Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.

CAN-14-1154

Cancer
Microenvironment and Immunology Research

Multivalent Forms of the Notch Ligand DLL-1

Enhance Antitumor T-cell Immunity in Lung
Cancer and Improve Efcacy of EGFR-Targeted
Therapy
Asel K. Biktasova1, Duafalia F. Dudimah2, Roman V. Uzhachenko2, Kyungho Park3,
Anwari Akhter4, Rajeswara R. Arasada4, Jason V. Evans4, Sergey V. Novitskiy1,
Elena E. Tchekneva4, David P. Carbone4, Anil Shanker2,5, and Mikhail M. Dikov4

Abstract
Activation of Notch signaling in hematopoietic cells by T cells enhanced expression of T-bet and phosphorylation of
tumors contributes to immune escape. T-cell defects in tumors Stat1/2. Adoptive transfer of T cells from DLL1-treated tumor-
can be reversed by treating tumor-bearing mice with multiva- bearing immunocompetent hosts into tumor-bearing SCID-
lent forms of the Notch receptor ligand DLL-1, but the immu- NOD immunocompromised mice attenuated tumor growth
nologic correlates of this effect have not been elucidated. Here, and extended tumor-free survival in the recipients. When com-
we report mechanistic insights along with the efcacy of com- bined with the EGFR-targeted drug erlotinib, DLL-1 signicant-
binational treatments of multivalent DLL-1 with oncoprotein ly improved progression-free survival by inducing robust
targeting drugs in preclinical mouse models of lung cancer. tumor-specic T-cell immunity. In tissue culture, DLL1 induced
Systemic DLL-1 administration increased T-cell inltration into proliferation of human peripheral T cells, but lacked prolifer-
tumors and elevated numbers of CD44CD62LCD8 mem- ative or clonogenic effects on lung cancer cells. Our ndings
ory T cells while decreasing the number of regulatory T cells and offer preclinical mechanistic support for the development of
limiting tumor vascularization. This treatment was associated multivalent DLL1 to stimulate antitumor immunity. Cancer Res;
with upregulation of Notch and its ligands in tumor-inltrating 75(22); 472841.
2015 AACR.

Introduction regulate the expression of various tissue-specic transcriptional

activators (2, 3).
Notch is a family of evolutionarily conserved transmembrane
An important role for Notch has been proposed in the mod-
receptors and ligands, and regulates a variety of processes in
ulation of T-cell differentiation and immune responses. Evidence
development and differentiation, including cell fate decisions
supports that NotchDLL1 interaction can upregulate T-bet, stim-
(1). The mammalian Notch family includes four cell-bound
ulate IFNg expression, and promote Th1 cell differentiation (4).
Notch receptors, Notch14, and ve Notch ligands DLL1, DLL3,
Conditional transgenic expression of Notch1 intracellular
DLL4, Jagged1, and Jagged2, which are also cell bound. Multiple
domain (ICD) in antigen-specic CD8 T cells induced a central
downstream Notch target genes, including Hes, Hey, and Deltex,
memory phenotype and increased cytotoxicity effects and gran-
zyme B levels (5). Gain-of-function studies indicate that Delta-
like Notch ligands (DLL) promote Th1 commitment of CD4 T
1
Department of Cancer Biology, Vanderbilt University Medical Center, cells (6, 7). By transactivating Th2-promoting target genes IL4 and
Nashville, Tennessee. 2Department of Biochemistry and Cancer Biol- Gata3, Notch can also promote Th2 cell differentiation (8, 9).
ogy, Meharry Medical College School of Medicine, Nashville, Tennes-
see. 3Department of Medicine, Vanderbilt University Medical Center,
Although controversial, the bias is that Jagged ligands are asso-
Nashville, Tennessee. 4Division of Medical Oncology, Department of ciated with Th2-promoting Notch function (6, 10). Unlike other
Internal Medicine, Ohio State University Medical Center, Columbus, ligands, DLL3 is unable to activate Notch in cultured cells and
Ohio. 5Vanderbilt-Ingram Cancer Center, Nashville, Tennessee. seems to inhibit Notch signaling (11).
Note: Supplementary data for this article are available at Cancer Research In vivo, overexpression or inhibition of Notch ligands on
Online (http://cancerres.aacrjournals.org/). antigen-presenting cells (APC) suggested that APC-bound ligands
A.K. Biktasova, D.F. Dudimah, R.V. Uzhachenko, A. Shanker, and M.M. Dikov might specify Th differentiation, with DLL and Jagged supporting
contributed equally to this article. Th1 and Th2 polarization, respectively (1215). In addition to
Corresponding Authors: Anil Shanker, Meharry Medical College School inuencing Th1 and Th2 differentiation, an immunosuppressive
of Medicine, WBSB 2005, 1005 Drive DB Todd Jr. Boulevard, Nashville, function of Notch ligands has also been identied. Expression of
TN 37208. Phone: 615-327-6460; Fax: 615-327-6442; E-mail: [email protected]; Jagged ligands by APCs or hematopoietic progenitors favored
and Mikhail M. Dikov, Ohio State University Medical Center, 460 W. 12th
generation of suppressive regulatory T cells (Treg) in vivo (1618).
Avenue, 484 BRT, Columbus, OH 43210. E-mail: [email protected]
Regulation of IL17 and RORgt gene promoters and activation of
doi: 10.1158/0008-5472.CAN-14-1154 Th17 differentiation has also been reported for Notch ligands

