Clinical Microbiology Reviews-2018-Corvec-e00064-17.full

REVIEW
crossm
Clinical and Biological Features of Cutibacterium (Formerly

Propionibacterium) avidum, an Underrecognized
Microorganism
Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

Stéphane Corveca,b
a
CHU Nantes, Service de Bactériologie-Hygiène Hospitalière, Nantes, France
b CRCINA, INSERM, U1232, Université de Nantes, Nantes, France
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
MICROBIOLOGY OF C. AVIDUM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Microbiota . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Genetic and Genomic Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
PATHOGENESIS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Virulence Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Lipase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Hyaluronate Lyase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Hemolysins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Biofilm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Other Virulence Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Host-Bacterium Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Immune Stimulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Antitumor Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
DIAGNOSTIC PROCEDURES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Conventional Microbial Cultures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Contamination Issues in Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Identification Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Serological and Precipitin Tests for Identification of Cutibacterium-Related Species . . . . 19
Cell Wall Analysis and Protein Pattern Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Susceptibility to Free Fatty Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Biochemical and Enzyme Profiles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Molecular Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
MALDI-TOF Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
CLINICAL PRESENTATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Breast Infections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Abdominal Infections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Splenic Abscesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Skin Abscesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Infective Endocarditis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
Bone and Joint Infections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Prostate Infections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Acne Lesions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Caries-Like Lesions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
ANTIMICROBIAL SUSCEPTIBILITY OF C. AVIDUM AND RELATED SPECIES . . . . . . . . . . . . . 33
ANTIBIOTIC TREATMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
ACKNOWLEDGMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 Published 30 May 2018
AUTHOR BIO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Citation Corvec S. 2018. Clinical and biological
features of Cutibacterium (formerly
Propionibacterium) avidum, an
underrecognized microorganism. Clin
SUMMARY The recent description of the genus Cutibacterium has altered the tax- Microbiol Rev 31:e00064-17. https://doi.org/10
onomy of Propionibacterium species. These organisms still belong to the genera of .1128/CMR.00064-17.
the skin coryneform group, and the most-studied species remains Cutibacterium ac- Copyright © 2018 American Society for
Microbiology. All Rights Reserved.
nes. Cutibacterium avidum is also a known skin commensal. This underrecognized
Address correspondence to
microorganism can, however, act as a pathogen after bacterial seeding and can be [email protected].
considered opportunistic, causing either superficial or deep/invasive infections. It can
July 2018 Volume 31 Issue 3 e00064-17 Clinical Microbiology Reviews cmr.asm.org 1

Corvec Clinical Microbiology Reviews
cause numerous infections, including but not limited to breast infections, skin ab-
scesses, infective endocarditis, and device-related infections. The ecological niche of
C. avidum is clearly different from that of other members of the genus: it is found in
the axillary region or at wet sites rather than in dry, exposed areas, and the number
of microorganisms increases during puberty. Historically, it has been used for its
ability to modulate the immune response and for its antitumor properties. Conven-
tional microbial culture methods and identification processes allow for its accurate
identification and characterization. Thanks to the modern omics tools used for phy-
logenomic approaches, understanding C. avidum pathogenesis (including host-
bacterium interactions and virulence factor characterization) is becoming easier, al-
lowing for more thorough molecular characterization. These analyses have revealed

that C. avidum causes diverse diseases mediated by multiple virulence factors. The
recent genome approach has revealed specific genomic regions within this species
that are involved in adherence and biofilm formation as well as fitness, survival, and
defense functions. Numerous regions show the presence of phages and horizontal
gene transfer. C. avidum remains highly sensitive to a broad spectrum of antibiotics,
such as ␤-lactams, fluoroquinolones, macrolides, and rifampin, although erythromy-
cin and clindamycin resistance has been described. A long-term treatment regimen
with a combination of antibiotics is required to successfully eliminate the remaining
adherent bacteria, particularly in the case of deep infections after debridement sur-
gery.
KEYWORDS Cutibacterium avidum, abscess, biofilm, immunostimulation, genomic,
identification, clinical infections, antibiotic resistance, biofilms, clinical infection,
genomics, genotypic identification, immune regulation, phenotypic identification
INTRODUCTION
T he genus Propionibacterium, initially described by Orla-Jensen in 1909 (1), is made

up of species that produce propionic acid during the fermentation process (2). It
belongs to the class Actinobacteria of the order Propionibacteriales (3). Following its
discovery in a patient with a chronic skin disease called “acne vulgaris,” Propionibac-
terium acnes, henceforth called Cutibacterium acnes, underwent a series of taxonomic
changes. It was successively placed in the genus Bacillus and then the genus Coryne-
bacterium (4, 5). However, in 1946, Douglas and Gunter were able to demonstrate that
this microorganism was more closely related to the Propionibacterium genus members
since, like other species of this genus, it ferments lactose to propionic acid in an
anaerobic atmosphere (6, 7). Recently, a significant taxonomic revision was proposed
by Scholz and Kilian, placing all Propionibacterium species from the skin microbiota
within a new genus (8). As a result, Propionibacterium is replaced by Cutibacterium for
P. avidum, P. acnes, P. granulosum, and the more recently discovered species P.
namnetense (9) and P. humerusii (10). Too often ignored or considered a contaminant
in cultures (especially in hemolin performance diphasic medium for blood cultures),
these species are the subject of renewed interest, with literature results on these
organisms increasing from 100 papers per year in 2000 to some 230 to 240 per year
since 2013.
In this review, I focus on Cutibacterium avidum, a Gram-positive, anaerobic-
aerotolerant rod which is a minor colonizer and resident of the human skin. Its
ecological niche is clearly different from that of other members of this genus due to a
specific tropism for wet areas (11). Although often defined as a commensal (12, 13), C.
avidum has likely been underestimated, unrecognized, and misidentified in more than
a few clinical cases. Despite its low frequency in human infections, numerous associated
or invasive infections have been reported: bone infections, infective endocarditis (IE),
breast infections, abdominal infections, prostate infections, and splenic and skin ab-
scesses have been described, sometimes with significant collateral damage for the
patient (14–17). In my experience, although one cannot avoid bias according to the
identification method used and the awareness of microbiologists, there has been a
July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 2

Cutibacterium avidum, an Underrecognized Microorganism Clinical Microbiology Reviews
clearly observed increase in C. avidum-positive samples over the past years. A plausible
explanation for the increased awareness of this microorganism is probably the intro-
duction of matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) tech-
nology in numerous laboratories. Most positive samples have been from blood cultures
and valve, bone, tissue, and deep perioperative specimens. Interestingly, most of them
were recovered from deep tissue samples taken from immunosuppressed patients or
patients receiving chemotherapy or radiotherapy.
In the 1970s, C. avidum was considered a less virulent bacterium or culture con-
taminant (3), but a specific dedicated preparation was used as an adjuvant in antitumor
therapies (18) or as an immunity stimulant (19). Later, following various studies, the key
role played by Cutibacterium species, especially C. avidum, in immune response mod-

ulation was demonstrated (20). Indeed, this species has been reported to be an
immunomodifier, stimulating the various cell subset populations involved in nonspe-
cific antibacterial, antiviral, or antitumor abilities (21).
Thereafter, biochemical characterization of the cell wall (22) and specific interactions
with reticuloendothelial cell systems (23), including various lymphocyte subpopula-
tions, demonstrated its medical significance. Following the culture-centric period,
which focused on developing new tools to better understand bacterial growth, me-
tabolism, and immunogenicity, we are now entering a new genomic era. This new
period will allow us to decipher virulence factors, genome organization, genome
plasticity, and specific features of several C. avidum clinical strains by using next-
generation sequencing (13, 14) and confirming hypotheses through various in vitro and
in vivo models. As a result, deeper genomic analyses and comparisons will provide key
information about the organism’s phylogeny and the potential existence of subpopu-
lations within this species. Such results have been described for C. acnes clinical strains,
with a correlation of different clusters or lineages involved in specific clinical entities.
Although a whole-genome synteny has been reported for propionibacteria originating
from the skin, a certain plasticity of the genome has revealed C. avidum-specific
genomic regions. This observation demonstrated the importance of these genomic
regions, which include exopolysaccharide gene synthesis functions, prophage regions,
and other gene-specific clusters involved in diverse fitness, survival, and defense
functions (14). Although only 2 of 14 complete genomes are publicly available, C.
avidum genomic studies remain relevant to furthering the understanding of the
pathogenesis of this organism.
In light of the growing number of C. avidum infections, I review both the physio-
pathology and the clinical/microbiological significance of C. avidum as the causative
agent of invasive and deep infections and provide an extended view of its biochemical
features, immune responsiveness, and culture and identification methods as well as a
genomic analysis based on publicly available data. This bacterium is therefore a new
rod that needs to be monitored as a potential opportunistic pathogen.
MICROBIOLOGY OF C. AVIDUM
Microbiota
Anaerobic diphtheroid bacteria are typical residents of the human sebaceous folli-
cles and are certainly almost all implicated in the pathogenicity of acne or other skin
disorders. Following multiplication in the lumen, they move from the follicles over the
sebum-covered skin surface, forming dense populations in sebum-rich areas of the skin
(24–26). The three most prevalent species (C. avidum, C. acnes, and C. granulosum) can
be found on the skin surfaces of both healthy individuals and acne patients. Their
spread and distribution are related to the skin region studied, patient age, and presence
of skin lesions (papules, pustules, and nodules), as previously reported (6, 27). However,
skin microbiome data did not focus on C. avidum’s relative abundance, leading to the
absence of available data about its prevalence and location. Recent comparative
genomics studies revealed distinct host-interacting traits of the three major human-
associated propionibacteria, including C. avidum. They have evolved specific charac-
teristics to interact with their human hosts, and their various surface properties

probably determine their colonizing capability and pathogenicity at the cutaneous

level and also at nonskin sites (13).
The ecological niche of C. avidum has been previously analyzed. C. avidum tends to
reside in regions containing eccrine sweat glands (28). It is most frequently detected in
wet areas (nares, axilla, groin, and rectum, which include areas populated with enteric
Gram-negative bacteria, suggesting an intestinal reservoir), compared to an oily and
wet distribution for C. granulosum (29). In contrast, C. acnes is the most prevalent and
ubiquitous in sebum-rich areas. Indeed, other Cutibacterium species, especially C. acnes
and C. granulosum, need skin surface lipids to be present and are involved in acne
vulgaris, whereas C. avidum is rarely or generally not associated with acne (28). Using
immunohistochemical staining of acne samples or confocal laser scanning microscopy

with specific antibodies, Jahns et al. reported the absence of C. avidum in all acne
samples, whereas C. acnes and C. granulosum were found as distinct populations within
the same follicles (30). While C. acnes requires reduced sulfur, which can be found in
vivo thanks to its constant availability in the form of sulfhydryl groups in the sebaceous
glands during keratinization (31), water plays a crucial role in C. avidum ecology (11).
Thus, the relationship between C. acnes and its nutritional substrate may provide this
microorganism a selective advantage in this ecological niche, whereas modifications in
the axillary environment, possibly changes of the moist environment and nutritive
substance availability, are key variables in the C. avidum ecology (11, 28). These local
modifications suggest the existence of various skin environments differing in moisture,
sweat density, sebaceous gland content, and extent of anaerobiosis (11). Consequently,
looking at the distribution of cutaneous Cutibacterium species, regional qualitative and
quantitative differences have been demonstrated. Thus, a real-time PCR technique for
accurate assessment of the presence of some Cutibacterium species (inhabitants of
human skin) has also been developed to detect the three main species (12).
Interestingly, the prevalence of C. avidum recovery increases with age and is
associated with the onset of puberty. This bacterium is clearly recovered less from
young females or males aged between 5 and 10 years (5% and 4.5%, respectively),
while female and male teenagers (15- to 16-year-olds) are colonized at 45% and 58%,
respectively. Females tend to be colonized earlier in life than males. In fact, it has been
suggested that increases in axillary moisture with age and in nutritive substance
availability may act as key points during the colonization process. This finding consti-
tutes a possible explanation for the most relevant routes of contamination during a
breast abscess (see references 15, 17, and 32).
Finally, C. avidum has also been reported to be present in the hen intestine (poultry
microbiota). This ecological niche had never been described previously, even though
the presence of other ex-Propionibacterium species, especially Acidipropionibacterium
acidipropionici, was known. Compared to the latter species, C. avidum seems to be
better adapted to the scarce nutrients of cecal water, most likely due to its proteolytic
activity (33).
Genetic and Genomic Analyses

Though propionic acid-producing bacteria were initially reported by von Freudren-
reich and Orla-Jensen at the beginning of the last century (1906) (1), in the 1970s DNA
similarities were investigated to classify the Propionibacterium species. Over the past
decade, in light of their frequency in human samples, various molecular typing tech-
niques have been promoted and used to classify subgroups of P. acnes/C. acnes clinical
strains. Phylotype determination was first performed by recA or tly gene sequencing
(34). Recently, Barnard et al. proposed a multiplex PCR method to identify the six major
phylotypes (IA1, IA2, IB, IC, II, and III) (35). To be more accurate, multilocus sequence
typing (MLST) methods were developed, leading to the differentiation of clusters within
P. acnes/C. acnes populations, called sequence types (ST) or clonal complexes (CC) (36,
37). Recently, a new method, single-locus sequence typing (SLST), has been endorsed
for identification of P. acnes/C. acnes subpopulations, with an excellent degree of
correlation (38). Moreover, over the past 3 years, a Propionibacterium taxonomy mod-


FIG 1 Genome relatedness of C. avidum to various Cutibacterium species, based on orthologous average
nucleotide identity (orthoANI) values. The dendrogram was created using Oat 0.93 software.
ification was proposed, with three subspecies for Cutibacterium acnes. Indeed, regard-
ing the different phylotypes and based on morphological features, it has been pro-
posed that phylotype I corresponds to C. acnes subsp. acnes, phylotype II to C. acnes
subsp. defendens, and phylotype III to C. acnes subsp. elongatum (39, 40).
Finally, in 2016, Scholz and Kilian proposed a reclassification of the genus Propi-
onibacterium as Cutibacterium for cutaneous propionibacteria, based on genomic evi-
dence (8). These changes were consistent with species habitats, genomic topology,
DNA G⫹C content, and peptidoglycan composition. Thus, Cutibacterium is the genus
for C. avidum, C. acnes, and C. granulosum and two recent Cutibacterium species, C.
namnetense and C. humerusii. For more information, the NamesforLife website can be
useful to better understand the Cutibacterium species (http://doi.namesforlife.com/10
.1601/tx.6610). In comparing the three species belonging to a homology group, C.
acnes, C. granulosum, and C. avidum, this work demonstrated a 62 to 63% GC content
in the DNA from the C. avidum group. In contrast, the C. acnes group revealed a 59 to
60% GC content and only 50% homology with strains belonging to the C. avidum
group, along with a low level of similarity with C. granulosum group strains (41).
More recently, genome availability among Cutibacterium species revealed several
differences. Compared with the C. acnes genome GenBank database, C. avidum genome
data remain poor. Indeed, there are only 14 of these genomes available in GenBank,
two of which are complete genomes, for C. avidum 44067 and DPC 6544. Consequently,
although more than 120 C. acnes genomes are available, in accord with their signifi-
cance in the medical field, including complete (n ⫽ 16), chromosome (n ⫽ 3), scaffold
(n ⫽ 71), and contig (n ⫽ 25) data, C. avidum and C. granulosum are still less studied
at the genomic level despite new and easy access to the genomic platform. Only 14 C.
avidum genomes (complete [n ⫽ 2], scaffold [n ⫽ 2], and contigs [n ⫽ 10]) and three
C. granulosum genomes (complete [n ⫽ 1] and contigs [n ⫽ 2]) are available.
Interestingly, phylogenetic analysis confirms the close relationship between C.
avidum and C. acnes, whereas C. granulosum is more distant, as shown in Fig. 1. Based
on orthologous average nucleotide identity (orthoANI), analysis of the genome relat-
edness of the five Cutibacterium species was performed and revealed a closer relation-
ship between C. acnes, C. namnetense, and C. humerusii (orthoANI values between 80.6
and 88.5%; the validated threshold for the species definition is 95%). In contrast, C.
avidum represents another branch of the dendrogram. The same observation is true for
the C. granulosum genome, which is much smaller (about 400 kb smaller), with fewer
than 1,865 genes, on average, coding for 1,592 proteins, whereas C. avidum and C.
acnes possess 2,362 and 2,357 genes, respectively, on average, and 2,207 and 2,267
proteins, respectively, on average (Table 1). The core genome shared between these
three species encodes almost 1,400 proteins. The genes corresponding to these pro-

TABLE 1 Genomic comparison and characteristics of Cutibacterium sp. genomes available in GenBank
Genome size
Species Strain (Mb) GC% No. of scaffolds No. of genes No. of proteins
Cutibacterium avidum 44067 2.52614 63.5 2,297 2,184
DPC 6544 2.72985 63.4 2,505 2,312
ATCC 25577 2.55396 63.3 7 2,290 2,200
MJR 7694 2.46715 63.4 16 2,235 2,118
TM16 2.53972 63.4 420 2,441 2,133
UCD-PD2 2.66729 63.4 51 2,442 2,304
T13 2.46397 63.4 15 2,223 2,109
T15 2.46279 63.4 14 2,223 2,110
T14 2.52207 63.4 9 2,290 2,184
CI878 2.54154 63.4 73 2,371 2,201
CI855 2.54166 63.5 51 2,376 2,190

CI853 2.53246 63.5 41 2,375 2,191
CI882 2.49707 63.4 66 2,309 2,129
CI828 2.47614 63.4 30 2,303 2,130
Cutibacterium granulosum NCTC11865 2.17524 64.2 1,857 1,749

TM11 2.14036 64.2 98 1,820 1,619
DSM 20700 2.12823 64.1 372 1,911 1,566
Cutibacterium acnes KPA171202 2.56026 60 2,371 2,277

SK137 2.49533 60.1 2,353 2,278
266 2.49458 60 2,332 2,250
66609 2.56028 60 2,378 2,282
ATCC 11828 2.48863 60 2,318 2,198
Type IA2 Pacn33 2.48962 60 2,327 2,241
Type IA2 Pacn17 2.52289 60.1 2,354 2,274
Type IA2 Pacn31 2.49877 60 2,324 2,246
C1 2.519 60.1 2,377 2,276
HL096PA1 2.54977 60.15 2,414 2,336
hdn-1 2.49456 60 2,327 2,247
KCOM 1861 2.52244 60 2,337 2,208
PA_15_1_R1 2.54796 60.15 2,409 2,324
PA_21_1_L1 2.56035 60 2,380 2,276
PA_15_2_L1 2.54001 60.1 2,382 2,300
PA_12_1_L1 2.49196 60 2,337 2,261
Cutibacterium namnetense NTS31307302 2.36966 60.5 24 2,188 2,052

SK182B-JCVI 2.43067 60.2 50 2,242 2,053
teins code for main metabolic pathways, in particular the propionate formation path-
way (characteristic of this genus), the respiratory chain, and fatty acid metabolism (13).
Each species has specific traits. C. avidum is characterized by the production of an
exopolysaccharide-like structure, encoded by a targeted region corresponding to a
species-specific genomic island. This region harbors more than 30 genes, including 19
encoding glycosyltransferases and enzymes implicated in mono- or polysaccharide
modification (14). In contrast, C. granulosum is able to produce pilus-like structures,
appendices that have never been described for either C. avidum or C. acnes. Rather, C.
acnes pathogenesis is likely associated with the secreted and surface-associated pro-
teins (42, 43). The detection of a genetic region encoding proteins or enzymes
implicated in exopolysaccharide synthesis and modification may explain the ability of
C. avidum strains not only to produce biofilm following adherence to foreign surfaces
or devices but also to evade phagocytosis, along with the ensuing reduced or limited
immune response. Consequently, C. avidum persistence in infection can be explained.
In fact, the chronic status (see the clinical section of this report) of several reported
cases can be ascribed to the lack of clearance by the immune system and to frequently
described abscess formation, especially after surgery. According to genomic analysis,
Cutibacterium species have developed different colonization/invasive strategies leading
to host interactions or damage. Depending on the genetic background, each species
appears to be able to become a potential pathogen, either at a cutaneous level or at
nonskin sites.


FIG 2 Overall similarities among the publicly available genomes of C. avidum, based on orthoANI values.
The dendrogram was created using Oat 0.93 software. Note that two of the genomes, for strains DPC
6544 and UCP-PD2, lie very close to the species boundary of 95% orthoANI, which suggests the presence
of a subtype population in the species.
The recent description of the genome sequence of C. avidum strain 44067, recov-
ered from a human skin abscess, revealed the presence of clustered regularly inter-
spaced short palindromic repeat (CRISPR) loci (44). This microbial adaptive immune
system has been described extensively for Streptococcus thermophilus (45). This pow-
erful system restricts the potential for foreign DNA acquisition and therefore limits
horizontal gene transfer. This system confers an immune response to invasion by a
variety of foreign mobile elements, such as viruses and plasmids (46). The C. avidum
CRISPR system is related to the C. acnes CRISPR system, mostly present in phylotype II
isolates. They belong to the same subtype (subtype 1-E), and their cas genes are
homologous, with 78% identity at the nucleotide level (GenBank accession number
NC021064 for C. avidum and GenBank accession number NC006085 for C. acnes).
Moreover, the C. avidum CRISPR sequence has two nucleotide substitutions compared
to the C. acnes sequence. For C. avidum ATCC 25577, the CRISPR sequence has
previously been reported, with 28 copies along the genome. This sequence appears in
58 copies in C. avidum DPC6544 and in 20 copies in C. avidum 44064, suggesting
potential diversity among C. avidum isolates. Moreover, a 100% match has also been
noticed in the C. granulosum NCTC11865 genome, with 31 copies. However, ATCC
25577 does not possess any of the 18 predicted spacers of isolate 44067. Interestingly,
regarding the spacers in C. avidum species, one matches a sequence within a gene
located on a plasmid (strain HL096PA1) or on a tight adherence locus-containing
mobile element (strain 15.1.P1) (46, 47). Further studies are needed to determine
whether spacers are predicted to target any known phages or other plasmids in C.
avidum strains.
Based on the same orthologous average nucleotide identity method reported
previously (48), a genomic comparison of the first nine Cutibacterium avidum genomes
available in the GenBank database was performed. Figure 2 shows a significant relat-
edness among seven genomes, with orthoANI values between 98.4 and 100%. I can,
however, also highlight the presence of a cluster including two genomes (C. avidum
DPC6544 and UCP-PD2), which might constitute a subpopulation or a subtype of this
species, as recently suggested (14, 49). These data should be investigated further with
more diverse and numerous clinical isolates to better investigate C. avidum pathoge-
nicity, but also the group or cluster delineation of strains with specific potential for
abscess formation, for example, or biofilm production, as previously reported for the C.
acnes phylotype groups (50–52).


