Abstract
Introduction:
Methicillin-resistant Staphylococcus aureus (MRSA) is an important clinical problem. In 2005, a livestock-associated MRSA clone was described, named CC398, being mostly associated with pigs, and causing colonization and infection in pigs and in related humans. The prevalence of these strains in food-producing pigs raised concerns about the possibility of MRSA-CC398 being a foodborne pathogen. The objective of this study was to investigate the presence of S. aureus and MRSA in 141 carcasses of pigs at three slaughterhouses of Portugal, discarded from the food chain by signs of infection, and to characterize the recovered isolates.
Methods:
S. aureus isolates were identified by matrix-assisted laser desorption/ionization time-of-flight and they were typed (spa, CC398-clone, and SCCmec). The study of antibiotic resistance and virulence genes, and the detection of immune evasion cluster genes and prophages were performed by PCR and sequencing.
Results:
Twenty-eight S. aureus were obtained from 141 samples (one/sample, 19.9%), being 22 MRSA and 6 methicillin-susceptible S. aureus (MSSA). All MRSA strains were typed as CC398 and were ascribed to three spa types (t011, t108, and t1451). The SCCmec detected differed according to the spa types of MRSA isolates (SCCmecV: t011 and t108; SCCmecIVa: t1451). The MSSA strains were classified as spa-t1491-ST1-CC1. All the strains contained a wide range of antimicrobial resistance genes, the resistance to tetracycline being the prevalent one. In contrast, the strains contained only a few virulence genes. Among the 6 integrases of phages tested, three were detected: SΦ1, SΦ2, and SΦ7, with variations between MRSA and MSSA strains.
Conclusions:
MRSA-CC398 is not only a habitual pig colonizer but also an opportunistic pathogen in these animals, and must be controlled at the level of producers and slaughterhouses because of its impact on public health.
Introduction
S
MRSA CC398 is the major LA-MRSA CC found in Europe and is mostly associated with colonization in pigs, not being especially pathogenic in these animals. It is also considered slightly pathogenic in humans, occurring in colonization without symptoms.5,6 Direct contact between humans and pigs has been recognized as a risk factor for colonization and infection by MRSA CC398 (ST398). Nevertheless, the cases of infection and colonization by MRSA CC398 have been reported not only in single cases with professionals in direct contact with livestock (farmers, veterinarians, or slaughterhouse workers) and their families, but also in the community and in hospitals. 7
The dissemination of MRSA CC398 in humans and the presence in pigs raised concerns about the presence of food MRSA for human feeding.8,9 Meat contamination may be due to improper handling and storage, and the production of heat-stable toxins by certain strains may be a cause of disease in humans. 9 The possibility of MRSA being a foodborne pathogen is of concern to the pork industry, mainly because of consumer's confidence. Therefore, the purpose of this research was to screen for the presence of S. aureus and MRSA in carcasses of pigs at the slaughterhouse level, discarded from the food chain by signs of infection, and to characterize the recovered isolates.
Materials and Methods
Samples and bacterial isolates
For a period of 4 months (July–September 2015), 141 samples from pigs and piglets (one per animal) of three slaughterhouses, which receive animals from different farms, were collected in the north and center of Portugal, which corresponded to animal carcasses rejected by signs of infection, and so, not included in the food chain. From slaughterhouse A, 38 samples were collected from piglets' abscesses. From slaughterhouse B, we collected 5 samples of piglets' abscesses and 44 of pigs' carcasses from osteomyelitis. From slaughterhouse C, 10 samples were collected from piglets' abscesses. Forty-four samples were excluded due to the fact that no S. aureus was found.
All samples were recovered with sterile swabs from abscesses and osteomyelitis during postmortem inspection of pig carcasses. All parts of the carcasses presenting these lesions were included in the study. Samples were maintained under refrigeration and were sent to the laboratory for processing. The samples were grown in Brain Heart Infusion (BHI) broth with 6.5% NaCl and then cultured on mannitol salt agar and oxacillin resistance screening agar base supplemented with 2 mg/L oxacillin (OXOID) to recover S. aureus and MRSA, respectively. Isolates with S. aureus morphology were identified by matrix-assisted laser desorption/ionization time-of flight (Biotyper©). Moreover, a multiplex PCR with primers nuc, 16S rDNA, and mecA was performed 10 for S. aureus identification and for detection of the main methicillin resistance gene.
