Methicillin-resistant Staphylococcus aureus (MRSA) ST398 in a farmer with skin lesions and in pigs of his farm: clonal relationship and detection of lnu(A) gene


Corresponding author: Carmen Torres, Área Bioquímica y Biología Molecular, Universidad de La Rioja, Madre de Dios 51, 26006 Logroño, Spain


Clin Microbiol Infect 2011; 17: 923–927


Skin infection associated with methicillin-resistant Staphylococcus aureus (MRSA)-ST398 was detected in a pig-farmer, and MRSA-ST398 isolates were also detected in nasal samples of the patient and of 11/12 pigs on his farm. Twelve MRSA isolates were obtained from skin lesions (n = 6) and nasal samples (n = 6) of the patient in two sampling moments and 11 MRSA isolates from nasal samples of pigs. They were typed as t011-SCCmecIVa-agrI and t108-SCCmecV-agrI (patient and pigs) and t588-SCCmecV-agrI (patient). The following resistance genes were detected (number isolates): tet(K) (1), tet(L) (23), tet(M) (13), erm(A) (13), erm(C) (13), msr(A) (11), lnu(A) (21), aph(2′′)-acc(6′) (3), ant(4′) (13), aph(3′) (12), dfrS1 (15) and dfrK (22). Seventeen human and animal MRSA-ST398 isolates showed indistinguishable PFGE patterns (A1-spa-t011 or B2-spa-t108) and similar phenotypic-genotypic characteristics, including the presence of the lnu(A) gene, associated with lincomycin resistance. Potential pig-to-human transference of ST398 is suggested in this study. The first detection of the lnu(A) gene in MRSA-ST398 is reported.

Methicillin-resistant Staphylococcus aureus (MRSA) of sequence type ST398 has been identified either colonizing or causing infections in animals and humans, and has been found to cause diseases related to occupational contact with pigs. The prevalence of colonization by these strains in those who work with pigs has been proved to be higher than in people without this risk factor [1]. MRSA-ST398 is characterized by its resistance to tetracycline, one of the main antibiotics used in pig farming. Moreover, an unusual resistance phenotype showing clindamycin resistance but erythromycin susceptibility has been reported in a few MRSA-ST398 isolates [2–5]. Recently, this phenotype has been associated in some strains with the presence of plasmid-borne resistance genes vga(A) or vga(C) [6]. The expression of lincomycin-modifying enzymes could confer resistance to lincomycin but not to erythromycin, and this mechanism is infrequent in S. aureus, and has not been reported so far in MRSA-ST398 isolates. The aim of this study was to perform the genetic characterization of 23 MRSA isolates (12 recovered from a patient with skin lesions and 11 from pigs on the farm where the patient worked) in order to determine the possible MRSA animal-to-human transmission, and to characterize the mechanism of lincomycin resistance detected in 21 of 23 of these MRSA isolates (12 from a human and nine from pigs).

A 54-year-old patient, who worked on a pig farm in Spain, presented skin lesions, diagnosed as psoriasis by the dermatologist. In order to know whether superinfection with bacteria or fungi could be also involved, three samples from lesions (scalp, dorsum of the nose and chin) and moreover two from both nasal ducts were obtained (October 2008, period 1) and submitted to microbiological analysis. Heavy growth of MRSA isolates was obtained in all five tested samples of the patient and they were identified by classical microbiological tests and confirmed by detection of nuc and mecA genes by PCR [7]. Five MRSA isolates (one per sample) were further characterized.

Nasal samples of two family members of the patient (wife and son) who did not live close to the farm and had no contact with pigs were also tested in October 2008, and they were negative for S. aureus and MRSA.

The patient was treated with an antiseptic gel for scalp and beard washing, a metilprednisole solution and a cream with betametasone and fusidic acid for cutaneous lesions, and nasal MRSA decolonization with mupirocin. The patient had some clinical improvement of the lesions in the next weeks, although they were not resolved, and he returned to a new control with his dermatologist 3 months later (January 2009, period 2). New samples were then obtained from three skin lesions (ear canal, hand and chin) and from both nasal ducts, and they were submitted to microbiological study. MRSA isolates were recovered again from skin samples (three isolates, one isolate per sample) and nasal samples (four isolates, two isolates per sample).

