A new phylogenetic group of Propionibacterium
Andrew McDowell,1Alexandra L. Perry,2Peter A. Lambert2
and Sheila Patrick1
1School of Medicine and Dentistry, Queen’s University, Belfast BT12 6BN, UK
2Pharmaceutical and Biological Sciences, School of Life and Heath Sciences, Aston University,
Birmingham B4 7ET, UK
Received 2 July 2007
Accepted 12 October 2007
Immunofluorescence microscopy-based identification of presumptive Propionibacterium acnes
isolates, using the P. acnes-specific mAb QUBPa3, revealed five organisms with an atypical
cellular morphology. Unlike the coryneform morphology seen with P. acnes types I and II, these
isolates exhibited long slender filaments (which formed large tangled aggregates) not previously
described in P. acnes. No reaction with mAbs that label P. acnes types IA (QUBPa1) and II
(QUBPa2) was observed. Nucleotide sequencing of the 16S rRNA gene (1484 bp) revealed the
isolates to have between 99.8 and 99.9% identity to the 16S rRNA gene of the P. acnes type IA,
IB and II strains NCTC 737, KPA171202 and NCTC 10390, respectively. Analysis of the
recA housekeeping gene (1047 bp) did reveal, however, a greater number of conserved
nucleotide polymorphisms between the sequences from these isolates and those from NCTC 737
(98.9% identity), KPA171202 (98.9% identity) and NCTC 10390 (99.1% identity). Phylogenetic
investigations demonstrated that the isolates belong to a novel recA cluster or lineage distinct
from P. acnes types I and II. We now propose this new grouping as P. acnes type III. The
prevalence and clinical importance of this novel recA lineage amongst isolates of P. acnes
remains to be determined.
Propionibacterium acnes belongs to the ‘high GC’ group of
Gram-positive bacteria and is found predominately in the
sebaceous gland-rich areas of the skin in adults (McGinley
et al., 1978). Although traditionally considered a relatively
non-pathogenic member of the resident human micro-
biota, an increasing number of studies have implicated P.
acnes as the agent responsible for various clinical condi-
tions and infections. In addition to its well described role in
inflammatory acne (Eady & Ingham, 1994), P. acnes is
emerging as an important pathogen in relation to medical
implant-related infections, such as those associated with
central nervous system shunts (Brook & Frazier, 1991),
silicone implants (Ahn et al., 1996), and prosthetic heart
valves and hip joints (Delahaye et al., 2005; Tunney et al.,
1998, 1999). Furthermore, P. acnes is responsible for
endophthalmitis, ocular and periocular infections (Aldave
et al., 1999; Clark et al., 1999; Horgan et al., 1999), as well
as periodontal and dental infections (Debelian et al., 1992;
LeGoff et al., 1997), and has been linked to synovitis–acne–
(Kotilainen et al., 1996), sarcoidosis (Eishi et al., 2002)
and prostate cancer (Cohen et al., 2005). The identification
of a wide range of putative virulence determinants within
the recently published genome sequence of the P. acnes
strain KPA17202 has further brought the pathogenic
potential of this organism into focus (Bruggemann et al.,
Two distinct phenotypes of P. acnes (types I and II), which
can be distinguished by serological agglutination tests and
cell-wall sugar analysis, have been known for over 30 years
(Johnson & Cummins, 1972). Additional studies have
shown that these biovars display differences in the
fermentation of sugar and sugar alcohols (Higaki et al.,
2000; Kishishita et al., 1979), as well as their susceptibility
to bacteriophage infection (Webster & Cummins, 1978).
Sequence analysis of the P. acnes recA gene has revealed
that types I and II correspond to phylogenetically distinct
clusters or lineages (McDowell et al., 2005). These two
clusters are, however, almost identical based on 16S rRNA
sequencing. Analysis of the recA gene has also identified a
subcluster of strains within P. acnes type I that have been
Abbreviations: IFM, immunofluorescence microscopy; NCTC, National
Collection of Type Cultures; ROH, Royal Orthopaedic Hospital.
