Phenotypic and genotypic characterization of antimicrobial resistance in Escherichia coli and salmonella isolated across the production cycle of integrated broiler chicken farms in Karnataka, India

Introduction

Broiler production is among the fastest-expanding segments of Indian agriculture, and antimicrobials continue to be used therapeutically, prophylactically and, in some systems, for growth promotion. This reliance has coincided with a global rise in antimicrobial resistance (AMR), which the World Health Organization has identified as one of the leading public health threats of the current era [1][2][3][4] . Escherichia coli and Salmonella are widely used sentinel organisms for AMR surveillance in food animals because they are ubiquitous commensals or pathogens, are readily recovered from diverse sample matrices, and reflect the selective pressure exerted by antimicrobial use in the production environment [6,7] . Cross-sectional surveys of AMR in broiler flocks are increasingly available from South Asia, but comparatively few studies have tracked resistance longitudinally across a full production cyclefrom day-old chick placement to slaughter agein a way that distinguishes environmental, feed-and water-borne sources of resistant bacteria from bird-associated ones. Understanding when, during the cycle, resistant organisms and resistance genes first appear, and where they persist, is central to designing targeted interventions, since resistance already present at placement points toward hatchery-or breeder-level transmission rather than on-farm antimicrobial pressure alone. This study was undertaken to (i) isolate and identify E. coli and Salmonella from two integrated broiler farms and one franchisee farm in Bengaluru Rural district, Karnataka, at defined intervals across a single crop cycle; (ii) determine the phenotypic antimicrobial susceptibility profiles of the recovered isolates; and (iii) screen isolates for 22 resistance genes spanning β-lactam, plasmid-mediated quinolone resistance (PMQR) and mobile colistin-resistance (mcr) determinants, so as to identify the principal drivers and reservoirs of AMR within the broiler production chain.

Materials and Methods

Study area, farms and sample collection

The study was based in the Department of Veterinary Microbiology, Veterinary College, Bengaluru (Karnataka Veterinary, Animal and Fisheries Sciences University, Bidar). Three commercial broiler farms in Bengaluru Rural districttwo integrated units and one franchisee farm affiliated to a commercial breeder companywere monitored through one complete crop cycle each, between September and October 2021, using proportionate random sampling. From each farm, five sample categoriesboot socks (inside and outside the shed), water (drinkers, tank, borewell and puddle sources), feed (bags and feeders), pooled faecal swabs (inside and outside the shed) and gastrointestinal contents (ileum and caecum, collected on day 35)were obtained on days 0, 7, 14, 21, 28 and 35, giving a total of 101 samples. Samples were transported to the laboratory under cold chain and processed within 24 h.

Isolation and molecular confirmation

Samples were pre-enriched in buffered peptone water; for Salmonella, pre-enrichment was followed by selective enrichment in tetrathionate brilliant green bile broth. Enriched cultures were streaked onto Eosin Methylene Blue agar (EMBA) and EMBA supplemented with cefotaxime (to enrich for cephalosporin-resistant/ESBL-producing E. coli), and onto Xylose Lysine Deoxycholate (XLD) and Brilliant Green agar (BGA) for Salmonella. Presumptive colonies were purified on nutrient agar and confirmed by colony PCR targeting the E. coli-specific uspA gene (884 bp) [9] and the Salmonella-specific hilA gene (784 bp) [10] . Confirmed isolates were archived in duplicate as 30% glycerol stocks at -20 °C.

Antimicrobial susceptibility testing

A representative subset of isolates (one E. coli and one cefotaxime-selected E. coli per positive sample, n = 117; and all 10 confirmed Salmonella isolates) was tested by the Kirby-Bauer disc diffusion method on Mueller-Hinton agar against eight antimicrobials from six classes commonly used in poultry: ampicillin (10 µg), cefotaxime/ceftiofur (30 µg), doxycycline (30 µg), gentamicin (10 µg), amikacin (30 µg), enrofloxacin (5 µg) and trimethoprim-sulfamethoxazole (1.25/23.75 µg). Zone diameters were interpreted against standard breakpoints.