In addition, nasal swabs from 12 pigs on the farm where the patient worked were randomly obtained (January 2009, period 2). They were inoculated in brain-heart-infusion-broth (BHI, BD) containing 6.5% NaCl and incubated at 35°C for 24 h, and later they were seeded on oxacillin-resistance-screening agar plates (ORSAB, Oxoid) with oxacillin (2 mg/L), and were incubated at 35°C for 36 h. MRSA was detected in 11 of these 12 animals (91.6%), and one MRSA isolate per animal was further characterized.

Antimicrobial susceptibility to 23 antibiotics was determined by the VITEK-2 system (bioMérieux, Marcy l’ Etoile, France) and by disk-diffusion assay in the 23 recovered MRSA isolates (12 from the patient and 11 from the pigs) following CLSI recommendations [8]. The tested antibiotics were: penicillin, oxacillin, cefoxitin, tetracycline, erythromycin, telithromycin, clindamycin, lincomycin, gentamicin, tobramycin, sulfamethoxazole-trimethoprim, ciprofloxacin, levofloxacin, vancomycin, teicoplanin, quinupristin/dalfopristin, linezolid, fosfomycin, fusidic acid, mupirocin, nitrofurantoin, rifampin and kanamycin. Staphylococcus aureus ATCC 29213 and Enterococcus faecalis ATCC 29212 were used as control strains. Lincomycin and clindamycin susceptibility was also studied by the agar dilution method [8,9]. CLSI breakpoints [8] were used for all antibiotics, with two exceptions: lincomycin breakpoints were considered as recommended by the Societé Française de Microbiologie [9], and mupirocin breakpoints as considered by Oliveria et al. 2007 [10]. The presence of erm(A), erm(B), erm(C), msr(A), lnu(A), lnu(B), lnu(C), lnu(D), cfr, vga(C), lsa(B), tet(K), tet(L), tet(M), tet(O), aph(2′′)-acc(6′), ant(4′), aph(3′), dfrS1, dfrD and dfrK antimicrobial resistance genes was studied by PCR and sequencing [7,11]. MRSA isolates of human and animal origin were typed (PFGE, MLST, SCCmec-, spa- and agr-typing) as previously described [7,12–14]. Characteristics of MRSA of human and animal origin are included in Table 1.

Table 1.   Characteristics of the MRSA-ST398 isolates recovered from the patient and from pigs on the farm
StrainOriginPeriodaspa/SCCmecagrPFGE patternMIC (mg/L)Antimicrobial resistance phenotype for non-betalactamsResistance genes detected
  1. TET, tetracycline; ERY, erythromycin; TEL, telithromycin; CLI, clindamycin; LIN, lincomycin; TOB, tobramycin; KAN, kanamycin; SXT, sulfamethoxazole-trimethoprim; CIP, ciprofloxacin, LVX, levofloxacin; I, intermediate. All isolates were susceptible to vancomycin, teicoplanin, quinupristin/dalfopristin, linezolid, fosfomycin, fusidic acid, mupirocin, nitrofurantoin and rifampin.aSamples: (1) first sampling period (October 2008), skin lesions and nasal samples of the patient; (2) second sampling period (January 2009), skin lesions of the patient and nasal samples of the patient and of the pigs.