The GenBank/EMBL/DDBJ accession numbers for the recA and 16s
RNA sequences of Propionibacterium acnes reported in this paper are
DQ672246–DQ672256 and DQ672257–DQ672261, respectively.
A figure of recA sequence alignments is available as supplementary data
with the online version of this paper.
Journal of Medical Microbiology (2008), 57, 218–224
218 47489G2008 SGMPrinted in Great Britain
designated type IB (McDowell et al., 2005; Valanne et al.,
2005). These organisms do not react with the previously
described mAb QUBPa1, specific for all other type I
organisms (known as type IA) (McDowell et al., 2005;
Valanne et al., 2005). However, variable labelling with the
type II mAb QUBPa2, ranging from no reaction to a weak
reaction, has been observed. In the latter case a significantly
reduced fluorescence intensity and a reduction in the
proportion of the bacterial population labelled is seen
(McDowell et al., 2005). Type IB organisms also display
differences from type IA strains in the production of
putative Christie–Atkins–Munch–Peterson (CAMP) factor
proteins (Valanne et al., 2005).
This paper describes the identification of a novel lineage of
P. acnes based on recA sequence analysis, which we now
formally propose as P. acnes type III. The isolates also
display differences from P. acnes types I and II in their cell
surface antigens and cellular morphology.
Bacterial strains. The P. acnes reference strains NCTC 737 (type IA,
ATCC6919) and NCTC 10390 (type II, ATCC12930) were obtained
from the National Collection of Type Cultures (NCTC) (London,
UK). A total of four isolates of P. acnes type III were recovered from
spine intervertebral disc material removed during microdiscectomy
procedures (for relief of severe sciatica), while one isolate was
recovered from a prosthetic hip joint removed during revision
arthroplasty. All surgeries were performed at the Royal Orthopaedic
Hospital (ROH), Birmingham, England. A further 95 isolates of P.
acnes (types I & II) were also included in our study for comparative
purposes. A total of 15 of these isolates were recovered from failed
prosthetic hip joints, as well as associated tissue, removed during
revision arthroplasties; while a further 34 isolates were isolated from
spine intervertebral disc material removed during microdiscectomy
procedures at ROH. A total of 13 isolates from routine blood cultures,
18 from acne lesions and 15 from normal skin were obtained from the
Queen Elizabeth Hospital, Birmingham,
Propionibacterium granulosum and Propionibacterium avidum recov-
ered from patients with acne lesions were a kind gift from Professor
Keith Holland. Ethical approval was obtained from the South
Birmingham Local Research Ethics Committee and informed patient
consent was obtained in all cases.
Bacterial culture. Strains were grown on anaerobic blood agar
(ABA) (CM0972; Oxoid). Cultures were incubated at 37 uC in an
anaerobic cabinet (MACS MG 1000; Don Whitley Scientific), in an
atmosphere of 80% N2, 10% CO2 and 10% H2. Isolates were
examined using the API ID 32A and API 20A biochemical
identification systems (bioMe ´rieux) in accordance with the manu-
facturer’s instructions. Fermentation analyses with the substrates
sorbitol, ribose and erythritol were carried out as previously described
(McDowell et al., 2005). Putative virulence factors, namely hae-
molytic, proteinase, lipase, lecithinase, DNase, elastase and hyalur-
onidase activities, were detected as reported by Balke & Weiss (1984)
and Spare et al. (2003).
Immunofluorescence microscopy (IFM). IFM was conducted on
multiwell slides as described previously (McDowell et al., 2005).
Strains were examined with the mAbs QUBPa1 and QUBPa2, which
label P. acnes type IA and II, respectively, as well as mAb QUBPa3,
which reacts with all P. acnes strains examined to date (Tunney et al.,
1999). Slides were read using a Leitz Dialux 20 fluorescence
microscope. Images were captured and bacterial cell lengths
determined using LUCIA G software following the manufacturer’s
instructions (Laboratory Imaging).
