Conserved genes in a path from commensalism to pathogenicity: comparative phylogenetic profiles of Staphylococcus epidermidis RP62A and ATCC12228
- Wu Wei†1, 2,
- ZhiWei Cao†2,
- Yu-Li Zhu3,
- XiaoJing Wang1, 2,
- GuoHui Ding1,
- Hao Xu2,
- PeiLin Jia1, 2,
- Di Qu3Email author,
- Antoine Danchin4Email author and
- YiXue Li1, 2Email author
© Wei et al; licensee BioMed Central Ltd. 2006
Received: 01 March 2006
Accepted: 10 May 2006
Published: 10 May 2006
Staphylococcus epidermidis, long regarded as an innocuous commensal bacterium of the human skin, is the most frequent cause of nosocomial infections associated with implanted medical devices. This conditional pathogen provides a model of choice to study genome landmarks correlated with the transition between commensalism and pathogenicity. Traditional investigations stress differences in gene content. We focused on conserved genes that have accumulated small mutation differences during the transition.
A comparison of strain ATCC12228, a non-biofilm forming, non-infection associated strain and strain RP62A, a methicillin-resistant biofilm clinical isolate, revealed consistent variation, mostly single-nucleotide polymorphisms (SNPs), in orthologous genes in addition to the previously investigated global changes in gene clusters. This polymorphism, scattered throughout the genome, may reveal genes that contribute to adaptation of the bacteria to different environmental stimuli, allowing them to shift from commensalism to pathogenicity. SNPs were detected in 931 pairs of orthologs with identical gene length, accounting for approximately 45% of the total pairs of orthologs. Assuming that non-synonymous mutations would mark recent evolution, and hence be associated to the onset of the pathogenic process, analysis of ratios of non-synonymous SNPs vs synonymous SNPs suggested hypotheses about possible pathogenicity determinants. The N/S ratios for virulence factors and surface proteins differed significantly from that of average SNPs. Of those gene pairs, 40 showed a disproportionate distribution of dN vs dS. Among those, the presence of the gene encoding methionine sulfoxide reductase suggested a possible involvement of reactive oxygen species. This led us to uncover that the infection associated strain was significantly more resistant to hydrogen peroxide and paraquat than the environmental strain. Some 16 genes of the list were of unknown function. We could suggest however that they were likely to belong to surface proteins or considered in priority as important for pathogenicity.
Our study proposed a novel approach to identify genes involved in pathogenic processes and provided some insight about the molecular mechanisms leading a commensal inhabitant to become an invasive pathogen.
Emerging diseases have been a matter of great concern recently: the sudden appearance of SARS and the recent spread of bird's flu raise the question of how microbes evolve into pathogens to human not only an important but also an urgent question to answer. Most microbes have a benign symbiotic relationship with humans and only cause infections to healthy individuals under limited conditions. Staphylococcus epidermidis is such an opportunistic pathogen which is a common member of the normal flora of our skin and mucous membranes. But on certain occasions, the presence of S. epidermidis as a contaminant of medical devices, or breach of the skin by trauma or inoculation needles, makes it emerge as a causative agent of infections . As biomedical devices are increasingly used in medical practice, a major complication due to S. epidermidis infection when using these devices is affecting several millions of patients worldwide each year .
The pathogenic process of foreign-body-associated infections with S. epidermidis is characterized by the ability of this species to colonize polymer surfaces by the formation of multilayered cell clusters, which are enveloped and protected by an amorphous slimy material to form a biofilm [3–5]. Most genetic and biochemical evidence has shown that polysaccharide intercellular adhesion (PIA) production , mediated by the icaABCD operon [7–9], is crucial in biofilm fomation. S. epidermidis does not produce components that are easily recognized as virulence factors, such as toxins or aggressive degradative exoenzymes . Genetic manipulation of S. epidermidis has been difficult so far, limiting efforts that would elucidate the molecular basic of its pathogenicity. The analysis of bacterial genome sequences provides us with an alternative way to investigate some of the constraints that might pave the way from commensalism to pathogenicity in this species.
