Ecology and Epidemiology of Campylobacter Jejuni in Broiler Chickens

Ecology and Epidemiology of Campylobacter jejuni in Broiler Chickens

A DISSERTATION SUBMITTED TO THE FACULTY OF THE UNIVERSITY OF MINNESOTA BY

Hae Jin Hwang

IN PARTIALL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

Dr. Randall Singer, Dr. George Maldonado

June 2019

Acknowledgements

I would like to sincerely thank my advisor, Dr. Randall Singer, for his intellectual guidance and support, great patience, and mentorship, which made this dissertation possible. I would also like to thank Dr. George Maldonado for his continuous encouragement and support. I would further like to thank my thesis committee, Dr. Richard Isaacson and Dr. Timothy Church, for their guidance throughout my doctoral training. I thank all my friends and colleagues I met over the course of my studies. I am especially indebted to my friends, Dr. Kristy Lee, Dr. Irene Bueno Padilla, Dr. Elise Lamont, Madhumathi Thiruvengadam, Dr. Kaushi Kanankege and Dr. Sylvia Wanzala, for their support and friendship. Heartfelt gratitude goes to my family, for always believing in me, encouraging me and helping me get through the difficult and stressful times during my studies. Lastly, I thank Sven and Bami for being the best writing companions I could ever ask for.

Abstract Campylobacteriosis, predominantly caused by Campylobacter jejuni, is a common, yet serious foodborne illness. With consumption and handling of poultry products as the most important risk factor of campylobacteriosis, reducing Campylobacter contamination in poultry products is considered the best public health intervention to reduce the burden and costs associated with campylobacteriosis. To this end, there is a need to improve our understanding of epidemiology and ecology of Campylobacter jejuni in poultry. The overall goal of this dissertation is to answer knowledge gaps regarding Campylobacter ecology and epidemiology in broiler chickens with an emphasis on the pre-harvest components of the broiler production system. The objectives included: (1) to assess current interventions and control measures practiced by the U.S. broiler industry to reduce Campylobacter contamination on broiler chicken products; (2) to investigate indirect selective pressures responsible for the persistence of fluoroquinolone-resistant

Campylobacter jejuni in broiler chickens; and (3) to describe the temporal changes in broiler litter bacterial microbiota with respect to bird age, flock cycle, antibiotic use and presence of Campylobacter on farms. For the first objective, we surveyed key stakeholders of the U.S. broiler industry, including poultry veterinarians, farm managers and processing plant managers. The survey respondents reported the use of various pre- and post-harvest interventions that are recommended by the USDA-FSIS to decrease

Campylobacter contamination in broiler chickens and on carcasses. Yet, the survey results also revealed the lack of understanding of Campylobacter epidemiology among the respondents, indicating that further education and training programs are necessary to improve understanding of Campylobacter among the key stakeholders which can lead to a reduction of Campylobacter contamination and improvement in food safety. To

investigate indirect selective pressures allowing persistence of Campylobacter jejuni in broiler chickens, we tested two hypotheses: limited bioavailability in poultry and activity of microcin B17. However, we found that the hypotheses tested were unlikely to be responsible for the persistence of fluoroquinolone-resistant Campylobacter jejuni despite the lack of fluoroquinolone use in the industry. Lastly, we investigated the temporal changes in the broiler litter microbiota by sampling seven commercial broiler farms over two flock cycles. We found that flock cycle and age of bird were two significant farm variables driving the bacterial diversity and composition of the litter microbiota.

Additionally, when the litter bacterial microbiota was compared between Campylobacter- positive and Campylobacter-negative samples, a lower abundance of Lactobacillus was observed among the positive samples. Overall, the work presented in this dissertation demonstrates collective efforts to improve our understanding of Campylobacter jejuni in broiler chickens using multidisciplinary approaches.

iii

Table of Contents Acknowledgements ...... i Abstract ...... ii Table of Contents ...... iv List of Tables ...... v List of Figures ...... vii Chapter One: Literature Review ...... 1 Chapter Two: Survey of the U.S. broiler industry regarding pre- and post-harvest interventions targeted to mitigate Campylobacter contamination on broiler chicken products ...... 16 Summary ...... 16 Introduction ...... 17 Methods ...... 20 Results ...... 22 Discussion ...... 29 Chapter Three: Investigation of indirect selective pressures conferring fitness advantages in fluoroquinolone-resistant Campylobacter jejuni in vitro ...... 50 Summary ...... 50 Introduction ...... 51 Materials and Methods ...... 53 Results ...... 63 Discussion ...... 65 Chapter Four: Characterization of temporal changes in the bacterial microbiota of conventional broiler litter ...... 80 Summary ...... 80 Introduction ...... 81 Methods ...... 83 Results ...... 88 Discussion ...... 95 Chapter Five: Conclusion ...... 141 Bibliography ...... 145 Appendices ...... 166 Appendix A. Survey Questionnaire...... 166

List of Tables Table 2.1. Descriptions of the processing plants and broiler farms that participated in the survey...... 36 Table 2.2. List of pre-harvest interventions and management practices implemented on- farm to reduce foodborne pathogen colonization in broiler chickens. (n* = total number of responses from broiler farm mangers, n† = total number of responses from poultry veterinarians)...... 37 Table 2.3. Opinions on the effectiveness of pre- and post-harvest interventions and management practices implemented on farm and in the processing plant to reduce Campylobacter contamination in broiler chickens and carcasses...... 38 Table 2.4. List of post-harvest interventions implemented in the processing plant to minimize microbial contamination on broiler chicken products...... 40 Table 2.5. New recommendations or changes that have been implemented within the past year to meet the USDA-FSIS performance standards for Campylobacter and Salmonella on raw poultry products by the U.S. broiler industry...... 42 Table 3.1. Oligonucleotide primers used in this study...... 72 Table 3.2. List of genes and their potential functions measured using RT-qPCR...... 73 Table 3.3. Characteristics of the FQ-R and FQ-S C. jejuni strains used in the study...... 74 Table 3.4. Comparison of growth kinetics between FQ-S and FQ-R C. jejuni using pairwise permutation statistical tests with 10,000 simulations on OD600...... 75 Table 4.1. Descriptions of the commercial broiler chicken farms enrolled in the study. 105 Table 4.2. Summary of the final constrained model that best describes the variation observed in the bacterial composition of the broiler litter samples...... 106 Table 4.3. Results of the analysis of multivariate homogeneity of groups dispersions among farm variables...... 107 Table 4.4. Summary of PERMANOVA based on Bray-Curtis dissimilarities showing the farm variables associated with the shifts in the bacterial composition in the broiler litter samples...... 108 Supplementary Table 4.1. Results of PERMANOVA analyses showing all models tested prior to picking the best model...... 121 Supplementary Table 4.2. Results of DESeq2 differential abundance testing showing ASVs that were differentially abundant in flock 2 relative to flock 1 within No- Antibiotic-Ever and conventional farms...... 124 Supplementary Table 4.3. Results of DESeq2 differential abundance testing showing ASVs that were differentially abundant in No-Antibiotic-Ever farms relative to the conventional farms over all time points...... 131

Supplementary Table 4.4. Results of DESeq2 differential abundance testing showing ASVs that were differentially abundant in Campylobacter-positive litter samples relative to Campylobacter-negative litter samples from week 3 to 7 weeks of bird age...... 137