2015 American Association for Cancer Research. (19). These data clearly conrm the immune modulatory function

4728 Cancer Res; 75(22) November 15, 2015

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

of Notch ligands. However, no information is available on the tumor was evaluated by MRI in vivo, as described previously
role of Notch ligand-specic signaling in antitumor immune (24). For this model, tumor recurrence was recorded when
effector functions. tumor volume exceeded by 30% the residual volume after
Our recent work revealed a mechanistic link in the molecular erlotinib treatment.
pathways underlying the tumor-induced perturbation of hemato-
poietic Notch signaling and demonstrated that altered expression DLL1 clusters and treatment regimen
of Notch ligands attenuated Notch signaling in the hematopoietic Mouse or human DLL1Fc fusion protein is composed of the
compartment of tumor-bearing host as a means of causing extracellular domain of mouse or human DLL1 and the Fc part of
immunosuppression. This Notch-mediated immune suppression mouse IgG2A or human IgG1, respectively. To form DLL1 clusters,
could be reversed by the enhanced DLL1-mediated Notch signal- DLL1-Fc, biotinylated anti-IgG antibodies, and NeutrAvidin
ing in hematopoietic microenvironment (2022). This predicted (Pierce) were mixed at a molar ratio of 1:4:10 in PBS, as described
a novel therapeutic approach based on the stimulation of Notch earlier (21, 26). As a control in all applications, Fc fragment of
signaling using soluble multivalent form of DLL1 to overcome mouse IgG2 (Sigma-Aldrich) was used instead of DLL1Fc.
cancer-associated immunosuppression, stimulate antitumor Mouse DLL1Fc and biotinylated donkey anti-mouse IgG anti-
immunity and attenuate tumor growth. bodies were from R&D Systems; human DLL1-Fc and biotinylated
In the present study, we evaluated the immunologic correlates goat anti-human IgG antibodiesfrom Enzo Life Sciences, Inc.
of the systemic activation of Notch signaling using clustered DLL1 Tumor-bearing mice received clustered DLL1 at doses of 0.15
and its efcacy in combination with oncogene-targeted treatment mg/kg (4 mg/injection) of DLL1-Fc protein in 100 mL of PBS i.p.
in the mouse lung cancer model. We show that DLL1-based every other day (length of treatment is indicated in the gure
therapy can induce robust tumor antigen-specic T-cell effector legends and Results section). The control group received control
and memory responses, enhance T-cell inltration into the tumor, clusters with Fc fragments instead of DLL1Fc protein. Twice
while decreasing Treg differentiation and tumor angiogenesis higher doses of clustered DLL1 were used in some experiments
without increasing the tumorigenic potential of cancer cells. Such with similar results suggesting dose saturation of the clustered
an activation of DLL1Notch signaling suppressed tumor growth DLL1 effects.
in wild-type mice as well as provided signicant therapeutic In mutant EGFR tumor model, mice were treated with clustered
benet following an adoptive T-cell transfer into tumor-bearing DLL1 or control clusters, as above, from days 12 to 28 after tumor
SCID-NOD mice. Combined with mutant EGFR-targeted treat- induction by doxycycline, whereas erlotinib was given during
ment by erlotinib, multivalent DLL1 signicantly improved pro- days 15 to 25 daily at a dose of 50 mg/kg, i.p., as previously
gression-free survival (PFS). This supports the potential therapeu- described (24).
tic utility of multivalent Notch ligand in cancer treatment settings. In separate experiments, nontumor mice Balb/c mice
received clustered DLL1 or control clusters injections every
Materials and Methods other day for total of three times. Hematopoietic tissues from
Cell lines these mice were collected on the second day after the last
The human lung cancer cell lines (H157, H460, HCC15, injection and evaluated for the expression of Notch receptors,
HCC1437, HCC1264, and HCC2469) and murine Lewis lung Notch ligands, and downstream Notch target genes Hes1, Hey1,
carcinoma (LLC) cell line were obtained from the ATCC; low- and Deltex by qRT-PCR.
passage (less than 10) cultures were used for the experiments.
D459 cells are murine broblasts malignantly (murine brosar- Immunologic assays
coma) transformed in our laboratory by transfection of human D459 cells have a dened mutant p53 antigenic peptide (FYQ-
Ras and mutant human p53 (21, 23). Our laboratory is the LAKTCPVQL, aa 128-139; ref. 27). Induction of antigen-specic
primary source of these cells, and we regularly go back to reference responses in this model was characterized by evaluation of IFNg-
stocks to ensure delity; routine sterility and Mycoplasma testing producing T cells, as follows: splenocytes or LN cells from D459
were performed regularly. tumor-bearing mice treated with clustered DLL1 or control clus-
ters were stimulated with 10 mmol/L of mutant p53 or control
Mice and tumor models peptide for 60 hours; IFNg intracellular staining was performed
Female Balb/c, C57BL/6, and SCID/NOD mice (7- to 8-week- using Mouse Intracellular Cytokine Staining Kit (BD Pharmingen)
old) were purchased from The Jackson Laboratory. Mutant EGFR according to the manufacturer's recommendations. Data were
tetracycline-inducible transgenic mouse line that expresses an acquired with FACSCalibur ow cytometer (BD Immunocytome-
L858R mutant human EGFR in lung epithelial cells was described try Systems). Gates were set on CD8 or CD8CD44CD62L
earlier and provided by Dr. William Pao (Vanderbilt University, cells. LLC cells also have a dened antigenic peptide MUT1
Nashville, TN; ref. 24). (spontaneously mutated connexin 37, FEQNTAQP (28, 29).
The animals were housed in pathogen-free units at the Splenocytes and lymph node (LN) cells (2.5  105 cells/well)
Vanderbilt University School of Medicine, in compliance with from LLC tumor-bearing mice treated with control or DLL1
the Institutional Animal Care and Use Committee regulations. clusters were stimulated with 10 mmol/L of MUT1 or control
To induce tumor, mice were inoculated s.c. in ank with 0.3  peptide for 48 hours and IFNg-producing cells were enumerated
106 D459 or LLC cells, as described previously (21, 25). For by ELISPOT assay (CTL) according to the manufacturer's protocol.
s.q. models, tumor volume was measured with calipers and For the mutant EGFR model, lungs were assessed for the inltra-
tumor tissues were weighed at the endpoint of the experi- tion by IFNg-producing cells and other immune cells. Lung single-
ments. In mutant EGFR mouse model, tumor growth was cell suspensions were prepared, as described previously (25).
induced and sustained for the length of the experiment by IFNg-producing cells were enumerated by intracellular staining
providing mice with doxycycline in chow and the size of lung and inltration by immune lineages was assessed by ow

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4729

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Biktasova et al.

cytometry (see below). CD45 cells for evaluation of Notch (Abcam); GAPDH (Santa Cruz Biotechnology). All antibodies
signaling were isolated from lung single-cell suspensions, as recognize epitopes in ICD of Notch.
described earlier (30). To determine the interaction of clustered DLL1 with different
Peptides were synthesized by the American Peptide Company, Notch ligands, mouse thymus lysate in the above buffer was
Inc. prepared and proteins were precipitated with a complex of
DLL1-Fc with anti-Fc antibodies bound to protein-G magnetic
Flow cytometry beads (Pierce); as a control, Fc fragment was used instead of DLL1-
Fluorochrome-labeled cell-surface marker or intracellular pro- Fc. Beads were then washed and mixed with SDS sample buffer for
tein specic antibodies were obtained from BD Bioscience Phar- the subsequent Western blot analysis of bound proteins.
mingen and eBioscience, Inc. For staining of cell surface markers, Note that as full-length Notch is composed of two noncova-
cells were incubated with the antibodies for 20 minutes on ice. For lently bound domains, on Western blot analysis in denaturing/
intracellular cytokines, FoxP3, Stat, or phospho-Stat (p-Stat) cells reducing conditions with antibodies to ICD epitopes, it appears as
were rst stained for lineage-specic markers, and then permea- ICD band of approximately 110 kD or 90 kD for Notch1 and 2 or
bilized for 20 minutes with BD xation/permeabilization kit and Notch3 and 4, respectively. In some cases, in cancer cells where
incubated with uorochrome-labeled or unlabeled specic anti- Notch expression is high, full-length Notch could also be seen
bodies for 30 minutes on ice. When unlabeled primary antibodies under these conditions.
were used, cells were washed, and then stained with uorochrome-
conjugated secondary antibodies. Matched uorochrome-conju- Immunohistochemistry
gated isotype IgG controls were used. Flow-cytometry data were Tumor tissue was extracted, xed in 10% formalin, embedded
acquired using a FACS LSR II (BD Immunocytometry) and ana- in parafn, and sectioned (5 mm). Slides were stained with
lyzed with FlowJo software (Tree Star). Nonviable cells were peroxidase-labeled antibodies to the CD3e (Santa Cruz Biotech-
excluded by using 7-amino actinomycin D. Antigen negativity was nology), CD11b, Gr1, or CD34 cell surface markers (Novus
dened as having the same uorescent intensity as the isotype Biologicals). The number of cells or tumor blood vessels identi-
control. ed by peroxidase substrate staining was counted using a Nikon
Eclipse E600 with a 40/1.0 NA Plan Apochromat oil objective
Adoptive T-cell transfer (Nikon Instruments) and with Olympus DP-11 digital camera
Splenocytes and tumor-draining LN cells from D459 tumor- (Olympus America). Images were processed using Meta-Morph
bearing mice were collected on day 25 after inoculation of D459 5.0.7 (Universal Imaging) and ImageJ (NIH, Bethesda, MD)
cells and mixed; then, 5  106 cells were injected into retro-orbital software.
plexus of SCID-NOD mice bearing palpable (34 mm) D459
tumors. Tumor growth was monitored and tumors weighted at Cell proliferation and colony-forming assays
the end of the experiment. Human peripheral blood was drawn from healthy consented
donors according to the protocol approved by the Vanderbilt
Expression levels of Notch receptors, ligands and downstream University School of Medicine Internal Review Board. Peripheral
targets, and transcription factors blood mononuclear cells (PBMC) were isolated by gradient
Quantitative RT-PCR (qRT-PCR) was used to quantify expres- centrifugation using Ficoll-Paque (GE Healthcare). Proliferation
sion of Notch downstream target genes, receptors and ligands as of T cells was measured using the intracellular dye carboxyuor-
well as T-bet, Gata3, RORgt, and FoxP3 transcription factors in escein succinimidyl ester (CFSE; Molecular Probes; Life Technol-
samples of mouse hematopoietic tissues or tumor cells using ogies). PBMC were stimulated with Dynabeads Human T-Acti-
primers described earlier (21, 31). RNA was extracted with an vator (anti-CD3, anti-CD28, and anti-CD137 antibodies coupled
RNeasy Mini Kit and possible genomic DNA contamination was to beads; Life Technologies), as recommended by the manufac-
removed by on-column DNase digestion using the RNasefree ture, with 0.5 mg/mL multivalent DLL1 (based on DLL1-Fc pro-
DNase set (Qiagen). cDNA was synthesized using the SuperScript tein) or control clusters for 4 days. T-cell proliferation was assessed
III Reverse Transcriptase Kit (Invitrogen). cDNA, iQ SYBR green by CFSE dilution in CD3 cells by ow cytometry.
supermix (Bio-Rad) and gene-specic primers (see in Supplemen- To evaluate the proliferation of human and murine cancer cells,
tary Table S1) were used in 20 mL PCR reactions as recommended cells were seeded in 96-well plates and cultured for 16 hours in
by the manufacturer. Amplication of endogenous b-actin or DMEM medium supplemented with 10% FBS. Then, 1 mg/mL
GAPDH was used as internal controls. clustered DLL1 (based on DLL1-Fc protein) or control clusters
were added and cells were cultured for additional for 24 hours; for
Western blot and ligand precipitation the last 2 hrs of incubation 1 mCi [3H-thymidine] per well was
Cells or tissues were lysed in a lysis buffer containing 20 added and [3H-thymidine] incorporation was assessed by liquid
mmol/L HEPES, 150 mmol/L NaCl, 10% glycerol, 1% Triton X- scintillation counting, as described earlier (33).
100, 1 mmol/L EGTA, and 1.5 mmol/L MgCl2 with set of Colony formation of human lung cancer cells in soft agar was
inhibitors, as described previously (32). Equal amounts of performed, as previously described (34), with 2,500 cells seeded
protein were mixed with SDS sample buffer and separated by in 6-well plate in DMEM with 10% FBS. Colonies were counted
7.5 or 10% SDS-PAGE, and transferred to polyvinylidene after 2 weeks.
diuoride membrane (Amersham Biosciences). The following
antibodies were used for detection: Notch1 (Cell Signaling Statistical analysis
Technology); Notch2 (Origene Technologies); Notch3 and Data were analyzed using the GraphPad Prism 4.0 software
DLL4 (Abcam); Notch4 (Millipore); Jag1 (Cell Signaling Tech- (GraphPad Software Inc.) and presented as mean  SEM. Com-
nology); Jag2 (Thermo Scientic); DLL1 (Sigma-Aldrich); DLL4 parisons between treatment and control groups were performed

