Abstract
Between December 2009 and November 2011, we collected 57 (12.3%) Staphylococcus aureus isolates from 464 pigs and 16 (30.8%) isolates from 52 farmers in the largest farm in Dakar. Fifty-one isolates (70%) belonged to four major multilocus sequence typing clonal complexes (CCs): CC152 (26.0%), CC15 (19.2%), CC5 (13.7%), and CC97 (10.9%). The CC variability among the pigs was similar to that observed among the farmers. Six isolates that were recovered only among pigs were resistant to methicillin (10.5%). They were assigned to the ST5-staphylococcal cassette chromosome mec type (SCCmec) IV (n=5) and ST88-SCCmec IV (n=1) clones. The luk-PV genes encoding Panton-Valentine leukocidin (PVL), present in 43 (58.9%) isolates overall, including all major CCs and the MRSA ST5-SCCmec IV clone, were highly prevalent compared to data from industrialized countries. This finding is of major concern with regard to the potential virulence of these strains.
Introduction
W
Materials and Methods
Study population
Approximately 300 fattening female and male pigs were present in the farm, with a turnover of approximately 150 pigs per month. About 50 pigs were randomly investigated every 3 months between December 2009 and November 2011. Nasal swabs were collected from pigs and farmers. For humans, a standardized specific questionnaire was completed to collect demographic data, medical history over the previous 12 months, and information about contact with pigs.
Microbiological analysis, DNA extraction, mecA detection, toxin gene profiling, and genotyping
The swabs were immediately placed at +4°C and analyzed within 4 h. Briefly, swabs were inoculated in a pre-enrichment medium consisting of brain-heart broth containing 7.5% NaCl. After overnight aerobic incubation at 37°C, selective enrichment was performed in brain-heart broth (bioMérieux) supplemented with colistin (4 mg/L). Twenty-five microliters of the selective enrichment broth was inoculated on specific agar plates (BBL CHROMagar™ Staph aureus and BBL CHROMagar™ MRSA; Becton Deckinson, Franklin Lakes, NJ). S. aureus identification, antimicrobial susceptibility, genomic extraction, screening in all S. aureus isolates for the mecA, sea, sec, sed, seh, sem, sep, ser, tst, eta, etb, luk-PV, and lukM genes, genotyping by MLST and spa typing, and SCCmec typing for mecA-positive strains were performed as previously described (Breurec et al., 2011a).
Statistical analysis
The chi-square test and Wilcoxon's test were used to compare categorical and continuous variables, respectively, using R software. Significance was assumed at p<0.05.
Results
S. aureus was recovered from 57 (12.3%) of the 464 pigs screened and from 16 (30.8%) of the 52 farmers (male/female ratio 1.08; median age, 45 years). Six isolates found only among pigs were resistant to methicillin (10.5%). In addition, strains were characterized by a high level susceptibility to most of the antibiotics tested, except for penicillin (100.0% of resistance), co-trimoxazole (47.9%), and tetracycline (27.4%; Table 1). Univariate analysis identified no risk factors for human S. aureus carriage (Table 2).
spa types associated with MRSA isolates (t311 correspond to ST5 staphylococcal cassette chromosome mec [SCCmec] IV and t3489 to ST88 SCmec IV).
All strains were susceptible to kanamycin, tobramycin, gentamicin, lincomycin, pristinamycin, fosfomycin, fusidic acid, pefloxacin, and vancomycin. Furthermore, all strains were susceptible to oxacillin and cefoxitin, except for six strains isolated from pigs.
etb, hlb, and lukM were not detected in any of the isolates.
hlb, lukM, sec, sed, and ser were not detected in any of the isolates.
spa CC, staphylococcal protein A clonal complexe; MLST CC, multilocus sequence typing clonal complex; MLST ST, multilocus sequence typing sequence type; MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-susceptible Staphylococcus aureus; eta, exfoliatin A; etb, exfoliatin B; luk-PV, Panton–Valentine leukocidin; se, staphylococcal enterotoxin; tst, toxic shock syndrome toxin; ERY, erythromycin; FOX, cefoxitin; OXA, oxacillin; RIF, rifampicin; PEN, penicillin; SXT, co-trimoxazole; TET, tetracycline.