Antimicrobial susceptibility
Antimicrobial susceptibility was tested by the disk-diffusion method. 11 Antimicrobials tested were the following ones (in μg/disk): penicillin G (10 U), cefoxitin (30), tetracycline (30), erythromycin (15), clindamycin (2), gentamicin (10), tobramycin (10), kanamycin (30), streptomycin (10), trimethoprim/sulfamethoxazole (1.25/23.75), linezolid (30), chloramphenicol (30), ciprofloxacin (5), fusidic acid (20), and mupirocin (5). Isolates displaying resistance to three or more antimicrobial classes were considered multidrug resistant (MDR).
Detection of virulence and antimicrobial resistance genes
All S. aureus isolates were screened for virulence and antimicrobial resistance determinants by PCR amplification using primers previously described. The resistance determinants tested were those conferring resistance to the following: beta-lactams (blaZ and mecA), macrolides and/or lincosamides and/or streptogramins [erm(A), erm(B), erm(C), erm(T), msr(A)/msr(B), linA/linA′, vga(A), vga(B), vga(C), vga(E), tetracycline (tet(M), tet(K), and tet(L)], aminoglycosides [aac(6′)-Ie-aph(2′′)-Ia, ant(4′)-Ia, and aph(3′)-IIIa], phenicols, and/or oxazolidinones (catpC194, catpC221, catpC223, fexA, and cfr) and trimethoprim (dfrA, dfrG, and dfrK).7,8,12 The virulence determinants tested encoded hemolysins (hla, hlb, hld, hlg, and hlg variant), Panton–Valentine leukocidin (lukPV), and exfoliatins (eta and etb). 12 In addition, the detection of gene scn, marker of the immune evasion cluster (IEC), and gene cna were tested in these isolates. 7 Moreover, three integrase phage profiles of the families SΦ1 to SΦ7 were tested by PCR. 13
Molecular typing
All S. aureus isolates were submitted to spa typing, by PCR and sequencing, and the sequences were analyzed with the program Ridom Staph-Type version 1.5.21 (Ridom GmbH). 14 The same happened to agr typing. 15 Moreover, selected S. aureus isolates were submitted to multilocus-sequence typing. The SCCmec-typing16,17 and a specific CC398-PCR 18 were performed on MRSA isolates. The diversity of prophages was analyzed in all S. aureus isolates, as previously reported. 13
Results
S. aureus isolates were recovered from 28 (from different animals) of the 141 samples analyzed (19.9%) and one isolate per sample was further characterized. Twenty-one of the positive samples corresponded to piglet abscesses and seven to osteomyelitis (Table 1). From the 28 S. aureus isolates, 22 were MRSA (78.6%) and 6 were methicillin-susceptible S. aureus (MSSA). All MRSA strains were typed as belonging to the lineage CC398 and were included in three different spa types (t011, t108, and t1451), t1451 being the prevalent one. The six MSSA strains were typed as spa t1491, and sequence type ST1 (included in CC1). The agr typing differentiated MRSA and MSSA strains: all MRSA strains belonged to agr type I and MSSA to agr type III. All isolates were negative for the PVL toxin genes (lukS/F-PV), cna gene, eta and etb genes, as well as for scn gene of the IEC system.
Genetic Lineages of the Twenty-Eight Staphylococcus aureus Recovered from Pig Carcasses at the Slaughterhouses
STs included in brackets were assumed according to the spa type.
MRSA, methicillin-resistant Staphylococcus aureus; MSSA, methicillin-susceptible S. aureus; ST, sequence type, CC, clonal complex.