FIG 3 Whole-genome comparison of the publicly available genomes of C. avidum. The comparison was performed using the BLAST algorithm and visualized
with BRIG 0.95 software. The complete genome of C. avidum 44067 was used as a reference sequence. Note the presence of multiple isolate-specific regions,
which indicates significant genome plasticity. A representative set of these regions (labeled 1 to 10) was chosen for further analysis and suggested an active
involvement of C. avidum in horizontal gene transfer (see Table 2 for details).
A detailed comparison of available C. avidum genomes was also conducted to

investigate the possible presence of species-specific traits as well as whether the
differences observed within the dendrogram are linked to horizontal gene transfer or
mobile genetic elements. I decided to use the complete genome of C. avidum 44067 as
a reference sequence. I clearly observed the presence of multiple isolate-specific
regions, indicating significant genome plasticity (Fig. 3). Interestingly, the 10 most
representative genomic regions investigated revealed the presence of transposases,
phages, or mobile genetic elements (Table 2). From this point forward, new isolates
with a typical clinical presentation should be investigated to study the impact of gene
content according to tropism (skin, bone, prostate, etc.) and to highlight the possibility
of phylogenetically distinct clusters within the C. avidum population. Lastly, a genomic

TABLE 2 Putative nature and distribution of a representative set of genomic regions specific to C. avidum 44067a
Region Coordinates Size (bp) Putative nature BLAST search result
1 52500–57400 4,900 Mobile element C. avidum
2 232700–234300 1,600 Acid phosphatase C. avidum, C. acnes, Bifidobacterium bifidum
3 309100–310400 1,300 Transposase C. avidum, C. acidipropionici
5 801000–830000 29,000 Phage C. avidum
7 1215800–1228000 12,200 Phage C. avidum
9 1792700–1806200 13,500 Horizontal gene transfer hot spot C. avidum
aCoordinates are specified according to the C. avidum 44067 genome numbering. The distribution of the genomic regions was determined using BLAST searches with

the nr/nt database of NCBI’s GenBank.
comparison should also be performed because of the recent report of a C. avidum strain
isolated from a feline anal sac (49), corresponding to the first description of Cutibac-
terium carriage in animals other than poultry.
PATHOGENESIS
Virulence Factors
Some virulence factors are presented here, but the reader should keep in mind that
these factors are predicted or putative virulence factors/host interaction factors. Indeed,
to the best of my knowledge, very few data (presented below) are available for C.
avidum strains compared to the literature on C. acnes. I found no molecular-level
investigations describing the creation of knockout C. avidum strains for putative
virulence factors in order to test their functions, properties, and interactions with
eukaryotic cells. Nevertheless, C. avidum is able to produce several biologically active
substances, including enzymes that break down mucopolysaccharides, such as hyal-
uronidase and neuraminidase (53). Propionicins and bacteriocins are also produced by
C. avidum strains, some of which are able to inhibit all the C. acnes or C. granulosum
strains tested (54).
Lipase
In Cutibacterium species, the triacylglycerol lipase gene is one of the most-studied
and most significant virulence factors. This extracellular enzyme is produced by most
Cutibacterium species. By comparing the C. avidum, C. acnes, and C. granulosum lipase
gene sequences, I found a higher degree of identity between the C. avidum and C.
acnes gehA genes, with 94% similarity. The GehA and GehB proteins, on the other hand,
share 96% and 86% similarities, respectively, between C. avidum and C. acnes se-
quences.
The lipase encoded by the gehA gene has a crucial impact by hydrolyzing triglyc-
erides of sebum, which partially explains acne pathogenesis. The release of irritating
free fatty acids within the pilosebaceous follicles can explain the inflammatory skin
disorder commonly observed in acne. Whether C. acnes and, to a lesser extent, C.
granulosum have clearly been involved in acne, the role of C. avidum remains unclear.
However, despite the different ecological niches, Nakamura et al. were able to use
gas-liquid chromatographic analysis to demonstrate higher activity of the C. avidum
ATCC 25577 lipase (3-fold more butyric acid produced) than that of the C. acnes ATCC
11827 lipase (55). Moreover, while the lipase activity of C. acnes has been reported to
be expressed during the early phase of growth, various investigations have revealed an
effect of substrates present in the medium on lipase activity expression. That is, lipase
expression is reduced by glucose addition (56).
In C. avidum strains, lipase is able to metabolize various substances, giving rise to the
free fatty acids responsible for bad odor, especially in axillary osmidrosis (58). To avoid
or limit these disadvantages, various substances (jumi-haidoku-to or unsei-in) are used
to decrease or suppress propionic and butyric acid production (57). Although C. avidum

is still considered a weak pathogen, the degradative properties of lipase may cause host
tissue and cell damage. Other Gram-positive bacteria with lipase activity have also been
described, such as Staphylococcus caprae (59) and Staphylococcus aureus USA300, for
which a contribution to colonization and persistence on human skin is likely to be the
most plausible role. Moreover, a lipase-deleted S. aureus strain revealed a decreased
ability to produce biofilm compared to that of the parent strain. Likewise, by using an
intraperitoneal infection model in mice, the same group demonstrated reduced for-
mation of peritoneal abscesses, probably correlated with a reduced bacterial load
compared to that with the wild type (60).
Other Gram-positive pathogens are able to produce lipase, such as bacteria of the
Burkholderia cepacia complex, with an impact in lung epithelial cell invasion (61), and

Candida parapsilosis, a skin commensal for which lipase modulates the immune re-
sponse of macrophages (62). Future studies clearly need to be performed to investigate
the pathogenesis of C. avidum infection, especially in liponecrotic breast abscess or skin
necrosis (15–17, 32).
Hyaluronate Lyase
Hyaluronate lyase, commonly referred to as hyaluronidase, is an extracellular en-
zyme involved in bacterial pathogenesis (63). This enzyme has been described for
several Gram-positive bacteria, such as Streptococcus pneumoniae and Streptococcus
pyogenes. Initially, it was described as a spreading factor due to its capacity to hydrolyze
extracellular matrix in eukaryotic hosts. One of its main substrates remains hyaluronan,
a major extracellular matrix component in vertebrates. CD44 is the hyaluronan cell
surface receptor molecule. This receptor is present on various cell types, and its
importance has been reported previously for the normal immune response process
(64).
The three main Cutibacterium species recovered from the skin can produce hyal-
uronidase. Even though C. acnes strains are the most active (68.8%), with clearly
measurable hyaluronidase activity, C. avidum (45%) and C. granulosum (33.3%) strains
are also positive, despite significantly lower activity (65). The impact of C. avidum
hyaluronate lyase is still poorly understood. C. acnes carries two hyaluronate lyase gene
alleles. The HYL-IB/II variant was reported as a highly active allele of hyaluronate lyase
resulting in total hyaluronic acid degradation. In contrast, the low activity of the HYL-IA
variant resulted in partial hyaluronic acid degradation (66). The first C. acnes hyaluro-
nate lyase gene (hyl gene in type IB/II strains) has been cloned and sequenced to better
understand its structure, binding domains, and homology. Nevertheless, in Escherichia
coli, the active enzyme was reported to be only cell associated, not secreted. The
2,256-bp gene encodes an 82-kDa enzyme with a leader sequence of 32 amino acids
that is cleaved to produce the extracellular form of the mature protein (67).
Further studies with C. avidum strains are needed to investigate the role of C. avidum
hyaluronate lyase in skin or deep infections, the impact of potential variants, its
involvement in penetrating the extracellular matrix found in the tissues, and its
relationship with the immune system, as recently suggested for C. acnes (66).
Hemolysins
Like many other microorganisms, C. avidum is able to produce hemolysins, leading
to significant beta-hemolysis on blood agar plates (Fig. 4) and resulting in potential side
effects in the host. In 1977, Hoeffler demonstrated the most active hemolytic activity of
C. avidum strains compared to those of C. acnes and C. granulosum strains, regardless
of erythrocyte origin (68). Nevertheless, sheep erythrocytes seemed to be the most
resistant. These differences in erythrocyte susceptibility may indicate that there are
several substances with distinct hemolytic specificity but also differences in erythrocyte
membrane composition.
In 1982, a hemolysin was partially purified from C. avidum 1148, isolated from the
skin of a healthy adult. The molecular mass was determined to be 32 kDa. The
hemolytic spectrum of hemolysin was established by use of different animal erythro-

FIG 4 Colony morphologies of C. avidum and related species on Schaedler agar plates after 3 days of incubation

in an anaerobic atmosphere.
cytes. Hemolysin activity was completely lost by heating to 60°C for 5 min (53). Various
treatments also caused a complete loss of activity (the protein is sensitive to digestion
by pronase, DNase, or chymotrypsin), whereas RNase, lysozyme, and lipase did not
affect hemolysin activity. Inhibition of hemolysin activity was observed with lecithin but
not with cholesterol, which may be a receptor or substrate for hemolysin. Lastly, this C.
avidum hemolysin is a thiol-activated (oxygen-labile) type of hemolysin, since mercap-
toethanol protects it from losing activity in the atmosphere (53).
Historically, the cohemolytic factor has been reported for C. acnes strains (69). The
CAMP reaction corresponds to the synergistic lysis of erythrocytes by the interaction of
an extracellular protein (CAMP factor [for Christie–Atkins–Munch-Petersen]) produced
by different species, especially Streptococcus agalactiae with the S. aureus sphingomy-
elinase C (␤-toxin). Previous investigations of potential hemolytic activity were per-
formed by adding different concentrations of washed animal erythrocytes to a blood
agar base (68). Some hemolytic ability differences were observed. For C. acnes, various
hemolysins were described (69). In order to study the role of each CAMP factor,
Sörensen et al. (70) conducted a study on CAMP-deleted strains. Valanne and col-
leagues (69) demonstrated the presence of the five genes encoding CAMP factors in C.
acnes subgroups IA, IB, and II. Interestingly, observed differences showed variations at
the expression level rather than missing genes (70). Moreover, when the camp2 or
camp4 gene was deleted, only the Δcamp2 strain revealed reduced hemolytic activity
with sheep erythrocytes, suggesting that this CAMP factor is the major active cohe-
molytic factor, but in the IA phylotype C. acnes genetic background. More recently, my
group revealed that only strains belonging to IA and IB phylotypes are hemolytic (71),
with different behaviors, as previously reported (37).
Concerning C. avidum, no such phylogenetic approach has been developed to
analyze differences. According to comparative genomic analysis, different CAMP
factors, especially the camp1, camp2, and camp4 genes, are missing in C. avidum
genomes. Nevertheless, C. avidum isolates contain two CAMP factor genes with
high degrees of similarity to the genes for CAMP factors 3 and 5 (HMPREF9153_0708
and HMPREF9153_1759 in strain ATCC 25577) (13). Thus, this characteristic is
routinely important for strain identification, since the hemolytic activity of C.
avidum strains is quite high compared to that of C. acnes (71, 72) (Fig. 4).
Biofilm
Bacterial biofilm is a structured bacterial community producing an exopolysaccha-
ride matrix that can allow adherence to foreign devices or viable cell surfaces. The
change in behavior (from planktonic cells to an aggregated state or sessile form) is
activated by a mechanism of chemical communication that differs between species
(73). Biofilm formation has been well studied and characterized for different clinical
strains of Cutibacterium species, especially C. acnes (6, 51, 74, 75). To study the ability
of C. acnes strains to produce such a structure, various experiments were performed in
vitro. These experiments used 96-well plates (76, 77), glass beads (75), or various

biomaterials, such as titanium, steel, and silicone (73, 78), as adhesion surfaces. These
experiments demonstrated that different C. acnes strains are able to form early or
mature biofilms on all these surfaces.
In 2009, Holmberg et al. showed that only invasive C. acnes strains could produce
biofilm, regardless of their phylotype, whereas strains isolated from healthy skin
produced less biofilm (77). The ability of C. acnes to create a biofilm is also influenced
by the presence of plasma and the surface roughness of the biomaterial (77, 79).
Although in vitro studies are important for analyzing the properties and extrinsic factors
influencing the ability of strains to produce biofilm, the clinical relevance of this ability
was demonstrated only recently by Tunney et al., by use of immunofluorescence
microscopy with a C. acnes monoclonal antibody or 16S rRNA gene amplification from

sonicate fluids and inflammation cell infiltration in tissues around the implanted device
(80). However, the various mechanisms and steps necessary for the synthesis of an
exopolysaccharide by C. acnes strains remain to be elucidated. Jahns et al. reported that
C. acnes biofilm composition is similar to that of other bacteria, comprising extracellular
DNA, proteins, and glycosylated residues (81). Furthermore, transcriptomic analysis
revealed that the genes coding for DNA replication, cell wall synthesis, and cell division
were underexpressed in sessile bacteria, whereas the gene encoding the virulence
factor CAMP 1 was overexpressed (81).
Conversely, very little clinical and fundamental information is currently available for
C. avidum strains and associated biofilm production ability. The recent description of
device-related infections by genomic analysis revealed new data for this species (14).
Indeed, the three C. avidum strains analyzed possess a gene involved in the exopoly-
saccharide synthesis pathway. Consequently, C. avidum pathogenicity may be linked to
this bacterial property leading to clinically significant infections, including abscess
formation, in the presence of foreign-body material, such as prosthetic joint devices.
However, according to the study of Achermann et al., no distinction between C. avidum
and C. acnes isolates was reported for the ability to produce a biofilm, despite the
presence of an exopolysaccharide island detected in all C. avidum clinical strains (72).
Future studies are clearly required to analyze the kinetics of biofilm formation and
biofilm maturation and the complex existence of biofilm for this species.
Biofilm formation seems to be regulated by the secretion of particular molecules, a
process referred to as “quorum sensing” (82, 83). Quorum sensing is a specific type of
communication between bacterial cells, based on a self-induced process: when the
chemical signals begin to accumulate in the area surrounding the environment of the
pathogen, the latter undergoes a series of physiological modifications that permit
biofilm formation (14). Thus, it may be interesting to determine if C. avidum strains are
unable to send or send only “safety” biomolecules when present in “controlled”
quantities under commensal conditions but become pathogenic and send “danger”
biomolecules via quorum sensing in the form of excess free fatty acid production,
which activates different Toll-like receptors (TLRs) (TLRs 2 and 4), as occurs in the C.
acnes bacterial community (84).
For this reason, genomic, molecular, biofilm kinetics, and endpoint studies are
required to investigate the role of biofilm in C. avidum infections. Indeed, in daily
practice, we clearly know that biofilm-related infections are difficult to eradicate (75,
85). Sessile C. avidum cells will be more resistant than their planktonic counterparts to
many of the most-prescribed antibiotics, and they are also able to resist antibodies and
evade the immune response. Finally, sessile cells seem to synthesize more lipases as
well as various molecules involved in quorum sensing, leading to potentially increased
inflammation and abscess formation (58, 76). New perspectives for studying biofilm
ability by confocal microscopy, by dynamic study of the biofilm bacterial population
along with investigations of specific intercellular communications or with animal
models (74, 75), are opening up research projects with this poorly known and too often
underestimated species.

Other Virulence Factors

While studying other enzymatic properties among a collection of different C. avidum
strains, Hoeffler demonstrated that this species is also able to produce gelatinase,
DNase, lecithinase, and a weak phosphatase (68). The role of gelatinase was previously
reported for other species, with the ability of Enterococcus faecalis to induce epithelial
cell permeability, though its deletion can also prevent biofilm prevention (86). Lecithi-
nase activity may also be involved in bacterial spread in culture. Such enzymes are
required during Listeria monocytogenes infection of human epithelial cells (87). Finally,
extracellular DNases in bacteria can contribute to physiology and pathogenesis by
using different significant pathways. In group A streptococci, the DNase Sda1 sup-
presses both the TLR9-mediated innate immune response and macrophage bactericidal

activity, whereas in S. aureus, biofilm susceptibility is time dependent: early biofilms
prove to be DNase susceptible, while advanced biofilms are more resistant (88, 89). All
these putative virulence factors in C. avidum strains should be investigated to better
understand host-bacterium interactions at the molecular level that lead to abscesses,
biofilm formation, or deep or device-related infections.
Host-Bacterium Interaction
For C. acnes, innate immune activation occurs at least partially via TLR2 as well as the
NLRP3 inflammasome (90); thus, C. avidum is also likely to interact with these receptors.
Nevertheless, C. avidum can also interact with its own environment via the production
of chemical compounds. Some types of biologically active compounds may have a
crucial role in C. avidum disease physiopathology, particularly that of skin abscesses
(see the clinical section of this review). In fact, within the skin microbiota but also inside
the pilosebaceous duct in moist areas, C. avidum PF77 is able to interact with Vero cells,
but much more so with skin fibroblasts, with a cytotoxic effect. Investigation of the
impact of C. avidum propionate production revealed a correlation between cytotoxicity
and propionate concentrations in samples (91). The cytotoxic effect could be related to
this production and cause damage to various human cells. Moreover, propionate and
other carboxylic acid salts (C1 to C5) were assessed at the same concentrations and
demonstrated various degrees of cytotoxicity depending on chain length. Similar
results have also been observed with certain clades of C. acnes strains incubated with
bone cells (52).
Similarly, proinflammatory molecule production by C. avidum was investigated
using culture supernatants in two different models: a guinea pig ileum preparation and
a rat fundic strip preparation. These tissue models can be considered models for the
receptors implicated in the inflammation cascade. This valuable tool showed that the
tissue response observed due to histamine, tryptamine, or short-chain fatty acid salts is
systematically dose dependent, leading to smooth muscle contraction or relaxation.
Interestingly, the levels of histamine-like or tryptamine-like activities measured in C.
avidum PF77 culture supernatants grown in the presence or absence of histidine or
tryptophan revealed less histamine-like activity but more tryptamine-like activity for C.
avidum than for C. acnes. The histamine, tryptamine, and short-chain fatty acids
produced by C. avidum may therefore play a role, especially in sebaceous follicles, as
biologically active substances leading to inflammation (92). Nevertheless, further stud-
ies should be conducted to better understand these biological effects in vivo, according
to the potential microenvironmental differences that can clearly exist between different
grades of normal and acne-prone skin (pH, oxygen tension, bacterial competition,
resource levels, etc.).
Immune Stimulation
This section and the following one summarize early studies focused on testing
immune stimulation and antitumor effects of different Cutibacterium strains, which
were probably less well characterized and identified at the molecular level than they
are now.

In 1976, the mitogenic activity of bacterial cells or cell wall fractions prepared from
various Cutibacterium species, including C. avidum ATCC 25577, was investigated. The
results suggested that the cell walls used acted as mitogens on both thymus-derived
lymphocytes (T cells) and bone marrow-derived lymphocytes (B cells) (19). This mito-
genic potential was suggested to be dose dependent: it was reported to be more active
with 100 ␮g than with 10 ␮g C. avidum cell wall preparations (93).
Lymphoreticular stimulatory properties of anaerobic coryneforms may vary remark-
ably due to cell wall composition, especially the composition of the peptidoglycan. A
first comparative study on the cell wall composition of some Cutibacterium-related
species revealed various lymphoreticular stimulatory activities. The cell wall of C.
granulosum strains, which is inactive in assays for such activation, was reported to

contain a larger amount of alanine than that for most of the C. avidum or C. acnes
strains investigated (94). Accordingly, significantly higher antibody titers toward the
acidic polysaccharide fraction are found in severe acne patients than in healthy
patients. These antibodies may cross-react with C. avidum serotype I or II. These
findings suggest that this external bacterial component may participate in the inflam-
mation of acne (95). Nevertheless, no recent studies have reported specificity and
antibody titer levels against C. avidum or C. acnes in acne or other patients.
The key role played by propionibacteria in modulating the immune response has
been reported based on different approaches. Indeed, the three main species, including
C. avidum, were described to be immunomodifiers stimulating the different cell subset
populations implicated in nonspecific antibacterial, antiviral, and antitumor resistances
(21). The higher activity for C. avidum was confirmed in parallel with antitumor activity
by Cummins and Linn (23). Later, the Pulverer group demonstrated the ability of C.
avidum KP-40 (a strain included in Pulverer group patent CA1197798A) to potentially
modify and enhance thymocyte/lymphocyte proliferation, maturation, emigration, and
activation (21).
Indeed, by using different models, different groups highlighted this ability. Com-
paring the effect of a killed C. avidum preparation to that of lipopolysaccharide on nitric
oxide synthase induction kinetics in a rat model, Dhouib et al. confirmed that Kupffer
cells were responsible for the uptake of bacteria, like other reticuloendothelial cells,
such as specialized macrophages (96). Moreover, several tests (chemiluminescence
responses of human leukocytes and mouse peritoneal macrophages, spleen enlarge-
ment, and pulmonary tumor colonization) revealed that the C. avidum KP-40 isolate
displayed clear immunostimulatory and antineoplastic activities (97–99).
Although there have been numerous surveys on the Cutibacterium active cell wall
element implicated in antitumor activity (see the next section), the active cell wall
structure responsible for protecting against viral infections remains unknown. As
previously described, Cutibacterium can stimulate the reticuloendothelial system, but it
can also trigger immune cell modifications, in particular macrophage activation, natural
killer cell activation, and interferon induction. In a mouse viral infection model, a
protective role of the Cutibacterium cell wall fraction in pretreatment was demonstrated
(100, 101).
At the veterinary level, using various models, the same groups described the
usefulness and probable positive economic impact on farms and animal breeding of
the potent granulopoiesis response after C. avidum immune stimulation. By use of 10
different test parameters, it was suggested that the role played by this preparation in
protection against bacterial and viral infections leads to improved conditions for animal
development and growth (faster weight gain) (102, 103). The adjuvant properties of C.
avidum KP-40 have also been reported for various swine models (104–106).
Thus, C. avidum proved to be a potent and sometimes excellent stimulator of the
macrophage-monocyte system and an inducer of endogenous interferons (107, 108).
This bacterium also seems to be able to act as an adjuvant with vaccines. C. avidum and
its closest relatives are clearly involved in inflammation (chemokines and cytokines)
(109) but can also interact with the immune system (macrophages, dendritic cells,
polymorphonuclear cells, and natural killer cells) (110–112). Consequently, different

clinical applications have been proposed. For example, it has been suggested that a
10-mg per os administration of C. avidum KP-40 (twice a day) for 7 days may restore the
downregulated immune response resulting from intense sports activities (113). Lastly,
this bacterium can modify the gut microbiota, adhere to intestinal epithelial cells, and
produce vitamin B12 as well as inhibiting negative microbiota by fatty acid release and
bacteriocin production (114).
These properties are most likely due to specific interactions at the molecular level.
Members of the Cutibacterium genus interact closely with immune cells via multiple
receptors, particularly TLR2, TLR4, and TLR9 (90, 109, 115, 116). These species are
triggers for innate immunity and can promote inflammation but can also mediate or
modulate the immune system in a complex equilibrium (117). It should be noted that,

