Salmonella is a food-borne zoonotic pathogen, which can be divided into more than 2,600 serotypes. The majority of these serotypes are zoonotic pathogens that infect humans and animals (1). The consequences of Salmonella infection in poultry include reduced productivity, increased mortality and elevated control costs, resulting in substantial economic losses for the poultry production industry (2). Reusable padding, water, feed, dust, infected people and contaminated equipment can be used as a repository for Salmonella in chicken houses (3). The horizontal transmission of pathogens among avian species in chickens can occur through contaminated food and water or via aerosols (4, 5). In some cases, Salmonella may become a permanent contaminant in the ventilation system, as it is difficult to clean and disinfect (2). The asymptomatic or symptomatic transmission of bacteria from infected animals can pollute the environment or directly infect others (6). For example, Salmonella pullorum is known to cause salmonellosis in poultry, resulting in significant economic losses for the poultry production industry. Healthy people can be infected with Salmonella by eating or handling contaminated, undercooked or raw chicken and products (7, 8). Furthermore, Salmonella has been identified as the primary bacterial cause of food-borne diseases on a global scale (9), thus posing a significant threat to food safety and public health (10). Consequently, effective measures to control Salmonella infection in poultry are imperative to minimize economic losses and human infections (11).

Antibiotics are widely used on poultry farms to promote growth and to prevent and treat bacterial infections. However, extensive and uncontrolled use of antibiotics can easily lead to the development of antibiotic-resistant bacteria (12). In addition, the development of drug resistance to key antimicrobial agents used in veterinary and human medicine poses a threat to the global healthcare system (13). In light of the mounting cognizance of AMR evolution in Salmonella, government entities and research organizations have initiated a longitudinal monitoring program, providing invaluable epidemiological data for risk assessment and the formulation of drug use guidelines (14).

Whole genome sequencing (WGS) has become a cost-effective and high-resolution genome analysis method, capable of providing significant information regarding antibiotic resistance genes (ARGs), genome mutation, MLST, and core genome MLST (cgMLST) (15–17). The World Health Organization has advocated for the implementation of genome-wide sequencing of Salmonella, to monitor the source and trend of transmission, and evaluate and monitor the effectiveness of prevention and control plans. The integration of WGS with epidemiological traceability data has been demonstrated to be a highly effective method for identifying the source of Salmonella infection. Furthermore, WGS has been demonstrated to facilitate the identification of determinants of AMR and the prediction of antibiotic resistance phenotype (18).

Despite the presence of homebred chicken farms in China, the epidemiology of Salmonella has received limited attention. To enhance comprehension of the dynamics of Salmonella in the context of homebred chicken farms, further study is required on the subject of Salmonella in chickens and the environment. The present study thus adopts a homebred chicken farm as its research object, employing epidemiological investigation and whole genome sequencing as its investigative tools. The primary objectives are to gain a comprehensive understanding of the epidemic situation and genetic relationship of Salmonella isolates and to elucidate the transmission law and epidemic characteristics of Salmonella.

The extensive and inappropriate utilization of antimicrobial agents has expedited the emergence and propagation of AMR (36). While AMR Salmonella infection has been reported on a global scale, the number of cases in developing countries is increasing at an alarming rate (37). The clinical use of antibiotics is a primary method of treating Salmonella infections. Chickens serve as natural Salmonella reservoirs, enabling zoonotic transmission through food chains. Salmonella infection in chickens and chicken farms is a significant concern and homebred chicken farms may play a significant role in the transmission of human salmonellosis (38). Therefore, it is necessary to detect Salmonella regularly in homebred chicken farms. Strict management measures, such as environmental disinfection, are effective in reducing the prevalence of Salmonella during the breeding and hatching processes.

In this study, a chicken farm in Jiangsu Province, where 229 cloacal swabs and 60 environmental swabs were collected in August 2024, and 60 strains of Salmonella were isolated and identified. Using WGS technology facilitates precise identification of the serotype of Salmonella. The predominant serotypes identified in this study is S. Enteritidis and S. Kentucky. As one of the most prevalent serotypes of non-typhoid Salmonella infection, SE in shelled eggs is widely regarded as the causative agent of large-scale foodborne outbreaks (39, 40). As previously reported, the average detection rate of SE in China was 1.5% from 2006 to 2012 and 15.1% from 2013 to 2021 (41). In this study, SE was identified as the most prevalent serotype, a finding that aligns with earlier research conducted in China (38, 41). The MLST model based on WGS showed that there were two kinds of ST in 60 Salmonella strains. The most prevalent STs were found to be ST11, followed by ST198, which is consistent with the findings of previous studies. Furthermore, the SE isolated from eggs and chickens also belongs to only one sequence type, namely ST11 (42, 43). Consequently, ST11 has been identified as the predominant sequence type, accounting for approximately 90% of the EnteroBase database of SE worldwide, which underscores the interconnectedness of poultry-associated AMR risks across continents (44).