Minimum inhibitory concentration for colistin

Because colistin susceptibility is not reliably assessed by disc diffusion, resistance to colistin sulphate was determined for all 117 E. coli and 10 Salmonella isolates by broth microdilution, with an MIC greater than 2 µg/mL classified as resistant.

creening for antimicrobial resistance genes

Genomic DNA extracted by boiling lysis was screened by multiplex/simplex PCR for 22 resistance genes spanning three antimicrobial classes: nine β-lactamase genes (blaTEM, blaSHV, blaOXA, blaNDM, and blaCTX-M universal and groups 1, 2, 8/25 and 9) [11] , eight PMQR/quinolone-resistance genes (qnrA, qnrB, qnrC, qnrD, qnrS, oqxAB, qepA and aac(6′)-Ib-cr), and five mobile colistin-resistance genes (mcr-1 to mcr-5) [12] . Each run included positive, negative and notemplate controls, and amplicons were resolved on 2% agarose gels against a 100 bp DNA ladder.

Statistical analysis

Data were analysed in R (RStudio). The association between phenotypic resistance and the presence of the corresponding resistance gene(s) was assessed for enrofloxacin, ampicillin, ceftiofur and colistin sulphate using Pearson's chi-square test, with p < 0.05 considered statistically significant.

Results

Isolation of E. coli and Salmonella

Of 101 samples processed, 342 E. coli isolates were recovered (287 on EMBA and 55 on EMBA + cefotaxime) and confirmed by uspA PCR, corresponding to an overall culturebased prevalence of 76.2%. Recovery was highest (100%) from feed-feeder swabs, externally collected faecal swabs and ileal/caecal contents, and lowest from water tank (12%) and borewell/puddle sources (33.3%); E. coli was not recovered from boot socks collected inside the shed or from borewell water in Farm 1. Ten hilA-PCR-confirmed Salmonella isolates were obtained, an overall prevalence of 9.9%. Salmonella was not detected in Farm 1, a newly established unit, while boot-sock swabs yielded the highest recovery (27.7%) in Farms 2 and 3, followed by ileal/caecal contents (6.6%). Neither organism was recovered from feed bags or from water sources other than drinkers (Table 1).

Antibiogram of E. coli

All 117 E. coli isolates tested were resistant to at least one antimicrobial. Resistance was highest to ampicillin (91.3%) and doxycycline (89.3%), followed by gentamicin (74.0%), enrofloxacin (72.0%), trimethoprim-sulfamethoxazole (56.0%) and amikacin (48.7%); ceftiofur retained the greatest activity, with only 13.3% resistance. The pattern was broadly consistent across the three farms, although Farm 1 recorded the highest resistance to gentamicin, enrofloxacin and amikacin (Table 2). Notably, resistance to most agents was already substantial in isolates recovered on day 0, before birds

Colistin resistance by minimum inhibitory concentration

Colistin sulphate resistance (MIC > 2 µg/mL) was recorded in 42.3% of E. coli isolates overall, ranging from 32% (Farm 2) to 57% (Farm 1). Among Salmonella isolates, colistin resistance was markedly higher, at 79% overall (75% in Farm 2 and 83% in Farm 3).

Antibiogram of Salmonella

All ten Salmonella isolates were resistant to at least one antimicrobial. Resistance was highest to doxycycline (92%) and colistin sulphate (76% by disc diffusion; 79% by MIC), followed by ampicillin and enrofloxacin (64% each), gentamicin (44%) and trimethoprimsulfamethoxazole/amikacin (24% each); all isolates remained sensitive to ceftiofur (Table 2b).

Prevalence of antimicrobial resistance genes

Sixteen of the 22 screened genes were detected among the E. coli isolates. qnrB (96.9%) and qepA (92.7%) were nearubiquitous, followed by blaTEM (72.0%), blaCTX-M universal (29.6%), qnrS (28.9%) and qnrD (28.6%); individual blaCTX-M groups, oqxAB, blaSHV, blaOXA, aac(6′)-Ib-cr and qnrA were present at lower frequencies (0.5-10.8%). Among the mcr genes, only mcr-3 was detected, at 6.3% overall (farm range 4.2-10.5%); mcr-1, -2, -4 and -5 were not found in any isolate. Gene positivity was already 45.5% among isolates recovered on day 0, declining marginally to 37.9% by day 35, and was concentrated in bootsock, faecal-swab and ileal/caecal isolates. In Salmonella, only β-lactamase and PMQR genes were detected; no mcr gene was found. blaTEM and blaSHV were present in all ten isolates (100%), while qnrD (68.3%), qnrB (61.6%), qnrA (15%) and oqxAB (5%) were the predominant quinolone-resistance determinants. None of the blaOXA, blaNDM or blaCTX-M variants, nor qnrC, qnrS or aac(6′)-Ibcr, were detected (Table 3).