List of Figures Figure 2.1. The approximate incidence of Campylobacter and Salmonella from internal microbiological testing using whole carcasses at processing plants...... 44 Figure 2.2. The number of broiler farms that are regularly tested positive for Campylobacter and Salmonella in the processing plant...... 45 Figure 2.3. Scores from a 5-point Likert scale showing the survey participants’ understanding of the changes made to the United States Department of Agriculture– Food Safety and Inspection Service performance standards for Campylobacter and Salmonella on raw poultry products...... 46 Figure 2.4. Comparison of understanding of on-farm ecology and epidemiology of Campylobacter in broiler chickens...... 47 Figure 2.5. Ranking of on-farm management strategies targeted to reduce Campylobacter colonization in broiler chickens based on their effectiveness...... 48 Figure 2.6. Scores from a 5-point Likert scale showing the survey participants’ familiarity with the risk factors associated with the increased Campylobacter contamination in broiler chickens and on carcasses...... 49 Figure 3.1. Relative expression ratio of iron acquisition and oxidative stress regulation genes in FQ-R and FQ-S C. jejuni strains over time...... 76 Figure 3.2. Relative expression ratios of iron acquisition and oxidative stress regulation genes in FQ-R C. jejuni relative to FQ-S C. jejuni under normal and iron-limiting conditions ...... 77 Figure 3.3. The effect of mccB17 on the growth of E. coli DH5α strains susceptible or resistant to mccB17 and ciprofloxacin...... 78 Figure 3.4. The effect of mccB17 on the growth of FQ-R and FQ-S C. jejuni...... 79 Figure 4.1. Sampling scheme ...... 109 Figure 4.2. Estimated levels (logCUF/gram) of total and fluoroquinolone-resistant Campylobacter spp. in the broiler litter...... 110 Figure 4.3. Comparison of alpha diversity of the broiler litter within and among farms...... 111 Figure 4.4. Comparison of alpha diversity of the broiler litter by age of bird and farm type over two consecutive flock cycles...... 112 Figure 4.5. Principal coordinate analysis plots of beta diversity measured by Jaccard and Bray-Curtis indices...... 113 Figure 4.6. Constrained redundancy analysis ordination plots showing the effect of flock cycles and age of bird after controlling for the farm effects...... 114 Figure 4.7. Changes in the microbial composition of the broiler litter at the phylum taxonomic level over time...... 115

vii

Figure 4.8. Changes in the relative abundance of families belong to Firmicutes over time...... 116 Figure 4.9. Top 50 most abundant ASVs among farms across time...... 117 Figure 4.10. Bacterial genera differentially abundant between two flock cycles within each farm type...... 118 Figure 4.11. Bacterial genera differentially abundant in NAE farms relative to conventional farms...... 119 Figure 4.12. Plot summarizing the differences observed in the bacterial composition between Campylobacter-positive and Campylobacter-negative litter samples...... 120 Supplementary Figure 4.1. Principal coordinate analysis ordination plots based on Jaccard similarities and Bray-Curtis dissimilarities showing the ASV membership and composition between front and middle-back sections of the broiler houses...... 139 Supplementary Figure 4.2. Principal coordinate analysis ordination plots showing dispersions of farm variables tested in PERMANOVA analysis...... 140

viii

Chapter One: Literature Review

1.1 General characteristics of Campylobacter spp.

Campylobacter spp. (Campylobacter) is a Gram-negative, facultative intracellular organism belonging to the class of Epsilonproteobacteria. While Campylobacter has been reported and identified as an animal pathogen since the early 1900s 1-3, due to its fastidious growth conditions, its importance as a causative agent of bacterial gastroenteritis illness was not established until the 1970s with the development of selective media that could isolate and support the growth of Campylobacter in the laboratory 4. Campylobacter was previously taxonomically grouped as Vibrio spp., and its infections in humans and animals were described as “related Vibrio infections” 5. In

1963, the genus Campylobacter was proposed by Véron and Sebald 6, and subsequently

Vibrio-like organisms, including Campylobacter jejuni (C. jejuni) and Campylobacter coli (C. coli) were regrouped as Campylobacter species in 1973 7.

Campylobacter is a fastidious organism, requiring very specific conditions to grow. As a thermophilic and microaerophilic organism, Campylobacter grows optimally at 42ºC, although minimal growth occurs at 31-32ºC 8,9, in an atmosphere containing

10 approximately 5% O2, 10% CO2 and 85% N2 . Another distinctive characteristic that separates Campylobacter from other enteric bacteria is its inability to metabolize glucose

11. Instead, Campylobacter utilizes amino acids and tricarboxylic acid cycle intermediates as carbon sources 11,12. Despite the metabolic limitations, Campylobacter is capable of colonizing a wide-variety of wild and domestic animal hosts 13. Avian species is the

preferred host species of Campylobacter, as it appears that its optimal growth is supported by the natural intestine temperature of avian species 14.

Campylobacter is sensitive to a number of environmental stressors, including exposure to atmospheric oxygen level 15,16, desiccation 17,18, as well as heat 19,20 and cold shocks 9.

However, the organism is capable of surviving extended periods of time in unfavorable environmental conditions using different survival mechanisms 21. Campylobacter has been shown to enter a viable but non-culturable (VBNC) state upon exposure to unfavorable conditions 22,23. The VBNC state is characterized by morphological changes, reduced metabolic rates and increased tolerance to chemical and physical stressors, accompanied by the loss of culturability in growth media 22,24,25. However, it is debatable whether VBNC Campylobacter cells can be resuscitated to colonize the host and/or initiate an infection in the host 26-29. Additionally, it has been shown that Campylobacter can form biofilm when exposed to aerobic conditions 30,31. The study by Ica et al. showed that once the biofilm is formed, Campylobacter cells may enter a VBNC state to protect the cells from detrimental conditions 32. However, Teh et al. argue that there is not enough evidence to conclude that the formation of biofilm in Campylobacter is a part of the survival mechanism; instead, the observed formation of biofilm may be a result of simple attachment to abiotic surfaces or food matrices 33. Nonetheless, the ability of

Campylobacter to survive outside the host and adapt to different environmental conditions demonstrates how Campylobacter can survive through the food production chain.

1.2 Campylobacter jejuni in human foodborne illness

Campylobacter is one of the most common causes of acute gastroenteritis in developed countries 34. Campylobacteriosis, an infection caused mostly by C. jejuni35, typically occurs as sporadic cases, although outbreaks of Campylobacteriosis have been reported previously 36-39. The majority of campylobacteriosis cases are asymptomatic or cause mild symptoms, such as watery and/or bloody diarrhea, fever, nausea and abdominal cramps 4. These mild clinical symptoms can persist for up to one week without requiring specific antibiotic treatments 4. However, more serious complications, such as pancreatitis, peritonitis, bacteremia, reactive arthritis and Guillain-Barré syndrome, can occur causing serious long-term sequelae 40. The exact mechanisms by which

Campylobacter infection causes these long-term sequela are unknown.

The estimated incidence of campylobacteriosis in U.S. is 14 cases per 100,000 people 41.

In the year 2015, 13 deaths were attributable to Campylobacter infections with 6,289 confirmed cases reported by the FoodNet surveillance system in the U.S 42. However, because of the asymptomatic nature of campylobacteriosis, the actual incidence is estimated to be much higher, even estimated to be 9 million cases per year in Europe 43.

The incidence of campylobacteriosis in developed countries typically peaks between July and August, which coincides with the high isolation levels of Campylobacter from poultry during the summer months 44. On the contrary, campylobacteriosis in developing countries is characterized as endemic and less severe with most cases being asymptomatic 45. Campylobacteriosis is estimated to be more prevalent in developing countries 45,46.

The infectious dose of C. jejuni is estimated to be low. A self-inflicted experiment by

Robinson showed that as little as 500 C. jejuni cells were able to induce diarrhea and

abdominal pains 47. Similarly, experimental studies showed that 800 to 105 CFU of C. jejuni cells were able to cause clinical symptoms in healthy adult volunteers 48,49.

