Skip Navigation

A Geographic Variant of the Staphylococcus aureus Panton-Valentine Leukocidin Toxin and the Origin of Community-Associated Methicillin-Resistant S. aureus USA300

  1. F. Patrick O'Hara1,
  2. Nicolas Guex4,
  3. J. Michael Word4,
  4. Linda A. Miller2,
  5. Julie A. Becker1,
  6. Stacey L. Walsh1,
  7. Nicole E. Scangarella3,
  8. Joshua M. West3,
  9. Ribhi M. Shawar3 and
  10. Heather Amrine-Madsen4
  1. 1Evolutionary and Structural Bioinformatics, GlaxoSmithKline, Collegeville, Pennsylvania
  2. 2Infectious Diseases Medicine Development Center, GlaxoSmithKline, Collegeville, Pennsylvania
  3. 3Department of Microbiology, GlaxoSmithKline, Collegeville, Pennsylvania
  4. 4Evolutionary and Structural Bioinformatics, GlaxoSmithKline, Research Triangle Park, North Carolina
  1. Reprints or correspondence: Dr. Heather Amrine-Madsen, Evolutionary and Structural Bioinformatics, GlaxoSmithKline, 5 Moore Dr., MAI-C2357, Research Triangle Park, NC 27709 (heather.a.madsen{at}gsk.com).
 
Next Section

Abstract

Background. The majority of recent community-associated methicillin-resistant Staphylococcus aureus (MRSA) infections in the United States have been caused by a single clone, USA300. USA300 secretes Panton-Valentine leukocidin (PVL) toxin, which is associated with highly virulent infections.

Methods. We sequenced the PVL genes of 174 S. aureus isolates from a global clinical sample. We combined phylogenetic reconstruction and protein modeling methods to analyze genetic variation in PVL.

Results. Nucleotide variation was detected at 12 of 1726 sites. Two PVL sequence variants, the R variant and the H variant, were identified on the basis of a substitution at nt 527. Of sequences obtained in the United States, 96.7% harbor the R variant, whereas 95.6% of sequences obtained outside the United States harbor the H variant; 91.3% of MRSA isolates harbor the R variant, and 91.3% of methicillin-susceptible strains harbor the H variant. A molecular model of PVL shows 3 mechanisms by which the amino acid substitution may affect PVL function.

Conclusions. All sampled PVL genes appear to share a recent common ancestor and spread via a combination of clonal expansion and horizontal transfer. US isolates harbor a variant of PVL that is strongly associated with MRSA infections. Protein modeling reveals that this variant may have functional significance. We propose a hypothesis for the origin of USA300.

Methicillin-resistant Staphylococcus aureus (MRSA) has been a major cause of nosocomial infections since it was first described in 1961 [1], and hospital-associated strains (HA-MRSA) are frequently resistant to multiple classes of antibacterials [2]. Beginning in the 1990s, outbreaks of community-associated MRSA (CA-MRSA) have become increasingly frequent, and some CA-MRSA strains have acquired multidrug resistance [3]. CA-MRSA infections in the United States are predominantly caused by a single epidemic clone, USA300 [4–6], which has multilocus sequence typing (MLST) sequence type (ST) 8, pulsed-field gel electrophoresis (PFGE) type USA300, and staphylococcal chromosomal cassette mec (SCCmec) type IV. First described in 2000, USA300 rapidly supplanted the original US CA-MRSA clone known as MW2 (PFGE type 400 and ST1) and has begun to displace traditional HA-MRSA strains [5, 7].

Both USA300 and MW2 express the Panton-Valentine leukocidin (PVL) genes [8, 9], which are associated with highly virulent infections of the skin and soft tissue as well as necrotizing pneumonia and necrotizing fasciitis [10, 11]. The PVL genes (lukF-PV and lukS-PV) encode a cytotoxin that has multiple functions: it causes concentration-dependent necrosis and apoptosis of human polymorphonuclear neutrophils (PMNs) [12], lysis of human monocytes and macrophages [13], activation of calcium channels [14], and global changes in gene transcription [15]. PVL has been shown to be sufficient to cause necrotizing pneumonia in a murine model [15]. CA-MRSA from other countries are genetically distinct but can also harbor PVL and SCCmec type IV [16]. To date, none of the sequenced S. aureus genomes publicly available represent PVL-positive methicillin-susceptible S. aureus (MSSA) strains.

The lukSF-PV genes are carried on a prophage, and there is in vitro evidence that the genes may be transferred from one strain to another via phage transduction [17]. Thus, PVL genes may be transmitted vertically within the same clones or horizontally among different clones. Although variation in the genetic backgrounds of strains of MRSA and MSSA has been investigated, no studies to date have focused on evaluating the genetic variation within the lukSF-PV genes in a large clinical sample. Observing patterns of genetic variation in lukSF-PV can contribute to our understanding of the population biology of MRSA and MSSA and give insight into how resistant and virulent strains emerge.