4730 Cancer Res; 75(22) November 15, 2015 Cancer Research

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

using one-way ANOVA followed by Dunnett's posttests. Compar- that systemic administration of clustered DLL1 in LLC tumor-
isons between two groups were performed using two-tailed bearing mice stimulated phosphorylation of Stat1 and Stat2
unpaired t tests. Survival curves were compared using the Man- transcription factors in CD4 T cells (Fig. 2A and B) that are
telHaenszel log rank test. Values were considered statistically associated with Th1 differentiation. Enhanced Stat1 signaling in
signicant when a P value was less than 0.05. CD4 T cells from DLL1-treated mice correlated with the increase
in the expression of T-beta mediator of transcriptional effects of
Stat1 on T-cell differentiation. Among the lineage-specic tran-
Results scription factors involved in the regulation of Th cell differenti-
Multivalent DLL1 interacts with Notch receptors and ation, only T-bet gene expression displayed signicant upregula-
upregulates hematopoietic Notch signaling in vivo tion, whereas expression of Gata3, RORgt, and FoxP3 genes, as
Activation of Notch receptor proteolytic cleavage and signaling analyzed in a pool of splenocytes and LN cells from treated LLC-
requires a multivalent interaction between Notch receptors and bearing mice, did not show any signicant change (Fig. 2C).
ligands, whereas soluble forms of ligands act as Notch inhibitors Statistically signicant upregulation in phosphorylation of Stat3,
(35). In this study, we used a multivalent or clustered form of responsible for the survival of activated T cells (22), was also
DLL1, which was a complex of DLL1IgG Fc fusion proteins with detected, thus suggesting improved T-cell survival (Fig. 2A).
biotinylated anti-Fc antibody and avidin (21), acting as an acti-
vator of Notch. Clustered DLL1 therapy improves antitumor T-cell function
Notch system appears to be very sensitive to modulation by its and memory
ligands. We performed ligand precipitation experiments to deter- We demonstrated earlier using different mouse models that
mine the Notch receptors that bind clustered DLL1. DLL1-Fcanti- therapeutic enhancement of DLL1/Notch signaling produces
Fc antibody complex or Fcanti-Fc antibody complex, as a con- signicant T-cellmediated attenuation of tumor growth (21).
trol, were bound to protein G magnetic beads and the beads were Here, we investigated whether such therapy is capable of enhanc-
added to the mouse thymus lysate to pull down the interacting ing tumor-specic immune responses and generating specic
Notch receptors. Western blot analysis of the precipitated proteins tumor-protective T-cell memory in lung tumor models, LLC and
revealed that all four Notch receptors interact with clustered DLL1, D459, where tumor-specic antigenic peptides have been iden-
thus suggesting that each of them could be involved in mediation tied, thus allowing the assessment of tumor-specic immune
of the observed effects of the enhanced DLL1 signal (Fig. 1A). responses.
To explore the effects of clustered DLL1 on hematopoietic Treatment of mice with clustered DLL1 or control cluster for 10
Notch system in vivo, clustered DLL1 was injected in healthy mice days after s.c. injection of LLC cells elicited strong antigen-specic
i.p. every other day for a total of three doses and Notch signaling CTL response to the endogenous LLC tumor antigen MUT1.
was evaluated on the second day after the last administration. Higher number of IFNg-secreting cells were noted in spleens and
qRT-PCR analysis demonstrated that such treatment sustained LNs of mice treated with DLL1 clusters than in the control group
signicantly elevated levels of Notch target genes (Fig. 1B). The after restimulation with tumor antigenic peptide MUT1 (Fig. 2D).
clustered DLL1 reagent seems to deliver activating DLL1 signals to This correlated with signicantly smaller tumor mass in clustered
all hematopoietic organs, as changes in the expression of one or DLL1-treated mice than in control clusters-treated animals (not
more Notch genes are detectable in all organs except LNs, which shown). These results suggest high efcacy of clustered DLL1 as an
could be due to the low vascularization/circulation of LNs or be an immunization adjuvant.
attribute of the Notch system response in LN cells. In D459 model, s.c. tumor appeared on days 7 to 8 after cell
Clustered DLL1 also altered receptor and ligand expression inoculation and developed rather slowly for an additional 10 to
patterns in these sites (Fig. 1C and D). The expression pattern of 12 days, after which tumor grows exponentially (Fig. 3A). Clus-
Notch receptors and ligands appears to be tissue specic. Bone tered DLL1 or control clusters were administered after tumors
marrow, blood, and spleen show signicantly increased Notch were established (tumor diameter 45 mm) from days 7 to 19
signaling as well as the expression of Notch ligands following every other day (Fig. 3A). Clustered DLL1 delayed tumor growth
clustered DLL1 administration (Fig. 1C and D). High levels of when compared with the control cluster (Fig. 3A). Immunologic
Notch ligand expression in these organs might associate with the parameters were examined on day 21 when the differences in
high number of myeloid cells, which are known to be a source of tumor size in control and clustered DLL1 groups were still
Notch ligands for the differentiating lymphocytes (6, 7). The insignicant. This excluded variations in the systemic immuno-
magnitude of Notch receptor expression changes is highest in logic effect due to tumors of differing sizes. Signicantly higher
the spleen and thymus, which corresponds to the high number of levels of T-cell activation marker CD25 and intracellular IFNg
lymphoid cells in these tissuesthe recipients of the activating production were observed in the splenic and LN CD8 T cells
signals from the ligands (Fig. 1C). The increased expression of following re-challenge with D459 tumor antigenicmutant p53
Notch receptors and ligands upon pharmacologic DLL1-mediated peptide (Fig. 3B). Moreover, multivalent DLL1 therapy resulted in
stimulation may result in the amplication of the initial signal. a signicant increase of splenic CD44CD62L CD8 T cells
This might explain why relatively low doses of clustered DLL1 characterized as central memory effector T cells (Fig. 3C and
produce signicant biologic effects. D). Among CD44CD62L CD8T cells, there were signicantly
more IFNg-producing T cells after re-stimulation with the cog-
Pharmacologic enhancement of DLL1-mediated Notch nate-mutant p53 peptide, thus indicating increased number and
signaling supports effector T-cell differentiation and survival in function of tumor-specic memory T cells (Fig. 3E). In addition to
tumor-bearing mice stimulating robust antigen-specic T-cell responses, systemic acti-
Notch signaling plays an important role in regulating differ- vation of DLL1/Notch signaling resulted in moderate, but statis-
entiation of naive CD4 T cells into distinct Th lineages. We found tically signicant reduction of the number of regulatory T cells in

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4731

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Biktasova et al.