Phenotype-genotype correlation

Chi-square analysis showed a significant association between phenotypic resistance and the presence of the corresponding resistance gene(s) for enrofloxacin (χ² = 54.576, p = 3.99 × 10⁻¹¹), ampicillin (χ² = 124.71, p < 2.2 × 10⁻¹⁶) and ceftiofur (χ² = 106.78, p < 2.2 × 10⁻¹⁶). No significant association was found for colistin sulphate (χ² = 3.723, p = 0.0537), indicating that a negative mcr genotype does not reliably predict phenotypic colistin susceptibility (Table 4).

Discussion

The 76.2% culture-based prevalence of E. coli recorded here is broadly consistent with earlier broiler-associated surveys, in which recovery has typically ranged between about 62% and 72% [13,15] . Salmonella prevalence (9.9%) was comparatively low, sitting within the wide range (roughly 10-50%) reported elsewhere depending on flock age, biosecurity practice and sample matrix [24,31] . The absence of Salmonella from Farm 1 a newly commissioned unittogether with its recovery predominantly from boot-sock swabs in the other two farms, agrees with reports identifying boot-sock or environmental sampling as a sensitive indicator of farm-level Salmonella contamination [25] .

Resistance to ampicillin and doxycycline dominated the phenotypic profile of both organisms, echoing regional surveys that report similarly high resistance to these older, widely used agents alongside comparatively preserved susceptibility to extended-spectrum cephalosporins such as ceftiofur. Sarker et al. [16] reported an overall E. coli resistance of 61.7% in Bangladeshi broilers, with near-complete resistance to ampicillin and tetracycline and colistin resistance of 48.7%figures that align closely with the present findings.

A notable observation was the substantial phenotypic and genotypic resistance already present on day 0, before birds had received any farm-level antimicrobial treatment. This points to sources of AMR that precede on-farm antimicrobial exposure: vertical transmission of resistant organisms and resistance genes from breeder flocks through the hatchery to newly placed chicks have been repeatedly implicated in comparable studies [27,28] . This finding reframes farm-level antimicrobial stewardship as necessary but not, on its own, sufficient; hatchery-and breeder-level surveillance is likely required to meaningfully reduce the resistance burden entering grow-out farms. The near-universal presence of qnrB and qepA in E. coli, and of blaTEM/blaSHV in Salmonella, mirrors the wide dissemination of plasmid-mediated quinolone resistance and TEM/SHV β-lactamases among poultry-associated Enterobacteriaceae, both of which are frequently carried on mobile genetic elements that facilitate horizontal transfer [18,19] . The comparatively restricted mcr profileonly mcr-3 detected, and only in E. coliis noteworthy given that colistin use in animal feed has been prohibited in India. Persistence of mcr-3-positive isolates and of phenotypic colistin resistance (42.3% in E. coli, 79% in Salmonella) despite this restriction suggests either residual off-label use, environmental persistence of resistant clones, or resistance maintained through co-selection by other antimicrobials, consistent with observations from other South Asian settings where regulatory bans have not eliminated colistin-resistant isolates [21,22] . The strong phenotype-genotype concordance observed for quinolones and β-lactams, contrasted with its absence for colistin, is consistent with the more complex and still incompletely catalogued genetic basis of colistin resistance, which includes chromosomal mechanisms not captured by an mcr-targeted panel. This underscores that phenotypic MIC testing, rather than genotypic screening alone, remains essential for reliable colistin-resistance surveillance. Taken together, these findings identify boot socks, faecal material and gastrointestinal contentsrather than feed or wateras the principal reservoirs of resistant E. coli and Salmonella within the farm environment, consistent with previous reports implicating litter and faecal material as major AMR reservoirs capable of subsequently contaminating agricultural land through litter reuse [17] .

