|Home | About | Journals | Submit | Contact Us | Français|
A set of degenerate PCR primers was designed and used to amplify and sequence about 75% of the catalase (kat) gene from each of 49 staphylococcal strains. In some strains of Staphylococcus xylosus, S. saprophyticus, and S. equorum, two catalase genes, katA and katB, were found. A phylogenetic tree was generated and showed diversities among 66 partial (about 900-bp) staphylococcal kat nucleotide sequences (including 17 sequences found in GenBank) representing 26 different species. The topology of this tree showed a distribution of staphylococcal species similar, but not identical, to those reported previously based on 16S rRNA, hsp60, sodA, rpoB, tuf, and gap genes. The kat gene sequences were less conserved than those of 16S rRNA, rpoB, hsp60, and tuf genes and slightly more conserved than those of the gap gene. Therefore, kat gene sequence analysis may provide an additional marker for inferring phylogenetic relationships of staphylococci. Moreover, the discrete nucleotide polymorphism revealed in this gene could be exploited for rapid, low-cost identification of staphylococcal species through PCR-restriction fragment length polymorphism (RFLP) analysis. In this study, a PCR-RFLP assay performed by using only the TaqI restriction enzyme was successfully developed for rapid unequivocal identification/differentiation, at species and subspecies levels, of coagulase-positive staphylococci (CPS). The assay was validated by testing the DNA from 100 staphylococcal strains, including reference and wild CPS strains isolated from different environments. This reliable, rapid, and low-cost approach (requiring about 6 h from DNA isolation to the achievement of results and <5 Euros for each strain tested) allowed unambiguous identification of all the strains assayed, including the newly described S. delphini and S. pseudintermedius CPS species.
Staphylococci are a group of microorganisms occurring widely in nature. As reported in the List of Prokaryotic Names with Standing in Nomenclature (www.bacterio.net) as of the full update in March 2006, the Staphylococcus genus includes 39 valid species, 11 of which are divided into two or more subspecies, resulting in more than 50 recognized systematic entities (15).
The genus includes both human and animal pathogens, generally coagulase-positive staphylococci (CPS) such as S. aureus, S. intermedius, S. delphini, S. hyicus, S. schleiferi subsp. coagulans, and S. pseudintermedius (12, 21, 29, 35), and coagulase-negative staphylococci (CNS) such as S. equorum, S. xylosus, S. carnosus, S. simulans, S. saprophyticus, S. succinus, S. warneri, S. vitulinus, S. pasteuri, S. epidermidis, and S. lentus.
Notwithstanding their valuable role in food fermentation (5-9, 23, 24), some CNS such as S. saprophyticus, S. epidermidis, and S. haemolyticus exhibit increasing abilities as opportunistic/emerging pathogens in colonizing animal and human tissues, due in part to the ability of these species (particularly S. epidermidis) to form protective biofilms and their ubiquitous occurrence in the environment (37). Finally, the occurrence of some pathogenic species (S. aureus, S. intermedius, and S. hyicus) in food is a potential public health hazard, since many strains can produce enterotoxins (4, 9, 31). Indeed, enterotoxin-producing food-associated CNS strains (S. carnosus, S. equorum, S. piscifermentans, and S. xylosus) were also recently described by Zell et al. (40). Despite the importance of accurate identification of staphylococcal species by microbiological laboratories, previous studies demonstrated the unreliability of phenotypic methods, including the Vitek 2 system (bioMérieux, Marcy l'Etoile, France), compared to different molecular techniques (5-8, 11) to identify staphylococci.
Due to the high sensitivity and specificity they provide, molecular markers are an alternative tool for accurate identification and classification of Staphylococcus species. Evaluations of 16S to 23S rRNA gene polymorphisms by PCR and PCR-denaturing gradient gel electrophoresis (DGGE) analyses of 16S ribosomal sequences have been assessed for the ability to achieve clear identification of strains within the Staphylococcus genus (6). Molecular assays targeting some housekeeping genes such as hsp60 (17), the 16S rRNA gene (14, 18), femA (36), tuf (22), gap (38, 39), sodA (26), rpoB (13), and dnaJ (20, 30) have been used for reliably identifying and classifying staphylococci. However, except for the sodA, hsp60, and nuc gene sequence analyses carried out by Sasaki et al. (29), which allowed some phenotypically identified S. intermedius strains to be reclassified as S. delphini and S. pseudintermedius, to our knowledge, no method allowing rapid identification and differentiation of CPS species including the recently described S. delphini and S. pseudintermedius is available to date.
