Introduction

Each year, diarrheal diseases impact an estimated 550 million people worldwide and are thought to be the cause of 33 million fatalities. In India, Salmonella is one of the most frequent pathogens for gastrointestinal and systemic diseases [1, 2]. Approximately 2600 serovars of Salmonella enterica subsp. enterica (Salmonella enterica) have been reported, which cause enteric infections in both animals and humans [3, 4]. Non-typhoidal Salmonella (NTS) associated with many food products, poultry meat and eggs, is recognized as a potent zoonotic pathogen linked with foodborne transmission [5,6,7]. Salmonella spp. serovars are equipped with various mechanisms to invade and counter the oxidative stress within the cells. They also harbor diverse virulence factors resulting in enteric and systemic clinical manifestations in the host [8,9,10,11]. In the recent years, emergence of virulent multidrug-resistant (MDR) isolates of Salmonella has become a matter of serious public health concern [11].

A recent study reported the predominance of serovars Typhimurium, Infantis, Kentucky, Gallinarum, and Virchow in Indian poultry (2005–2019) [12]. A comparative study on poultry Salmonella strains representing 17 states (2011–2016) also involving the major egg-producing and consumer states reported Typhimurium, Gallinarum, and Enteritidis as the predominant serovars [13]. With an overall Salmonella prevalence of 3.5%, the predominant serovars reported were Enteritidis (68.1%) and Typhimurium (31.8%) from poultry and poultry products from Karnataka state. The study also reported 72.7% isolates as MDR (≥3 antimicrobial class) with the highest resistance was observed for polymyxin-B (81.8%) followed by nalidixic acid (72.7%) [14]. The emergence of serovar Agona among poultry farms in addition to commonly reported serovars was recently reported from the union territory of Jammu and Kashmir [15]. The use of antibiotic growth promoters (AGP) in poultry feed is unregulated in India. Testing of antimicrobial residues in 70 chicken meat samples meant for human consumption in New Delhi, India, indicated 40% positivity, with the predominant ones indicated were norfloxacin (20%), ciprofloxacin (14.3%), doxycycline (14.3%), oxytetracycline (11.4%), and chlortetracycline (1.4%) [16].

DNA-based fingerprinting techniques such as Enterobacterial Repetitive Intergenic Consensus PCR (ERIC-PCR) are rapid and sensitive assays and can be easily performed in small-scale laboratories [17]. Hence, many researchers have attempted genotyping of Salmonella isolates to infer their persistence and spread in different geographical regions [18, 19]. Thus, continuous monitoring of the emergence and spread of virulent AMR Salmonella serovars is essential to prepare for establishing preventive and control strategies. Despite the high prevalence of salmonellosis in India, there are limited reports on the detailed characterization of Salmonella strains associated with poultry. Considering these facts, the present study was envisaged to study AMR, virulence, and genotypic characteristics of Salmonella strains recovered from poultry in India during the last four years.

Materials and methods

Bacterial strains and serotyping

Twenty-six (n=26) Salmonella strains maintained at the repository of National Salmonella Centre-Veterinary (NSC-Vet), Indian Veterinary Research Institute were used in the present study. The strains were recovered from poultry (broilers) during a period ranging from 2019 to 2022 recovered from four different states viz., Uttar Pradesh, Uttarakhand, Andaman and Nicobar Islands, and Jammu and Kashmir, India. Among these, n = 12 strains were also reported in a previous study from Jammu and Kashmir, India [15]. The strains were revived using brain heart infusion (BHI) broth after incubation for 12–18 h at 37 °C following the sub-culturing on Hektoen Enteric (HE) agar. Green-colored colonies with black centers were presumptively identified as Salmonella strains and were subsequently sub-cultured on nutrient agar for downstream analysis. Further, biochemical characterization was carried out using catalase, triple sugar iron, urease, citrate, and motility indole lysine (MIL) tests. Salmonella serotyping was performed using agglutination test and specific antisera (SSI Diagnostica A/S, Denmark) according to the White-Kauffmann Le-minor (WKL) scheme [20]. Molecular serotyping of the strains was also performed using serovar-specific PCR as described earlier [12]. The genomic DNA was extracted using QIAamp DNA Mini Kit (Qiagen, USA). All PCR assays in the present study were carried out in 25 μL reaction mixture containing 2.5 μL of DNA template, 12.5 μL of 2x master mix (DreamTaq Green PCR Master Mix, Thermo Scientific™), 0.5 μL each of forward and reverse primers (10 pmol/μL), and nuclease-free water to make up the volume. The primers, amplicon lengths, and annealing temperatures used for molecular serotyping PCR are mentioned in Supplementary Table 1. The amplified PCR products were electrophoresed on 2% agarose gels containing ethidium bromide and visualized using the UV gel documentation system (Alpha Imager, Germany).

