Antibiogram Profiles and Risk Factors for Multidrug Resistance of Salmonella enterica Recovered from Village Chickens (Gallus gallus domesticus Linnaeus) and Other Environmental Sources in the Central and Southern Peninsular Malaysia

The emergence of multidrug resistance (MDR), including colistin resistance, among Enterobacteriaceae recovered from food animals poses a serious public health threat because of the potential transmission of these resistant variants to humans along the food chain. Village chickens or Ayam Kampung are free-range birds and are preferred by a growing number of consumers who consider these chickens to be organic and more wholesome. The current study investigates the antibiogram profiles of Salmonella isolates recovered from village chicken flocks in South-central Peninsular Malaysia. A total of 34 isolates belonging to eight serotypes isolated from village chickens were screened for resistance towards antimicrobials including colistin according to the WHO and OIE recommendations of critical antibiotics. S. Weltevreden accounted for 20.6% of total isolates, followed by serovars Typhimurium and Agona (17.6%). The majority of isolates (73.5%) demonstrated resistance to one or more antimicrobials. Eight isolates (23.5%) were resistant to ≥3 antimicrobial classes. Colistin resistance (minimum inhibitory concentrations: 4–16 mg/L) was detected among five isolates (14.7%), including S. Weltevreden, S. Albany, S. Typhimurium, and Salmonella spp. Univariable analysis of risk factors likely to influence the occurrence of MDR Salmonella revealed that the flock size, poultry production system, and use of antibiotics in the farm were not significantly (p > 0.05) associated with MDR Salmonella. The current study highlights that MDR Salmonella occur at a lower level in village chickens compared to that found in live commercial chickens. However, MDR remains a problem even among free-range chickens with minimal exposure to antibiotics.


Introduction
Increasing antibiotic resistance among foodborne pathogens is an emerging problem of global health importance [1] and the overuse of antibiotics in food animal production has been reported as one of its major drivers [2]. Colistin is a critically important antimicrobial in veterinary medicine [2], and is considered the drug of last resort against the emergent multidrug resistant (MDR) Gram-negative bacterial infections in humans, especially the carbapenem-resistant Enterobacteriaceae [3]. However, the overuse of colistin in the animal industry is reported to play an important role in the global emergence of colistin resistance [2]; as such, its use in livestock is being reconsidered in order to preserve drug efficacy [4,5]. Consequently an increasing number of consumers are choosing to consume

Antimicrobial Resistance and Antibiogram Profiles of the Salmonella Isolates
Tables 1 and 2 illustrate the isolate antibiotic resistance profile, multiple antibiotic resistance (MAR) index, and resistance phenotypes. Among the isolates, 26.5% (n = 9) were susceptible to all antibiotics, while 73.5% (n = 25) were resistant to at least one tested antibiotic. Multidrug resistance was displayed by eight isolates (23.5%). Ciprofloxacin (100%), gentamicin (97.1%), norfloxacin (97.1%), cefotaxime (97.1%), and ceftiofur (97.1%) were effective against most isolates. The highest level of resistance was observed for tetracycline (35.3%) and streptomycin (35.3%; Table 1).  3 show the frequency distribution of isolates resistant to commonly used antimicrobials according to the various sample sources, farms utilizing antimicrobials for treatment, and farm production systems. Generally, higher percentages of resistance were observed among isolates from farms that use antibiotics as compared to those farms that did not, and from the free-range system compared to the other systems. Most isolates found to manifest antimicrobial resistance were recovered from cloacal samples ( Table 2). Table 3 presents the distribution and percentages of MDR Salmonella recovered from cloacal swabs, drinking water, and flies caught at the farm.
MIC, minimum inhibitory concentration (mg/L) according to the MIC-strip microbroth dilution method, 2 MAR, multiple antibiotic resistance index, MAR index = No. of resistance antibiotic types/total number of antibiotic types tested, 3 R/I/S, Resistant, Intermediate or Susceptible according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines (http://www.eucast.org/clinical breakpoints/) (Te, Tetracyclines; Amp, Ampicillin; S, Streptomycin; W, Trimethoprim; S3, Sulfonamides; Amc, Amoxicillin-clavulanate; Na, Nalidixic acid; Ct, Chloramphenicol; Ctx, Cefotaxime; Cn, Gentamicin; F, Nitrofurantoin; K, Kanamycin). Figures 1-3 show the frequency distribution of isolates resistant to commonly used antimicrobials according to the various sample sources, farms utilizing antimicrobials for treatment, and farm production systems. Generally, higher percentages of resistance were observed among isolates from farms that use antibiotics as compared to those farms that did not, and from the freerange system compared to the other systems. Most isolates found to manifest antimicrobial resistance were recovered from cloacal samples ( Table 2). Table 3 presents the distribution and percentages of MDR Salmonella recovered from cloacal swabs, drinking water, and flies caught at the farm.         Table 2). The two former serovars originated from free range (n = 3), while the latter two were from semi-intensive (n = 2) production systems. All but S. Typhimurium were MDR (Table 2).