In this study, we evaluated the catalase (kat) gene as a new target for phylogenetic analysis of staphylococci. Catalase is a heme-containing enzyme involved in dismutation of hydrogen peroxide generated during cellular metabolism to water and molecular oxygen. In S. aureus, a correlation between catalase activity and virulence has been observed, suggesting a role for catalase in defensive mechanisms against the oxygen radicals produced by macrophages (28). Additionally, a recent study (25) showed that the S. aureus catalase is a major factor in S. aureus defense against Streptococcus pneumoniae due to neutralization of secreted H2O2 produced by the latter microorganism. Sanz et al. (28) showed that catalase deficiency in S. aureus subsp. anaerobius is associated with natural loss-of-function mutations within the structural gene. Therefore, catalase-negative staphylococcal strains may also harbor the catalase gene. Barrière et al. (3) described the katA gene of S. xylosus C2a and supposed the presence of a second catalase gene (katB) in this strain. In fact, the katA-deficient mutant of S. xylosus constructed by these authors still exhibited catalase activity. After analysis of polymorphism within the kat genes of 26 different staphylococcal species, we designed and successfully applied a molecular assay allowing rapid and unequivocal identification and differentiation of CPS species and subspecies.
Forty-six reference strains (including type strains and previously identified strains) representing 25 different staphylococcal species were used in this study (Fig. (Fig.1).1). Moreover, to validate the implemented katA PCR-restriction fragment length polymorphism (RFLP) approach for rapid identification of CNS, other staphylococcal strains from different environments were also analyzed (Table (Table1).1). Working cultures were grown in brain heart infusion broth (Oxoid Ltd., Basingstoke, Hampshire, United Kingdom) at 37°C. Before DNA extraction, cultures were streaked onto brain heart infusion agar plates and grown overnight at 37°C.
DNA extraction from a single colony was carried out by using the InstaGene matrix under the conditions described by the supplier (Bio-Rad Laboratories, Hercules, CA).
Degenerate primers KDN-For (AARGGWTCHGGWGCWTTYGG) and KDN-Rev (TGTTCRAARTARTTRTCRTCATC) were selected from among a highly conserved region of staphylococcal strain katA gene sequences available in the GenBank and EMBL databases (1); the sequences were from S. saprophyticus, S. warneri, S. epidermidis, S. haemolyticus, S. xylosus, S. aureus subsp. anaerobius, and S. aureus subsp. aureus (Fig. (Fig.1).1). The primers amplify an internal fragment of 1,114 bp of the katA gene, from position 164 to position 1277 (nucleotide numbering with respect to the katA gene of S. xylosus C2a [accession no. AJ295151]).
PCR amplifications were performed with a total volume of 50 μl, including 5 to 10 μl (25 to 50 ng) of target DNA, 5 μl of Taq DNA polymerase 10× buffer (Invitrogen S.R.L., Milan, Italy), 2.5 μl of 50 mM MgCl2, 0.5 μl of deoxynucleoside triphosphate (dNTP) mix (25 mM [each] dNTPs), 0.2 μl of each primer (0.1 mM), and 0.5 μl of a Taq DNA polymerase solution (5 U/μl; Invitrogen S.R.L.). PCR thermal conditions consisted of an initial denaturing step (95°C for 3 min), 40 amplification cycles (60 s at 94°C, 60 s at 52°C, and 90 s at 72°C), and one final step at 72°C for 10 min. The PCR amplification fragments were resolved by agarose (1.5%, wt/vol) gel electrophoresis at 100 V for 2 h. The gel was stained with ethidium bromide, and the bands were visualized under UV illumination at 254 nm.
The PCR products of the expected length were purified from the agarose gel by using a QIAquick gel extraction kit (Qiagen S.p.A., Milan, Italy) and quantified by comparison with lambda/HindIII molecular marker fragments (Invitrogen S.R.L.).
Before sequencing, restriction digestion of the purified PCR products was performed in a total volume of 50 μl by using about 300 ng of DNA and 30 U of CfoI restriction endonuclease enzyme (Promega Italia S.R.L., Milan, Italy) so as to evaluate the specificity of the amplified fragment and/or the presence of spurious PCR products (see Results).