Additionally, these experimental studies showed that attack rates increased with the higher doses ingested 48,49. The typical incubation period of C. jejuni is estimated to be 2

– 4 days, with shorter incubation periods followed by infections with higher doses of C. jejuni 48,49.

Because campylobacteriosis is often self-limiting, patients typically recover without specific treatments 36. Replacement of lost fluids and electrolytes is recommended for mild cases 50. However, antibiotic treatment may be indicated for severe cases, especially for at-risk subpopulations, including the elderly, children and those with compromised immune systems 4,51. Macrolides are a common choice of antibiotic for treating campylobacteriosis, especially in children in which the use of fluoroquinolones, another alternative antibiotic, is contraindicated 50,52.

A number of source attribution studies have identified consumption or handling of raw or undercooked poultry products as the most important risk factor associated with sporadic cases of campylobacteriosis 53,54. The high levels of Campylobacter colonization poultry can maintain as well as the high consumption of poultry products worldwide indicate poultry as the most important source of Campylobacter in humans 55. Additional risk factors associated with campylobacteriosis are: consumption of unpasteurized dairy products 53,56, direct contact with farm animals or pets 57, swimming in contaminated water 58,59, and drinking untreated water 53,59. Moreover, a recent study by Ricotta et al. suggests that a large proportion of U.S. campylobacteriosis may be attributable to travel to developing countries 60. Between 2005 and 2011, approximately 18% of

campylobacteriosis cases in the U.S. were associated with international travel, and these infections were more likely to be caused by antibiotic resistant Campylobacter compared to domestically acquired infections 60. The sources of foreign-acquired campylobacteriosis have not been identified, however, it is likely that similar sources

(e.g., consumption of contaminated poultry products and/or cross-contaminated foods) are responsible for the infection 60.

Because poultry is the major source of sporadic campylobacteriosis, reducing

Campylobacter contamination in poultry products is considered the most effective strategy available to reduce the public health burden of campylobacteriosis in humans.

However, contamination control measures targeted to poultry production system have not been effective. In fact, Iceland and New Zealand, two countries that were able to make a drastic reduction in the incidence of campylobacteriosis, demonstrate that campylobacteriosis is a multifaceted problem requiring multidisciplinary measures 61-63.

In Iceland, a national epidemic of campylobacteriosis was declared in 1998, a year after campylobacteriosis became a reportable disease 61. The incidence of campylobacteriosis rose from 12.2 to 116 per 100,000 population within a 4-year period, and the availability of fresh poultry products was attributable to the epidemic 62. In order to control the epidemic, several initiatives were implemented to multiple levels of the food production chain 61. Surveillance and monitoring system were implemented at pre- and post-harvest, and products produced from Campylobacter-positive flocks were required to be frozen before sold on the market to reduce the number of viable Campylobacter on the products

61. On-farm biosecurity and general hygiene of farms and processing plants were enhanced through producer education programs 61. Consumer education campaigns

focused on improving awareness of campylobacteriosis and reducing the risky food handling and cooking behaviors in households 61,62. These targeted efforts resulted in a

72% reduction in the incidence of campylobacteriosis in Iceland 64. Similar initiatives that targeted both pre- and post-harvest sectors as well as retails and consumers were implemented in New Zealand, resulting in a 54% reduction in the average incidence of campylobacteriosis 63. The reduction seen in these two countries are remarkable, yet, it is questionable if the similar successes can be achieved in other countries 64.

1.3 Campylobacter jejuni in broiler chickens production

1.3.1 Pre-Harvest

In the U.S., the commercial poultry industry is vertically integrated; a single firm owns breeder farms, hatcheries, processing plants and feed mills, and independent farmers work under contract with the firm to grow broiler chickens until the birds reach the market weight, which takes about 6 to 7 weeks 65. During the grow-out and processing stages, Campylobacter can enter the production chain and contaminate either flocks on farm or carcasses at the processing plant.

C. jejuni is generally considered a commensal organism in broiler chickens due to the absence of inflammatory response induced by C. jejuni colonization and inability of chickens to clear C. jejuni out from the intestines 66. As such, there is no clinical signs associated with C. jejuni colonization in broiler chickens. In broiler chickens, C. jejuni colonization takes place in the lower gastrointestinal tract, and is thought to persist in the intestine by undergoing rapid replication in the mucus followed by invasion and evasion from cecal crypt, thereby avoiding the expulsion from the intestine 66,67. Broiler chickens

can harbor high levels of C. jejuni, up to 109 CFU of C. jejuni per gram of cecal contents

67.

Once C. jejuni is introduced into a flock, it is expected that the whole flock would become colonized with C. jejuni within a couple days 68. However, several studies pointed out that not all chickens become colonized with Campylobacter within a positive flock, and the levels of colonization may vary largely within flocks 69-71. The C. jejuni colonization in broiler chickens is not observed until 2-3 weeks of broiler age 72. This colonization timing coincides with waning maternal immunity which suggests that maternal antibodies protect broiler chicks from C. jejuni colonization in the early life 73.

Additionally, the decrease in the shedding of Campylobacter as broilers age has been reported 74-76. For example, a study by Dipineto et al. showed that the level of

Campylobacter colonization decreased over time in laying hens, resulting in some birds to be free of Campylobacter 76. This observed phenomenon suggests that matured immunity and/or development of stable intestinal microflora may help to clear

Campylobacter colonization in older birds.

The potential of vertical transmission of Campylobacter in broiler chickens has been investigated extensively 77-82. An experiment by Shanker et al. showed that unless C. jejuni was directly inoculated into albumin, the colonization did not initiate naturally even when the egg shells were challenged with C. jejuni 78. In addition, several studies have failed to identify same genotypes of C. jejuni in parent flocks and their offspring 79-

82. It appears that maternal immunity against Campylobacter prevents vertical transmission and colonization of Campylobacter in newly hatched chicks 73.

The major mode of transmission of Campylobacter in broiler chickens is horizontal transmission 72. Many epidemiological studies have focused on identifying sources of

Campylobacter in poultry farms, and several potential sources have been identified.

These sources are: insects 81,83,84, rodents 85, wild birds 86, other farm animals on poultry farms 87, environment around and within the poultry farm (e.g., soil, puddle, and drinking water) 88-91, and residual Campylobacter population from previous flocks 68,92. However, many studies have failed to detect Campylobacter from the potential sources until the flock was colonized, failing to demonstrate the directionality of Campylobacter exchange. This failure could be due to the fact that 1) many studies had inadequate study design and sampling schemes, 2) previous studies fail to identify unknown sources of

Campylobacter around and within the poultry farm, 3) the current laboratory techniques cannot resuscitate and culture Campylobacter in the VBNC state, or 4) the current genotyping techniques are not sensitive enough to differentiate Campylobacter strains due to high frequency of DNA recombination observed in Campylobacter genome 93,94.

Based on the source attribution studies, a number of on-farm risk factors associated with

Campylobacter colonization in broiler flocks have been identified. Most of these risk factors are related to having improper biosecurity and hygiene measures on the farm 95-99, such as not disinfecting drinking water, a lack of anteroom and boot dips, and a lack of proper pest controls to prevent entry of wild reservoirs of Campylobacter into the farm.

Other on-farm risk factors are seasonality (summer vs. other seasons) 97,98,100,101, reusing litter 101,102, duration of downtime period (period between flocks to clean and disinfect broiler houses) 84,103, thinning (partial depopulation) practice 104,105, general conditions of broiler houses 98, and age of broilers 98.

