Multiresistant Bacteria Isolated from Chicken Meat in Austria

Multidrug resistant bacteria (MDR bacteria), such as extended spectrum beta-lactamase (ESBL) Enterobacteriaceae, methicillin resistant Staphylococcus aureus (MRSA), and vancomycin-resistant Enterococci (VRE), pose a challenge to the human health care system. In recent years, these MDR bacteria have been detected increasingly outside the hospital environment. Also the contamination of food with MDR bacteria, particularly of meat and meat products, is a concern. The aim of the study was to evaluate the occurrence of MDR bacteria in chicken meat on the Austrian market. For this study, 50 chicken meat samples were analysed. All samples originated from chickens slaughtered in Austrian slaughterhouses and were marked as produced in Austria. Samples were analysed for the presence of ESBL Enterobacteriaceae, methicillin resistant Staphylococci and VRE. Resistance genes of the isolated bacteria were characterised by PCR and sequencing. In the present study 26 ESBL producing E. coli, five mecA gene harbouring Staphylococci (but no MRSA), and four VRE were detected in chicken meat samples of Austrian origin. In 24 (48%) of the samples no ESBL Enterobacteriaceae, MRSA, methicillin resistant coagulase negative Staphylococcus (MRCNS) or VRE could be detected. None of the samples contained all three types of investigated multiresistant bacteria. In concordance to previous studies, CTX-M-1 and SHV-12 were the dominant ESBL genes.

sporadic infections in hospitals. VRE are among the first documented antibiotic resistant bacteria with primary origin in animal farming. The rise of VRE was caused by the use of the glycopeptide avoparcin as a growth promoter, starting from 1975. After that it was used extensively, mainly in broiler and pig farming. But, as avoparcin confers cross resistance to vancomycin, the (mis)use of avoparcin resulted in the selection of VRE until it was finally banned in animal farming in the whole European Union in 1996. A study carried out in Denmark reported a decline from 80% to 5% in the occurrence of VRE isolated from broilers in the period from 1995 to 1998. In Central Europe the presence of VRE isolated from human patients is stable at the level of 1% since 2000. Several studies show local rates up to 15%. This seems to be more related to nosocomial outbreaks [23][24][25][26][27][28]. The aim of this study was to investigate the presence of ESBL Enterobacteriaceae, methicillin resistant Staphylococci (MRSA and MRCNS) and VRE in chicken meat products slaughtered in Austria and the determination of (genetic) characteristics of the isolated strains. Samples were taken from the four main supermarket chains and from local butchers, in order to get an impression of chicken meat contamination with MDR bacteria.

Samples
In the course of this study 50 samples of chicken meat (chm-sample-1 to chm-sample-50) were analysed. All samples originated from chickens slaughtered in Austrian slaughterhouses and were marked as produced in Austria. From January to March 2012, fresh meat samples (without skin) from four different supermarket chains as well as butchers were bought in the city of Graz. The samples were transported to the laboratory in a cooling-box at 4 °C and the investigation process was started immediately after arrival in the laboratory.

Strain Isolation and Detection
The preparation of the food samples was done according to ISO 6887-2:2003 [29]. Two enrichment methods were chosen for ESBL and MRSA: the first with peptone broth, the second with thioglycolate bouillon. VRE enrichment was done with enterococcosel (Becton Dickinson Austria, Vienna, Austria). 25 g of the meat samples were randomly selected, mixed with 225 mL peptone broth 1% (Oxoid Ltd, Basingstoke, England) and homogenized in a Stomacher ® 400 Lab Blender for two minutes. The solution was poured into an Erlenmeyer flask and enriched overnight (16 to 24 hours) in an incubator shaker (Innova 4000, New Brunswick Scientific, Enfield, USA) at 150 rpm at 37 °C . After that, a decimal dilution series from peptone broth and enterococcosel enrichment up to 10 −3 was performed. 0.1mL from the appropriate dilution was inoculated on chromeID TM ESBL (bioMé rieux, Marcy-l'Etoile, France), chromeID TM VRE (bioMé rieux) and oxacillin resistance screening agar (OXA agar, Oxoid Ltd.). ChromeID TM ESBL agar was incubated for 24 hours at 37 °C ; chromeID TM VRE and OXA agar for 48 h at 37 °C . Colonies were assessed as described in the manufacturer's manual and identified with Vitek MS (bioMerieux) [30]. In the second enrichment method, 1 g of each meat sample was inoculated in 5 mL thioglycolate bouillon and incubated overnight (16 to 24 h) at 37 °C . A sterile cotton swab was dipped into the thioglycolate bouillon and inoculated on the selective agar plates mentioned above. Bacteria were identified as described above. ESBL-positive E. coli were confirmed by CLSI Screening and Confirmatory Tests [31]. Resistance to the glycopeptides vancomycin and teicoplanin was confirmed by Etest (bioMé rieux) according to the manufacturer's instructions.

