salivary glands have differ- 13.2 Traditional Surgical
ent viscosity of saliva that reflects the histologic Interventions
subtype of acinar cells within its lobules. The
parotid gland consists of mostly serous subtype 13.2.1 Nonneoplastic
acini and secretes watery saliva. ~25 % of the and Inflammatory Salivary
daily saliva production is from the parotid gland. Gland Disease
The sublingual gland and minor salivary glands
consist of mostly mucous acini and secrete vis- Nonneoplastic salivary gland diseases include
cous saliva. These glands together comprise of wide differential diagnosis including infectious,
2–4 % of the daily saliva production. The sub- inflammatory, obstructive, traumatic, and
mandibular gland acini are a mixture of serous radiation-induced etiologies. These disease pro-
and mucous types, and therefore, the gland cesses more commonly involve the major sali-
secretes an intermediate viscosity saliva which vary glands and may involve either the salivary
contributes ~70 % of the daily saliva production gland parenchyma or the ducts. Some conditions
[2, 17]. Mucinous cells are found surrounding may be a condition of a systemic disease process.
the lumen of the acini while serous acinar cells Presentation may be acute, chronic, or recurrent
are organized at the end of the acinus to form and may be present in both adult and pediatric
a serous demilune [2]. Viscosity is also affected populations.
by the stimulating factor for saliva production.
Parasympathetic innervation stimulates a less 13.2.1.1 Acute Suppurative
viscous and watery type of saliva while sympa- Sialadenitis
thetic stimulation produces a thick, low volume, Acute suppurative sialadenitis is a condition in
viscous saliva. Parasympathetic innervation which retrograde bacterial contamination of the
uses the neurotransmitter acetylcholine binding salivary ducts from the microflora of the oral
to muscarinic receptors to stimulate salivation. cavity causes an acute infection of the ducts and
Sympathetic stimulation uses the neurotransmit- salivary gland. The submandibular gland is the
ter norepinephrine binding to adrenergic recep- most common (the original description is correct
tors [4, 8, 15]. as far as I know. In addition, the following sen-
Saliva consists of electrolytes, proteins, and tences make more sense if the parotid gland is
other molecules. The acinus generates the fluid listed as the most common gland) salivary gland
component of saliva in the form of an isotonic affected by bacterial sialadenitis due to increased
solution. Sodium and chloride ions are resorbed viscosity and the decreased concentration of
in the proximal ductal network, while potas- the antibacterial lysosomes, IgA antibodies,
sium and bicarbonate ions are secreted. and sialic acid in parotid gland vs. the subman-
Electrolyte reabsorption and secretion involves dibular, sublingual, and minor salivary glands.
active transport processes. The majority of the Submandibular and sublingual gland saliva also
protein component is secreted at the level of the contains glycoproteins that have been shown to
acinus; however, the secretory duct also con- competitively inhibit bacterial attachment on the
tributes protein molecules. The end product is a epithelium of salivary ducts [18–21]. Suppurative
hypotonic solution with pH 6–7. Salivary flow bacterial sialadenitis has been associated with
rates also affect electrolyte composition as patients undergoing major abdominal or hip sur-
slower flow rates allow more time for sodium gery in their postoperative hospital course. These
and chloride resorption. However, increased infections may also be associated with a preexist-
flow rates stimulate increased bicarbonate ing malignancy or head and neck infection [19,
22]. The inciting events to a bacterial infection of imaging (e.g. CT scan) is helpful to confirm loca-
the salivary glands are reduced flow and salivary tion and extent of the purulence collection before
stasis due to obstruction due to a sialolith, foreign an incision and drainage is performed. Image-
body, injury, or other factors that reduce flow. guided drainage with CT or ultrasound is a mini-
Predisposing conditions include dehydration, mally invasive method but may not be an option
periodontal disease, immunodeficiency, diabetes in some institutions. If surgical drainage is
mellitus, neurodegenerative disease, systemic required for a parotid abscess, a modified Blair
autoimmune conditions, cystic fibrosis, radiation incision can be used for exposure, and blunt dis-
injury, chemotherapy, and medications that cause section toward the abscess is oriented in the
reduced salivary flow as a side effect [18–20]. direction of the facial nerve branches to avoid
The clinical presentation begins with pain and injury. The parotid fascia must be incised parallel
rapid, diffuse enlargement of the salivary gland. to the facial nerve branches. For a submandibular
Palpation reveals tenderness, warmth, and indu- abscess, the transcervical approach is the most
ration. A stone may be identified with bimanual direct method; the marginal mandibular nerve
palpation and may be the predisposing factor for must either be protected with the Hayes-Martin
recurrent infections. Purulence may be expressed maneuver or identified and protected. A surgical
from the papilla of the involved by pressing on the drain may be placed [18].
gland and sent for culture. Polymicrobial infections
are common; however, Staphylococcus aureus 13.2.1.2 Viral Sialadenitis
is reported to be the most causative organism. Viral sialadenitis is similar to bacterial sialad-
Streptococcus pneumonia, Streptococcus pyogenes, enitis and is thought to develop more commonly
Streptococcus viridans, Haemophilus influenzae, by the hematogenous route more often than ret-
and Escherichia coli are other aerobic organisms rograde ductal migration. Mumps is the most
that have been cultured. Anaerobic organisms common form of viral sialadenitis. The infec-
responsible for infections include Bacteroides tious agent is paramyxovirus which typically
species, Peptostreptococcus, Prevotella species, affects the parotid gland. The infection histori-
Fusobacterium species, and Burkholderia pseudo- cally occurred most frequently in children and
mallei [18, 19, 21]. Computerized tomography (CT) the incidence has decreased after routine vacci-
and ultrasound are used to evaluate for an abscess nation. Recently, an increasing number of young
or a sialolith. Sialography is contraindicated due to adults are being diagnosed with the infection [18,
the risk of exacerbation of the infection. 19]. Viral sialadenitis is nonsuppurative unless a
The treatment of acute bacterial sialadenitis bacterial superinfection occurs. After infection
consists of antibiotics, frequent gland massage, via the respiratory tract, the virus enters an incu-
sialogogues, hydration, electrolyte repletion, and bation period of several weeks. The clinical pre-
the application of heat packs. The causative fac- sentation includes a nonspecific viral prodrome
tor must be addressed if identified. Medications of fever, myalgia, and malaise. Salivary gland
that reduce salivary flow must be discontinued. enlargement presents within the first week and
Broad-spectrum antibiotic therapy directed is often bilateral. Other manifestations of mumps
against gram-positive organisms and anaerobes include orchitis, myocarditis, and aseptic men-
can be narrowed once culture results are avail- ingitis. The virus may rarely cause sensorineu-
able. Sialendoscopy is contraindicated during an ral hearing loss. Acute viral sialadenitis (usually
acute infection but can address sialoliths or other parotitis) can also be caused by cytomegalovirus,
obstructive etiologies after the infection is coxsackie viruses A and B, lymphocytic cho-
treated; this technique is discussed in another riomeningitis virus, enteric cytopathic human
section of this chapter. If treatment does not orphan virus, and influenza virus. The subman-
improve symptoms within 2–3 days, an abscess dibular gland is rarely involved. Diagnosis can be
or antibiotic resistance must be considered [18– made with viral serology or isolation of the virus
20, 23]. In the event of an abscess, diagnostic through cerebrospinal fluid. Symptomatic treat-
ment is usually sufficient for acute viral sialad- species). Strictures and ductal abnormalities caus-
enitis, and surgery is rarely indicated [18, 19].. ing obstruction can develop [18, 19, 25, 26].
Human immunodeficiency virus (HIV) can Treatment is similar to acute sialadenitis and con-
cause diffuse enlargement of the salivary glands sists of gland massage, sialogogues, hydration,
(most often the parotid gland), which is referred and application of heat. Antistaphylococcal anti-
to as HIV-associated salivary gland disease (HIV- biotic therapy can be started after a culture is
SGD). Associated symptoms can include xerosto- obtained from the parotid duct. Conservative
mia and lymphadenopathy. Treatment includes treatment is almost always sufficient although
antiviral drugs, sialogogues, and oral hygiene. parotidectomy can be considered in refractory
HIV can also predispose patients to benign cases. Ductal ligation and tympanic neurectomy
lymphoepithelial cysts, Kaposi sarcoma, and have also been described; however, these are
lymphoma of the salivary glands. The lympho- rarely performed [19, 25, 26]. Angioplasty bal-
epithelial cysts can usually be managemed with loon catheters have also been used for stricture
needle aspiration for symptomatic but temporary dilations by interventional radiology under fluo-
relief, or sclerotherapy. Surgical intervention with roscopic control [25, 27–29]. Sialendoscopic
extracapsular dissection or superficial parotidec- techniques may be used to address strictures; this
tomy is reserved for refractory disease [18, 19]. technique is discussed later in this chapter.
13.2.1.3 Sialadenitis in the Pediatric 13.2.1.4 Chronic Sialadenitis

Population Chronic sialadenitis is a condition in which there
Neonatal suppurative parotitis is an uncommon are recurrent episodes of inflammation and pain
but reported condition that occurs most often in in the major salivary glands; the parotid gland is
male and preterm neonates; dehydration appears the most commonly involved gland. Symptoms
to be the inciting factor. Clinical presentation are worse with eating. Similar to acute sialadeni-
consists of fever, irritability, anorexia, failure to tis, salivary stasis, reduced salivary flow rates,
thrive, gland swelling, and erythema of the over- ductal obstruction (with a sialolith or other for-
lying skin. Bilateral glands may be involved. eign body), systemic disease, or dehydration are
Infection may originate from oral flora or hema- possible predisposing factors. Sialolithiasis is the
togenous dissemination of bacteria. A number of most common cause [18, 19]. Repeated episodes
pathogens can be responsible, S. aureus being the of acute sialadenitis cause permanent structural
most common. E. coli, Pseudomonas aerugi- changes including acinar destruction, ductal ecta-
nosa, and group B Streptococcus species have sia, and fibrosis. the gland becomes enlarged
been reported. Antibiotics are the mainstay of exacerbations, and saliva is difficult to express
therapy with drainage or surgical intervention for from the duct. Xerostomia and change in salivary
refractory cases [18, 19, 24]. content (altered electrolyte composition,
Juvenile recurrent parotitis is a nonsuppurative increased immunoglobulins with IgG predomi-
inflammatory condition in which the parotid nance, albumin, transferrin, increased lysozyme
gland periodically enlarges with associated ten- concentrations) develop in long-standing disease
derness, fever, and malaise. It is the most common [18, 19]. Gland atrophy can occur, and firm,
salivary gland disease of childhood after mumps. fibrotic areas of the gland may be palpated. These
The condition may be unilateral or less commonly firm areas must be ruled out for malignancy.
bilateral. The peak incidence is between 3 and Ductal strictures can form and cause obstruction.
6 years of age. Episodes occur every 3–4 months. CT and ultrasound can help to further character-
The etiology of juvenile recurrent parotitis is ize gland structure and identify non-palpable
unclear, and multiple etiologies have been pro- sialoliths, while sialography (traditional and MRI
posed, including congenital duct malformation sialography) can characterize ductal architecture.
(ectasia), immunologic deficiencies, and infec- Treatment includes massage, sialogogues, heat,
tious causes (Staphylococcus and Streptococcus and hydration. Several procedural interventions
have been described for symptomatic manage- inorganic substance that allows salt precipitation
ment including ductal papilla dilation, sialo- in the setting of salivary stasis or reduced flow [18,
dochoplasty, ductal steroid injection, ductal 19, 34]. Due to the more alkaline and viscous
ligation, and tympanic neurectomy [18, 19]. properties of the submandibular gland saliva, the
Interventional radiology techniques to dilate duc- submandibular duct is reported to be the most sus-
tal strictures under fluoroscopy have been ceptible to sialolith formation. The duct is also
reported [27–29]. Surgical extirpation of the long and saliva flows against the force of gravity.
gland can be considered when all other treatment Calculi occur in chronic sialadenitis patients as
modalities fail to sufficiently relieve symptoms. well as patients with gout. Calculi may be the pre-
Sialendoscopy (described below) is a developing disposing factor in acute suppurative sialadenitis.
treatment modality that can delay or prevent the Symptoms include postprandial pain and swelling
need for open surgical intervention. as well as a history of acute suppurative sialadeni-
Benign lymphoepithelial lesions (LE lesions) tis [18, 19].
can develop in the setting of long-standing chronic Sialography, CT, and ultrasound and MRI sia-
disease. LE lesions are characterized by a lympho- lography can be used for diagnosis although cal-
plasmacytic infiltrate, acinar atrophy, and ductal culi less than 2 mm may be missed by imaging
metaplasia leading to the development of epimyo- [18]. Plain films are more useful for submandibu-
epithelial islands [19, 30]. This condition is well lar stones, which are usually radio-opaque unlike
described in patients with Sjögren’s syndrome and parotid stones. Virtual MRI endoscopy is a new
has been termed Mikulicz’s disease. LE lesions are modification of MRI that allows a three-
usually asymptomatic enlargements unless they dimensional endoscopic view of the ductal sys-
become infected which may require drainage or tem. Treatment may be conservative and consists
surgical removal. Kuttner’s tumor (chronic scle- of gland massage, sialogogues, hydration, and
rosing sialadenitis) is a similar process occurring observation for spontaneous passage. This is
in the submandibular gland characterized by a often successful for small (<2 mm) sialoliths [18,
firm, painless swelling associated with areas of 19, 34, 35]. Procedural interventions are consid-
gland atrophy. Kuttner’s tumor differs histologi- ered for refractory cases; the best approach
cally (lymphoid infiltrate and discrete tubular depends on the location, size, and shape of the
structures with regularly aligned nuclei) from LE sialolith. Transoral sialolith removal can be
lesions. Patients with benign LE lesions and attempted; however, gland extirpation may be
Kuttner’s tumor must be monitored for develop- required. Stenson’s duct calculi can be approached
ment of ductal carcinoma [18, 19, 30, 31]. transorally if the calculus is medial to the masse-
ter muscle. Shockwave lithotripsy and sialendos-
13.2.1.5 Sialolithiasis copy (discussed below) are being increasingly
Sialolithiasis is the development of calculi in the used. Combined approaches with endoscopy and
salivary gland ductal system. Sialolithiasis either transoral or external approaches have been
accounts for 50 % of major salivary gland diseases shown to be successful [18, 19]. Interventional
[27, 32]. They occur most frequently in the sub- radiology techniques under fluoroscopy have
mandibular gland (80 %) followed by the parotid been described as well; the first sialolith removed
gland (20 %) and sublingual gland (1 %) [18, 19, via basket under fluoroscopy was reported by
33]. Minor salivary gland stones are rare and are Kelly in 1991 [29]. Coronary angioplasty bal-
most often in the upper lip or buccal mucosal loon, embolectomy catheters, and wire loop
glands. Sialolithiasis occurs more frequently in snares have also been used to remove stones.
men. The calculi are composed of c alcium phos- Capaccio’s literature review revealed that fluoro-
phate and calcium carbonate and are mixed with scopic guided sialolith removal was reasonable
organic molecules including glycoproteins, muco- for mobile stones in proximal and middle sub-
polysaccharides, and cellular debris [19, 33]. The mandibular ductal system as well as parotid
nidus for calculi development is believed to be an stones [27–29] .
13.2.1.6 Granulomatous Diseases food products (including milk) and domestic or