Enhancing biosecurity and general hygiene practices are regarded as the most effective on-farm intervention at preventing the introduction of Campylobacter to broiler flocks.

However, even the most strict on-farm biosecurity measures seem to fail to prevent introduction of Campylobacter in broiler flocks 44. Furthermore, these measures only seem effective in Scandinavian countries where the prevalence of Campylobacter in broiler chickens is already low, and harsh climatic conditions may prevent persistence of certain insect and rodent populations that may carry Campylobacter 106. Other on-farm interventions, such as feed and water acidification have yielded limited and inefficacious results 107. Likewise, acidifying litter treatments to make the litter environment less favorable to Campylobacter and other foodborne pathogens, such as Salmonella, seem to have limited effects on the levels of Campylobacter in broiler chickens 108,109. Currently, there are no cost-effective and reliable on-farm interventions available to reduce the prevalence and/or levels of Campylobacter in broiler chickens.

Other novel approaches to reduce Campylobacter colonization on farm include use of bacteriocins 110-113, vaccination 114-116 and probiotics 117-120. Of these, vaccination of broiler chickens against Campylobacter has shown the most promising results. Neal-

McKinney et al. showed that intramuscular injection of Campylobacter vaccine developed using C. jejuni surface-exposed colonization proteins showed a 2-log reduction in the level of C. jejuni colonization in 20-day old chicks 114. Likewise, a study by Wyszyńska et al. used attenuated Salmonella as a vector to deliver Campylobacter immunogenic surface proteins 115. This vaccine trial showed that oral vaccination of 4- week old broiler chickens greatly reduced the prevalence and levels of C. jejuni colonization 115. However, these novel interventions are under development and require

further research to address several key issues, such as optimizing the best delivery system and testing efficacy under field settings. Until these interventions become commercially available, producers are left with no choice but to rely on enhancing biosecurity of the poultry farms.

1.3.2 Post-Harvest

The primary processing of broiler chickens typically involves the following steps: transportation from farm to processing plant, live haul, bleeding, scalding, picking

(defeathering), evisceration, and chilling.

Transportation of a flock from the farm to the processing plant poses a great risk for cross-contamination of Campylobacter both within and between flocks 121-123.

Epidemiological studies have shown that improper cleaning and disinfection of transportation modules and vehicles can result in Campylobacter-negative birds to become contaminated with the residual Campylobacter found on these equipment during the transportation. The exterior of broilers can become contaminated with Campylobacter

124. This catching and transportation of broiler chickens event is even more crucial in

European countries where thinning (i.e., early or late slaughtering of a subset of a flock to meet different market weight demands) is commonly practiced, which greatly increases the risk of introducing Campylobacter to the remaining broilers unless strict biosecurity measures are implemented 125.

Throughout the processing steps, the levels of Campylobacter on carcasses can change markedly. Although complete elimination of Campylobacter on carcasses during processing is hard to achieve, reduction of the levels of Campylobacter on carcasses is

possible, while the cross-contamination of negative carcasses is possible as well. The steps that are known to increase the foodborne pathogen cross-contamination are scalding126, picking126,127 and evisceration128.

Scalding involves immersion of carcasses into hot water to open up feather follicles so feathers can be easily picked during picking step. High fecal materials found on feathers can contaminate scalding tanks, leading to cross-contamination between carcasses 129.

During the picking or defeathering step, feathers and the uppermost layer of the skin are removed by picking fingers prior to evisceration 130. Fecal materials can be released when the picking fingers rub against the carcasses, again leading to cross-contamination between carcasses 126,131. Evisceration is a step that removes internal organs and any processing defects from the carcasses prior to chilling. Because high loads of

Campylobacter is typically found in the intestine of broiler chickens, rupturing of the intestine can release Campylobacter, contaminating carcasses 132. Any carcasses with visible contamination on carcasses are inspected and re-processed. However, reprocessing does not necessarily decrease the bacterial levels on the carcasses 133. As such, preventing the release of Campylobacter from contaminated carcasses and direct contacts between contaminated and clean carcasses is essential.

During each processing step, implementation of proper antimicrobial and hygienic measures are crucial to reduce the levels of Campylobacter found on the carcasses and to limit the cross-contamination between carcasses. The United State Department of

Agriculture - Food Safety Inspection Service (USDA-FSIS) recommends frequent rinsing of the processing machinery to reduce the buildup of contaminants and bacteria 134.

Addition of antimicrobial agents, such as chloride and peroxides is recommended in rinse

water and chilling tanks as well 134,135. Implementing counter-current flow of water is also recommended, such that carcasses are progressively exposed to cleaner water 136.

Maintaining a proper pH in water tanks to prevent the growth of Campylobacter and other foodborne pathogens is another recommended practice 134. Finally, improving maintenance of processing machines and equipment have been shown to reduce cross- contamination between carcasses 134.

1.4 Antimicrobial resistance in Campylobacter spp.

Antibiotics are a type of antimicrobials that either kill or slow the growth of bacteria.

Since the discovery of penicillin by Alexander Flaming in 1928 137, antibiotics have greatly improved infectious disease prevention and treatment in human and veterinary medicines. Antibiotics were first introduced in food animal production as growth promoters in the 1940s 138. Since then, the overuse of antibiotics in agriculture has caused the rise in the antimicrobial resistant foodborne pathogens 139,140. With the increasing public health concern of the transmission of antibiotic resistant foodborne pathogens to humans via the food chain, the use of antibiotics that are medically important to humans as growth promoters in food animal production has been banned in the U.S. in 2017 141.

Isolates of Campylobacter resistant to various antibiotics have been reported in both developing and developed countries. Interestingly, Campylobacter is intrinsically resistant to a variety of antibiotics, including vancomycin, trimethoprim, rifampicin and bacitracin, although the exact mechanisms of these intrinsic resistance are unclear 142,143.

Besides these antibiotics, the most common antibiotics Campylobacter is resistant to are fluoroquinolone (FQ), macrolides and tetracycline 143,144.

The prevalence of Campylobacter resistant to macrolides is reported to be low, less than

10% in the U.S. and other developed countries 145,146, whereas the increased macrolide resistance among Campylobacter isolates has been reported in developing countries 147.

Most macrolide resistance has been reported in C. coli isolated from swine, as tylosin, a macrolide antibiotic, was frequently used as growth promoter in swine production 145.

The resistance to macrolides seems to be low in Campylobacter isolated from other animal origins, including broiler chickens 148. The development of macrolide resistance in

Campylobacter is thought to be a slow process, as the rate of the emergence of macrolide resistance is estimated to be low, requiring a prolonged macrolide exposure 149. In addition, a high fitness cost associated with the macrolide resistance makes the resistance phenotype less stable, allowing Campylobacter to lose the resistance phenotype quickly in the absence of the selection pressure 143,149.

Unlike macrolide, a steady increase in the FQ resistance among Campylobacter isolates has been reported worldwide 150. The FQ was first introduced to poultry production in the

1980s to control respiratory diseases 151. However, since then the prevalence of FQ resistant Campylobacter isolated from poultry and humans has increased dramatically 145.

A study by Nachamkin et al. showed a significant increase (less than 5% to 40.5%) in the

FQ resistance rate of C. jejuni isolated from patients in health care system since 1992, with isolates exhibiting high levels of resistance 152. A recent study conducted in China showed that almost 100% of the Campylobacter isolates from chicken showed resistance to FQs during 2008 and 2014 153.