Here, we analyze sequences of the lukSF-PV genes from 174 isolates of S. aureus collected from global clinical trials of subjects with uncomplicated skin infections. We demonstrate that all US MRSA isolates in the sample possess a unique geographic variant of PVL. Furthermore, we show that the geographic variant is distinguished by an amino acid substitution. Using molecular modeling, we provide evidence that this substitution may have functional significance and may contribute to the success of USA300. Our data allow us to suggest a hypothesis for the origin of the USA300 clone.

Previous SectionNext Section

Methods

Bacterial isolates. Single S. aureus isolates were obtained during the baseline visit from the skin specimens of adult and pediatric patients enrolled in 3 global clinical trials of uncomplicated skin infections. All isolates were collected in 2004–2005 from investigator sites in North America (including 13 US states), Germany, Russia, India, South Africa, Peru, and Costa Rica. Isolates were stored frozen at −70°C and subcultured on Luria-Bertani medium for testing. Multiplex polymerase chain reaction (PCR) technology was used to characterize all S. aureus isolates on the basis of the presence or absence of the PVL and mecA genetic determinants [18]. Using a PCR-based test for the lukSF-PV genes, we determined that 177 isolates were PVL positive. We could amplify the full fragment from 174 of these isolates, so these 174 isolates (57 MRSA and 117 MSSA) were included in the present study. MLST was performed on all isolates, as described by Enright et al. [19]. The MLST sequences were compared with those in the public database (http://www.mlst.org) to generate STs, and the STs were grouped into clonal complexes by use of the eBURST analysis tools [20, 21]. A clonal complex is defined as a group of STs in which each ST shares at least 6 of 7 identical loci with at least 1 other ST in the group. Clonal complexes are named after the most common ST in the complex.

Sequencing of the lukSF-PV genes. An internal fragment of the lukSF-PV coding region starting 124 bp downstream of the start of lukS-PV and ending 69 bp upstream of the end of lukF-PV was amplified by PCR, and both strands were sequenced. The following primers were used for amplification and sequencing: PVL1F (5′-ggtgatggcgctgaggtagtcaaa-3′), PVL2F (5′-ggcgctgaggtagtcaaaagaacag-3′), PVL1R (5′-ctgtatgattttcccaatcaacttc-3′), PVLint3R (5′-gtattggaaaggccacctcattgc-3′), PVLint2R (5′-caaattcacttgtatctcctgagcc-3′), PVLint2F (5′-caactgcaacatcagattccgataag-3′), PVLint1R (5′-ctccaccataagaataacctaccg-3′), and PVLint3F (5′-gggaccatatggcagagatagttatc-3′). Sequences have been submitted to GenBank under the accession numbers EF571668–EF571841.

Phylogenetic analyses. Contigs were assembled using Sequencer (version 4.5; Gene Codes), homologous DNA sequence alignments were generated using ClustalX (version 1.83) [22], and sequences were analyzed and manually edited in GeneDoc (version 2.6) [23]. Sequences of lukSF-PV from the published genomes of the CA-MRSA strains USA300 (CP000255 region: 1546170–1548087) and MW2 (BA000033 region: 1529381–1531298) were collected from GenBank and aligned for use as comparators. Statistical parsimony (haplotype) networks were generated using TCS (version 1.21) [24], with gaps treated as missing data. The statistical parsimony connection limit was set at 99%.

Molecular modeling. A crude model of the lukSF-PV pore complex as an octamer was built from the crystal structure of the related heptameric α-hemolysin pore (protein data bank identifier 7ahl [25]) using Swiss-PdbViewer [26] and MAGE [27] by superimposing the crystal structures of the lukF-PV monomer (1pvl [28]) and the lukS-PV monomer (1t5r [29]) onto chain B of the α-hemolysin structure.

Previous SectionNext Section

Results

PVL sequence variation among S. aureus isolates. A 1726-bp fragment comprising 90% of the full sequence of the lukSF-PV genes from 174 isolates was amplified by PCR, and the products were sequenced. The sequence is highly conserved, but nucleotide variation is seen at 12 sites (figure 1). The lukS-PV sequence is more variable than the lukF-PV sequence; 8 of the 12 variable sites are found in lukS-PV. Eight of the 12 variable sites contain minor variants; the minor allele was present in only 1–5 sequences. Four of these minor variable sites result in amino acid change. At the remaining 4 variable sites, a major variant was observed in >5 sequences. One of these major variable sites results in an amino acid change.