Analysis of Risk Factors for MDR Salmonella
Simple logistic regression analysis of the risks factors associated with the development of MDR Salmonella revealed that none of the factors investigated (including flock size, poultry production system, and use of antibiotics) were not significantly associated (p > 0.05) with MDR Salmonella (Table 4).

Discussion
Food safety issues, such as the presence of residual chemicals and antibiotic resistance, have resulted in an increased willingness among consumers to pay more for organically raised foods. Many Malaysians consume raw village chicken eggs because they believe it is medicinal, nutritious, safe, and antibiotic-free [6]. Salmonella isolates found in this study were susceptible to most of the antibiotics tested. High sensitivity levels were demonstrated against gentamicin, ciprofloxacin, cefotaxime, norfloxacin, and ceftiofur. This finding is consistent with similar studies that report a minimum level of resistance against the aforementioned antibiotics from local commercial chickens and chicken carcasses [23][24][25][26]. Complete susceptibility against cefotaxime, ciprofloxacin, and gentamicin was also demonstrated by Salmonella isolates recovered from Spanish broiler flocks [27]. In the present study, we found the highest levels of resistance were against tetracyclines (35.3%), streptomycin (35.3%), sulfonamides (29.4%), and trimethoprim (20.6%). In contrast, these levels were much lower than those reported against the same antibiotics (34-100%) for Salmonella isolates originating from commercial chickens in Malaysia [24,26]. The highest levels of resistance manifested by the isolates from the aforementioned studies were against tetracycline (100%), ampicillin (100%), clindamycin (100%), and ciprofloxacin (83%). Moreover our Salmonella isolates were susceptible to ciprofloxacin in contrast to that reported elsewhere as being between 30.8% and 96% [28][29][30][31].
Other studies around the world have shown the presence of antimicrobial resistant Salmonella recovered from backyard or free-range chickens. In most of these studies, resistance was demonstrated against ampicillins and tetracyclines [32][33][34][35][36]. Eight (23.5%) of the Salmonella isolates in this study were resistant to three or more antimicrobial agents (MDR; Table 3). This is comparatively lower than those reported in previous studies in commercial chickens in Malaysia, where a 100% and 75% of Salmonella recovered from cloacal swabs and chicken carcasses/products [26] respectively exhibited MDR [37]. In China, a similar level of MDR (26.3%) of Salmonella was also reported among free-range chickens [35].
Elsewhere, in commercial poultry as well as in retail chicken meats, or other related products, reports of the frequency of MDR Salmonella vary widely. For example, MDR Salmonella has been found in 80% of chicken carcasses and other related products in Egypt [28], with a similar level found in retail chicken meats in China [29]. Conversely, Spanish broiler flocks [27], as well as poultry houses [38] and broiler chicken farms [39] in Brazil reported MDR Salmonella as lower than 20%. MDR has been well documented among epidemiologically important serovars, such as S. Typhimurium, which can exhibit resistance to up to 13 antibiotics [25], and S. Enteritidis, which can demonstrate resistance to up to six antibiotics [24]. In our study, isolates of the aforementioned serotypes were not MDR (Table 3).
In Iran, a similar study found that S. Typhimurium isolates recovered from backyard chickens exhibited resistance against tetracycline, sulfamethoxazole, and trimethoprim [36].
The level of MDR found in this study, although meaningfully lower than those reported in commercial chickens locally and elsewhere, is worrying given that antibiotics are not routinely used in the production of village chickens. However, recent reports of the discovery of antibiotic resistance among animals in geographically remote locations where commercial antibiotics have not been applied may help to explain our findings [40]. The practice of free-range farming, allowing birds to roam freely, may result in their exposure to natural environmental hazards, such as untreated water and soil, which have been well documented to harbor drug-resistant foodborne pathogens [41][42][43][44]. Recent reports suggest that wild birds and gulls, creatures not exposed to the selective pressure of antibiotic use, have also been found to harbor MDR organisms in light of their continuous exposure to the natural environment [45][46][47][48]. It is also possible that feeding village chicken table scraps, which may contain resistant bacteria or materials, has additionally contributed to our observation of the phenomena. We found isolated MDR Salmonella from the cloacal swabs, drinking water and flies (Table 3) Table 2). All of these isolates originated from free-range and semi-intensive production systems where chickens spend much of their time free grazing. Since the first report of the plasmid-mediated colistin resistance gene (mcr-1) from China in 2015, reports on resistance have emerged from more than 30 countries across several continents. These resistant isolates were recovered from several sources, such as environmental samples, food-producing animals, ready-to-eat foods, fruits, and vegetables, and humans [51][52][53][54][55][56][57][58][59][60]. Colistin resistance has been reported worldwide [58]; nevertheless, the burden of colistin resistance is highest in Asian countries. For example, several reports have documented the identification of MCR-1-producing E. coli isolates from samples of chickens and chicken meat, pigs and piglets, cattle, calves, turkeys, and humans in Cambodia [61,62], South Korea and China [53,63,64], Japan [65,66], Laos [67], Nepal [68], Pakistan [69,70], India [71], Thailand [67,72], and Vietnam [73,74]. In 2015, a study in Laos reported a possible clonal transmission of colistin-resistant E. coli between a domesticated pig and a human [67]. Although the direction of the transmission cannot be ascertained, the findings indicate that the fluidity of resistant agent transmission between species is of animal and public health significance.