Granulomatous diseases of the head and neck wild animals. Clinical presentation is classically
may involve the salivary glands and the lym- described as a neck mass with rapid enlargement,
phatic networks associated with the glands. violaceous overlying skin changes, and resis-
Granulomatous infections can invade salivary tance to initial antibiotic therapy. Cervical lymph-
gland parenchyma in advanced cases. The most adenopathy is common. The infection may
commonly discussed granulomatous infectious progress to an abscess that may spontaneously
diseases of the head and neck are tuberculous and drain and form a sinus tract. Diagnosis with FNA
nontuberculous mycobacterial disease, cat biopsy is controversial and carries the risk of fis-
scratch disease, toxoplasmosis, and actinomyco- tula tract formation. Cultures take up to 6 weeks
sis. Noninfectious granulomatous disease to result and may be negative. Antibiotic therapy
includes sarcoidosis and Sjögren’s syndrome. may also require weeks to months of treatment
Mycobacterium tuberculosis is the pathogen and may not be effective. Complete gland exci-
associated with tuberculosis, which can manifest sion is therefore considered and can serve as
as cervicofacial lymphadenopathy. Although sali- definitive treatment. If the parotid gland is
vary gland involvement is rare, it is reported in involved, superficial and/or deep parotidectomy
immigrants from underdeveloped countries as with facial nerve preservation must be performed
well as immunocompromised patients. Infection [18, 19].
can be primary by way of ductal migration from Cat scratch disease is a local infection that
the oral or oropharyngeal saliva or lymphoid tis- originates at the scratch site with ensuing granu-
sue or can be secondary with either lymphatic or lomatous lymphadenitis in the draining lymph
hematogenous spread. The intraglandular lymph nodes. Bartonella henselae is the pathogen and is
nodes of the parotid gland may become sites of a gram-negative bacillus that is usually spread to
latent infection. The parotid gland is the most the skin laceration from the scratch or bite of a
common gland affected. Submandibular gland household cat. The upper extremity is the most
infection is more common in systemic and dis- common site of infection followed by the head
seminated tuberculosis. The infection can present and neck. Head and neck infection can involve
as an inflammatory lesion that mimics sialadeni- the lymph nodes associated with the parotid
tis or can present as a mass that masquerades as a gland or the submandibular gland. The infection
neoplasm. Diagnosis involves purified protein starts as a pustule at the site of scratch or bite and
derivative (PPD) skin test, chest x-ray, and fine progresses to local and regional l ymphadenopathy
needle aspiration (FNA) of lesions. FNA cytol- over 1–2 weeks. Erythema and lymphadenitis
ogy may reveal characteristic granulomatous frequently develops and may progress to abscess
inflammation with epithelioid histiocytes. formation with spontaneous drainage. Antibody
Samples may be sent for acid fast staining. In detection for B. henselae or PCR detection is
cases in which the diagnosis is uncertain or is used for diagnosis. Bacilli may be visible in tis-
resistant to antibacterial therapy, the involved sue specimens with Warthin-Starry silver stain-
glands are excised [18–20]. ing. Culture requires 6 weeks due to the slow
PPD skin test may be negative in non- growth of the organism. The infection is usually
tuberculosis mycobacterium (NTM) infections self-limiting, and antibiotic therapy is reserved
that more commonly present with cervicofacial for patients with advanced or systemic spread of
lymphadenopathy. These infections are usually disease. Surgical excision, like tuberculous disease,
localized without systemic signs or symptoms. is reserved for infections that fail to resolve.
The most common NTM infections are caused by Resolution may take several months. Parinaud’s
M. kansasii, M. scrofulaceum, M. avium- oculoglandular syndrome is an atypical presenta-
intracellulare, and M. bovis. These infections are tion of cat scratch disease characterized by uni-
encountered in children less than 5 years of age, lateral granulomatous conjunctivitis with
and the pathogens are carried in soil, water, and ipsilateral cervicofacial or salivary gland lymph
node involvement. Parotid involvement with with another autoimmune disorder). The disease is
facial nerve palsy has been reported [18, 19]. more commonly seen in women during the fourth
Toxoplasmosis is caused by the organism and fifth decade of life. Exam reveals xerostomia,
Toxoplasma gondii and rarely involves the sali- dental caries, and possible oral candidiasis.
vary glands. Domestic cats are the host for this Systemic manifestations include arthritis, pneumo-
organism. The pathogen is transferred through nitis, skin rash, myositis, and other complaints.
ingestion of infected meat or through cat feces. Ocular exam may reveal decreased tear secretion
Hematogenous dissemination can spread the dis- (may be evaluated with Schirmer test), lacrimal
ease to the intraparotid lymph nodes or the perip- gland enlargement, enlarged conjunctival vessels,
arotid lymph nodes. Antibiotic therapy is usually corneal damage, and pericorneal injection. These
sufficient even in advanced cases; surgery is findings are characteristic of keratoconjunctivitis
reserved for large suppurative lesions [18, 19]. sicca. The Imaging by CT or MRI can reveal calci-
Actinomycosis is caused by the organism fication in involved salivary glands. Sialography
Actinomycosis species (A. israelii, A. bovis, and A. may reveal sialectasis. Histopathology reveals lym-
naeslundii), a gram-positive, nonacid fast bacilli. phocytic infiltration of gland tissue starting with
The microscopic appearance is similar to mycobac- the ducts and progressing to destroy and replace
teria and fungi given the branching, filamentous acinar tissue, which in turn reduces the salivary
appearance. A. israelii is commonly found as part of gland function. Laboratory tests include the detec-
the oropharyngeal lymphoid tissue flora and in cari- tion of autoantibodies against RNA/protein com-
ous dentition [19]. Cervicofacial infection is the plexes Ro (SS-A) and La (SS-B) in addition to
most common presentation and is caused by inva- rheumatoid factor (RF) and ANA (antinuclear anti-
sion of the organism after trauma or poor oral body). SS patients are also at increased risk for
hygiene. Retrograde ductal migration may explain developing lymphoma. Diagnosis is aided by
salivary gland infection although direct invasion biopsy of a minor salivary gland of the labial
into parotid or submandibular gland tissue is also mucosa. Biopsy may be performed in the clinic set-
possible. Infection of the salivary gland is character- ting: several lobes of minor salivary gland tissue
ized by painless enlargement of the gland with must be sampled (collected, or biopsied, if authors
chronic purulent drainage. The disease can progress like) and examined. Diagnostic criteria have been
to form cutaneous drainage tracts. Fibrotic changes established and involve the presence of signs and
and soft tissue destruction cause induration of the symptoms of keratoconjunctivitis sicca, symptoms
gland upon palpation. Microscopic examination of of xerostomia and signs of decreased salivary gland
tissue samples or swabs reveals the characteristic function, salivary gland biopsy results, and pres-
sulfur granules in the presence of branching fila- ence of Ro and La antibodies. The presence of
mentous, gram-positive rods. Long term (minimum another autoimmune disorder such as systemic
6 weeks) antibiotic therapy is sufficient for limited lupus erythematous or rheumatoid arthritis sug-
disease, but surgical extirpation is required in the gests secondary SS. Treatment includes symptom-
presence of fistulous tracts and for cases refractory atic treatment to protect the eyes and the teeth with
to antibiotics [18–20]. eye lubricants, eye patches, saliva substitutes, den-
tal care and oral hygiene, and pilocarpine [18, 19,
13.2.1.7 Noninfectious Inflammatory 30, 36].
Disease Sarcoidosis is a granulomatous disorder with a
Sjögren’s syndrome (SS) is an autoimmune disor- wide range of systemic manifestations involving
der characterized by autoimmune destruction of multiple organ systems. Common presenting
exocrine glands. B- and T-cell-mediated damage symptoms include cough, dyspnea, weight loss,
causes symptoms including xerostomia, dry eyes erythema nodosum, arthralgias, and myalgias.
(foreign body sensation in the eye), dysphagia, and Salivary gland involvement is rare but presents as
enlargement of the salivary glands. Sjögren’s syn- gland swelling. Uveoparotid fever is a manifesta-
drome can be primary or secondary (associated tion of sarcoidosis characterized by uveitis, parotid
gland enlargement, and facial paralysis. The onstrate secretion of saliva in the wound. If the
parotid gland enlargement can last months but is duct is transected, a salivary stent or catheter is
self-limited. Minor salivary gland biopsy, as in SS, placed in the duct, and primary end-to-end anas-
may aid diagnosis. Corticosteroids are used for tomosis is performed. The catheter or stent is left
treatment in uveoparotid fever and are effective for in place to allow healing for 2 weeks [18]. The
resolution of the facial paralysis [18, 19, 37]. duct may also be rerouted and sutured into the
oral cavity. Serial dilations may be required after
13.2.1.8 Radiation-Induced repair to prevent strictures and stenosis. Salivary
Sialadenitis gland parenchyma lacerations can be closed pri-
Radiation-induced xerostomia is a well-known marily with interrupted sutures. Sialoceles or a
complication of radiotherapy for head and neck salivary cutaneous fistula may develop shortly
cancer. Radiation dosages greater than 20–30 Gy after the repair. Serial drainage and a pressure
predispose glands to lipid peroxidase injury, dressing can conservatively manage these condi-
enzyme spillage, and cell lysis [38]. Injury begins tions. Botox injections have been used to decrease
with an acute inflammatory reaction that leads to the salivary production and allow fistula resolu-
acinus destruction with continued irradiation. tion. If the fistula fails to resolve, ductal injury
Strictures and kinks can form in the ducts and can must be suspected [41]. The ductal system can be
cause duct obstruction. Increased incidence of evaluated with sialography or MRI sialography.
pleomorphic adenomas and malignant salivary If the above management fails to resolve the fis-
gland neoplasms have been reported in patients tula, gland excision can be considered [18, 42].
with radiation exposure. Ductal injuries are less common in the subman-
Iodine-131 treatment for thyroid malignancy dibular and sublingual glands; however, the
may cause dose-dependent sialadenitis. The approach to repair is similar as described above.
sodium-potassium-chloride transporter in salivary Patients with penetrating facial trauma must
gland tissue concentrates radioactive iodine to lev- also be assessed for facial nerve injury. Physical
els that can cause parenchymal damage. The exam may reveal weakness or complete paraly-
parotid glands are most commonly affected fol- sis. Nerve stimulation can further characterize
lowed by the submandibular glands. Sialendoscopy the injury. Facial nerve injuries that lie posterior
(described below) has revealed ductal stenosis, to a line drawn from the lateral canthus to the
mucous plugs, and other findings of chronic mental foramen must be repaired immediately. If
inflammation. Sialendoscopy has been used to the injury lies anterior to this line, the injury can
provide symptomatic relief by allowing ductal irri- likely be observed for recovery [18].
gation and/or steroid instillation [18, 39, 40].
13.2.1.10 Cysts and Ranula
13.2.1.9 Trauma Cystic lesions of the salivary glands occur most
Traumatic injury to the salivary glands, ducts, or often in the parotid gland and may be congenital or
associated nerves requires surgical exploration acquired. Congenital cysts include dermoid cysts,
and repair. Penetrating or laceration injuries to branchial cleft cysts (typically first branchial cleft
the parotid gland place the duct and facial nerve cyst), and congenital duct cysts. Dermoid cysts
at risk. Blunt trauma may cause hematomas that consist of keratinizing squamous epithelium with
require drainage to prevent fibrosis or superinfec- associated dermal appendages; these cysts must be
tion. Penetrating injuries posterior to the anterior completely excised [18, 43]. First branchial cleft
border of the masseter muscle must be evaluated cysts are rare and typically present within the
for ductal injury due to the proximity of the duct parotid gland. They are classified as type I (ecto-
to the skin. A probe may be placed transorally dermal derived duplication of the external auditory
and the duct may be assessed through the wound. canal) or type II (ectoderm and mesoderm derived
The proximal end of a lacerated duct may be cyst or fistula). These lesions may become repeat-
identified by gland massage, which should dem- edly infected in which case they must be excised
when there is no active infection to allow clear dis- ranulas but have a true epithelial lining. Although
section of the cyst and its tract [44]. The tracts are marsupialization is the classic treatment for sialo-
always intimately associated with the facial nerve celes, simple sialodochostomy has been reported
and superficial parotidectomy with facial nerve as a safe and effective method of treatment for
monitoring, and preservation is often required. congenital sialocele associated with an imperfo-
Congenital duct cysts can be diagnosed and further rate submandibular or sublingual duct [51].
characterized by sialography. Intervention is not
warranted unless the cyst becomes infected [18].
Acquired cysts of the salivary glands include 13.2.2 Neoplasm
posttraumatic cysts, postinfectious cysts, neo-
plasms, benign LE cysts (described above), The major salivary glands originate from
mucoceles, and sialectasis with duct obstruction epithelial invaginations from oral ectoderm dur-
(due to sialoliths vs. other etiology). Unless the ing the 6th week of gestation, as described above.
cyst is associated with a neoplasm or becomes The ingrowths develop into the ductal system.
infected, no intervention is warranted as long as The acinus drains into the intercalated duct,
the cyst is asymptomatic. which in turn drains into the striated duct fol-
Mucoceles form due to extravasation of lowed by the excretory duct [2].
mucous; mucous retention cysts are true cysts and There are two theories of tumorigenesis for
are lined by epithelium. Both of these phenomena neoplasms of the salivary glands: the multicellu-
usually occur in minor salivary glands on the labial lar theory and the bicellular reserve theory [8,
mucosa, buccal mucosa, and ventral tongue. 52]. The multicellular theory states that each type
Treatment for symptomatic mucoceles or retention of neoplasm is derived from a differentiated cell
cysts is accomplished by complete excision or of origin within the salivary gland (Warthin’s
marsupialization [18]. A ranula is a large muco- tumors and oncocytic tumors arise from striated
cele that arises from the sublingual gland from a duct, acinic cell tumors arise from acinic cells,
ruptured duct or acinus. It presents as a cystic mass etc.). The bicellular theory states that all primary
on the floor of the mouth. If the ranula continues to salivary gland neoplasms originate from the basal
increase in size, it can dissect through a congenital or stem cell of either the excretory duct or the
dehiscence of the mylohyoid muscle or in between intercalated duct cells (adenomatous tumors such
the mylohyoid and hyoglossus muscles into the as pleomorphic adenomas and oncocytic tumors
submandibular space and present as a neck mass originate from the intercalated duct, while tumors
[45–47]. This is referred to as a “plunging ranula.” with an epidermoid component such as squa-
Surgical intervention involves either marsupializa- mous cell carcinoma and mucoepidermoid carci-
tion of a small ranula, surgical excision of the noma originate from the excretory duct) [8, 52].
ranula, or attempts at inducing fibrosis that would Several etiologic factors have been linked with
prevent reformation. Methods to induce fibrosis salivary gland neoplasms including environmen-
include laser vaporization and sclerosing agents tal factors such as radiation exposure (Warthin’s
[18, 45, 46, 48]. An outpatient method of inducing tumor), viruses (EBV and lymphoepithelial car-
fibrosis involves placing several sutures into the cinoma), tobacco use (Warthin’s tumor), expo-
ranula to allow drainage with subsequent suture sure to silica dust and nitrosamines, diet, and
removal once adequate fibrosis has been achieved genetic factors [8, 13].
(Fig. 13.3). This relatively new technique has been The majority of salivary gland tumors
termed “micro-marsupialization.” The concept was (approximately 70 %) arise in the parotid and the
introduced in 1995; however, its safety and effi- majority of parotid gland tumors are benign
cacy have been under recent investigation [49, 50]. (approximately 80 %). Ten percent of tumors
An imperforate submandibular or sublingual arise in the submandibular gland, and the ratio of
duct orifice may also present as an intraoral cystic benign to malignant tumors is similar to the
swelling. These congenital sialoceles may mimic parotid gland. Twenty percent of salivary gland
a b
c d
Fig. 13.3 Photographs depicting suture marsupialization tial lesion. In image (c), repeat suture marsupialization
of a right-sided intraoral ranula. In image (a), the lesion is has been performed. Image (d) demonstrates the resolu-
depicted in the right floor of the mouth; this lesion was tion of the lesion; the sutures have been removed 2 weeks
masupialized with suture but recurred. Image (b) depicts after placement (Reprinted from Ref. 50)
the recurrent intraoral ranula located posterior to the ini-
tumors arise in minor salivary glands, and benign tumors. Obstruction of the duct may cause
50–75 % of these tumors are malignant [13, 53]. rapid swelling or predispose the gland to sialadeni-
Most salivary gland tumors in adults are benign. tis. Cutaneous malignancy of the scalp or facial
Salivary gland tumors in the pediatric population skin may also metastasize to the intraparotid or
are far less common than in adults; however, the periparotid lymph nodes. Benign tumors are typi-
majority of pediatric salivary gland tumors are cally mobile and well defined. Tumors may origi-
malignant. Other lesions that may present in the nate from the superficial or deep lobe of the parotid
salivary glands of the pediatric population include and may present on the face or in the neck or may
hemangiomas, vascular malformations, and lym- occupy the parapharyngeal space and present as
phatic malformations [8, 13, 24]. intraoral swelling. Malignant tumors are more
Tumors of the parotid gland present as painless likely to be fixed to surrounding tissues and cause
swelling; the rate of enlargement is often slow for facial nerve paresis. Malignant tumors are more
likely to be associated with regional and cervical the parotid and can extend to the deep lobe into
lymphadenopathy [8, 13, 54, 55]. the parapharyngeal space. Pleomorphic ade-
Tumors of the submandibular gland are less com- noma of the minor salivary glands can occur on
mon but present as a mobile mass in the subman- the palate, the labial mucosa (more commonly
dibular triangle of the neck. Malignant lesions may the upper lip), or parapharyngeal space. The
be fixed to surrounding structures and may cause consistency of the tumor is typically smooth and
tongue weakness or numbness from perineural rubbery in texture. Encapsulation is present;
spread. Lower lip weakness may suggest involve- however, it may be incomplete with “pseudo-
ment of the marginal mandibular branch of the facial pod” extensions of the tumor [8, 13]. Due to the
nerve. Tumors of the sublingual gland are rare and risk of recurrence and the presence of pseudo-
may present as a floor of mouth mass. Clinical pre- pod extensions, complete surgical resection
sentations of minor salivary gland tumors depend on with a margin of normal tissue is performed.
the location of the gland; the most common sites of This may entail partial or superficial parotidec-
presentation are the palate and the parapharyngeal tomy with facial nerve preservation. Rarely,
space [8, 13, 54, 55]. pleomorphic adenoma has been reported to
Diagnosis of a parotid or submandibular gland metastasize to the bone, lungs, skin, and other
tumor can be obtained by fine needle aspiration regions of the head and neck. Recurrence or
(FNA) biopsy. This may be guided by ultrasound metastasis is attributed to either leaving residual
if the mass is indiscrete or difficult to visualize. tumor in the surgical bed or rupture of the tumor
Complications of FNA biopsy include local during excision. Malignant transformation is
infection, hemorrhage, infarction, fibrosis, and rare and is termed “carcinoma ex pleomorphic
tumor seeding. Mukunyadzi et al. noted that FNA adenoma” [8, 13].
biopsy with a 25 G needle is not only safe but
also allowed the surgeon to obtain an adequate Warthin’s Tumor
diagnosis without tumor seeding [8, 56]. Warthin tumor, also known as papillary cystade-
The extent and type of imaging of salivary gland noma lymphomatosum, is the second most com-
tumors depends on the size, location, and suspicion mon (5–10 %) benign neoplasm of the salivary
for malignancy. Small and well-defined and palpa- glands. The most common site for Warthin’s
ble tumors may require ultrasound evaluation or tumor is the parotid gland. Smoking is a known
may not require any imaging; however, larger tumors risk factor and the tumor is more common in men.
and suspicion for malignancy require workup with The tumor may be multicentric; it may present
CT, ultrasound with color Doppler, positron emis- bilaterally in up to 12 % of patients [57]. The
sion topography, or even MRI in advanced tumors tumor is slow growing, painless, and smooth in
with perineural invasion [8, 13, 54, 55]. appearance with a well-defined capsule. A cross-
section often reveals multiple cystic spaces with
13.2.2.1 Benign Salivary Gland brown mucoid fluid. Histology is characteristic of
Neoplasms a projection of a double-layered papillary epithe-
Pleomorphic Adenoma lium with lymphoid stroma into cystic spaces.
Pleomorphic adenoma, which is also known as Treatment of Warthin’s tumor is complete surgi-
“benign mixed tumor,” is the most common cal excision. Recurrence may be attributed to
(65–75 %) salivary gland tumor. It is most often undiagnosed multicentricity [8, 13, 55].
found in the parotid gland followed by the sub-
mandibular gland and the minor salivary glands. Oncocytoma
Pleomorphic adenoma is a slow-growing and Oncocytomas represent 1 % of salivary gland
painless tumor that contains both mesenchymal tumors that present almost exclusively in the
and epithelial components; the mesenchymal parotid gland. It presents as a painless mass and
stroma varies between tumors [8, 13, 55]. These is firm, encapsulated, and rubbery in consistency.
tumors may originate in the superficial lobe of Histologically, the tumor contains granular
eosinophilic cells with abundant, hyperplastic malignant tumor of the parotid gland is muco-
mitochondria and indented nuclei [13]. Complete epidermoid carcinoma, it is the second most
resection in an extracapsular fashion is sufficient common malignant neoplasm of the subman-
treatment. Oncocytomas of the minor salivary dibular gland after adenoid cystic carcinoma.
glands may be locally invasive and have potential Mucoepidermoid carcinoma usually occurs
to destroy adjacent tissues despite their benign after the third decade of life with a female pre-
nature. Surgical excision is the preferred treat- dominance [13]. These tumors are classified as
ment [8, 13, 55]. either high grade or low grade based on histo-
logical findings. Low-grade tumors contain
Basal Cell Adenoma mucoid as well as epidermal cell components
Basal cell adenoma represents 2–3 % of salivary and rarely metastasize, while high-grade
gland tumors and occurs most often in the parotid tumors are predominated by epidermoid cells
gland but has been reported in the submandibular and have a high propensity to metastasize.
and minor salivary glands. It typically affects patients Low-grade tumors are usually small, can be
in their seventh to eighth decade of life and affects encapsulated, and contain mucinous fluid.
women more commonly than men. Basal cell ade- High-grade tumors are usually solid and may
nomas are encapsulated and can present in four dis- have no encapsulation. The prognosis is worse
tinct histological patterns (solid, tubular, trabecular, for high-grade mucoepidermoid carcinoma.
and membranous). Minor salivary gland basal cell Surgical resection is recommended for low-
adenomas may lack a capsule. Surgical resection is grade tumors; the neck is not treated in the clin-
the treatment of choice. The membranous subtype is ically N0 neck due to the low incidence of
nodular in appearance and can display multicentric- nodal metastasis [13, 54]. High-grade tumors
ity, which increases the risk of recurrence after surgi- are treated with complete surgical resection,
cal resection. Basal cell adenomas appear similar and elective neck dissection is usually per-
histologically to adenoid cystic carcinoma; however, formed due to the higher rate (21 %) of occult
they do not invade surrounding tissues or adjacent nodal metastasis [58]; this is often followed by
nerves [8, 13, 55]. adjuvant radiation therapy.
Other Benign Neoplasms Adenoid Cystic Carcinoma