Figure 1
View larger version:
Figure 1

The lukSF-PV sequence variants found across 174 clinical isolates of Staphylococcus pneumoniae. Shown is a summary of nucleotide variation observed among lukSF-PV sequences from methicillin-resistant and methicillin-susceptible S. aureus and comparison of the sequence with the reported USA300 lukSF-PV sequence. The thick horizontal line at the top represents the sequence from the USA300 genome. Short vertical lines above and below indicate relative positions of variable sites. The nucleotide residue from the USA300 genome is indicated. Longer vertical lines indicate the 5′ and 3′ ends of the sequenced fragment. The unique haplotypes observed in our sample are represented by the thick horizontal lines below. Short vertical lines indicate the positions of sites at which the haplotype differs from that of USA300. The alternate nucleotide residues are indicated, and substitutions resulting in an amino acid change are indicated in bold. Nos. to the left of the horizontal lines indicate the frequency of the haplotype observed. At right, the haplotypes are bracketed according to how they are grouped into 1 of 4 major haplotype groups.

We identified 13 unique haplotypes among the sampled sequences. The sequence of the most common haplotype is identical to that of USA300. The sequences can be assigned to 4 major haplotype groups on the basis of variation at positions 527, 663, and 1396 of the nucleotide sequence. Sixty-three sequences have a G at position 527, which results in an arginine (R) at aa 176. The remaining 111 have an A at 527, resulting in a histidine (H) at aa 176. These will be referred to hereafter as the “R variant” and the “H variant,” respectively. The H variant can be further broken into groups H1, H2, and H3.

Variation at aa 176 correlates strongly with the geography of the isolate harboring the PVL sequence and is also correlated with the presence of mecA (table 1). Of the 63 R variant sequences, 58 are from the United States. Of the remaining 111 sequences, 109 are found outside the United States. The H1 and H2 variants are commonly found in South Africa and India, whereas the H3 variant is common in Europe (Germany and Russia). The R variant is strongly associated with presence of mecA; 53 of 63 isolates are mecA positive. The opposite is true for isolates carrying theHvariant. Only 5 of 111 isolates carrying the H variant of PVL also carried mecA.

Figure 2
View larger version:
Figure 2

Statistical parsimony network depicting the phylogenetic relationships among 174 lukSF-PV sequences. Ellipses indicate individual alleles, with the size proportional to the frequency of the allele. The box indicates the predicted ancestral allele. Lines connecting ellipses indicate single-nucleotide substitutions. Ellipses are color coded according to the major haplotype variants: yellow, H1; green, H2; blue, H3; purple, R. Haplotypes of note are annotated to give additional genetic and population data.

Figure 3
View larger version:
Figure 3

Model of the Panton-Valentine leukocidin (PVL) octamer. Top (A) and side (B) views of the crude model of the LukS-PV (blue)—LukF-PV (orange) octamer show the position of H176 (148 in the LukS-PV crystal structure and in our model) in red.

Figure 4
View larger version:
Figure 4

Proposed structural significance of arginine at position 176(148). A, Close-up of the interaction between the histidine (H variant) at position 176(148) of LukS-PV, which may or may not carry a positive charge, and the negatively charged carboxy-terminus of LukF-PV. B, Proposed strong ionic interaction between the positively charged arginine (R variant) at position 176(148) and the carboxy-terminus. For clarity, only 1 LukF (orange) and 1 LukS (blue) monomer “head” are represented. The “stalk” of the octamer is represented in dark gray.

Table 1
View larger version:
Table 1

Correlation of Panton-Valentine leukocidin (PVL) sequence variants with methicillin resistance, geography, and clonal complexes.

After performing MLST and eBURST analyses, we were able to assign each isolate to a clonal complex (table 1). These analyses reveal that 57 of 63 isolates harboring the R variant of PVL are found in the same clonal complex, CC8. The H variant is frequently found in CC121 and CC22 but is dispersed among several clonal complexes. H1 is found predominantly in South Africa and India and is normally associated with CC22. H2 is also common in South Africa and India but is associated with CC121. H3 is common in Europe and is associated with CC121.

Recent evolutionary history of PVL and the origins of PVL-positive MRSA. To reconstruct the evolutionary history of the PVL genes, we used TCS to estimate a phylogenetic network of 174 PVL sequences (figure 2). Branches indicating a single nucleotide substitution connect ellipses representing the PVL alleles sampled in this study. All PVL sequences are connected on the network at a 99% confidence level without the need to include hypothetical missing alleles or loops, pointing to the absence of any ambiguities in connecting the sampled haplotypes. This indicates that the relationships among the individual PVL alleles are fully resolved. On the basis of statistical parsimony, the H variant is predicted to be the ancestor of the R variant. This is consistent with the fact that the H variant has a broader geographic distribution, is spread among several clonal complexes, and contains more genetic variation (11 variant genotypes vs. 2 for the R variants). Statistical parsimony predicts that the H2 variant is the likely ancestral allele that gave rise to the other H variants. This sequence is identical to other published sequences, including USA1100 from Japan and A980470 from France. CC121, which is predominantly MSSA, serves as the reservoir for the H2 variant. Through a combination of clonal expansion and horizontal exchange, the H2 variant spread and gave rise to the H3 and H1 variants, which are found in most PVL-positive isolates around the world; these isolates are predominantly mecA negative. A single nucleotide substitution differentiates the H3 variant from the rare R variant, which is represented by a single sequence among the sampled isolates. This sequence is identical to the lukSF-PV sequence found in the MW2 strain, and the isolate is ST1 and mecA positive, just like MW2. The phylogenetic history of PVL represented in the network allows us to reconstruct the origins of the PVL-positive CA-MRSA strains MW2 and USA300. The MW2 strain originated when a PVLnegative strain from CC1 obtained PVL from a PVL-positive CA-MSSA strain from CC121, either before or after the acquisition of mecA by CC1. The PVL in MW2 was then transferred to a CC8 strain harboring SCCmec type IV, leading to the formation of the USA300 strain. This strain rapidly became the predominant CA-MRSA strain in the United States between its discovery in 2000 and the time of this study (2004–2005).