A B 0.2 **
Precipitation ** *

Ratio to -Actin, (%)

Control : + - + - + - + -
0.15 Control
Clust DLL1: - + - + - + - +
* Clust DLL1
kD
250 0.1 **
150 *
100
0.05
75 *

BM Blood Spleen Thymus LN

**
C 0.5 Control
**
Clust DLL1
0.4
Ratio to -Actin (%)

0.3 **
**
0.2 ** *

0.1

*
0
L1

1
2

L1

1
2

L1

1
2

L1

1
2
LN
L1

1
2
L4

L4

L4

L4

L4
Jag
Jag

Jag
Jag

Jag
Jag

Jag
Jag

Jag
Jag
DL

DL

DL

DL

DL
DL

DL

DL

DL

DL
BM Blood Spleen Thymus LN

D 2 **
Control

Clust DLL1
1.5
Ratio to -Actin (%)

**
** *
1
**
*
0.5 *
*
0
No 1
No 2
No 3
4

No 1
No 2
No 3
4

No 1
No 2
No 3
4

No 1
No 2
No 3
4

No 1
No 2
No 3
4
tch
tch
tch
tch

tch
tch
tch
tch

tch
tch
tch
tch

tch
tch
tch
tch

tch
tch
tch
tch
No

No

No

No

No

BM Blood Spleen Thymus LN

Figure 1.
Clustered DLL1 binds to four Notch receptors, upregulates Notch signaling, and modulates expression of hematopoietic Notch genes in vivo. A, precipitation of
Notch receptors from mouse thymus lysate by DLL1-Fc/anti-Fc antibody or Fc/anti-Fc antibody (control) complexes bound to protein G beads; precipitated
proteins were separated by Western blot analysis and visualized using antibodies to Notch 1, 2, 3, or 4. BD, mRNA expression of downstream Notch target
genes Hes1 and Hey1 (B), Notch ligands (C), and Notch receptors (D) in hematopoietic organs of mice treated with clustered (clust) DLL1. Mice received
three injections of clustered DLL1 or control clusters i.p. every 2 days. Gene expression was evaluated on the second day after the last injection by qRT-PCR.
Mean  SEM, 68 mice per group,  , P < 0.05;   , P < 0.01. BM, bone marrow.

4732 Cancer Res; 75(22) November 15, 2015 Cancer Research

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

Control
A B
Figure 2. 0.2%
Enhancement of DLL1-mediated Notch
8 ** Control
signaling promotes T-cell
differentiation and survival in tumor- Clust DLL1
6
bearing host and elicits tumor antigen

)ROGFKDQJH
specic CTL responses in tumor- **

CD4
bearing mice. LLC tumor-bearing mice
4
were treated with clustered DLL1 or
control clusters i.p. every 2 days for * Clust DLL1
10 days. Gene expression or Stat 2
phosphorylation was evaluated by 1.2%
qRT-PCR or intracellular protein
immunouorescence staining, 0
respectively, in a pool of splenocytes
and LN cells; expressed as fold increase
in clustered DLL1-treated mice over the
control clusters group. A, changes in

phospho-Stat proteins in gated CD4
p-Stat1
cells following intracellular ow-
cytometry staining. B, representative
phospho-Stat1 versus CD4 dot plot. Control
C, expression of transcription factors C D 
regulating T-cell differentiation Clust DLL1
assessed by qRT-PCR. Mean  SEM; 5 3 **
Control
*
mice per group;  , P < 0.05;   , P < 0.01.  Control
D, IFNg-producing cells to endogenous 2.5 Clust DLL1
LLC tumor antigen MUT1 were

Number
2 
)ROGFKDQJH

enumerated by ELISPOT assay in a

mixture of splenocytes and LN cells of
LLC tumor-bearing mice treated with 1.5 
control or DLL1 clusters following Clust DLL1
restimulation in vitro with tumor 1
antigenic peptide MUT1 for 48 hours. 
Mice received clustered DLL1 or control 0.5
treatment for 10 days immediately after

injection of LLC cells. Mean  SEM; 0
5 mice per group;  , P < 0.05. Tbet Gata3 RORt FoxP3 MUT1-specific IFN-
producing cells

the spleen of treated animals (Fig. 3F). The combination of these ably higher (2.65-fold) T-cell inltration into tumors as
effects might have contributed to the observed inhibitory effect on assessed by CD3e immunostaining of D459 tumor sections
tumor growth. (Fig. 4C), a factor known to correlate with the improved
Induction of DLL1-induced T-cell effector memory and pro- prognosis in human patients (36). In this model, no signicant
tective immunity was further conrmed in the adoptive T-cell differences were found in the number of tumor-inltrating
transfer experiments. A total lymphocyte fraction from a pool Gr1 or CD11b myeloid cells between clustered DLL1-treated
of splenocytes and tumor-draining LN cells, to have a higher and control groups (data not shown). D459 tumors staining
frequency of tumor antigen-specic T cells, from D459 tumor- with endothelial marker CD34 revealed signicantly decreased
bearing Balb/c mice treated with clustered DLL1 or control vascularization of tumors in multivalent DLL1-treated animals
clusters were transferred i.v. into SCID-NOD mice bearing than in control animals (Fig. 4D). This result is in line with the
palpable D459 tumors. Lymphocytes transferred from clustered observation that DLL1-induced Notch signaling has suppres-
DLL1-treated donors, but not from the control-treated animals, sive effect on tumor growth in B16 melanoma model due to the
signicantly attenuated tumor growth in SCID-NOD mice attenuated vascularization (37).
(Fig. 4A and B). These data suggest that the antiangiogenic effect of multiva-
These data strongly suggest that the multivalent DLL1-medi- lent DLL1 therapy together with the enhanced antitumor T-cell
ated Notch activation possesses functional capacity to induce responses contribute to tumor-inhibitory effects in therapeutic
tumor-specic T-cell responses and memory, resulting in the settings.
signicant therapeutic benet in tumor models. They imply
strong association of the DLL1Notch axis in regulation of the Clinical and immunologic effectiveness of the multivalent
T-cellmediated antitumor immunity. DLL1 in combination with mutant EGFR oncogenetargeted
therapy associates with the enhanced Notch signaling and
Increased tumor inltration by immune cells and decreased improved immune responses
tumor vascularization in mice treated with clustered DLL1 We tested multivalent DLL1 therapy in combination with
Additional effects of the pharmacologic DLL1-mediated mutant EGFR oncogenetargeted inhibition in the EGFRL858R
Notch activation in tumor-bearing host associate with remark- transgenic mouse model. Treatment of patients with tumors

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4733

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Biktasova et al.

A B 100
Control Clust DLL1

1,400 Control peptide

Control 80
p53 peptide
Tumor volume, (mm3)

1,200

% of Max
60
6.42 6.36 8.17

1,000
Clust DLL1 40
8.3 54.2 9.41
20

800 0
0 1 2 3 4 0 1 2 3 4 0 1 2 3 4
10 10 10 10 10 10 10 10 10 10 10 10 10 10 10

600 ** CD25 Isotype

100 Figure 3.
400 * 80 Attenuation of tumor growth by

% of Max
clustered DLL1 correlates with the
200 60

7.83 8.26 7.94

40
improved antitumor T-cell immunity.
10.1 62.8 8.68
0 Mice inoculated with D459 tumor cells
20

0 19 21 23 25 were treated with clustered DLL1 or

0

Days 10
0
10
1
10
2
10
3 4
10 10
0
10
1
10
2
10
3 4
10 10
0
10
1
10
2
10
3
10
4 control clusters i.p. every 2 days from
IFN Isotype days 7 (after tumors reached 45 mm)
to 19. A, D459 tumor growth. Mean 
SEM; 8 mice per group;  , P < 0.05;

C D , P < 0.01. B, surface expression of
80 Control activation marker CD25 and
intracellular staining for IFNg in
* Clust DLL1
cultures of splenic CD8 T cells
CD44

60 isolated from mice on day 21 following

Cells, %

restimulation in vitro with p53 D459

tumor antigenic or control peptide
40 for 60 hours. Numbersmean
uorescence intensity for control
20 (left) and cognate (right ) peptide-
CD62L stimulated cells. Representative
histograms of total 4. C, proportion of

0 memory CD44 CD62L CD8 T cells in
CD44+CD62L+ splenocytes of mice. D, representative
CD8+ T cells CD44 versus CD62L ow cytometry

dot plot on gated splenic CD8 T cells
from a total of 5 mice. E, increased
number of IFNg-producing T cells
E F
within splenic CD44 CD62L memory
20 * Control 3
CD8 T-cell population. F, decreased
Control
Clust DLL1 number of regulatory T cells in spleen
2.5 Clust DLL1 of clustered DLL1 compared with
15
control cluster-treated animals.
Cells, %

2
Cells, %

C, E, F, mean  SEM; 5 mice per group;

10 1.5
* 
, P < 0.05.