There are a number of other benign tumors of the Adenoid cystic carcinoma is the second most
salivary glands such as canalicular adenomas common parotid gland malignancy but is the
(presents in minor salivary glands), oncocytic pap- most common malignancy of the submandibular
illary cystadenoma (most often in larynx), myo- gland and the minor salivary glands. The tumor
epithelioma, sialadenoma papilliferum, inverted may be partially encapsulated (or without a cap-
ductal papilloma, and others that are outside the sule) and appears histologically as basaloid epi-
scope of this chapter. The treatment of the majority thelium arranged in cribriform, solid (worst
of these tumors is surgical resection [8, 13, 55]. prognosis), and tubular patterns with an eosino-
philic stroma. Although adenoid cystic carci-
noma is a slow-growing tumor, the tumor
13.2.2.2 Malignant Salivary Gland infiltrates surrounding tissue and demonstrates
Tumors perineural invasion with resultant facial nerve
Mucoepidermoid Carcinoma palsy. Local recurrences after resection and dis-
Mucoepidermoid carcinoma is the most com- tant metastasis to the lung are not uncommon.
mon malignant salivary gland neoplasm Surgical resection with postoperative radiation is
(approximately 30–35 % of all malignant sali- typically recommended. Occult metastasis is rare
vary gland neoplasms) [13]. The most common and elective neck dissection for the N0 neck is
site for mucoepidermoid carcinoma is the not performed. If the neck is clinically positive,
parotid gland. Although the most common the overall survival is lower [13, 54].
Acinic Cell Carcinoma The facial nerve can be preserved if the tumor
Acinic cell carcinoma most commonly affects has not invaded the neural tissue; intraoperative
women after the fourth decade of life and most frozen sections can assess for tumor margins in
often presents in the parotid masses. The tumor the nerve tissue. Nerve grafting should be per-
can be multicentric and can present bilaterally in formed for sacrificed nerves. Neck dissections
the parotid masses. The tumor is well encapsu- are performed in the clinically positive neck, and
lated and contains both serous acinar cells and elective neck dissections are performed in the
acinar cells with a clear appearing cytoplasm. clinically N0 neck for high-grade tumors [13,
Several histologic patterns are possible (cystic, 54]. The extent and specific indications for neck
papillary, vacuolated, follicular), and the cells dissections are outside the scope of this chapter.
stain positive on periodic acid-Schiff (PAS) stain. The most common approach to the parotid
Treatment is surgical resection; adjuvant radiation gland is via a modified facelift (modified Blair)
is performed with facial nerve involvement, neck incision or a preauricular incision that curves
metastasis, skin involvement, or other poor prog- along a skin crease into the neck approximately
nostic indicators. For histologically high-grade 2 cm below the angle and border of the mandible.
lesions, elective neck dissection is performed. A skin flap is raised anteriorly in a level superfi-
Local recurrences can present years after treat- cial to the parotid fascia until the masseter muscle
ment [13, 54, 55]. is encountered. The greater auricular nerve is
identified and preserved if possible in the event
Other Malignant Neoplasms that nerve grafting is required. The posterior
Other malignant salivary gland neoplasms include branch of the nerve can usually be preserved to
adenocarcinoma (minor salivary glands and mainatin sensation to the ear lobe; the anterior
parotid gland), polymorphous low-grade adeno- branch is often sacrificed. The tail of the parotid is
carcinoma, carcinoma ex-pleomorphic adenoma dissected from the sternocleidomastoid muscle.
(derived from pleomorphic adenoma), primary The posterior belly of the digastric muscle can be
squamous cell carcinoma (most often in subman- identified here and can serve as a landmark for
dibular gland), undifferentiated carcinomas, sali- facial nerve identification. Blunt dissection is then
vary duct carcinomas, sarcomas, lymphomas, and performed to separate the tragal cartilage from the
others. The treatment of these lesions is complete parotid gland tissue. This reveals the tragal
surgical resection with neck dissection for high pointer, which guides the surgeon in localizing
grade lesions [13, 54, 55]. the facial nerve (1 cm medial). If the tumor inter-
feres with identifying the nerve, a distal branch
13.2.2.3 Surgery of the Parotid Gland may be traced in a retrograde fashion to find the
The facial nerve branches divide the parotid into main nerve trunk [3, 5, 6, 13, 54]. Fibrosis from
arbitrary superficial and deep “lobes” as it prior surgery, radiation, or other anatomic distor-
courses through the parotid gland parenchyma. tion may prevent adequate identification of the
Benign neoplasms and low-grade, well-encap- nerve; the mastoid cavity may then be drilled to
sulated malignancies in the superficial lobe are find the intratemporal facial nerve, which is then
treated with a superficial parotidectomy with followed to its extratemporal course. The nerve is
preservation of the facial nerve branches. A total identified and traced anteriorly to its main
parotidectomy involves resecting the deep branches, separating the superficial and deep
parotid tissue as well; this is reserved for malig- lobes of the gland [3, 13, 54]. Janes was the first
nancies of the deep lobe, high-grade tumors, or surgeon to describe the identification process for
tumors with nodal metastasis. Cutaneous malig- the facial nerve trunk in 1940 [3, 59]. The nerve
nancies of the scalp or the face with nodal may then be mobilized if the deep lobe of the
metastasis to the parotid gland or high risk for gland is to be removed. The posterior auricular
nodal metastasis also require parotidectomy [3, artery, external carotid artery, and retromandibu-
6, 13, 54]. lar vein may be encountered during superficial
Fig. 13.4 Surgical
anatomy of the parotid
gland and relationship
to facial nerve
(Reprinted from
Ref. 54)
“Pointer” tragal
cartilage
Facial nerve
Digastric muscle
Masseter
Sternocleidomastoid
parotidectomy, and the internal carotid artery and seroma, infection, skin flap necrosis, trismus due
internal jugular vein are likely encountered during to inflammation or fibrosis of the masseter mus-
deep lobe removal [3, 13]. Figure 13.4 depicts the cle, development of a sialocele, and facial nerve
surgical approach to the facial nerve during a paralysis. Facial nerve paralysis is usually tem-
parotid gland dissection [54]. porary, and permanent facial nerve paralysis
Intraoperative frozen sections may be sent to occurs in less than 4 % of parotidectomies for
assess for extent of disease. If the tumor invades benign disease in which the nerve was identified
the facial nerve, the nerve may need to be traced and preserved [13, 61, 62]. The nerve may be
proximally into the temporal bone to obtain nega- stretched, compressed, or injured due to thermal
tive margins. Nerve reconstruction after nerve energy from electrocautery, or ischemia from
sacrifice is performed by primary repair or with extensive dissection. Postoperative edema of the
nerve grafting using the great auricular nerve or nerve may contribute to paresis and some sur-
the sural nerve from the lower extremity [13, 54]. geons administer postoperative steroids to reduce
Large parotid tumors may extend into the para- edema. Continuous facial nerve monitoring with
pharyngeal space. Tumors described as “dumbbell EMG is used to allow intraoperative nerve stimu-
tumors” may involve both the superficial and deep lation and to warn the surgeon when the nerve is
lobes as they straddle the mandibular ramus and in close proximity [3, 13].
stylomandibular ligament. Parapharyngeal space Late complications include Frey’s syndrome,
tumors can be removed either through a transoral tumor recurrence, and poor cosmesis due to
approach or via transcervical approach, which either scarring or loss of tissue bulk. Frey’s
may require division of the styloid process, stylo- syndrome is a well-known complication and
mandibular ligament, stylohyoid ligament, and is also referred to as “gustatory sweating” or
associated muscles, or even mandibulotomy for “auriculotemporal nerve syndrome” [3, 8, 13,
increased access. The transoral approach is usually 62, 63]. Frey’s syndrome was first described
reserved for well-encapsulated benign tumors due in 1853 by Baillarger, and the pathophysiol-
to the limited access and exposure [3, 13, 54, 60]. ogy was described by Frey in 1923 [64, 65].
Complications of parotid gland surgery can be This complication is thought to occur due to
classified as early and late complications. Early aberrant reinnervation of nerve fibers from
complications include bleeding, hematoma or postganglionic parasympathetic fibers of the
parotid gland (which use acetylcholine as a encountered and can be preserved. The lin-
neurotransmitter) to the sweat glands and tran- gual nerve is identified and mobilized from the
sected postganglionic sympathetic fibers to the gland. The hypoglossal nerve is often identified
sweat glands (which also use acetylcholine as at this point and preserved. Wharton’s duct is
a neurotransmitter). This causes sweating and identified and ligated [3, 8, 13]. Figure 13.5
flushing of the cheek skin as a parasympathetic depicts the classic lateral cervical approach to
response during salivation. Using a thicker skin the submandibular gland [13].
flap and less extensive parotid dissection may Complications of submandibular gland sur-
reduce the incidence of Frey’s syndrome. Using gery include bleeding or hematoma, seroma,
fascial flaps, muscle flaps, or synthetic material infection, scarring, injury to the marginal man-
as a barrier has also been described [3, 13, 63]. dibular branch of the facial nerve, injury to the
10 % of patients have symptomatic Frey’s syn- lingual nerve, or injury to the hypoglossal nerve.
drome [13]. Symptomatic treatments include; Temporary lower lip paresis can occur with sim-
botox injections, topical antiperspirants, topi- ilar injury mechanisms to the facial nerve dur-
cal anticholinergics, or tympanic neurectomy. ing parotid surgery. Tongue weakness and
Postoperative radiation has decreased the inci- tongue hypesthesias can occur from hypoglos-
dence of Frey’s syndrome [3]. Sialoceles are sal nerve and lingual nerve injury, respectively
subcutaneous saliva collections that can be [3, 8, 13].
managed with observation, needle aspiration,
or a pressure dressing [3]. Botox injections may 13.2.2.5 Surgical Approach
decrease salivation to promote resolution [3]. to the Sublingual Gland
Frey’s syndrome was first described in 1853 The sub lingual gland is approached in a transoral
by Baillarger, and the pathophysiology was fashion. A linear incision can be made parallel to
described by Frey in 1923 [64, 65]. the ipsilateral mandible. The gland can be bluntly
dissected from adjacent structures such as the
13.2.2.4 Surgery of the Submandibular Wharton’s duct and the lingual nerve. The supe-
Gland rior and medial aspects of the gland can be dis-
Surgical resection of the submandibular gland sected such that the gland can be peeled from the
is typically confined to the submandibular tri- sublingual space with blunt dissection. Injuries to
angle unless an extensive malignant neoplasm the lingual nerve and Wharton’s duct may occur
extends to surrounding structures. The subman- when attempting gland removal. A floor of mouth
dibular gland is typically approached in a trans- hematoma may occur and compromise the
cervical fashion, although submental, transoral, patient’s airway if hemostasis is not adequately
retroauricular, and endoscopic-assisted/endo- acquired at the time of surgery [3].
scopic robot-assisted approaches have been
described. The transcervical approach improves
direct access to the gland. An incision is made 13.3 A
dvances in Salivary Gland
1.5–2 cm below the inferior border of the man- Surgery
dible along a neck skin crease. A subplatysmal
flap is raised superiorly, and the marginal man- 13.3.1 Sialendoscopy
dibular nerve is identified and preserved or sim-
ply preserved with the Hayes-Martin maneuver Sialendoscopy is a relatively novel technique that
as described previously. The fascia investing provides visualization into the salivary ducts via
the submandibular gland is incised to expose a small-caliber endoscope. The technology was
the gland. The facial artery is encountered and originally invented for the purpose of diagnosis
ligated. The mylohyoid muscle is then retracted but is currently used to treat a variety of nonneo-
anteriorly to expose the anterior aspect of the plastic salivary gland pathologies while allowing
gland. The nerve to the mylohyoid may be gland preservation [66, 67]. The endoscope is
Facial vein
ligated and
divided
Marginal
mandibular
nerve
Tumor mass
Incision
Facial artery
ligated and
divided
Posterior belly Hypoglossal

