Food Is Reservoir of MDR Salmonella: Prevalence of ESBLs Profiles and Resistance Genes in Strains Isolated from Food

Salmonella spp. are among the most frequent causes of foodborne diseases, and the increasing occurrence of MDR strains is an additional cause for concern. In the three-year period 2019–2021, we collected Salmonella spp. strains isolated from different food categories analysed in the context of Regulation (EC) No 2073/2005 in order to assess their antibiotic susceptibility profiles and ESBL production. To determine the susceptibility profiles and identify MDR strains, we used the Kirby–Bauer method to test 17 antibiotics. Double-disc and PCR testing then allowed us to assess the production of ESBLs and the presence of beta-lactamase resistance genes. Phenotypic tests showed that 36 out of 67 strains were MDR and 52.7% of these were ESBL producers. Finally, molecular investigations conducted on ESBL-producing strains revealed the presence of blaSHV, blaCTX-M and blaTEM genes. Our results confirmed the prevalence of S. Infantis, an MDR strain and ESBL producer, in chicken meat. This suggests that further research on the prevalence of antibiotic resistance genes (ARGs) in foodborne strains is needed, especially from a One Health perspective.


Introduction
Salmonellosis is a commonly reported gastrointestinal infection in humans, and an important cause of foodborne outbreaks. In the European Union (EU) in 2019, the number of confirmed salmonellosis cases was 87,923; in 2020, the number was 57,702, which was the lowest recorded number since 2007 because of the impacts of the withdrawal of the United Kingdom from the EU and the COVID-19 pandemic [1]. The main route of infection is ingestion of food or water contaminated with Salmonella spp., Gram-negative, facultative anaerobic bacilli belonging to the Enterobacteriaceae family [2,3]. Salmonella is ubiquitous in the human food chain and is one of the most important foodborne pathogens in the world. In particular, S. Enteritidis, S. Typhimurium, monophasic S. Typhimurium, S. Infantis and S. Derby are the five serotypes most commonly involved in human infections [1]. In the EU, microbiological food controls carried out in the context of Regulation (EC) No. 2073/2005 found the highest percentages of Salmonella-positive samples in egg products, poultry meat and poultry products, which are the most critical sources of Salmonella spp. transmission to humans [1,4].
Although salmonellosis is generally self-limited and usually does not require specific treatment, antibiotic therapy with quinolones, beta-lactams, aminoglycosides, tetracyclines or sulfamethoxazole-trimethoprim is necessary in severe cases [5]. However, the overuse of antibiotics has contributed to the selection of MDR Salmonella strains, i.e., resistant simultaneously to three or more classes of antibiotics, including those most commonly prescribed for the treatment of salmonellosis [6]. The spread of MDR Salmonella represents a significant health problem, as it causes longer hospitalisations, prolonged illnesses and higher mortality rates than susceptible strains [7,8]. The World Health Organization estimates that of the 100,000 cases of salmonellosis each year, a large number are caused by MDR Salmonella [9], with the majority acquired through the consumption of contaminated food of animal origin, particularly beef, pork and poultry products [10,11].
In Enterobacteriaceae such as Salmonella, the main mechanism of resistance to betalactams is the acquisition of genes (bla gene) that encode for beta-lactamase hydrolytic enzymes, which inactivate the antibiotic [12]. Extended-spectrum beta-lactamases (ESBLs), which hydrolyse first-, second-, and third-generation penicillins and cephalosporins, are encoded by genes belonging to the TEM, SHV, and CTX-M families, including multiple variants of the bla TEM , bla SHV and bla CTX-M genes [13]. These ESBL genes have been identified in bacteria isolated from animals and food products of animal origin [7,14], as well as from other types of foods, such as seafood [15], raw vegetables [16] and ready-to-eat (RTE) foods [17], suggesting the possible role of the food production chain as a reservoir for this group of bacteria. Indeed, factors such as selective pressure in animal and environmental microbiomes, the circulation of bacteria between animals and environment and ineffective food safety management can contribute to the presence and persistence of antibiotic-resistant bacteria (ARB) and antibiotic resistance genes (ARGs) in the food production context [18].
The aim of this study was to evaluate the MDR potential of Salmonella strains isolated in the period from January 2019 to December 2021 from food samples analysed in the context of the Regulation (EC) No 2073/2005 [4]. Furthermore, for every MDR Salmonella strain, ESBL production and ESBL gene presence were determined by double-disc diffusion and PCR tests, respectively.
Isolation according to ISO 6579-1:2017 was performed, and strains were then identified by biochemical enzymatic assays and serotyping, according to the Kauffmann-White-Le Minor scheme (Supplementary Materials Table S1) [19].
Interpretation of inhibition zones and classification of isolates as susceptible (S), intermediate (I) or resistant (R), was done in accordance with CLSI guidelines [20].