Functional implications of the histidine to arginine substitution at position 176. Substitution of arginine for histidine changes the physiochemical properties of the site. The side chain of histidine contains an aromatic imidazole ring that can make 2 hydrogen bonds and that, because it has a pKa that ranges from ∼6 to 7, is either neutral or positively charged depending on the local environment [30]. The side chain of arginine is considerably longer, ending in a guanidinium group that can make 5 hydrogen bonds and is always positively charged under physiological conditions (pKa ∼ 12.5 [31]).

Because Jayasinghe and Bayley [32] have proposed that the LukFS-PV pore complex is an octamer consisting of alternating LukF-PV and lukS-PV monomers, we constructed a crude model of such a pore (see Methods for details) to assess the functional relevance of the mutation H176R (H148R in the crystal structure of lukS-PV and in our model). On the basis of this model, it appears that the mutation is located at the surface of LukS-PV, in its interaction surface with lukF-PV (figure 3A and 3B), suggesting that changes at this site may have functional consequences for the PVL toxin and lending itself to several intriguing hypotheses.

First, H176(H148) is predicted to be located near a region of the LukF-PV binding partner with the sequence GNIxSG, a motif conserved among members of the LukF-PV family except for residue x, which is Y (tyrosine) in LukF-PV and N (asparagine) in HlgB and LukD, and both residues are capable of accepting hydrogen bonds. In our model, position x is the LukF-PV residue closest to H176(H148). Perhaps the tyrosine side chain is large enough to form hydrogen bonds with both histidine and arginine, whereas the shorter asparagine side chain can interact only with arginine. It is therefore possible that the arginine allele enables LukS-PV to interact with additional partners besides LukF-PV (e.g., LukD and HlgB) and thus increases the diversity and numbers of pores formed, essentially acting as an amplifier.

Assuming that the mechanism of pore assembly for LukSF-PV is similar to that of α-hemolysin [33], the octamer is assembled at the surface of the membrane after receptor recognition and before the conformational change that results in the formation of the pore itself. Interestingly, position 176(148) is on the same side and near the F163-P175 loop, which corresponds to loop M174-N193 of α-hemolysin, shown to interact specifically with caveolin-1 at the cell surface [34]. Assuming this loop would play a similar role in LukS-PV, position 176(148) may influence recognition of LukS-PV cognate membrane receptor and help initiate oligomer recruitment. Indeed, LukS-PV has been shown to possess a high specific affinity for a unique receptor on PMNs and monocytes [35].

A final hypothesis for functional change at position 176(148) involves an interaction between Arg176(Arg148) and the 3′ end of the LukF-PV subunit. Because of its location in a β sheet, the arginine could adopt a nearly fully extended side chain conformation (the most frequently observed conformation for an arginine located in a β sheet, according to Lovell et al. [36]). This conformation is compatible with the LukS-PV structure, and, after moving M300 to an allowed region of the Ramachandran plot in the model, it could allow the positively charged Arg176(Arg148) to form a strong ionic interaction with the negatively charged carboxy-terminus of LukF-PV (figure 4A). The side chain of His176(His148) does not reach as far and may not be positively charged (figure 4B). It is worth noting that the last 2 residues of monomeric LukF-PV (M300 and S301) have high B factors (indicating increased thermal motion) relative to any other residues of LukF-PV. Therefore, it is possible that the mutation affects the ability of LukS-PV to interact with LukF-PV, thus influencing pore stability or the rate of pore formation.