digastric muscle nerve
Lingual
Maximum removal nerve
of Wharton’s duct Postganglionic
nerve fibers Submandibular
ganglion
Mylohyoid
retracted
Total excision
of gland
Wharton’s
duct
Fig. 13.5 Surgical approach to the submandibular gland depicting the Hayes-Martin maneuver (Reprinted from Ref. 13)
passed into the Stenson’s duct or Wharton’s duct, The endoscopes can range in size from 0.8 to
and saline is irrigated through the endoscope to 1.6 mm in diameter although some studies recom-
fill the lumen and distend the salivary ductal tree. mend limiting the caliber to 1.2 mm to avoid iatro-
Katz was the first to describe salivary endoscopy genic injury [67]. The endoscopes most commonly
and endoscopic anatomy in 1991 [10]. The mas- used today are semirigid although Katz first
seteric bend of Stensen’s duct was characterized described the use of a flexible endoscope [35].
by endoscopic anatomy. The optical resolu- Atienza and López-Cedrún recently performed a
tion has since improved; the sialendoscopists of systematic review of the management of obstruc-
today can perform procedures to treat a variety of tive salivary disorders with sialendoscopy and
conditions. concluded that sialendoscopy is both safe and
Fig. 13.6 Instruments a b
used in Sialendoscopy.
(a) sialendoscope,
modular;
(b) sialendoscope, all in
one; (c) biopsy and
grasping forceps; d
(d) forceps; (e) bougies;
(f) probes; (g) dilatator; c
f
(h) stone extractor;
(i) microdrill; (j) balloon
catheter (Courtesy of e
Karl Storz Ref. [68])
g
h
i
j
effective for the treatment of obstructive salivary thotomy, or ductal dissection can be performed for
gland disorders. Four thousand one hundred thirty- larger stones. Stones that are 8 mm or greater typi-
four sialendoscopic procedures were performed in cally require a combined surgical approach [27,
Atienza’s review [67]. Figure 13.6 depicts typical 34, 67]. Figure 13.7 depicts endoscopic images
instruments used for sialendoscopy [68]. during endoscopic sialolith removal [35].
Sialendoscopy and sialendoscopic interven- Inflammatory disorders such as radiation- and
tions are typically outpatient procedures and can radioiodine-induced sialadenitis and autoim-
be performed either under local or general anes- mune sialadenitis may also be treated with sialen-
thesia. The sialendoscope is introduced into the doscopy irrigation with instillation of steroids
duct although serial dilation of the papilla and [67]. Strictures can be found in diseases such as
duct with lacrimal probes may be required for Sjögren’s syndrome, radiation-induced sialadeni-
insertion. The lumen is irrigated with normal tis, and juvenile recurrent parotitis; balloon dila-
saline which distends the ductal tree and allows tion and steroid instillation may be performed for
both visualization and room for the endoscope these conditions. 1 mm balloons that dilate to
and instruments to pass. One port is required for 3 mm are available for stricture dilation. Acute
saline irrigation. Larger endoscopes contain an sialadenitis is a contraindication for sialendos-
instrumentation port for forceps, wire baskets, copy as the risk for ductal perforation increases
micro-drills, balloons, stents, or laser interven- in this setting [34, 35, 66, 67]. Sialendoscopy has
tions. Medications such as steroids may also be also been used in the pediatric population for
instilled into the lumen. Patients are instructed to juvenile recurrent parotitis, Sjögren’s syndrome,
massage their glands postoperatively. Stents are and other acquired or congenital strictures [69].
typically removed several weeks after the proce- Atienza’s review of sialendoscopy for treatment
dure [35, 66, 67]. of obstructive disorders reports a success rate of
Atienza’s review revealed that sialolithiasis is 76 % for all sources of obstruction. In this sys-
involved in 66 % of the patients that undergo inter- tematic review, success was defined by resolution
ventional sialendoscopy [67]. Marchal recom- of obstruction with no symptoms upon patient
mends that sialoliths less than 3 mm in the parotid follow-up. The success rate increases to 96 %
gland and less than 4 mm for the submandibular when sialendoscopic intervention was combined
gland can safely be removed endoscopically [34]. with another surgical approach (papillotomy,
Laser lithotripsy or combined endoscopic and transoral incisions, incisions through parotid fas-
transoral maneuvers such as papillotomy, sialoli- cia, external incisions) [67].
a b c
Fig. 13.7 Endoscopic sialolith removal using a wire basket. (a) Sialendoscopic view of sialolith in Wharton’s duct; (b)
sialolith engaged in wire basket; (c) view of duct after sialolith removal (Reprinted from Ref. 35)
The most common complication of sialendos- view during an endoscopic-assisted transoral sub-
copy is post-procedural glandular swelling (typi- mandibular sialadenectomy [75].
cally resolves within 48 h), perforation of the In contrast, the role of robotic surgery has
duct, and injuries to adjacent blood vessels and become increasingly significant since its first
nerves. Other complications reported have been application in the field of otolaryngology in
postoperative stenosis due to structural failure of 2002 at the Medical College of Georgia [70, 79].
the duct or papilla; the studies in Atienza’s review Advantages of surgical robotics include increased
often cited the use of a stent to compensate for precision, three-dimensional magnification,
this complication. Failure to remove stones or improved articulation, and possibly improved
failure of equipment (such as the wire basket surgical ergonomics [80]. Transoral robotic sur-
for stone retrieval) was also reported. 4.6 % of gery is becoming an established part of the head
glands were excised after a sialendoscopic proce- and neck oncologic surgeon’s armamentarium.
dure in Atienza’s review [67]. Robotic capabilities in head and neck surgery are
continuing to be developed, and the robotic surgi-
cal procedures have been well documented in the
13.3.2 Minimally Invasive oral cavity, oropharynx, larynx, skull base and
and Robotic Surgery otologic procedures, thyroidectomy, and salivary
gland excision [70, 80, 81].
Endoscopic surgery has had limited use in neck Robotic surgery has been described for sali-
surgery due to the lack of anatomic space for vary gland excision, sialolith excision, and ranula
instrumentation and the need for high insufflation excision [81]. Terris et al. used a cadaver model
pressures for the neck [70]. Multiple minimally to perform endorobotic submandibular gland
invasive approaches to the submandibular gland excision in 6 cadavers and 11 total glands; they
have been reported, and several authors report the reported faster procedure times compared to neck
benefits of minimal scarring with an endoscopic- endoscopic surgery alone (Fig. 13.9) [82].
assisted submandibular sialadenectomy through a Lee et al. performed a prospective study com-
number of different incisions in the neck, hairline, paring robot-assisted and endoscopic-assisted
retroauricular, facelift incision, and modified submandibular sialadenectomy [83]. They con-
facelift approaches [71–78]. A video-assisted cluded that the early postoperative outcomes
approach to submandibular gland sialadenectomy were comparable and that patients in both cohorts
may yield excellent results with minimal scarring; were satisfied with their cosmesis. Although
this approach has yet to become a widely accepted more convenient for the surgeon, the robot did
and routinely practiced technique at most major not give the surgeon any clinical advantage over
institutions. Figure 13.8 depicts an endoscopic the endoscope.
Fig. 13.8 Intra
operative photograph
of endoscopic-assisted
approach to transoral
left submandibular
sialadenectomy. D
submandibular duct, L
lingual nerve, SLG
sublingual gland
(Reprinted from
Ref. 75)
case series in 2015 describing robot-assisted