Antimicrobial susceptibility testing

The antibiotic susceptibility testing was performed using the Kirby-Bauer disk diffusion method [21] as per the Clinical and Laboratory Standards Institute guidelines [22]. Fourteen (n = 14) antimicrobial agents (Himedia, India) were tested at the following concentrations: cefotaxime + clavulanic acid (CEC, 30/10 μg), ceftazidime (CAZ, 10 μg), ceftazidime + clavulanic acid (CAC, 30/10 μg), streptomycin (S, 10 μg), levofloxacin (LE, 5 μg), ciprofloxacin (CIP, 5 μg), nitrofurantoin (NIT, 300 μg), tetracycline (TE, 30 μg), doripenem (DOR, 10 μg), ertapenem (ERT, 10 μg), meropenem (MER, 10 μg), imipenem (IMP, 10 μg), nalidixic acid (NA, 30 μg), kanamycin (K, 30 μg), amoxicillin/clavulanic acid (AMC, 30, 20/10), aztreonam (AT, 30 μg), trimethoprim-sulphathoxazole/co-trimoxazole (COT, 25 μg), and trimethoprim (TR, 5 μg). The extended spectrum beta-lactamase (ESBL) production in the Salmonella strains was determined by double disk diffusion test [23]. Escherichia coli strain ATCC 25922 and Klebsiella pneumoniae strain ATCC 700603 were used as the quality control strains. The zone of inhibition around the antibiotic disks was measured in mm and compared with the CLSI clinical break points (CLSI, 2018).

For genotypic resistance profiling, resistance genes were selected against antibiotics for which maximum strains were showing resistance. Molecular detection was carried out using polymerase chain reaction targeting the resistance genes like qnrA, qnrB, qnrC, qnrD, qnrS, Ib-cr, and qepA for quinolones [24,25,26,27,28], tetA, tetB, and tetC genes for tetracycline [29], and dfrA gene for trimethoprim [30]. The genomic DNA extracted using QIAamp DNA Mini Kit (Qiagen, USA) was used as the template DNA. PCR conditions and product visualization were carried out as mentioned in earlier section. The primers, amplicon lengths, and annealing temperatures used for PCR are mentioned in Supplementary Table 1.

Virulence profiling

The Salmonella strains were investigated by PCR for the presence of virulence genes using standard PCR protocols. Uniplex PCR targeting six virulence genes invA, agfA, lpfA, hilA, sopE, and spvC were performed. [31,32,33,34,35]. PCR conditions and product visualization were carried out as mentioned in the earlier section. The primers, amplicon lengths, and annealing temperatures used for PCR are mentioned in Supplementary Table 1.