Because of the broad-spectrum activity of FQs and its wide use in treatment of various infectious diseases, including campylobacteriosis, FQ resistance in Campylobacter can

complicate the treatment 154. Several studies have shown that campylobacteriosis caused by FQ-resistant Campylobacter was likely caused by FQ-resistant Campylobacter coming from contaminated poultry products. For example, a study by Kassenborg et al. showed that patients who had FQ-resistant campylobacteriosis were less likely to have taken FQs before they were diagnosed with the infection compared to those who were diagnosed with FQ-susceptible campylobacteriosis 155. Additionally, the study results showed that those patients with FQ-resistant campylobacteriosis were ten times more likely to have eaten poultry products seven days prior to the onset of the infection compared to the matched healthy controls, suggesting that the consumption of poultry products that were contaminated with FQ-resistant Campylobacter was a likely source of the infection. As such, the use of FQs in U.S. poultry industry was banned by the U.S.

Food and Drug Administration in 2005. Nonetheless, this action has not resulted in a decrease in FQ-resistant Campylobacter in poultry or in humans in the U.S. This phenomenon suggests that FQ-resistant Campylobacter can persist in broiler chickens without the direct selective pressures, and FQ resistance may increase fitness of

Campylobacter, as shown by the in vivo and in vitro studies by Luo et al. 156 and Han et al 157.

The acquisition of FQ resistance in Campylobacter is demonstrated to be rapid; several in vivo and in vitro studies have shown that less than 24hrs is required to detect FQ-resistant

C. jejuni after the administration of the drug 158,159. The development of FQ resistance in other enteric bacteria, such as Salmonella and Escherichia coli, requires stepwise accumulation of point mutations in gyrA and parC genes 160,161. However, in

Campylobacter, single point mutation in gyrA is sufficient to confer resistance to FQs 162.

A substitution of isoleucine for threonine at position 86 in the gyrA is the most frequently reported point mutation in Campylobacter 163,164. This point mutation in gyrA has been shown to reduce negative supercoiling of DNA which alters DNA topology during transcription, replication and recombination 157. Alterations in DNA supercoiling influence several promoter functions and ultimately transcriptomic expressions in vivo and in vitro 165,166. In fact, a recent study that compared transcriptomic expressions between isogenically-paired (i.e., theoretically the only genetic difference between FQ- resistant and -susceptible strain is the point mutation in gyrA) FQ-resistant and FQ- susceptible C. jejuni showed differential gene expressions in vitro 167. This study results suggest that reduced negative supercoiling due to FQ resistance enhances iron uptake and regulation systems in C. jejuni in the absence of FQs. Nonetheless, the ecological drivers of FQ resistance dynamics within Campylobacter population and risk factors associated with emergence and persistence of FQ-resistant Campylobacter in U.S. broiler production are not well understood.

Chapter Two: Survey of the U.S. broiler industry regarding pre- and post-harvest interventions targeted to mitigate Campylobacter contamination on broiler chicken products

Summary

Campylobacter is one of the most commonly reported foodborne pathogens in the U.S.

Because poultry is considered the major source of Campylobacter in humans, reducing

Campylobacter contamination on poultry products is the most important and effective public health strategy to reduce the burden of campylobacteriosis in humans. To better understand the current pre- and post-harvest Campylobacter control strategies implemented by the U.S. broiler industry, a comprehensive survey was administered to key stakeholders of the U.S. broiler industry. The survey results indicate that while the

U.S. broiler industry is following the USDA- FSIS recommended guidelines to control

Campylobacter contamination in broiler flocks and on carcasses, the majority of the participants felt that current interventions are not effective at reducing Campylobacter contamination, especially on-farms. Additionally, findings from the survey suggest the lack of understanding of risk factors associated with the increased risk of Campylobacter colonization in broiler flocks. Continued educational and training programs of key stakeholders of the U.S. broiler industry are needed to inform the industry about progress being made around the mitigation of Campylobacter in the broiler production system and that Campylobacter is a multifaceted program that requires efforts from both the pre- and post-harvest sectors.

Introduction

Campylobacter spp. (Campylobacter), the causative agent of campylobacteriosis, is one of the most common foodborne pathogens reported globally 168,169. Handling or consuming poultry products is considered a major risk factor for campylobacteriosis

106,170. Most campylobacteriosis cases are asymptomatic or mild, causing diarrhea, nausea and abdominal cramps 52. However, severe infection can occur in certain at-risk populations, including the elderly, children or those with compromised immune system, requiring antibiotic treatments 171.

The United States Department of Agriculture (USDA) – Food Safety and Inspection

Service (FSIS) has established Salmonella and Campylobacter performance standards for raw poultry parts and Not-Ready-To-Eat (NRTE) comminuted poultry products to reduce risks posed by Salmonella and Campylobacter contamination on broiler chicken and turkey products to consumers. The performance standards are intended to assess a processing establishment’s ability to effectively address Salmonella and Campylobacter contamination during slaughter and the grinding processes.

In 2015 and 2016, FSIS introduced new performance standards for Salmonella and

Campylobacter as well as the new routine, random sampling scheme for the performance standard verification testing in which a moving window approach is used to measure the establishment’s process control continuously 172,173. The new performance standards were proposed to meet the Healthy People 2020 goals of reducing human illnesses caused by

Salmonella by about 25% and by Campylobacter by about 33% by the year of 2020 173.

The change in sampling scheme from the set-based sampling program (i.e., samples

collected during a defined number of sequential days of production) to a moving window approach (i.e., sequential samples collected randomly given the fixed timeframe of 52- weeks) should allow FSIS to continuously monitor the prevalence of Salmonella and

Campylobacter during the processing while accounting for seasonal variations 173.

Additionally, the FSIS announced that they would be posting the verification testing results on its website for public knowledge. This information includes the names of poultry slaughter establishments and their performance category (consistent process control, variable process control and highly variable process control). The FSIS anticipate that web-posting of the verification testing results would encourage establishments to improve Salmonella and Campylobacter control performances during the processing, which in turn, would result in a reduction in human illnesses associated with Salmonella and Campylobacter 173.

While the FSIS has announced that it will not take actions against establishments exceeding the new Campylobacter performance standards in raw poultry products yet 174,

U.S. broiler production companies will be under pressure once the performance standards are enforced. This pressure will especially be difficult because currently there are few, if any, commercially available on-farm interventions specifically targeted to reduce

Campylobacter in broiler chickens other than improving biosecurity of a farm to prevent entry and further spread of Campylobacter in flocks 175. There have been several attempts to develop Campylobacter poultry vaccines 114,115,176, however, none are commercially available yet. Likewise, use of on-farm antimicrobial interventions, such as litter, water and feed acidifiers has shown limited effects at reducing the levels of Campylobacter in broiler chickens 108,177,178. Additionally, there are no reliable microbiological testing

methods to detect and to enumerate Campylobacter on farm and in the processing plant

179. Lack of such laboratory testing methods makes it difficult to survey Campylobacter in a flock and to carry out logistic processing (i.e., processing a Campylobacter-free flock before Campylobacter-positive flock) to reduce potential cross-contamination between flocks.

A successful mitigation of the risk posed by Campylobacter contamination on broiler chicken products will require reduction of Campylobacter levels in broiler chickens on farm and further reduction of Campylobacter contamination on broiler carcasses during processing stages using various antimicrobial interventions as suggested by quantitative risk assessment studies 180,181. In order to do that, pre- and post-harvest interventions need to be tailored specifically to Campylobacter with consideration of Campylobacter ecology and epidemiology in broilers, as demonstrated by Iceland and New Zealand.

These two countries were able to achieve a significant reduction in the incidence of campylobacteriosis associated with raw poultry products by adopting a combination of pre- and post-harvest interventions 61,63. This demonstrates that campylobacteriosis is a multifaceted problem, requiring multidisciplinary approaches.