Staphylococcus epidermidis is one of the most commonly isolated bacterial pathogens in hospitals because of its large numbers and ubiquitous distribution. Environmental S. epidermidis strains differ significantly in their invasive and ability to form biofilm. Two complete genomes of S. epidermidis strains have been sequenced, one is ATCC 12228, a non-biofilm forming/non-infection associated strain, and another one is RP62A (also called ATCC35984), an infectious and biofilm forming strain [11, 12]. Both strains are highly similar in genome sequence, and we tried to see whether some particular features might be correlated with the surviving ability of the organisms in certain environment, allowing us to uncover processes that might be important for the transition between a commensal life style, and a pathogenic behavior. In previous comparative genomics studies, the most obvious differences between these genome pairs revealed the loss or gain of large DNA segments . Further comparative studies of S. epidermidis with other staphylococcal species indicated that the majority of the genes unique to a given strain or species is also related to the presence or absence of prophages and genomic islands. In the absence of other criteria to define virulence, some of these genes were proposed to be the main causes of pathogenicity and virulence . However, acquisition of those genes has repeatedly been reported to be insufficient to trigger a pathogenic response . For example, it has been discovered that the ATCC12228 and RP62A strains share most of the possible pathogenic factors. As a case in point, the cap operon encoding the polyglutamate capsule and being recognized as a major virulence factor in Bacillus anthracis, has been found to be integrated into the genomes of both S. epidermidis RP62A and ATCC12228, while the latter is a non-infectious strain [12, 15]. In addition to gross differences displayed as genomic islands and unique genes, Single Nucleotide Polymorphisms (SNPs) have also been found to widely exist in the chromosomes of the two strains, especially in some genes with cell envelope functions . Therefore, pathogenicity appears to be a complex phenomenon which could be accounted for not only by large genetic differences, but also by small variations in gene contents. This prompted us to perform a detailed SNPs comparative analysis between S. epidermidis ATCC12228 and RP62A, complementing the previous approaches used to differentiate pathogenic strains from non-pathogenic ones [16–18]. Assuming that pathogens have evolved recently , the amount of small genetic variations that would be accumulated in the genome sequences might not have had time to undergo purifying selection, thus revealing some of the genes that are important for pathogenic processes. Analyzing those variations, mostly SNPs, we expected to find genes bearing landmarks of recent evolution, and which may contribute to the pathogenesis of the bacterium. In this paper, we focused on the systematic SNP analysis between a pair of infectious and non-infectious strains of S. epidermidis, in particular by comparing synonymous and non-synonymous mutations in orthologous gene pairs. The genes displaying higher dN/dS rate, including genes of unknown function, were analyzed in relation with function possibly associated to pathogenicity.
Small variations between the genomes of two S. epidermidis strains
The S. epidermidis RP62A and ATCC12228 chromosomes are 2,616,530 bp and 2,499,279 bp long, respectively [11, 12]. Beside large segments coding for genes present in one genome and absent in the other, many small scale genetic variations which affected an individual gene or a small numbers of genes (< 10 CDSs) were dispersed throughout nearly all of both chromosomes, revealing extensive divergence between both species. The comparison of the insertions and deletions (indels) differentiating the two genome sequences showed that such events generally involved small-scale variations (see Additional file 1). In addition to indels, a total of 10,297 SNPs were found in the genome of S. epidermidis ATCC12228 when compared to that of strain RP62A . These small-scale variations, mostly SNPs, might contribute to the different phenotypes of the two strains. We therefore focused on the SNPs of the orthologous genes to explore how they could contribute to adaptation of the bacteria to various environmental stimuli allowing them to shift from commensalism to pathogenicity.
Comparison of the CDSs of S. epidermidis ATCC 12228 and S. epidermidis RP62A
S. epidermidis ATCC12228a
S. epidermidis RP62Aa
Total CDS number
CDSs categorized by sequence variations
Orthologs with identical DNA sequences
Orthologs with SNPs and the same lengthb
Orthologs with insertions/deletions
Unique compared to the other S. epidermidisc
Unique compared to all staphylococcal
Pathogenicity-related genes and highly expressed proteins genes evolve differently
Comparison of SNP frequency and distribution into different functional groups
P I c
Genes potentially involved in pathogenicity
Non-synonymous substitutions significant group (p < 0.10) of Orthologs with SNP Pairs potential involves in pathogenicity
S. epidermidis ATCC12228
S. epidermidis RP62A
Annotation from ATCC12228
fosfomycin resistance protein fofB
peptide methionine sulfoxide reductase
conserved hypothetical protein
glycerol uptake facilitator
probable ATP-dependent DNA helicase RecQ
choline transport ATP-binding protein
conserved hypothetical protein
conserved hypothetical protein
YheB (surface protein)
glycine betaine aldehyde dehydrogenase gbsA
DNA topoisomerase III topB
50S ribosomal protein L3
30S ribosomal protein S12
putative deoxyribodipyrimidine photolyase
glycine betaine aldehyde dehydrogenase gbsA
conserved hypothetical protein
conserved hypothetical protein
YqgE (MultiDrug Efflux proteins)
phenol soluble modulin beta 1
conserved hypothetical protein
YkyA (surface protein)
conserved hypothetical protein
dihydropteroate synthase chain A synthetase
conserved hypothetical protein
conserved hypothetical protein
conserved hypothetical protein
In larger Bacterial clades
low-affinity inorganic phosphate transporter
conserved hypothetical protein
conserved hypothetical protein
conserved hypothetical protein
Although the sequencing technique is quite accurate, the error frequency in a finished sequence is thought to be one error (frameshift or base substitution) in 103 to 105 bases . To confirm the SNP sites identified in this study, we chose the top 13 pairs of mostly non-synonymous substitutions proteins and carried out PCR experiments on their cognate gene to validate the corresponding sequences. Sequences of the PCR products confirmed all the SNP sites we identified from the orthologs analysis.
Finally, genes are often grouped in operons, forming functionally consistent transcription units. If our hypothesis for the detection of genes important for pathogenesis is correct, then we should expect that a mutation bias would generally span entire operons when the corresponding genes participate to the process of pathogenicity. We found that genes making potential operons often displayed a collective increase in non-synonymous mutations (see Additional file 5). This is the case, for example, of potA (se0797) and potB (se0798), which code for the spermidine/putrescine ABC transporter.