ERIC profiling

Genotyping of the Salmonella strains was performed by Enterobacterial Repetitive Intergenic Consensus (ERIC) PCR using the primer pairs ERIC-F (5′-ATG TAA GCT CCT GGG GAT TCA C-3′) and ERIC-R (5′-AAG TAA GTG ACT GGG GTG AGC G-3′) [36]. The cycling conditions were as follows: initial denaturation at 95 °C for 7 min, followed by 30 cycles of denaturation at 90 °C for 30 s, annealing at 52 °C for 1 min, and extension at 65 °C for 8 min, and a final extension at 65 °C for 16 min [37]. The PCR products were separated using 2% agarose containing ethidium bromide. After electrophoresis, the gel images were captured and profiles were assigned manually.

Results

Bacterial strains and serotyping

Based on both the conventional and molecular serotyping, the strains were confirmed as belonging to the serovars Typhimurium (10/26), Infantis (5/26), Virchow (3/26), Kentucky (4/26), and Agona (4/26). The details of strains with respect to their serovar, host, source, farm, year of isolation, and geographical locations are shown in Supplementary Table 2.

Antimicrobial susceptibility testing

Twenty out of 26 strains (76.9%) were resistant to ≥1 of the tested antimicrobial drugs. The antimicrobial resistance profiles obtained for the strains are shown in Table 1. The antibiotic sensitivity testing (ABST) results revealed most of the strains resistant to tetracycline (TE) (n = 17/26; 65.38%) followed by nalidixic acid (NA) (n = 16/26; 61.53%) and trimethoprim (TR) (n = 8/26; 30.76%). In the present study, 61.53% (n = 16/26) of the strains were MDR isolates. None of the strains (0%; n = 26/26) were susceptible for all antibiotics, whereas 15.38% (n = 4/26) was resistant to one antibiotic, and another 15.38% (n = 4/26) was resistant to two different antibiotic classes. All the strains belonging to the serovars Infantis (5/5) and Virchow (3/3) were 100% MDR, whereas 75% (3/4) among Kentucky strains was MDR. However, only one Salmonella Typhimurium strain (n = 1/10; 10%) was identified as MDR. The two strains representing the serovar Infantis and one Typhimurium strain exhibited resistance to maximum numbers [6] of antimicrobial drugs, whereas the remaining Typhimurium strains were susceptible to most of the antimicrobial drugs. All strains except one representing serovars Kentucky (n = 3) and all Agona strains (n = 4) were resistant to levofloxacin, whereas among them 6 were resistant to ciprofloxacin. None of the Salmonella strains were identified as ESBL producers based on the double disk diffusion assay. Other than this, amoxicillin/clavulanic acid (AMC) resistance was exclusive to serovar Kentucky strains, and CIP resistance was exclusive to serovar Agona and Kentucky strains, whereas NIT resistance was predominant for serovar Infantis strains and K resistance predominant for serovar Infantis and Kentucky strains. All the above mentioned resistance profiles were also observed within strains belonging to the same geographical location (Jammu and Kashmir) and in most cases represented the same farms. On the other hand, the two Typhimurium strains representing two different farms from the same region differed for their AMR profiles. The serovar-wise resistance profiles are shown in Fig. 1A.

Table 1 Antimicrobial resistance profile of Salmonella strains. Among all strains (n = 26), 20 showed resistance to at least one antimicrobial agent tested. The highest number of strains showed resistance to tetracycline followed by nalidixic acid and trimethoprim
Full size table
Fig. 1
figure 1

Antimicrobial resistance and virulence pattern among Salmonella strains: serovar wise antimicrobial resistance (A) and virulence genes (B) are shown in the form of a bar diagram. The occurrence of viulence and AMR is represented as percentage calculated by considering each serovar seperately The number of strains respresenting each serovar was Typhimurium - 10, Infantis - 5, Kentucky - 4, Virchow - 3 and Agona - 4

Full size image

Out of the 26 Salmonella strains, 17 were resistant to TE in the disk diffusion assay. In accordance with the phenotypic assay, 16 out of the 17 tetracycline-resistant Salmonella strains (94.44%) were positive for the tetA (tetracycline efflux pump) gene. Thus, a good correlation between the phenotypic and genotypic expression of TE resistance was observed among the Salmonella strains. On the other hand, among the quinolone (levofloxacin and ciprofloxacin) (7/26; 26.92%) and trimethoprim (n = 8; 30.76%) resistant strains, corresponding resistant genes were not detected based on the genes targeted.