sialolithotomy with sialendoscopy (RASS) for
the management of large (>5 mm) hilar subman-
dibular gland sialoliths [85]. Twenty-two
patients underwent this procedure, and success
(defined as gland preservation with absence of
symptom recurrence) was reported in 100 % of
the subjects. This cohort was compared to a his-
torical cohort in Razavi’s study that consisted of
patients that underwent sialolithotomy via a
combined sialendoscopy/traditional transoral
approach for which the success rate was 75 %.
Although further investigation and prospective
Fig. 13.9 Photograph depicting trochar placement for
endorobotic submandibular gland excision (Reprinted studies are warranted, these results suggest that
from Ref. 82) the safety and efficacy of robot-assisted sialoli-
thotomy is excellent. Surgical robotics may
Walvekar reported the first case in which a eventually become an important adjunct to
submandibular gland megalith (19 × 11 mm) sialendoscopy. Walvekar et al. also reported the
was removed using a combination approach with first removal of a floor of mouth ranula using the
sialendoscopy to localize and trap the sialolith surgical robot [86].
while transoral robotic surgery was used to
remove the sialolith [84]. Razavi published a
13.3.3 Procedural Interventions 13.3.3.2 Salivary Gland Transfer

for Xerostomia Salivary gland transfer is a relatively new tech-
nique that has been developed to address postra-
Xerostomia affects the majority of patients who diation xerostomia as well as dry eyes and
undergo primary or adjuvant radiation therapy keratoconjunctivitis sicca [93–95]. Autologous
for head and neck cancer. Salivary glands are transplantation of both major and minor salivary
radiosensitive and gland destruction leads to glands has been described for these indications
hyposalivation [87, 88]. Decreased saliva causes [96, 97]. In patients undergoing radiotherapy for
the patient to experience xerostomia, dysphagia, head and neck cancer, salivary gland transfer may
and dysarthria while also predisposing the patient be performed at the time of surgical intervention
to dental caries and local oral and dental infec- in anticipation of postoperative radiation. Jha
tions. Traditional therapy for postradiation xero- et al. reported the first submandibular gland
stomia is a combination of strict oral hygiene, transfer to the submental space for shielding prior
saliva substitutes, fluoride agents, pilocarpine, to radiotherapy [98]. Wu et al. performed a sys-
and sialogogues. Amifostine has been used as a tematic review of the literature containing 369
cytoprotectant [89]. patients who underwent submandibular gland
transfer before radiotherapy in the included stud-
13.3.3.1 Acupuncture ies [99]. Both stimulated and unstimulated sali-
Acupuncture is an adjuvant alternative medicine vary flow rates were noted to be much higher in
modality extensively described in the treatment patients who underwent the intervention vs.
of xerostomia due to radiation as well as patients who either received pilocarpine or no
Sjögren’s syndrome. Although the mechanisms other intervention. They concluded that subman-
are not clearly elucidated, acupuncture has been dibular gland transfer is highly effective in the
shown to increase salivary flow in patients with prevention of postradiation xerostomia without
Sjögren’s syndrome as well as radiation xerosto- serious adverse effects.
mia [88, 90, 91]. Acupuncture does not appear to Major and minor salivary gland transfer tech-
be a widely available or accepted treatment given niques have also been performed for the purposes
the lack of standardization in its technique as of eye lubrication in the setting of dry eyes and
well as prospective randomized trials evaluating keratoconjunctivitis sicca [96, 97, 100]. Figure
its efficacy. The existing studies consist of small 13.10 depicts a schematic diagram of the four
sample sizes with a variety of technical varia- possible salivary gland transfers for xerophthal-
tions including needle location, needle stimula- mia [101]. The composition of saliva and tears is
tion, needle depth, number of treatments, and fairly similar, and the digestive component of
frequency of treatments [88]. Li et al. acknowl- saliva and presence of amylase have not been
edged the lack of standardization and proposed found to be destructive to the ocular surface [96].
an acupuncture protocol for patients with radia- Mucosal grafts containing salivary gland tissue
tion-induced xerostomia [91]. Zhuang et al. per- for dry eyes were first described by Murube in
formed a systematic review of the literature 1998 [102]; labial minor salivary glands were
depicting acupuncture as a treatment modality described to significantly reduce dry eye symp-
for radiation-induced xerostomia and acknowl- toms despite the minor differences in composi-
edged that there is insufficient evidence to deter- tion and increased viscosity. The graft is sutured
mine its safety or efficacy [88]. Furness et al.’s to the undersurface of the eyelid. The grafts
Cochrane review concluded that there is low appear to be 90 % viable although further pro-
quality evidence that acupuncture affects xero- spective studies are warranted [97].
stomia symptoms greater than placebo [92]. Limitations noted in major gland transfer
Therefore, routine use of acupuncture for radia- include potential gland necrosis, hypersecretion,
tion-induced xerostomia is not recommended at and donor site morbidity such as facial nerve
this time. injury. Surgical options described in animal and
a b
c d
Fig. 13.10 Diagrammatic illustration depicting three dif- vary glands; (b) Transposition of the parotid gland duct;
ferent surgical techniques of major salivary gland transfer (c) sublingual gland transplantation; (d) submandibular
for xerophthalmia. (a) Transplantation of the minor sali- gland transplantation. (Reprinted from Ref. 102)
human studies include transposition of Stenson’s blepharitis, corneal calcifications, and increased
duct to the inferior fornix, free transplantation of bacterial load in the conjunctival sac in canine
the sublingual gland to the conjunctival fornix studies [96, 103]. Epiphora may lead to increased
without microvascular anastomosis, and free trans- eye wiping and subsequent keratitis. Sublingual
plantation of the submandibular gland with or gland transplantation is not performed due to the
without microvascular anastomosis and implanta- high rate of necrosis given that the gland is trans-
tion of Wharton’s duct into the upper temporal for- planted as a free graft without vascular supply [96].
nix [96, 97]. Parotid gland transfer or Stensen’s Submandibular gland transplantation with
duct transposition has been reported to produce microvascular anastomosis appears to have the
copious amounts of tearing (saliva) and epiphora most advantages to the ophthalmologic surgeon.
as a gustatory response. Complications include This procedure was first described by Murube-del
Castillo in 1986, and several other authors have options include physical therapy, medications
replicated this technique [94, 96, 102, 104–108]. (glycopyrrolate, scopolamine), botox injection,
The seromucinous nature of submandibular gland treatment of gastroesophageal reflux, and radia-
secretions simulate the seromucinous lacrimal tion to the glands [109, 110]. Surgical options
secretions. The gustatory reflex of epiphora is not include gland excision, tympanic neurectomy,
present due to intraoperative denervation during and duct ligation. A combination of the above
the procedure [96]. Major salivary gland trans- procedures involving multiple glands may be
plantation to the ocular tissues requires a team of performed. Salivary duct repositioning or
ophthalmologists as well as head and neck sur- rerouting is also a well-known technique to

geons. After resection of the submandibular gland address sialorrhea. The parotid or submandibular
along with the duct, a surrounding cuff of mucosa ducts are rerouted to the posterior oropharynx or
at the papilla, and facial artery and vein glandular elsewhere in the oral cavity mucosa to avoid
branches, the vessels are anastomosed to branches more definitive procedures such as gland exci-
of the superficial temporal artery or preauricular sion. Duct rerouting has allowed preservation of
vessels. The gland is placed in a pocket created in salivation with reduction of drooling [110].
the temporalis muscle, and the duct is tunneled Hockstein states that the most definitive surgical
subcutaneously to the conjunctival fornix. therapy is bilateral parotid duct ligation with
Prospective studies have revealed improved bilateral submandibular gland excision [109].
Schirmer’s test, fluorescein break-up time, use of Reed et al. performed a meta-analysis of the sur-
artificial tears, and discomfort. Epiphora and epi- gical management of drooling and noted that
thelial edema were common complications for there is no single procedure that is agreed upon as
successful transplantations [96]. Geerling and the most effective [110]. Large, directly compar-
Sieg note that of the three major salivary glands, ative studies depicting the safety and efficacy of
autologous submandibular gland transplantation the above procedures are required to identify a
is the only procedure that can currently be recom- procedure that may be universally performed by
mended in humans [96]. otolaryngologists and maxillofacial surgeons.
13.3.4.1 Salivary Duct Repositioning