Previous SectionNext Section

Discussion

Despite being gathered from isolates with diverse histories, including clonal complex membership, geography, and patient demographics, lukSF-PV sequences are highly conserved at the nucleotide level, indicating that they all share a recent common ancestor. The observed patterns of divergence of lukSF-PV sequences, combined with MLST analysis, suggest that genes encoding PVL can rapidly spread via a combination of 2 mechanisms: clonal expansion and horizontal gene transfer. The prevalence of identical R variant sequences in the United States, 90% of which are associated with CC8, reveals that these PVL genes have primarily spread rapidly via a single, highly successful clone. A similar pattern is observed in the H3 variant in Europe, where 83% of sampled isolates are associated with CC121. In contrast, the H1 variant is most often associated with CC22, but CC22 isolates account for only 47% of the H1 variant sequences observed in this study. H1 variant sequences are also found in 8 other clonal complexes. This pattern suggests that CC22 serves as the primary reservoir for H1 variant genes that spread via horizontal gene transfer to other clonal complexes at a relatively high rate. The spread of highly ecologically successful virulent clones is an established property of the population biology of S. aureus [37, 38]. The exchange of mobile genetic elements, particularly when associated with virulence and resistance determinants, is also predicted to be a significant force driving the evolution of the accessory genome [39]. The lukSF-PV gene has been shown to be located within the sequence of a prophage, and the capacity for phage-mediated transfer has been demonstrated in vitro [17]. Our survey of lukSF-PV sequence variation provides evidence supporting both of these mechanisms.

The geographic distribution of the major variants of lukSF-PV is striking. The presence of arginine at position 176 marks 58 of 60 isolates sampled from the United States, including all MRSA collected from the United States. Interestingly, this identical variant is found to be broadly distributed in the United States, including states across the continental United States as well as Hawaii. This variant is associated with a particularly virulent strain of CA-MRSA known as the USA300 clone, which has been shown to have a wide geographic distribution in the United States [4]. Outside the United States, 106 of 111 sampled isolates have histidine at residue 176. All but 5 of these samples are MSSA, and all non-US MRSA strains collected in this study harbor the H variant. Together with the observation that most CC8 isolates are MRSA, this suggests that the acquisition of PVL in a US clone of MRSA led to the development of the highly successful and virulent USA300 strain.

Previous investigations into the evolutionary history of MRSA have concluded that genetically distinct lineages have acquired SCCmec multiple times in independent horizontal transfer events [40, 41]. Data presented here indicates that multiple lineages possess lukSF-PV genes that all derive from a recent common ancestor, indicating that independent horizontal transfer events of PVL also contribute to the evolution of S. aureus. Furthermore, we show evidence that lukSF-PV may be acquired by strains that are predominantly MRSA (e.g., CC8), whereas previous models have predicted that CA-MRSA originated with the acquisition of SCCmec type IV by PVL-positive MSSA [42, 43]. Countries in which the frequency of MRSA is still low may benefit from analysis of MSSA strains that are PVL positive in order to anticipate the spread of lukSF-PV to MRSA clones.

In addition to its utility as an epidemiologic marker, the substitution at aa 176 may have functional implications for the PVL toxin. Molecular modeling suggests that the substitution could enable lukS-PV to form pores with subunits other than lukF-PV, amplifying the rate of pore formation. It could also alter interaction with the yet-unknown cognate cell-surface receptor, required for competent pore formation. Of particular interest is the hypothesis that an arginine at site 176 can extend further than a histidine and thus can interact with the 3′ end of the LukF-PV protein, stabilizing the LukFS-PV interaction. A more stable interaction between the LukS and LukF subunits could allow faster and more efficient pore formation, which could be critical to toxicity, because there is evidence that PVL function is concentration dependent [12]. This proposal is well supported by the molecular model and implies that the R variant is more fit than the H variant.

The relationship between PVL and the fitness of pathogenic clones is unclear at this time, but the connection between PVL and virulence has been the subject of considerable study. PVL is strongly associated with highly virulent strains [44], particularly those that cause necrotizing pneumonia [45]. Indeed, PVL-positive strains and purified PVL toxin have been shown to be sufficient to cause pneumonia in mice [15]. The addition of PVL to a relatively avirulent strain significantly increases PMN apoptosis [12], but other factors affect PMN lysis in more virulent strains [46]. Although the association between PVL and serious skin abscesses in humans [11] is at odds with the observation that PVL does not significantly affect abscesses in mice [46], it should be noted that PVL does not target mouse neutrophils [47] and that staphylococcal virulence in mice generally requires higher concentrations of enterotoxins than those needed in humans [48]. The observation that PVL significantly alters gene expression in S. aureus [15] suggests that PVL may affect the virulence and fitness of the organism in diverse ways. The clinical relevance of PVL in human infections and the functional implications of sequence variants of PVL warrant further investigation.

Analysis of the phylogenetic network of lukSF-PV sequences in light of what we know about their geography, clonality, and mecA status allows us to reconstruct the evolutionary history of the R variant of PVL present in US MRSA, including the USA300 strain. The H variant is older than the R variant. This is supported by its broader geographic distribution, its distribution among several clonal complexes, and its increased sequence diversity relative to that of the R variant. The older variant is more frequently associated with MSSA, whereas the newer variant is associated with CA-MRSA, suggesting that PVL-positive MSSA has acted as a reservoir for PVL genes that are subsequently incorporated into CA-MRSA lineages. Statistical parsimony predicts that the R variants are derived from the H3 variant, which is found in isolates that are all mecA negative, and predominantly CC121. A single nucleotide substitution led to the R variant found in the MW2 lineage, and this PVL was transferred into CC8, leading to the development of USA300, which spread rapidly in the United States.