since 1988, only C. acnes has been shown to induce granulocyte activating factor
release as a possible mediator of inflammation, a signature of its potential role in the
chronic inflammatory disease acne, in which C. avidum is less involved (see the
information on ecological niches in the earlier sections) (118).
Antitumor Activity
In the 1960s, anaerobic coryneforms were used extensively as antitumor agents
following the discovery that, in mice, killed preparations were able to stimulate cells of
the reticuloendothelial system (18). A selection of C. acnes type I or II strains, C.
granulosum strains, C. avidum strain 4982, and control strains were examined for the
ability to inhibit tumor growth. Interestingly, C. avidum was marginally more effective
than the other strains at high-dose stimulation. In fact, this species stimulated effector
cells which were cytotoxic at high effector/target cell ratios. A relatively good correla-
tion between the drop in packed cell volume, splenomegaly, and the antitumor activity
of the C. avidum strain was observed. In contrast, no correlation was reported with this
model between antitumor activity and the extent of the inflammatory peritoneal
exudate induced after 4 days. These strains, which have been compared serologically
but also in terms of their biological activity, stimulated tumor immunity. This was only
an observation, however, especially for the C. avidum strain, and it would be premature
to speculate on a potential role for the soluble, cell-attaching polysaccharide antigen
(18).
Working with a significant panel of anaerobic coryneform strains, Cummins and Linn
confirmed the reticulostimulating properties of species (C. avidum, C. acnes, and C.
granulosum, but also classical Propionibacterium species, such as Propionibacterium
freudenreichii, Propionibacterium thoenii, and Propionibacterium jensenii) from a wide
variety of both geographic and clinical sources (23). Even though a considerable degree
of variation was observed (strain and mouse host variabilities), almost all the vaccines,
regardless of origin, produced splenomegaly in the recipient mice. C. avidum and C.
acnes type II were the most active as analyzed by the spleen weight ratio. As reported
earlier by McBride et al., the C. granulosum strains were either not active or less active
(18, 23). These results were later confirmed by Ko et al. by testing of immunostimulatory
potency by use of the spleen enlargement responses to different Cutibacterium species.
They found that 82%, 12%, and 0% of C. acnes, C. avidum, and C. granulosum strains,
respectively, displayed stimulatory activity (119). Moreover, the restriction of tumor
growth and the prolongation of mean survival time were observed only with strains
capable of inducing splenomegaly (⬎200 mg in a mouse model).
Differences in the immunostimulatory potencies of the species, correlated with
various resistances to degradation by macrophages and the prolonged presence of
nondisintegrated bacterial membrane elements in these cells, were accountable for the
reticuloendothelial stimulatory effect (119). For vaccines, investigations into the nature
of the reticulostimulatory substance by use of different inhibitory treatments (alanine
content) also suggested that the cell wall peptidoglycan and lipid fractions are the
active material. Nevertheless, a strong adjuvant effect (indirect effect) on B cell re-
sponses was noticed, and the primary action was on macrophages (120). The enhance-
ment of C. avidum antitumor activity was also confirmed to be macrophage dependent

in the model and depended on the restoration of T cell function in combination with
neurotropin (121). Later, the C. avidum KP-40 strain demonstrated immunostimulatory
activity, leading to the possible prevention of metastatic lung and liver colonization in
a mouse tumor model (122).
Nevertheless, in 1980, Roszkowski et al. revealed the ability of strains of C. avidum,
C. acnes, and C. granulosum to stimulate lymphocytes in vitro to similar degrees (123).
This group demonstrated that the same species injected in multiple doses of 1
mg/mouse appeared effective at retarding the growth of Sarcoma-180 in mice, with a
prolonged survival of Sarcoma-180-bearing mice (124). Subsequently, in 1985, Hojo et
al. demonstrated a dose effect for limiting BC47 tumor growth in a rat model, leading
to cytotoxic lymphocyte action following intraperitoneal injection of the C. avidum

strain (125).
In summary, C. avidum, like other Cutibacterium species, not only can stimulate
several cells largely involved in the immune system (antitumor activity) but also can
modulate this system at both the local level (especially at the skin level) and the
systemic level, with involvement of dendritic cells, macrophages, and NK cells.
According to in vitro and in vivo model results obtained during the early 1990s, the
Pulverer group in Germany decided to move from the bench to the bedside with
prospective randomized studies including patients with gastrointestinal tumors (212).
No effect was observed regarding postoperative complications or tumor recurrence
rates between the control and treatment groups (212), but after a preoperative
intravenous (i.v.) infusion of 10 mg of the whole C. avidum KP-40 preparation, they
demonstrated enhanced neopterin secretion in urine, with persistence of this bio-
marker. Indeed, neopterin is considered to be an early signal of the cellular immune
response (126). They also highlighted that only a lymphocyte subset population (the
natural killer cell population) was statistically increased following infusion of the same
preparation (same dose) (110, 111). This immunotherapy approach with the C. avidum
KP-40 preparation and its therapeutic added value remain debatable according to the
type of tumor treated, the strain selected, the route and timing of the administration
scheme, and the potential adverse events (127, 128).
DIAGNOSTIC PROCEDURES
Conventional Microbial Cultures
The present section summarizes studies focused on nutritional requirements, espe-
cially examining the impacts of amino acids, Tween 80, glucose, pH, and oxygen on C.
avidum growth in conventional microbial cultures.
In 1978, Ferguson and Cummins developed a basal medium and improved the
growth rate of Cutibacterium spp. by adding various components (129). The nutritional
requirements of the three main Cutibacterium species, including C. avidum types I and
II, were consequently determined. The vitamin requirements for these anaerobic bac-
teria are consistent with the propionic acid fermentation pathway for carbohydrates
found in conventional propionibacteria, leading to propionic acid production.
According to the growth rates observed with numerous modifications of the basal
medium used, the nutritional requirements and the products of their metabolic path-
ways are quite similar. For more details, an excellent summary analyzing the effects of
selected nutrients on growth is provided in Table 2 of reference 129. C. avidum strains
can grow in a glucose-supplemented basal salts medium, pantothenate, biotin, thia-
mine, and 12 different amino acids. Six more amino acids should be added for C.
granulosum and C. acnes (asparagine, leucine, lysine, proline, threonine, and valine).
Nicotinamide had no effect on C. avidum growth. Other nutrients that were not
absolutely necessary factors, though they greatly enhanced the rate of growth, in-
cluded purines, sodium L-lactate, pyruvate, ␣-ketoglutarate, and, to a lesser degree,
Tween 80, a potential oleic acid source (129).
A Tween 80 concentration of 0.005% (vol/vol) resulted in the highest growth rate for
C. avidum strains, whereas 0.1% was better for the other species. This result is consistent
with the conditions encountered in their natural habitats (11, 28). Interestingly, this

concentration may play a role in the recovery rate for clinical skin samples, and
preincubation of the swab in a 0.1% Tween 20 – 0.15 M NaCl broth can improve
Cutibacterium detection, though it decreases the growth of coagulase-negative staph-
ylococci. Moreover, depending on the incubation conditions (anaerobic atmosphere or
not), colony growth may vary according to the medium used (blood agar, brucella agar,
Schaedler agar, or chocolate agar) and the incubation period (51).
In the 1980s, the effects of glucose concentration were investigated by Cunliffe and
coworkers, using different criteria: biomass, maximum specific growth rate, and extra-
cellular enzyme production by C. avidum and related species (130). Biomass increased
with glucose concentrations of up to 0.3% (wt/vol) for C. avidum, and then biomass
remained constant or decreased slightly. Under these conditions, lipase activity re-

mained virtually unchanged, while it was higher at 0.4%, and glucose concentrations of
up to 0.5% reduced this activity. Under glucose-free conditions, C. avidum gave
approximately 4-fold higher initial yields than those of C. granulosum. Consequently,
and unlike C. granulosum, C. avidum (along with C. acnes) displayed the highest
maximum specific growth rate when the strains were grown in a medium free of
glucose. Therefore, C. avidum seems to be the most nutritionally adaptable Cutibacte-
rium species. Indeed, this species grew well under glucose-free conditions, and its
biomass index in response to glucose supplementation was higher than those for other
species.
In contrast, it should be noted that C. avidum produced 20-fold less lipase per unit
of biomass than that produced by other species. Consequently, in light of their own
niches and the protein-rich and moist skin environment, C. avidum relies on proteinases
to scavenge for amino acids and peptides as a carbon/energy source, whereas C.
granulosum is dependent on other carbon/energy sources (basically saccharolytic
metabolism) (130).
To determine the impact of pH modifications, but also to mimic the reality of
microenvironmental skin changes, especially within sebaceous follicles, this variable
was similarly investigated. Using C. avidum, maximum biomass production was
achieved at pH 6.0. The same pH was considered to be optimum for the maximum
specific growth rate. According to potential enzyme denaturation (i.e., lipase for C.
avidum) at low or high pH values (⬍4.5 and ⬎6.5), lipase production was found to be
highest at pH 5.5. With C. acnes lipase, lyase, and phosphatase or C. granulosum lipase
and hyaluronate lyase, the same narrower pH range was reported (131). This pH value
is the most frequently encountered on the skin, regardless of location, thus explaining
the growth of Cutibacterium species in this specific microenvironment. Interestingly, at
this pH, lipase can hydrolyze sebum triglycerides produced by the sebaceous glands,
which may explain in part how this specific microbial ecology can affect skin homeo-
stasis at different levels, e.g., free fatty acid production, triglyceride degradation,
bacterial density, immune system dysregulation, and inflammation (109).
In the same way, oxygen concentration variations were investigated for various
Cutibacterium species. While the presence of oxygen reduced the maximum specific
growth rate for all species, C. avidum was unaffected by the presence of oxygen and
was the best species at adapting its growth to aerobic conditions (129, 132). This
difference is clearly observed with clinical strains in routine laboratory diagnosis.
Indeed, C. avidum can easily grow aerobically. Consequently, lipase production by C.
avidum was constant across all oxygen conditions tested, whereas production was
dramatically reduced for C. acnes and C. granulosum. The same observation was made
for both species with regard to hyaluronate lyase production (132). These different
investigations, with some modifications to growth conditions and their impacts on
growth rate (133), show that extracellular enzyme production may constitute an
excellent argument due to the role of these species in acne or deep infections, with or
without a device. Indeed, bacterial multiplication, but also the role of an enzyme such
as lipase or hyaluronate lyase, may be crucial in acne pathogenesis and Cutibacterium
species device-related infections (134).

To summarize, in daily practice, skin swabs, deep tissues, or biopsy specimens can
be used to isolate Cutibacterium species. These can be inoculated onto plates contain-
ing a medium consisting of 3% tryptone soy broth, 1% yeast extract, 1.2% bacterio-
logical agar, and 0.5% Tween 80 (28). As for C. acnes or C. granulosum, other media can
also be inoculated, such as Schaedler agar or chocolate agar plates or enrichment
broth, especially thioglycolate broth for clinical samples. All should be incubated
anaerobically for at least 7 to 10 days at 37°C and, depending on the type of infection,
for up to 14 days, especially for bone and joint infections (51). Interestingly, C. avidum
grows faster than C. acnes, most likely due to its metabolism (C. avidum is more aerobic
than C. acnes) and to the microenvironment (sebum, skin pH, etc.) (6, 51, 72).
Contamination Issues in Culture

The clinical significance of a culture positive for Cutibacterium species in different
cases (orthopedic and device-related infections, catheter shunt infections, breast spec-
imens, etc.) remains unclear, both concerning its potential role as a true pathogen
(infection), a real contaminant, or a skin invader due to skin incision (commensal
presence in deep tissue) and its potential role as a coinfectant (135). There are no
available guidelines for culture interpretation. How many media should be used, and
how long should the media be incubated? How many colonies are needed to distin-
guish contamination from infection? In the orthopedic field, Schaedler or thioglycolate
broths are usually incubated for up to 14 days (136). Recently, Bossard et al. reported
that shortening of the culture incubation time to 7 days would result in a 21.4%
increased rate of false-negative diagnosis; thus, they recommend that perioperative
samples be incubated for at least 10 days to detect Cutibacterium species and to
distinguish contamination from true infection (137). This delay of 10 or 11 days was also
proposed by Frangiamore et al. to separate early and late growth (138).
C. avidum and C. acnes infections are defined as the isolation of Cutibacterium
species from at least two or more specimens, with the presence of typical perioperative
findings and/or local signs of infection (139). Nevertheless, Cutibacterium species seem
likely to be spread throughout the surgical field from the subdermal layer after the skin
incision via soft tissue handling by the surgeon and instruments, even during primary
shoulder arthroplasty (140, 141). On the other hand, to assess the Cutibacterium species
catheter colonization rate, 1,000 vascular catheters were treated by the roll-plate
method, but after conventional aerobic incubation, all primary culture plates were
reincubated for anaerobe detection. Significant numbers of C. acnes cells were detected
in these catheter samples (for 39 patients, representing 14.7% of all positive catheters
[most likely by contamination of the device, with noninfection or bloodstream infec-
tion]). These data suggest that methodological shortcomings impair the proper detec-
tion of Cutibacterium spp. as a hidden and possible cause of contamination or catheter-
related infections (142).
Therefore, to avoid misinterpretation or overinterpretation of positive cultures
(contamination), new approaches are necessary to minimize this contact and to reduce
the colonization rate (new antibiotic prophylaxis and new strategies for skin prepara-
tion). Different risk factors have been pointed out: male sex, previous surgeries, and
preoperative corticosteroid injections are associated with a higher likelihood of bacte-
rial growth in culture. More clinical studies are needed to confirm and define the proper
role of these risk factors (143, 144). Moreover, the relative risk of obtaining a positive
Cutibacterium culture was 2-fold greater for the anterolateral approach than for the
deltopectoral approach in total shoulder arthroplasty (135). Finally, considering the
clinical meaning of unexpected positive Cutibacterium cultures, in almost 25% of cases,
unexpected culture had no clinical significance, and persistent infection was demon-
strated in only 10% of cases (143, 145).
Identification Methods
In the 1970s and 1980s, C. avidum strains were identified by colony appearance:
glossy, rust colored, and sometimes mucoid on blood agar or Schaedler agar plates

(146). Confirmation was obtained by testing indole production, hemolysin production,

and nitrate reduction (28).
Serological and Precipitin Tests for Identification of Cutibacterium-Related

Species
Before the modern era with MALDI-TOF technology, several teams developed aggluti-
nation tests to separate and identify the different species or groups of species within the
Cutibacterium group (147–149). In 1975, Cummins proposed a serological identification
method using trichloroacetic acid extracts. This method was evaluated on 142 human
skin strains, including C. avidum, C. acnes, and C. granulosum strains. An excellent
agreement between serological results and the fermentation patterns of the strains was
reported (150). Serological reactions were generally well developed and easier to

interpret under standard conditions for C. avidum and C. acnes than for C. granulosum.
The existence of two serotypes was confirmed for both C. avidum and C. acnes strains,
even though investigation of a broad collection of C. avidum isolates could not be
performed. The majority of the strains provided and analyzed were recovered from
suppurative lesions. It was suggested that these microorganisms are not common in
the skin microbiota, although they certainly do occur there.
Cell Wall Analysis and Protein Pattern Analysis

Among anaerobic coryneform bacteria, different Cutibacterium species were initially
defined by DNA homology relationships and cell wall analysis. Cummins confirmed
these different species and demonstrated that C. avidum and C. acnes can be subdi-
vided further into two subtypes (150). Later, the protein patterns obtained after
Coomassie brilliant blue staining of polyacrylamide gels proved to be highly distinctive
and reproducible. Consequently, including different C. avidum strains of various origins
(hip sinus, pelvic abscess, inguinal abscess, wound swab earlobe cyst, and normal skin),
the protein patterns confirmed the earlier results of DNA homology and immunodif-
fusion studies, indicating a close relationship between C. avidum and C. acnes, along
with the possibility of distinguishing both serotypes.
Moreover, malate dehydrogenase proteins from the three species had different
motilities, with strong, well-defined single bands. Enzymes from C. avidum type I and II
strains could not be distinguished. This method was distinctive for each species,
especially to differentiate subtypes between C. avidum and C. acnes (22).
Polyacrylamide gel electrophoresis (PAGE) of whole-cell proteins of different Cutibacte-
rium isolates of various origins demonstrated no particular protein profile, regardless of
origin or patient (healthy skin patient or acne patient). This investigation confirmed the
proximity between C. avidum and C. acnes, with close or identical protein patterns,
whereas C. granulosum can clearly be distinguished by PAGE (151).
The C. avidum ATCC 25577 cell wall polysaccharide composition consists mainly of
galactose and mannose, whereas glucose, galactose, and mannose are found for C.
acnes ATCC 11829, C. acnes C7, and C. granulosum ATCC 25564 (19). Moreover, the
analysis of the cell wall composition of various Cutibacterium spp. revealed a new
component: 2,3-diamino-2,3-dideoxyglucuronic acid. This component makes up be-
tween 3 and 5% of the total cell wall polysaccharides of the closely related species C.
avidum and C. acnes, regardless of type (152). On the other hand, it appears to be
present in the C. granulosum cell wall, but in much smaller amounts, as also revealed
by genomic comparisons (see the section on genomic analysis).
Concerning cell wall composition, all the strains have some combination of galac-
tose, glucose, and mannose as cell wall sugars, and most have alanine, glutamic acid,
glycine, and L-␣-␧-diaminopimelic acid as amino acids of peptidoglycan. However, a
few strains that already have some differences in the C. acnes and C. avidum homology
groups have meso-diaminopimelic acid and minimal amounts of glycine (41). These
data gave rise to what was probably the first description of C. acnes types I and II based
on antiserum analysis, leading to the first phylogenetic study of these types. Later,
other methods for discriminating both serotypes, based on bacteriophage and fermen-

tation typing as well as immunofluorescence assay with polyclonal antisera, were also
reported (34). Unfortunately, to the best of my knowledge, the lack of antisera for C.
avidum prevented any further investigation with these methods, and now the genomic
era will bring deeper information on and new insights into the population diversity of
C. avidum.
Susceptibility to Free Fatty Acids

As discussed above, Cutibacterium species produce lipase, considered to be one of
the main virulence factors. A complex mixture of various fatty acids with different chain
structures and unsaturation positions can be detected on the human skin surface.
These compounds from sebum triglycerides are hydrolyzed by lipase in pilosebaceous
follicles. This local chemical reaction may have a significant impact on the physiopa-

thology of inflammatory skin lesions linked to specific interactions with the host
immune system (24, 109, 117). Ko et al. focused on differences in the susceptibilities of
C. avidum and related species to a selection of free fatty acids. MIC values for the
different strains tested demonstrated unsaturated acids to be more inhibitory than
saturated ones. Octadeca-9,12,15-trienoic acid appeared to be the most inhibitory acid.
In contrast, decanoic acid had the lowest bacteriostatic activity. C. avidum appeared to
be more resistant than the C. granulosum strains to octadeca-9,12,15-trienoic acid.
Finally, C. avidum strains appeared to be more susceptible than C. granulosum and C.
acnes strains to decanoic and dodecanoic acids.
Fatty acid analysis may explain the changes in Cutibacterium density and distribu-
tion in both healthy and sick patients, especially acne patients. Interestingly, these data
can be used for bacterial identification, and their analysis may also serve as an
additional characteristic for Cutibacterium classification, as recently reported (8, 9).
Although cellular fatty acid composition analysis is restricted to reference labora-
tories and probably not possible in routine practice, this identification step can be
exploited to discriminate between different profiles among closely related species.
Thus, C. avidum reveals a significant proportion of Ci15:0 (153), as recently described for
the new species Cutibacterium namnetense (9).
Biochemical and Enzyme Profiles

As early as 1977, Hoeffler described the production of toxins and enzymes by
Cutibacterium-related species, including C. avidum, C. acnes, and C. granulosum, to
determine whether or not there are differences in their profiles as a means of identi-
fication, and possibly as a virulence marker (68).
With a collection of 33 C. avidum isolates (from normal human skin), 40 C. acnes
isolates (from human hair, acne lesions, and nonspecific infections of the cervicofacial
region), and 18 C. granulosum isolates (from acne lesions and normal human skin), the
production of chondroitin sulfatase, hyaluronidase, DNase, phosphatase, and lecithi-
nase was tested.
Chondroitin sulfatase was not detected in most C. avidum strains, whereas 68% of
C. acnes strains were found to be producers. Cutibacterium species phosphatase activity
was shown to be weakly positive or negative. In contrast, most C. avidum strains
exhibited DNase and gelatinase activities, but most of them were not producers of
hyaluronidase and lecithinase (only 18% and 27% produced these enzymes, respec-
tively) (68).
Looking at the regional distribution of cutaneous Cutibacterium strains, McGinley et
al. demonstrated that all 52 clinical C. avidum strains produced gelatinase and hydro-
lyzed casein. In contrast, C. granulosum strains did not metabolize these compounds,
and most C. acnes strains produced indole (87%) (11).
In a recent study of the physiological and functional characteristics of Cutibacterium
strains of the poultry microbiota, the biochemical properties reported to characterize
the C. avidum isolates studied were as follows: catalase activity, gelatin liquefaction, and
esculin hydrolysis. The main substrates used to identify the isolates (100%) by acid
formation were as follows: erythritol, D-galactose, N-acetyl-D-glucosamine, maltose,

TABLE 3 Biochemical characteristics of C. avidum compared to the other Propionibacterium species from the skin environmenta
Result
C. avidum C. acnes C. granulosum C. namnetense
Characteristic CCUG 36754T CCUG 1794T CCUG 32987T CCUG66358T
Catalase ⫹ ⫹ ⫹ ⫹
Beta-hemolysisb ⫹⫹⫹ ⫹⫹ ⫺ ⫹
Acid produced from:
Glucose ⫹ ⫺ ⫹ ⫹
Ribose ⫹ ⫹ ⫺ ⫹
Xylose ⫺ ⫺ ⫺ ⫺
Mannitol ⫺ ⫹ ⫹ ⫺
Maltose ⫹ ⫺ ⫹ ⫺
Lactose ⫺ ⫺ ⫺ ⫺

Sucrose ⫹ ⫺ ⫹ ⫺
Glycogen ⫺ ⫺ ⫺ ⫺
Raffinose ⫺ ⫺ ⫹ ⫺
Production of:
Indole ⫺ ⫹ ⫺ ⫹
Nitrate reductase ⫺ ⫹ ⫺ ⫹
␣-Galactosidase ⫹ ⫺ ⫺ ⫺
␤-Galactosidase ⫺ ⫹ ⫺ ⫹
␣-Glucosidase ⫹ ⫺ ⫹ ⫹
␤-Glucosidase ⫹ ⫺ ⫺ ⫺
Arginine arylamidase ⫹ ⫹ ⫺ ⫺
Proline arylamidase ⫹ ⫹ ⫹ ⫺
Phenylalanine arylamidase ⫹ ⫺ ⫺ ⫺
Leucine arylamidase ⫹ ⫺ ⫺ ⫺
Tyrosine arylamidase ⫹ ⫺ ⫺ ⫺
Glycine arylamidase ⫹ ⫹ ⫺ ⫹
Histidine arylamidase ⫺ ⫺ ⫺ ⫺
Glutamyl glutamic acid arylamidase ⫹ ⫺ ⫺ ⫺
Hydrolysis of:
Esculin ⫹ ⫺ ⫺ ⫺
Gelatin ⫹ ⫹ ⫺ ⫹
aCCUG, Culture Collection of the University of Gothenburg (Swedish culture collection). All these strains are type strains. C. avidum CCUG 36754 belonged to the
Pasteur Institute (unknown origin; Prevot collection), C. acnes CCUG 1794 was recovered from an acne lesion in human facial skin, C. granulosum CCUG 32987 was
isolated from human skin, and C. namnetense CCUG 66358 was recovered from a bone infection.
bThe number of plus symbols reflects the relative level of hemolysis. For C. acnes CCUG 1794T, the hemolysis level is ⫹⫹, but for other strains of this species, the
level depends on the phylotype (phylotypes II and III are not hemolytic).
mannitol, mannose, sucrose, and trehalose (33). Table 3 provides a summary of the key
differences in terms of biochemical characteristics between the Cutibacterium species.
In daily practice, routine tests were performed, such as Gram staining of suspected
orange colonies with significant beta-hemolysis growing on Schaedler or chocolate
blood agar plates and testing for positive reactions for catalase, esculin, and CAMP
factor. Moreover, a negative indole spot can distinguish C. avidum from other Cutibac-
terium spp. These tests can be oriented toward routine identification (screening tests).
Nevertheless, although these parameters were useful at one time, they are now
considered obsolete due to the potential risk of reading errors by the operator. Thus,
with the advent of MALDI-TOF technology, definitive and accurate identification is
achieved quickly, and hence the other methods have been abandoned or restricted to
use in reference laboratories.
Molecular Identification
Molecular methods have revolutionized bacterial identification and have high-
lighted discrepancies, potential discordant results, and errors in phenotypic iden-
tifications that, until recently, were the only ones available. Bacterial molecular
identification is based on the sequence analysis of a species-specific target gene.
The sequence obtained is then run against various publicly available databases via
the Internet. Depending on the chosen target, different software applications can be

used, such as the GenBank database (https://blast.ncbi.nlm.nih.gov/Blast.cgi), RIDOM