 C1676Chin1t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXT-CIPI-LVXItet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′);, aph(3′), dfrS1, dfrK
 C1677Dorsum of nose1t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXT-CIPI-LVXItet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1675Scalp1t011/IVaIA2>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1673Nasal1t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXT-CIPI-LVXItet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1674Nasal1t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1870Ear canal2t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1871Chin2t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1 , dfrK
 C1872Hand2t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1856Nasal2t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A),lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1857Nasal2t108/VIB1>128>128TET-ERY-TEL-CLI-LIN-TOB-SXTtet(L), erm(A), erm(C), lnu(A), dfrK
 C1859Nasal2t108/VIB20.5>128TET-LIN-SXTtet(L), lnu(A), dfrK
 C1858Nasal2t588/VIC>128>128TET-ERY-TEL-CLI-LIN-TOB-SXTtet(K), tet(L), erm(A), erm(C), msr(A), lnu(A), dfrS1, dfrK
 C1842Nasal-12t011/IVaIA1>128>128TET-ERY-TEL-CLI-LIN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), ant(4′), aph(3′), dfrS1, dfrK
 C1844Nasal-22t011/IVaIA3>128>128TET-ERY-TEL-CLI-LIN-GEN-TOB-KANI-SXTtet(L), tet(M), erm(A), erm(C), msr(A), lnu(A), aph(2′′)-acc(6′), ant(4´), aph(3′), dfrS1, dfrK
 C1849Nasal-32t011/IVaIA40.1251TET-GEN-TOB-KAN-SXTtet(L), tet(M), aph(2′′)-acc(6′), ant(4′), aph(3′), dfrS1, dfrG, dfrK
 C1850Nasal-42t011/IVaIA50.1251TET-GEN-TOB-KANtet(L), tet(M), aph(2′′)-acc(6′), ant(4′), aph(3′)
 C1840Nasal-52t108/VIB20.5>128TET-LIN-SXT-CIPI-LVXItet(L), lnu(A), dfrS1, dfrK
 C1841Nasal-62t108/VIB20.5>128TET-LIN-SXTtet(L), lnu(A), dfrK
 C1843Nasal-72t108/VIB20.5>128TET-LIN-SXTtet(L), lnu(A), dfrK
 C1845Nasal-82t108/VIB20.5>128TET-LIN-SXTtet(L), lnu(A), dfrK
 C1846Nasal-92t108/VIB20.5>128TET-LIN-SXT-CIPI-LVXItet(L), lnu(A), dfrS1, dfrK
 C1847Nasal-102t108/VIB20.5>128TET-LIN-SXTtet(L), lnu(A), dfrK
 C1848Nasal-112t108/VIB20.5>128TET-LIN-SXTtet(L), lnu(A), dfrK

All five MRSA isolates recovered from the patient in period 1 before treatment (three from skin lesions and two from nasal samples) and four of the seven isolates obtained after the treatment (three from lesions and one from nasal sample) were typed as MRSA-ST398-t011, SCCmec-IVa and agr-I and showed resistance to tetracycline, erythromycin, telithromycin, clindamycin, lincomycin, tobramycin, kanamycin and sulfamethoxazole-trimethoprim (Table 1).

The remaining three MRSA nasal isolates obtained from the patient after treatment were typed as ST398-t108-SCCmecV-agrI (two isolates) and ST398-t588-SCCmecV-agrI (one isolate), and showed different antimicrobial resistance phenotypes and genotypes (Table 1).

Eleven MRSA isolates were recovered from pigs. Seven of them were typed as ST398-t108-SCCmecV-agrI and four MRSA isolates of pig origin were typed as ST398-t011-SCCmecIVa-agrI (Table 1).

Twenty-one of 23 MRSA isolates obtained in this study (12/12 isolates from the patient and 9/11 isolates of animal origin) presented a high level of resistance to lincomycin and harboured the lnu(A) gene. Thirteen of these isolates showed resistance to erythromycin/clindamycin/lincomycin (ST398-t011-SCCmecIVa-agrI, ST398-t108-SCCmecV-agrI and ST398-t588-SCCmecV-agrI) and harboured the erm(A) and erm(C) genes, in addition to the lnu(A) gene; the remaining eight isolates presented an unusual erythromycin/clindamycin-susceptibility and lincomycin-resistance phenotype (ST398-t108-SCCmecV-agrI) and lacked erm genes (Table 1).