Nucleotide sequence analysis. Nucleotide sequence analysis was
performed on 16S rRNA and recA genes. The 16S rRNA gene
(1484 bp) was amplified using the universal primers UFPL and URPL
(LiPuma et al., 1999), while recA was amplified with the primers PAR-
1 and PAR-2, which are directed to downstream and upstream
flanking sequences of the recA ORF, respectively, and generate a
1201 bp amplicon (McDowell et al., 2005). PCR products were
verified by electrophoresis on 1% (w/v) agarose gels and duplicate
samples for each gene pooled before purification with a QIAquick
PCR purification kit (Qiagen) (McDowell et al., 2005). Sequencing
reactions were performed using ABI PRISM ready reaction terminator
Biosystems) according to the manufacturer’s instructions and the
samples analysed on an ABI PRISM 3100 genetic analyser capillary
electrophoresis system (Perkin-Elmer Applied Biosystems). Raw
sequences from both DNA strands were obtained by using the
appropriate forward and reverse primers. Internal sequencing primers
were also used to facilitate determination of the larger 16S rRNA gene
sequence. Initial sequences were screened using the Basic Local
Alignment Search Tool (BLAST; www.ncbi.nlm.nih.gov) to confirm
Phylogenetic analysis. Multiple nucleotide sequence alignments
(with no gaps) of the recA gene (1047 bp) from all five atypical
isolates, as well as complete recA sequences for previously published
strains representative of types I and II (McDowell et al., 2005), were
constructed using CLUSTAL W (Thompson et al., 1994). Phylogenetic
analyses were conducted using the genetic distance-based neighbour-
joining algorithms within MEGA version 3.1 (http://www.megasoft-
ware.net/) and phylogenetic trees constructed using the Jukes–Cantor
matrix model. The sequence input order was randomized and
bootstrapping resampling statistics were performed using 100 datasets
for each analysis.
Nucleotide sequence accession numbers. Nucleotide sequences
for 16S rRNA and recA gene sequences generated during this study
were submitted to GenBank and the accession numbers DQ672257–
DQ672261 and DQ672246–DQ672256 generated, respectively.
RESULTS AND DISCUSSION
During routine IFM-based identification of clinical isolates
presumptively identified as P. acnes by biochemical
analysis, five organisms (designated Asn10–Asn14) with
an atypical morphology were identified after labelling with
the mAb QUBPa3, which reacts with a carbohydrate or
glycolipid-containing antigen on the surface of all P. acnes
cells (Tunney et al., 1999). Four of these isolates were
recovered from spine intervertebral disc material (excised
disc protrusion) removed during microdiscectomy proce-
dures for severe sciatica, while one was from a prosthetic
hip arthroplasty. Compared to the classical coryneform
morphology normally seen with types I and II (i.e. clubs,
‘tadpole’ forms and short bifid forms), these isolates
consisted of individual cells of variable length and long
slender filaments that formed very large tangled aggregates
A new phylogenetic group of P. acnes
(Figs 1 and 2). This was also observed upon Gram staining
of the cells (not illustrated). No labelling with the mAbs
QUBPa1 and QUBPa2, which react with a proteinaceous
and carbohydrate/glycolipid-containing antigen on types IA
and II (McDowell et al., 2005), respectively, was detected for
any of the isolates. This provides evidence that the cell
surface antigens recognized by these mAbs either are not
present, or are structurally distinct to those found on types I
and II giving rise to a different set of epitopes.
Analysis of cells from one of these isolates (Asn12),
captured from three different fields of view of an IFM
image using a 6100 objective, revealed individual cells and
filaments that varied in length from 1.20 to 21.8 mm, with
widths ranging from 0.71 to 0.96 mm. However, much
longer filaments, which could be difficult to identify due to
a greater tendency to form or to be associated with large
aggregates, may also be present. For comparison, analysis
of individual cells of NCTC 737 (type IA) and NCTC 10390
(type II) in a similar manner revealed lengths that ranged
from 0.84 to 2.56 mm and 0.89 to 2.80 mm, respectively,
with widths ranging from 0.63 to 0.88 mm and 0.66 to
0.86 mm, respectively. While filaments have not been
described before in P. acnes, slender or fine branching
filaments (5–20 mm or greater in length) are a character-
istic feature of Propionibacterium propionicum (previously
Arachnia propionica), along with short irregular rods of
variable length that are commonly arranged in pairs (Holt
et al., 1994). Unlike the atypical P. acnes isolates described
in this study, however, P. propionicum does not show any
reaction with the mAb QUBPa3 (A. McDowell & S.