The DNA sequences were determined by the dideoxy chain termination method by using a DNA sequencing kit (Perkin-Elmer Cetus, Emeryville, CA) and primer KDN-For. The sequences were analyzed by MacDNasis Pro version 3.0.7 (Hitachi Software Engineering Europe S.A., Olivet, France), and research on DNA similarity was performed with the GenBank and EMBL databases (1).
Phylogenetic analysis was carried out using MEGA version 4.0 (33) after multiple-sequence alignment of data by CLUSTAL W 1.8 (34). Distance matrix and neighbor-joining methods (27) were applied for tree construction.
DNA sequence similarity analysis was performed by BioEdit version 7.0.9 (www.mbio.ncsu.edu/BioEdit/bioedit.html).
From the highly conserved region of S. aureus, S. delphini, S. hyicus, S. intermedius, S. pseudintermedius, and S. schleiferi katA gene sequences found in GenBank or determined during this study (Fig. (Fig.1),1), two oligonucleotide primers were selected: CPSK1F (CARAAYAACTGGGATTTCTGGAC) and CPSK6R (GCATCRCCRTAWGAGAATAAACG). Targeting positions 487 to 509 and 1031 to 1009 of the katA gene of S. aureus subsp. aureus Mu50 (BA000017) allowed amplification of fragments of 544 bp. PCR amplifications were performed with a total volume of 50 μl, including 5 μl of target DNA, 5 μl of Taq DNA polymerase 10× buffer (Invitrogen), 2.5 μl of 50 mM MgCl2, 0.5 μl of dNTP mix (25 mM [each] dNTPs), 0.2 μl of each primer (0.1 mM), and 0.2 μl of Taq DNA polymerase solution (5 U/μl; Invitrogen S.R.L.). PCR thermal conditions consisted of an initial denaturing step (95°C for 3 min) and 40 amplification cycles: a denaturing step for 10 s at 95°C and an annealing-extension step for 45 s at 56°C. After amplification, 15-μl samples of PCR mixtures were tested by agarose (1.5%, wt/vol) gel electrophoresis at 100 V for 1 h. The remaining part (30 μl) of the PCR product was digested in a total volume of 50 μl by 20 U of TaqI restriction endonuclease (GE Healthcare, Milan, Italy) at 65°C for 2 h. Restriction fragments were resolved by agarose (2%, wt/vol) gel electrophoresis at 100 V for 2 h.
The sequence of the katB gene of S. xylosus DSM 20266T determined in this study was deposited in GenBank under accession number AY702101.
Partial sequences (<900 bp) of the kat genes from 49 staphylococcal strains were determined by using degenerate PCR primers designed during this study. Initially, for only 46 of 49 staphylococcal strains available, it was possible to directly sequence more than 900 bp of the purified 1,114-bp KDN-For/KDN-Rev PCR product. Gene sequences obtained from S. saprophyticus strains P72K3, P52K3, and GB1, S. warneri strains DSM 20316 and P-98M5, S. epidermidis strains DSM 1798 and DSM 20044, S. haemolyticus strain DSM 20263, and S. aureus strains ATCC 27664 and DSM 20231 displayed a high level of similarity to katA gene sequences from strains of the same species reported in the database (Fig. (Fig.1),1), confirming that the amplified products were really part of the katA gene.
It was not possible to achieve direct sequencing of the purified 1,114-bp KDN-For/KDN-Rev PCR product from S. xylosus DSM 20266T, S. equorum subsp. linens DSM 15097T, and S. saprophyticus subsp. saprophyticus DSM 20229T strains. In fact, in analyzing the CfoI KDN-For/KDN-Rev restriction pattern of S. xylosus DSM 20266T, we observed that in some cases the sum of the molecular sizes of the restriction fragments was about 2,200 bp, i.e., double the expected length (Fig. (Fig.2,2, pattern C). This result was confirmed when 1 μg of PCR product was digested with 30 U of CfoI in 16 h at 37°C, excluding the hypothesis of partial digestion of the PCR product. Moreover, 23 of the 31 S. xylosus strains analyzed showed restriction profiles similar to that of DSM 20266T (data not shown). The presence of a second kat gene (katB) in strain S. xylosus DSM 20266T, similar to the one described by Barrière et al. (3), was indicated by TaqI restriction digestion of the KDN-For/KDN-Rev PCR product, and the full nucleotide sequence (accession number AY702101) was determined by inverse PCR experiments, i.e., digesting total DNA with TaqI and using KBSxFI2 (CCTGACGAAGCAGCGAAAAT) and KBSxFI4 (ACGCGTTCACCTTTGTCGTT) as inverse primers in the PCR.