(http://www.ridom.de/seqsphere/), or the BIBI database (https://umr5558-bibiserv.univ
-lyon1.fr/lebibi/lebibi.cgi).
In the 1970s, Woese et al. proposed the use of the first molecular clock scale based
on 16S rRNA gene analysis as a phylogenetic tool (154). The 16S rRNA gene sequencing
method can be performed routinely in microbiology laboratories either for phyloge-
netic analysis (9) or for the description of species not available in phenotypic or
MALDI-TOF databases (see below) (155, 156). However, using this target, multiple cases
of probable clinical misidentifications versus conventional culture-based identifications
across a wide range of bacteria have been reported (157). Indeed, depending on the
species or genus analyzed, 16S rRNA gene sequencing does not warrant an accurate

delineation of bacterial species, especially for the Cutibacterium or Propionibacterium
species with bacterial subgroups or complexes (Fig. 5A).
Consequently, the current species definition by molecular identification proposes a
16S rRNA gene divergence of 1.3%, and a genus divergence cutoff of 2% has been
validated (about 97% similarity). For more information, see the review by Georgiades
and Raoult (158). A second important remark is that regarding the size of the sequence
analyzed, the bigger the better. In other words, the more 16S rRNA gene domains are
analyzed, the better the resulting species identifications will be.
Finally, before genomic comparison and the revolution of the genomic era, other
molecular targets could be used to avoid a misleading species definition. The following
targets have been used to identify a broad range of Gram-positive and -negative patho-
gens: rpoB, encoding the ␤-subunit of RNA polymerase; gyrB, encoding the ␤-subunit of
DNA gyrase; sodA, encoding superoxide dismutase; hsp60, encoding a heat shock protein;
recA, encoding a bacterial recombination protein; and tuf, encoding an elongation factor.
By using rpoB and gyrB in the routine lab (Fig. 5B and C), my colleagues and I can
clearly distinguish all the members of the genus Cutibacterium with a greater discrim-
inatory power than that for 16S rRNA gene sequencing (9). Moreover, we have
demonstrated the ability of gyrB sequencing to distinguish C. avidum from C. nam-
netense and the different lineages of the C. acnes clinical strains misidentified by the
phenotypic method prior to the MALDI-TOF revolution.
Interestingly, as suspected with genomic analysis (see above), we can clearly see two
subpopulations of C. avidum strains, highlighting the probable existence of two sub-
types of C. avidum strains by gyrB analysis (Fig. 5C), with group 1 comprising the ATCC
25577 and 44067 strains and group 2 comprising the TM16 and DPC6544 strains.
MALDI-TOF Identification
For the past decade, matrix-assisted laser desorption ionization–time of flight mass
spectrometry (MALDI-TOF MS) has allowed quick and easy bacterial identification (159).
This strategy is relevant for both pathogenic and commensal bacteria. It can therefore
be performed directly on blood culture bottles, saving time and money by identifying
strict pathogens and contaminant bacteria (160). This tool is therefore useful and
effective for detection of Gram-positive bacteria, especially Cutibacterium spp.
Among the 544 anaerobes identified by MALDI-TOF, La Scola et al. reported an
unexpectedly high frequency of Cutibacterium spp. in sinus samples, representing 67%
of anaerobes recovered during sinusitis. C. avidum was the least frequent species, after
C. granulosum (n ⫽ 2) and C. acnes (n ⫽ 20) (161). Using another system, called
Andromas, clinical Cutibacterium strains were correctly identified to the species level,
including 18 C. avidum strains (162). Finally, only three C. avidum strains involved in a
prosthetic joint infection were reported by Peel et al., working on the ability of MALDI-TOF
to quickly identify and distinguish relevant pathogens versus culture contaminants (163).
Thanks to MALDI-TOF technology, C. avidum is increasingly found routinely in
laboratories and is probably misidentified by biochemical profile comparison. Regard-
ing the different spectra, several peaks are discriminant for better distinguishing
Cutibacterium species. Moreover, even though the possibility of identifying slow-
growing anaerobes with high accuracy and speed at low cost compared to those of

FIG 5 (A) Neighbor-joining (NJ) tree for the genera Cutibacterium, Propionibacterium, and Pseudopropionibacterium based
on 16S rRNA gene sequences, including those for 8 strains of Cutibacterium avidum. Twenty-three 16S rRNA gene
sequences selected from the GenBank database were aligned by using MEGA6 (www.megasoftware.net). Accession
numbers are indicated after strain names. The evolutionary history was inferred using the neighbor-joining method. The
(Continued on next page)

conventional procedures is relevant in daily practice, this tool also has the ability to
perform species typing (spectrum analysis and comparison) and to provide information
concerning the main phylogenetic groups of a species (164).
Further studies are needed to increase and develop the database of spectra with the
different MALDI-TOF systems available and then to investigate the potential diversity in
the C. avidum population as well as within the C. granulosum species. This would indeed
be relevant, since the strains are likely different depending on the site of isolation. I
provide an example of the four main spectra obtained with Vitek MALDI-TOF MS
(bioMérieux) in Fig. 6. Peaks and mass comparisons are also possible with various free
software applications, such as Mass. Using Vitek MALDI-TOF MS, one can separate the
four main Cutibacterium species by using specific peaks and masses. Thus, according to

the bioMérieux algorithm, C. avidum is characterized by the following 8 masses
compared to other Cutibacterium spp., with excellent discriminatory power and good to
excellent intensity: 3,685.4, 4,762, 5,125.9, 6,526.4, 6,776.6, 6,818.7, 7,369, and 13,045
Da. A detailed analysis is provided in Table 4, with 7, 13, and 11 specific masses with
high discriminatory power for C. acnes, C. granulosum, and C. namnetense, respectively,
as follows: for C. acnes, 3,518.4, 4,984.5, 7,182, 8,668, 8,690.7, 13,576.9, and 15,058.2 Da;
for C. granulosum, 3,481, 4,421, 4,789.9, 5,258.4, 5,688.8, 6,682.6, 6,709.7, 6,805, 6,985,
7,886, 8,646.3, 8,792, and 9,576.7 Da; and for C. namnetense, 3,412.2, 3,472, 3,972, 3,993,
4,767.5, 5,865, 6,766.9, 6,822.7, 7,287, 8,683.5, and 9,530 Da.
CLINICAL PRESENTATION
C. avidum infections may be expected following a skin break or loss of skin integrity.
This bacterium can also be recovered after surgery, especially after breast, cardiac, or
device-related (e.g., orthopedic) procedures. The prerequisites for these types of infections,
in particular infection adjacent to moist areas, increased body mass index (BMI), presence
of foreign-body devices, abscess development, etc., have been identified. A detailed review
of the specific clinical entities previously reported in the literature is discussed in this
section, including treatment options and highlighting the need for debridement in the case
of abscess. All C. avidum case reports are summarized in Table 5.
Breast Infections
Aesthetic (reduction mammoplasty) or reconstructive breast implant surgeries after
FIG 5 Legend (Continued)

figure shows the optimal tree, for which the sum of the branch lengths for 16S rRNA genes was 0.54511947. The
percentages of replicate trees in which the associated taxa clustered together in the bootstrap test (1,000 replicates) are
shown next to the branches. The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary
distances used to infer the phylogenetic tree. The evolutionary distances were computed using the Kimura 2-parameter
method and expressed as numbers of base substitutions per site. Codon positions included were 1st, 2nd, 3rd, and
noncoding positions. All ambiguous positions were removed for each sequence pair. There were a total of 1,302 positions
in the final data set. Evolutionary analyses were conducted in MEGA6. (B) Neighbor-joining tree for the genera Cutibac-
terium, Propionibacterium, and Pseudopropionibacterium based on rpoB gene sequences, including those for 6 strains of
Cutibacterium avidum. Thirsteen rpoB gene sequences selected from the GenBank database were aligned by using MEGA6
(www.megasoftware.net). Accession numbers are indicated after strain names. The evolutionary history was inferred using
the neighbor-joining method. The figure shows the optimal tree, for which the sum of the branch lengths for rpoB genes
was 0.60848730. The percentages of replicate trees in which the associated taxa clustered together in the bootstrap test
(1,000 replicates) are shown next to the branches. The tree is drawn to scale, with branch lengths in the same units as those
for the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were computed using the
Kimura 2-parameter method and expressed as numbers of base substitutions per site. Codon positions included were 1st,
2nd, 3rd, and noncoding positions. All ambiguous positions were removed for each sequence pair. There were a total of
3,264 positions in the final data set. Evolutionary analyses were conducted in MEGA6. (C) Neighbor-joining tree for the
genera Cutibacterium, Propionibacterium, and Pseudopropionibacterium based on gyrB gene sequences, including those for
4 strains of Cutibacterium avidum. Fourteen gyrB gene sequences selected from the GenBank database were aligned by
using MEGA6 (www.megasoftware.net). Accession numbers are indicated after strain names. The evolutionary history was
inferred using the neighbor-joining method. The figure shows the optimal tree, for which the sum of the branch lengths
for gyrB genes was 1.76236673. The percentages of replicate trees in which the associated taxa clustered together in the
bootstrap test (1,000 replicates) are shown next to the branches. The tree is drawn to scale, with branch lengths in the
same units as those for the evolutionary distances used to infer the phylogenetic tree. The evolutionary distances were
computed using the Kimura 2-parameter method and expressed as numbers of base substitutions per site. Codon
positions included were 1st, 2nd, 3rd, and noncoding positions. All ambiguous positions were removed for each sequence
pair. There were a total of 2,214 positions in the final data set. Evolutionary analyses were conducted in MEGA6.

FIG 6 Main spectra obtained by Vitek MALDI-TOF MS for C. avidum, C. granulosum, C. acnes, and C.
namnetense. The abscissa axis corresponds to the mass m/z (Da) and the ordinate axis to the peak intensity
(arbitrary units).

TABLE 4 Cutibacterium species identification by Vitek MALDI-TOF MS, with measured

masses and their discriminating powers and intensities according to bioMérieux
algorithms
Species Measured mass (Da) Discriminating power Intensity
C. avidum 3,143.3 ⫹⫹⫹ ⫹
3,404 ⫺⫺⫺ ⫺
3,460 ⫺⫺⫺ ⫹
3,685.4 ⫹⫹⫹ ⫹⫹
4,021 ⫹ ⫹
4,329 ⫺⫺⫺ ⫹
4,374 ⫺ ⫹⫹
4,485.1 ⫹ ⫹
4,762 ⫹⫹⫹ ⫹⫹
⫹⫹⫹ ⫹⫹

5,125.9
5,679 ⫹⫹⫹ ⫹
5,733 ⫹ ⫹⫹
5,792 ⫺⫺⫺ ⫹
6,380 ⫹ ⫹
6,490 ⫹⫹⫹ ⫺
6,526.4 ⫹⫹⫹ ⫹⫹
6,659.7 ⫹⫹⫹ ⫺
6,776.6 ⫹⫹⫹ ⫹⫹
6,918.7 ⫹⫹⫹ ⫹⫹
6,937 ⫺⫺⫺ ⫺
7,369 ⫹⫹⫹ ⫹⫹⫹
11,622 ⫹⫹⫹ ⫹
13,045 ⫹⫹⫹ ⫹⫹
C. acnes 3,290 ⫺⫺⫺ ⫹

3,404 ⫹ ⫹
3,424 ⫺⫺⫺ ⫹
3,519.4 ⫹⫹⫹ ⫹⫹
4,023 ⫺⫺⫺ ⫹
4,077 ⫺⫺⫺ ⫹⫹
4,346 ⫺⫺⫺ ⫹
4,374 ⫺⫺⫺ ⫹⫹⫹
4,396 ⫺⫺⫺ ⫹
4,484 ⫺⫺⫺ ⫹
4,694.4 ⫹⫹⫹ ⫹
4,779 ⫺⫺⫺ ⫹
4,984.5 ⫹⫹⫹ ⫹⫹
5,172 ⫺⫺⫺ ⫹⫹⫹
5,191 ⫺⫺⫺ ⫹
6,377 ⫺⫺⫺ ⫺
6,787 ⫺⫺⫺ ⫹⫹
7,182 ⫹⫹⫹ ⫹⫹⫹
8,668 ⫹⫹⫹ ⫹⫹
8,690.7 ⫹⫹⫹ ⫹⫹
9,964 ⫹ ⫹⫹⫹
13,576.9 ⫹⫹⫹ ⫹
15,058.2 ⫹⫹⫹ ⫹⫹⫹
C. granulosum 3,231.4 ⫹⫹⫹ ⫹

3,341 ⫹⫹⫹ ⫹
3,355.5 ⫹⫹⫹ ⫹
3,481 ⫹⫹⫹ ⫹⫹⫹
3,500.2 ⫹ ⫹
4,324 ⫹ ⫹⫹
4,397.3 ⫹ ⫹⫹
4,421 ⫹⫹⫹ ⫹⫹
4,789.9 ⫹⫹⫹ ⫹⫹
4,849.3 ⫹⫹⫹ ⫹
5,067.6 ⫹⫹⫹ ⫹
5,113 ⫹⫹⫹ ⫹
5,212.9 ⫹⫹⫹ ⫹
5,258.4 ⫹⫹⫹ ⫹⫹
5,688.8 ⫹⫹⫹ ⫹⫹
(Continued on next page)

TABLE 4 (Continued)
Species Measured mass (Da) Discriminating power Intensity
5,794.2 ⫹ ⫺
6,462 ⫹⫹⫹ ⫹⫹
6,682.6 ⫹⫹⫹ ⫹⫹
6,709.7 ⫹⫹⫹ ⫹⫹
6,781.5 ⫹ ⫹
6,805 ⫹⫹⫹ ⫹⫹
6,962 ⫺⫺⫺ ⫹⫹⫹
6,985 ⫹⫹⫹ ⫹⫹
7,886 ⫹⫹⫹ ⫹⫹⫹
8,266 ⫹ ⫹
8,646.3 ⫹⫹⫹ ⫹⫹
8,792 ⫹⫹⫹ ⫹⫹

9,222.4 ⫹⫹⫹ ⫹
9,257.9 ⫹⫹⫹ ⫹
9,576.7 ⫹⫹⫹ ⫹⫹
10,133 ⫹⫹⫹ ⫹
10,220 ⫹⫹⫹ ⫺
C. namnetense 3,177.7 ⫹⫹⫹ ⫹

3,247 ⫹⫹⫹ ⫹
3,290.2 ⫹ ⫹⫹
3,412.2 ⫹⫹⫹ ⫹⫹
3,425.6 ⫹ ⫹⫹
3,462.8 ⫹ ⫹
3,472 ⫹⫹⫹ ⫹⫹
3,501 ⫺⫺⫺ ⫹
3,957.9 ⫹⫹⫹ ⫹
3,972 ⫹⫹⫹ ⫹⫹⫹
3,993 ⫹⫹⫹ ⫹⫹
4,010 ⫹⫹⫹ ⫹
4,079 ⫹ ⫹⫹
4,343 ⫹ ⫹⫹
4,376 ⫺⫺⫺ ⫹⫹⫹
4,513 ⫹⫹⫹ ⫹
4,767.5 ⫹⫹⫹ ⫹⫹
4,781.6 ⫹ ⫹⫹
5,172 ⫹ ⫹⫹
5,193.1 ⫹ ⫹⫹
5,733 ⫺⫺⫺ ⫹
5,865 ⫹⫹⫹ ⫹⫹
6,405 ⫹⫹⫹ ⫹
6,766.9 ⫹⫹⫹ ⫹⫹⫹
6,822.7 ⫹⫹⫹ ⫹⫹
6,943 ⫹ ⫹⫹⫹
6,965.6 ⫹ ⫹⫹
7,287 ⫹⫹⫹ ⫹⫹⫹
8,260 ⫺⫺⫺ ⫺
8,683.5 ⫹⫹⫹ ⫹⫹
8,721.4 ⫹⫹⫹ ⫹
9,322 ⫹⫹⫹ ⫹
9,530 ⫹⫹⫹ ⫹⫹
9,556 ⫹⫹⫹ ⫹
9,964 ⫺⫺⫺ ⫺
10,067.6 ⫹⫹⫹ ⫹
10,383 ⫹⫹⫹ ⫹
12,801 ⫹⫹⫹ ⫺
carcinoma are potentially vulnerable to infections, in particular S. aureus, anaerobe, and

Cutibacterium infections, despite the increase in standardized preventive measures. An
initial case of breast infection due to C. avidum was reported in 2005 by Panagea et al.
following a breast reduction surgery (32). Several studies revealed predisposing con-
ditions or risk factors for the occurrence of Cutibacterium infection but no specific
characteristic correlates for C. avidum infection. In the first case, the only predisposing
factor was surgery (165).

TABLE 5 Clinical characteristics of the main C. avidum infections described in the literaturea
Presence of Signs and symptoms
Reference Patient no. Age (yr) Sex C. avidum samples Diagnosis fever (°C) of infected site/joint BMI
72 1 to 12 Median, 61 7 F, 5 M Synovial fluid, tissue 10 hip and 2 shoulder 4 of 12 Pain (n ⫽ 12), wound Median, 34
(45–81) cultures 3 to 8, and prosthesis infection patients secretion or sinus (27.9–40.6)
sonication fluid tract (n ⫽ 6)
culture
178 13 85 F 3 sets of blood culture Infective endocarditis Yes (38.3) Aortic stenosis NA
bottles
14 14 87 F Synovial fluid and 5 Abscess formation No Redness, intense pain 38
tissue biopsy anterior to the left
specimens hip

14 15 75 M Superficial wound Hip prosthesis infection NA Loosening of the cup 37
culture and 6 biopsy
specimens plus
synovial fluid
14 15 75 M Synovial fluid and 7 Hip prosthesis infection NA Increase of pain and 37

tissue biopsy and abscess discomfort
specimens formation below the
acetabulum floor
17 16 37 F 1 purulent fluid sample Liponecrosis, breast No No

abscess
15 16 36 F 1 purulent fluid sample Liponecrosis, breast No No inflammatory

abscess syndrome
175 17 26 M Drainage fluid Skin inguinal abscess No No NA
167 17 46 F 1 purulent fluid sample Abdominal No Pain 30.1

parietoplasty
16 18 56 F 2 deep swabs Skin necrosis No Generalized induration NA

and erythema
16 18 56 F Repeat swabs Wound deterioration NA Generalized induration NA
and erythema
185 19 78 F Joint purulent fluid and Hip septic arthritis No Pain 30.5
blood culture bottle
32 20 40 F Pus samples Breast abscess No Induration and NA

erythematous
breasts
168 21 70 M Abscess pus Perianal abscess NA Painful left buttock NA
mass and ruptured
epidermal cyst with
abscess formation
173 22 79 M Splenic abscess Cardiac infarction Yes (38.5) Pain NA
184b 23 67 M Different bone samples Sacroiliitis, psoas Yes (38.4) Chronic lumbago and NA
and biopsy abscess, and sciatica
specimens plus osteomyelitis
blood
169c 24 61 M Collection of fluid Splenic abscess Yes (39.1) No NA
under sonographic
guidance
aM, male; F, female; NA, not available; DAIR, debridement-antibiotics-irrigation-retention; WBC, white blood cell; AMC, amoxicillin-clavulanate; AMX, amoxicillin; CEF,
cephalothin; CIP, ciprofloxacin; CLI, clindamycin; MEM, meropenem; MTR, metronidazole; MXF, moxifloxacin; RIF, rifampin; SXT, trimethoprim-sulfamethoxazole; TZP,
piperacillin-tazobactam; VAN, vancomycin.
bFor this study (surgical repair of an inguinal hernia), the resulting erythrocyte sedimentation rate (ESR) was 60 mm/h.
cFor this study (coronary artery bypass surgery), the resulting ESR was 90 mm/h.
Later, the same case was described by two teams, from Switzerland and Austria,
revealing the role of this common skin colonizer, which is well known as a resident of
the pilosebaceous follicle unit (15, 17). The strain was the only etiology of an ulceration
of the inframammary region with induration and putrid secretion. In this case, the
proximity of the axillary region may have played a role in wound contamination

TABLE 5 (Continued)
Predisposing conditions WBC concn CRP concn
and risk factors (g/liter) (mg/liter) Resistance Surgical treatment Treatment Outcome
Degenerative joint disorder NA 50 (4.6–200) None except one isolate DAIR (n ⫽ 5), one-stage CLI (n ⫽ 1), levofloxacin plus RIF Good outcome
(n ⫽ 9), trauma (n ⫽ 2), resistant to exchange (n ⫽ 2), (n ⫽ 4), ciprofloxacin plus RIF except for
head necrosis clindamycin and later two-stage exchange (n ⫽ 2), clindamycin plus RIF DAIR
(n ⫽ 1) to rifampin plus a (n ⫽ 4), only medical (n ⫽ 2) strategy
fluoroquinolone treatment (n ⫽ 1)
Hypertension diabetes NA 86 None Delayed cardiac surgery Daptomycin plus ampicillin plus Death
mellitus ceftriaxone
Surgeries, corticoid 16.3 155 None Moore dorsolateral 3 doses of cloxacillin at 2 g i.v., Favorable
treatment, obesity approach then CLI (300 mg 3 times
per day) plus RIF (600 mg
per day) (1 mo), then
phenoxymethylpenicillin at 2 g
3 times per day (2 mo)

Surgery, obesity NA 64 None Muscle-sparing Watson- CLI at 600 mg 3 times per day (1 Failure
Jones approach wk) and then CLI at 300 mg 3
(anteriorly curved times per day (5 wk)
skin incision)
Surgery, obesity 47.2 in 180 None Revision two-stage TZP 3 times per day (1 wk) plus Favorable
synovial exchange benzylpenicillin at 1 g i.v. 3
fluid times per day (3 wk); TZP at 4
g 3 times per day plus VAN at
1.5 g twice a day and then
SXT at 800 mg 3 times daily
plus AMX at 1 g 3 times daily
Surgery 9.1 16 Clindamycin Abscess drainage i.v. AMC at 2.2 g 3 times per day Favorable
(15 days) and p.o. AMC at 750
mg 3 times per day (1 mo)
Surgery, smoker 11 6 Clindamycin Abscess drainage i.v. AMC at 2.2 g 3 times per day Favorable
(15 days) and p.o. AMC at 750
mg 3 times per day (1 mo)
Seborrheic dermatitis skin 11.1 26 None Abscess drainage i.v. CEF at 1 g 6 times daily plus Favorable
alteration MTR at 500 mg 3 times daily
(5 days) and then oral
cephalexin (7 days)
Prosthetic abdominal 13 162 None Surgical evacuation p.o. AMC at 3 g per day Favorable
surgery, overweight with abscess
drainage
Surgery 10.3 7.1 NA Necrosis debridement i.v. MEM plus p.o. MXF (2 wk) Failure
Surgery NA NA NA Further fat MXF (2 wk) and then AMC (4 wk) Favorable
debridements
Intra-articular 6.5 185 None Arthrotomy plus lavage i.v. CEF (6 g per day) plus RIF Favorable
glucocorticoid (1,800 mg per day)
treatment, overweight,
permanent humidity of
the inguinal folds
Surgery 22.5 135 None Purulent aspiration i.v. flucloxacillin plus penicillin G Favorable
and p.o. AMC (1 mo)
Cirrhotic patient 7.5 NA NA Abscess drainage Ampicillin plus cloxacillin (1 wk) Favorable
hemorrhoidectomy
Cardiac catheterization, 13.1 NA NA Laparotomy with NA NA

diabetes, hypertension abscess resection
Surgical repair of an 16.6 NA NA Bone biopsy and psoas i.v. AMC at 8.8 g per day Favorable but
inguinal hernia abscess debridement followed by CIP at 1 g per day sacroiliitis
for 6 mo sequela
Coronary artery bypass 5.6 NA None Laparotomy and spleen NA NA

surgery resection
(representing a difference from bone and joint infections [see below]). In this context,
and despite the absence of foreign-body material, the development of C. avidum led to
severe infection by virtue of its ability to adhere but also to grow and create abscesses,
as reported previously (Table 5).
Moreover, the decrease in inoculum by excellent debridement and a prolonged
course of antibiotics may play a crucial role in the successful treatment of severe breast
infections following reduction mammoplasty (16). Persistence and biofilm production
ability were suspected by Mak et al. on comparing the genomes of several Cutibacte-
rium species and were later confirmed by Wildeman et al. by specific and extended
genomic analysis of C. avidum genomes (13, 14).