Antimicrobial Susceptibility Testing
For all identified Enterobacteriaceae, Staphylococcus spp and Enterococcus spp susceptibility testing was performed as recommended by the European Committee on Antimicrobial Susceptibility testing (EUCAST). The inhibition zone diameters were interpreted according to EUCAST guidelines, except Enterobacteriaceae tested for tetracycline, chloramphenicol and nalidixic acid, which were evaluated in conformity with Clinical Laboratory Standards Institute (CLSI) guidelines. There a no interpretation guidelines for zone diameters of these three antibiotics according to EUCAST (23,24).

Detection of Resistance Genes
PCR detection and gene identification were carried out for three different β-lactamase gene families, blaTEM, blaSHV and blaCTX-M (including subgroup 1, 2 and 9). PCR and sequencing procedures were performed as described previously [32,33]. Conditions for PCR: initial denaturation at 94 °C for 5 min; 35 cycles at 95 °C for 30 s, 52 °C for 45 s, and 72 °C for 60 s; and final incubation for 10 min at 72 °C . Taq DNA polymerase and dNTPs from QIAGEN (Hilden, Germany) were used. Sequences were compared to NCBI database. The detection of the mecA genes was performed as described previously [34] and the detection of the vancomycin resistance genes (vanA/vanB) was performed by real time PCR with the Light cycler VRE Detection Kit (Roche, Branchburg, New Jersey, USA) [35].

ESBL Enterobacteriaceae
In 21 samples 26 different ESBL E. coli were detected; no other ESBL positive Enterobacteriaceae species were recovered. In five samples two phenotypically and genotypically different ESBL E. coli could be isolated. Three different ESBL genes were responsible for (ESBL) resistance in these isolates: Twelve isolates had genes encoding for CTX-M-1 enzyme, twelve for SHV-12 and two for SHV-2. The gene for non ESBL TEM-1 enzyme was also detected in three strains (two times in combination with CTX-M-1 and one time with SHV-12) ( Table 1).  All isolates were susceptible to the tested aminoglycosides, β-lactam/β-lactamase inhibitor, carbapenems, ciprofloxacin and tigecycline. Highest resistance rates were observed for tetracycline in 23 (88.46%) of the isolates and nalidixic acid in 18 (78.26%) isolates. All isolates were resistant against the third generation cephalosporine cefotaxime. In addition, the SHV-12 positive isolates (46.15% of isolated ESBL E. coli) showed non-susceptibility to third generation ceftazidime (10 µg ). Nine isolates (34.61%), all positive for the CTX-M-1 gene, were resistant against the fourth generation cephalosporin cefepime (Table 1).

MRSA and MRCNS
All samples were tested negative for MRSA. However, five coagulase negative Staphylococcus strains, recovered from MRSA screening Agar, were positive for the mecA gene (10 % of the samples). They were identified as Staphylococcus vitulinus (two), Staphylococcus sciuri (two) and Staphylococcus fleurettii (one) ( Table 1). Beside resistance to penicillin and cefoxitin, none of these strains showed additional resistance.