As shown in Fig. Fig.22 by the CfoI KDN-For/KDN-Rev restriction patterns for S. equorum subsp. linens DSM 15097T (pattern B) and S. saprophyticus DSM 20229T (ATCC 15305; pattern F), these strains also produced additional fragments, suggesting again the presence of two catalase genes. By applying the approach described above for S. xylosus DSM 20266T, it was possible to determine partial sequences of both catalases of S. equorum and S. saprophyticus. The S. equorum subsp. linens DSM 15097T katA gene clustered with S. xylosus katA genes, while katB clustered with S. saprophyticus katA and S. xylosus katB genes (Fig. (Fig.1).1). The katA gene sequence of S. saprophyticus DSM 20229T determined during this study was identical to that already described for S. saprophyticus ATCC 15305 (accession number AP008934), while katB revealed about 98% similarity to katA genes of S. xylosus strains (Fig. (Fig.11).
A total of 66 partial Staphylococcus kat sequences (including 17 sequences found in the GenBank and EMBL databases [www.ncbi.nlm.nih.gov]) representing 26 different species were compared. The identity matrix (developed by BioEdit) based on comparison of the katA gene sequences of 30 Staphylococcus strains revealed that the interspecies sequence similarity ranged from 0.40 to 0.93. The latter value was calculated for S. carnosus DSM 20501 and S. condimenti DSM 11674, as well as for S. pseudintermedius LMG 22219 and S. intermedius DSM 20273. Sequence similarity values of >0.96 were found for strains belonging to subspecies of the same species: S. aureus subsp. anaerobius MVF213 and S. aureus subsp. aureus DSM 20231 (0.985), S. schleiferi subsp. schleiferi DSM 4807 and S. schleiferi subsp. coagulans DSM 6628 (0.96), and S. succinus subsp. casei DSM 15096 and S. succinus subsp. succinus DSM 14617 (0.96).
Evolutionary divergence between the 66 kat gene sequences was estimated by using MEGA version 4.0 (33). The number of base differences per site obtained by averaging over all sequence pairs was 0.197 (standard error, 0.01). The dendrogram resulting from the neighbor-joining analysis of 66 kat sequences is shown in Fig. Fig.1.1. The topology of the kat gene-based tree was similar to those of trees obtained by analyzing the rpoB, sodA, tuf, gap, 16S rRNA, and hsp60 genes, recently published by Ghebremedhin et al. (16). Sequences from strains of the same species clustered very close (bootstrap values, 99 to 100%), with the exception of the S. saprophyticus DSM 20229 katB gene, clustering close to the katA genes of S. xylosus strains (Fig. (Fig.1),1), due mainly to the presence of two catalase genes in these species. We named the previously known kat gene katA and the one discovered in this study katB. Based on the analyses of katA from S. saprophyticus and katB from S. xylosus, these species are not very close.
Finally, different clusters with significant bootstrap values (>90%) could be identified (Fig. (Fig.1):1): (i) the S. saprophyticus group, including S. saprophyticus, S. xylosus, and S. equorum (bootstrap value, 99%, based on the katA and katB genes); (ii) the S. carnosus group, including S. carnosus, S. condimenti, and S. simulans (bootstrap value, 98%); (iii) the S. intermedius group, including S. intermedius, S. pseudintermedius, and S. delphini (bootstrap value, 99%); (iv) the S. epidermidis/S. aureus group, including S. epidermidis, S. capitis, S. caprae, S. haemolyticus, S. pasteuri, S. warneri, and S. aureus (bootstrap value, 93%); and (v) the S. sciuri group, including S. sciuri, S. vitulinus, and S. lentus (bootstrap value, 100%).
In silico restriction (computer-aided) endonuclease analysis of all Staphylococcus kat partial sequences was performed by MacDNasis Pro (version 3.0.7.) software.
TaqI PCR-RFLP in silico pattern analysis may allow identification and differentiation of the CPS species S. aureus, S. delphini, S. intermedius, S. pseudintermedius, S. schleiferi subsp. coagulans, and S. hyicus (Table (Table2)2) .