Conclusion

This production-cycle-wide survey shows that resistant E. coli and Salmonella, together with a substantial burden of transferable resistance genes, are already present in broiler flocks at placement and persist throughout the grow-out period, implicating vertical and environmental transmission alongside farm-level antimicrobial use as drivers of AMR. Boot-sock, faecal and gastrointestinal samples consistently carried the highest resistance-gene loads, making them practical sentinel matrices for routine surveillance. Given the near-universal presence of β-lactam and PMQR determinants, and continuing phenotypic colistin resistance despite regulatory restrictions, integrated interventionshatcherylevel monitoring, stricter farm biosecurity, prudent antimicrobial stewardship and continued genomic surveillanceare needed to curb the spread of AMR along the broiler production chain.

Acknowledgements

The authors thank the management and staff of the participating broiler farms for permitting access for sample collection, and the Department of Veterinary Microbiology, Veterinary College, Bengaluru, Karnataka Veterinary, Animal and Fisheries Sciences University, Bidar, for providing the laboratory facilities used in this study.

References

1. Satapathy D, Sharma A, Paswan JK, Sarkar S, Varun TK. Economic broiler farming: scope and limitations. Indian Farmer. 2017;4(5):393-405.
2. Chowdhury R, Islam KMS, Khan MJ, Karim MR, Haque MN, Khatun M, et al. Effect of citric acid, avilamycin, and their combination on the performance, tibia ash, and immune status of broilers. Poultry Science. 2009;88(8):1616-1622.
3. Van Boeckel TP, Brower C, Gilbert M, Grenfell BT, Levin SA, Robinson TP, et al. Global trends in antimicrobial use in food animals. Proceedings of the National Academy of Sciences of the United States of America. 2015;112(18):5649-5654.
4. Hayden D, Hedman KA, Vasco K, Zhang L. A review of antimicrobial resistance in poultry farming within lowresource settings. Animals. 2020;10(8):1264.
5. Eurosurveillance Editorial Team. WHO member states adopt global action plan on antimicrobial resistance. Eurosurveillance. 2015;20(21):21137.
6. Aarestrup FM, Bager F, Jensen NE, Madsen M, Meyling A, Wegener HC. Resistance to antimicrobial agents used for animal therapy in pathogenic, zoonotic and indicator bacteria isolated from different food animals in Denmark. APMIS. 1998;106(1-6):745-770.
7. Tadesse DA, Zhao S, Tong E, Ayers S, Singh A, Bartholomew MJ, et al. Antimicrobial drug resistance in Escherichia coli from humans and food animals, United States, 1950-2002. Emerging Infectious Diseases. 2012;18(5):741-749.
8. Eller C, Simon S, Miller T, Frick JS, Prager R, Rabsch W, et al. Presence of β-lactamases in extended-spectrum cephalosporin-resistant Salmonella enterica of 30 different serovars in Germany, 2005-2011. Journal of Antimicrobial Chemotherapy. 2013;68(9):1978-1981.
9. Chen J, Griffiths MW. Cloning and sequencing of the gene encoding universal stress protein from Escherichia coli O157:H7 isolated from a Jack-in-a-Box outbreak. Letters in Applied Microbiology. 1999;29(2):103-107.
10. Pathmanathan SG, Cardona-Castro N, Sanchez-Jimenez MM, Correa-Ochoa MM, Puthucheary SD, Thong KL. Simple and rapid detection of Salmonella strains by direct PCR amplification of the hilA gene. Journal of Medical Microbiology. 2003;52(9):773-776.
11. Dallenne C, Da Costa A, Decré D, Favier C, Arlet G. Development of a set of multiplex PCR assays for the detection of genes encoding important β-lactamases in Enterobacteriaceae. Journal of Antimicrobial Chemotherapy. 2010;65(3):490-495.
12. Buelow E, Bayjanov JR, Majoor E, Willems RJ, Bonten MJ, Schmitt H, et al. Limited influence of hospital wastewater on the microbiome and resistome of wastewater in a community sewerage system. FEMS Microbiology Ecology. 2018;94(7):fiy087.