13.3.4 Procedural Interventions Salivary duct repositioning is used to address
for Sialorrhea sialorrhea and xerophthalmia and to prevent post-
operative salivary duct obstruction from oral can-
Sialorrhea is the term for excessive salivation in cer resection or salivary calculi. Parotid salivary
both children and adults with neurologic impair- duct repositioning is described often in the litera-
ment. Patients with neurologic impairment suffer ture as a surgical procedure to address sialorrhea.
from a defect in their oral and oropharyngeal The procedure is most often described in pediat-
phases of swallowing which causes pooling of ric populations, and submandibular duct reposi-
saliva. Other causes for sialorrhea include oral tioning is the most commonly described.
inflammation, gastroesophageal reflux, medica- Puraviappan et al. performed a prospective study
tion side effects and toxins, or anatomic abnor- in which the efficacy of submandibular duct relo-
malities of the oral cavity and oropharynx cation was assessed in eight children with cere-
(tonsillar hypertrophy, macroglossia) [109]. bral palsy using a visual analogue score by the
Hypersecretion of saliva in combination with patient’s parents. They reported that seven of
poor oropharyngeal and facial muscle control eight patients had significant reduction in drool-
and dysphagia leads to pooling of saliva in the ing and reported parent satisfaction in all patients
oral cavity, oropharynx, and larynx. Patients with [111]. De et al. reported outcomes for subman-
sialorrhea often suffer from dehydration, chapped dibular duct relocation for 56 pediatric patients;
lips, and are socially marginalized due to the odor drooling was significantly reduced in 49 cases,
and appearance of excess saliva. Nonsurgical and parental satisfaction was noted to be high.
The main complication reported was ranula for- Stevens-Johnsons syndrome but not in ocular
mation in five cases. They conclude that duct pemphigoid (these patients failed to improve).
repositioning is a significant means to improve Complications reported may include duct
quality of life in pediatric patients with sialorrhea obstruction, dislocation, and ductal contraction,
[112]. Panarese et al. reported outcomes for 37 which have the potential to cause entropion or
pediatric patients and noted that 76.5 % of ectropion. Gustatory epiphora is a manifestation
patients had long-term control of sialorrhea, and due to the nature of the parasympathetic
the authors also concluded that the procedure is innervation as described previously. Prospective
safe and successful and improves quality of life studies are warranted to further evaluate the effi-
in the majority of patients [113]. Uppal et al. per- cacy and rate of complications.
formed a retrospective review of 23 neurologi-
cally impaired children and noted an overall
improvement in drooling in 20 patients (13 Salivary Duct Repositioning for Head and
patients with complete cessation of drooling); Neck Cancer
reported complications were ranula, submandib- Salivary duct repositioning has also been used in
ular gland swelling (three transient, two which the setting of oral cancer. Salivary gland swelling
required gland excision). Three patients were and pain may occur in the postoperative period,
reported to have a poor outcome, and they noted which may be confused for a tumor recurrence.
that these patients had the most severe oral-motor Duct relocation may prevent this potential false-
dysfunction [114]. Katona et al. performed sub- positive diagnosis and decrease postoperative sali-
mandibular duct relocations on 14 young adults vary gland colic. Stenson’s duct rerouting has been
and children using high-frequency radiosurgery reported as a means for gland preservation without
techniques; 79 % of patients achieved a satisfac- compromising cancer resection [120, 121]. The
tory decrease in sialorrhea. Katona et al. also salivary duct can be repositioned even in the setting
reported decreased operative time with high-fre- of oral cancer reconstruction with a free flap by
quency radiosurgery and endorsed its safety and routing the duct through the free flap [122].
efficacy [115]. Sakakibara reported that repositioning of
Wharton’s duct could lower the likelihood of post-
Salivary Duct Repositioning for operative obstructive complications. Mehta et al.
Xerophthalmia performed parotid duct relocation in buccal mucosa
Parotid duct relocation, as described in the previ- cancer resection in 562 patients and reported a
ous section, has also been used to treat xeroph- markedly reduced incidence of postoperative sialo-
thalmia due to several etiologies (autoimmune, cele and parotitis [121].
inflammatory). The duct is rerouted either tran-
sorally or extraorally to the conjunctival fornix;
an external approach was first described by
Filatov and Chevaljev in 1951 [96, 116–118]. 13.4 T
he Future: Salivary Gland
Zhang et al. reported outcomes on 40 cases in Regeneration
which parotid duct transposition was performed
for xerophthalmia, 82.5 % of patients had tearing The minimally invasive and gland-sparing proce-
postoperatively, and vision was improved in dures described above are relatively recent inno-
72.5 % of patients [119]. The etiologies for dry vations in the history of head and neck surgery.
eye in this studies included Stevens-Johnson syn- The future of salivary gland surgery is promising
drome, ocular pemphigoid, and alkali eye burn. and may build on the principles of gland sparing
They concluded that parotid duct transplantation techniques. Regenerative medicine is a rapidly
is a simple and easy procedure that should be emerging field of research and will certainly
considered in patients with dry eyes caused by impact the surgical management of salivary gland
disease. The molecular and genetic mechanisms onstrated that coculture of embryonic mouse sub-
of salivary gland biogenesis and development are mandibular gland cells resulted in better-developed
becoming increasingly elucidated to allow for epithelial structures (acinar-like aggregations)
experimentation with human salivary gland than monoculture of embryonic mouse salivary
regeneration. Salivary gland stem cells are being gland cells. This highlights the significance of the
characterized and will play a large role in thera- stem cell niche and suggests that induced pluripo-
peutic salivary gland regeneration. tent stem cells may be able to accelerate to regen-
Salivary gland regeneration for the purposes eration and development of salivary glands. These
of restoring function in xerostomia and irradiated studies provide hope that the functional regenera-
salivary glands is a major focus of research. tion of salivary glands will soon be possible in
Mouse models have allowed researchers to patients. It is unclear at this time how the stem cell
propose several methods of salivary gland regen- microenvironment in human salivary glands is
eration [7]. Approaches to regeneration are gene affected after radiation, surgery, or in the setting
therapy with viral vectors, stem cell therapy, and of both mild and severe autoimmune disease.
replacement of native gland tissue with bioengi- Future studies may further investigate the impact
neered salivary glands [7, 123]. Viral vectors of these variables on stem cell niche and the
have been used to express water channels (aqua- potential for human application.
porins) into the ductal epithelium via intraductal There is also active research interest in the use
injection of the vector [124]. Bone marrow stem of bioengineered cells and tissue for functional
cells have been transplanted into irradiated mouse organ restoration. Ogawa et al. performed ortho-
salivary glands; the cells secreted a factor which, topic transplantation of bioengineered salivary
acted in a paracrine fashion to regenerate epithe- gland germ cells into gland-deficient mice and
lia and increased salivary secretions, provided demonstrated functional regeneration of mature
cell protection, increased vascularity, and induced salivary glands that produced saliva in response to
the upregulation of biomarkers responsible for pilocarpine and gustatory stimulation, protected
cell regeneration [125–127]. The Coppes lab has against bacterial infection, and improved swallow-
demonstrated that the transplantation of cells ing [132]. Synthetic extracellular matrix has been
expressing Kit (a tyrosine kinase growth factor proposed to serve as a scaffold for implanted cells
receptor) into mouse salivary glands induced the to form epithelium and other glandular compo-
functional regeneration of gland epithelium. nents. Molecular components of the extracellular
Autologous gland transplantation in humans with matrix regulate cell proliferation and develop-
Kit+ salivary gland cells biopsied prior to irradia- ment; a variety of synthetic extracellular matrix
tion, and then reimplantation postradiation may scaffolds may one day be designed to customize
be a therapeutic implication of these findings [7, cellular polarity and function. In vitro regeneration
128, 129]. of human salivary gland tissue by culturing human
Stem cells (including embryonic and other salivary gland cells in three dimensions in a colla-
types) have complex interaction patterns with gen and matrigel construct has been described.
their microenvironment, also termed “stem cell Single human gland cells are proliferated and
niche”; stem cells affect the microenvironment assembled into both acinar and ductal structures
and differentiate under the influence of extrinsic [133]. Cells may be cultured in this fashion for
factors. Stem cells have been proposed to reside ultimate implantation into in vivo salivary gland
outside the ducts of salivary glands. The respec- tissue [134, 135].
tive salivary gland niche likely impacts the differ- Three-dimensional printing in resin has been
entiation of salivary gland stem cells [130]. Ono used in murid models to replicate the anatomic
et al. attempted to regenerate salivary gland cells shape of soft tissue organs such as the salivary
by coculturing embryonic salivary glands and gland based on three-dimensional MRI recon-
induced pluripotent stem cells [131]. They dem- structions [136]. 3D printing of functional salivary
glands has yet to be developed; however, this tech- tures and kinks will continue to require interven-
nology could help shape synthetic matrices for tion; parenchymal regeneration may be of no use
cellular growth and biogenesis of functional sali- without a functioning collecting duct and drain-
vary glands. These methods will be further age system. Adjunctive procedures such as sialen-
explored and validated in nonhuman models; doscopic dilation of ducts or instillation of
human clinical trials may one day incorporate sev- anti-inflammatory or growth factors may one day
eral of these techniques for salivary gland support gland regeneration. Sialoceles and sali-
regeneration. vary fistulas may develop in cases of aberrant
repair or regeneration, as in the setting of trauma.
The ability to replicate the three-dimensional
13.4.1 Implications for the Head anatomy of salivary glands in humans is also
and Neck Surgeon unclear. The malignant potential of newly gener-
ated or bioengineered tissue is also unknown.
The study of functional restoration of human sali- Future studies will need to evaluate the feasibility
vary gland tissue is in its infancy; it has yet to of salivary gland regeneration of human salivary
directly translate into the routine care for patients gland tissue in the setting of prior surgery and the
with xerostomia. Discoveries in regenerative abovementioned conditions.
medicine may one day allow partial or complete
regeneration of atrophic glands by implantation of Conclusion
salivary gland cells or implantation of entire sali- The salivary glands may harbor a variety of
vary glands generated in vitro. The head and neck conditions including obstructive, inflamma-
surgeon will need to be aware of the implications tory, infectious, and neoplastic disorders.
of these potential treatments. Studies that success- Surgical management of salivary gland dis-
fully describe the regeneration of salivary gland ease requires an in-depth understanding of the
tissue are carried out in a controlled and favorable anatomy, physiology, and common disease
environment. Head and neck surgeons often per- processes involving salivary gland tissue. The
form salivary gland surgery on patients with history of surgical intervention for salivary
altered anatomy, infected salivary glands, glands gland disorders in otolaryngology and maxil-
containing neoplasms, atrophic glands, and lofacial surgery features an evolution of tech-
fibrotic salivary glands. Each of these conditions niques from definitive and invasive procedures
affects the procedural technique and level of dif- such as gland extirpation to minimally inva-
ficulty for the surgeon. Regeneration after stem sive procedures that may preserve the glands
cell or bioengineered tissue implantation may not such as salivary gland duct surgery and sialen-
be successful in human salivary glands that were doscopy. Salivary duct repositioning has
irradiated in vivo or atrophied after an inflamma- allowed gland preservation in patients with
tory or obstructive process. The irradiated or atro- sialorrhea and can provide symptomatic relief
phic gland’s blood supply may be scant due to to patients with xerostomia. An increasing
dense fibrosis. This may prevent the proliferation number of surgical approaches to salivary
of regenerative cells in humans. Fibrosis also pre- gland extirpation have also developed with a
vents the expansion of tissue, which may limit the trend toward smaller skin incisions and the
growth of regenerating salivary glands. Patients pioneering of endoscopic and robotic tech-
with poor nutritional status and wound healing niques. It is difficult to predict how the expand-
capabilities may also benefit to varying degrees ing applications of regenerative medicine will
compared with healthy patients. The stem cell impact the future of salivary gland surgery.
niche may also vary depending on the location Head and neck surgeons must be aware of the
within the histologic architecture of the gland; technological and procedural advances in sali-
this is important when deciding the anatomic vary gland treatment in order to efficiently
location for cell or tissue implantation. Duct stric- incorporate new techniques into practice.
Reference 16. Edgar WM. Saliva and dental health. Clinical impli-