By coupling Alu with CfoI or TaqI PCR-RFLP in silico patterns, all staphylococcal species could be differentiated (Table (Table2).2). In addition, CfoI and TaqI also allowed intraspecific polymorphism to be revealed. In particular, CfoI PCR-RFLP in silico analysis provided discrimination between S. aureus susbp. aureus and S. aureus susbp. anaerobius, as well as S. succinus subsp. casei and S. succinus subsp. succinus, whereas by using TaqI, S. schleiferi subsp. schleiferi and S. schleiferi subsp. coagulans could be distinguished (Table (Table22).
To set up a rapid method allowing identification and differentiation of CPS strains, we designed universal primers CPSK1F and CPSK6R, allowing the amplification of a 544-bp region of katA containing polymorphic TaqI restriction sites of the various CPS strains. As shown in Fig. Fig.3,3, clear differentiation of all species was achieved by using the PCR-RFLP approach. However, to further validate and confirm the usefulness of this assay, the PCR-RFLP patterns displayed by strains listed in Table Table11 were also determined and analyzed. Based on their PCR-RFLP patterns, strains were identified as follows. Among 56 strains received as S. aureus, 53 were confirmed to be S. aureus, 2 (SAT-51 and SAT-52) were identified as S. delphini, and 1 (SAT-48) was identified as S. intermedius. Among five strains received as S. delphini, three were confirmed to be S. delphini and two (h-2c and P-27B) were identified as S. intermedius. Among 21 strains received as S. intermedius, 10 were confirmed to be S. intermedius, 1 strain (SAT-42) displayed a mixed S. aureus/S. intermedius pattern, 7 were identified as S. delphini, and 3 were identified as S. pseudintermedius. Identification of strains received as S. pseudintermedius, S. schleiferi, and S. hyicus was confirmed. One strain received as S. chromogenes (Ter-24N) was identified as S. delphini. Finally, some strains received as Staphylococcus spp. and showing patterns (designated X, Y, and Z in Table Table1)1) different from those of CPS reference strains based on sequencing of the katA gene (KDN-For/KDN-Rev PCR fragment) were identified as S. epidermidis (strains ED-1 and ED-2), S. chromogenes (strain SH4/03), and Staphylococcus sp. (strain SH3/03, showing the highest level of similarity, 88%, to S. hyicus). To confirm the identification, some strains reclassified on the basis of their TaqI PCR-RFLP patterns were subjected to sequencing of the katA gene (544-bp fragment). As indicated in Fig. Fig.4,4, comparison of the resulting sequences with those from reference strains yielded findings in agreement with those of the PCR-RFLP analysis.
Moreover, the same strains were also subjected to partial sequencing of the gap gene (700-bp fragment) as described by Ghebremedhin et al. (16). As indicated in Fig. Fig.5,5, identifications obtained by comparison of the resulting gap sequences with those from reference strains found in GenBank fit perfectly with the identifications obtained by PCR-RFLP analysis and katA gene sequencing. The K1224 strain (of equine origin), originally identified as S. intermedius by traditional identification methods, was recently reclassified as S. delphini based on rpoB sequencing (showing 99.38% identity to S. delphini rpoB) (K. Becker, Münster University, Germany, personal communication).
Due to the very high interspecies sequence similarity (90 to 99%) displayed by staphylococcal species, the use of 16S rRNA gene sequence analysis has been questioned in studies at the species level. Better results have been obtained by comparing sequences of other housekeeping genes such as hsp60, sodA, rpoB, tuf, and gap (16).
In this study, we evaluated the kat gene as a new target for phylogenetic analysis of staphylococci and identification and differentiation of staphylococcal isolates at the species level.
Degenerate primers designed during this study were able to coamplify the two catalase genes of S. xylosus DSM 20266. Furthermore, by inverse PCR experiments, we were able to determine the full nucleotide sequence of katB from this microorganism. Two catalase genes were also found in S. equorum and S. saprophyticus. S. xylosus, S. equorum, and S. saprophyticus are the main staphylococcal species occurring in naturally fermented meat products (5-8). S. xylosus is widely used as a starter culture in association with lactic acid bacteria for process improvement (19). The presence of two catalase genes in the above-listed staphylococcal species may be a defense mechanism against hydrogen peroxide produced by lactic acid bacteria during fermentation.
The kat gene sequences were less conserved (with interspecies sequence similarity ranging from 40 to 93%) than those of the genes rpoB (71.6 to 93.6% similar), hsp60 (74 to 93% similar), and tuf (86 to 97% similar) and slightly more conserved than those of the gap gene (24 to 96% similar) (16).