1
5
0.5
0 0
IFN-producing Treg cells
CD44+CD62L+ CD8+ T cells

bearing activating EGFR mutations with EGFR inhibitors repre- mutant leads to the development of lung adenocarcinomas in 2 to
sents an example of successful oncogenic pathwaytargeted ther- 3 weeks after doxycycline induction with erlotinib treatment
apy. EGFR gene in-frame deletions in exon 19 and L858R muta- causing rapid tumor regression (24). In our regimen, during the
tion in exon 21 constitute nearly 90% of the lung adenocarcinoma doxycycline tumor induction, mice received two injections of
somatic-activating mutations and have been associated with clustered DLL1, and then a combination of erlotinib with clus-
sensitivity and rapid clinical response to the EGFR tyrosine kinase tered DLL1 followed by two more injections of clustered DLL1;
inhibitors (TKI) getinib and erlotinib (38, 39). However, in most mice then were left untreated (Fig. 5A). The control group received
of responding patients, the cancer resumes detectable growth control clusters instead of multivalent DLL1. Tumor size was
within several months (38, 40). We tested whether integrating monitored at different time points by MRI with tumor recurrence
the multivalent DLL1-based immunotherapy with oncogene-tar- determined when the volume of tumor exceeded the residual
geted TKI would induce sustained immune responses and long- tumor volume after erlotinib treatment by 30%. EGFRL858R
lasting remission in sensitive tumors. We used a tetracycline- mutant mice were highly responsive to the clustered DLL1 com-
inducible transgenic mouse line that expresses an L858R-mutant bination therapy, as seen by the decreased lung tumor burden and
human EGFR in lung epithelial cells (24). The expression of EGFR signicantly improved PFS (Fig. 5A and B).

4734 Cancer Res; 75(22) November 15, 2015 Cancer Research

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

A 800 B
Control donor Control
1 donor

Tumor volume, (mm3)

600
Clust DLL1 donor
Clust DLL1
0.8 donor

**

Weight, (g)
400 0.6
**
0.4
Figure 4.
200
Multivalent DLL1 therapy elicits tumor * 0.2
antigen-specic T-cell memory,
enhances tumor inltration by T cells,
and attenuates tumor angiogenesis.
D459 tumor growth (A) and weight (B)
0 0
in SCID-NOD mice that received 0 16 18 21 Tumor weight
lymphocyte transfer from donor mice Days
bearing D459 tumor and treated with
clustered DLL1 or control clusters.
6
A total of 5  10 cells of the total
lymphocyte fraction from a pool of
C 140 * Control
RBC-depleted splenocytes and tumor- Clust DLL1
draining LN cells were harvested from
120 CD3+ cells
&HOOQXPEHU

donor mice at day 21 (see Fig. 3) and 100

transferred into SCID-NOD mice
bearing palpable (34 mm) D459 80
tumors at day 5. Mean  SEM; 5 SCID-
NOD mice per group;  , P < 0.05; 60

, P < 0.01. C and D, immunostaining of
tumor tissues with CD3e (C) or CD34
40
(D) antibodies. D459 tumor-bearing 20
mice were treated with DLL1 or control
clusters, as in Fig. 3, and tissue 0
sections were prepared on day 21. CD3+ cells Control Clust DLL1
Representative images and the

numbers of inltrating CD3e T cells or

CD34 tumor vessels are presented. D 120 Control
Mean  SEM; 5 mice per group; 10 elds
CD34+ endothelial cells
1XPEHURIYHVVHOV

on two nonadjacent sections were 100 Clust DLL1

counted for each sample;  , P < 0.05.
80
60
*
40
20
0
CD34+ cells Control Clust DLL1

Analysis of the hematopoietic Notch signaling, protein expres- found remarkably increased numbers of INFg-producing T cells
sion, and immunologic parameters revealed that the observed and CD11bCD11chigh dendritic cells (DC), despite the moderate
therapeutic effects correlated with the enhanced Notch signaling decrease in the total inltrating CD3 T cells. Worth noting is also
and improved immune responses (Fig. 6). Treatment with mul- the increased number of CD19 B cells (Fig. 6C). The data suggest
tivalent DLL1 signicantly upregulated the expression of down- that the enhancement of DLL1/Notch signaling provides benet
stream Notch targets Hes1, Hey1, and Deltex1 in lung-inltrating in combination treatment with oncogene-targeted drugs due to
immune cells of tumor-bearing EGFRL858R transgenic animals as the improved antitumor immunity.
well as enhanced the expression of splenic Delta-like ligands, Jag2
and Notch1, 2, and 3 receptors, thus apparently reversing previ- DLL1-Notch signaling enhances human peripheral T-cell
ously observed tumor-induced deciency in hematopoietic Notch proliferation without directly stimulating tumorigenicity
signaling and ligand expression (Fig. 6A and B; ref. 21). Admin- In human adults, peripheral T cells play a key role in
istration of clustered DLL1 resulted in signicant alterations in the mediating immune responses. We thus tested whether multi-
numbers of immune cells inltrating the diseased lungs. We valent DLL1 would have direct effect on human peripheral

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4735

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Biktasova et al.

Control Clust DLL1

A &OXVW '//GD\V

(UORWLQLEGD\V

Erlotinib+Control Control Clust DLL1

200 Erlotinib+Clust DLL1

160 Control Clust DLL1

Tumor volume, (mm3)

Figure 5.
Multivalent DLL1 signicantly
120 improves PFS in combination with
EGFR oncogenetargeted treatment
in the EGFRL858R transgenic mouse
Control Clust DLL1 model. Transgenic EGFRL858R mice
80
with induced lung tumors were treated
with erlotinib in combination with
** clustered DLL1 or control clusters, as
40 * shown in A. A, lung tumor (white
* opacities) growth was evaluated by
MRI and volume quantied. Insets,
0 representative MRI images at the
15 25 40 50 60 Days corresponding time points. B, PFS;
recurrence was determined when
tumor volume exceeded by 30%
B 120
Progression-free survival, (%)

Control residual tumor volume after erlotinib

treatment. Mean  SEM; 8 mice per
100 Clust DLL1 group;  , P < 0.05;   , P < 0.05.

80

60

40

20

0
0 10 20 30 40 Days

T-cell function. PBMCs from human donors were stimulated with abnormalities were noted, nor was there any substantial changes
beads-coupled CD3, CD28, and CD137 antibodies with or with- in the numbers of red or white blood cells, lymphocytes or
out multivalent DLL1 for 4 days. Proliferation of gated CD3 T platelets counts in the peripheral blood following DLL1 treat-
cells, as assessed by CFSE dilution, demonstrated that clustered ments (data not shown).
DLL1 enhanced proliferation of human peripheral T cells (Fig. 7A).
The pleiotropic functions of Notch and complex effect of
interference with this signaling pathway raise legitimate safety Discussion
concerns regarding systemic activation of Notch signaling by the T-cell immune surveillance against tumors is well established.
multivalent DLL1. We assessed the effect of this reagent on However, induction of tumor-induced deciencies in T-cell dif-
tumorigenic properties of different human lung and mouse cancer ferentiation and function is a fundamental mechanism for tumor
cells. Various tumor cell lines that we tested expressed Notch escape from the host immune system. We reported earlier a
receptors (Fig. 7B) and showed varying kinetics and levels of RNA previously unidentied mechanism for tumor-associated defects
expression of target genes, Hes1 and Hey1 following culture with in T lymphocytes mediated by the alteration of the expression
mouse or human multivalent DLL1 (Supplementary Fig. S1). pattern of Notch ligands and reduced Notch signaling in the
However, of high clinical signicance is the fact that this activated hematopoietic compartment. Selective systemic activation of
signaling did not translate into the increased proliferation or Notch signaling by a multivalent form of DLL1 resulted in
clonogenicity of tumor cells (Fig. 7C and D). Rather, clustered signicant attenuation of tumor growth in a T-celldependent
DLL1 had antiproliferative and/or anticlonogenic effect on some manner in tumor models (21). The current study elucidates the
tumor cells (H157, H460, HCC2429 and H460, H1437, immunologic consequences of the pharmacologic enhancement
respectively; Fig. 7C and D). Moreover, DLL1-treated mice of DLL1 signaling and tests the hypothesis that the multivalent
showed no clinically abnormal behavior or any difference in DLL1-based immunotherapy would benet the oncogene-tar-
body or organ weight compared with the control mice. No gross geted treatments.