Significance was assumed at p<0.05.
The 73 isolates belonged to 22 spa types that were separated by the Based Upon Repeat Pattern (BURP) algorithm into eight spa clonal complexes (CCs) and seven singletons (groups represented by a single spa type). Each spa CC corresponded to a single multilocus sequence typing (MLST) CC. Seventy percent belonged to four major MLST CCs: CC152 (26.0%), CC15 (19.2%), CC5 (13.7%), and CC97 (10.9%). The CC variability among the pigs was similar to that observed among the farmers (Table 1). MRSA isolates (n=6) found only among the pigs were assigned to the ST5-staphylococcal cassette chromosome mec type (SCCmec) IV (n=5, spa type t311) and ST88-SCCmec IV (n=1, spa type t3489) clones.
The luk-PV genes encoding Panton–Valentine leukocidin (PVL) were present in 49.1% (n=36) and 43.8% (n=7) of porcine and human isolates, respectively (10 different spa-types, regardless of the host, including all major CCs and the MRSA ST5-SCCmec IV clone).
Discussion
MRSA prevalence among pigs was far lower than that reported in industrialized countries. This may reflect the lesser use of antibiotics for animal growth and therapy in Senegal (unpublished data), as well as the absence of evidence of the ST398 MRSA clone common in pigs in developed countries (Espinosa-Gongora et al., 2012; Morcillo et al., 2012; van der Wolf et al., 2012). To the best of our knowledge, no other data are available on the prevalence of MRSA carriage among pigs and pig farmers in Africa.
The MRSA clones found among the pigs, ST5-SCCmec IV and ST88-SCCmec IV, have already been described as major MRSA lineages responsible for human infections in Dakar (Breurec et al., 2011a). As no MRSA clones were recovered from the farmers, longitudinal epidemiological studies among pigs, farmers and family members are needed to investigate the interspecies transmission of these clones. The presence of luk-PV genes encoding PVL in ST5 MRSA is potentially of particular concern as the presence of this toxin has been linked to deep abscesses, severe necrotizing pneumonia, and severe bone and joint infections in human epidemiological studies (Breurec et al., 2011a).
Apart from CC97, the major MSSA CCs (CC15 and CC152) found among the pigs and the farmers have already been described as major MSSA lineages responsible for human infections in Dakar. Lineage CC97 is a pandemic bovine S. aureus lineage frequently responsible for mastitis but rarely detected in swine (Aires-de-Sousa et al., 2007).
Our data, together with those of a previous study on human infections in Dakar (Breurec et al., 2011b), indicate that Senegal has a far higher prevalence of PVL-positive S. aureus than industrialized countries (0–2% of human or swine nasal carriage isolates) (O'Hara et al., 2008), whatever the origin of the isolates. The reasons for such a high prevalence are currently unclear, but the increased virulence of these strains, combined with factors that increase the risk of interindividual transmission (e.g., poor hygienic and sanitary conditions, and overcrowding), enhances the transmission potential of S. aureus (Massey et al., 2006).
Conclusion
These data indicate a low prevalence of MRSA strains among pigs and pig farmers. Ongoing MRSA surveillance in animals, farmers, and family members is needed to study the transmission between species and detect changes in epidemiology.
Footnotes
Acknowledgments
We thank Fatou Bintou Dieye and Clotaire Rafai (Institut Pasteur, Dakar, Senegal), as well as Helene Meugnier and Michele Bes (National Reference Center for Staphylococci, Lyon, France) for their technical help. We are grateful to David Young for editorial assistance. We also thank all the veterinarians and farmers involved in this study, particularly Mouhamadou M. Diagne, Algor Cissé, and Jean Robert Diatta. This study was supported by local funds from Institute Pasteur in Dakar.
Disclosure Statement
No competing financial interests exist.