ESBL Production Evaluation by Double-Disc Test
The double-disc test (DDT) was conducted on 36 MDR Salmonella strains to phenotypically assess ESBL production. Discs containing cephalosporins (cefotaxime 30 µg, ceftazidime 30 µg, cefepime 30 µg) were placed next to a disc with clavulanic acid (30 µg amoxicillin-clavulanic acid), as recommended by EUCAST [21]. When zones of inhibition around any of the cephalosporin discs were increased or there was a 'keyhole' in the direction of amoxicillin-clavulanic acid disc, the test was considered positive.

Detection of Beta-Lactamase Genes
The beta-lactamase gene detection was conducted on the 19 strains that were found by the double-disc test to be ESBL-producing. Bacterial DNA was extracted using 100 µL of PrepMan™ ultra Sample Preparation Reagent (Thermo Fisher Scientific, Waltham, MA, USA), according to the procedure recommended by the manufacturer. Real Time PCR reactions were performed using 10 ng of DNA template and 0.5 µM of the forward and reverse primers listed in Table 1, for a total volume of 25 µL of 1X of Advanced Universal SYBR Green Supermix (Bio-Rad Laboratories, Hercules, CA, USA), in order to amplify bla TEM , bla CTX-M , bla SHV and bla OXA genes. The amplification program included an initial denaturation at 94 • C for 10 min, followed by 32 cycles of 94 • C for 30 s, 60 • C for 30 s, 72 • C for 15 s, and a final extension at 72 • C for 10 min. Subsequently, 10 µL of the PCR product were used for electrophoresis on 2% E-Gel™ Go! Agarose Gels (Thermo Fisher Scientific, Waltham, MA, USA) to determine the size of the product. In each Real Time PCR reaction, a positive and a negative control were used. The positive one was represented by DNA belonging to a strain of Salmonella in which the presence of the bla gene was previous confirmed by sequencing; the negative control was represented by a Not Template Control (NTC), in which the reaction volume with DNase free water was obtained. S. Infantis was the predominant serotype (48%), present in 32 poultry meat samples. Instead, S. Typhimurium (9%), S. Derby (6%) and S. Enteritidis (3%) serotypes were found to have a low prevalence (Figure 2  S. Infantis was the predominant serotype (48%), present in 32 poultry meat samples. Instead, S. Typhimurium (9%), S. Derby (6%) and S. Enteritidis (3%) serotypes were found to have a low prevalence ( Figure 2). S. Infantis was the predominant serotype (48%), present in 32 poultry meat samples. Instead, S. Typhimurium (9%), S. Derby (6%) and S. Enteritidis (3%) serotypes were found to have a low prevalence ( Figure 2).

Antibiotic Susceptibility and ESBL Production Test Results
Antibiotic susceptibility testing conducted on the 67 Salmonella strains showed the absence of resistance in 24 of these strains, whereas 43 strains (64%) were resistant to one or more of the tested antibiotics. Supplementary Materials Table S1 provides an overview of these strains and their resistances.
An MDR profile was found in 36 strains that showed resistance to three (n = 4), four (n = 22) and five (n = 10) antibiotic classes (Supplementary Materials Table S1). Specifically,

Antibiotic Susceptibility and ESBL Production Test Results
Antibiotic susceptibility testing conducted on the 67 Salmonella strains showed the absence of resistance in 24 of these strains, whereas 43 strains (64%) were resistant to one or more of the tested antibiotics. Supplementary Materials Table S1 provides an overview of these strains and their resistances.
An MDR profile was found in 36 strains that showed resistance to three (n = 4), four (n = 22) and five (n = 10) antibiotic classes (Supplementary Materials Table S1). Specifically, the most frequent MDR profiles were: aminoglycosides, beta-lactams, quinolones, sulphonamides and tetracyclines; resistance to these was found in eight S. Infantis, one S. Salamae and one S. Kentucky. Resistance to beta-lactams, quinolones, sulphonamides and tetracyclines was found in nine S. Infantis and one S. Cerro.
Finally, the double-disc test allowed detection of ESBL production in 19 strains. Indeed, for these strains, an increase in the zones of inhibition in the direction of amoxicillin or clavulanic acid was recorded around the tested cephalosporins (Table 2).