Currently, it is unclear whether the current approaches taken by the U.S. broiler industry to mitigate Campylobacter contamination problems on broiler chicken products are effective and sufficient. Therefore, a comprehensive survey of the pre- and post-harvest interventions adopted by the U.S. broiler industry was conducted targeting the key stakeholders of the industry, including poultry veterinarians, broiler farm managers or supervisors and processing facility managers and food safety or quality assurance team

members. The objective of the survey was twofold: 1) to assess the effectiveness of the current pre- and post-harvest interventions implemented by the U.S. broiler industry to mitigate Campylobacter contamination on broiler chicken products, in part to meet the new FSIS performance standards, and 2) to gain better insight into the stakeholder’s perception and knowledge about Campylobacter ecology and epidemiology in broiler chickens and the recent changes made to the performance standards. The results of this survey will aid the industry by informing Campylobacter control strategies that have been attempted in the past and currently being practiced, as well as identifying knowledge gaps regarding Campylobacter ecology and epidemiology in broiler chickens for future education and research opportunities.

Methods

A cross-sectional survey was conducted from January to June 2018. The survey was created using web-based survey software (Qualtrics, Provo, UT, USA). The survey was targeted to key stakeholders in the U.S. broiler industry who are currently employed at vertically integrated broiler production companies located in U.S. These key stakeholders included broiler farm managers or supervisors, poultry veterinarians, processing facility managers and food safety or quality assurance team members. A hyperlink to the online survey was distributed the National Chicken Council (NCC) and the U.S. Poultry & Egg

Association (USPOULTRY). Announcements were also made at professional and commodity meetings. All participation was voluntary, and no compensation was given to any participants. A complete print-version of the survey is included in the Appendix A.

All surveys were anonymously conducted, and no personal or company-specific information was collected. The survey was largely divided into three sections. The first section focused on collecting basic information about the farm or processing facility at which each participant is currently employed. These questions included the number of chickens slaughtered weekly, number of farms under supervision, use of microbial tests to survey Campylobacter and Salmonella on the farm and in the processing plant, and approximate incidence of Campylobacter in the processing plant, as well as the

Campylobacter and Salmonella status (regularly tested positive or negative) of the participant’s establishment.

The second part of the survey focused on gauging participant’s understanding of the changes made to the USDA-FSIS Salmonella and Campylobacter performance standards on poultry products in 2015 and 2016. Participants were asked to list any recent changes that have been made within the past year on the farm and/or in the processing plant to reduce Campylobacter contamination on broiler chicken products and to meet the FSIS performance standards. To compare the broiler industry’s mitigation strategies for

Campylobacter versus those applied to other foodborne pathogens, similar questions were asked about Salmonella control. Additionally, to gauge participant’s knowledge about Campylobacter ecology and epidemiology in broiler chickens, questions about the modes of transmission between chickens, potential sources of Campylobacter on farms, and familiarity with risk-factors associated with the increased contamination of

Campylobacter in broiler chickens were asked. Likewise, participants were asked to rank five different on-farm interventions and management practices (use of antibiotics,

management of drinking water, litter, feed and biosecurity) in the order of effectiveness at reducing Campylobacter colonization in broiler chickens.

The last sections of the survey focused on pre- and post-harvest mitigation strategies adopted by the U.S. broiler industry. The last set of questions participants received depended on the job title they answered at the beginning of the survey. Specifically, poultry veterinarians and broiler farm managers were asked to record any on-farm interventions implemented to reduce Campylobacter colonization in broiler chickens, including antimicrobial control strategies applied to drinking water, litter and feed and biosecurity on farm. Processing plant managers were asked to record antimicrobial control strategies applied on scalding, picking (defeathering), evisceration, chilling stages and second processing. All descriptive statistical analyses were performed in RStudio

(ver. 1.1.419) 182 using R version 3.5.0 183. All visualization of the data was performed using licorice (ver. 0.1) 184 and ggplot2 (ver. 3.1.0) 185 R packages.

Results

A total of 81 surveys was initiated and 56 responses (69%) were completed. Twenty surveys (36%) were completed by processing facility managers, food safety or quality assurance team members. The remaining surveys were completed by 18 (32%) broiler farm supervisors and 18 (32%) poultry veterinarians. The median number of birds slaughtered weekly in the companies represented by the respondents was 727,500 with a range between 80,000 and 60,000,000. A majority of the processing plants practiced both primary and secondary processing (Table 2.1). Only one respondent answered that their processing plant practices logistic slaughter on a subset of incoming flocks, while the

remaining respondents answered they do not practice logistic slaughter because they do not know the Campylobacter or Salmonella status of incoming flocks. The number of farms supervised by broiler farm managers ranged from 16 to 350 with a median of 120.

The number of farms under supervision of poultry veterinarians ranged from 200 to 2000 with a median of 500.

Eighty percent (n=16) of the respondents reported that their processing plants conduct their own microbiological tests to survey Campylobacter and Salmonella contamination on poultry products (Table 2.1). Based on the results of the microbiological tests, more than half of the processing plants tested positive for Campylobacter and Salmonella

(Figure 2.1). Unlike at the processing plant, 11% (n=2) of the poultry veterinarians and none of the broiler farm managers reported that the farms under their supervision conduct on-farm microbiological tests to determine Campylobacter status of the flocks (Table

2.1). In contrast, a slightly higher percentage of the participants (33% of the poultry veterinarians and 56% of the broiler farm managers) responded that on-farm microbiological tests are conducted to detect Salmonella in the flocks. Fifty-six percent

(n=10) and 88% (n=16) of poultry veterinarians answered that the broiler farms under their supervision regularly test positive for Campylobacter and Salmonella in the processing plant, respectively (Figure 2.2). Notably, 61% (n=11) of broiler farm managers did not know if the farms under their supervision regularly tested positive for

Campylobacter in the processing plant. However, broiler farm managers were more aware of Salmonella status of the farms under their supervision. Thirty-nine percent

(n=7) answered that their supervised farms regularly tested positive for Salmonella in the processing plant while 11% (n=2) did not know.

All of the processing plants had validated interventions throughout the slaughter processes to specifically reduce Salmonella contamination on broiler products (Table

2.1). However, only 33% (n=6) answered that they have implemented validated interventions to reduce Campylobacter contamination during the processing steps.

Likewise, a higher percentage of broiler farm managers and poultry veterinarians answered that the farms under their supervision have farm-specific biosecurity management plans to reduce Salmonella colonization in broiler chickens on the farms as compared to Campylobacter.

Pre- and post-harvest interventions implemented to reduce microbial contamination

Table 2.2 shows the list of pre-harvest interventions and management practices adopted to reduce foodborne colonization in broiler chickens on-farm. Sanitation of drinking water and administration or application of acidifiers to drinking water and litter to lower pH and to make these environments unfavorable for microbial growth were common management practices implemented on-farm. Some of the other management practices recorded were windrowing of litter between flocks to reduce growth of foodborne pathogens and use of antibiotics in feed. When asked if these on-farm interventions and management practices are effective at reducing Campylobacter colonization in broiler chickens, a majority of the participants answered that they were either not effective or did not know (Table 2.3). One of the comments made by a poultry veterinarian was that many of these on-farm interventions may be helpful but their effectiveness against

Campylobacter is likely marginal. Particularly, most of the broiler farm managers answered “Don’t know” to these opinion questions. Additionally, the participants thought

that improvements in biosecurity of farms does not help to reduce Campylobacter colonization in broiler chickens.