Abdominal Infections
Acute abdominal infections are usually caused by Enterobacteriaceae, enterococci,
and anaerobes, such as Bacteroides fragilis (166). In 2013, Janvier et al. reported an
abdominal wall and intraperitoneal abscess by C. avidum as a complication of abdom-
inal parietoplasty (167). In this case, I can hypothesize that the microorganism was
hidden in skin folds and that infection began due to contiguity with the scar. Lastly, and
again in this case, the authors specified the presence of an abscess requiring debride-
ment surgery, which is quite often reported for the different deep infection cases
published (Table 5). A perianal abscess caused by C. avidum in a cirrhotic patient was
also reported and had a favorable outcome after surgical drainage (168).
Splenic Abscesses

C. avidum was first reported as the causative agent of a splenic abscess in 1986 (169).
This disease remains rare, involving mainly S. aureus or Enterobacteriaceae. In the 1980s,
C. acnes was also described as a potential etiologic agent (170, 171), and more recently,
a new case was reported, involving multiple splenic nodules with fever in an immu-
nocompromised patient with chronic lymphocytic leukemia (172). Despite the absence
of hyperleukocytosis, the splenic abscess was positive only for C. avidum. This species
should therefore be taken into account, especially after surgical explorations. Thus, C.
avidum can be responsible for serious deep infectious diseases, including abscesses,
most likely linked to its ability to develop inflammation and to recruit immune cells
leading locally to a cytokine storm.
A second case of a C. avidum splenic abscess was reported after a cardiac cathe-
terization complication. This disease remains uncommon but may occur in patients
with sickle hemoglobinopathies, trauma, bacteremia, splenic infarction, hematoma, or
intravenous drug use. For the second patient, the etiology of the splenic abscess was
unclear, with different potential hypotheses, including undetected C. avidum endocar-
ditis (a recent case was published [see below]) or a predisposing splenic lesion
(hematoma, infarction, or cyst) (173). In both cases, debridement surgery was the key
to success.
Skin Abscesses
Only one case of skin abscess, without any risk factors in a young immunocompe-
tent subject, has been reported, especially without previous surgery. Interestingly, a
recurrence of this abscess was noticed despite a 7-day course of ciprofloxacin treat-
ment, but the emergence of fluoroquinolone resistance in Cutibacterium spp. was
recently reported (174). A lavage drainage was performed during surgery to reduce the
inoculum, and the patient was treated for 2 weeks with antibiotics, with a favorable
clinical outcome. The inguinal localization may explain the infection, which had a
purulent aspect. Nevertheless, no obvious cause except for broken skin integrity can
explain the occurrence of such an infection (175).
Infective Endocarditis
There have been several reports of infective endocarditis (IE) due to C. avidum. In
2001, an initial case of C. avidum IE was reported for a patient with a history of aortic
valve prosthesis. Blood cultures were positive in 7.8 days, with a clinical diagnosis of
vegetation regarding the transesophageal echocardiography (TEE) investigation and an
abscess of the aortic ring (176). A second IE case report revealed an elevated C-reactive
protein (CRP) level (up to 200 mg/liter) with a positive C. avidum blood culture
(unknown time to positivity) in a 70-year-old woman who had undergone biological
aortic valve replacement surgery 3 months previously. TEE showed multiple vegeta-
tions on the biological aortic valve and a perivascular abscess (177). The last case of C.
avidum IE was recently reported by a Spanish group. An elderly woman with a
permanent pacemaker implantation underwent bioprosthetic aortic valve replacement
due to aortic stenosis. Eighteen months later, a perivalvular abscess was diagnosed, and
three sets of blood culture bottles were positive. Following a complicated postopera-

tive course, the patient died 48 h after surgery, highlighting the fact that C. avidum can
lead to host tissue damage (178).
Therefore, C. avidum IE should be suspected in patients with a history of prosthetic
heart valves, even if unspecific symptoms are present. Even though the incidence of
Cutibacterium IE remains rare (approximately 1%), it should not be disregarded, as it can
be linked to abscess formation, as noticed for breast infections. Lastly, although
Cutibacterium species, which are constituents of the skin microbiota, are too often
considered nonvirulent microorganisms, the overall mortality rate is relatively high, at
almost 20 to 25% (179). In the three recent C. granulosum IE reports, it was also
systematically noticed that extensive decalcification, abscess formation, and valvular
destruction are in fact trends of Cutibacterium species IE (180, 181).

More recently, the same observation was reported for 24 cases of C. acnes IE. The
authors highlighted that C. acnes remains an underrecognized cause of prosthetic valve
IE. Among the 24 cases, almost exclusively prosthetic valves were affected. Surgery was
required in most cases, and only males with advanced heart disease were found to have
IE due to C. acnes. Moreover, it was almost systematically associated with high rates of
local invasion and embolic complications, as reported for C. avidum and C. granulosum
(182). Therefore, as previously written, and even though this Cutibacterium infectious
episode remains rare, it constitutes a serious medical entity, and valve DNA sequencing
(with careful sample handling) identifies Cutibacterium spp. as the cause of the aortic
endocarditis in most cases, whereas otherwise these cases would have been catego-
rized as culture negative. Routine use of extended-duration incubation of blood
cultures should be recommended (i.e., more than 7 days) when Cutibacterium IE is
suspected and can be expected to identify C. avidum in patients whose IE would
otherwise have been misdiagnosed.
The increasing reports of infections associated with this bacterium may be explained
by improvements in diagnostic microbiology techniques and by the increased number
of foreign-body devices implanted in patients.
Bone and Joint Infections

Due to its conventional niche, C. avidum is an unusual player in bone and joint
infections, particularly in hip prostheses. Nevertheless, in the event of a high BMI with
skin folds, the growth and development of C. avidum near the surgical wound can lead
to contiguous inoculation of this uncommon bacterium. A potential association with
the anterior surgical approach, which has been performed in recent years, has been
mentioned (183). This observation was also reported by Achermann et al., but without
any proof (72).
Estoppey et al. reported a case of C. avidum sacroiliitis and osteomyelitis, highlight-
ing the fact that this species may act as a potential etiology of septic arthritis and
osteomyelitis, although C. acnes remains the leader of such anaerobic infections (184).
In this case, the bacterium was recovered not only from several bone samples and
biopsy specimens but also from blood. This severe C. avidum infection did not reveal
any specific risk factors (prosthesis, iatrogenic immunosuppression, or surgery), as
noted by Brook and Frazier (165).
In 2008, Million et al. reported a case of septic arthritis of the hip with C. avidum
bacteremia following intra-articular treatment for hip osteoarthritis (185). Although C.
avidum is considered a contaminant or colonizer, note that in these last two cases the
deep infection was accompanied by bacteremia. Even though the potential pathogenic
properties of C. avidum (see the virulence factor section) are not fully understood, this
microorganism appears to be capable of causing severe deep infections, even with a
delay in disease expression as often described for Cutibacterium (6, 51, 75, 186). More
recently, Wildeman et al. reported C. avidum as a new etiological agent of prosthetic hip
joint infection (14).
Several cases in the literature have revealed the ability of this microorganism to
cause clinically relevant infections, including abscess formation (72). In the cases
described above, the BMI was circa 38 and 37. This raises the following initial question:

how can we be sure about the antisepsis process with chlorhexidine? Another hypoth-
esis may consider the impact of the anteriorly curved skin incision approach. Indeed,
this strategy can be considered a risk factor, since the incision is closer to the groin than
in other approaches. Nevertheless, Ilchmann et al. demonstrated no increased risk
factors for the direct anterior approach performed by two highly experienced surgeons
(187). Moreover, perioperative antibiotic prophylaxis should be taken into account in
the prevention of infections due to Cutibacterium spp. No consensus guidelines exist at
this time, especially for total shoulder arthroplasties, for which there have been a
number of Cutibacterium-related publications in recent years (188–190).
Finally, what about monotherapy treatment and the pharmacokinetics/pharmaco-
dynamics (PK/PD) approach (mode of administration, dose, and frequency but also

absorption, distribution, metabolism, and elimination) to be sure about penetration
into bone tissue and the ability to eradicate bacteria growing in a biofilm? Indeed,
genomic analysis revealed specific sequences compared to those of other Cutibacte-
rium species and the ability to produce a biofilm structure according to the description
of a gene cluster encoding exopolysaccharide production, suspected to be involved in
strong adherence, as previously described for C. acnes (74). Interestingly, and recently
in my experience, it has been observed that C. avidum can be involved in shoulder
prosthesis infection, probably leading to a chronic status (years) according to its ability
to resist antibiotics and to evade the immune system. Henceforth, C. avidum should
clearly be considered a virulent pathogen causing, in particular, hip prosthesis infection
(after hip arthroplasty surgery using an anterior surgical approach) in obese patients
(risk factor), with pain and wound secretion (183). Treatment combining one-stage
exchange with prolonged antibiotic therapy allows a favorable outcome. In contrast, for
all patients treated with either an initial debridement-antibiotics-irrigation-retention
(DAIR) procedure or antibiotic alone, treatment failed, necessitating a two-stage revi-
sion of the prosthesis, with a final good clinical outcome (72).
Prostate Infections
Though Cutibacterium spp. have clearly been reported and associated with inflam-
mation, their role in several diseases besides acne (see below), such as prostate
carcinoma or sarcoidosis, has also been highlighted in various studies. C. acnes was the
organism most frequently recovered from radical prostatectomy specimens, reaching
rates of 23% to 60% in cases of prostate carcinoma (191). Nevertheless, C. avidum was
represented in 7% of prostate tissue samples (192). From an experimental point of view,
these species may contribute to prostate carcinogenesis by locally inducing high-level
inflammation (193). Indeed, using the HACAT or RWPE1 prostate epithelial cell line,
cytokine secretion was upregulated using the TLR2–NF-␬B signaling pathway. Cocul-
ture with C. acnes also revealed different pathways (STAT3, COX2-prostaglandin, and
plasminogen-matrix metalloproteinase pathways) (194).
According to updated knowledge with the next-generation approach, it would be
interesting to investigate the roles of the various C. avidum clusters recently described
for other infections, such as bone and joint infections (14). Indeed, one may hypoth-
esize that different clusters of C. avidum are associated with different localizations
and/or infections. More studies are required to better characterize the numerous C.
avidum strains in order to reveal any specificities or a possible tropism as described for
C. acnes. No overlap was observed between C. acnes lineages involved in acne (CC18,
CC3, and CC28) and strains isolated from opportunistic infections belonging to other
clusters (CC53 and CC60) (13, 192).
Acne Lesions
Compared to the significant amount of literature on the involvement of C. acnes in acne,
C. avidum remains a rare or exceptional player in this skin disease. Its presence has
nevertheless been reported in combination with C. acnes and C. granulosum. This species
should therefore not be forgotten during discussions and should be included as a potential
etiological factor for acne (195). More recently, an investigation of bacterial colonization and

keratinocyte proliferation in hair follicles from control (n ⫽ 117) and acne (n ⫽ 26) skin
patients revealed that none of the samples were positive for C. avidum (30). In fact, as
previously suggested, C. avidum is more probably detected in the wet body areas (196). Its
ecological niche may explain the behavior observed when skin innate immunity is explored
in vitro, which is different from those of different C. acnes types, such as phylotypes III, II, IC,
IA1, and IB. Therefore, C. avidum has much less inflammatory potential, inducing only tissue
inhibitor metallopeptidase 2 (TIMP-2) and matrix metallopeptidase 13 (MMP-13) expression,
according to a recent study by my group (109). Importantly, and unlike the C. acnes strains,
C. avidum decreased TLR2 expression, downregulated anti-inflammatory transforming
growth factor beta (TGF-␤) expression, and had no effect on proinflammatory tumor
necrosis factor alpha (TNF-␣) expression.

Caries-Like Lesions
Because the pH in the mouth is near 6, the best pH for C. avidum growth, plaque pH
can lead to C. avidum development, especially after a sugar challenge or exposure.
Cell-to-cell interaction with various bacterial species, such as Streptococcus mutans, can
lead to caries. Consequently, repeated exposure to high-starch and carbohydrate diets
gives rise to metabolically active plaque, leading to distinct lesions. The fermentation
of sucrose, sorbitol, or xylitol by C. avidum can result in the formation of caries-like
lesions in enamel (197). Although plaque acidogenicity does not necessarily reflect
cariogenicity, food components, and especially sugar retention, may account for in-
creased cariogenicity at plaque retention sites in the mouth.
ANTIMICROBIAL SUSCEPTIBILITY OF C. AVIDUM AND RELATED SPECIES

In 1976, Hoeffler et al. reported the first major study focusing on the antimicrobial
susceptibility of Cutibacterium avidum and related species, especially C. acnes, C.
granulosum, and Corynebacterium minutissimum (198). Seventy-three strains were
tested against 32 antimicrobial components from different families. No acquired resis-
tance was observed, regardless of the species, and the lowest MICs were reported for
␤-lactams, rifampin, erythromycin, clindamycin, and minocycline (198, 199). Neverthe-
less, among the aminoglycosides, kanamycin was the least active, and most C. avidum
strains were (naturally) resistant to metronidazole and colistin.
One point should be highlighted: the European guidelines for antibiotic suscepti-
bility testing do not provide any rules or critical breakpoints for interpreting anaerobe
testing in the latest version (http://www.eucast.org/clinical_breakpoints/). Conversely,
the CLSI guidelines propose various critical breakpoints, as summarized in Table 6.
However, as early as 1989, erythromycin-resistant Cutibacterium species in antibiotic-
treated acne patients were investigated, and a potential association with therapeutic failure
was discussed (200). Among the strains isolated from the 52 subjects enrolled in the study,
63 strains were resistant to erythromycin, including five C. avidum strains (8% versus
66.6% for C. acnes and 25.4% for C. granulosum). At that time, no molecular investiga-
tion was conducted to characterize the genetic support of the resistance compared to
that of C. acnes. Later, numerous studies, especially on C. acnes, revealed that eryth-
romycin resistance is usually due to methylation of the 23S ribosomal subunit (muta-
tion in the 23S rRNA gene, usually at position 2058 or 2059), with a high- or low-level
resistance pattern.
In contrast, and to the best of my knowledge, among C. avidum isolates recovered
in different contexts (from skin, blood, tissue, prostate, or abdomen) (201), macrolide
resistance due to point mutations remains rare compared to that in C. acnes clinical
strains, and no ermX gene has yet been described (51). Interestingly, the strains from
the various published case reports (Table 5) were fully susceptible to antibiotics, except
for clindamycin in two cases (15, 17) with high-level resistance to clindamycin, as
previously reported (51, 202). The same observation was reported for clinical strains
isolated during bone and joint infections (203). Consequently, one recommendation is
that clindamycin should be tested when used in clinical treatment to avoid treatment
failure.

TABLE 6 CLSI anaerobe breakpoints of interest for treatment of Cutibacterium avidum

infections
MIC interpretative breakpoint
(mg/liter)
Drug class Antimicrobial agent Susceptible Intermediate Resistant
Penicillins Penicillin ⱕ0.5 1 ⱖ2
Ampicillin ⱕ0.5 1 ⱖ2
␤-Lactam–␤-lactam inhibitor Amoxicillin-clavulanate ⱕ4/2 8/4 ⱖ16/8

combinations Piperacillin-tazobactam ⱕ32/4 64/4 ⱖ128/4
Cephems Cefotetan ⱕ16 32 ⱖ64

Cefoxitin ⱕ16 32 ⱖ64

Ceftriaxone ⱕ16 32 ⱖ64
Cefotaxime ⱕ16 32 ⱖ64
Carbapenems Ertapenem ⱕ4 8 ⱖ16

Imipenem ⱕ4 8 ⱖ16
Meropenem ⱕ4 8 ⱖ16
Tetracyclines Tetracycline ⱕ4 8 ⱖ16
Fluoroquinolones Moxifloxacin ⱕ2 4 ⱖ8
Lincosamides Clindamycin ⱕ2 4 ⱖ8
Phenicols Chloramphenicol ⱕ8 16 ⱖ32
To limit the emergence of antibiotic resistance due to resistant mutant selection,

various alternative treatments have been proposed. Whether or not tetracyclines are
still used, this bacteriostatic antibiotic family remains active against Cutibacterium
species (less than 10 to 15% resistance) (204). Even in the acne field, low-dose
tetracyclines are still used for their anti-inflammatory properties (205), but the C.
avidum clinical strains (from various origins) remain susceptible to this family. Based on
experience with Cutibacterium species over 8 years in the microbiology laboratory,
more than 1,900 Cutibacterium strains have been isolated, and all C. avidum strains were
fully susceptible to all the antibiotics tested, with the exception of one macrolide-
resistant strain recovered from a healthy skin patient. Otherwise, tetracycline can also
have a slight effect on the extracellular lipase produced by C. avidum. Lipase production
is twice as sensitive to tetracycline as that of total cellular and extracellular proteins
(206).
Another component, benzoyl peroxide, seems to be quite active and bactericidal,
without resistance (207). This compound displayed excellent activity in vitro when
tested on C. avidum PF77. This powerful oxidizing agent is active because it releases
highly reactive oxygen species that can disrupt the bacterial cell wall (132). Recently, to
optimize the surgical process and reduce the local Cutibacterium sp. burden, especially
in cases of total shoulder arthroplasty, local antibiotic treatment and combination with
benzoyl peroxide demonstrated an excellent effect in reducing the risk of delayed and
biofilm-associated Cutibacterium infections, which are frequent in males during the
posterolateral surgery approach (208, 209).
ANTIBIOTIC TREATMENT
To date, there is no gold standard in terms of treatment of Cutibacterium infections
(85). Various strategies have been developed, though the therapeutic approaches
proposed are usually based on tradition and guided by personal and expert experience.
Although daily routine practices vary, some consensus can be reported in light of the
knowledge base for C. acnes infections. Recently, Shah et al. proposed different
therapeutic schemes for cases of prosthetic joint infection (i.e., device-related infection),
as follows: first-line therapy, penicillin at 20 to 24 million units given i.v. daily, either as
a continuous infusion or in 6 divided doses, or ceftriaxone at 2 g given i.v. daily; and

second-line therapy, clindamycin at 600 to 900 mg given i.v. every 8 h or vancomycin

at 15 mg/kg of body weight given i.v. every 12 h. The recommended duration of
therapy is 4 to 6 weeks (210). Interestingly, different groups of experts usually recom-
mend prolonged treatment with different antibiotics, in particular combinations in-
cluding rifampin (75, 211), with excellent bone distribution and diffusion through
biofilms, along with a fluoroquinolone, amoxicillin, or clindamycin. In case of abscesses,
initial surgery with excellent debridement should be performed, and a course of 2 to 4
weeks seems to be linked to a favorable outcome. Nevertheless, as far as the literature
(Table 5) is concerned, there again appears to be no consensus. Therefore, like the case
for C. acnes infection, further studies are clearly needed to optimize C. avidum infection
treatment, and even though this microorganism remains rare, it is most likely under-

estimated (85).
Indeed, with respect to device-related infection or abscess infection (high inoculum),
new in vitro and especially in vivo (animal models) investigations are required to allow
proposal of an optimal regimen and its duration. Recently, Achermann et al. reported
on 12 hip prosthetic infection treatments. Following surgical debridement or prosthesis
exchange, all patients were treated intravenously for approximately 2 weeks with a
␤-lactam (or vancomycin in the event of allergies), followed by oral therapy including
either clindamycin, clindamycin plus rifampin, or a fluoroquinolone plus rifampin. Oral
treatment and its duration were determined according to MIC values and the surgical
approach adopted, respectively. Thus, antibiotics were given for 3 months for the DAIR
approach as well as one-stage exchange and for 6 weeks for two-stage exchange (72).
Zeller et al. reported a similar approach with either clindamycin or cefazolin combined with
rifampin administered for 12 weeks after a one-stage exchange for 11 chronic C. avidum hip
infections (183). According to these experiences, the DAIR approach seems to be linked to
poor outcomes, whereas combining a one-stage surgical approach with prolonged anti-
biotic treatment allows favorable outcomes.
In my opinion, and based on my experience, rifampin, amoxicillin, ceftriaxone,
levofloxacin, and maybe moxifloxacin, as well as daptomycin or linezolid (in cases of
mixed infection with multidrug-resistant coagulase-negative staphylococci), remain the
drugs of choice, in combination for 3 months, to eradicate Cutibacterium biofilm- or
device-related infection.
In any event, based on my experience, although C. avidum is clearly encountered
less frequently in daily routine, these strains are more susceptible to all antibiotics than
potentially occurring C. acnes strains resistant to fluoroquinolones, macrolides, tetra-
cyclines, co-trimoxazole, and rifampin (51).
CONCLUSIONS
This review focused on C. avidum infections, which are an emergent medical issue
due to the increased use of implants, especially orthopedic implants but also breast
implants, and depending on the ecological niche of wet areas of human skin. Numer-
ous clinical cases have been detailed reporting different features. This underrecognized
species, and most likely its various subtypes suggested by molecular identification
methods, may cause more infections in the future and will require closer attention,
especially after surgery or in the presence of a skin abscess.
Over the past few years, focus has been placed on improved diagnostic procedures
and accurate identification methods (molecular or spectral). With the genomic era, new
insights will be investigated in terms of pathogenesis and host-bacterium interactions,
which might explain how this bacterium was able to be used as an immunomodulator
several decades ago.
To better understand C. avidum physiopathogenesis and virulence, its bacterial
adherence, invasiveness, and biofilm formation ability, along with phage and mobile
genetic element (e.g., plasmid) acquisition and gene transfer, should be investigated
further to determine their roles.