Patrick, unpublished data). Filaments that are associated
with branching and hyphal formation are also character-
istic of other members of the Actinomycetales, such as the
Nocardiodes, and the less closely related Actinomyces and
Rhodococcus (Holt et al., 1994). It is interesting to note that
in addition to resembling Actinomyces israelii morpholo-
gically, P. propionicum can also cause actinomycosis, a
chronic, subacute suppurating granulomatous infection
(Hall, 2006). Whether these P. acnes strains also share any
similarities with P. propionicum and Actinomyces spp. in
terms of pathogenicity remains to be determined. Their
ability to form filaments may impact on the potential for
colonization, particularly in relation to tissue penetration
and biofilm formation. It is possible that type III isolates
have remained unidentified as P. acnes due to their atypical
morphology, as would be observed upon routine Gram
Nucleotide sequence and phylogenetic analysis
Systematic analysis of all five atypical P. acnes isolates was
initially conducted by direct nucleotide sequencing of a
1484 bp fragment of the 16S rRNA gene amplified with the
universal 16S rRNA-based primers UFPL and URPL. The
GenBank accession numbers for these 16S sequences are
listed in Table 1. Upon CLUSTAL W analysis, sequences
obtained for all the isolates had between 99.8 to 99.9%
identity to the previously published sequences for NCTC
KPA171202 (type IB, GenBank accession no. NC_006085)
and NCTC 10390 (type II, GenBank accession no.
AY642044), thus confirming the very close relationship
between these organisms and types I and II. In contrast,
when a 1473 bp sequence stretch of the 16S rRNA gene
accession no. AB042288)
Fig. 1. Micrographs of P. acnes (after growth on ABA plates)
immunolabelled with the mouse IgG mAb QUBPa3 and a FITC-
conjugated goat anti-mouse IgG antibody (magnification ?100):
(a) Asn10 (type III), (b) Asn12 (type III), (c) NCTC 737 (type IA),
(d) NCTC 10390 (type II).
Fig. 2. Micrograph of a large filamentous aggregate from the P.
acnes type III isolate Asn12 (after growth on an ABA plate).
Filaments were immunolabelled with the mouse IgG mAb QUBPa3
and a FITC-conjugated goat anti-mouse IgG antibody (magnifica-
A. McDowell and others
220Journal of Medical Microbiology 57
from all five isolates was aligned and compared with the
sequence from P. propionicum strain DSM 43307 (GenBank
accession no. AJ003058) much greater differences were
observed (95.8–95.9% identity). Although sequencing of
the 16S rRNA gene is still considered the ‘gold-standard’
for investigating the phylogenetic relationship between
bacterial organisms, potential problems associated with its
ability to resolve the relationships between closely related
species, due to an extremely low rate of neutral mutation,
have been recognized (Vandamme et al., 1996). In contrast,
protein-encoding genes with housekeeping functions, such
as recA, often provide a better foundation for bacterial
systematics and differentiation of closely related organisms
(Eisen, 1995; Mahenthiralingam et al., 2000). This is due to
a higher neutral mutation rate within such genes, which is
a consequence of the redundancy of the genetic code
resulting in synonymous substitutions at the third codon
position that have no effect on the function of the resulting
protein. Since we previously found that analysis of the recA
housekeeping gene provided much greater insight into the
phylogenetic relationship between P. acnes types I and II
compared to 16S rRNA (McDowell et al., 2005), it was a
natural and pertinent step to further investigate potential
phylogenetic differences between these five atypical isolates
of P. acnes and types I and II based on this locus.
