Community-Associated Methicillin-Resistant Staphylococcus aureus Epidemic Clone USA300 in Isolates from Florida and Washington

Recent publications have suggested that community-acquired methicillin-resistant Staphylococcus aureus (CA-MRSA) infections are becoming commonplace and that particular S. aureus clones feature prominently in epidemics occurring in the United States (1, 2, 6, 7). A specific MRSA clone that is multilocus sequence type 8 (ST8) and pulsotypes, designated USA300 by the Centers for Disease Control and Prevention (6), has been identified in community outbreaks of MRSA wound infections in California, Georgia, Texas, Pennsylvania, Mississippi, and Minnesota (1, 4, 7). USA300 carries the type IV staphylococcal methicillin resistance gene cassette (SCCmec IV) (6) and produces Panton-Valentine leucocidin, a virulence factor that is linked to cases of community-acquired necrotizing pneumonia (3, 5, 10). In the present study, we examined community-associated MRSA isolates collected from individuals in Florida and Washington State with a variety of disease etiologies by pulsed-field gel electrophoresis (PFGE) for the presence of the epidemic USA300 clone.

A total of 299 community-associated methicillin-resistant S. aureus organisms isolated from wounds, noses, sputa, blood, and other sources were used in this study. CA-MRSA isolates were requested from two hospitals in Washington, one hospital in Florida, and a reference laboratory in Florida. All isolates were collected from patients treated on an outpatient basis, within 48 h of consultation, and only one isolate was obtained per patient. The control strain, a USA300 (NRS384), was obtained from the Network on Antimicrobial Resistance in S. aureus.

PFGE was performed as previously reported for enterococcal isolates (8) with the following changes for S. aureus. Overnight cultures of S. aureus were grown in BBL Trypticase soy broth (Becton Dickinson, Sparks, MD). Lysostaphin (Sigma, St. Louis, MO) was added to the cell lysis buffer (6 mM Tris-HCl, 1 M NaCl, 100 mM EDTA, 0.5% Brij-58, 0.2% sodium deoxycholate, 0.5% sodium lauroyl sarcosine) at a concentration of 3 U/ml, and plugs were lysed at 37°C, followed by overnight incubation in a solution containing 0.5 M EDTA (pH 9), 1% sodium lauroyl sarcosine, and 50 μg/ml proteinase K at 50°C. Plugs were digested in SmaI overnight, melted at 69°C, and loaded onto a 1% Seakem Gold (Cambrex BioScience, Rockland, ME) gel. PFGE was performed using a DR-II CHEF Mapper (Bio-Rad, Hercules, CA) using the following parameters: 200 V, 14°C, 5.3-s initial switch, 34.9-s final switch, and 20-h run time. Data were analyzed using BioNumerics (Applied Math, Sint-Martens, Belgium). Dendrograms were derived from unweighted-pair group method using arithmetic averages (UPGMA) and based on Dice coefficients. PCR for the pvl genes was performed as previously reported (5).

A number of studies have demonstrated an epidemiologic link between the USA300 epidemic clone and CA-MRSA infections (1, 4, 6, 7). These studies have demonstrated the presence of this isolate in wound infections in six states (1, 4, 6, 7). In our study, we used PFGE to examine CA-MRSA isolates from Florida and Washington State for the presence of the USA300 clone. While our collection of 299 isolates contained 39 isolates from Washington, the majority (260, 87%) of the isolates were collected in Florida. USA300, as designated by the CDC, includes two closely related PFGE patterns, or pulsotypes (6). In the present study, we found that 44% (17 of 39) of our Washington isolates and 42% (111 of 260) of our Florida CA-MRSA isolates have the same pulsotypes as USA300, using SmaI digestion, demonstrating the presence of the epidemic clone in these two states (Fig. 1). Although the control strain represents only one of two possible USA300 patterns, a one-band difference as shown in Fig. 1 (CBD739) is still considered identical by widely accepted criteria (9). Furthermore, the CBD739 pulsotype appears to be identical to the alternative USA300 pattern represented in publication (6) but not available as a control strain.

Previous studies have demonstrated that the USA300 isolates contain the Panton-Valentine leucocidin (pvl) genes which have been epidemiologically linked to severe skin infections and necrotizing pneumonia (3, 5, 10). The pvl genes encode membrane toxins which target leukocytes, and therefore, not surprisingly, the studies of USA300 have primarily involved isolates from wound infections (1, 4, 7). To expand the investigation of the spread of this epidemic clone, we included in our pulsed-field analysis, in addition to wound isolates, organisms isolated from other sites (Table 1). Using BioNumerics software (Fig. 1), we found that the majority of CA-MRSA isolates were isolated from wounds, and the majority of these isolates (64%) belonged to the USA300 pulsotypes (Table 1). Although USA300 pulsotypes were present in different types of infections, e.g., blood (12%), nose (15%), sputum (10%), and other (5%) (Fig. 1), they accounted for a small number of these isolates. PCR results demonstrated that 100% of our USA300 isolates carry the pvl locus (data not shown). In addition to having methicillin resistance, 97% of the USA300 isolates tested were resistant to macrolides, and 22% were resistant to clindamycin (Vicki Luna, personal communication). Multilocus sequence typing performed on select USA300 isolates confirmed that all are ST8 as previously reported (6) (data not shown). These results further confirm earlier studies that showed that there is a strong correlation between the USA300 pulsotypes and wound infections (1, 4, 7). Additionally, these results demonstrate that while these pulsotypes are not unique to wound infections, they are uncommon in the other S. aureus infections included in this study.

The remaining 149 isolates on which PFGE was performed included 1 isolate that was a match to the USA400 epidemic clone, several isolates that are closely related to USA100, and several isolates with unique patterns (data not shown). The majority of the USA100-like isolates in our collection were obtained from nasal swabs, and 99% of these isolates were PCR negative for the pvl locus, but it is unknown if they were present as a result of infection or colonization.

In conclusion, the epidemic CA-MRSA USA300/ST8 clone, as previously reported in six states, is also present in Florida and Washington. We have further strengthened the evidence for an epidemiological link between USA300 and wound infections by demonstrating that these pulsotypes cause the majority of CA-MRSA wound infections in our database and all are positive for the pvl genes. We have also demonstrated the presence of USA300 in isolates from blood, sputum, nose, and conjunctiva (Table 1), although they accounted for a minority of the isolates.

• Open in new tab
• Download powerpoint

FIG. 1.

Dendrogram of SmaI-PFGE of community-associated S. aureus isolates. These isolates represent five disease etiologies and were isolated from two states as indicated. USA300 strains are characterized by two closely related patterns as shown above by the following two isolate types: USA300 (control strain obtained from the Network on Antimicrobial Resistance in S. aureus) and CBD739.

• Copyright © 2006 American Society for Microbiology