cations of saliva: report of a consensus meeting. Br
Dent J. 1990;169:96–8. doi:10.1038/sj.bdj.4807284.
1. Lydiatt DD, Bucher GS. The historical evolution of
17. Mandel ID. The role of saliva in maintaining oral
the understanding of the submandibular and sublin-
homeostasis. J Am Dent Assoc. 1989;119:298–304.
gual salivary glands. Clin Anat. 2012;25:2–11.
18. Walvekar RR, Bowen MA. Nonneoplastic diseases
doi:10.1002/ca.22007.
of the salivary glands. In: Eibling DE, Newlands SD,
2. Elluru RG. Physiology of the salivary glands. In:
editors. Bailey’s head neck surgery – otolaryngol.
Haughey BH, Robbins KT, editors. Cummings oto-
5th ed. Baltimore: Lippincott Williams & Wilkins;
laryngol, Head Neck Surg. 5th ed. Philadelphia:
2014. p. 702–16.
Mosby-Elsevier; 2010. p. 1134–42.
19. Rogers J, McCaffrey TV. Inflammatory disorders of
3. Nadershah M, Salama A. Removal of parotid, sub-
the salivary glands. In: Haughey BH, Robbins KT,
mandibular, and sublingual glands. Oral Maxillofac
editors. Cummings otolaryngol, Head Neck Surg.
Surg Clin North Am. 2012;24:295–305. doi:10.1016/j.
5th ed. Philadelphia: Mosby-Elsevier; 2010.
coms.2012.01.005.
p. 1151–61.
4. Walvekar RR, Loehn BC, Wilson MN. Anatomy and
20. Brook I. Diagnosis and management of parotitis. Arch
physiology of the salivary glands. In: Eibling DE,
Otolaryngol Head Neck Surg. 1992;118:469–71.
Newlands SD, editors. Bailey’s head neck surgery –
doi:file://Z:\References\Text Files\00000003991.txt
otolaryngol. 5th ed. Baltimore: Lippincott Williams
21. Brook I. Aerobic and anaerobic microbiology of
& Wilkins; 2014. p. 691–700.
suppurative sialadenitis. J Med Microbiol. 2002;51:
5. Carlson GW. The salivary glands. Embryology, anat-
526–9.
omy, and surgical applications. Surg Clin North Am.
22. Lary BG. Postoperative suppurative parotitis. Arch
2000;80:261–73. xii
Surg. 1964;89:653–5.
6. Du Toit DF, Nortje C. Salivary glands: applied anat-
23. Perry RS. Recognition and management of acute
omy and clinical correlates. SADJ. 2004;59:65–74.
suppurative parotitis. Clin Pharm. 1985;4:566–71.
doi:10.1017/S000748530002229X.
24. Iro H. Salivary gland diseases in children. GMS Curr
7. Holmberg KV, Hoffman MP. Anatomy, biogenesis
Top Otorhinolarynol. 2014;13:1–30. doi:10.3205/
and regeneration of salivary glands. Monogr Oral
cto000109.
Sci. 2014;24:1–13. doi:10.1159/000358776.
25. Capaccio P, Sigismund PE, Luca N, Marchisio P,
8. Calzada GG, Hanna EY. Benign neoplasms of the
Pignataro L. Modern management of juvenile recur-
salivary glands. In: Haughey BH, Robbins KT, edi-
rent parotitis. J Laryngol Otol. 2012;126:1254–60.
tors. Cummings otolaryngol, Head Neck Surg. 5th ed.
doi:10.1017/S0022215112002319.
Philadelphia: Mosby-Elsevier; 2010. p. 1162–77.
26. Schneider H, Koch M, Künzel J, Gillespie MB,
9. Frommer J. The human accessory parotid gland: its
Grundtner P, Iro H, et al. Juvenile recurrent parotitis:
incidence, nature, and significance. Oral Surg Oral
a retrospective comparison of sialendoscopy versus
Med Oral Pathol. 1977;43:671–6.
conservative therapy. Laryngoscope. 2014;124:451–
10. Katz P. Endoscopy of the salivary glands. Ann
5. doi:10.1002/lary.24291.
Radiol (Paris). 1991;34:110–3.
27. Capaccio P, Torretta S, Ottavian F, Sambataro G,
11. Kwak HH, Park HD, Youn KH, Hu KS, Koh KS,
Pignataro L. Modern management of obstructive
Han SH, et al. Branching patterns of the facial nerve
salivary diseases. Acta Otorhinolaryngol Ital Organo
and its communication with the auriculotemporal
Uff Della Soc Ital Di Otorinolaringol E Chir Cerv-
nerve. Surg Radiol Anat. 2004;26:494–500.
Facc. 2007;27:161–72.
doi:10.1007/s00276-004-0259-6.
28. Drage NA, Brown JE, Escudier MP, McGurk
12. Cannon CR, Replogle WH, Schenk MP. Facial nerve
M. Interventional radiology in the removal of sali-
in parotidectomy: a topographical analysis.
vary calculi. Radiology. 2000;214:139–42.
Laryngoscope. 2004;114:2034–7. doi:10.1097/01.
doi:10.1148/radiology.214.1.r00ja02139.
mlg.0000147943.13052.62.
29. Kelly IM, Dick R. Technical report: Interventional
13. Oh YS, Russel MS, Eisele DW. Salivary gland neo-
sialography: dormia basket removal of Wharton’s
plasms. In: Gourin CG, Pou AM, editors. Bailey’s
duct calculus. Clin Radiol. 1991;43:205–6.
head neck surgery – otolaryngol. 5th ed. Baltimore:
30. Ma Q, Song H. Diagnosis and management of lympho-
Lippincott Williams & Wilkins; 2014. p. 1760–87.
epithelial lesion of the parotid gland. Rheumatol Int.
14. Zenk J, Hosemann WG, Iro H. Diameters of the
2011;31:959–62. doi:10.1007/s00296-010-1617-9.
main excretory ducts of the adult human subman-
31. Cheuk W, Chan JKC. Kuttner tumor of the subman-
dibular and parotid gland: a histologic study. Oral
dibular gland fine-needle aspiration cytologic findings
Surg Oral Med Oral Pathol Oral Radiol Endod.
of seven cases. Am J Clin Pathol. 2002;117:103–8.
1998;85:576–80.
doi:10.1309/G9T3-22MH-Q7KL-G2DL.
15. Humphrey SP, Williamson RT. A review of
32. Kenefick JS. Some aspects of salivary gland disor-
saliva: normal composition, flow, and function.
ders. Proc R Soc Med. 1975;68:283–5.
J Prosthet Dent. 2001;85:162–9. doi:10.1067/
33. Sigismund PE, Zenk J, Koch M, Schapher M, Rudes
mpr.2001.113778.
M, Iro H. Nearly 3,000 salivary stones: some clinical
and epidemiologic aspects. Laryngoscope. cal outcomes and safety from a phase II clinical
2015;125:1879–82. doi:10.1002/lary.25377. trial. Head Neck. 2015;37:197–201. doi:10.1002/
34. Marchal F, Dulguerov P. Sialolithiasis management: hed.23579.
the state of the art. Arch Otolaryngol Head Neck Surg. 51. Rosow DE, Ward RF, April MM. Sialodochostomy
2003;129:951–6. doi:10.1001/archotol.129.9.951. as treatment for imperforate submandibular duct: a
35. Singh PP, Gupta V. Sialendoscopy: introduction, systematic literature review and report of two cases.
indications and technique. Indian J Otolaryngol Int J Pediatr Otorhinolaryngol. 2009;73:1613–5.
Head Neck Surg. 2014;66:74–8. doi:10.1007/ doi:10.1016/j.ijporl.2009.03.007.
s12070-013-0675-1. 52. Dardick I, Burford-Mason AP. Current status of histo-
36. Kagami H, Wang S, Hai B. Restoring the function of genetic and morphogenetic concepts of salivary gland
salivary glands. Oral Dis. 2008;14:15–24. tumorigenesis. Crit Rev Oral Biol Med. 1993;4:639–
doi:10.1111/j.1601-0825.2006.01339.x. 77. doi:10.1177/10454411930040050201.
37. Denny MC, Fotino AD. The Heerfordt-Waldenström 53. Adelstein DJ, Rodriguez CP. What is new in the
syndrome as an initial presentation of sarcoidosis. management of salivary gland cancers?. 2011;23.
Proc (Bayl Univ Med Cent). 2013;26:390–2. doi:10.1097/CCO.0b013e328344f59c.
38. Saarilahti K, Kouri M, Collan J, Hämäläinen T, 54. Sunwoo JB, Lewis Jr JS, McJunkin J, Sequeira
Atula T, Joensuu H, et al. Intensity modulated radio- S. Malignant neoplasms of the salivary glands. In:
therapy for head and neck cancer: evidence for pre- Haughey BH, Robbins KT, editors. Cummings oto-
served salivary gland function. Radiother Oncol. laryngol, Head Neck Surg. 5th ed. Philadelphia:
2005;74:251–8. doi:10.1016/j.radonc.2004.11.004. Mosby-Elsevier; 2010. p. 1178–99.
39. Kim JW, Han GS, Lee SH, Lee DY, Kim 55. Guzzo M, Locati LD, Prott FJ, Gatta G, McGurk
Y-M. Sialoendoscopic treatment for radioiodine M, Licitra L. Major and minor salivary gland
induced sialadenitis. Laryngoscope. 2007;117:133– tumors. Crit Rev Oncol Hematol. 2010;74:134–48.
6. doi:10.1097/01.mlg.0000247776.72484.62. doi:10.1016/j.critrevonc.2009.10.004.
40. Caglar M, Tuncel M, Alpar R. Scintigraphic evaluation of 56. Mukunyadzi P, Bardales RH, Palmer HE, Stanley
salivary gland dysfunction in patients with thyroid cancer MW. Tissue effects of salivary gland fine-needle aspi-
after radioiodine treatment. Clin Nucl Med. 2002;27:767– ration. Does this procedure preclude accurate histo-
71. doi:10.1097/01.RLU.0000031607.42515.B3. logic diagnosis? Am J Clin Pathol. 2000;114:741–5.
41. Hemenway WG, Bergstrom L. Parotid duct fistula: a doi:10.1309/NR4U-70L7-MBDG-DXTL.
review. South Med J. 1971;64:912–8. 57. Ethunandan M, Pratt CA, Higgins B, Morrison A,
42. Lewkowicz AA, Hasson O, Nahlieli O. Traumatic Umar T, Macpherson DW, et al. Factors influenc-
injuries to the parotid gland and duct. J Oral ing the occurrence of multicentric and “recurrent”
Maxillofac Surg. 2002;60:676–80. doi:10.1053/ Warthin’s tumour: a cross sectional study. Int J Oral
joms.2002.33118. Maxillofac Surg. 2008;37:831–4. doi:10.1016/j.
43. Work WP. Cysts and congenital lesions of the parotid ijom.2008.05.001.
gland. Otolaryngol Clin North Am. 1977;10:339–43. 58. Ghosh-Laskar S, Murthy V, Wadasadawala
44. Aronsohn RS, Batsakis JG, Rice DH, Work T, Agarwal J, Budrukkar A, Patil N, et al.
WP. Anomalies of the first branchial cleft. Arch Mucoepidermoid carcinoma of the parotid gland:
Otolaryngol. 1976;102:737–40. factors affecting outcome. Head Neck. 2011;33:497–
45. Huang S-F, Liao C-T, Chin S-C, Chen I-H. Transoral 503. doi:10.1002/hed.21477.
approach for plunging ranula – 10-year experience. 59. Heeneman H. Identification of the facial nerve in
Laryngoscope. 2010;120:53–7. doi:10.1002/ parotid surgery. Can J Otolaryngol. 1975;4:145–51.
lary.20674. 60. Casani AP, Cerchiai N, Dallan I, Seccia V,
46. Harrison JD. Modern management and pathophysi- Franceschini SS, Unit O. Benign tumours affecting
ology of ranula: literature review. Head Neck. the deep lobe of the parotid gland: how to select the
2010;32:1310–20. doi:10.1002/hed. optimal surgical approach. Acta Otorhinolaryngol
47. Zhao YF, Jia Y, Chen XM, Zhang WF. Clinical Ital. 2015;35:80–7.
review of 580 ranulas. Oral Surg Oral Med Oral 61. Koch M, Zenk J, Iro H. Long-term results of morbidity after
Pathol Oral Radiol Endod. 2004;98:281–7. parotid gland surgery in benign disease. Laryngoscope.
doi:10.1016/j.tripleo.2004.01.013. 2010;120:724–30. doi:10.1002/lary.20822.
48. Lai JB, Poon CY. Treatment of ranula using carbon 62. Guntinas-Lichius O, Gabriel B, Klussmann JP. Risk
dioxide laser – case series report. Int J Oral of facial palsy and severe Frey’s syndrome after con-
Maxillofac Surg. 2009;38:1107–11. doi:10.1016/j. servative parotidectomy for benign disease: analysis
ijom.2009.04.024. of 610 operations. Acta Otolaryngol. 2006;126:1104–
49. Delbem AC, Cunha RF, Vieira AE, Ribeiro LL. 9. doi:10.1080/00016480600672618.
Treatment of mucus retention phenomena in chil- 63. Li C, Zhang Q, Li L, Shi Z. Interventions for the
dren by the micro-marsupialization technique: case treatment of Frey’s syndrome. Cochrane Database of
reports. Pediatr Dent. n.d.;22:155–8. Systematic Reviews. 2015;7:CD009959.
50. Woo SH, Chi JH, Kim BH, Kwon SK. Treatment of 64. Baillarger M. Mémoire sur l’oblitération du canal
intraoral ranulas with micromarsupialization: clini- Sténon. Gaz Médicale Paris. 1853;23:194–7.
65. Frey L. Le syndrome du nerf auriculo-temporal. Rev 80. Maan ZN, Gibbins N, Al-Jabri T, D’Souza AR. The
Neurol. 1923;2:97–104. use of robotics in otolaryngology-head and neck sur-
66. Marchal F, Becker M, Dulguerov P, Lehmann W. gery: a systematic review. Am J Otolaryngol Head
Interventional sialendoscopy. Laryngoscope. 2000; Neck Med Surg. 2012;33:137–46. doi:10.1016/j.
110:318–20. doi:10.1097/00005537-200002010-00026. amjoto.2011.04.003.
67. Atienza G, López-Cedrún JL. Management of 81. Haus BM, Kambham N, Le D, Moll FM, Gourin
obstructive salivary disorders by sialendoscopy: C, Terris DJ. Surgical robotic applications in oto-
a systematic review. Br J Oral Maxillofac Surg. laryngology. Laryngoscope. 2003;113:1139–44.
2015;53:507–19. doi:10.1016/j.bjoms.2015.02.024. doi:10.1097/00005537-200307000-00008.
68. Karl Storz GmbH and Co. KG G. Telescopes and 82. Terris DJ, Haus BM, Gourin CG, Lilagan PE. Endo-
instruments for sialendoscopy. KARL STORZ robotic resection of the submandibular gland in
GmbH & Co. KG, Germany. Catalog of Oral and a cadaver model. Head Neck. 2005;27:946–51.
Maxillofacial Surgery. 3rd ed. 2012. doi:10.1002/hed.20273.
69. Bruch JM, Setlur J. Pediatric sialendos- 83. Lee HS, Park DY, Hwang CS, Bae SH, Suh MJ, Koh
copy. Adv Otorhinolaryngol. 2012;73:149–52. YW, et al. Feasibility of robot-assisted submandibu-
doi:10.1159/000334474. lar gland resection via retroauricular approach: pre-
70. Gourin CG, Terris DJ. Surgical robotics in otolar- liminary results. Laryngoscope. 2013;123:369–73.
yngology: expanding the technology envelope. Curr doi:10.1002/lary.23321.
Opin Otolaryngol Head Neck Surg. 2004;12:204–8. 84. Walvekar RR, Tyler PD, Tammareddi N, Peters
doi:00020840-200406000-00010 [pii] G. Robotic-assisted transoral removal of a subman-
71. Baek C-H, Jeong H-S. Endoscope-assisted sub- dibular megalith. Laryngoscope. 2011;121:534–7.
mandibular sialadenectomy: a new minimally doi:10.1002/lary.21356.
invasive approach to the submandibular gland. 85. Razavi C, Pascheles C, Samara G, Marzouk
Am J Otolaryngol. 2006;27:306–9. doi:10.1016/j. M. Robot-assisted sialolithotomy with sialendos-
amjoto.2005.11.018. copy for the management of large submandibular
72. Chen M-K. Minimally invasive endoscopic resection gland stones. Laryngoscope. 2016;126:345–51.
of the submandibular gland. Oper Tech Otolaryngol doi:10.1002/lary.25396.
Neck Surg. 2008;19:33–5. doi:10.1016/j. 86. Walvekar RR, Peters G, Hardy E, Alsfeld L,
otot.2008.03.001. Stromeyer FW, Anderson D, et al. Robotic-
73. Song CM, Jung YH, Sung M-W, Kim KH. Endoscopic assisted transoral removal of a bilateral floor of
resection of the submandibular gland via a hairline mouth ranulas. World J Surg Oncol. 2011;9:78.
incision: a new surgical approach. Laryngoscope. doi:10.1186/1477-7819-9-78.
2010;120:970–4. doi:10.1002/lary.20865. 87. Emami B, Lyman J, Brown A, Coia L,
74. Parente Arias PL, Fernández Fernández MM, Varela Goitein M, Munzenrider JE, et al. Tolerance
Vázquez P, de Diego Muñoz B. Minimally invasive of normal tissue to therapeutic irradiation. Int
video-assisted submandibular sialadenectomy: surgi- J Radiat Oncol Biol Phys. 1991;21:109–22.
cal technique and results from two institutions. Surg doi:10.1016/0360-3016(91)90171-Y.
Endosc. 2015; doi:10.1007/s00464-015-4604-0. 88. Zhuang L, Yang Z, Zeng X, Zhua X, Chen Z, Liu
75. Beahm DD, Peleaz L, Nuss DW, Schaitkin B, L, et al. The preventive and therapeutic effect of
Sedlmayr JC, Rivera-Serrano CM, et al. Surgical acupuncture for radiation-induced xerostomia in
approaches to the submandibular gland: a review of patients with head and neck cancer: a systematic
literature. Int J Surg. 2009;7:503–9. d oi:10.1016/j. review. Integr Cancer Ther. 2013;12:197–205.
ijsu.2009.09.006. doi:10.1177/1534735412451321.
76. Ruscito P, Pichi B, Marchesi P, Spriano G. Minimally 89. Kałuzny J, Wierzbicka M, Nogala H, Milecki
invasive video-assisted submandibular siaload- P, Kopeć T. Radiotherapy induced xerostomia:
enectomy: a preliminary report. J Craniofac Surg. mechanisms, diagnostics, prevention and treat-
2007;18:1142–7. ment – evidence based up to 2013. Otolaryngol Pol.
77. Kim HS, Chung SM, Pae SY, Park HS. Endoscope 2014;68:1–14. doi:10.1016/j.otpol.2013.09.002.
assisted submandibular sialadenectomy: the 90. Hackett KL, Deane KHO, Strassheim V, Deary V,
face-lift approach. Eur Arch Otorhinolaryngol. Rapley T, Newton JL, et al. A systematic review of non-
2011;268:619–22. doi:10.1007/s00405-010-1392-y. pharmacological interventions for primary Sjögren’s
78. De Virgilio A, Park YM, Kim WS, Lee SY, Seol syndrome. Rheumatology (Oxford). 2015;54:2025–
JH, Kim S-H. Robotic sialoadenectomy of the sub- 32. doi:10.1093/rheumatology/kev227.
mandibular gland via a modified face-lift approach. 91. Li L, Tian G, He J. The standardization of acu-
Int J Oral Maxillofac Surg. 2012;41:1325–9. puncture treatment for radiation-induced xerosto-
doi:10.1016/j.ijom.2012.04.008. mia: a literature review. Chin J Integr Med. 2015;
79. Terris DJ. Surgical robotics in otolaryngology. In: doi:10.1007/s11655-015-2145-y.
Flint PW, editor. Cummings Otolaryngol, Head 92. Furness S, Worthington H V, Bryan G, Birchenough
Neck Surg. 5th ed. Philadelphia: Mosby-Elsevier; S, McMillan R. Interventions for the management of
2010. p. 38–44. dry mouth: topical therapies. Cochrane Database Syst
Rev. 2011;(12):CD008934. doi:10.1002/14651858. fer for management of xerophthalmia. Zhonghua