4736 Cancer Res; 75(22) November 15, 2015 Cancer Research

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

A 0.25 Control
**

Ratio to -actin (%)

0.2 * Clust DLL1

0.15

0.1
*
0.05

Figure 6. 0
Clustered DLL1 enhances Notch Hes1 Hey1 Deltex1
signaling, modulates expression of
Notch genes in hematopoietic
compartment, and provides
B
immunologic benet in combination
kD
with EGFR inhibition in the EGFRL858R
transgenic lung cancer mouse model. 250
DLL1 DLL4
Lung tumors were induced and mice 150
treated with erlotinib in combination
with clustered DLL1 or control clusters, GAPDH GAPDH 100
as in Fig. 5. For immunologic assays
and analysis of lung-inltrating 75
immune cells, single-cell suspensions
Jag1 Jag2
from lungs were prepared on day 50
GAPDH
after tumor induction. A, mRNA
expression of downstream Notch GAPDH GAPDH

target genes in CD45 cells isolated Control: + + + +
from mouse lungs. B, protein Clust DLL1: + + + +
expression of Notch ligands and
receptors in splenocytes. C, proportion
of lung inltrating immune cell C
Proportion of CD45+ cells (%)

lineages. Mean  SEM, n 5; 40 50 15 15


, P < 0.05;   , P < 0.01. **
* 40
Control
30 * 10 10
* Clust DLL1
30
20
20
5 5
10 10
0 0 0 0
CD19+ B cells CD3+ T cells IFN-producing CD11b+CD11c+ DC
CD3+ T cells

Notch system appears to be highly responsive to the mod- Stimulation of Notch signaling with multivalent DLL1 has
ulation by its ligand. The effects included not only increased signicant positive effect on T-cellmediated antitumor immu-
downstream signaling but also a selective upregulation of nity. It induces tumor antigen-specic IFNg production by
Notch family receptor and ligand expression in the hemato- CD8 T lymphocytes, increases the pool of central-memory
poietic organs. These results suggest the potential existence of CD8 T cells and enhances IFNg synthesis within central mem-
an autocrine amplication loop in the Notch system, where the ory CD8 T-cell population. These results imply the capacity of
initial receptorligand signal is further amplied via upregula- DLL1 to induce effector T cells in vivo. Differentiation of T cells
tion of the Notch system components. It would be clinically into effector and memory subsets is dependent on the status of
important to consider such autocrine amplication of Notch Stat activation, including their phosphorylation. Activation of
signaling from a potential therapeutic intervention point, as Stat1, Stat2, Stat4, and Stat6 promotes Th1-type response,
studies indicate that the effect of Notch modulation might be activation of Stat5 signaling pathway drives Th2 differentiation
dose-dependent (41, 42). Our experiments revealed that clus- whereas Stat3 supports Th17 stemming in activated T cells via
tered DLL1 binds to all four Notch receptors. Additional inves- inhibition of T-bet and stimulation of RORgt transcription
tigations would be required to identify the role of each of the factor (30). The DLL1Notch axis appears to be actively
receptors as well as their potential heterodimeric interaction involved in the regulation of Stat signaling by enhancing phos-
with the multivalent ligand in mediating the observed effects phorylation of Stat1 and Stat2. This supports Th1 prole of T
and to determine whether they are DLL1-specic and how other cells together with the upregulation of T-bet involved in the
ligands modulate Notch gene expression. transcriptional effects of Stat1 on T-cell differentiation and IFNg

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4737

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Biktasova et al.

A Clustered DLL1 Control

100 Control
Clustered DLL1
80

60

%
40

*
20

0 0 20 40 %
100 101 102 103 104 Cells
CFSE

B
kD kD
250
250

150 150

100 100

75 75

C
300 Cont 8,000 400 8,000 800
DLL1
400 2,000 2,000
dpm

200
* 4,000 * 200 2,000 400 *
200 1,000 1,000
100

D459 LLC H157 H460 H1264 H1437 HCC15 HCC2429

D
150 Cont 200 40 1,000 1,000
400 1,500
Number of colonies

DLL1
40

100 1,000

100 200 20
* 500
* 500
20
50 500

D459 LLC H157 H460 H1264 H1437 HCC15 HCC2429

Figure 7.
DLL1-induced Notch signaling enhances human peripheral T-cell proliferation without any protumorigenic effects on cancer cells in vitro. A, PBMCs from human

donors were stimulated with beads-coupled CD3, CD28, and CD137 antibodies with or without multivalent DLL1 for 4 days. Gated CD3 T-cell proliferation
was assessed by CFSE dilution. Representative histogram overlays (higher peak, clustered DLL1) as well as day 4 total cell yields are shown. Mean  SEM;
n 4;  P < 0.05. B, expression of Notch1 and Notch3 in mouse and human cancer cell lines assessed by Western blotting. Notch1 (left) and Notch3 (right)
bands at approximately 110 kD and 90 kD, respectively, correspond to ICD; higher bands at approximately 250 kD represent full-length Notch (see also note for
Western blot analysis in Materials and Methods). C, cell proliferation measured by [3H]-thymidine incorporation. D, colony formation in soft agar evaluated
after 2 weeks. Mean  SEM; n 4. Both PBMC and cancer cells were cultured with clustered DLL1 or control clusters at 1 mg/mL of DLL1-Fc protein.