For an extended period, the primary approach to disease prevention and control in China has relied on the utilization of antibiotics. This strategy has contributed to the high prevalence of drug-resistant bacteria and has presented significant challenges to prevention efforts. This study examined the antibiotic resistance patterns of isolates, with all isolates exhibiting resistance to a minimum of two different antibiotics. The isolates demonstrated high resistance levels to macrolides and penicillin, followed by compound glycosides. The resistance rate to erythromycin is more than 95%, which is similar to the previous research results (45, 46). This suggests that resistance to erythromycin is likely an intrinsic property of Salmonella, precluding its use for treating salmonellosis in chickens. These trends align with WHO reports of >50% AMR Salmonella in low/middle-income countries' poultry sectors. This phenomenon may be attributed to the predominance of AMP as the primary treatment for Salmonella, accompanied by an escalating trend in antibiotic resistance, which ultimately leads to the development of high-level drug resistance. The present study demonstrates the antibiotic resistance phenotypes were largely consistent with homologous ARGs. In this study, 29 types of ARGs were identified, with 11 types accounting for the highest percentage of 28.33%. The results showed that SK carried more ARGs than SE. In recent years, SK has become a common NTS associated with human infection (32, 33). SK is resistant to a variety of antibiotics (including aminoglycosides, beta lactams, sulfamethoxazole and tetracycline), and carries a variety of resistance genes such as aac(3)-Id, aada7, bla_TEM−1, sul1 and tet(A) (34, 35). Some isolates subsequently became highly resistant by obtaining resistance genes encoding extended spectrum bla_CTX−M, cephalosporinase (CMY) or carbapenem lactase (oxa-48, vim and NDM). These genes are often associated with IncC and IncI1 plasmids. Furthermore, the most prevalent mutations of ARGs are gyrA and parC. Mutation analysis demonstrated that parC (T57S) and gyrA (S83F) exhibited the highest mutation frequency, while prior reports substantiated that parC (T57S) was the most prevalent mutation in quinone resistance-determining regions (QRDR) (47). In addition, strains exhibiting mutations in both gyrA (S83F) and parC (T57S and S80I) demonstrated a high level of fluoroquinolone (FQ) resistance (48). Notably, resistance to quinolones and fluoroquinolones, particularly ciprofloxacin resistance, is a growing problem (49). Moreover, this study also found that 25% of the strains carried four mutations at the same time, including gyrA (D87N)+gyrA (S83F)+parC (S80R)+parC (T57S) and gyrA (D87G)+gyrA (S83F)+parC (S80R)+parC (T57S) (Figure 4B).

Despite the clear evidence of Salmonella infection in clinically affected chicken, both environmental swabs and cloacal swabs from asymptomatic contacts yielded negative results. This discrepancy may be attributed to several factors. Firstly, the implementation of stringent and routine biosecurity measures, such as frequent disinfection of facilities, equipment, and feeders, could effectively reduce the bacterial load in the environment to levels below the detection limit of culture methods. Secondly, inherent limitations of the sampling methodology, including the selection of specific sampling sites, the volume of material collected, and the potentially low and patchy distribution of pathogens in the environment, might have contributed to a false-negative outcome. Furthermore, the timing of sampling is critical; it is plausible that extensive sanitation procedures were initiated following the initial disease outbreak, thereby temporarily eliminating environmental contamination. Similarly, the negative results from all cloacal swabs of asymptomatic chicken suggest that, at the time of sampling, the infection may have been relatively contained within the clinically ill cohort and had not yet established widespread subclinical carriage or shedding within the broader flock. This pattern could reflect the early stage of the outbreak or the effectiveness of initial control measures in limiting horizontal transmission. While these negative results prevent us from mapping a complete transmission chain, they are nonetheless informative. They highlight the challenges in detecting intermittent or low-level shedding and environmental contamination, and underscore the critical importance of robust and persistent biosecurity protocols in breaking the cycle of infection, even when pathogens are not readily detectable in the immediate environment.

While this study provides comprehensive genomic insights, several constraints should be considered when interpreting the results. The exclusive focus on a homebred chicken farm from Jiangsu Province, while providing detailed regional data, limits the generalizability of findings to other regions with different agricultural practices. Moreover, this study is a cross-sectional survey that reveals the possibility of transmission, but cannot prove dynamic transmission events.

In summary, 60 strains of Salmonella were collected from 229 cloacal swabs and 60 environmental swabs from a chicken farm in a homebred chicken farm in Jiangsu Province in August 2024. The main serotypes are S. Enteritidis and S. Kentucky, and ST11 and ST198 are the corresponding ST types respectively. The potential risk of food-borne infection caused by AMR Salmonella is emphasized, and we should pay close and continuous attention to Salmonella isolated from animal-derived foods. Our findings provide baseline data for priority interventions to ensure food safety and public health.