Longitudinal study highlights patterns of Salmonella serovar co-occurrence and exclusion in commercial poultry production

Amy T Siceloff^1,2Doug Waltman³Christian E Gunning⁴Sean P Nolan⁵Pejman Rohani^4,6,7Nikki W Shariat^1,2,8*

• ¹Poultry Diagnostic and Research Center, College of Veterinary Medicine, University of Georgia, Athens, Georgia, United States
• ²Department of Microbiology, University of Georgia, Athens, Georgia, United States
• ³Georgia Poultry Laboratory Network, Gainsville, Georgia, United States
• ⁴Odum School of Ecology, University of Georgia, Athens, Georgia, United States
• ⁵Nolan Integrated Pest Control and Management (NIPCAM) Group, Watkinsville, United States
• ⁶Center for the Ecology of Infectious Diseases, Odum School of Ecology, University of Georgia, Athens, United States
• ⁷Department of Infectious Diseases, College of Veterinary Medicine, University of Georgia, Athens, Georgia, United States
• ⁸Center for Food Safety, College of Agricultural and Environmental Sciences, University of Georgia, Griffin, Georgia, United States

Recent advances in next-generation sequencing approaches have revealed that Salmonella often exists in multiserovar populations, with important implications for public health as time and resource constraints limit serovar characterization by colony-based isolation methods. It is important to characterize Salmonella population dynamics to then understand how the microbial ecology influences serovar evolution and thus, animal and human health outcomes. Chicken remains the leading source of foodborne Salmonella outbreaks in the U.S., despite reductions in contamination at the product level, underscoring the need for targeted control strategies. This study aimed to survey multiserovar Salmonella populations in broiler breeder flocks and monitor fluctuations throughout production. Deep serotyping was performed on environmental breeder samples collected over two years as part of a surveillance program. About 18% (104/568) of samples contained multiple serovars, with serovar Kentucky negatively associated with other serovars, often excluding them. Longitudinal sampling across two commercial complexes over 65 weeks included pullet and breeder farms. Environmental samples were collected via premoistened boot socks and rodent bait boxes, with on-farm rodents captured. Salmonella prevalence in pullet flocks was 17% (11/64), while 41% (135/330) of breeder samples were positive, peaking at 38 weeks of age. Rodents showed 35% (17/49) positivity in gastrointestinal samples and 9% (3/33) in bait station swabs, with six serovars identified, three of which were shared with flocks. Our cross-sectional and longitudinal Salmonella surveillance highlights the complexity of serovar interactions with further work required to elucidate the mechanisms of competitive exclusion.