While an important topic for human health, pathogenicity is not the most prevalent development process for living organisms. Often, this particular way of life happens as a new ecological niche (a possible host) is colonized by organisms that were previously indifferent or simple commensals. Previous work concluded, in general, that genome segments, often named pathogenicity islands, were the landmarks of pathogenic processes . These islands are generally supposed to be the result of Horizontal Gene Transfer (HGT). HGT is a general process of gene acquisition by bacteria, where it may be associated to control of the background mutation level . In Firmicutes, HGT is often the result of phage infection . Bacteriophages can subsequently remain in the genome as functional (SPbeta in B. subtilis, for example) or more or less defective prophages (PBX, Skin, and similar elements). In the case of S. epidermidis, this type of analysis led Gill et al. to conclude that HGT was the major contribution to pathogenicity in the pathogenic strain they analyzed . However, acquisition of virulence genes has repeatedly been found to be insufficient to trigger a pathogenic response. As a case in point, the cap operon, encoding the presumably virulence-associated polyglutamate capsule is present in both the commensal and pathogenic S. epidermidis strains . The absence/presence of the ica operon that produces a biofilm exopolysaccharide was taken as the clearest genetic difference contributing to the different phenotype of two S. epidermidis strains [8, 9]. However, the prevalence of icaADBC in commensal strains did not differ from that in invasive strains, indicating that other factors should been involved in pathogenesis . All this indicates that important information about the development of pathogenic processes has been overlooked by the standard approaches focusing on HGT analysis. The small-scale variations which pepper the chromosomes of the commensal and pathogenic strains were not explored in detail. In order to gain further insight into the pathogenicity process, we studied the Single Nucleotide Polymorphisms (SNPs) between the orthologous gene pairs of the pathogenic and the commensal S. epidermidis strains of interest.
SNP analysis rests obviously on the quality of the genome sequences determination. The error frequency in a finished sequence has rarely been precisely measured. A common belief is that anywhere between one error in 103 to 105 bases, depending on the project. In order to evaluate the impact of sequencing errors on our analysis, approximately 23 kb of DNA sequence was re-sequenced after PCR amplification of 13 pairs of orthologs from both strains. 11 new SNP sites (5 and 6 in strains ATCC12228 and RP62A, respectively) were retrieved when the new sequences were compared to published genome sequences. If we take those new sites as sequencing errors, and not PCR errors or recent mutations, then the average error frequency is 4.8 errors in 104 bases. This shows that, while we may have missed some interesting genes, our approach is validated with the present available quality of genome sequences.
Analysis of the ratios of total non-synonymous SNPs vs total synonymous SNPs showed that N/S ratios for virulence factors and surface proteins differed significantly from that of total SNPs. While random mutations and evolutionary drift should not favor one type of mutation over the other one, genes related to pathogenic processes may have evolved recently due to positive selection to fit the pathogenic conditions . Their evolution could be compared to that of genes belonging to translation, ribosomal structure and biogenesis (labeled according to the COG classification), which are submitted to purifying selection because of the general optimization of the translation process in the course of evolution. As expected, the latter were mostly submitted to synonymous mutations. This is remarkable and consistent with the interpretation of Drummond and colleagues who argued that highly expressed proteins evolve slowly because of the severe selective pressure on highly expressed proteins to avoid misfolding even when they are mistranslated . This interpretation further supports the hypothesis that non synonymous mutations may be the landmark of proteins involved in pathogenic processes.
The percentages of identical vs. non-identical genes among the orthologs of those groups (virulence factors, surface proteins, translation-related genes and total genes) were also compared (see Additional file 9). Interestingly, the ratio of non-identical vs. identical genes coding for surface proteins differs significantly from that in translation-related genes and total genes, following the same trend as the ratio of non-synonymous vs. synonymous substitutions. This is not the case for virulence factors. The percentage of non-identical vs. identical genes reflected differences at the nucleic acids level, while the non-synonymous vs. synonymous differences emphasized the DNA mutations that result in a change protein sequences. Proteins are the ultimate functional entities in an organism, and this makes it reasonable to focus on the DNA mutations that change the coded amino acids, rather than silent mutations.
Of the 931 SNPs orthologous pairs identified in this study, a total of 40 genes showed a disproportionate distribution of dN/dS. Most are probably significant, as we consistently found that when present the imbalanced dN/dS distribution extended to genes common to an operon. That they may play important roles in pathogenic processes is suggested by the presence in the list of genes coding for proteins that are likely to influence pathogenicity, such as phenol-soluble modulin (PSM) family peptides (SE0847) or the Clp protease (SE0674). PSMs, a class of surfactant peptides with proinflammatory and putative biofilm-inhibitory properties, presumably represent key factors controlling the switch between the colonization and disseminative stages of the pathogen. Suppression of production of PSMs in the biofilm stage enables cells to stick together and to evade the host immune defense . ATP-dependent Clp proteases are involved in regulation processes by proteolysis in many bacteria. They consist of a proteolytic subunit, ClpP, which confers substrate specificity through association with ATPase subunits. Signature-tagged mutagenesis screening experiments have demonstrated that ATP-dependent proteases are key factors in bacterial adaptation to environmental stress, including ROS . Little is known about clp gene regulation in pathogenic bacteria, despite the fact that many of these genes play important roles in virulence, such as ClpX in Staphylococcus aureus  and ClpB in Leishmania sp. . It seems plausible that these bifunctional chaperones/proteases may also modulate the activity of other virulence factors  and it is therefore interesting to have observed that a member of this class is evolving fast in the S. epidermidis pair studied here. Genes involved in osmoprotection which may respond to high salt conditions in cell agglomerations on human skin, such as gbsA (glycine betaine aldehyde dehydrogenase) were also found among the non-synonymous significant genes.