Using a diversity of approaches, including protein modeling and phylogenetics, we have revealed a geographic variant of the PVL toxin containing an amino acid substitution that may have functional importance. Hopefully, the insights that have been gained into the evolutionary history of current CA-MRSA strains such as USA300 will help to predict and even mitigate the emergence of future strains.

Previous SectionNext Section

Acknowledgments

We thank the GlaxoSmithKline United States—based sequencing facility, under the management of Ganesh Sathe, for its efforts on this project.

Previous SectionNext Section

Footnotes

  • Potential conflicts of interest: none reported.

  • Financial support: GlaxoSmithKline.

  • Received January 25, 2007.
  • Accepted May 22, 2007.
Previous Section
 

References

  1. ↵
    1. Jevons MP
    . “Celbenin”-resistant staphylococci. Br Med J 1961;1:124-5.
    FREE Full Text
  2. ↵
    1. Okuma K,
    2. Ikawaka K,
    3. Turnidge JD,
    4. et al
    . Dissemination of new methicillin-resistant Staphylococcus aureus clones in the community. J Clin Microbiol 2002;40:4289-294.
    Abstract/FREE Full Text
  3. ↵
    1. Ellis MW,
    2. Hospenthal DR,
    3. Dooley DP,
    4. Gray PJ,
    5. Murray CK
    . Natural history of community-acquired methicillin-resistant Staphylococcus aureus colonization and infection in soldiers. Clin Infect Dis 2004;39:971-9.
    Abstract/FREE Full Text
  4. ↵
    1. Moran GJ,
    2. Krishnadasan A,
    3. Gorwitz RJ,
    4. et al
    . Methicillin-resistant S. aureus infections among patients in the emergency department. N Engl J Med 2006;355:666-74.
    CrossRefMedline
  5. ↵
    1. Tenover FC,
    2. McDougal LK,
    3. Goering RV,
    4. et al
    . Characterization of a strain of community-associated methicillin-resistant Staphylococcus aureus widely disseminated in the United States. J Clin Microbiol 2006;44:108-18.
    Abstract/FREE Full Text
  6. ↵
    1. Chavez-Bueno S,
    2. Bozdogan B,
    3. Katz K,
    4. et al
    . Inducible clindamycin resistance and molecular epidemiologic trends of pediatric community-acquired methicillin-resistant Staphylococcus aureus in Dallas, Texas. Antimicrob Agents Chemother 2005;49:2283-8.
    Abstract/FREE Full Text
  7. ↵
    1. Seybold U,
    2. Kourbatova EV,
    3. Johnson JG,
    4. et al
    . Emergence of community-associated methicillin-resistant Staphylococcus aureus USA300 genotype as a major cause of health care—associated blood stream infections. Clin Infect Dis 2006;42:647-56.
    Abstract/FREE Full Text
  8. ↵
    1. Baba T,
    2. Takeuchi F,
    3. Kuroda M,
    4. et al
    . Genome and virulence determinants of high virulence community-acquired MRSA. Lancet 2002;359:1819-27.
    CrossRefMedlineWeb of Science
  9. ↵
    1. Francis JS,
    2. Doherty MC,
    3. Lopatin U,
    4. et al
    . Severe community-onset pneumonia in healthy adults caused by methicillin-resistant Staphylococcus aureus carrying the Panton-Valentine leukocidin genes. Clin Infect Dis 2005;40:100-7.
    Abstract/FREE Full Text
  10. ↵
    1. Dufour P,
    2. Gillet Y,
    3. Bes M,
    4. et al
    . Community-acquired methicillin-resistant Staphylococcus aureus infections in France: emergence of a single clone that produces Panton-Valentine leukocidin. Clin Infect Dis 2002;35:819-24.
    Abstract/FREE Full Text
  11. ↵
    1. Miller LG,
    2. Perdreau-Remington F,
    3. Rieg G,
    4. et al
    . Necrotizing fasciitis caused by community-associated methicillin-resistant Staphylococcus aureus in Los Angeles. N Engl J Med 2005;352:1445-53.
    CrossRefMedlineWeb of Science
  12. ↵
    1. Genestier AL,
    2. Michallet MC,
    3. Prevost G,
    4. et al
    . Staphylococcus aureus Panton-Valentine leukocidin directly targets mitochondria and induces Bax-independent apoptosis of human neutrophils. J Clin Invest 2005;115:3117-27.
    CrossRefMedlineWeb of Science
  13. ↵
    1. Gladstone GP,
    2. Van Heyningen WE
    . Staphylococcal leucocidins. Br J Exp Pathol 1957;38:123-37.
    MedlineWeb of Science
  14. ↵
    1. Baba Moussa L,
    2. Werner S,
    3. Colin DA,
    4. et al