Table 2.4 summarizes post-harvest antimicrobial interventions and management practices implemented in the processing plants to minimize bacterial contamination on broiler carcasses. Most processing plants adopted the processing practices and antimicrobial interventions recommended by the FSIS compliance guidelines for controlling

Salmonella and Campylobacter in poultry products. Use of counter flow of process water, administration of antimicrobial agents in wash or dip and multiple carcass washes between processing stages were commonly practiced by the processing plants. When asked if the current processing practices are effective at reducing Campylobacter contamination on carcasses, a majority of the processing plant managers responded ‘Yes’

(Table 2.3). One of the processing plant managers commented that none of the antimicrobial interventions and processing management practices are working to reduce

Campylobacter contamination on carcasses, while another suggested that interventions implemented to reduce Salmonella contamination are working to reduce Campylobacter contamination on carcasses as well. Notably, only a 5% of processing plant managers responded that the transportation modules are cleaned between flocks to reduce cross- contamination of Campylobacter and Salmonella (Table 2.4), despite the fact that more than half of the participants (56%) agreed that cleaning the transportation module, vehicles and catching equipment is helpful in reducing cross-contamination of

Campylobacter between flocks.

Understanding and achieving USDA-FSIS performance standards on raw poultry products

A 5-point Likert scale with the options ‘Extremely familiar’, ‘Very familiar’,

‘Moderately familiar’, ‘Slightly familiar’, and ‘Not familiar at all’ was used to determine the participant’s familiarity with the changes made to the FSIS performance standards for

Campylobacter and Salmonella on poultry products. Familiarity varied greatly among the three job types (Figure 2.3), showing that plant managers were most familiar with the changes made to the FSIS performance standards while broiler farm managers were the least familiar with the changes. A total of 85% (n=17) of processing plant managers answered they were either very familiar or extremely familiar with the new changes made to the FSIS performance standards. In contrast, none of the broiler farm managers were very familiar or extremely familiar with the changes. Fifty-six percent (n=10) of the broiler farm managers answered they were not familiar at all or are slightly familiar with the changes. About half of the poultry veterinarians (44%, n=8) answered that they are either very familiar or extremely familiar with the changes.

Table 2.5 summarizes the efforts taken by the broiler companies to meet the new performance standards. Participants were asked if their company had made any changes, such as amending or implementing new control strategies or interventions to pre- and/or post-harvest processes within the past year. Most of the processing plants and broiler farms (73%, n=41) recently made changes to their establishments to meet the new performance standards for Salmonella (Table 2.5). Of the changes made, implementation of new post-harvest interventions in the processing plant to reduce Salmonella contamination on final products was the most common change made by the

establishment. Nonetheless, most establishments implemented or improved interventions and management strategies at both pre- and post-harvest processes to control Salmonella.

Only one participant responded that he/she did not know of any changes made to the establishment.

Contrary to the efforts taken to control Salmonella, very few establishments made changes to mitigate Campylobacter contamination (Table 2.5). Thirty-four percent

(n=19) responded that their establishment had made changes within the past year to pre- and/or post-harvest processes to meet the Campylobacter performance standards. Similar to Salmonella control, implementation of new post-harvest interventions in the processing plant was the most common change made, followed by implementation of in- plant microbiological testing to check Campylobacter contamination levels during processing. However, very few pre-harvest changes were implemented to reduce

Campylobacter colonization on farm. Additionally, 17% (n=7) responded that they did not know if their establishment made any changes regarding Campylobacter controls.

Knowledge about Campylobacter in broiler chickens

Six true/false questions were designed to elicit knowledge about Campylobacter epidemiology in broiler chickens. Ideas about Campylobacter transmission were often correct, but sometimes misguided. A large fraction of the participants was aware that

Campylobacter is readily transmitted between birds via horizontal transmission (Figure

2.4). Seventy percent (n=39) of the participants answered that Campylobacter can be transmitted between birds via feces. Additionally, seventy-one percent (n=40) answered that Campylobacter can be transmitted to birds from contaminated farm equipment.

Nonetheless, a majority of the processing plant managers (65%) and poultry veterinarians

(50%) responded that Campylobacter can be transmitted through vertical transmission from laying hens to broiler chicks, a belief that is not supported in the scientific literature and is not widely accepted.

Seventy percent (n=39) believed that effective on-farm biosecurity alone cannot eliminate

Campylobacter from broiler farms. A mixed result was observed when participants were asked if they think the incidence of Campylobacter decreases during second processing.

Seventy-two percent (n=13) of the broiler farm managers answered that they do not know, while 65% of the plant managers answered ‘Yes’ to the question. For the poultry veterinarians, the response was equally divided among the three choices (28%, 28% and

33% for ‘True’, ‘False’, and ‘Don’t know’, respectively).

When asked to rank the five on-farm interventions in the order of effectiveness to reduce

Campylobacter colonization in broiler chickens on farm, all three job types answered that the use of antibiotics was the least effective intervention (Figure 2.5). The most effective on-farm intervention or management practice against Campylobacter ranked by broiler farm managers and plant managers was biosecurity. Poultry veterinarians, on the other hand, ranked management of litter as the most effective management practice for

Campylobacter colonization.

Lastly, familiarity with the risk factors associated with increased Campylobacter contamination on-farm and in the processing plant varied markedly among the three job types (Figure 2.6). Notably, the plant managers were more familiar with on-farm risk

factors than the factors in the processing plant. Likewise, more poultry veterinarians answered that they were more familiar with the factors in the processing plant than on farm. In contrast, broiler farm managers were the least familiar with both risk factors among the three job types.

Discussion

This survey was design to assess the current pre- and post-harvest interventions and control strategies implemented at broiler production establishments to reduce

Campylobacter contamination on raw broiler products. Overall, the findings show that both processing plants and broiler farms were following most of the USDA-FSIS recommended guidelines which were published to aid the U.S. broiler industry to meet the Campylobacter and Salmonella performance standards on raw poultry products imposed by the USDA-FSIS. A majority of the processing plants adopted antimicrobial interventions and sanitary practices at multiple processing stages. These interventions included the addition of antimicrobial agents in processing water, use of counter-flow of processing water and multiple carcass washes between processing stages. Likewise, the broiler farms implemented strict biosecurity measures and antimicrobial interventions, including sanitization of drinking water and application of acidifiers to feed and litter to prevent and reduce bacterial growth.

One of the crucial practices emphasized in the USDA-FSIS recommended guidelines is maintaining strict biosecurity on farm, during transportation of birds from farm to slaughter and in the processing plant 186. However, more than half of the broiler farm managers and poultry veterinarians responded that enhanced on-farm biosecurity does not

help to reduce Campylobacter colonization in broiler chickens, while many ranked biosecurity as the most effective on-farm management strategy to tackle Campylobacter.

This discrepancy may have stemmed from the fact that although maintaining strict on- farm biosecurity is considered one of the best on-farm management strategies available to reduce the introduction of Campylobacter in broiler chickens, however, it is not always effective 125,187. Numerous epidemiological studies have identified use of partial depopulation 105,188, lack of proper rodent and pest management 72,97,99, inadequate hygiene barriers 95,97,189 and lack of proper hygiene and/or biosecurity training of farm personnel 190,191 as potential risk factors for the presence of Campylobacter in broiler flocks. Yet, these studies have also pointed out that even implementing the most severe biosecurity controls does not guarantee complete prevention of Campylobacter colonization. In addition, maintaining strict biosecurity is often expensive; the cost and time it takes to implement biosecurity barriers to poultry houses and equipment as well as the training of farm personnel may not be cost-effective 44,192. Despite the lack of

Campylobacter-specific on-farm interventions, it should be emphasized that a combination of enhanced biosecurity and other antimicrobial interventions should be used together to reduce the risk of Campylobacter contamination in broiler chickens. Use of antimicrobial interventions in drinking water and litter, for example, can help prevent further spread of Campylobacter in birds and can complement on-farm biosecurity measures.