Surface proteins play a fundamental role in the interaction between the bacterial cell and its environment. In the present case, several of the 40 non-synonymous genes, in particular some genes with unknown function, have features suggesting that they code for surface proteins under the pressure of escaping the host immune system. Interestingly, SE1527 and SE0790 are orthologous to surface proteins experimentally identified in one of Group A Streptococcus (GAS) strains (SPY0792 and SPY2018) . It may also be significant that two ribosomal proteins, for which there is also some evidence of extracellular functions, were also found in their surface proteome. This makes RpsL (SE0309) and RplC (SE1824), the two ribosomal proteins which have more non-synonymous substitutions than expected in our analysis more interesting. RpsL, protein S12 of the small subunit of the ribosome is known to control accuracy of translation, and this might indeed play a role in pathogenicity . Further work is required to investigate this aspect as we can expect that some of the genes identified are simply the result of random deviation from the average.
Yao et al have used microaray-based genome-wide comparison of clinical and commensal strains to identify putative virulence factors in S. epidermidis . 39 genes were found to be more frequent among clinical strains than commensal strains. Interestingly, we find that after detailed comparison our 40 high-non-synonymous substitutions genes have no overlap with those 39 high-frequent genes in clinical strains. This is probably not unexpected, for the following reasons. Firstly, we looked for conserved genes rather than genes that would differ in a variety of strains, and our study involves only orthologous pairs of genes in the two stains, while Yao et al focused on those genes which showed disproportionate difference in distribution between clinical and commensal strains. Second, our 40 high non-synonymous substitution genes were identified because they carry SNPs, small differences which may not be detected by hybridization with the oligonucleotides of the microarray.
Under laboratory growth conditions, with plenty of metabolites available, bacterial gene expression is typically dominated by the highly expressed genes involved in transcription, protein biosynthesis, maturation and folding. In contrast, as shown for example in a study by Rollenhagen and Bumann, most of the genes highly expressed in Salmonella enterica cells recovered from the caecum differ considerably from those highly expressed in bacteria located in the spleen . The overall functional profile of highly expressed genes suggests a marked shift in transcriptional activity upon change in growth environment. In this respect, some pathogenicity related genes are likely to be highly expressed during infection to fight against the host. Interestingly, we found that one of the non-synonymous significant genes of the present study, gene se2269 encoding dihydropteroate synthase and recovered as a virulence-related antigen of a Gram-positive fish pathogen , was significantly more expressed in RP62A than in ATCC12228 (unpublished data).
Since pathogenicity is a recent form of lifestyle for an organism in evolution, we expected that genes involved in the process would show a significant mutation pattern . Furthermore, we could expect that many of the corresponding mutations would not yet be adapted to the "self" of the bacteria, as they would result from non-synonymous mutations in codons, leading to alteration of the polypeptide sequence of the corresponding genes. Our observations substantiated this hypothesis, showing a significant number of genes, often grouped in operons, affected by non-synonymous mutations. In parallel – and this can be seen as an internal validation of our hypothesis – we observed that genes of the translation machinery, expected to be expressed at a high level (and hence under a "mutator" pressure) were indeed mutated, but that the corresponding mutations were usually of the synonymous class. This is easily accounted for by the enormous selection pressure operating on the translation machinery (made of proteins interacting with each other and with highly conserved RNAs), which cannot easily accommodate mutations resulting in alteration of the corresponding proteins. Even if important for pathogenicity, these genes would become invisible because the functional constraints would result in a purifying selection process. However not all genes can be equally submitted to purifying selection and our approach reveals several genes of the general cell machinery, as possibly important for pathogenicity. As a case in point, we found one gene (se1042) coding for peptide methionine sulfoxide reductase, involved in protection against ROS, which could probably be such an important candidate for adaptation to pathogenicity [24, 25]. As the clpB gene found in our non-synonymous set also may point to adaptation to ROS, we explored the differential resistance to H2O2 and paraquat of the environmental species as compared to the pathogen (Figure 2 and 3, see Additional file 6 and file 7) and our experiments substantiated the interest of our approach to identify specific pathways of evolution to pathogenicity. In the same way a (presumably divalent) cation transporter SE0773 might be involved in the scavenging of divalent metals such as magnesium, manganese, iron or cobalt during infection. Gene se2128 might code for polyphosphate-AMP phosphotransferase, an observation that would support interest for the poorly explored role of polyphosphate in cells and its possible involvement in pathogenicity . Others are involved in biofilm formation, an important contribution to virulence in several pathogenic bacteria , as they considerably limit the success of both antibiotic treatment and the human immune defense. Gene expression profiling of the S. epidermidis biofilm was analyzed and some significant metabolic shift was found between the planktonic and the biofilm modes of growth .