    . Discoupling the Ca2+-activation from the pore-forming function of the bi-component Panton-Valentine leucocidin in human PMNs. FEBS Lett 1999;461:280-6.
    CrossRefMedlineWeb of Science
  15. ↵
    1. Labandiera-Rey M,
    2. Couzon F,
    3. Boisset S,
    4. et al
    . Staphylococcus aureus Panton-Valentine leukocidin causes necrotizing pneumonia. Science 2007;315:1130-3.
    Abstract/FREE Full Text
  16. ↵
    1. Vandenesch F,
    2. Naimi T,
    3. Enright MC,
    4. et al
    . Community-acquired methicillin-resistant Staphylococcus aureus carrying Panton-Valentine leukocidin genes: worldwide emergence. Emerg Infect Dis 2003;9:978-84.
    MedlineWeb of Science
  17. ↵
    1. Narita S,
    2. Kaneko J,
    3. Chiba J,
    4. et al
    . Phage conversion of Panton-Valentine leukocidin in Staphylococcus aureus: molecular analysis of a PVLconverting phage, phiSLT. Gene 2001;268:195-206.
    CrossRefMedlineWeb of Science
  18. ↵
    1. West J,
    2. Ray R,
    3. Scangarella N,
    4. Shawar R
    . Abstracts of the 45th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, DC: American Society for Microbiology; 2005. Antibacterial activity of retapamulin against 109 MRSA skin infection isolates with characterized SCCmec and PVL genotypes recovered from phase III clinical trials [abstract F-2054].
  19. ↵
    1. Enright MC,
    2. Day NP,
    3. Davies CE,
    4. Peacock SJ,
    5. Spratt BG
    . Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. J Clin Microbiol 2000;38:1008-15.
    Abstract/FREE Full Text
  20. ↵
    1. Feil EJ,
    2. Li BC,
    3. Aanensen DM,
    4. Hanage WP,
    5. Spratt BG
    . eBURST: inferring patterns of evolutionary descent among clusters of related bacterial genotypes from multilocus sequence typing data. J Bacteriol 2004;186:1518-30.
    Abstract/FREE Full Text
  21. ↵
    1. Spratt BG,
    2. Hanage WP,
    3. Li B,
    4. Aanensen DM,
    5. Feil EJ
    . Displaying the relatedness among isolates of bacterial species: the eBURST approach. FEMS Microbiol Lett 2004;241:129-34.
    CrossRefMedlineWeb of Science
  22. ↵
    1. Higgins D,
    2. Thompson J,
    3. Gibson T
    . CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 1994;22:4673-80.
    Abstract/FREE Full Text
  23. ↵
    1. Nicholas KB,
    2. Nicholas HB,
    3. Deerfield DW II.
    . GeneDoc: analysis and visualization of genetic variation. EMBNEW.NEWS 1997;4:14.
  24. ↵
    1. Clement M,
    2. Posada D,
    3. Crandall K
    . TCS: a computer program to estimate gene genealogies. Mol Ecol 2000;9:1657-60.
    CrossRefMedline
  25. ↵
    1. Song L,
    2. Hobaugh MR,
    3. Shustak C,
    4. Cheley S,
    5. Bayley H,
    6. Gouaux JE
    . Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore. Science 1996;274:1850-1.
    FREE Full Text
  26. ↵
    1. Guex N,
    2. Peitsch MC
    . SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 1997;18:2714-23.
    CrossRefMedlineWeb of Science
  27. ↵
    1. Richardson DC,
    2. Richardson JS
    . The kinemage: a tool for scientific communication. Protein Sci 1992;1:3-9.
    MedlineWeb of Science
  28. ↵
    1. Pedelacq JD,
    2. Maveyraud L,
    3. Prevost G,
    4. et al
    . The structure of a Staphylococcus aureus leucocidin component (LukF-PV) reveals the fold of the water-soluble species of a family of transmembrane pore-forming toxins. Structure 1999;7:277-87.
    Medline
  29. ↵
    1. Guillet V,
    2. Roblin P,
    3. Werner S,
    4. et al
    . Crystal structure of leucotoxin S component: new insight into the staphylococcal beta-barrel poreforming toxins. J Biol Chem 2004;279:41028-37.
    Abstract/FREE Full Text
  30. ↵
    1. Edgcomb SP,
    2. Murphy KP
    . Variability in the pKa of histidine side-chain correlates with burial within proteins. Proteins 2002;49:1-6.
    CrossRefMedlineWeb of Science
  31. ↵
    1. Lehninger AL,
    2. Nelson DL,
    3. Cox MM
    . Principles of biochemistry. 2nd ed. New York: Worth Publishers; 1993. p. 13.
  32. ↵
    1. Jayasinghe L,
    2. Bayley H
    . The leukocidin pore: evidence for an octamer with four LukF subunits and four LukS subunits alternating around a central axis. Protein Sci 2005;14:2550-61.