4738 Cancer Res; 75(22) November 15, 2015 Cancer Research

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

production, but not the other transcription factors regulating Th increased tumor inltration by T cells and DCs. Implying
cell differentiation (31). Stat1 sustains CD8 T-cell memory safety of the enhanced hematopoietic DLL1/Notch signaling
pool by increasing cell survival after their antigen-specic acti- was our observation that mice overexpressing DLL1 in bone
vation (32). This explains the higher rate of CD8 central- marrow appeared normal and did not display any behavioral,
memory T cells after clustered DLL1 treatment. Although clus- tissue, or hematopoietic abnormalities (21). In another study,
tered DLL1 also elevated Stat3 phosphorylation, alterations in DLL1-mediated signaling was implicated in the inhibition of
RORgt expression were marginal and not statistically signicant melanoma growth due to the attenuated vascularization (37).
in our study. Overall, the analysis of transcription factors dis- It is also important to note that inactivating Notch mutations
plays prevalence of Stat1/Stat2/T-bet/Th1 nexus over the other are being discovered in cancers, suggesting that the Notch
types of T lineage differentiation. In addition to the regulation pathway could have an important tumor-suppressor role
of T-cell differentiation, multivalent DLL1 in combination with (47). For therapeutic applications, a short-term regimen of
TCR stimulation demonstrated ability to accelerate proliferation multivalent DLL1 might be sufcient to boost immune sys-
of T cells in human PBMC. Thus, stimulation of Notch signaling tem and induce tumor-specic immune responses. Combina-
with multivalent DLL1 could be a potent therapeutically rele- tions of immune stimulatory multivalent DLL1 with other
vant approach to promote anti-tumor Th1 and CD8 T-cell therapies associated with the release of tumor antigens holds
effector functions via modulation of the expression of transcrip- promise to be effective in inducing long-lasting immune
tion factors associated with T-cell differentiation, regulation of responses.
Stat signaling, and enhanced T-cell proliferation. Multiple studies in recent years have called into question the
Our results are in agreement with the notion that the Notch use of Notch inhibitors to treat cancer because of an increased risk
system regulates T-cell differentiation and lineage commitment of endothelial cell tumors seen in animal models (48). Our
by providing the instructive signals during the induction of studies have shown that downregulation of Notch signaling in
antigen-specic responses and with the fact that most gain-of- the host may promote evasion of the immune system by tumors.
function experiments suggest that high expression of Delta-like Data presented here suggest that, instead of blocking the Notch
ligands promotes Th1 type responses (6, 7). Although somewhat pathway, ligand-specic and controlled restoration of the Notch
controversial, strong evidence implicated Notch1 and 2 in the signaling would benet antitumor immunity and provide clinical
induction of antitumor immunity, including induction of tumor- benet. These data underscore the novel role of DLL1/Notch,
specic CTL and central memory T cells (5, 43). Together with most likely, Notch1 and 2 signaling in the induction of T-cell
these data, our results point to the functional axis DLL1/Notch1 antitumor immune responses.
and/or Notch2 as an immunotherapeutic target for activation and Successful application of multivalent DLL1 predicted by our
validate the clinically relevant approach to effectively induce studies opens a venue for exploration of a novel set of thera-
effector T-cell differentiation and T-cellmediated immunity crit- peutics based on the multivalent forms of Notch ligands
ical for tumor rejection. for modulation of Notch signaling under various pathologic
The strong immune stimulatory effect of pharmacologically conditions.
enhanced DLL1-mediated Notch signaling supports the con-
cept that multivalent DLL1 could be used as a novel immu-
Disclosure of Potential Conicts of Interest
notherapeutic to induce robust immune responses, provide D.P. Carbone reports receiving a commercial research grant from Bristol
effective tumor surveillance, and prolong tumor-free survival Myers Squibb and is a consultant/advisory board member for Merck, Genen-
when combined with tumor oncogene-targeted therapies. In tech/Roche, Boehringer Ingelheim, Novartis, and Pzer. No potential conicts
our studies with the erlotinib treatment of the experimental of interest were disclosed by the other authors.
mutant EGFR-dependent lung cancer, the hypothesis was that
the correction and stimulation of the host immune system by Authors' Contributions
Notch activation before and during the massive tumor cell Conception and design: A.K. Biktasova, D.P. Carbone, A. Shanker, M.M. Dikov
killing by EGFR inhibitor would elicit strong effector and Development of methodology: A.K. Biktasova, D.F. Dudimah, K. Park,
memory T-cell responses. This would provide signicant clin- A. Akhter, S.V. Novitskiy, E.E. Tchekneva, D.P. Carbone, A. Shanker, M.M. Dikov
ical benet by immune-mediated elimination of residual and Acquisition of data (provided animals, acquired and managed patients,
circulating tumor cells/cancer stem cells and/or by rejection of provided facilities, etc.): A.K. Biktasova, D.F. Dudimah, R.V. Uzhachenko,
K. Park, A. Akhter, R.R. Arasada, S.V. Novitskiy, A. Shanker
recurrent tumors via eliciting effective T-cell memory. Indeed,
Analysis and interpretation of data (e.g., statistical analysis, biostatistics,
data suggest that stronger immune responses elicited by com- computational analysis): A.K. Biktasova, D.F. Dudimah, R.V. Uzhachenko,
bination treatment effectuated sustained tumor destruction K. Park, A. Akhter, E.E. Tchekneva, D.P. Carbone, A. Shanker, M.M. Dikov
and extended the PFS. Writing, review, and/or revision of the manuscript: A.K. Biktasova,
Growing evidence shows that pleiotropic functions of D.F. Dudimah, R.V. Uzhachenko, K. Park, J.V. Evans, D.P. Carbone, A. Shanker,
Notch can be tumor suppressive or oncogenic depending on M.M. Dikov
Administrative, technical, or material support (i.e., reporting or organizing
the cellular context in both solid tumors and hematologic
data, constructing databases): A.K. Biktasova, K. Park, D.P. Carbone,
malignancies (4446). Our data suggest the therapeutic safety A. Shanker, M.M. Dikov
of enhancement of DLL1/Notch signaling by systemic admin- Study supervision: D.P. Carbone, A. Shanker, M.M. Dikov
istration of the multivalent DLL1 reagent. The experiments
with multiple human lung and mouse tumor cells demon-
Acknowledgments
strated that clustered DLL1 increases neither proliferation nor
The authors thank Elena M. Dikova for editing the article and D. Luke,
clonogenic potential of cancer cells. In vivo studies revealed D. Dulcinea, and Dr. M.C. Thounaojam for their help with the article and gure
an antitumor effect of this reagent associated with decreased preparation; they also thank the Vanderbilt University Institute of Imaging
tumor angiogenesis, improved T-cell differentiation and Sciences for the assistance with the small animal MRI.

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4739

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Biktasova et al.

Grant Support The costs of publication of this article were defrayed in part by the
This work was supported by NIH grants R01CA138923 (M.M. Dikov) and payment of page charges. This article must therefore be hereby marked
RO1CA175370 (M.M. Dikov and D.P. Carbone), Ohio State University Drug advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate
Development Institute grant (M.M. Dikov), Dallapezze Fund (M.M. Dikov this fact.
and D.P. Carbone), Pilot Project in Lung Cancer SPORE P50CA90949
(M.M. Dikov and A. Shanker), Meharry Clinical and Translational Research
Center Pilot Grant U54 MD007593 (A. Shanker), U54 CA163069 (A. Shanker), Received April 17, 2014; revised August 6, 2015; accepted August 12, 2015;
and SC1 CA182843 (A. Shanker). published OnlineFirst September 24, 2015.