A 1201 bp fragment containing the recA gene was
successfully amplified from all five atypical isolates, using
the previously described primers PAR-1 and PAR-2
(McDowell et al., 2005), and sequenced. The GenBank
accession numbers for these recA sequences are listed in
Table 1. Other closely related Propionibacterium species,
such as P. granulosum, do not show any reaction with this
particular primer pair (A. McDowell & S. Patrick,
unpublished data). Upon CLUSTAL W analysis, the recA
sequences (1047 bp) for these isolates showed distinct
differences from those previously published for NCTC 737
(GenBank accession no. AY642055, 98.9% identity),
KPA171202 (GenBank accession no. NC_006085, 98.9%
identity) and NCTC 10390 (GenBank accession no.
AY642061, 99.1% identity). Strains of P. acnes type I (IA
& IB) and II differ in 10 highly conserved regions within the
recA gene (99% identity). For our group of atypical isolates,
the recA gene sequences were found to contain a
combination of four type I-specific polymorphisms (com-
mon to IA and IB), six type II-specific polymorphisms and
five polymorphisms unique to the group (the sequence
alignments are shown in Supplementary Fig. S1 available
with the online journal). To investigate the phylogenetic
relationship between these five isolates and types I and II, a
recA phylogenetic tree was constructed based on the
nucleotide sequences (Fig. 3). Previously published recA
sequences for types IA, IB and II were included in the
analysis for comparison (McDowell et al., 2005). Strains of
P. acnes types I, II and this novel grouping formed highly
distinct branches within the tree supported by bootstrap
values of 98% or 100%. This demonstrates that the atypical
isolates identified by IFM represent a new or novel recA
phylogenetic cluster or lineage, which we propose as P. acnes
type III. To facilitate further studies of the type III grouping
the isolate Asn12 (recovered from intervertebral disc
material) will be deposited in the NCTC bacterial cell bank
as a representative of this novel phylogenetic cluster.
These results again highlight the potential limitations to the
use of the 16S rRNA locus alone in understanding bacterial
phylogeny. This may be especially important for studies
that attempt to identify novel bacterial groupings or
phylotypes within microbial communities, such as those
present on the skin, using culture-independent 16S rRNA-
based methods (Dekio et al., 2005). For such investigations
the use of protein-encoding genes with housekeeping
functions may provide more valuable information for
Table 1. Nucleotide sequence analysis of P. acnes isolates
with atypical morphology
Isolate GenBank 16S rRNA
Fig. 3. Unrooted phylogenetic tree of P. acnes based on the
complete recA gene sequence, illustrating the three recA lineages
of the organism. Multiple sequence alignments were performed on
the recA gene sequences from type III isolates and published
sequences representative of type IA, IB and II isolates (McDowell
et al., 2005) (withno gaps
Bootstrapping resampling statistics was performed using 100
datasets, with bootstrap values shown on the arms of the tree. The
type status for the different strains analysed is also shown.
in thealigned sequence).
A new phylogenetic group of P. acnes
certain taxa (Palys et al., 1997). Indeed, we have also
constructed an accurate phylogeny of P. acnes (congruent
with recA phylogeny) based on sequence analysis of
putative virulence genes (McDowell et al., 2005; Valanne
et al., 2005). This demonstrates that where rates of
sequence divergence are appropriate, non-housekeeping
gene sequences can also provide data that may be useful for
bacterial systematics. Although P. acnes isolates can be
assigned to the type III phylogenetic grouping based on
analysis of their cellular morphology (morphovars), we
would recommend sequencing of the recA gene for
unambiguous classification. On the basis of previous
DNA–DNA hybridization studies (Johnson & Cummins,
1972), types I and II would appear to represent distinct
sequence clusters within P. acnes rather than novel species,
at least based on current definitions of a bacterial species
(Vandamme et al., 1996; Cohan, 2002). Similar experi-
ments will now have to be carried out with isolates from
the type III grouping to confirm if this is also the case for
Unequivocal identification of the type III isolates as P.
acnes (99.9% identity) was observed on the basis of
analysis with the API ID32A and API 20A biochemical
galleries. All type III isolates were positive for indole and
nitrate reduction characteristic of P. acnes, but not P.
granulosum or P. avidum (Holt et al., 1994). The isolates
were negative for sorbitol and erythritol fermentation, but
could ferment ribose (biotype 4), as well as glucose and
glycerol. The production of acid from mannose was
variable between the isolates. In keeping with other P.
acnes strains, the type III isolates did not produce acid from
lactose, salicin, xylose, maltose, arabinose, cellobiose,
melezitose, raffinose or rhamnose, and were negative for
urease and b-glucosidase activity. All type III organisms
were positive for catalase activity, but arginine dihydrolase
activity was variable.