    CrossRefMedlineWeb of Science
  33. ↵
    1. Kawate T,
    2. Gouaux E
    . Arresting and releasing staphylococcal alphahemolysin at intermediate stages of pore formation by engineered disulfide bonds. Protein Sci 2003;12:997-1006.
    CrossRefMedlineWeb of Science
  34. ↵
    1. Pany S,
    2. Vijayvargia R,
    3. Krishnasastry MV
    . Caveolin-1 binding motif of alpha-hemolysin: its role in stability and pore formation. Biochem Biophys Res Commun 2004;322:29-36.
    CrossRefMedlineWeb of Science
  35. ↵
    1. Gauduchon V,
    2. Werner S,
    3. Prevost G,
    4. Monteil H,
    5. Colin DA
    . Flow cytometric determination of Panton-Valentine leucocidin S component binding. Infect Immun 2001;69:2390-5.
    Abstract/FREE Full Text
  36. ↵
    1. Lovell SC,
    2. Word JM,
    3. Richardson JS,
    4. Richardson DC
    . The penultimate rotamer library. Proteins 2000;40:389-408.
    CrossRefMedlineWeb of Science
  37. ↵
    1. Stemper ME,
    2. Shukla SK,
    3. Reed KD
    . Emergence and spread of community-associated methicillin-resistant Staphylococcus aureus in rural Wisconsin, 1989 to 1999. J Clin Microbiol 2004;42:5673-80.
    Abstract/FREE Full Text
  38. ↵
    1. Aires de Sousa M,
    2. Sanches IS,
    3. Ferro ML,
    4. et al
    . Intercontinental spread of a multidrug-resistant methicillin-resistant Staphylococcus aureus clone. J Clin Microbiol 1998;36:2590-6.
    Abstract/FREE Full Text
  39. ↵
    1. Lindsay J,
    2. Holden MT
    . Staphylococcus aureus: superbug, super genome? Trends Microbiol 2004;12:378-85.
    CrossRefMedlineWeb of Science
  40. ↵
    1. Musser JM,
    2. Kapur V
    . Clonal analysis of methicillin-resistant Staphylococcus aureus strains from intercontinental sources: association of the mec gene with divergent phylogenetic lineages implies dissemination by horizontal transfer and recombination. J Clin Microbiol 1992;30:2058-63.
    Abstract/FREE Full Text
  41. ↵
    1. Deurenberg RH,
    2. Vink C,
    3. Kalenic S,
    4. Friedrich AW,
    5. Bruggeman CA,
    6. Stobberingh EE
    . The molecular evolution of methicillin-resistant Staphylococcus aureus. Clin Microbiol Infect 2007;13:222-35.
    CrossRefMedlineWeb of Science
  42. ↵
    1. Boyle-Vavra S,
    2. Daum RS
    . Community-acquired methicillin-resistant Staphylococcus aureus: the role of Panton-Valentine leukocidin. Lab Invest 2007;87:3-9.
    CrossRefMedlineWeb of Science
  43. ↵
    1. Robinson DA,
    2. Kearns AM,
    3. Holmes A,
    4. et al
    . Re-emergence of early pandemic Staphylococcus aureus as a community-acquired meticillinresistant clone. Lancet 2005;365:1256-8.
    CrossRefMedlineWeb of Science
  44. ↵
    1. Melles DC,
    2. Gorkink RF,
    3. Boelens HA,
    4. et al
    . Natural population dynamics and expansion of pathogenic clones of Staphylococcus aureus. J Clin Invest 2004;114:1732-40.
    CrossRefMedlineWeb of Science
  45. ↵
    1. Gillet Y,
    2. Issartel B,
    3. Vanhems P,
    4. et al
    . Association between Staphylococcus aureus strains carrying gene for Panton-Valentine leukocidin and highly lethal necrotising pneumonia in young immunocompetent patients. Lancet 2002;359:753-9.
    CrossRefMedlineWeb of Science
  46. ↵
    1. Voyich JM,
    2. Otto M,
    3. Mathema B,
    4. et al
    . Is Panton-Valentine leukocidin the major virulence determinant in community-associated methicillinresistant Staphylococcus aureus? J Infect Dis 2006;194:1761-70.
    Abstract/FREE Full Text
  47. ↵
    1. Fischetti V,
    2. Novick R,
    3. Ferretti J,
    4. Portnoy D,
    5. Rood J
    1. Bohach GA
    . Staphylococcus aureus exotoxins. In: Fischetti V, Novick R, Ferretti J, Portnoy D, Rood J, editors. Gram-positive pathogens. 2nd ed. Washington, DC: American Society for Microbiology; 2006. p. 464-77.
  48. ↵
    1. Tarkowski A,
    2. Collins LV,
    3. Gjertsson I,
    4. et al
    . Model systems: modeling human staphylococcal arthritis and sepsis in the mouse. Trends Microbiol 2001;9:321-6.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. J Infect Dis. (2008) 197 (2): 187-194. doi: 10.1086/524684
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most