In this paper, a SNPs comparative approach was developed to identify conserved genes possibly involved in pathogenicity by measuring their selective pressure in the gene pairs of non-infection associated S. epidermidis strain ATCC12228 and biofilm-forming strain RP62A. Our approach identified new genes that may be involved in pathogenesis, including some genes with unknown function. These results may provide fresh insight into discovering the genes that determine the success of S. epidermidis as an opportunistic pathogen. Complementing the previous methodologies which mainly focused on horizontal gene transfer, extensive SNPs investigations on more pathogens will facilitate our understanding of the path from commensalism to pathogenicity, a crucial prerequisite for designing therapeutic interventions directed to control pathogen infections.
Orthologous genes identification
For the identification of orthologous genes, all predicted CDSs from the S. epidermidis RP62A and ATCC12228 genomes were searched against each other locally using BLASTP . Those genes that matched a non-self genomic sequence at P value of < = 10-5, and identity > = 35%, matching at least 75% of the length of both query and subject sequences were considered homologous (non strain-specific genes). Of these homologous pairs, the bidirectional-best match was defined as an orthologous group. Identical nucleic acid sequence pairs, SNPs and insertion/deletions of the genes' pairs of orthologs between the two S. epidermidis genomes were identified using BLASTN.
The Nei-Gojobori (NG) method  was applied for estimating dS and dN of the SNPs pairs of S. epidermidis strains with minor modifications. The ratio of transition to transversion changes (R) was obtained by counting the total numbers of transitions to transversions that were observed in the entire set of orthologous pairs of the two species and adjusted using Kimura's 2-parameters method . The adjusted R computed from our data was 2.330294 and used in the following computation. We also applied the one-tail Z (or t∞) test to conduct the statistical test of the positiveness of dN – dS. The estimated dS and dN, standard error (S.E.), Z-scores and their corresponding p-values were presented in the additional file 5. To explore the distribution of dN/dS, we applied the transformation f (k) = log(k + 0.001) as described in other works[20, 47].
H2O2 and paraquat sensitivity analysis
We grew the S. epidermidis strains at different concentration of H2O2 and measured the growth curve of the strains by reading OD595 every one hour. Overnight bacterial cultures at (OD595 approximately 1.2) were added to fresh TSB(Tryptone Soya Broth) at a 1:100 dilution, and different concentrations of H2O2 (0, 1%, 5%, 10% and 15%) were added at the same time. Cells were grown at 37°C, agitated at, 220 rpm and the absorbancy of the culture (OD595) was followed every one hour.
MIC (minimal inhibitory concentration) assay of paraquat (sigma) by broth dilution was performed according to NCCLS (National Committee for Clinical Laboratory Standards) . The bacterial inoculum was prepared using a 3–4 h broth culture of each isolates adjusted to a turbidity equivalent to a 0.5 McFarland standard, diluted in CAMHB to achieve a final concentration of 5 × 105 CFU/ml in the test tube. Broth not containing an antimicrobial agent is inoculated as a control for organism viability, and E.coli ATCC 25922 was used as a test quality control strain. The bacteria were cultured in MH broth (OXOID) with different concentration of paraquat (from 1 mM to 8 mM) and incubated at 35°C for 16–20 h. The MIC was defined as the lowest concentration of paraquat giving complete inhibition of visible growth. The growth of bacteria incubated with lower concentration paraquat than MIC was measured by OD600.
In order to get PCR fragments containing SNPs, we designed primers by Primer Premier 5.0. The primers sequences are listed in Additional file 8. We used Pfu DNA polymerase (Tiangen Biotechnology Co., Ltd) to ensure high-fidelity synthesis. Amplifications were carried out in a thermocycler (GeneAmp PCR system 9700) through the following temperature program: 1 cycle of 5 min at 94°C; 30 cycles of 30 s at 94°C, 60 s at 55°C, and 60 s at 72°C; and finally 1 cycle at 72°C for 7 min. The PCR products were purified by agarose gel DNA purification kit (TaKaRa Biotechnology Co., Ltd) and sequenced by Shanghai Invitrogen Biotechnology Co., Ltd.
number of non-synonymous substitutions per site
number of synonymous substitutions per site
Coding DNA sequence
We thank Zhongming Zhao for comments on the manuscript and useful discussion. This work was supported by Key Program of Basic Research of Shanghai (No. 04QMX1450, 03XD14018, 02DJ14002), the 863 Hi-Tech Program of China (No. 2003AA231011, 2004BA711A21, 2004AA223080) and the State Key Program of Basic Research of China (No. 2003CB715900, 2004CB518606, 2002CB512803). AD acknowledges the Blastsets program for support.