CD008934.pub2. Yan Ke Za Zhi. 1998;34:388–90.
93. Jha N, Seikaly H, McGaw T, Coulter L. Submandibular 109. Hockstein NG, Samadi DS, Gendron K, Handler
Salivary Glanf Transfer Prevents Radiation-Induced SD. Sialorrhea: a management challenge. Am Fam
Xerostomia. International Journal of Radiation Physician. 2004;69:2628–34.
Oncology Biology Physics. 2000;46:7-11. 110. Reed J. Surgical management of drooling. Arch
94. MacLeod AM, Robbins SP. Submandibular gland Otolaryngol Head Neck Surg. 2009;135:924–31.
transfer in the correction of dry eye. Aust N Z doi:10.1542/gr.15-2-19.
J Ophthalmol. 1992;20:99–103. 111. Puraviappan P, Dass DB, Narayanan P. Efficacy of
95. Geerling G, Sieg P, Bastian GO, Laqua relocation of submandibular duct in cerebral palsy
H. Transplantation of the autologous submandibular patients with drooling. Asian J Surg. 2007;30:209–
gland for most severe cases of keratoconjunctivitis 15. doi:10.1016/S1015-9584(08)60024-X.
sicca. Ophthalmology. 1998;105:327–35. 112. De M, Adair R, Golchin K, Cinnamond
96. Geerling G, Sieg P. Transplantation of the major MJ. Outcomes of submandibular duct relocation: a
salivary glands. Dev Ophthalmol. 2008;41:255–68. 15-year experience. J Laryngol Otol. 2003;117:821–
doi:10.1159/000131094. 3. doi:10.1258/002221503770716287.
97. Geerling G, Raus P, Murube J. Minor salivary gland 113. Panarese A, Ghosh S, Hodgson D, McEwan J, Bull
transplantation. Dev Ophthalmol. 2008;41:243–54. PD. Outcomesofsubmandibularductre-implantationfor
doi:10.1159/000131093\r131093. [pii] sialorrhoea. Clin Otolaryngol Allied Sci. 2001;26:143–
98. Jha N, Seikaly H, McGaw T, Coulter 6. doi:10.1046/j.1365-2273.2001.00439.x.
L. Submandibular salivary gland transfer pre- 114. Uppal HS, De R, D’Souza AR, Pearman K, Proops
vents radiation-induced xerostomia. Int J Radiat DW. Bilateral submandibular duct relocation for
Oncol Biol Phys. 2000;46:7–11. doi:10.1016/ drooling: an evaluation of results for the Birmingham
S0360-3016(00)80143-1. Children’s Hospital. Eur Arch Otorhinolaryngol.
99. Wu F, Weng S, Li C, Sun J, Li L, Gao 2003;260:48–51. doi:10.1007/s00405-002-0516-4.
Q. Submandibular gland transfer for the preven- 115. Katona G, Csákányi Z, Lorincz A, Gerlinger
tion of postradiation xerostomia in patients with I. Bilateral submandibular duct relocation by high-
head and neck cancer: a systematic review and frequency radiosurgery. Eur Arch Otorhinolaryngol.
meta-analysis. Oral Oncal. 2015;610041:70–86. 2008;265:1103–8. doi:10.1007/s00405-008-0604-1.
doi:10.1159/000371854. 116. Crawford B. Parotid duct transplantation for ocular
100. Su J-Z, Liu X-J, Liu D-G, Ren W-G, Yu xerosis. Trans Aust Coll Ophthalmol. 1970;2:92–5.
G-Y. Sialography of the transplanted submandibular 117. Filatov V, Chevaljev V. Surgical treatment of par-
gland. Ocul Surf. 2014;12:215–20. doi:10.1016/j. enchymatous ophthalmoxerosis. J Opthal. 1951;3:
jtos.2014.02.005. 131–7.
101. Geerling G, Sieg P, Raus P. Salivary gland transplan- 118. Pierce M, Goldberg J, Brooks C. A direct approach
tation. In: Essentials Ophthalmol. Berlin: Springer for transposition of the parotid duct. Arch
Berlin Heidelberg; 2007. p. 208 .Figure 14.1 Ophthalmol. 1960;64:566–70.
102. Murube-del-Castillo J. Transplantation of salivary 119. Zhang H, Zhou Z, Chen Z, Zhao C. Management of
gland to the lacrimal basin. Scand J Rheumatol the dry eye with parotid duct transplantation: a sum-
Suppl. 1986;61:264–7. mary on 40 cases. Yan Ke Xue Bao. 1995;11:67–9.
103. Gelatt K. Canine lacrimal and nasolacrimal diseases. 120. Longo B, Germano S, Laporta R, Belli E, Santanelli
2nd ed. London: Lea & Febiger; 1991. F. Stensen duct relocation after cheek mucosa
104. Sieg P, Geerling G, Kosmehl H, Lauer I, Warnecke tumor resection. J Craniofac Surg. 2012;23:e250–1.
K, von Domarus H. Microvascular submandibular doi:10.1097/SCS.0b013e31824ef809.
gland transfer for severe cases of keratoconjunctivi- 121. Mehta S, Agrawal J, Dewan AK, Pradhan T. Parotid
tis sicca. Plast Reconstr Surg. 2000;106:554–60. duct relocation in buccal mucosa cancer resection.
105. Yu GY, Zhu ZH, Mao C, Cai ZG, Zou LH, Lu L, et al. J Craniofac Surg. 2014;25:1746–7. doi:10.1097/
Microvascular autologous submandibular gland trans- SCS.0000000000000998.
fer in severe cases of keratoconjunctivitis sicca. Int 122. Sakakibara A, Minamikawa T, Hashikawa K,
J Oral Maxillofac Surg. 2004;33:235–9. doi:S0901- Sakakibara S, Hasegawa T, Akashi M, et al. Does sal-
5027(02)90438-8 [pii]\r10.1006/ijom.2002.0438 ivary duct repositioning prevent complications after
106. Kumar PA, Hickey MJ, Gurusinghe CJ, O’Brien tumor resection or salivary gland surgery? J Oral
BM. Long term results of submandibular gland Maxillofac Surg. 2015;73:1003–7. doi:10.1016/j.
transfer for the management of xerophthalmia. Br joms.2014.12.006.
J Plast Surg. 1991;44:506–8. 123. Redman RS. On approaches to the functional res-
107. Paniello RC. Submandibular gland transfer for toration of salivary glands damaged by radiation
severe xerophthalmia. Laryngoscope. 2007;117:40– therapy for head and neck cancer, with a review of
4. doi:10.1097/01.mlg.0000246953.44163.81. related aspects of salivary gland morphology and
108. Jia G, Wang Y, Lu L, Wang X, Li Z. Reconstructive development. Biotech Histochem. 2008;83:103–30.
lacrimal gland with free submandibular gland trans- doi:903247450 [pii]\r10.1080/10520290802374683
124. Baum BJ, Alevizos I, Zheng C, Cotrim AP, Liu S, 130. Lombaert IMA, Hoffman MP. Epithelial stem/
McCullagh L, et al. Early responses to adenoviral- progenitor cells in the embryonic mouse subman-
mediated transfer of the aquaporin-1 cDNA for dibular gland. Front Oral Biol. 2010;14:90–106.
radiation-induced salivary hypofunction. Proc Natl doi:10.1159/000313709.Epithelial.
Acad Sci U S A. 2012;109:19403–7. doi:10.1073/ 131. Ono H, Obana A, Usami Y, Sakai M, Nohara K,
pnas.1210662109. Egusa H, et al. Regenerating salivary glands in
125. Lombaert IMA, Wierenga PK, Kok T, Kampinga the microenvironment of induced pluripotent
HH, Coppes RP. Cancer therapy: preclinical mobi- stem cells. Biomed Res Int. 2015;2015:293570.
lization of bone marrow stem cells by granulocyte doi:10.1155/2015/293570.
colony-stimulating factor ameliorates radiation- 132. Ogawa M, Oshima M, Imamura A, Sekine Y,
induced damage to salivary glands. Clin Cancer Ishida K, Yamashita K, et al. Functional salivary
Res. 2006;12:12–5. doi:10.1158/1078-0432. gland regeneration by transplantation of a bioen-
CCR-05-2381. gineered organ germ. Nat Commun. 2013;4:2498.
126. Tran SD, Liu Y, Xia D, Maria OM, Khalili S, Wang doi:10.1038/ncomms3498.
RW-J, et al. Paracrine effects of bone marrow soup 133. Joraku A, Sullivan CA, Yoo J, Atala A. In-vitro recon-
restore organ function, regeneration, and repair in stitution of three-dimensional human salivary gland
salivary glands damaged by irradiation. PLoS One. tissue structures. Differentiation. 2007;75:318–24.
2013;8:e61632. doi:10.1371/journal.pone.0061632. doi:10.1111/j.1432-0436.2006.00138.x.
127. Pringle S, Van Os R, Coppes RP. Concise review: 134. Sun T, Zhu J, Yang X, Wang S. Growth of min-
adult salivary gland stem cells and a potential ther- iature pig parotid cells on biomaterials in vitro.
apy for xerostomia. Stem Cells. 2013;31:613–9. Arch Oral Biol. 2006;51:351–8. doi:10.1016/j.
doi:10.1002/stem.1327. archoralbio.2005.10.001.
128. Feng J, van der Zwaag M, Stokman MA, van Os 135. Cantara SI, Soscia DA, Sequeira SJ, Jean-Gilles
R, Coppes RP. Isolation and characterization of RP, Castracane J, Larsen M. Selective functional-
human salivary gland cells for stem cell transplan- ization of nanofiber scaffolds to regulate salivary
tation to reduce radiation-induced hyposalivation. gland epithelial cell proliferation and polarity.
Radiother Oncol. 2009;92:466–71. doi:10.1016/j. Biomaterials. 2012;33:8372–82. doi:10.1016/j.
radonc.2009.06.023. biomaterials.2012.08.021.
129. Nanduri LSY, Maimets M, Pringle SA, van der 136. Cecchini MP, Parnigotto M, Merigo F, Marzola P,
Zwaag M, van Os RP, Coppes RP. Regeneration Daducci A, Tambalo S, et al. 3D printing of rat sali-
of irradiated salivary glands with stem cell marker vary glands: the submandibular-sublingual complex.
expressing cells. Radiother Oncol. 2011;99:367–72. Anat Histol Embryol. 2014;43:239–44. doi:10.1111/
doi:10.1016/j.radonc.2011.05.085. ahe.12074.

S.G. Development & Regeneration

Uploaded by

Document Informationclick to expand document information

Document Informationclick to expand document information

Copyright:

Available Formats

S.G. Development & Regeneration

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

S.G. Development & Regeneration

Uploaded by

Copyright:

Available Formats

Seunghee Cha Editor

Advances in Research and

ISBN 978-3-319-43511-4 ISBN 978-3-319-43513-8 (eBook)

Library of Congress Control Number: 2017933465

© Springer International Publishing Switzerland 2017

Printed on acid-free paper

This Springer imprint is published by Springer Nature

to restore secretory function. I am proud to present their collective efforts and

Gainesville Seunghee Cha, DDS, PhD

Part I Updates on Salivary Gland Development

1 Implications of Salivary Gland Developmental Mechanisms

Part II Glandular Damage and Cell Replacement Therapy

4 Histologic Changes in the Salivary Glands

Part III Bioengineering of Salivary Glands

7 Current Cell Models for Bioengineering Salivary Glands . . . . . 133

8 Matrix Biology of the Salivary Gland:

Part IV Therapeutic Considerations for Restoration of

11 Regulation of Salivary Secretion. . . . . . . . . . . . . . . . . . . . . . . . . . 207

Sergio Aguilera, MD INDISA Clinic, Santiago, Chile

Mary C. Farach-Carson, PhD BioScience Research Collaborative,

Robert S. Redman, DDS Oral and Maxillofacial Diagnostic Science,

Increasing our knowledge of how organs develop

© Springer International Publishing Switzerland 2017 3

In order to appropriately translate informa- o­ utlined in depth in the following chapters. In

1.2.1 Morphological Development

r­ eestablishment of the balance between parasym-

Inflammation following radiation is an acute 1.5 Future Prospects

developed. Clearly, reestablishing crosstalk 8. Beetz I, Steenbakkers RJ, Chouvalova O, Leemans

Melinda Larsen, Petko Bogdanov, Ravi Sood,

M. Larsen, PhD (*) • H.R. Kwon, PhD

© Springer International Publishing Switzerland 2017 23

between embryonic submandibular salivary glands grown in vivo and

2.1 Introduction tems biology with reductionist approaches is

2.2 Current Reductionist Numerous transgenic mouse models have identi-

Number of genes 2000

Ex-vivo annotation percentages in

In-vivo annotation percentages in Metabolic process: ex-vivo: 9.3 % in-vivo: 9.4 %

Fig. 2.3 Comparison 0.4

Fig. 2.4 Comparison of

included processes have

2.3.3 Systems-Level Analysis c­left-enriched genes, genes were partitioned into

during branching morphogenesis, which merits subpopulations required for mesenchymal-epi-

Blood vessel development

Sensory organ development

Positive regulation of response

Table 2.2 Putative endothelial-expressed cleft-enriched genes

Table 2.3 Putative neuronal-expressed cleft-enriched genes

I. Castro, PhD • M.-J. Barrera, PhD • U. Urzúa, PhD

© Springer International Publishing Switzerland 2017 45

mucins to extracellular matrix (ECM) support evidence on a self-­

3.1 Structure, Biosynthesis mucins. The second class consists of membrane-­

Table 3.1 Classification and tissue expression of mucins

viscoelastic and high bacteria-agglutinating GSTAPPAHGVTSAPDTRPAP sequence [54]

s­ uggested due to the coincident onset of glandu- 3.3.1 R

4.1 Introduction 83, 100, 101]. Without ­diligent ­professional

© Springer International Publishing Switzerland 2017 75

4.2 Salivary Gland 4.3.1 I llustrations of Normal

Gainesville Seunghee Cha, DDS, PhD

Part IV Therapeutic Considerations for Restoration of

In order to appropriately translate informa- o utlined in depth in the following chapters. In

1.2.1 Morphological Development

r eestablishment of the balance between parasym-

Inflammation following radiation is an acute 1.5 Future Prospects

2.1 Introduction tems biology with reductionist approaches is

2.2 Current Reductionist Numerous transgenic mouse models have identi-

2.3.3 Systems-Level Analysis cleft-enriched genes, genes were partitioned into

mucins to extracellular matrix (ECM) support evidence on a self-

3.1 Structure, Biosynthesis mucins. The second class consists of membrane-

s uggested due to the coincident onset of glandu- 3.3.1 R

4.1 Introduction 83, 100, 101]. Without diligent professional

4.2 Salivary Gland 4.3.1 I llustrations of Normal

d istances into tissues than electromagnetic radia- 4.4.2 R

5.4 Adipose Tissue-Derived

salivaryfunction observed following salivary tumorigenic. However, the components which

intraglandular delivery or systemic delivery, result- 6.2.2 Intrinsic Factors

stem cell (iHepSC) 3,5-diethoxycarbonyl-1,4- capacities of both

7.1 Introduction Hyposalivation is associated with the follow-

7.2.5 SMG-C6 and SMG-C10 7.2.6 Par-C10 and Par-C5

7.3.2 Salivary Spheres 7.4 alivary Gland-Derived Stem

7.5 Scaffolds References

8.1 otivation for Engineering

d evelopment delivered by perlecan’s heparan sul- composed of cytoplasmic scaffolding proteins

Other lung decellularization examples d ecellularized with Triton X/SDC-based decel-

9.1 Introduction surgery of complex renal cell carcinomas in

RP technology. The model material is robust, 9.2.1.2 PolyJet Modeling

9.4 D Bioprinting and Salivary

11.1 Introduction by the three pairs of major glands (parotid, sub-

thought of food. However, few scientists have 11.4 Neural Connections

c hallenges. First, the sorting of the transgene 12.4 Genetic Medicine

13.1 I ntroduction and Historical

13.2.1.3 Sialadenitis in the Pediatric 13.2.1.4 Chronic Sialadenitis

13.2.1.6 Granulomatous Diseases food products (including milk) and domestic or

13.3.3 Procedural Interventions 13.3.3.2 Salivary Gland Transfer

13.3.4.1 Salivary Duct Repositioning