ACKNOWLEDGMENTS
This work was supported by an internal grant.
I am grateful to Stanimir Kambarev for his technical help with the genome figure. I also
thank the bioMérieux R&D department for their precious and valuable help in recovering
spectra, masses, and intensities regarding Cutibacterium species identification using Vitek
MS technology. I thank my microbiologist colleagues, especially Pascale Bémer from the
Nantes University hospital for her help in the diagnosis of bone and joint infections, in
particular low-grade Cutibacterium infections, and Marie-Emmanuelle Juvin for helping with
the microbiological data from the database. Finally, I thank Amir Khammari, Brigitte Dréno,
and Andrej Trampuz for their constant support and helpful discussions.

REFERENCES
1. Orla-Jensen S. 1909. Die hauptlinien der natürlichen bakterienssystems. the cause of severe breast infection following reduction mammo-
Zentralbl Bakteriol Parasitenkd Infektionskrankh Hyg Abt 2:305–346. plasty. ANZ J Surg 78:318 –319. https://doi.org/10.1111/j.1445-2197
2. Whitman WB, Goodfellow M, Kämpfer P, Busse H-J, Trujillo ME, .2008.04450.x.
Ludwig W, Suzuki K-I, Parte A (ed). 2012. The Actinobacteria. 17. Kritikos A, Pagin M, Borens O, Voide C, Orasch C. 2015. Identification of
Bergey’s manual of systematic bacteriology, 2nd ed, vol 5. Springer- Propionibacterium avidum from a breast abscess: an overlooked etiol-
Verlag, New York, NY. ogy of clinically significant infections. New Microbes New Infect 4:9 –10.
3. Patrick S, McDowell A. 2012. Genus I. Propionibacterium, p 1138 –1156. https://doi.org/10.1016/j.nmni.2014.12.001.
In Whitman WB, Goodfellow M, K̈ampfer P, Busse H-J, Trujillo ME, 18. McBride WH, Dawes J, Dunbar N, Ghaffar A, Woodruff MF. 1975. A
Ludwig W, Suzuki K-I, Parte A (ed), Bergey’s manual of systematic comparative study of anaerobic coryneforms. Attempts to correlate
bacteriology, 2nd ed. Springer-Verlag, New York, NY. their anti-tumour activity with their serological properties and ability to
4. Cummins CS, Johnson JL. 1974. Corynebacterium parvum: a synonym stimulate the lymphoreticular system. Immunology 28:49 –58.
for Propionibacterium acnes? J Gen Microbiol 80:433– 442. https://doi 19. Azuma I, Taniyama T, Sugimura K, Aladin A, Yamamura Y. 1976. Mito-
.org/10.1099/00221287-80-2-433. genic activity of the cell walls of mycobacteria, nocardia, corynebacte-
5. Gilchrist T. 1900. A bacteriological and microscopical study of over ria and anaerobic coryneforms. Jpn J Microbiol 20:263–271. https://doi
three hundred vesicular and pustular lesions of the skin, with a .org/10.1111/j.1348-0421.1976.tb00987.x.
research upon the etiology of Acne vulgaris. Johns Hopkins Hosp 20. Siwicki AK, Krzyzanowski J, Bartoszcze M, Mizak Z, Paluch S, Szmigielski
Rep 9:409 – 430. S, Jeljaszewicz J, Pulverer G. 1998. Adjuvant properties of killed Propi-
6. Achermann Y, Goldstein EJC, Coenye T, Shirtliff ME. 2014. Propionibac- onibacterium avidum KP-40 in vaccination of dogs against canine par-
terium acnes: from commensal to opportunistic biofilm-associated im- vovirosis. Dtsch Tierarztl Wochenschr 105:186 –190.
plant pathogen. Clin Microbiol Rev 27:419 – 440. https://doi.org/10 21. Ko HL, Winkler C, Beuth J, Tunggal L, Buss G, Jeljaszewicz J, Klippel KF,
.1128/CMR.00092-13. Pulverer G. 1995. Influence of Propionibacterium avidum KP-40 on the
7. Douglas HC, Gunter SE. 1946. The taxonomic position of Corynebacte- proliferation, maturation, emigration and activity of thymocytes and
rium acnes. J Bacteriol 52:15–23. monocytes. Zentralbl Bakteriol Int J Med Microbiol 282:86 –91. https://
8. Scholz CFP, Kilian M. 2016. The natural history of cutaneous propi- doi.org/10.1016/S0934-8840(11)80800-5.
onibacteria, and reclassification of selected species within the genus 22. Gross CS, Ferguson DA, Cummins CS. 1978. Electrophoretic protein
Propionibacterium to the proposed novel genera Acidipropionibacte- patterns and enzyme mobilities in anaerobic coryneforms. Appl Envi-
rium gen. nov., Cutibacterium gen. nov. and Pseudopropionibacterium ron Microbiol 35:1102–1108.
gen. nov. Int J Syst Evol Microbiol 66:4422– 4432. https://doi.org/10 23. Cummins CS, Linn DM. 1977. Reticulostimulating properties of killed
.1099/ijsem.0.001367. vaccines of anaerobic coryneforms and other organisms. J Natl Cancer
9. Aubin GG, Bémer P, Kambarev S, Patel NB, Lemenand O, Caillon J, Inst 59:1697–1708. https://doi.org/10.1093/jnci/59.6.1697.
Lawson PA, Corvec S. 2016. Propionibacterium namnetense sp. nov., 24. Ko HL, Heczko PB, Pulverer G. 1978. Differential susceptibility of
isolated from a human bone infection. Int J Syst Evol Microbiol 66: Propionibacterium acnes, P. granulosum and P. avidum to free fatty
3393–3399. https://doi.org/10.1099/ijsem.0.001204. acids. J Investig Dermatol 71:363–365. https://doi.org/10.1111/1523
10. Butler-Wu SM, Sengupta DJ, Kittichotirat W, Matsen FA, Bumgarner RE. -1747.ep12556296.
2011. Genome sequence of a novel species, Propionibacterium humer- 25. Barnard E, Shi B, Kang D, Craft N, Li H. 2016. The balance of metag-
usii. J Bacteriol 193:3678. https://doi.org/10.1128/JB.05036-11. enomic elements shapes the skin microbiome in acne and health. Sci
11. McGinley KJ, Webster GF, Leyden JJ. 1978. Regional variations of cuta- Rep 6:39491. https://doi.org/10.1038/srep39491.
neous propionibacteria. Appl Environ Microbiol 35:62– 66. 26. Grice EA, Kong HH, Conlan S, Deming CB, Davis J, Young AC, NISC
12. Globa AG, Alekseev II, Arzumanian VG, Zaborova VA, Guridova AA. Comparative Sequencing Program, Bouffard GG, Blakesley RW, Murray
2014. The use of real time PCR for quantitative determination of some PR, Green ED, Turner ML, Segre JA. 2009. Topographical and temporal
propionic bacteria residing on human skin. Biomed Khim 60:372–377. diversity of the human skin microbiome. Science 324:1190 –1192.
https://doi.org/10.18097/pbmc20146003372. https://doi.org/10.1126/science.1171700.
13. Mak TN, Schmid M, Brzuszkiewicz E, Zeng G, Meyer R, Sfanos KS, 27. Lomholt HB, Kilian M. 2014. Clonality and anatomic distribution on the
Brinkmann V, Meyer TF, Brüggemann H. 2013. Comparative genomics skin of antibiotic resistant and sensitive Propionibacterium acnes. Acta
reveals distinct host-interacting traits of three major human-associated Derm Venereol 94:534 –538. https://doi.org/10.2340/00015555-1794.
propionibacteria. BMC Genomics 14:640. https://doi.org/10.1186/1471 28. Nordstrom NK, Noble WC. 1984. Colonization of the axilla by Propi-
-2164-14-640. onibacterium avidum in relation to age. Appl Environ Microbiol 47:
14. Wildeman P, Brüggemann H, Scholz CFP, Leimbach A, Söderquist B. 1360 –1362.
2016. Propionibacterium avidum as an etiological agent of prosthetic 29. Benediktsdóttir E, Hambraeus A. 1982. Dispersal of non-sporeforming
hip joint infection. PLoS One 11:e0158164. https://doi.org/10.1371/ anaerobic bacteria from the skin. J Hyg (Lond) 88:487–500. https://doi
journal.pone.0158164. .org/10.1017/S0022172400070340.
15. di Summa PG, Yvon A, Larcher L, Raffoul W, Koch N. 2015. Propionibac- 30. Jahns AC, Eilers H, Ganceviciene R, Alexeyev OA. 2015. Propionibacte-
terium avidum infection following breast reduction: high morbidity rium species and follicular keratinocyte activation in acneic and normal
from a low-virulence pathogen. J Surg Case Rep 2015:rjv002. https:// skin. Br J Dermatol 172:981–987. https://doi.org/10.1111/bjd.13436.
doi.org/10.1093/jscr/rjv002. 31. Nielsen PA. 1983. Role of reduced sulfur compounds in nutrition of
16. Levin B, Foo H, Lee A, Gottlieb T. 2008. Propionibacterium avidum as Propionibacterium acnes. J Clin Microbiol 17:276 –279.

32. Panagea S, Corkill JE, Hershman MJ, Parry CM. 2005. Breast abscess and software for calculating average nucleotide identity. Int J Syst Evol
caused by Propionibacterium avidum following breast reduction Microbiol 66:1100 –1103. https://doi.org/10.1099/ijsem.0.000760.
surgery: case report and review of the literature. J Infect 51:e253– e255. 49. Dahms PA, Martin AL, Ganz HH, Eisen JA, Coil DA. 2017. Draft genome
https://doi.org/10.1016/j.jinf.2005.04.005. sequence of Propionibacterium avidum strain UCD-PD2 isolated from a
33. Argañaraz-Martínez E, Babot JD, Apella MC, Perez Chaia A. 2013. Phys- feline anal sac. Genome Announc 5:e00034-17. https://doi.org/10.1128/
iological and functional characteristics of Propionibacterium strains of genomeA.00034-17.
the poultry microbiota and relevance for the development of probiotic 50. Sampedro MF, Piper KE, McDowell A, Patrick S, Mandrekar JN, Rouse
products. Anaerobe 23:27–37. https://doi.org/10.1016/j.anaerobe.2013 MS, Steckelberg JM, Patel R. 2009. Species of Propionibacterium and
.08.001. Propionibacterium acnes phylotypes associated with orthopedic im-
34. McDowell A, Valanne S, Ramage G, Tunney MM, Glenn JV, McLorinan plants. Diagn Microbiol Infect Dis 64:138 –145. https://doi.org/10.1016/
GC, Bhatia A, Maisonneuve J-F, Lodes M, Persing DH, Patrick S. 2005. j.diagmicrobio.2009.01.024.
Propionibacterium acnes types I and II represent phylogenetically dis- 51. Aubin GG, Portillo ME, Trampuz A, Corvec S. 2014. Propionibacterium
tinct groups. J Clin Microbiol 43:326 –334. https://doi.org/10.1128/JCM acnes, an emerging pathogen: from acne to implant-infections, from
.43.1.326-334.2005. phylotype to resistance. Med Mal Infect 44:241–250. https://doi.org/10
35. Barnard E, Nagy I, Hunyadkürti J, Patrick S, McDowell A. 2015. Multiplex .1016/j.medmal.2014.02.004.

touchdown PCR for rapid typing of the opportunistic pathogen Propi- 52. Aubin GG, Baud’huin M, Lavigne J-P, Brion R, Gouin F, Lepelletier D,
onibacterium acnes. J Clin Microbiol 53:1149 –1155. https://doi.org/10 Jacqueline C, Heymann D, Asehnoune K, Corvec S. 2017. Interaction of
.1128/JCM.02460-14. Cutibacterium (formerly Propionibacterium) acnes with bone cells: a step
36. McDowell A, Gao A, Barnard E, Fink C, Murray PI, Dowson CG, Nagy I, toward understanding bone and joint infection development. Sci Rep
Lambert PA, Patrick S. 2011. A novel multilocus sequence typing 7:42918. https://doi.org/10.1038/srep42918.
scheme for the opportunistic pathogen Propionibacterium acnes and 53. Fujimura S, Ko HL, Pulverer G, Jeljaszewicz J. 1982. Hemolysin of
characterization of type I cell surface-associated antigens. Microbiology Propionibacterium avidum. Zentralbl Bakteriol Mikrobiol Hyg A 252:
157:1990 –2003. https://doi.org/10.1099/mic.0.049676-0. 108 –115.
37. Lomholt HB, Kilian M. 2010. Population genetic analysis of Propi- 54. Ko HL, Pulverer G, Jeljaszewicz J. 1978. Propionicins, bacteriocins pro-
onibacterium acnes identifies a subpopulation and epidemic clones duced by Propionibacterium avidum. Zentralbl Bakteriol Orig A 241:
associated with acne. PLoS One 5:e12277. https://doi.org/10.1371/ 325–328.
journal.pone.0012277. 55. Nakamura M, Kametani I, Higaki S, Yamagishi T. 2003. Identification of
38. Scholz CFP, Jensen A, Lomholt HB, Brüggemann H, Kilian M. 2014. A Propionibacterium acnes by polymerase chain reaction for amplification
novel high-resolution single locus sequence typing scheme for mixed of 16S ribosomal RNA and lipase genes. Anaerobe 9:5–10. https://doi
populations of Propionibacterium acnes in vivo. PLoS One 9:e104199. .org/10.1016/S1075-9964(03)00061-1.
https://doi.org/10.1371/journal.pone.0104199. 56. Edwards HD, Anderson RC, Taylor TM, Miller RK, Hardin MD, Nisbet DJ,
39. Dekio I, Culak R, Misra R, Gaulton T, Fang M, Sakamoto M, Ohkuma Krueger NA, Smith SB. 2013. Interactions between oil substrates and
M, Oshima K, Hattori M, Klenk H-P, Rajendram D, Gharbia SE, Shah glucose on pure cultures of ruminal lipase-producing bacteria. Lipids
HN. 2015. Dissecting the taxonomic heterogeneity within Propi- 48:749 –755. https://doi.org/10.1007/s11745-013-3793-3.
onibacterium acnes: proposal for Propionibacterium acnes subsp. 57. Higaki S, Morohashi M, Yamagishi T. 2003. Anti-lipase activity of
acnes subsp. nov. and Propionibacterium acnes subsp. elongatum Unsei-in against Propionibacterium avidum in the human axilla. Int J
subsp. nov. Int J Syst Evol Microbiol 65:4776 – 4787. https://doi.org/ Antimicrob Agents 21:597–599. https://doi.org/10.1016/S0924
10.1099/ijsem.0.000648. -8579(03)00089-X.
40. McDowell A, Barnard E, Liu J, Li H, Patrick S. 2016. Proposal to reclassify 58. Higaki S, Nakamura M, Kitagawa T, Morohashi M, Yamagishi T. 2001.
Propionibacterium acnes type I as Propionibacterium acnes subsp. acnes Effect of lipase activities of Propionibacterium granulosum and Propi-
subsp. nov. and Propionibacterium acnes type II as Propionibacterium onibacterium acnes. Drugs Exp Clin Res 27:161–164.
acnes subsp. defendens subsp. nov. Int J Syst Evol Microbiol 66: 59. d’Ersu J, Aubin GG, Mercier P, Nicollet P, Bémer P, Corvec S. 2016.
5358 –5365. https://doi.org/10.1099/ijsem.0.001521. Characterization of Staphylococcus caprae clinical isolates involved in
41. Johnson JL, Cummins CS. 1972. Cell wall composition and deoxyribo- human bone and joint infections, compared with goat mastitis isolates.
nucleic acid similarities among the anaerobic coryneforms, classical J Clin Microbiol 54:106 –113. https://doi.org/10.1128/JCM.01696-15.
propionibacteria, and strains of Arachnia propionica. J Bacteriol 109: 60. Hu C, Xiong N, Zhang Y, Rayner S, Chen S. 2012. Functional character-
1047–1066. ization of lipase in the pathogenesis of Staphylococcus aureus. Biochem
42. Mak TN, Fischer N, Laube B, Brinkmann V, Metruccio MME, Sfanos KS, Biophys Res Commun 419:617– 620. https://doi.org/10.1016/j.bbrc
Mollenkopf H-J, Meyer TF, Brüggemann H. 2012. Propionibacterium .2012.02.057.
acnes host cell tropism contributes to vimentin-mediated invasion and 61. Mullen T, Markey K, Murphy P, McClean S, Callaghan M. 2007. Role
induction of inflammation. Cell Microbiol 14:1720 –1733. https://doi of lipase in Burkholderia cepacia complex (Bcc) invasion of lung
.org/10.1111/j.1462-5822.2012.01833.x. epithelial cells. Eur J Clin Microbiol Infect Dis 26:869 – 877. https://
43. Brüggemann H, Lomholt HB, Kilian M. 2012. The flexible gene pool of doi.org/10.1007/s10096-007-0385-2.
Propionibacterium acnes. Mob Genet Elem 2:145–148. https://doi.org/ 62. Tóth A, Németh T, Csonka K, Horváth P, Vágvölgyi C, Vizler C,
10.4161/mge.21204. Nosanchuk JD, Gácser A. 2014. Secreted Candida parapsilosis lipase
44. Ordögh L, Hunyadkürti J, Vörös A, Horváth B, Szucs A, Urbán E, modulates the immune response of primary human macrophages.
Kereszt A, Kondorosi E, Nagy I. 2013. Complete genome sequence of Virulence 5:555–562. https://doi.org/10.4161/viru.28509.
Propionibacterium avidum strain 44067, isolated from a human skin 63. Duran-Reynals F. 1942. Tissue permeability and the spreading factors in
abscess. Genome Announc 1:e00337-13. https://doi.org/10.1128/ infection: a contribution to the host:parasite problem. Bacteriol Rev
genomeA.00337-13. 6:197–252.
45. Barrangou R, Fremaux C, Deveau H, Richards M, Boyaval P, Moineau S, 64. Jedrzejas MJ. 2001. Pneumococcal virulence factors: structure and func-
Romero DA, Horvath P. 2007. CRISPR provides acquired resistance tion. Microbiol Mol Biol Rev 65:187–207. https://doi.org/10.1128/MMBR
against viruses in prokaryotes. Science 315:1709 –1712. https://doi.org/ .65.2.187-207.2001.
10.1126/science.1138140. 65. Höffler U. 1979. Production of hyaluronidase by propionibacteria from
46. Brüggemann H, Lomholt HB, Tettelin H, Kilian M. 2012. CRISPR/cas loci different origins. Zentralbl Bakteriol Orig A 245:123–129.
of type II Propionibacterium acnes confer immunity against acquisition 66. Nazipi S, Stødkilde-Jørgensen K, Scavenius C, Brüggemann H. 2017. The
of mobile elements present in type I P. acnes. PLoS One 7:e34171. skin bacterium Propionibacterium acnes employs two variants of hyal-
https://doi.org/10.1371/journal.pone.0034171. uronate lyase with distinct properties. Microorganisms 5:E57. https://
47. Fitz-Gibbon S, Tomida S, Chiu B-H, Nguyen L, Du C, Liu M, Elashoff D, doi.org/10.3390/microorganisms5030057.
Erfe MC, Loncaric A, Kim J, Modlin RL, Miller JF, Sodergren E, Craft N, 67. Steiner B, Romero-Steiner S, Cruce D, George R. 1997. Cloning and
Weinstock GM, Li H. 2013. Propionibacterium acnes strain populations in sequencing of the hyaluronate lyase gene from Propionibacterium
the human skin microbiome associated with acne. J Investig Dermatol acnes. Can J Microbiol 43:315–321. https://doi.org/10.1139/m97-044.
133:2152–2160. https://doi.org/10.1038/jid.2013.21. 68. Hoeffler U. 1977. Enzymatic and hemolytic properties of Propionibac-
48. Lee I, Kim YO, Park S-C, Chun J. 2016. OrthoANI: an improved algorithm terium acnes and related bacteria. J Clin Microbiol 6:555–558.