- von Eiff C, Peters G, Heilmann C: Pathogenesis of infections due to coagulase-negative staphylococci. Lancet Infect Dis. 2002, 2 (11): 677-685. 10.1016/S1473-3099(02)00438-3.PubMedView ArticleGoogle Scholar
- Mack D, Becker P, Chatterjee I, Dobinsky S, Knobloch JK, Peters G, Rohde H, Herrmann M: Mechanisms of biofilm formation in Staphylococcus epidermidis and Staphylococcus aureus: functional molecules, regulatory circuits, and adaptive responses. Int J Med Microbiol. 2004, 294 (2–3): 203-212. 10.1016/j.ijmm.2004.06.015.PubMedView ArticleGoogle Scholar
- Peters G, Locci R, Pulverer G: Microbial colonization of prosthetic devices. II. Scanning electron microscopy of naturally infected intravenous catheters. Zentralbl Bakteriol Mikrobiol Hyg [B]. 1981, 173 (5): 293-299.Google Scholar
- Christensen GD, Simpson WA, Bisno AL, Beachey EH: Adherence of slime-producing strains of Staphylococcus epidermidis to smooth surfaces. Infect Immun. 1982, 37 (1): 318-326.PubMedPubMed CentralGoogle Scholar
- Gotz F: Staphylococcus and biofilms. Mol Microbiol. 2002, 43 (6): 1367-1378. 10.1046/j.1365-2958.2002.02827.x.PubMedView ArticleGoogle Scholar
- Heilmann C, Gerke C, Perdreau-Remington F, Gotz F: Characterization of Tn917 insertion mutants of Staphylococcus epidermidis affected in biofilm formation. Infect Immun. 1996, 64 (1): 277-282.PubMedPubMed CentralGoogle Scholar
- Heilmann C, Schweitzer O, Gerke C, Vanittanakom N, Mack D, Gotz F: Molecular basis of intercellular adhesion in the biofilm-forming Staphylococcus epidermidis. Mol Microbiol. 1996, 20 (5): 1083-1091.PubMedView ArticleGoogle Scholar
- Galdbart JO, Allignet J, Tung HS, Ryden C, El Solh N: Screening for Staphylococcus epidermidis markers discriminating between skin-flora strains and those responsible for infections of joint prostheses. J Infect Dis. 2000, 182 (1): 351-355. 10.1086/315660.PubMedView ArticleGoogle Scholar
- Li H, Xu L, Wang J, Wen Y, Vuong C, Otto M, Gao Q: Conversion of Staphylococcus epidermidis strains from commensal to invasive by expression of the ica locus encoding production of biofilm exopolysaccharide. Infect Immun. 2005, 73 (5): 3188-3191. 10.1128/IAI.73.5.3188-3191.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Vuong C, Otto M: Staphylococcus epidermidis infections. Microbes Infect. 2002, 4 (4): 481-489. 10.1016/S1286-4579(02)01563-0.PubMedView ArticleGoogle Scholar
- Zhang YQ, Ren SX, Li HL, Wang YX, Fu G, Yang J, Qin ZQ, Miao YG, Wang WY, Chen RS, Shen Y, Chen Z, Yuan ZH, Zhao GP, Qu D, Danchin A, Wen YM: Genome-based analysis of virulence genes in a non-biofilm-forming Staphylococcus epidermidis strain (ATCC 12228). Mol Microbiol. 2003, 49 (6): 1577-1593. 10.1046/j.1365-2958.2003.03671.x.PubMedView ArticleGoogle Scholar
- Gill SR, Fouts DE, Archer GL, Mongodin EF, Deboy RT, Ravel J, Paulsen IT, Kolonay JF, Brinkac L, Beanan M, Dodson RJ, Daugherty SC, Madupu R, Angiuoli SV, Durkin AS, Haft DH, Vamathevan J, Khouri H, Utterback T, Lee C, Dimitrov G, Jiang L, Qin H, Weidman J, Tran K, Kang K, Hance IR, Nelson KE, Fraser CM: Insights on evolution of virulence and resistance from the complete genome analysis of an early methicillin-resistant Staphylococcus aureus strain and a biofilm-producing methicillin-resistant Staphylococcus epidermidis strain. J Bacteriol. 2005, 187 (7): 2426-2438. 10.1128/JB.187.7.2426-2438.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Glass JI, Belanger AE, Robertson GT: Streptococcus pneumoniae as a genomics platform for broad-spectrum antibiotic discovery. Curr Opin Microbiol. 2002, 5 (3): 338-342. 10.1016/S0959-4388(02)90328-4.PubMedView ArticleGoogle Scholar
- Rohde H, Kalitzky M, Kroger N, Scherpe S, Horstkotte MA, Knobloch JK, Zander AR, Mack D: Detection of virulence-associated genes not useful for discriminating between invasive and commensal Staphylococcus epidermidis strains from a bone marrow transplant unit. J Clin Microbiol. 2004, 42 (12): 5614-5619. 10.1128/JCM.42.12.5614-5619.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Kocianova S, Vuong C, Yao Y, Voyich JM, Fischer ER, DeLeo FR, Otto M: Key role of poly-gamma-DL-glutamic acid in immune evasion and virulence of Staphylococcus epidermidis. J Clin Invest. 2005, 115 (3): 688-694. 10.1172/JCI200523523.PubMedPubMed CentralView ArticleGoogle Scholar