Detection of Beta-Lactamase Genes
Genes responsible for beta-lactamase activity in 19 ESBL-producing Salmonella strains were screened by PCR. The presence of beta-lactamase genes was detected in all tested strains, confirming the phenotypic results of ESBL production tests (Table 3). The most frequently identified genes were bla SHV and bla CTX-M , which were present in 68.4% and 47.3% of strains, respectively. Furthermore, the bla TEM gene was harboured by only one strain, while bla OXA was not detected. Specifically, nine strains harboured only the bla SHV gene, six strains harboured only the bla CTX-M gene, three strains harboured the bla CTX-M and bla SHV genes together, and one strain harboured the bla TEM and bla SHV genes together.

Discussion
Salmonella spp. are among the most frequent causes of foodborne diseases, and the increasing occurrence of MDR strains is an additional cause for concern. Thus, in the three-year period 2019-2021, we collected Salmonella spp. strains isolated from different food categories analysed in the context of Regulation (EC) No 2073/2005 [4], in order to assess their antibiotic susceptibility profiles and ESBL production.
Our data show that among the different food categories analysed, poultry meat was a relevant source of Salmonella. Moreover, regarding poultry meat, it is possible to note that the prevalence of Salmonella significantly increased over the three-year period, rising from 40% in 2019 to 71.4% in 2021; the prevalent serovar was S. Infantis (48%).
We performed a screening test using the Kirby-Bauer method to estimate the antibiotic susceptibility profiles of these strains, and we found a very high rate of strains showing at least one phenotypic resistance (64%). Among these, the highest rates of resistance were found against sulphonamides (43.2%), a class of antibiotics used in severe Salmonella infections, but also against nalidixic acid (47.7%) and kanamycin (31.3%). In addition, a high percentage of strains showed resistance to tetracyclines (50.7%), despite the fact that, in 2006, the European Union, in an attempt to counteract this trend, imposed a ban on the non-therapeutic use of antibiotics of human importance, such as tetracyclines, in farm animal feed. However, resistance to these drugs in Salmonella from food samples continues to be of concern [8,23]. This observation may be related to the human manipulation of these kinds of foods [24].
Of the strains tested, 53.7% showed an MDR profile with resistance to four or five classes in the majority of strains. These data are alarming, not only because of the real risk for consumers of becoming infected with an MDR strain, but also because many of these strains showed resistance to antibiotic classes important in human medicine, such as beta-lactamases. Thus, in order to obtain a complete overview of the resistance profiles of all the MDR strains isolated, we conducted a double-disc test (DDT) for ESBL phenotype detection. This test is one of the four different methods for confirming the ESBL phenotype recommended by EUCAST [21]. Despite the EFSA 2018/2019 report's observation of resistance to third-generation cephalosporins at the overall low levels of 1.8% and 1.2% for cefotaxime and ceftazidime, respectively, for Salmonella spp., our experiment indicated that 52% of all MDR strains had an ESBL phenotype [8]. Finally, because these phenotypes could be conferred by several ARGs [25], the detection of beta-lactamase genes was performed in order to confirm phenotypic pattern. The PCRs we conducted allowed us to identify at least one gene encoding for β-lactamase enzymes in each strain that had an ESBL profile ( Table 3). The bla CTX-M gene was present in 9 out of 19 ESBL strains, and in three of these, it was in association with the bla SHV gene, which was found to be the most prevalent gene among our isolates, because of its detection in 12 out of 19 ESBL strains. The bla CTX-M genes encode for extended-spectrum β-lactamases (ESBLs) frequently identified in Gram-negative pathogens. These types of enzymes are active against cephalosporins and monobactams (but not cephamycins or carbapenems), and are currently of great epidemiological and clinical interest [26]. The bla SHV gene has been identified mainly in Enterobacteriaceae causing nocosomal infections, but also in isolates from different contexts (human, animal and environment) [27,28]. Probably originating from a chromosomal penicillinase of Klebsiella pneumoniae, SHV β-lactamases currently comprise a large number of allelic variants, including extended-spectrum β-lactamases (ESBLs), non-ESBLs and several unclassified variants [29]. Our isolates showed an ESBL phenotype, so we have probably identified bla SHV genes encoded for extended-spectrum β-lactamases.
These data are certainly alarming, since all of our strains came from food samples, particularly poultry, intended for human consumption. Indeed, although cooking these products may reduce the risk of foodborne disease, ARGs can resist high temperatures and, once ingested, can be transferred to the gut microbiota and confer resistance to other bacteria [30]. Therefore, our data are in line with the latest EFSA recommendations, which confirm how important it is in the monitoring and surveillance of antibiotic resistance (AMR) to assess the presence of ARGs in foodborne strains, especially in a One Health approach that recognises the circularity of human, animal and environmental health.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/microorganisms10040780/s1. Table S1: Analysed strains and their phenotypic resistances. Funding: This research was funded by IZS SI 07/20 RC: "Study on the correlation between pesticides and antimicrobial resistance in bacteria isolated from plants".