13. Ghanbarpour R, Sami M, Salehi M, Ouromiei M. Phylogenetic background and virulence genes of Escherichia coli isolates from colisepticemic and healthy broiler chickens in Iran. Tropical Animal Health and Production. 2011;43(1):153-157.
14. Díaz-Sánchez S, Sánchez S, Ewers C, Höfle U. Occurrence of avian pathogenic Escherichia coli and antimicrobial-resistant E. coli in red-legged partridges (Alectoris rufa): sanitary concerns of farming. Avian Pathology. 2012;41(4):337-344.
15. Kevin N. Prevalence and antimicrobial susceptibility of Escherichia coli and Salmonella isolates from feeds, litter and cloacal swabs from broiler chicken in Kalerwe and Kasubi markets. Health Science Journal. 2020;15(7):796.
16. Sarker MS, Mannan MS, Ali MY, Bayzid M, Ahad A, Bupasha ZB. Antibiotic resistance of Escherichia coli isolated from broilers sold at live bird markets in Chattogram, Bangladesh. Journal of Advanced Veterinary and Animal Research. 2019;6(3):272-277.
17. Bhushan C, Khurana A, Sinha R, Nagaraju M. Antibiotic resistance in poultry environment: spread of resistance from poultry farm to agricultural field. New Delhi: Centre for Science and Environment; 2017.
18. Cui M, Xie M, Qu Z, Zhao S, Wang J, Wang Y, et al. Prevalence and antimicrobial resistance of Salmonella isolated from an integrated broiler chicken supply chain in Qingdao, China. Food Control. 2016;62:270-276.
19. Chuanchuen R, Padungtod P. Antimicrobial resistance genes in Salmonella enterica isolates from poultry and swine in Thailand. Journal of Veterinary Medical Science. 2009;71(10):1349-1355.
20. Rahman MM, Husna A, Elshabrawy HA, Alam J, Runa NY, Badruzzaman ATM, et al. Isolation and molecular characterization of multidrug-resistant Escherichia coli from chicken meat. Scientific Reports. 2020;10(1):1-11.
21. Uddin MB, Hossain SM, Hasan M, Alam MN, Debnath M, Begum R, et al. Multidrug antimicrobial resistance and molecular detection of mcr-1 gene in Salmonella species isolated from chicken. Animals. 2021;11(1):206.
22. Waghamare RN, Paturkar AM, Vaidya VM, Zende RJ, Dubal ZN, Dwivedi A, et al. Phenotypic and genotypic drug resistance profile of Salmonella serovars isolated from poultry farm and processing units located in and around Mumbai city, India. Veterinary World. 2018;11(12):1682-1688.
23. Rahmani M, Peighambari SM, Svendsen CA, Cavaco LM, Agersø Y, Hendriksen RS. Molecular clonality and antimicrobial resistance in Salmonella enterica serovars Enteritidis and Infantis from broilers in three northern regions of Iran. BMC Veterinary Research. 2013;9(1):66.
24. Bhuvaneswari M, Shanmughapriya S, Natarajaseenivasan K. Prevalence of multidrug-resistant (MDR) Salmonella Enteritidis in poultry and backyard chicken from Tiruchirappalli, India. Journal of Microbiology. 2015;5(2):28-35.
25. Berghaus RD, Thayer SG, Law BF, Mild RM, Hofacre CL, Singer RS. Enumeration of Salmonella and Campylobacter spp. in environmental farm samples and processing plant carcass rinses from commercial broiler chicken flocks. Applied and Environmental Microbiology. 2013;79(13):4106-4114.
26. Manic M, Nikolic S, Marjanovic V, Antanasijevic S, Raicevic Z, Petrovic M. Resistance testing of Salmonella spp. bacteria isolated from samples derived from poultry on antimicrobial drugs. Journal of Dairy, Veterinary & Animal Research. 2016;3(2):67.
27. Petersen A, Christensen JP, Kuhnert P, Bisgaard M, Olsen JE. Vertical transmission of a fluoroquinoloneresistant Escherichia coli within an integrated broiler operation. Veterinary Microbiology. 2006;116:120-128.
28. Börjesson S, Ny S, Löfmark S, Englund S, Ringman M, Bergström J, et al. Community carriage of ESBLproducing Escherichia coli is associated with strains of low pathogenicity: a Swedish nationwide study. Journal of Antimicrobial Chemotherapy. 2016;71(11):3173-3176.
29. Cui M, Zhang P, Li J, Sun C, Song L, Zhang C, et al. Prevalence and characterization of fluoroquinoloneresistant Salmonella isolated from an integrated broiler chicken supply chain in China. Frontiers in Microbiology. 2019;10:2244.