In addition to studies focused on understanding the
metabolic profile of type III organisms, we also investigated
possible differences in the production of various virulence
factors between all type III isolates and 95 P. acnes isolates
representing types I (n575) and II (n520) (Table 2). These
isolates, which were recovered from different sources at the
ROH and Queen Elizabeth Hospital, had been routinely
identified to the level of type I or II on the basis of sugar
fermentation profiles (Higaki et al., 2000), random
amplification of polymorphic DNA fingerprints (Perry
et al., 2003) and, for some samples, IFM (with QUBPa1
and QUBPa2) (McDowell et al., 2005). All type III isolates
were positive for lipase activity, but variable in their
production of proteinases and hyaluronidase. They were,
however, negative for lecithinase activity, as well as a- and
b-haemolytic activities. Type I isolates were found to be
variable for a- and b-haemolysis, while type II isolates were
negative for b-haemolysis, but did display variable a-
haemolytic activity. The production of lipase, lecithinase,
proteinases and hyaluronidase activities was variable for type
I and II isolates (Table 2). No DNase or elastase activity was
detected in any of the P. acnes isolates examined. PCR
analysis of type III isolates revealed the presence of the co-
haemolysin or CAMP factor gene family, as well as the tly
gene encoding a putative haemolysin/cytotoxin (not illu-
strated), which have already been described in types I and II
(Valanne et al., 2005; McDowell et al., 2005). Further studies
with a greater range of isolated type III strains will be
required to confirm all these observations. Also, continued
investigation of the virulence profile of the various P. acnes
types is warranted, especially as distinct phylogenetic groups
of an organism can display differences in their pathogenic
potential (Ishii et al., 2007).
In conclusion, we have identified a novel phylogenetic
grouping or lineage of P. acnes that we propose as type III.
The identification of a third phylogenetic cluster in P. acnes
further challenges our understanding of this organism,
highlights potential caveats in the use of only one isolate
type in laboratory studies of P. acnes virulence, and raises
the possibility that other phylogenetic groups of the
organism may exist.
We acknowledge Mr Gisli Einarsson and Dr Isaac Chen for technical
assistance. A.M. was funded by a programme grant from the
Northern Ireland Health and Personal Social Services Research and
Development Office (NIPHSS R and D Office). A.P. was funded by
the University Hospital, Birmingham NHS Trust.
Table 2. Production of virulence factors by P. acnes types I, II and novel group III
Phylogenetic group No. of isolates producing each putative virulence factor (percentage total of each type)*
*All isolates negative for DNase and elastase activity.
A. McDowell and others
222 Journal of Medical Microbiology 57
Ahn, C. Y., Ko, C. Y., Wagar, E. A., Wong, R. S. & Shaw, W. W. (1996).
Microbial evaluation: 139 implants removed from symptomatic
patients. Plast Reconstr Surg 98, 1225–1229.
Aldave, A. J., Stein, J. D., Deramo, V. A., Shah, G. K., Fischer, D. H. &
Maguire, J. I. (1999). Treatment strategies for postoperative Propioni-
bacterium acnes endophthalmitis. Ophthalmology 106, 2395–2401.
Balke, E. & Weiss, R. (1984). A simple method for rapid detection of
bacterial hyaluronidase in K hyaluronate-containing gel. Zentralbl
Bakteriol Mikrobiol Hyg [A] 257, 317–322 (in German).
Propionibacterium species. Rev Infect Dis 13, 819–822.
I.& Frazier,E. H.(1991).
Bruggemann, H., Henne, A., Hoster, F., Liesegang, H., Wiezer, A.,
Strittmatter, A., Hujer, S., Durre, P. & Gottschalk, G. (2004). The
complete genome sequence of Propionibacterium acnes, a commensal
of human skin. Science 305, 671–673.