Virulence profiling

All the Salmonella strains were screened for the presence of virulence-associated genes, namely, invA, hilA, agfA, lpfA, sopE, and spvC. All the strains were found to carry a minimum of 3 virulence genes. The prevalence of virulence genes varied among strains. The highest prevalence was observed for invA, hilA, agfA, and lpfA (100%; n = 26/26). Strain wise distribution of virulence genes is shown in Table 2. Serovar wise distribution of virulence genes is shown in Fig. 1B. The strains were categorized into 3 virulence profiles based on the presence of virulence genes. Strains of profile P1 were having genes representing invA, agfA, hilA, and lpfA; P2 had virulence genes invA, agfA, hilA, lpfA, and sopE; P3 had virulence genes inv A, agfA, hilA, lpfA, and spvC. The frequency of occurrence of the virulence profile among the strains was P1 (50%), P2 (42.3%), and P3 (7.6%). The strains of the serovars Infantis (n = 5), Agona (n = 4), and Kentucky (n = 4) exhibited P1 virulence profile, whereas Salmonella Virchow (n = 3) showed P2 virulence profile. On the other hand, the strains of Typhimurium serovars (n = 10) manifested both P2 and P3 virulence profile. Out of the 10 Salmonella Typhimurium strains, 8 showed P2 profile and the 2 exhibited P3 profile. Salmonella serovars Infantis and Agona having the same geographical origin (Jammu and Kashmir) revealed similarity in virulence profile. Salmonella Kentucky (n = 4) strains from two different origins (Uttarakhand and Andaman and Nicobar Island) exhibited similar virulence profile. All the strains representing serovars Infantis, Kentucky, and Agona did not show the presence of spvC and sopE genes. Eight strains representing serovar Typhimuirum and three strains representing serovar Virchow harbored sopE gene, whereas spvC gene was unique to two Typhimurium strains isolated from suspected clinical cases from poultry.

Table 2 Virulence profiles among Salmonella strains. The strains represented 3 virulence profiles based on the presence of the virulence genes. The strains (n = 10), representing serovar Typhimurium indicated P2 and P3 virulence profiles
Full size table

ERIC profiling

PCR fingerprints of 26 Salmonella strains were obtained by ERIC-PCR. The ERIC-PCR profile was used to infer the genetic relatedness among isolates and to correlate with their serovar and geographical/farm origins. Strains with similar ERIC PCR fingerprints were assigned unique profile numbers (P1 to P10). In general, the strains representing the same serovar and belonging to same farm clustered together. The profile P1 represented Agona, P2 and P3 represented Kentucky, P4 and P5 represented Virchow and Infantis, P7–P10 represented Typhimurium strains. Two strains originating from the same geographical region (Jammu and Kashmir) within the serovar Kentucky formed a closely related cluster, whereas another strain belonging to a different geographical origin (Andaman and Nicobar Islands) formed separate cluster. Strains representing serovar Typhimurium represented 4 profiles, with two profiles (P9 and P10) representing strains from the same geographical origins (Uttarakhand). The remaining two strains from the same geographical origins (Uttar Pradesh) represented unique profiles (P7 and P8), but were isolated from different farms. The genetic diversity analysis of Salmonella strains using ERIC-PCR fingerprints along with the serovar information and geographical locations is depicted in Fig. 2.