Source of the Isolates
The design of this study, sampling, and data collection, and the process of isolation and identification have been described in our previous publication [75]. Briefly, isolates and data were obtained from a cross sectional study involving 35 village chicken farms across 4 states in central and Southern Peninsular Malaysia. Table 5 shows the distribution of Salmonella serotypes analyzed in this study. All work was carried out at the Veterinary Public Health Laboratory, Faculty of Veterinary Medicine, Universiti Putra Malaysia.
The diameter of the zone of inhibition (mm) were interpreted according to the criteria of the The Clinical and Laboratory Standards Institute (CLSI) [76]. Strains were subsequently evaluated according to the CLSI breakpoints as susceptible, intermediate, or resistant. Digital Vernier calipers were used to measure the diameter of the zones of inhibition. S. Typhimurium (ATCC 29213) and S. Enteritidis (ATCC 25922) were used as reference strains for antibiotic disc control. A resistant isolate was defined as an isolate resistant to one or more of the agents tested, whereas isolates resistant to three or more classes of antimicrobials were classified as multidrug resistant (MDR) [79]. The multiple antibiotic resistance (MAR) index was defined as the proportion formed by the number of antibiotic types to which a particular isolate displayed resistance, to the total number of antibiotics to which the isolate had been evaluated for susceptibility [80,81]. MAR is a good tool for assessing health risk, and is used to determine whether an isolate originates from a region of high or low antibiotic usage [81]. An MAR index of >0.2 reflects a bacteria from a high risk source of contamination where several antibiotics or growth promoters are used, whereas <0.2 represent those from a low risk source or source with less antibiotic use [80,81].
MAR index = a/b, where "a" is the number of antibiotics to which the isolates were resistant, and "b" is the total number of antibiotics to which the isolate was exposed [80,81].

Data Analysis
Data generated from the study was subjected to descriptive analysis using MS Excel (version 2011) to obtain percentages and proportions. SPSS (version 22.0, IBM, Armonk, NY: IBM Corp.) was used for all analyses. Chi-square, Fisher's exact test, or simple logistic regression was used for the univariable exploratory analysis to identify risk factors associated with the outcome variable. The outcome variable Antibiotics 2020, 9, 701 9 of 13 was the presence or absence of multidrug resistant isolates (Salmonella isolates manifesting resistance to ≥3 classes of antimicrobial agents). The statistical significance level was set at α = 0.05.

Conclusions
The present study found that the level of resistance to antibiotics among isolates recovered from local village chickens was much lower than those found in commercial chickens and their products. Therefore, village chickens could be an alternative to those consumers seeking to reduce exposure to antibiotic resistant pathogens via the food chain. The findings also highlight the presence of MDR isolates, including those resistant to colistin, despite minimal antibiotic usage in free-range chicken production systems. We conclude that prolonged exposure to the natural environment plays a critical role in the transmission of resistance. Nevertheless, the role of the environment in the perpetuation and transmission of agents of resistance requires further study in light of emerging global trends in free-range food animal production.