VRE
VRE were found in three of 50 samples (6%). A total of four Enterococcus strains, different in their van genes, were recovered and analysed in detail. All isolates were identified as Enterococcus faecium and harboured the vanA gen whereas one isolate (VRE-4) was additionally vanB positive. All isolates showed highly similar resistance patterns. They were all resistant to vancomycin, and susceptible to tigecycline and linezolid. Only one isolate was susceptible to teicoplanin (VRE-4) and one to amoxicillin (VRE-2) ( Table 1).

Discussion
Besides the international trade of farm animals, intensive antibiotic use is an important factor for the spread of livestock-associated, multiresistant bacteria. Presence of E. coli ESBL in poultry and poultry meat products is documented by many studies from all over the world. High contamination rates with ESBL Enterobacteriaceae, up to 93.3%, are reported [6,36,37]. Austria and surrounding countries do not seem to observe such high rates of MDR bacteria in meat although in some cases such as in Austria, data are extremely limited. The study from 2012 Springer and Bruckner with samples from 2009 revealed a little lower ESBL frequency (35.9% of chicken meat samples) but with a very low SHV frequency [38]. In contrast, recent German studies by Campos et al. [39] and Kola et al. [40] detected SHV-12 as second or most common ESBL enzymes in poultry meat. Frequencies of total ESBL E. coli Therefore the results of the present study concerning SHV-12 and CTX-M-1 are in concordance with those of other European countries. Both genes are very common in animal farming. SHV-12 is known to be associated mainly with poultry and CTX-M-1 is widespread in mammals. In contrast to other studies, TEM-52 and CTX-M-14, also common in farm animals and meat products, were not detected here [36,37,39,40]. Two reasons for the absence of these genes might be the relatively small sample size and the exclusive Austrian origin of the poultry.
In concordance with previous studies, the resistance against tetracycline is shown to be common in farm animals and meat product samples because tetracycline is known to be still in use in livestock breeding. In contrast to this, in the present study fluoroquinolone resistance was nearly absent, although fluoroquinolones are frequently used antibiotics in human healthcare in Austria. This finding can not only be explained with the connection of this resistance to the occurrence of CTX-M-15 because reports from non-CTX-M-15 ESBL E. coli from Austria show that fluoroquinolone resistance is present in human and also environmental isolates [7,38,[41][42][43].
LA-MRSA are known to be associated essentially with mammalian farm animals but was also isolated from poultry and poultry meat products. Contamination of meat with HA-or CA-MRSA, possibly caused due to the lack of hygiene during meat production, is also documented. Hence the absence of MRSA in this study was surprising [44][45][46]. But the mecA gene was present in coagulase negative Staphylococci in animal farming years before the first LA-MRSA was reported. The detected MRCNS species in this study are also in concordance with previous studies [47][48][49]. Although avoparcin was banned in animal feeding in the European Union more than two decades ago, VRE are still present in the meat production. The persistence of VRE over a prolonged time period is potentially caused by coselection with other food additives [24,25,50].

Conclusions
ESBL E. coli was the most common type of multiresistant bacteria observed in this study. Testing revealed 21 samples out of 50 (42%) as positive for ESBL. No MRSA could be detected but five samples were positive for MRCNS. VRE could be recovered from three samples (6%). None of the samples contained all three types of multiresistant bacteria, but three samples harboured E. coli ESBL together with MRCNS, and in two samples E. coli ESBL emerged additionally to VRE. In 24 (48%) samples no ESBL Enterobacteriaceae, MRSA, MRCNS or VRE could be detected. This study had a very limited sample size, and thus limited power to infer the prevalence of antimicrobial resistance in chicken meat in Austria. Although informative as a starting point, especially considering the lack of Austrian data, further work on this matter in Austria should occur.

Author Contributions
Gernot Zarfel, Herbert Galler, Gebhard Feierl, and Franz F. Reinthaler conceptualized the manuscript's focus and analyses. Peter Pless took samples and provided background information of samples. Herbert Galler, Christian Petternel and Josefa Luxner performed the bacterial isolation, microbiological experiments, and analyses. Gernot Zarfel and Josefa Luxner performed molecular biology experiments. Clemens Kittinger, Doris Haas, and Andrea J. Grisold edited the manuscript.

Conflicts of Interest
The authors declare no conflict of interest.