Clark, W. L., Kaiser, P. K., Flynn, H. W., Belfort, A., Miller, D. & Meisler,
D. M. (1999). Treatment strategies and visual acuity outcomes in
chronic postoperative Propionibacterium acnes endophthalmitis.
Ophthalmology 106, 1665–1670.
Cohan, F. M. (2002). What are bacterial species? Annu Rev Microbiol
Cohen, R. J., Shannon, B. A., McNeal, J. E., Shannon, T. & Garrett,
K. L. (2005). Propionibacterium acnes associated with inflammation in
radical prostatectomy specimens: a possible link to cancer evolution?
J Urol 173, 1969–1974.
Debelian, G. J., Olsen, I. & Tronstad, L. (1992). Profiling of
Propionibacterium acnes recovered from root canal and blood during
and after endodontic treatment. Endod Dent Traumatol 8, 248–254.
Dekio, I., Hayashi, H., Sakamoto, M., Kitahara, M., Nishikawa, T.,
Suematsu, M. & Benno, Y. (2005). Detection of potentially novel
bacterial components of the human skin microbiota using culture-
independent molecular profiling. J Med Microbiol 54, 1231–1238.
Delahaye, F., Fol, S., Celard, M., Vandenesch, F., Beaune, J., Bozio, A.
& de Gevigney, G. (2005). Propionibacterium acnes infective endo-
98, 1212–1218 (in French).
Eady, E. A. & Ingham, E. (1994). Propionibacterium acnes – friend or
foe? Rev Med Microbiol 5, 163–173.
Eisen, J. A. (1995). The RecA protein as a model molecule for
molecular systematic studies of bacteria: comparison of trees of RecAs
and 16S rRNAs from the same species. J Mol Evol 41, 1105–1123.
Eishi, Y., Suga, M., Ishige, I., Kobayashi, D., Yamada, T., Takemura, T.,
Takizawa, T., Koike, M., Kudoh, S. & other authors (2002).
Quantitative analysis of mycobacterial and propionibacterial DNA in
lymph nodes of Japanese and European patients with sarcoidosis. J Clin
Microbiol 40, 198–204.
Hall, V. (2006). Anaerobic actinomycetes and related organisms. In
Principles and Practice of Clinical Bacteriology, pp. 575–586. Edited by
S. H. Gillespie & P. M. Hawkey. Chichester: Wiley.
Higaki, S., Kitagawa, T., Kagoura, M., Morohashi, M. & Yamagishi, T.
(2000). Correlation between Propionibacterium acnes biotypes, lipase
activity and rash degree in acne patients. J Dermatol 27, 519–522.
Holt, J. G., Krieg, N. R., Sneath, P. H. A., Staley, J. T. & Williams, S. T.
(editors) (1994). Bergey’s Manual of Determinative Bacteriology, 9th
edn. Baltimore, MD: Williams and Wilkins.
Horgan, S. E., Matheson, M. M., McLoughlin-Borlace, L. & Dart, J. K.
(1999). Use of a low nutrient culture medium for the identification of
bacteria causing severe ocular infection. J Med Microbiol 48, 701–703.
Ishii, S., Meyer, K. P. & Sadowsky, M. J. (2007). Relationship between
phylogenetic groups, genotypic clusters and virulence gene profiles of
Escherichia coli strains from diverse human and animal sources. Appl
Environ Microbiol 73, 5703–5710.
Johnson, J. L. & Cummins, C. S. (1972). Cell wall composition and
deoxyribonucleic acid similarities among anaerobic coryneforms,
classical propionibacteria, and strains of Arachnia propionica.
J Bacteriol 109, 1047–1066.
Kishishita, M., Ushijima, T., Ozaki, Y. & Ito, Y. (1979). Biotyping of
Propionibacterium acnes isolated from normal human facial skin. Appl
Environ Microbiol 38, 585–589.
Kotilainen, P., Merilahti-Palo, R., Lehtonen, O. P., Manner, I.,
Helander, I., Mottonen, T. & Rintala, E. (1996). Propionibacterium
acnes isolated from sternal osteitis in a patient with SAPHO
syndrome. J Rheumatol 23, 1302–1304.
