High Occurrence of Methicillin-Resistant Staphylococcus aureus in Horses at Slaughterhouses Compared with Those for Recreational Activities: A Professional and Food Safety Concern?

Introduction

Methicillin-resistant Staphylococcus aureus (MRSA) is a widespread and well-known pathogen, able to cause serious infections in humans both in hospitals and in the community (Kluytmans-VandemBergh and Kluytmans, 2006). In addition, MRSA clones are associated with farm and companion animals, and their zoonotic potential is well known (Leonard and Markey, 2006). The presence of MRSA in horses has important implications both for veterinary medicine and for public health. In fact, MRSA could be responsible for several types of infections in horses (Weese et al., 2006b). The colonization of horses by MRSA could result in infections in humans working closely with them. There are several reports of outbreaks of MRSA infections in horses admitted to veterinary hospitals and in staff working with horses (O'Mahony et al., 2005). The most prevalent genotype isolated from horses was CC398; this genotype is also responsible for infections in humans, suggesting the circulation of MRSA between horses and humans (Van den Eede et al., 2013). The spread of MRSA in horses intended for meat production could result in colonization and infection of slaughterhouse operators, and represents a potential food safety concern (EFSA, 2009). Italy is the EU's biggest horse meat consumer; about 84,063 horses are slaughtered every year and horse meat production is ∼21,466 tons per year (ISTAT, 2009). In addition, people in some Italian regions still consume raw horsemeat (www.cibo360.it/alimentazione/cibi/carne/cavallo.htm). The aim of this report was to investigate (1) the prevalence and characteristics of MRSA in horses used for recreational purposes and for meat production from two regions of southern Italy (Apulia and Basilicata), and (2) the prevalence and characteristics of MRSA in humans working in close contact with horses.

Materials and Methods

Results

Discussion

MRSA is an emerging pathogen in horses. It was previously associated with sporadic infections in horses treated in veterinary hospitals (Weese et al., 2005; Garcia-Alvarez et al., 2012), and its spread in this animal species has now become a global issue (Weese and Van Duijkeren, 2010). The first MRSA isolated from horses belonged to the clonal complex 8 (CC8), a clone of human origin, spread mainly in horses in Canada and the United States (Weese and Van Duijkeren, 2010). Subsequently, an MRSA isolate of CC398, a clone typically associated with livestock, was isolated in horses. The spread of MRSA in horses has been widely investigated in several European countries and in North America, and surveys have been conducted on horses on farms and those treated in veterinary hospitals (Baptiste et al., 2005; Busscher et al., 2006; Weese et al., 2006a; Van den Eede et al., 2009). Although the presence of MRSA in farm animals and in humans in Italy has been widely documented (Pan et al., 2009; Battisti et al., 2010; Normanno et al., 2015), very few investigations have focused on horses and personnel working with them (De Martino et al., 2010; Mallardo et al., 2013; Carfora et al., 2016). In our study, the isolation rate of MRSA recorded in slaughtered horses was significantly higher than in those used for recreational purposes and racehorses, probably due to the rearing system of horses intended for meat production and the higher antimicrobial selective pressure present during rearing stages. In particular, the isolation rate recorded in Spanish and Polish horses reflects the high prevalence found in horses slaughtered in abattoirs. In fact, the prevalence of MRSA in French and Italian horses was low, and similar to values recorded in horses used for recreational purposes and in racehorses. To our knowledge, no prevalence data are available on MRSA in horses from Spain and Poland, although the isolation of MRSA in hospitalized horses has been reported once in Spain (Gómez-Sanz et al., 2014). As for the people working closely with horses, the percentage of MRSA isolation was higher in the slaughterhouse staff than in workers of farms, riding stables, and racecourses. This finding might be associated with slaughterhouse staff being more exposed to MRSA and to a higher level of MRSA contamination than workers at farms, riding stables, and racecourses. In fact, it is widely reported that when the prevalence of MRSA in animals is high, people in close contact with them are at a much greater risk of colonization and subsequent infection (EFSA, 2009). The antimicrobial susceptibility tests have shown that human and horse isolates are resistant to a wide range of antimicrobials of different families. Therefore, they have the characteristics of multiresistant isolates, similar to what has been reported for MRSA isolates of animal and human origin worldwide (Weese and van Duikeren, 2010). Regarding genetic characterization of the isolates, ST398 is the prevalent genotype in horses and in staff working closely with them. MRSA isolates belonging to ST398 are the most widespread in Europe in horses; ST398/t011 strains were isolated in Austria, Belgium, Spain, and the Netherlands (Cuny et al., 2008; van Duijkeren et al., 2010; Van den Eede et al., 2012; Gómez-Sanz et al., 2014), and MRSA ST398 is typically associated with livestock (LA-MRSA). These strains have been shown to cause serious infection in humans, mainly in people coming into close contact with animals for work (Cuny et al., 2010; Köck et al., 2013). Handling and/or consumption of food products contaminated by CC398 could be also risk factors for infection (EFSA, 2009). In addition, transmission of a MRSA ST398/t011 from a horse to human has been demonstrated in the Netherlands (van Duijkeren et al., 2011). In slaughterhouses, samples from both horses and workers were MRSA positive, whereas samples taken from the horses at stables and racecourses were MRSA negative, although some of the human nasal swabs tested positive. In our study, human and horse MRSA ST398 were isolated from samples collected at the same sampling point at M2: whereas these isolates presented the same ST and SCCmec, they actually belonged to different spa-types, therefore transmission from horses to humans or vice versa does not seem likely. The other genotype was ST1, identified in three isolates: from a slaughtered Spanish horse, a slaughterhouse operator at M1, and from a racecourse operator; all isolates had spa-type t127 and SCCmec IVa. MRSA ST1/t127 strains are typically associated with human infections, and cases have been reported in the United States, Germany, and the United Kingdom. In addition, these strains were among the prevalent clones isolated in human cases of invasive MRSA infections in Europe (Otter et al., 2009; Monaco et al., 2013). MRSA ST1/t127 was isolated from horses in different veterinary hospitals (Cuny et al., 2008; Loncaric et al., 2014), and from pigs, cows, sheep, and goats in Italy (Battisti et al., 2010; Cortimiglia et al., 2015; Caruso et al., 2016). In our study, human and horse MRSA ST1 were isolated from swabs taken at the same sampling point at M1. Unlike what was found at slaughterhouse M2, the two isolates showed the same genetic characteristics. Therefore, transmission of the strain from horses to humans and vice versa might be plausible.