Fig. 2
figure 2

ERIC PCR profiling of Salmonella strains. This figure depicts the ERIC-PCR profiles of Salmonella strains correlating with their serovars and geographical origins

Full size image

Discussion

Salmonellosis is a significant threat to public health worldwide. Globally, an estimated 93 million enteric infections occur annually because of non-typhoidal Salmonella (NTS) infections, causing 155,000 deaths [38]. NTS-associated infection is considered to be a neglected emerging enteric infection in India [39]. Unhygienic practices during the slaughtering of animals, improper handling, and transportation of raw meat all increase the likelihood of contamination of meat products with NTS [40]. The pathogenic potential and AMR of Salmonella can change over time and hence need to be routinely investigated for implementing appropriate control measures. Considering these facts, in the present study, a total of twenty-six Salmonella strains of poultry origin were subjected to antimicrobial susceptibility testing, virulence profiling, and strain typing based on ERIC-PCR profiling.

Among all, 20 strains (76.92%) were found to be resistant to at least one antimicrobial drug. The highest number of strains was resistant to tetracycline (TE) (65.38%). Earlier studies reported TE resistance ranging from 36.3 to 100% for Salmonella strains associated with poultry in India [41,42,43,44]. Resistance to tetracycline associated with broiler chicken from other countries like South Africa (93%) and Brazil (83%) was also reported in similar ranges [45]. Among all tetracycline resistance mechanisms, efflux pump tet gene classes A and B are recognized as the most common genes associated with resistance in Salmonella [46]. In the present study, tetA was found to be present in 94% of the resistant isolates. Significant association of tetA gene (56% to 100%) for tetracycline resistance in poultry Salmonella strains has been reported in India [41, 43, 44].

Treatment regime using fluoroquinolones may fail in patients infected with Salmonella spp. resistant to nalidixic acid (NA) [47, 48]. The present study identified nalidixic acid resistance among 61.53% of the isolates. Corroborating resistance percentage (56.25%) was reported by an earlier study on poultry associated Salmonella in Rajasthan, India [42]. Strains representing different geographical origin strains were also indicated with time-dependent changes (90.63 to 46.43% from 1990 to 2017) in their susceptibility pattern for NA [41].

The current study detected both trimethoprim and co-trimoxazole (COT) resistance in 34 % and ciprofloxacin (CIP) resistance in 23.07% strains. An earlier study reported 25.00% and 15.62% strains resistant to trimethoprim and ciprofloxacin, respectively, among a total of 32 strains [42]. However, none of the resistant strains were positive for respective AMR genes targeted by PCR in the study. Thus, it may be assumed that the phenotypic resistance may be mediated by point mutations or other mechanisms that are not explored in the study.

Presently, there is no national database or surveillance data for estimating the use of antimicrobials in health and veterinary sectors in India [49]. The antimicrobial drugs of choice for the treatment of human NTS infections are chloramphenicol, ampicillin, trimethoprim-sulphamethoxazole (cotrimoxazole), fluoroquinolones, and extended spectrum cephalosporins [41, 50]. However, antibiotic growth promoters (AGP) often included in animal feeds to promote growth can contribute to development of AMR bacteria. The AGPs reported for common use in Indian poultry sector are oxytetracycline, chlortetracycline, bacitracin, furazolidone, enrofloxacin, cephalosporins, ciprofloxacin, and tylosin [51]. Thus, the higher resistance for TE, NA, and CIP observed in the study correlates the possibility of high AGP usage in chicken. A high probability of resistance to NA, CIP, and TE in bacteria associated with Indian poultry sector was reported earlier based on studies in E. coli [52].

With respect to serovar association to any particular antibiotic resistance, we observed K, NIT, CIP, and AMC resistance associated to certain serovars such as Infantis, Agona, and Kentucky. In most cases, the strain represented the same farms and hence indicates these AMR profiles can be influenced by farm level managements including the source of poultry stock procurement to the antimicrobial usages within farm. Antimicrobial usage (AMU) pattern at farm levels correlating with high levels of resistance to the same antibiotic in Salmonella was reported in a recent study [53].