- Herd M, Kocks C: Gene fragments distinguishing an epidemic-associated strain from a virulent prototype strain of Listeria monocytogenes belong to a distinct functional subset of genes and partially cross-hybridize with other Listeria species. Infect Immun. 2001, 69 (6): 3972-3979. 10.1128/IAI.69.6.3972-3979.2001.PubMedPubMed CentralView ArticleGoogle Scholar
- Stabler RA, Marsden GL, Witney AA, Li Y, Bentley SD, Tang CM, Hinds J: Identification of pathogen-specific genes through microarray analysis of pathogenic and commensal Neisseria species. Microbiology. 2005, 151 (Pt 9): 2907-2922. 10.1099/mic.0.28099-0.PubMedView ArticleGoogle Scholar
- Doumith M, Cazalet C, Simoes N, Frangeul L, Jacquet C, Kunst F, Martin P, Cossart P, Glaser P, Buchrieser C: New aspects regarding evolution and virulence of Listeria monocytogenes revealed by comparative genomics and DNA arrays. Infect Immun. 2004, 72 (2): 1072-1083. 10.1128/IAI.72.2.1072-1083.2004.PubMedPubMed CentralView ArticleGoogle Scholar
- Hommais F, Pereira S, Acquaviva C, Escobar-Paramo P, Denamur E: Single-nucleotide polymorphism phylotyping of Escherichia coli. Appl Environ Microbiol. 2005, 71 (8): 4784-4792. 10.1128/AEM.71.8.4784-4792.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Drummond DA, Bloom JD, Adami C, Wilke CO, Arnold FH: Why highly expressed proteins evolve slowly. Proc Natl Acad Sci U S A. 2005, 102 (40): 14338-14343. 10.1073/pnas.0504070102.PubMedPubMed CentralView ArticleGoogle Scholar
- Jansen R, Gerstein M: Analysis of the yeast transcriptome with structural and functional categories: characterizing highly expressed proteins. Nucleic Acids Res. 2000, 28 (6): 1481-1488. 10.1093/nar/28.6.1481.PubMedPubMed CentralView ArticleGoogle Scholar
- Dorus S, Vallender EJ, Evans PD, Anderson JR, Gilbert SL, Mahowald M, Wyckoff GJ, Malcom CM, Lahn BT: Accelerated evolution of nervous system genes in the origin of Homo sapiens. Cell. 2004, 119 (7): 1027-1040. 10.1016/j.cell.2004.11.040.PubMedView ArticleGoogle Scholar
- Weinstock GM: Genomics and bacterial pathogenesis. Emerg Infect Dis. 2000, 6 (5): 496-504.PubMedPubMed CentralView ArticleGoogle Scholar
- Weissbach H, Resnick L, Brot N: Methionine sulfoxide reductases: history and cellular role in protecting against oxidative damage. Biochim Biophys Acta. 2005, 1703 (2): 203-212.PubMedView ArticleGoogle Scholar
- Ezraty B, Aussel L, Barras F: Methionine sulfoxide reductases in prokaryotes. Biochim Biophys Acta. 2005, 1703 (2): 221-229.PubMedView ArticleGoogle Scholar
- Wen ZT, Suntharaligham P, Cvitkovitch DG, Burne RA: Trigger factor in Streptococcus mutans is involved in stress tolerance, competence development, and biofilm formation. Infect Immun. 2005, 73 (1): 219-225. 10.1128/IAI.73.1.219-225.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Groisman EA, Casadesus J: The origin and evolution of human pathogens. Mol Microbiol. 2005, 56 (1): 1-7. 10.1111/j.1365-2958.2005.04564.x.PubMedView ArticleGoogle Scholar
- Medigue C, Rouxel T, Vigier P, Henaut A, Danchin A: Evidence for horizontal gene transfer in Escherichia coli speciation. J Mol Biol. 1991, 222 (4): 851-856. 10.1016/0022-2836(91)90575-Q.PubMedView ArticleGoogle Scholar
- Kunst F, Ogasawara N, Moszer I, Albertini AM, Alloni G, Azevedo V, Bertero MG, Bessieres P, Bolotin A, Borchert S, Borriss R, Boursier L, Brans A, Braun M, Brignell SC, Bron S, Brouillet S, Bruschi CV, Caldwell B, Capuano V, Carter NM, Choi SK, Codani JJ, Connerton IF, Danchin A: The complete genome sequence of the gram-positive bacterium Bacillus subtilis. Nature. 1997, 390 (6657): 249-256. 10.1038/36786.PubMedView ArticleGoogle Scholar
- Rocha EP, Danchin A: An analysis of determinants of amino acids substitution rates in bacterial proteins. Mol Biol Evol. 2004, 21 (1): 108-116. 10.1093/molbev/msh004.PubMedView ArticleGoogle Scholar
- Yao Y, Sturdevant DE, Otto M: Genomewide analysis of gene expression in Staphylococcus epidermidis biofilms: insights into the pathophysiology of S. epidermidis biofilms and the role of phenol-soluble modulins in formation of biofilms. J Infect Dis. 2005, 191 (2): 289-298. 10.1086/426945.PubMedView ArticleGoogle Scholar