Multiresistance patterns (resistance to at least three classes of antimicrobial agents) were exhibited in all strains (Table 2). Twenty-six isolates were resistant to tetracycline, the tet(M) gene being present in almost all strains, followed by tet(K) and tet(L). The erm(C) or msr(A) genes were present in the 13 macrolide/lincosamide-resistant strains. The gene aac(6′)Ie-aph(2′′)-Ia, conferring resistance to gentamicin, tobramycin, and kanamycin, was identified in 15 strains, while specific genes for tobramycin, kanamycin, or streptomycin resistance [ant(4′)-Ia, aph(3′)-IIIa and ant(6′)-Ia, respectively] were detected in only a few strains. Resistance to chloramphenicol was identified in 17 strains, mediated by catpC194, catpC221, catpC223, or fexA genes. Eleven strains showed resistance to ciprofloxacin and three to SXT, which harbored different combinations of dfrA, dfrG, and dfrK genes.
Antimicrobial Resistance Phenotype, Genotype, Virulence Genes, and Phages Detected in Isolates of Pig Carcasses at the Slaughterhouses
PEN, penicillin; FOX, cefoxitin; TET, tetracycline; ERY, erythromycin; CLI, clindamycin; GEN, gentamicin; TOB, tobramycin; KAN, kanamycin; STR, streptomycin; SXT, trimethoprim/sulfamethoxazole; CHL, chloramphenicol.
All MRSA strains harbored the same group of virulence genes (hla, hlb, hld, and hlg), and all MSSA strains also presented the same virulence profile (hla, hlb, hld, and hlgy).
MRSA strains contained the SΦ1 and/or SΦ2 phages, evidencing differences in relation to the spa type. In addition, all six MSSA strains contained the SΦ2 and SΦ7 integrase phages (Table 2).
Discussion
The detection of MRSA of the lineage CC398 in samples of pig carcasses with signs of infection at Portuguese slaughterhouses (15.6%) is concerning, even though they were discarded from the food chain. Studies reporting the widespread detection of MRSA CC398 in pigs in farms 19 point to the potential relevance of this clone as a pathogen to consider in these food-producing animals. 20 Furthermore, MRSA CC398 has also been detected in other food-producing animals such as rabbits and turkeys.21,22 The possibility of dissemination of MRSA CC398 in farms and slaughterhouses to other animals or humans seems to be possible considering the frequent presence of this microorganism in the pig environment.
Although the importance of contaminated meat with antibiotic-resistant strains is not yet well recognized in the food chain and public health, the presence of the MRSA CC398 in swine products has already been mentioned. It seems possible that the MRSA transmission is by management and eating of contaminated meat.
Animals can act as reservoir for this microorganism, making possible the transmission to other animals, as well as to humans in close contact with pigs. 23
The spa types t011, t108, and t1451, which were detected in our study, are very common among MRSA CC398 strains in pigs.8,24–27 The SCCmec type detected differed according to the spa types of MRSA isolates: t011 and t108 corresponded to SCCmec V, and t1451 to SCCmec IVa. Both SCCmec structures are the most frequently detected among MRSA CC398 isolates in other studies.8,28,29
The MSSA strains detected in this study corresponded to spa type t1491, associated with the sequence type ST1 and CC1. The ST1/CC1 is considered an LA-MRSA clone and can be found among MRSA isolates of pig origin.30,31 This clone is also frequently found in humans, which may demonstrate the transmission between humans and pigs. 27 Other studies had previously found MSSA ST1 to be associated with spa t1491 in pigs. 32
Most of the MRSA CC398 isolates recovered in this study, as well as the MSSA, showed an MDR phenotype containing a wide range of antimicrobial resistance genes, including some unusual ones (Table 2). The content of so many antimicrobial resistance genes among the MRSA and MSSA isolates of pig origin, initially destined for food production, could reflect the high pressure exerted by the use of antimicrobials in this animal species. No phenotypic resistances were detected for linezolid, mupirocin, and fusidic acid. It is important to point out the presence of unusual genes detected among the pig MRSA isolates, as in the case of the Inu(A) and Inu(B) genes, which are mostly associated with clonal lineages of animals. 