69. Valanne S, McDowell A, Ramage G, Tunney MM, Einarsson GG, O’Hagan 88. Kiedrowski MR, Crosby HA, Hernandez FJ, Malone CL, McNamara JO,
S, Wisdom GB, Fairley D, Bhatia A, Maisonneuve J-F, Lodes M, Persing Horswill AR. 2014. Staphylococcus aureus Nuc2 is a functional, surface-
DH, Patrick S. 2005. CAMP factor homologues in Propionibacterium attached extracellular nuclease. PLoS One 9:e95574. https://doi.org/10
acnes: a new protein family differentially expressed by types I and II. .1371/journal.pone.0095574.
Microbiology 151:1369 –1379. https://doi.org/10.1099/mic.0.27788-0. 89. Uchiyama S, Andreoni F, Schuepbach RA, Nizet V, Zinkernagel AS. 2012.
70. Sörensen M, Mak TN, Hurwitz R, Ogilvie LA, Mollenkopf HJ, Meyer TF, DNase Sda1 allows invasive M1T1 group A Streptococcus to prevent
Brüggemann H. 2010. Mutagenesis of Propionibacterium acnes and TLR9-dependent recognition. PLoS Pathog 8:e1002736. https://doi.org/
analysis of two CAMP factor knock-out mutants. J Microbiol Methods 10.1371/journal.ppat.1002736.
83:211–216. https://doi.org/10.1016/j.mimet.2010.09.008. 90. Lheure C, Grange PA, Ollagnier G, Morand P, Désiré N, Sayon S, Corvec
71. Corvec S, Luchetta J, Aubin GG. 2015. Is hemolysis a clinical marker of S, Raingeaud J, Marcelin A-G, Calvez V, Khammari A, Batteux F, Dréno
Propionibacterium acnes orthopedic infection or a phylogenetic B, Dupin N. 2016. TLR-2 recognizes Propionibacterium acnes CAMP
marker? Am J Orthop 44:E61–E62. factor 1 from highly inflammatory strains. PLoS One 11:e0167237.
72. Achermann Y, Liu J, Zbinden R, Zingg PO, Anagnostopoulos A, Barnard https://doi.org/10.1371/journal.pone.0167237.
E, Sutter R, Li H, McDowell A, Zinkernagel AS. 2018. Propionibacterium 91. Allaker RP, Greenman J, Osborne RH, Gowers JI. 1985. Cytotoxic activity
avidum—a virulent pathogen causing hip periprosthetic joint infection. of Propionibacterium acnes and other skin organisms. Br J Dermatol

Clin Infect Dis 66:54 – 63. https://doi.org/10.1093/cid/cix665. 113:229 –235. https://doi.org/10.1111/j.1365-2133.1985.tb02069.x.
73. Parsek MR, Singh PK. 2003. Bacterial biofilms: an emerging link to 92. Allaker RP, Greenman J, Osborne RH. 1987. The production of inflam-
disease pathogenesis. Annu Rev Microbiol 57:677–701. https://doi.org/ matory compounds by Propionibacterium acnes and other skin organ-
10.1146/annurev.micro.57.030502.090720. isms. Br J Dermatol 117:175–183. https://doi.org/10.1111/j.1365-2133
74. Achermann Y, Tran B, Kang M, Harro JM, Shirtliff ME. 2015. Immuno- .1987.tb04114.x.
proteomic identification of in vivo-produced Propionibacterium acnes 93. Scott MT. 1972. Biological effects of the adjuvant Corynebacterium
proteins in a rabbit biofilm infection model. Clin Vaccine Immunol parvum. I. Inhibition of PHA, mixed lymphocyte and GVH reactivity. Cell
22:467– 476. https://doi.org/10.1128/CVI.00760-14. Immunol 5:459 – 468.
75. Furustrand Tafin U, Corvec S, Betrisey B, Zimmerli W, Trampuz A. 2012. 94. Adlam C, Reid DE. 1977. Comparative studies on the cell wall compo-
Role of rifampin against Propionibacterium acnes biofilm in vitro and in sition of some anaerobic coryneforms of varying lympho-reticular stim-
an experimental foreign-body infection model. Antimicrob Agents Che- ulatory activity. Dev Biol Stand 38:115–120.
mother 56:1885–1891. https://doi.org/10.1128/AAC.05552-11. 95. Dalen A, Hellgren L, Iversen OJ, Vincent J. 1980. Antibodies against
76. Coenye T, Peeters E, Nelis HJ. 2007. Biofilm formation by Propionibac- extractable components from Propionibacterium acnes in humans with
terium acnes is associated with increased resistance to antimicrobial and without acne vulgaris. Arch Dermatol Res 269:253–259. https://doi
agents and increased production of putative virulence factors. Res .org/10.1007/BF00406418.
Microbiol 158:386 –392. https://doi.org/10.1016/j.resmic.2007.02.001. 96. Dhouib M, Gendrault JL, Lugnier AA. 1995. Kinetics of nitric oxide
77. Holmberg A, Lood R, Mörgelin M, Söderquist B, Holst E, Collin M, synthase induction by Propionibacterium avidum and lipopolysac-
Christensson B, Rasmussen M. 2009. Biofilm formation by Propionibac- charide. Eur J Pharmacol 293:439 – 445. https://doi.org/10.1016/0926
terium acnes is a characteristic of invasive isolates. Clin Microbiol Infect -6917(95)90064-0.
15:787–795. https://doi.org/10.1111/j.1469-0691.2009.02747.x. 97. Beuth J, Ko HL, Pulverer G. 1988. The role of hepatic lectins and the
78. Bayston R, Ashraf W, Barker-Davies R, Tucker E, Clement R, Clayton J, activity of the mononuclear phagocyte system in systemic Listeria
Freeman BJC, Nuradeen B. 2007. Biofilm formation by Propionibacte- monocytogenes infection in Balb/c mice. Med Microbiol Immunol (Berl)
rium acnes on biomaterials in vitro and in vivo: impact on diagnosis and 177:47– 49.
treatment. J Biomed Mater Res A 81:705–709. https://doi.org/10.1002/ 98. Pulverer G, Roszkowski W, Beuth HJ, Ko HL, Quie P. 1988. Granulocyte
jbm.a.31145. activating factor released from Propionibacterium acnes. A possible
79. Qi X, Gao J, Sun D, Liang W, Wan Y, Li C, Xu X, Gao T. 2008. Biofilm mediator of inflammation in acne vulgaris. Zentralbl Bakteriol Mikrobiol
formation of the pathogens of fatal bacterial granuloma after trauma: Hyg A 270:246 –251.
potential mechanism underlying the failure of traditional antibiotic 99. Ohshima Y, Ko HL, Beuth J, Burrichter H, Oette K, Pulverer G. 1991.
treatments. Scand J Infect Dis 40:221–228. https://doi.org/10.1080/ Activation of mononuclear immune cells in response to staphylococcal
00365540701632998. lipoteichoic acid. Zentralbl Bakteriol Int J Med Microbiol 275:374 –381.
80. Tunney MM, Patrick S, Curran MD, Ramage G, Anderson N, Davis RI, https://doi.org/10.1016/S0934-8840(11)80303-8.
Gorman SP, Nixon JR. 1999. Detection of prosthetic joint biofilm infec- 100. Ohashi M, Amagai T, Ushijima T, Imanishi J, Kishida T, Ozaki Y. 1983.
tion using immunological and molecular techniques. Methods Enzymol Mode of protection of mice against herpes simplex virus type 2 infec-
310:566 –576. https://doi.org/10.1016/S0076-6879(99)10044-2. tion by Propionibacterium. Microbiol Immunol 27:601– 609. https://doi
81. Jahns AC, Eilers H, Alexeyev OA. 2016. Transcriptomic analysis of Pro- .org/10.1111/j.1348-0421.1983.tb00621.x.
pionibacterium acnes biofilms in vitro. Anaerobe 42:111–118. https:// 101. Kobus M, Luczak M, Sobiczewska E, Szmigielski S, Jeljaszewicz J, Pul-
doi.org/10.1016/j.anaerobe.2016.10.001. verer G. 1997. Protection and therapy of experimental herpesvirus
82. Jakobsen TH, Tolker-Nielsen T, Givskov M. 2017. Bacterial biofilm con- infections in mice with immunomodulating Propionibacterium avidum
trol by perturbation of bacterial signaling processes. Int J Mol Sci KP-40 and/or acyclovir. Zentralbl Bakteriol Int J Med Microbiol 285:
18:E1970. https://doi.org/10.3390/ijms18091970. 445– 449. https://doi.org/10.1016/S0934-8840(97)80011-4.
83. Irie Y, Parsek MR. 2008. Quorum sensing and microbial biofilms. Curr 102. Markowska-Daniel I, Pejsak Z, Furowicz A, Czernomysy-Furowicz D,
Top Microbiol Immunol 322:67– 84. Szmigielski S, Jeljaszewicz J, Pulverer G. 1991. Prophylactic and thera-
84. Lwin SM, Kimber I, McFadden JP. 2014. Acne, quorum sensing and peutic application of Propionibacterium avidum KP-40 in swine and
danger. Clin Exp Dermatol 39:162–167. https://doi.org/10.1111/ced calves with acute enzootic bronchopneumonia. Dtsch Tierarztl
.12252. Wochenschr 98:384 –387.
85. Corvec S, Aubin GG, Bayston R, Ashraf W. 2016. Which is the best 103. Markowska-Daniel I, Pejsak Z, Szmigielski S, Jeljaszewicz J, Pulverer G.
treatment for prosthetic joint infections due to Propionibacterium 1992. Stimulation of granulopoiesis in pregnant swine and their off-
acnes: need for further biofilm in vitro and experimental foreign-body spring by Propionibacterium avidum KP-40. Br Vet J 148:133–145.
in vivo studies? Acta Orthop 87:318 –319. https://doi.org/10.3109/ https://doi.org/10.1016/0007-1935(92)90105-A.
17453674.2016.1162037. 104. Markowska-Daniel I, Pejsak Z, Szmigielski S, Jeljaszewicz J, Pulverer G.
86. Maharshak N, Huh EY, Paiboonrungruang C, Shanahan M, Thurlow L, 1992. Adjuvant properties of Propionibacterium avidum KP-40 in vacci-
Herzog J, Djukic Z, Orlando R, Pawlinski R, Ellermann M, Borst L, Patel nation against endemic viral and bacterial infections. I. Swine immu-
S, Dotan I, Sartor RB, Carroll IM. 2015. Enterococcus faecalis gelatinase nized with live attenuated Aujeszky’s disease virus vaccine and exper-
mediates intestinal permeability via protease-activated receptor 2. In- imentally infected with virulent viruses. Zentralbl Bakteriol Int J Med
fect Immun 83:2762–2770. https://doi.org/10.1128/IAI.00425-15. Microbiol 277:529 –537.
87. Flores-Díaz M, Monturiol-Gross L, Naylor C, Alape-Girón A, Flieger A. 105. Markowska-Daniel I, Pejsak Z, Szmigielski S, Sokolska G, Jeljaszewicz J,
2016. Bacterial sphingomyelinases and phospholipases as virulence Pulverer G. 1992. Adjuvant properties of Propionibacterium avidum
factors. Microbiol Mol Biol Rev 80:597– 628. https://doi.org/10.1128/ KP-40 in vaccination against endemic viral and bacterial infections. III.
MMBR.00082-15. Swine immunized with live attenuated Erysipelothrix rhusiopathiae vac-

cine and experimentally infected with virulent strains R203 and R270B terium avidum in combined with neurotropin in tumor bearing mice.
of E. rhusiopathiae. Zentralbl Bakteriol Int J Med Microbiol 277:547–553. Gan To Kagaku Ryoho 10:2389 –2392.
106. Markowska-Daniel I, Pejsak Z, Szmigielski S, Sokolska G, Jeljaszewicz J, 122. Beuth J, Ko HL, Uhlenbruck G, Pulverer G. 1987. Combined immunos-
Pulverer G. 1992. Adjuvant properties of Propionibacterium avidum timulation (Propionibacterium avidum KP 40) and anticoagulation (hep-
KP-40 in vaccination against endemic viral and bacterial infections. II. arin) prevents metastatic lung and liver colonization in mice. J Cancer
Swine immunized with inactivated Haemophilus pleuropneumoniae Res Clin Oncol 113:359 –362. https://doi.org/10.1007/BF00397719.
vaccine and experimentally infected with different virulent serotypes of 123. Roszkowski W, Szmigielski S, Ko HL, Janiak M, Wrembel JK, Pulverer G,
H. pleuropneumoniae. Zentralbl Bakteriol Int J Med Microbiol 277: Jeljaszewicz J. 1980. Effect of three strains of Propionibacteria (P. granu-
538 –546. losum, P. avidum, P. acnes) and cell-wall preparations on lymphocytes
107. Markowska-Daniel I, Pejsak Z, Szmigielski S, Jeljaszewicz J, Pulverer G. and macrophages. Zentralbl Bakteriol A 246:393– 404.
1993. Prophylactic application of Propionibacterium avidum KP-40 in 124. Roszkowski W, Kobus M, Luczak M, Ko HL, Szmigielski S, Laskowska B,
swine with acute experimental infections. II. Bacterial infections: pleu- Pulverer G, Jeljaszewicz J. 1980. Action of three species of Propionibac-
ropneumonia and swine erysipelas. Dtsch Tierarztl Wochenschr 100: teria (P. granulosum, P. acnes, P. avidum) on experimental tumor sys-
185–188. tems in mice. Zentralbl Bakteriol A 246:405– 414.
108. Markowska-Daniel I, Pejsak Z, Szmigielski S, Jeljaszewicz J, Pulverer G. 125. Hojo H, Endoh YS, Hashimoto Y. 1985. Cytotoxic lymphocytes in rat

1993. Prophylactic application of Propionibacterium avidum KP-40 in tumor in situ: effect of intraperitoneal injections of Propionibacterium
swine with acute experimental infections. I. Viral infections—Aujeszky’s avidum. Jpn J Cancer Res 76:53– 60.
disease and classical swine fever. Dtsch Tierarztl Wochenschr 100: 126. Peters KM, Bornhofen B, Grundmann R, Ko HL, Beuth J, Waters W,
149 –151. Pichlmaier H, Pulverer G. 1990. Neopterin as a marker of aspecific
109. Jasson F, Nagy I, Knol AC, Zuliani T, Khammari A, Dréno B. 2013. immunostimulation with Propionibacterium avidum KP-40 in patients
Different strains of Propionibacterium acnes modulate differently the with gastrointestinal tumors. A prospective randomized study. Med
cutaneous innate immunity. Exp Dermatol 22:587–592. https://doi.org/ Klin (Munich) 85:421– 424.
10.1111/exd.12206. 127. Isenberg J, Stoffel B, Wolters U, Beuth J, Stützer H, Ko HL, Pichlmaier H.
110. Beuth J, Ko HL, Peters KM, Bornhofen B, Pulverer G. 1990. Behaviour 1995. Immunostimulation by propionibacteria— effects on immune
of lymphocyte subsets in response to immunotherapy with Propi- status and antineoplastic treatment. Anticancer Res 15:2363–2368.
onibacterium avidum KP-40 in cancer patients. Zentralbl Bakteriol Int 128. Pulverer G, Ko HL, Tunggal L, Beuth J, Jeljaszewicz J. 1994. Combined
J Med Microbiol 273:386 –390. https://doi.org/10.1016/S0934 immunomodulation (Propionibacterium avidum KP-40) and lectin
-8840(11)80442-1. blocking (D-galactose) prevents liver tumor colonization in BALB/c-
111. Peters KM, Pfeiffer R, Bornhofen B, Ko HL, Beuth J, Grundmann R, mice. Zentralbl Bakteriol Int J Med Microbiol 281:491– 494. https://doi
Pulverer G. 1990. Comparative study on lymphocyte subpopulations in .org/10.1016/S0934-8840(11)80336-1.
cancer patients after immunostimulation with propionibacteria and in 129. Ferguson DA, Cummins CS. 1978. Nutritional requirements of anaero-
bic coryneforms. J Bacteriol 135:858 – 867.
renal transplant patients after combined immunosuppression. Ar-
130. Greenman J, Holland KT, Cunliffe WJ. 1981. Effects of glucose concen-
zneimittelforschung 40:1162–1166.
tration on biomass, maximum specific growth rate and extracellular
112. Toyokawa H, Inaba M, Takai S, Satoi S, Beuth J, Ko HL, Matsui Y, Kwon
enzyme production by three species of cutaneous Propionibacteria
A-H, Kamiyama Y, Ikehara S. 2002. Enhancement of circulating dendritic
grown in continuous culture. J Gen Microbiol 127:371–376.
cell activity by immunomodulators (OK432 and KP-40). Anticancer Res
131. Greenman J, Holland KT, Cunliffe WJ. 1983. Effects of pH on biomass,
22:2137–2145.
maximum specific growth rate and extracellular enzyme production by
113. Pottkämper M, Randerath O, Beuth J, Pulverer G. 1996. Influence of
three species of cutaneous Propionibacteria grown in continuous cul-
Propionibacterium avidum KP-40 on the cellular immune system after
ture. J Gen Microbiol 129:1301–1307.
intensive sport activity. Zentralbl Bakteriol Int J Med Microbiol 284:
132. Cove JH, Holland KT, Cunliffe WJ. 1983. Effects of oxygen concentration
367–371. https://doi.org/10.1016/S0934-8840(96)80112-5.
on biomass production, maximum specific growth rate and extracellu-
114. Jan G, Belzacq A-S, Haouzi D, Rouault A, Métivier D, Kroemer G, Brenner
lar enzyme production by three species of cutaneous Propionibacteria
C. 2002. Propionibacteria induce apoptosis of colorectal carcinoma cells
grown in continuous culture. J Gen Microbiol 129:3327–3334.
via short-chain fatty acids acting on mitochondria. Cell Death Differ 133. Greenman J, Holland KT. 1985. Effects of dilution rate on biomass
9:179 –188. https://doi.org/10.1038/sj.cdd.4400935. and extracellular enzyme production by three species of cutaneous
115. Tchaptchet S, Gumenscheimer M, Kalis C, Freudenberg N, Hölscher C, Propionibacteria grown in continuous culture. J Gen Microbiol 131:
Kirschning CJ, Lamers M, Galanos C, Freudenberg MA. 2012. TLR9- 1619 –1624.
dependent and independent pathways drive activation of the immune 134. Scholz CFP, Brüggemann H, Lomholt HB, Tettelin H, Kilian M. 2016.
system by Propionibacterium acnes. PLoS One 7:e39155. https://doi.org/ Genome stability of Propionibacterium acnes: a comprehensive study of
10.1371/journal.pone.0039155. indels and homopolymeric tracts. Sci Rep 6:20662. https://doi.org/10
116. Kim J, Ochoa M-T, Krutzik SR, Takeuchi O, Uematsu S, Legaspi AJ, .1038/srep20662.
Brightbill HD, Holland D, Cunliffe WJ, Akira S, Sieling PA, Godowski PJ, 135. Hudek R, Sommer F, Kerwat M, Abdelkawi AF, Loos F, Gohlke F. 2014.
Modlin RL. 2002. Activation of Toll-like receptor 2 in acne triggers Propionibacterium acnes in shoulder surgery: true infection, contami-
inflammatory cytokine responses. J Immunol 169:1535–1541. https:// nation, or commensal of the deep tissue? J Shoulder Elbow Surg
doi.org/10.4049/jimmunol.169.3.1535. 23:1763–1771. https://doi.org/10.1016/j.jse.2014.05.024.
117. Dreno B, Gollnick HPM, Kang S, Thiboutot D, Bettoli V, Torres V, Leyden 136. Société de Pathologie Infectieuse de Langue Française (SPILF), Collège des
J, Global Alliance to Improve Outcomes in Acne. 2015. Understanding Universitaires de Maladies Infectieuses et Tropicales (CMIT), Groupe de
innate immunity and inflammation in acne: implications for manage- Pathologie Infectieuse Pédiatrique (GPIP), Société Française d’Anesthésie
ment. J Eur Acad Dermatol Venereol 29(Suppl 4):S3–S11. https://doi et de Réanimation (SFAR), Société Française de Chirurgie Orthopédique et
.org/10.1111/jdv.13190. Traumatologique (SOFCOT), Société Française d’Hygiène Hospitalière
118. Pulverer G, Ko HL, Beuth J, Roszkowski K, Roszkowski W, Jeljaszewicz J. (SFHH), Société Française de Médecine Nucléaire (SFMN), Société Française
1988. Studies on a biological response modifying LYSAT from Propi- de Médecine Physique et de Réadaptation (SOFMER), Société Française de
onibacterium avidum KP 40. Zentralbl Bakteriol Mikrobiol Hyg A 270: Microbiologie (SFM), Société Française de Radiologie (SFR-Rad), Société
237–245. Française de Rhumatologie (SFR-Rhu). 2010. Recommendations for
119. Ko HL, Roszkowski W, Jeljaszewicz J, Pulverer G. 1981. Comparative bone and joint prosthetic device infections in clinical practice (prosthesis,
study on the immunostimulatory potency of different Propionibacte- implants, osteosynthesis). Med Mal Infect 40:185–211. https://doi.org/10
rium strains. Med Microbiol Immunol (Berl) 170:1–9. https://doi.org/10 .1016/j.medmal.2009.12.009.
.1007/BF02123791. 137. Bossard DA, Ledergerber B, Zingg PO, Gerber C, Zinkernagel AS, Zbin-
120. Howard JG, Christie GH, Scott MT. 1973. Biological effects of Coryne- den R, Achermann Y. 2016. Optimal length of cultivation time for
bacterium parvum. IV. Adjuvant and inhibitory activities on B lympho- isolation of Propionibacterium acnes in suspected bone and joint infec-
cytes. Cell Immunol 7:290 –301. tions is more than 7 days. J Clin Microbiol 54:3043–3049. https://doi
121. Toge T, Nakane H, Tanada M, Seto Y, Hamamoto S, Itagaki E, Tomoda .org/10.1128/JCM.01435-16.
S, Hattori T. 1983. Enhancement of antitumor activity of Propionibac- 138. Frangiamore SJ, Saleh A, Grosso MJ, Alolabi B, Bauer TW, Iannotti JP,

Ricchetti ET. 2015. Early versus late culture growth of Propionibacterium microbiology laboratories by matrix-assisted laser desorption
acnes in revision shoulder arthroplasty. J Bone Joint Surg Am 97: ionization–time of flight mass spectrometry. J Clin Microbiol 45:
1149 –1158. https://doi.org/10.2106/JBJS.N.00881. 2156 –2161. https://doi.org/10.1128/JCM.02405-06.
139. Lavergne V, Malo M, Gaudelli C, Laprade M, Leduc S, Laflamme P, 160. Thomin J, Aubin GG, Foubert F, Corvec S. 2015. Assessment of four
Rouleau DM. 2017. Clinical impact of positive Propionibacterium acnes protocols for rapid bacterial identification from positive blood culture
cultures in orthopedic surgery. Orthop Traumatol Surg Res 103: pellets by matrix-assisted laser desorption ionization–time of flight
307–314. https://doi.org/10.1016/j.otsr.2016.12.005. mass spectrometry (Vitek® MS). J Microbiol Methods 115:54 –56. https://
140. Phadnis J, Gordon D, Krishnan J, Bain GI. 2016. Frequent isolation of doi.org/10.1016/j.mimet.2015.05.010.
Propionibacterium acnes from the shoulder dermis despite skin prepa- 161. La Scola B, Fournier P-E, Raoult D. 2011. Burden of emerging anaerobes
ration and prophylactic antibiotics. J Shoulder Elbow Surg 25:304 –310. in the MALDI-TOF and 16S rRNA gene sequencing era. Anaerobe
https://doi.org/10.1016/j.jse.2015.08.002. 17:106 –112. https://doi.org/10.1016/j.anaerobe.2011.05.010.
141. Falconer TM, Baba M, Kruse LM, Dorrestijn O, Donaldson MJ, Smith MM, 162. Farfour E, Leto J, Barritault M, Barberis C, Meyer J, Dauphin B, Le Guern
Figtree MC, Hudson BJ, Cass B, Young AA. 2016. Contamination of the A-S, Leflèche A, Badell E, Guiso N, Leclercq A, Le Monnier A, Lecuit M,
surgical field with Propionibacterium acnes in primary shoulder arthro- Rodriguez-Nava V, Bergeron E, Raymond J, Vimont S, Bille E, Carbon-
plasty. J Bone Joint Surg Am 98:1722–1728. https://doi.org/10.2106/ nelle E, Guet-Revillet H, Lécuyer H, Beretti J-L, Vay C, Berche P, Ferroni