Le Goff, A., Bunetel, L., Mouton, C. & Bonnaure-Mallet, M. (1997).
Evaluation of root canal bacteria and their antimicrobial susceptibility
in teeth with necrotic pulp. Oral Microbiol Immunol 12, 318–322.
LiPuma, J. J., Dulaney, B. J., McMenamin, J. D., Whitby, P. W., Stull,
T. L., Coenye, T. & Vandamme, P. (1999). Development of rRNA-
based PCR assays for identification of Burkholderia cepacia complex
isolates recovered from cystic fibrosis patients. J Clin Microbiol 37,
Mahenthiralingam, E., Bischof, J., Byrne, S. K., Radomski, C., Davies,
J. E., Av-Gay, Y. & Vandamme, P. (2000). DNA-based diagnostic
approaches for identification of Burkholderia cepacia complex,
Burkholderia vietnamiensis, Burkholderia multivorans, Burkholderia
stabilis, and Burkholderia cepacia genomovars I and III. J Clin
Microbiol 38, 3165–3173.
McDowell, A., Valanne, S., Ramage, G., Tunney, M. M., Glenn, J. V.,
McLorinan, G. C., Bhatia, A., Maisonneuve, J. F., Lodes, M. & other
authors (2005). Propionibacterium acnes types I and II represent
phylogenetically distinct groups. J Clin Microbiol 43, 326–334.
of cutaneous propionibacteria. Appl Environ Microbiol 35, 62–66.
Palys, T., Nakamura, L. K. & Cohan, F. M. (1997). Discovery and
classification of ecological diversity in the bacterial world: the role of
DNA sequence data. Int J Syst Bacteriol 47, 1145–1156.
Perry, A.L.,Worthington, T.,Hilton, A.C.,Lambert, P.A.,Stirling, A.J. &
Elliott, T. S. J. (2003). Analysis of clinical isolates of Propionibacterium
acnes by optimised RAPD. FEMS Microbiol Lett 228, 51–55.
Spare, M. K., Tebbs, S. E., Lang, S., Lambert, P. A., Worthington, T.,
Lipkin, G. W. & Elliott, T. S. J. (2003). Genotypic and phenotypic
properties of coagulase negative staphylococci causing dialysis
catheter related sepsis. J Hosp Infect 54, 272–278.
Thompson, J. D., Higgins, D. G. & Gibson, T. J. (1994). CLUSTAL W:
improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, position-specific gap penalties and
weight matrix choice. Nucleic Acids Res 22, 4673–4680.
Tunney, M. M., Patrick, S., Gorman, S. P., Nixon, J. R., Anderson, N.,
Davis, R. I., Hanna, D. & Ramage, G. (1998). Improved detection of
infection in hip replacements. J Bone Joint Surg Br 80, 568–572.
Tunney, M. M., Patrick, S., Curran, M. D., Ramage, G., Hanna, D.,
Nixon, J. R., Gorman, S. P., Davis, R. I. & Anderson, N. (1999).
Detection of prosthetic hip infection at revision arthroplasty by
immunofluorescence microscopy and PCR amplification of the
bacterial 16S rRNA gene. J Clin Microbiol 37, 3281–3290.
Valanne, S., McDowell, A., Ramage, G., Tunney, M. M., Einarsson,
G. G., O’Hagan, S., Wisdom, G. B., Fairley, D., Bhatia, A. & other
authors (2005). CAMP factor homologues in Propionibacterium
acnes: a new protein family differentially expressed by types I and II.
Microbiology 151, 1369–1379.
A new phylogenetic group of P. acnes
Vandamme, P., Pot, B., Gillis, M., Devos, P., Kersters, K. & Swings, J.
(1996). Polyphasic taxonomy, a consensus approach to bacterial
systematics. Microbiol Rev 60, 407–438.
Webster, G. F. & Cummins, C. S. (1978). Use of bacteriophage typing
to distinguish Propionibacterium acnes types I and II. J Clin Microbiol
A. McDowell and others
224 Journal of Medical Microbiology 57