References
1. Fiuza UM, Arias AM. Cell and molecular biology of Notch. J Endocrinol hematopoiesis associated with elevated levels of VEGF. Blood 2007;
2007;194:45974. 110:62431.
2. Kopan R, Ilagan MX. The canonical Notch signaling pathway: unfolding the 21. Huang Y, Lin L, Shanker A, Malhotra A, Yang L, Dikov MM, et al. Resus-
activation mechanism. Cell 2009;137:21633. citating cancer immunosurveillance: selective stimulation of DLL1-Notch
3. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, et al. Reciprocal signaling in T cells rescues T-cell function and inhibits tumor growth.
developmental pathways for the generation of pathogenic effector TH17 Cancer Res 2011;71:612231.
and regulatory T cells. Nature 2006;441:2358. 22. Novitskiy SV, Csiki I, Huang Y, Johnson DH, Harth EM, Carbone DP, et al.
4. Maekawa Y, Tsukumo S, Chiba S, Hirai H, Hayashi Y, Okada H, et al. Anti-vascular endothelial growth factor treatment in combination with
Delta1-Notch3 interactions bias the functional differentiation of activated chemotherapy delays hematopoietic recovery due to decreased prolifera-
CD4 T cells. Immunity 2003;19:54959. tion of bone marrow hematopoietic progenitor cells. J Thorac Oncol
5. Sierra RA, Thevenot P, Raber PL, Cui Y, Parsons C, Ochoa AC, et al. Rescue of 2010;5:14105.
notch-1 signaling in antigen-specic CD8 T cells overcomes tumor- 23. Lee CT, Ciernik IF, Wu S, Tang DC, Chen HL, Truelson JM, et al. Increased
induced T-cell suppression and enhances immunotherapy in cancer. Can- immunogenicity of tumors bearing mutant p53 and P1A epitopes after
cer Immunol Res 2014;2:80011. transduction of B71 via recombinant adenovirus. Cancer Gene Ther
6. Amsen D, Antov A, Flavell RA. The different faces of Notch in T-helper-cell 1996;3:23844.
differentiation. Nat Rev Immunol 2009;9:11624. 24. Politi K, Zakowski MF, Fan PD, Schonfeld EA, Pao W, Varmus HE.
7. Amsen D, Blander JM, Lee GR, Tanigaki K, Honjo T, Flavell RA. Instruction Lung adenocarcinomas induced in mice by mutant EGF receptors
of distinct CD4 T helper cell fates by different notch ligands on antigen- found in human lung cancers respond to a tyrosine kinase inhibitor
presenting cells. Cell 2004;117:51526. or to downregulation of the receptors. Genes Dev 2006;20:
8. Tanigaki K, Tsuji M, Yamamoto N, Han H, Tsukada J, Inoue H, et al. 1496510.
Regulation of alphabeta/gammadelta T-cell lineage commitment and 25. Novitskiy SV, Ryzhov S, Zaynagetdinov R, Goldstein AE, Huang Y, Tikho-
peripheral T-cell responses by Notch/RBP-J signaling. Immunity 2004; mirov OY, et al. Adenosine receptors in regulation of dendritic cell
20:61122. differentiation and function. Blood 2008;112:182231.
9. Amsen D, Antov A, Jankovic D, Sher FA, Radtke F, Souabni A, et al. Direct 26. Heinzel K, Benz C, Bleul CC. A silent chemokine receptor regulates
regulation of Gata3 expression determines the T helper differentiation steady-state leukocyte homing in vivo. Proc Natl Acad Sci U S A 2007;
potential of Notch. Immunity 2007;27:8999. 104:84216.
10. Krawczyk CM, Sun J, Pearce EJ. Th2 differentiation is unaffected by Jagged2 27. Gabrilovich DI, Ishida T, Nadaf S, Ohm JE, Carbone DP. Antibodies to
expression on dendritic cells. J Immunol 2008;180:79317. vascular endothelial growth factor enhance the efcacy of cancer immu-
11. Ladi E, Nichols JT, Ge W, Miyamoto A, Yao C, Yang LT, et al. The divergent notherapy by improving endogenous dendritic cell function. Clin Cancer
DSL ligand Dll3 does not activate Notch signaling but cell autonomously Res 1999;5:296370.
attenuates signaling induced by other DSL ligands. J Cell Biol 2005; 28. Chen JQ, Xiu QY, Shen C, Yan ZM. [Treatment of spontaneous metastatic
170:98392. lung cancer with interleukin-12 gene-modied dendritic cells vaccine]. Ai
12. Skokos D, Nussenzweig MC. CD8-DCs induce IL-12-independent Th1 Zheng 2002;21:132831.
differentiation through Delta 4 Notch-like ligand in response to bacterial 29. Mandelboim O, Bar-Haim E, Vadai E, Fridkin M, Eisenbach L. Identica-
LPS. J Exp Med 2007;204:152531. tion of shared tumor-associated antigen peptides between two spontane-
13. Kassner N, Krueger M, Yagita H, Dzionek A, Hutloff A, Kroczek R, et al. ous lung carcinomas. J Immunol 1997;159:60306.
Cutting edge: plasmacytoid dendritic cells induce IL-10 production in T 30. Ryzhov S, Novitskiy SV, Zaynagetdinov R, Goldstein AE, Carbone DP,
cells via the Delta-like-4/Notch axis. J Immunol 2010;184:5504. Biaggioni I, et al. Host A(2B) adenosine receptors promote carcinoma
14. Liotta F, Frosali F, Querci V, Mantei A, Fil L, Maggi L, et al. Human growth. Neoplasia 2008;10:98795.
immature myeloid dendritic cells trigger a TH2-polarizing program via 31. Yang XO, Panopoulos AD, Nurieva R, Chang SH, Wang D, Watowich SS,
Jagged-1/Notch interaction. J Allergy Clin Immunol 2008;121:10005 e8. et al. STAT3 regulates cytokine-mediated generation of inammatory
15. Sun J, Krawczyk CJ, Pearce EJ. Suppression of Th2 cell development by helper T cells. J Biol Chem 2007;282:935863.
Notch ligands Delta1 and Delta4. J Immunol 2008;180:165561. 32. Quigley M, Huang X, Yang Y. STAT1 signaling in CD8 T cells is required for
16. Vigouroux S, Yvon E, Wagner HJ, Biagi E, Dotti G, Sili U, et al. their clonal expansion and memory formation following viral infection in
Induction of antigen-specic regulatory T cells following overexpres- vivo. J Immunol 2008;180:215864.
sion of a Notch ligand by human B lymphocytes. J Virol 2003;77: 33. Feoktistov I, Ryzhov S, Goldstein AE, Biaggioni I. Mast cell-mediated
1087280. stimulation of angiogenesis: cooperative interaction between A2B and
17. Yvon ES, Vigouroux S, Rousseau RF, Biagi E, Amrolia P, Dotti G, et al. A3 adenosine receptors. Cir Res 2003;92:48592.
Overexpression of the Notch ligand, Jagged-1, induces alloantigen-specic 34. Jones CA, Tsukamoto T, O'Brien PC, Uhl CB, Alley MC, Lieber MM. Soft
human regulatory T cells. Blood 2003;102:381521. agarose culture human tumour colony forming assay for drug sensitivity
18. Kared H, Adle-Biassette H, Fois E, Masson A, Bach JF, Chatenoud L, et al. testing: [3H]-thymidine incorporation vs. colony counting. Br J Cancer
Jagged2-expressing hematopoietic progenitors promote regulatory T-cell 1985;52:30310.
expansion in the periphery through notch signaling. Immunity 35. Heinzel K, Benz C, Martins VC, Haidl ID, Bleul CC. Bone marrow-
2006;25:82334. derived hemopoietic precursors commit to the T-cell lineage only
19. Keerthivasan S, Suleiman R, Lawlor R, Roderick J, Bates T, Minter L, et al. after arrival in the thymic microenvironment. J Immunol 2007;178:
Notch signaling regulates mouse and human Th17 differentiation. 85868.
J Immunol 2011;187:692701. 36. Jochems C, Schlom J. Tumor-inltrating immune cells and prognosis: the
20. Huang Y, Chen X, Dikov MM, Novitskiy SV, Mosse CA, Yang L, potential link between conventional cancer therapy and immunity. Exp
et al. Distinct roles of VEGFR-1 and VEGFR-2 in the aberrant Biol Med 2011;236:56779.

4740 Cancer Res; 75(22) November 15, 2015 Cancer Research

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent DLL1 in Cancer Immunotherapy

37. Zhang JP, Qin HY, Wang L, Liang L, Zhao XC, Cai WX, et al. Over- ation and in vivo marrow repopulating ability of cord blood cells. Blood
expression of Notch ligand Dll1 in B16 melanoma cells leads to reduced 2005;106:26939.
tumor growth due to attenuated vascularization. Cancer Lett 2011; 43. Sugimoto K, Maekawa Y, Kitamura A, Nishida J, Koyanagi A, Yagita H, et al.
309:2207. Notch2 signaling is required for potent antitumor immunity in vivo.
38. Wang Y, Schmid-Bindert G, Zhou C. Erlotinib in the treatment of advanced J Immunol 2010;184:46738.
nonsmall cell lung cancer: an update for clinicians. Ther Adv Med Oncol 44. Lobry C, Oh P, Aifantis I. Oncogenic and tumor suppressor functions of
2012;4:1929. Notch in cancer: it's NOTCH what you think. J Exp Med 2011;208:19315.
39. Pao W, Ladanyi M, Miller VA. Erlotinib in lung cancer. N Engl J Med 45. Klinakis A, Lobry C, Abdel-Wahab O, Oh P, Haeno H, Buonamici S, et al.
2005;353:173941. A novel tumour-suppressor function for the Notch pathway in myeloid
40. Pao W, Miller VA, Politi KA, Riely GJ, Somwar R, Zakowski MF, et al. leukaemia. Nature 2011;473:2303.
Acquired resistance of lung adenocarcinomas to getinib or erlotinib is 46. Roodhart JM, He H, Daenen LG, Monvoisin A, Barber CL, van Amersfoort
associated with a second mutation in the EGFR kinase domain. PLoS Med M, et al. Notch1 regulates angio-supportive bone marrow-derived cells:
2005;2:e73. relevance to chemoresistance. Blood 2013;122:14353.
41. Gale NW, Dominguez MG, Noguera I, Pan L, Hughes V, Valenzuela DM, 47. Stransky N, Egloff AM, Tward AD, Kostic AD, Cibulskis K, Sivachenko A,
et al. Haploinsufciency of delta-like 4 ligand results in embryonic lethality et al. The mutational landscape of head and neck squamous cell carcinoma.
due to major defects in arterial and vascular development. Proc Natl Acad Science 2011;333:115760.
Sci U S A 2004;101:1594954. 48. Yan M, Callahan CA, Beyer JC, Allamneni KP, Zhang G, Ridgway JB,
42. Delaney C, Varnum-Finney B, Aoyama K, Brashem-Stein C, Bernstein ID. et al. Chronic DLL4 blockade induces vascular neoplasms. Nature
Dose-dependent effects of the Notch ligand Delta1 on ex vivo differenti- 2010;463:E67.

www.aacrjournals.org Cancer Res; 75(22) November 15, 2015 4741

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.
 Published OnlineFirst September 24, 2015; DOI: 10.1158/0008-5472.CAN-14-1154

Multivalent Forms of the Notch Ligand DLL-1 Enhance Antitumor

T-cell Immunity in Lung Cancer and Improve Efficacy of
EGFR-Targeted Therapy
Asel K. Biktasova, Duafalia F. Dudimah, Roman V. Uzhachenko, et al.

Cancer Res 2015;75:4728-4741. Published OnlineFirst September 24, 2015.

Updated version Access the most recent version of this article at:
doi:10.1158/0008-5472.CAN-14-1154

Supplementary Access the most recent supplemental material at:

Material http://cancerres.aacrjournals.org/content/suppl/2015/10/08/0008-5472.CAN-14-1154.DC1

Cited articles This article cites 48 articles, 27 of which you can access for free at:
http://cancerres.aacrjournals.org/content/75/22/4728.full#ref-list-1

Citing articles This article has been cited by 2 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/75/22/4728.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, contact the AACR Publications Department at
[email protected].

Downloaded from cancerres.aacrjournals.org on September 20, 2017. 2015 American Association for Cancer Research.