A 19-year spanning research study (2000 to 2018) conducted in India unveiled the slow emerging trends of antimicrobial resistance patterns associated with NTS serovars isolated from human feces [50]. A study involving 271 Salmonella isolates representing the period 1990 to 2017 from poultry, farm animals, and environmental sources similarly indicated a rise in antibiotic resistance for most of the tested antibiotics except cephalosporins and carbapenems [41]. All these factors emphasize the importance of NTS serovars and associated AMR risks in all sectors.

Salmonella serovars during the course of time can lose or acquire virulence factors as a result of adaptation to new hosts or environments [54]. In the present study, the presence of six virulence genes was investigated. None of the Salmonella strains was found to harbor all 6 virulence genes, whereas all the strains harbored the invA, hilA, agfA, and lpfA genes. Among these, invA and hilA genes are part of the Salmonella pathogenicity island 1 (SPI-1) and hence are required for host epithelial cell invasion in pathogenic Salmonella. On the other hand, aggregative fimbriae (agfA) and long polar fimbriae (lpf) are reported for involvement in colonization and virulence [55]. The agfA gene is associated with adhesion and biofilm formation [56], whereas lpfA gene is involved in adhesion to surfaces and epithelial cells, an essential prior stage of biofilm formation [57]. This hence indicates the essentiality of these genes for the survival of Salmonella in poultry gastrointestinal tracts and associated environments. Similar ubiquitous presence of invA, hilA, agfA, and lpf genes among all the involved poultry strains representing serovar Typhimurium and Enteritidis was reported by earlier studies [11, 54, 57].

In the present study, the sopE gene was detected in 11 (42%) and the spvC gene was detected in two (6.79%) strains. The spv (Salmonella plasmid virulence) operon is a highly conserved region that attenuates intestinal inflammation, promotes bacterial dissemination, and results in systemic infection [58]. In the present study, the presence of these genes in serovar Typhimurium strains isolated from clinical cases probably indicates their role in the septicemic manifestation of disease. Salmonella outer proteins (Sop) encoded by sop genes are the effector molecules of type-III secretion system (TTSS) which are involved in the early stages of Salmonella infection. Several isoforms of the sop genes have been identified (sopA–sopE) [59]. For the study strains, sopE gene was unique to all the Salmonella Virchow strains [3] and to those Salmonella Typhimurium strains [5] lacking the spvC gene. Similar variation among Salmonella Typhimurium strains with respect to carriage of spvC and spoE genes was reported from poultry strains in earlier studies [60]. The possible reason could be that the association of these genes differs within the genome as the spvC gene is a plasmid-borne virulence gene, whereas spoE gene is a prophage-related virulence gene [61].

In the present study, ERIC profiles indicated strain clustering correlating their serovar and geographical/farm origins. A unique profile mostly involved strains representing a particular serovar. Exceptions were observed for serovar Typhimurium and Kentucky were strains represented multiple profiles. In most cases, a different profile was observed when the strain had a different geographical or farm origin. Multiple ERIC profiles [5] within a single serovar are reported while involving 22 strains of serovar Gallinarum. The strains involved in a single profile were also belonging to different geographical locations and years of isolation [62]. Similarly, ERIC profiling of 45 poultry Salmonella isolates representing six serovars was indicated with 8 major profiles correlating their serovar and host origins, whereas 6 strains had unique profiles and remained unclustered in phylogeny. Among these, serovars Kentucky and Enteritidis strains were also involved in unrelated clusters [63]. This hence indicates strains representing a particular serovar usually cluster together, but often can also be represented in multiple clusters or form unique clusters.

Conclusion

To conclude, the present study characterized 26 strains of Salmonella belonging to serovars Typhimurium, Virchow, Kentucky, Infantis, and Agona. The strains exhibiting both MDR phenotype and maximum virulence genes may have the potential to evolve into a dominant clone with high zoonotic potential. Salmonellosis in humans due to such strain may overwhelm the current therapeutic regimes resulting in high treatment costs and fatality. The genetic diversity among the strains regardless of the serovars further necessitates continuous monitoring and surveillance of Salmonella strains among the poultry industry.