Three unrelated pulsotypes (A–C) were identified after analysis of PFGE-patterns of all human and animal MRSA isolates [15]. In addition, pulsotype A comprised five subtypes (A1–A5) and pulsotype B two subtypes (B1 and B2) (Fig. 1). All MRSA isolates with spa-type t011 were ascribed to pulsotype-A, those with t108 to pulsotype-B, and the MRSA isolate t588 to pulsotype-C (Fig. 1 and Table 1). It is important to remark that MRSA isolates typed as PFGE/A1-spa-t011 and PFGE/B2-spa-t108 were recovered both from pigs and the patient (nasal or skin lesion samples), suggesting the possible animal-to-human MRSA transmission. All lnu(A)-containing MRSA isolates with the erythromycin/clindamycin-susceptible and lincomycin-resistance phenotype were ascribed to the pulsotype B2.

Figure 1.

 PFGE patterns of chromosomal DNA digested with the ApaI enzyme of MRSA isolates recovered from the patient and some animals in this study. Line 1, t011 (pulsotype A2); lines 2–6, t011 (pulsotype A1); line 7, t011 (pulsotype A3); line 8, t011 (pulsotype A4); line 9, t011 (pulsotype A5); line 10, t108 (pulsotype B1); lines 11–17, t108 (pulsotype B2); line 18, t588 (pulsotype C).

MRSA isolates of human and animal origin were tested by PCR for the genes encoding the Panton-Valentine-leukocidin, toxic-shock-syndrome-toxin 1, and exfoliative-toxin A, B and D [7,16], and negative results were obtained in all cases. The presence of PVL is not frequently found among MRSA-ST398 [17]. However, the lnu(A) gene has been previously detected in two MRSA PVL-positive isolates [18].

Therefore, the MRSA isolates investigated in this study were recovered from a human patient and from some pigs on his farm. Therapy with topical compounds (fusidic acid or mupirocin) might improve clinical symptoms but the efficacy could be low in people with close contact with MRSA-colonized animals [19], as is corroborated in this study. People who are in contact with pigs seem to have a significant bacterial load of MRSA [1] and can serve as a reservoir, which may have important epidemiological and clinical implications. Moreover, most pigs tested in this study (11/12, 91.7%) were MRSA-ST398 carriers. These results seem to be usual according to the high frequencies of detection recently found among slaughter pigs in Spain [7], and also to the high frequencies revealed in farm dust samples in some European countries, such as Spain and Germany [20].

The detection of the gene lnu(A) in MRSA-ST398 isolates from a human and some pigs is a novel observation. This gene encodes a lincomycin-nucleotidyltransferase and confers varying levels of resistance to different lincosamides such as pirlimycin and lincomycin [21], but not to clindamycin. In our study, this gene conferred high level of lincomycin resistance, as MIC was >128 mg/L in these isolates and some isolates showed also clindamycin resistance by the presence of erm genes. The lnu(A) gene was widely spread in the farm sampled, being detected in MRSA-ST398 isolates with different spa-types and pulsotypes. This gene is uncommon, and it has been generally found in strains resistant to both macrolides and lincosamides [11,22]. Lincomycin has occasionally been used in the animal industry and lnu(A) has already been detected in methicillin-susceptible S. aureus (MSSA) isolates from bovine origin [23]. However, as far as we know, it has not been reported in MRSA of the ST398 lineage.

In summary, the animal-to-human transmission of MRSA-ST398 seems evident in this study. Indistinguishable PFGE patterns and similar phenotypic and genotypic characteristics have been detected in MRSA isolates obtained from a farmer and from some of the pigs on the farm where he worked. The transmission of MRSA-ST398 isolates containing lnu(A) and other resistance genes between animals and humans could compromise treatment of infections caused by this microorganism. In addition, the high diversity of resistance genes harboured by most of the human and animal isolates is of great interest.


These data were presented, in part, at the XIII Annual Meeting of the Spanish Society of Infectious Diseases and Clinical Microbiology, Sevilla, Spain, 3-5 June 2009 and at the 20th European Congress of Clinical Microbiology and Infectious Diseases, Vienna, Austria, 10–13 April 2010.

Transparency Declarations

This work was partially supported by Project SAF2009-08570 from the Ministerio de Ciencia e Innovación of Spain and FEDER. Carmen Lozano has a fellowship from the Ministerio de Educación y Ciencia of Spain; Elena Gómez-Sanz has a fellowship from the Gobierno de la Rioja of Spain. There are no conflicts of interest.