The kat-based tree indicates the divergence of staphylococcal species and was well supported for all the strains analyzed during this study. In addition, with both kat genes and a bootstrap value of >90%, the staphylococcal species were divided into five well-supported clusters: the S. saprophyticus group, the S. carnosus/S. simulans group, the S. intermedius group, the S. epidermidis/S. aureus group, and the S. sciuri group. Similar but not identical results were obtained by Ghebremedhin et al. (16), who analyzed gap genes and divided staphylococcal species into only four well-supported clusters: the S. sciuri group, the S. hyicus/S. intermedius group, the S. haemolyticus/S. simulans group, and the S. epidermidis/S. aureus group. Therefore, kat gene analysis may represent an additional marker for inferring staphylococci phylogenetics.
katA genes display a high level of restriction endonuclease polymorphism, offering good opportunities for rapid, accurate species-level identification of staphylococcal isolates. However, as indicated by our in silico analysis of katA genes from only reference strains of 26 staphylococcal species, sequential use of two endonucleases, first AluI and then CfoI or TaqI, is needed to achieve identification of staphylococcal isolates at the species and subspecies levels. In contrast, by selecting the suitable katA region to analyze, we demonstrated with our evaluation of about 100 wild strains that TaqI PCR-RFLP alone can discriminate among CPS species.
Several studies (11, 29) have highlighted biases in phenotypically differentiating the various CPS species. Sasaki et al. (29), for example, reclassified some phenotypically identified S. intermedius isolates as S. pseudintermedius or S. delphini by sequence analyses of sodA, hsp60, and nuc genes. Commercial kits for identifying S. delphini and S. pseudintermedius, the most recently described CPS species, are not available. Therefore, at present the most appropriate approach to reliably identify the latter two species entails sequencing of more than one housekeeping gene: gap, sodA, hsp60, and/or nuc.
In this study, we designed a robust, rapid, and low-cost approach (requiring about 6 h from DNA isolation to the production of results and <5 Euros per strain tested) to identify and differentiate CPS species, based on TaqI restriction endonuclease analysis of the 544-bp PCR-amplified katA gene fragment (TaqI PCR-RFLP analysis).
Our strategy is similar to other PCR-RFLP methods based on gap, dnaJ, and pta genes (2, 20, 38, 39). However, Yugueros et al. (38, 39) analyzed only three CPS species (S. aureus, S. intermedius, and S. delphini), Hauschild and Stepanovic (20) obtained good results by using two restriction enzymes sequentially but did not analyze S. delphini strains, and Bannoehr et al. (2) were unable to differentiate strains of S. delphini, S. intermedius, and S. schleiferi.
Our PCR-RFLP technique was able to differentiate all CPS species. This approach was validated by unambiguous identification of more than 100 strains, including reference and wild CPS strains isolated from different sources. Owing to its specificity, manageability, and rapidity, the kat-based PCR-RFLP approach proposed in this study can be considered a valid strategy for rapid identification of CPS strains at the species level.
In conclusion, the high variability of the kat nucleotide sequences allows discrimination among closely related species, opens new possibilities for rapid, reliable identification of staphylococci, and offers good opportunities to develop assays based on hybridization (probe design), PCR (primer design), or DNA chip (microarray design) technologies. Upon comparing our results with the recent findings of Ghebremedhin et al. (16), the greater usefulness of kat gene analysis than of 16S rRNA, rpoB, hsp60, and tuf gene analyses is evident. Moreover, for several species of bacteria, nucleotide sequence analysis of multiple-protein-encoding loci has led to reliable phylogenies that have improved our understanding of the population structure. A multigenic approach fulfils the recent recommendations of the ad hoc committee for the reevaluation of the definition of bacterial species (32). The kat gene may be considered an excellent molecular marker for inferring the taxonomy and phylogeny of members of the genus Staphylococcus and may represent a good candidate in multilocus schemes designed to identify and characterize staphylococci.
We thank G. Lina of the CNTS (Lyon, France), K. Becker of the IMM (Münster, Germany), K. Hiramatsu of the DICS (Tokyo, Japan), and A. Ianieri of the DSA (Teramo, Italy) for kindly providing strains used in this study. Thanks are also due to Giancarlo Ciao for his technical assistance.
This work was supported partly by an Italian MIUR grant.
Published ahead of print on 4 November 2009.