33 The bifunctional enzyme, codified by the aac(6′)-Ie-aph(2′′)-Ia gene, was also frequently detected in this study, conferring resistance to the three aminoglycosides tested. This aac(6′)-Ie-aph(2′′)-Ia is the most commonly found aminoglycoside resistance gene in S. aureus, 34 along with aph(3′)-IIIa and ant(6′) genes. The cat genes, frequent among our isolates, were previously reported as important mechanisms of chloramphenicol resistance by others. 35 The fexA gene, found only in one strain, also confers resistance to florfenicol (a chloramphenicol derivative), which is used only in veterinary medicine. 36 Among the MSSA isolates detected in this study, five showed resistance to penicillin, although the blaZ gene was not found, suggesting the presence of other blaZ homologous variants not detected by the used PCR. 37 Four MRSA isolates showed an atypical lincosamide-resistant/macrolide-susceptible phenotype, which has been increasingly reported among ST398 MRSA from swine and has also been reported in Portugal. 38 This type of resistance is associated with vga genes, which mediate resistance to lincosamides or to lincosamides and pleuromutilins. 39 Furthermore, decreased susceptibility to lincosamides has been attributed to ATP-binding cassette transporters encoded by vga genes. 40
The high frequency of antimicrobial resistance genes among the MRSA CC398 isolates contrasts with the low detection of virulence genes, a fact that has already been described in other studies. 29 In this sense, the hemolysins were the unique virulence genes detected among these isolates, whereas the genes encoding the PVL or ETA or ETB toxins were not detected in these strains. The presence of hemolysins in this type of strains, although not usual, has already been detected in the swine population in Belgium. 25 Among the six integrases of phages tested, only three were found in this study: SΦ1, SΦ2, and SΦ7. In MRSA CC398 isolates, the SΦ1 and SΦ2 phages, frequently found in our study, are partially detected in literature. 41 In the MSSA ST1 strains, SΦ2 and SΦ7 were found. SΦ7 associated with European MRSA CC398 may have moved horizontally to clade P and contributed to human adaptation. In addition, this phage may also be considered a marker of human-to-human transmission ability in clade P isolates. 42
Conclusions
This study revealed the presence of MRSA CC398 and MSSA (20% and 8%, respectively) in pig carcasses at slaughterhouse level, with MDR phenotypes although with a lack of virulence genes.
To investigate the presence of MRSA strains in swine during the slaughter process, the sampling must be performed along the procedure. The majority of the positive MRSA animals do not show clinical signs and for this reason, normally they were not separated from the non-MRSA-carriers resulting in a possible cross-contamination during transport and in lairage. The high temperatures used in the carcass treatment eliminate most of the pathogenic strains, whereas some vestigial MRSA can be remaining in meat via fecal contamination or during evisceration. In this sense, the correct prophylactic measures based on animal hygiene and health, either before slaughter or during this process, should be considered a priority, avoiding the spread of bacterial strains in the food chain.
Footnotes
Disclosure Statement
No competing financial interests exist.
Funding Information
This work was supported by the Associate Laboratory for Green Chemistry-LAQV, which is financed by national funds from the Foundation for Science and Technology/Ministério da Ciência, Tecnologia e Ensino Superior (MCTES) (UID/QUI/50006/2019), and by the project NORTE-01-0145-Fundo Europeu de Desenvolvimento Regional-030101 “CAREBIO2,” funded by the European Regional Development Fund through the NORTE 2020 (Northern Regional Operational Program) and the Foundation for Science and Technology. V.S. is grateful to the Foundation for Science and Technology (Fundação para a Ciência e Tecnologia) for her PhD grant (SFRH/BD/137947/2018). The work performed in the University of La Rioja was supported by project SAF2016-76571-R from the Agencia Estatal de Investigación (AEI) of Spain and the Fondo Europeo de Desarrollo Regional (FEDER) of EU. L. Ruiz-Ripa has a predoctoral fellowship from the University of La Rioja, Spain; O.M. Mama has a predoctoral fellowship from Mujeres por Africa-Universidad de La Rioja.