JBJS.15.01133. A, Nassif X, Join-Lambert O. 2012. Evaluation of the Andromas matrix-
142. Martín-Rabadán P, Gijón P, Alcalá L, Rodríguez-Créixems M, Alvarado N, assisted laser desorption ionization–time of flight mass spectrometry
Bouza E. 2008. Propionibacterium acnes is a common colonizer of system for identification of aerobically growing Gram-positive bacilli. J
intravascular catheters. J Infect 56:257–260. https://doi.org/10.1016/j Clin Microbiol 50:2702–2707. https://doi.org/10.1128/JCM.00368-12.
.jinf.2008.01.012. 163. Peel TN, Cole NC, Dylla BL, Patel R. 2015. Matrix-assisted laser
143. Foruria AM, Fox TJ, Sperling JW, Cofield RH. 2013. Clinical meaning of desorption ionization time of flight mass spectrometry and diag-
unexpected positive cultures (UPC) in revision shoulder arthroplasty. J nostic testing for prosthetic joint infection in the clinical microbi-
Shoulder Elbow Surg 22:620 – 627. https://doi.org/10.1016/j.jse.2012.07 ology laboratory. Diagn Microbiol Infect Dis 81:163–168. https://doi
.017. .org/10.1016/j.diagmicrobio.2014.11.015.
144. Mook WR, Klement MR, Green CL, Hazen KC, Garrigues GE. 2015. The 164. Nagy E, Urbán E, Becker S, Kostrzewa M, Vörös A, Hunyadkürti J, Nagy
incidence of Propionibacterium acnes in open shoulder surgery: a con- I. 2013. MALDI-TOF MS fingerprinting facilitates rapid discrimination of
trolled diagnostic study. J Bone Joint Surg Am 97:957–963. https://doi phylotypes I, II and III of Propionibacterium acnes. Anaerobe 20:20 –26.
.org/10.2106/JBJS.N.00784. https://doi.org/10.1016/j.anaerobe.2013.01.007.
145. Kim SJ, Kim JH. 2014. Unexpected positive cultures including isolation 165. Brook I, Frazier EH. 1991. Infections caused by Propionibacterium
of Propionibacterium acnes in revision shoulder arthroplasty. Chin Med species. Rev Infect Dis 13:819 – 822. https://doi.org/10.1093/clinids/
J (Engl) 127:3975–3979. 13.5.819.
146. Marples RR, McGinley KJ. 1974. Corynebacterium acnes and other 166. Sartelli M, Weber DG, Ruppé E, Bassetti M, Wright BJ, Ansaloni L, Catena
anaerobic diphtheroids from human skin. J Med Microbiol 7:349 –357. F, Coccolini F, Abu-Zidan FM, Coimbra R, Moore EE, Moore FA, Maier RV,
https://doi.org/10.1099/00222615-7-3-349. De Waele JJ, Kirkpatrick AW, Griffiths EA, Eckmann C, Brink AJ, Mazuski
147. Linzenmeier G. 1954. Serological study of anaerobic Corynebacteria by JE, May AK, Sawyer RG, Mertz D, Montravers P, Kumar A, Roberts JA,
agglutinin method. Ann Inst Pasteur 87:572–579. Vincent J-L, Watkins RR, Lowman W, Spellberg B, Abbott IJ, Adesunk-
148. Linzenmeier G. 1957. Serology of anaerobic corynebacteria. II. Serology anmi AK, Al-Dahir S, Al-Hasan MN, Agresta F, Althani AA, Ansari S,
and pathogenetic significance of Corynebacterium acnes. Zentralbl Bak- Ansumana R, Augustin G, Bala M, Balogh ZJ, Baraket O, Bhangu A,
teriol Orig 170:85–90. Beltrán MA, Bernhard M, Biffl WL, Boermeester MA, Brecher SM, Cherry-
149. Pulverer G, Ko HL. 1973. Fermentative and serological studies on Bukowiec JR, Buyne OR, Cainzos MA, Cairns KA, Camacho-Ortiz A,
Propionibacterium acnes. Appl Microbiol 25:222–229. Chandy SJ, Che Jusoh A, Chichom-Mefire A, Colijn C, Corcione F, Cui Y,
150. Cummins CS. 1976. Identification of Propionibacterium acnes and re- Curcio D, Delibegovic S, Demetrashvili Z, De Simone B, Dhingra S, Diaz
lated organisms by precipitin tests with trichloroacetic acid extracts. J JJ, Di Carlo I, Dillip A, Di Saverio S, Doyle MP, Dorj G, Dogjani A, Dupont
Clin Microbiol 2:104 –110. H, Eachempati SR, Enani MA, Egiev VN, Elmangory MM, Ferrada P,
151. Nordstrom KM. 1985. Polyacrylamide gel electrophoresis (PAGE) of Fitchett JR, Fraga GP, Guessennd N, Giamarellou H, Ghnnam W, Gkiokas
whole-cell proteins of cutaneous Propionibacterium species. J Med G, Goldberg SR, Gomes CA, Gomi H, Guzmán-Blanco M, Haque M,
Microbiol 19:9 –14. https://doi.org/10.1099/00222615-19-1-9. Hansen S, Hecker A, Heizmann WR, Herzog T, Hodonou AM, Hong S-K,
152. Cummins CS, White RH. 1983. Isolation, identification, and synthesis of Kafka-Ritsch R, Kaplan LJ, Kapoor G, Karamarkovic A, Kees MG, Kenig J,
2,3-diamino-2,3-dideoxyglucuronic acid: a component of Propionibac- Kiguba R, Kim PK, Kluger Y, Khokha V, Koike K, Kok KYY, Kong V, Knox
terium acnes cell wall polysaccharide. J Bacteriol 153:1388 –1393. MC, Inaba K, Isik A, Iskandar K, Ivatury RR, Labbate M, Labricciosa FM,
153. Bernard KA, Bellefeuille M, Ewan EP. 1991. Cellular fatty acid composi- Laterre P-F, Latifi R, Lee JG, Lee YR, Leone M, Leppaniemi A, Li Y, Liang
tion as an adjunct to the identification of asporogenous, aerobic SY, Loho T, Maegele M, Malama S, Marei HE, Martin-Loeches I, Marwah
gram-positive rods. J Clin Microbiol 29:83– 89. S, Massele A, McFarlane M, Melo RB, Negoi I, Nicolau DP, Nord CE,
154. Woese CR, Fox GE, Zablen L, Uchida T, Bonen L, Pechman K, Lewis BJ, Ofori-Asenso R, Omari AH, Ordonez CA, Ouadii M, Pereira Júnior GA,
Stahl D. 1975. Conservation of primary structure in 16S ribosomal RNA. Piazza D, Pupelis G, Rawson TM, Rems M, Rizoli S, Rocha C, Sakakhushev
Nature 254:83– 86. https://doi.org/10.1038/254083a0. B, Sanchez-Garcia M, Sato N, Segovia Lohse HA, Sganga G, Siribum-
155. Aubin GG, Bémer P, Guillouzouic A, Crémet L, Touchais S, Fraquet N, rungwong B, Shelat VG, Soreide K, Soto R, Talving P, Tilsed JV, Timsit
Boutoille D, Reynaud A, Lepelletier D, Corvec S. 2011. First report of a J-F, Trueba G, Trung NT, Ulrych J, van Goor H, Vereczkei A, Vohra RS,
hip prosthetic and joint infection caused by Lactococcus garvieae in a Wani I, Uhl W, Xiao Y, Yuan K-C, Zachariah SK, Zahar J-R, Zakrison TL,
woman fishmonger. J Clin Microbiol 49:2074 –2076. https://doi.org/10 Corcione A, Melotti RM, Viscoli C, Viale P. 2016. Antimicrobials: a global
.1128/JCM.00065-11. alliance for optimizing their rational use in intra-abdominal infections
156. Aubin GG, Haloun A, Treilhaud M, Reynaud A, Corvec S. 2013. Gallibac- (AGORA). World J Emerg Surg 11:33. https://doi.org/10.1186/s13017
terium anatis bacteremia in a human. J Clin Microbiol 51:3897–3899. -016-0089-y.
https://doi.org/10.1128/JCM.01638-13. 167. Janvier F, Delacour H, Larréché S, Abdalla S, Aubert P, Mérens A. 2013.
157. Srinivasan R, Karaoz U, Volegova M, MacKichan J, Kato-Maeda M, Abdominal wall and intra-peritoneal abscess by Propionibacterium avi-
Miller S, Nadarajan R, Brodie EL, Lynch SV. 2015. Use of 16S rRNA dum as a complication of abdominal parietoplasty. Pathol Biol (Paris)
gene for identification of a broad range of clinically relevant bac- 61:223–225. https://doi.org/10.1016/j.patbio.2013.01.008.
terial pathogens. PLoS One 10:e0117617. https://doi.org/10.1371/ 168. Wang TKF, Woo PCY, Yuen KY. 2002. Perianal abscess caused by
journal.pone.0117617. Propionibacterium avidum in a cirrhotic patient. New Microbiol 25:
158. Georgiades K, Raoult D. 2010. Defining pathogenic bacterial species in 239 –242.
the genomic era. Front Microbiol 1:151. https://doi.org/10.3389/fmicb 169. Dunne WM, Kurschenbaum HA, Deshur WR, Dee TH, Samter TG, Wil-
.2010.00151. liams JE, Zabransky RJ. 1986. Propionibacterium avidum as the etiologic
159. Carbonnelle E, Beretti J-L, Cottyn S, Quesne G, Berche P, Nassif X, agent of splenic abscess. Diagn Microbiol Infect Dis 5:87–92. https://
Ferroni A. 2007. Rapid identification of staphylococci isolated in clinical doi.org/10.1016/0732-8893(86)90096-9.

170. Gangahar DM, Delany HM. 1981. Intrasplenic abscess: two case reports patient risk factors, and efficacy of perioperative prophylaxis. J Shoul-
and review of the literature. Am Surg 47:488 – 491. der Elbow Surg 25:846 – 852. https://doi.org/10.1016/j.jse.2015.09.033.
171. Gekowski KM, Lopes R, LiCalzi L, Bia FJ. 1982. Splenic abscess caused by 191. Davidsson S, Mölling P, Rider JR, Unemo M, Karlsson MG, Carlsson J,
Propionibacterium acnes. Yale J Biol Med 55:65– 69. Andersson S-O, Elgh F, Söderquis B, Andrén O. 2016. Frequency and
172. Kiritani S, Kaneko J, Aoki T, Sakamoto Y, Hasegawa K, Sugawara Y, typing of Propionibacterium acnes in prostate tissue obtained from men
Kokudo N. 2013. Multiple splenic nodules with fever: a case of splenic with and without prostate cancer. Infect Agent Cancer 11:26. https://
abscess due to Propionibacterium acnes. Clin J Gastroenterol doi.org/10.1186/s13027-016-0074-9.
6:434 – 437. https://doi.org/10.1007/s12328-013-0427-5. 192. Mak TN, Yu S-H, De Marzo AM, Brüggemann H, Sfanos KS. 2013.
173. Vohra A, Saiz E, Chan J, Castro J, Amaro R, Barkin J. 1998. Splenic Multilocus sequence typing (MLST) analysis of Propionibacterium acnes
abscess caused by Propionibacterium avidum as a complication of isolates from radical prostatectomy specimens. Prostate 73:770 –777.
cardiac catheterization. Clin Infect Dis 26:770 –771. https://doi.org/10 https://doi.org/10.1002/pros.22621.
.1086/517127. 193. Drott JB, Alexeyev O, Bergström P, Elgh F, Olsson J. 2010. Propionibac-
174. Nakase K, Sakuma Y, Nakaminami H, Noguchi N. 2016. Emergence of terium acnes infection induces upregulation of inflammatory genes and
fluoroquinolone-resistant Propionibacterium acnes caused by amino cytokine secretion in prostate epithelial cells. BMC Microbiol 10:126.
acid substitutions of DNA gyrase but not DNA topoisomerase IV. https://doi.org/10.1186/1471-2180-10-126.

Anaerobe 42:166 –171. https://doi.org/10.1016/j.anaerobe.2016.10.012. 194. Fassi Fehri L, Mak TN, Laube B, Brinkmann V, Ogilvie LA, Mollenkopf H,
175. Bentorki AA, Gouri A, Yakhlef A, Touaref A, Bensouilah T, Gueroudj A. Lein M, Schmidt T, Meyer TF, Brüggemann H. 2011. Prevalence of
2013. Propionibacterium avidum cutaneous abscess in a young immu- Propionibacterium acnes in diseased prostates and its inflammatory and
nocompetent. Ann Biol Clin (Paris) 71:703–706. https://doi.org/10.1684/ transforming activity on prostate epithelial cells. Int J Med Microbiol
abc.2013.0909. 301:69 –78. https://doi.org/10.1016/j.ijmm.2010.08.014.
176. Vetromile F, Malfante P, Fuentes N, Bosio A. 2001. Endocarditis bacte- 195. Lentze I, Ko HL, Höffler U. 1979. Differentiation of various Propionibac-
riana por Propionibacterium avidum: une primer caso en la literatura? terium species from acne-vulgaris efflorescences. Hautarzt 30:242–247.
Rev Hosp Priv Comunidad 4:37–39. 196. Grice EA, Segre JA. 2011. The skin microbiome. Nat Rev Microbiol
177. Braun DL, Hasse BK, Stricker J, Fehr JS. 2013. Prosthetic valve endocar- 9:244 –253. https://doi.org/10.1038/nrmicro2537.
ditis caused by Propionibacterium species successfully treated with 197. Gallagher IH, Pearce EI. 1983. The fermentation of sucrose, sorbitol, and
coadministered rifampin: report of two cases. BMJ Case Rep 2013: xylitol by Propionibacterium avidum, resulting in the formation of
bcr2012007204. https://doi.org/10.1136/bcr-2012-007204. caries-like lesions in enamel. N Z Dent J 79:75–79.
178. Loureiro-Amigo J, Pons S, Sierra M, Meije Y. 2017. Prosthetic valve with 198. Hoeffler U, Ko HL, Pulverer G. 1976. Antimicrobiol susceptibility of
infective endocarditis caused by Propionibacterium avidum. A case Propinibacterium acnes and related microbial species. Antimicrob
report. Enferm Infecc Microbiol Clin 35:196 –197. https://doi.org/10 Agents Chemother 10:387–394. https://doi.org/10.1128/AAC.10.3.387.
.1016/j.eimc.2016.08.010. 199. Höffler U, Niederau W, Pulverer G. 1980. Susceptibility of cutaneous
Propionibacteria to newer antibiotics. Chemotherapy 26:7–11. https://
179. Sohail MR, Gray AL, Baddour LM, Tleyjeh IM, Virk A. 2009. Infective
doi.org/10.1159/000237877.
endocarditis due to Propionibacterium species. Clin Microbiol Infect
200. Eady EA, Cove JH, Holland KT, Cunliffe WJ. 1989. Erythromycin resistant
15:387–394. https://doi.org/10.1111/j.1469-0691.2009.02703.x.
propionibacteria in antibiotic treated acne patients: association with
180. Yedidya I, Goldberg E, Sharoni R, Sagie A, Vaturi M. 2015. Infective
therapeutic failure. Br J Dermatol 121:51–57.
endocarditis caused by Propionibacterium granulosum. Isr Med Assoc J
201. Olsson J, Davidsson S, Unemo M, Mölling P, Andersson S-O, Andrén O,
17:642– 643.
Söderquist B, Sellin M, Elgh F. 2012. Antibiotic susceptibility in prostate-
181. Armstrong RW, Wuerflein RD. 1996. Endocarditis due to Propionibacte-
derived Propionibacterium acnes isolates. APMIS 120:778 –785. https://
rium granulosum. Clin Infect Dis 23:1178 –1179. https://doi.org/10.1093/
doi.org/10.1111/j.1600-0463.2012.02905.x.
clinids/23.5.1178.
202. Oprica C, Löfmark S, Lund B, Edlund C, Emtestam L, Nord CE. 2005.
182. Banzon JM, Rehm SJ, Gordon SM, Hussain ST, Pettersson GB, Shrestha
Genetic basis of resistance in Propionibacterium acnes strains isolated
NK. 2017. Propionibacterium acnes endocarditis: a case series. Clin
from diverse types of infection in different European countries. Anaer-
Microbiol Infect 23:396 –399. https://doi.org/10.1016/j.cmi.2016.12.026.
obe 11:137–143. https://doi.org/10.1016/j.anaerobe.2005.01.005.
183. Zeller VA, Letembet VA, Meyssonnier VA, Heym B, Ziza JM, Marmor SD. 203. Khassebaf J, Hellmark B, Davidsson S, Unemo M, Nilsdotter-
2018. Cutibacterium (formerly Propionibacterium) avidum: a rare but avid Augustinsson Å, Söderquist B. 2014. Antibiotic susceptibility of Propi-
agent of prosthetic hip infection. J Arthroplasty 2018:S0883- onibacterium acnes isolated from orthopaedic implant-associated infec-
5403(18)30135-9. https://doi.org/10.1016/j.arth.2018.02.008. tions. Anaerobe 32C:57– 62. https://doi.org/10.1016/j.anaerobe.2014.12
184. Estoppey O, Rivier G, Blanc CH, Widmer F, Gallusser A, So AK. 1997. .006.
Propionibacterium avidum sacroiliitis and osteomyelitis. Rev Rhum Engl 204. Dumont-Wallon G, Moyse D, Blouin E, Dréno B. 2010. Bacterial resis-
Ed 64:54 –56. tance in French acne patients. Int J Dermatol 49:283–288. https://doi
185. Million M, Roux F, Cohen Solal J, Bréville P, Desplaces N, Barthas J, .org/10.1111/j.1365-4632.2009.04270.x.
Nguyen Van J-C, Rajzbaum G. 2008. Septic arthritis of the hip with 205. Garrido-Mesa N, Zarzuelo A, Gálvez J. 2013. Minocycline: far beyond
Propionibacterium avidum bacteremia after intraarticular treatment for an antibiotic. Br J Pharmacol 169:337–352. https://doi.org/10.1111/
hip osteoarthritis. Joint Bone Spine 75:356 –358. https://doi.org/10 bph.12139.
.1016/j.jbspin.2007.06.012. 206. Chopra I, Hacker K. 1989. Effects of tetracyclines on the production
186. Portillo ME, Corvec S, Borens O, Trampuz A. 2013. Propionibacterium of extracellular proteins by members of the Propionibacteriaceae.
acnes: an underestimated pathogen in implant-associated infections. FEMS Microbiol Lett 51:21–24. https://doi.org/10.1111/j.1574-6968
Biomed Res Int 2013:804391. https://doi.org/10.1155/2013/804391. .1989.tb03412.x.
187. Ilchmann T, Zimmerli W, Bolliger L, Graber P, Clauss M. 2016. Risk of 207. Walsh TR, Efthimiou J, Dréno B. 2016. Systematic review of antibiotic
infection in primary, elective total hip arthroplasty with direct anterior resistance in acne: an increasing topical and oral threat. Lancet Infect
approach or lateral transgluteal approach: a prospective cohort study Dis 16:e23– e33. https://doi.org/10.1016/S1473-3099(15)00527-7.
of 1104 hips. BMC Musculoskelet Disord 17:471. https://doi.org/10 208. Dizay HH, Lau DG, Nottage WM. 2017. Benzoyl peroxide and clindamy-
.1186/s12891-016-1332-0. cin topical skin preparation decreases Propionibacterium acnes coloni-
188. Levy O, Iyer S, Atoun E, Peter N, Hous N, Cash D, Musa F, Narvani AA. zation in shoulder arthroscopy. J Shoulder Elbow Surg 26:1190 –1195.
2013. Propionibacterium acnes: an underestimated etiology in the https://doi.org/10.1016/j.jse.2017.03.003.
pathogenesis of osteoarthritis? J Shoulder Elbow Surg 22:505–511. 209. Sabetta JR, Rana VP, Vadasdi KB, Greene RT, Cunningham JG, Miller SR,
https://doi.org/10.1016/j.jse.2012.07.007. Sethi PM. 2015. Efficacy of topical benzoyl peroxide on the reduction of
189. Matsen FA, Russ SM, Bertelsen A, Butler-Wu S, Pottinger PS. 2015. Propi- Propionibacterium acnes during shoulder surgery. J Shoulder Elbow
onibacterium can be isolated from deep cultures obtained at primary Surg 24:995–1004. https://doi.org/10.1016/j.jse.2015.04.003.
arthroplasty despite intravenous antimicrobial prophylaxis. J Shoulder 210. Shah NB, Tande AJ, Patel R, Berbari EF. 2015. Anaerobic prosthetic joint
Elbow Surg 24:844– 847. https://doi.org/10.1016/j.jse.2014.10.016. infection. Anaerobe 36:1– 8. https://doi.org/10.1016/j.anaerobe.2015.08
190. Koh CK, Marsh JP, Drinković D, Walker CG, Poon PC. 2016. Propionibac- .003.
terium acnes in primary shoulder arthroplasty: rates of colonization, 211. Sendi P, Zimmerli W. 2012. Antimicrobial treatment concepts for or-

thopaedic device-related infection. Clin Microbiol Infect 18:1176 –1184. Preoperative immunostimulation with Propionibacterium avidum KP-40
https://doi.org/10.1111/1469-0691.12003. in patients with gastric carcinoma: a prospective randomized study.
212. Peters KM, Beuth J, Ko HL, Pulverer G, Kluger J, Grundmann R. 1990. Onkologie 13:124 –127.
Stéphane Corvec, Pharm.D., Ph.D., is cur-

rently Associate Professor in the microbi-
ology laboratory at the Nantes University
hospital and in the Faculty of Medicine
(Nantes, France). After obtaining his Ph.D.
in 2000 by working on Gram-negative
bacillus resistance, in 2011 he joined the
University of Lausanne (Lausanne, Switzer-

land) to work with A. Trampuz (CHUV, Lau-
sanne, Switzerland) on implant-associated
infections, mainly prosthetic joint infec-
tions. He has been involved in several clinical and epidemiological
studies in collaboration with experts in the field of implant-
associated infections in France, especially in the CRIOGO network, or
with international collaborations. Upon his return to Nantes, in 2013,
he started a collaboration with the dermatologist research unit (B.
Dréno), especially working on Cutibacterium species, and he joined the
INSERM U1232 team in 2016. His research is centered on the patho-
genesis of Cutibacterium species and their phylogeny, antibiotic resis-
tance, and biofilm formation. He has coauthored over 145 peer-
reviewed papers and is currently working on Cutibacterium species in
acne and implant-related infections. He also organized the 36th Euro-
pean Bone and Joint Infection Society meeting in Nantes, September
2017.

Clinical Microbiology Reviews-2018-Corvec-e00064-17.full

Uploaded by

Copyright:

Available Formats

Clinical Microbiology Reviews-2018-Corvec-e00064-17.full

Uploaded by

Document Information

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

Clinical Microbiology Reviews-2018-Corvec-e00064-17.full

Uploaded by

Copyright:

Available Formats

REVIEW

Clinical and Biological Features of Cutibacterium (Formerly

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 Clinical Microbiology Reviews cmr.asm.org 1

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

T he genus Propionibacterium, initially described by Orla-Jensen in 1909 (1), is made

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 2

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 3

probably determine their colonizing capability and pathogenicity at the cutaneous

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

Genetic and Genomic Analyses

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 4

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 5

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

Cutibacterium granulosum NCTC11865 2.17524 64.2 1,857 1,749

Cutibacterium acnes KPA171202 2.56026 60 2,371 2,277

Cutibacterium namnetense NTS31307302 2.36966 60.5 24 2,188 2,052

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 6

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 7

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

A detailed comparison of available C. avidum genomes was also conducted to

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 8

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 9

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 10

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 11

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 12

Other Virulence Factors

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 13

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 14

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 15

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 16

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 17

Contamination Issues in Culture

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 18

(146). Conﬁrmation was obtained by testing indole production, hemolysin production,

Serological and Precipitin Tests for Identiﬁcation of Cutibacterium-Related

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

Cell Wall Analysis and Protein Pattern Analysis

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 19

Susceptibility to Free Fatty Acids

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

Biochemical and Enzyme Proﬁles

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 20

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest

July 2018 Volume 31 Issue 3 e00064-17 cmr.asm.org 21

used, such as the GenBank database (https://blast.ncbi.nlm.nih.gov/Blast.cgi), RIDOM

Downloaded from http://cmr.asm.org/ on September 24, 2020 by guest