Regarding the genetic virulence markers detected in our isolates, we can summarize some considerations. All the isolates that carried the icaA gene were pvl negative and some were potentially enterotoxigenic. This suggests that although the pvl-negative isolates are generally less virulent than the Panton-Valentine leukocidin (PVL) producers, these isolates could be of concern for the meat industry. These strains can synthesize a biofilm that promotes resistance and persistence of the strain in the food industry, and produce some SEs, alone or in combination (SEH, SEG, SEM, and SEN). However, the foodborne risk appears quite limited, because the SE most frequently involved in staphylococcal food poisoning (SFP) outbreaks is SEA (Balaban and Rasooly, 2000), and the role of the recently discovered SEs in food safety is largely unknown. On the other hand, SEH-producing strains have been involved in SFP outbreaks (Ikeda et al., 2005), thus, the risk of foodborne intoxication linked to our findings must be carefully assessed. MRSA isolated from humans and animals show great variability in antimicrobial susceptibility, and this is of concern in human and veterinary medicine. In this study, the MRSA isolates from both horses and humans were resistant to between 9 and 13 antimicrobials. In our study, all the ST398 isolates, generally recognized as a livestock-associated clone (LA-MRSA), showed resistance to tetracycline; on the other hand, the only strain susceptible to tetracycline was the ST1/t127/IVa isolated from a human source. This finding reinforces the hypothesis that the excessive use of tetracycline in animal husbandry, especially on pig farms, could lead to the selection of clones carrying resistance (de Neeling et al., 2007).

In conclusion, the results of our study show that while a low MRSA prevalence is recorded in horses and workers in contact with them at farms, riding stables, and racecourses, a high prevalence was found in slaughterhouses, where a high prevalence was detected in imported horses. In addition, the detection in horses of MRSA belonging to the human transmissible genotypes (ST398, ST1) and the isolation of strains with identical genetic profiles in a horse and a worker at the same slaughterhouse, confirm that there is a potential zoonotic risk from direct contact with animals, or from the handling and consumption of raw horse meat, which is a widespread habit in some European countries.