- Robertson GT, Ng WL, Foley J, Gilmour R, Winkler ME: Global transcriptional analysis of clpP mutations of type 2 Streptococcus pneumoniae and their effects on physiology and virulence. J Bacteriol. 2002, 184 (13): 3508-3520. 10.1128/JB.184.13.3508-3520.2002.PubMedPubMed CentralView ArticleGoogle Scholar
- Mei JM, Nourbakhsh F, Ford CW, Holden DW: Identification of Staphylococcus aureus virulence genes in a murine model of bacteraemia using signature-tagged mutagenesis. Mol Microbiol. 1997, 26 (2): 399-407. 10.1046/j.1365-2958.1997.5911966.x.PubMedView ArticleGoogle Scholar
- Schirmer EC, Glover JR, Singer MA, Lindquist S: HSP100/Clp proteins: a common mechanism explains diverse functions. Trends Biochem Sci. 1996, 21 (8): 289-296. 10.1016/0968-0004(96)10038-4.PubMedView ArticleGoogle Scholar
- Bae T, Banger AK, Wallace A, Glass EM, Aslund F, Schneewind O, Missiakas DM: Staphylococcus aureus virulence genes identified by bursa aurealis mutagenesis and nematode killing. Proc Natl Acad Sci U S A. 2004, 101 (33): 12312-12317. 10.1073/pnas.0404728101.PubMedPubMed CentralView ArticleGoogle Scholar
- Rodriguez-Ortega MJ, Norais N, Bensi G, Liberatori S, Capo S, Mora M, Scarselli M, Doro F, Ferrari G, Garaguso I, Maggi T, Neumann A, Covre A, Telford JL, Grandi G: Characterization and identification of vaccine candidate proteins through analysis of the group A Streptococcus surface proteome. Nat Biotechnol. 2006, 24 (2): 191-197. 10.1038/nbt1179.PubMedView ArticleGoogle Scholar
- Bjorkman J, Samuelsson P, Andersson DI, Hughes D: Novel ribosomal mutations affecting translational accuracy, antibiotic resistance and virulence of Salmonella typhimurium. Mol Microbiol. 1999, 31 (1): 53-58. 10.1046/j.1365-2958.1999.01142.x.PubMedView ArticleGoogle Scholar
- Yao Y, Sturdevant DE, Villaruz A, Xu L, Gao Q, Otto M: Factors characterizing Staphylococcus epidermidis invasiveness determined by comparative genomics. Infect Immun. 2005, 73 (3): 1856-1860. 10.1128/IAI.73.3.1856-1860.2005.PubMedPubMed CentralView ArticleGoogle Scholar
- Rollenhagen C, Bumann D: Salmonella enterica highly expressed genes are disease specific. Infect Immun. 2006, 74 (3): 1649-1660. 10.1128/IAI.74.3.1649-1660.2006.PubMedPubMed CentralView ArticleGoogle Scholar
- Hirono I, Yamashita H, Park CI, Yoshida T, Aoki T: Identification of genes in a KG – phenotype of Lactococcus garvieae, a fish pathogenic bacterium, whose proteins react with antiKG – rabbit serum. Microb Pathog. 1999, 27 (6): 407-417. 10.1006/mpat.1999.0316.PubMedView ArticleGoogle Scholar
- Wright BE: Stress-directed adaptive mutations and evolution. Mol Microbiol. 2004, 52 (3): 643-650. 10.1111/j.1365-2958.2004.04012.x.PubMedView ArticleGoogle Scholar
- Brown MR, Kornberg A: Inorganic polyphosphate in the origin and survival of species. Proc Natl Acad Sci U S A. 2004, 101 (46): 16085-16087. 10.1073/pnas.0406909101.PubMedPubMed CentralView ArticleGoogle Scholar
- Costerton JW, Stewart PS, Greenberg EP: Bacterial biofilms: a common cause of persistent infections. Science. 1999, 284 (5418): 1318-1322. 10.1126/science.284.5418.1318.PubMedView ArticleGoogle Scholar
- Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ: Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 1997, 25 (17): 3389-3402. 10.1093/nar/25.17.3389.PubMedPubMed CentralView ArticleGoogle Scholar
- Zhang J, Rosenberg HF, Nei M: Positive Darwinian selection after gene duplication in primate ribonuclease genes. Proc Natl Acad Sci U S A. 1998, 95 (7): 3708-3713. 10.1073/pnas.95.7.3708.PubMedPubMed CentralView ArticleGoogle Scholar
- Kimura M: A simple method for estimating evolutionary rates of base substitutions through comparative studies of nucleotide sequences. J Mol Evol. 1980, 16 (2): 111-120. 10.1007/BF01731581.PubMedView ArticleGoogle Scholar
- Wall DP, Hirsh AE, Fraser HB, Kumm J, Giaever G, Eisen MB, Feldman MW: Functional genomic analysis of the rates of protein evolution. Proc Natl Acad Sci U S A. 2005, 102 (15): 5483-5488. 10.1073/pnas.0501761102.PubMedPubMed CentralView ArticleGoogle Scholar
- National Committee For Clinical Laboratory Standards: Performance standards for antimicrobial susceptibility testing; fourteenth Informational supplement M100-S14. NCCLS. 2004, Villanova, PA, USAGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.