It is important to note that a majority of the processing plant managers responded that they do not clean transportation modules, vehicles and/or catching equipment between flocks, even though they believed that doing so can help to reduce cross-contamination of

Campylobacter between flocks. A number of studies has shown that a Campylobacter- free flock can be contaminated with Campylobacter via cross-contamination during transportation by coming in contact with catching and transportation equipment that are not properly cleaned 121,122,193. The bird’s external surface, such as feathers, feet or skin, can be contaminated by Campylobacter when these body parts come in contact with the contaminated surfaces 124. As such, contaminated catching and transportation equipment can significantly increase the chance of cross-contamination happening between and within flocks. Uncleaned, contaminated catching equipment and transportation vehicles and equipment are serious breaches of biosecurity. This is a key area that needs to be reinforced with training of farm personnel and processing plant managers to reduce potential cross-contamination of Campylobacter.

The survey results revealed that there are gaps in the understanding of Campylobacter ecology and epidemiology in broiler chickens among the participants. A majority of the survey participants were aware of the fact that Campylobacter can be transmitted horizontally between birds by feces or from different farm environments. However, many participants answered that Campylobacter can also be transmitted vertically from laying hens (or breeder flocks) to broiler chicks. The possibility of vertical transmission of

Campylobacter in broiler chickens has been investigated extensively 77-79,194. These studies have shown that vertical transmission of Campylobacter is unlikely to be a source of Campylobacter in broiler flocks and conclude that environment as a primary source of

Campylobacter 77-79,194. Additionally, the lack of Campylobacter colonization in the early life of broiler chicks, which is thought to be due to maternal immunity, further supports the unlikeliness of vertical transmission of Campylobacter in broiler flocks 95. To

promote proper control of Campylobacter colonization in broiler chickens, on-farm control efforts should focus on preventing the horizontal and environmental transmission of Campylobacter, rather than on vertical transmission.

A great disparity in the knowledge of risk factors associated with increased colonization and contamination of Campylobacter in broiler chickens and on carcasses was observed among the processing plant managers, broiler farm managers and poultry veterinarians who responded. Processing plant managers self-reported to be the most familiar with both on-farm and processing plant risk factors. Likewise, they were more familiar with the changes made to the USDA-FSIS performance standards for Campylobacter on raw poultry products. This latter result is not surprising, as processing plant managers are constantly informed of their establishments’ performance standards by the USDA-FSIS and are in a position to amend their Hazard Analysis and Critical Control Point (HACCP) plan. As such, they are more familiar with the changes made to the performance standards and also with potential risk factors that may contaminate incoming flocks with

Campylobacter or other foodborne pathogens, such as Salmonella. It is important to emphasize that control of Campylobacter requires an integrated system of pre- and post- harvest efforts which would require producers, broiler farm managers and poultry veterinarians to be knowledgeable about potential risk factors associated with increased presence of Campylobacter on-farm and understand what preventative measures need to be taken to minimize such contamination. Reinforced education of broiler farm managers and poultry veterinarians on important on-farm risk factors associated with increased prevalence of Campylobacter is needed to complement the post-harvest interventions.

While both Salmonella and Campylobacter are important foodborne bacterial pathogens impacting public health, the survey results show that current foodborne pathogen control efforts are more focused on controlling Salmonella than Campylobacter. Establishments were more likely to implement interventions or biosecurity measures specifically targeted to reduce Salmonella than to Campylobacter. Establishments were more likely to make changes to bacterial contamination control strategies within the past year to meet the

Salmonella performance standards. Establishments were also more likely to conduct microbiological testing to detect the presence of Salmonella in flocks than

Campylobacter. There are several possible reasons for the discrepancy observed between the efforts taken to control Salmonella versus Campylobacter. First, FSIS has temporarily suspended Campylobacter performance standards verification testing as of August 2018

195. The agency is testing out a new detection method using enrichment media to resuscitate injured Campylobacter cells, and will be establishing new performance standards based on a new detection scheme 195. While the FSIS will continue to collect samples and share Campylobacter testing results with establishments, establishments are less likely to put additional efforts to control Campylobacter contamination on broiler products. Second, the availability of Salmonella-specific on-farm interventions, particularly Salmonella vaccines, allows establishments to reduce prevalence of

Salmonella in broiler flocks, and subsequently on broiler products. Vaccinating breeders and layers against Salmonella is believed to play an important role in reducing

Salmonella prevalence in broiler flocks 196-198. Several studies have suggested that vaccinating breeders and layers against Salmonella reduces prevalence of Salmonella in broiler flocks by reducing both the horizontal and vertical transmission of Salmonella

197,199-203. Lastly, Salmonella is relatively easier to detect and culture in the laboratory than Campylobacter. Campylobacter’s fastidious growth conditions, coupled with its ability to enter a viable-but-non-culturable (VBNC) state upon exposure to unfavorable growth conditions 204,205, makes it difficult to culture and isolate Campylobacter from complex organic samples such as feces of chicken, broiler litter or carcasses using conventional culture methods 206.

This study was the first qualitative survey of the U.S. broiler industry attempting to assess currently implemented pre- and post-harvest interventions and management strategies targeted to mitigate Campylobacter contamination in broiler chicken products and to gain insights on industry stakeholder’s understanding of Campylobacter in broiler chickens.

The study had some limitations that need to be addressed. First, the study had a relatively small sample size and a low response rate. In addition, the participants were not required to answer all questions in the survey. Because the predicted time to complete the survey was approximately 30 minutes, the number of the questions that were not fully answered by the participants increased towards the end of the survey. Particularly, there was a higher percentage of poultry veterinarians and broiler farm managers who did not complete the survey compared to the processing plant managers. Another limitation of the survey was that some of the key questions regarding Campylobacter biology in broiler chickens were not included in the survey. For example, the survey did not include a question about the average age at which birds are slaughtered; age at slaughter seems to be related to the degree of Campylobacter colonization and shedding in birds 75,207. While the exact mechanisms of such phenomenon is unknown, older chickens tend to be colonized with lower levels of Campylobacter than broiler chickens which are

slaughtered at 6 - 7 weeks of age. It is speculated that maturation of the bird’s immune system as the bird gets older may help to reduce Campylobacter colonization and shedding in birds 208. Establishments that slaughter birds at older ages or grow slow- growing breeds of chickens may not have the same level of Campylobacter contamination. These respondents might also have been more likely to report that the interventions listed in the survey are not effective or even required to reduce

Campylobacter colonization in broiler chickens.

Table 2.1. Descriptions of the processing plants and broiler farms that participated in the survey. Processing plant manager/Food safety or quality assurance team members (n=20) Median number of birds slaughtered per week 727500 (Range: 80000 ~ 60000000) Number of plants practicing primary slaughter 80% (n=16) (2 no responses; two plants practice primary processing only) Number of plants practicing secondary slaughter 80% (n=16) (2 no responses; two plants practice secondary processing only) Number of plants practicing logistic slaughter 5% (n=1) Number of plants conducting microbiological testing to survey Campylobacter and 80% (n=16) (2 no responses) Salmonella on poultry products Number of plants with validated Campylobacter interventions 33% (n=6) Number of plants with validated Salmonella interventions 100% (n=20) Broiler manager / Farm supervisor (n=18) Median number of farms under supervision 120 (Range: 16 ~ 350) Number of companies conducting on-farm microbiological testing for Campylobacter 0% (n=0) Number of companies conducting on-farm microbiological testing for Salmonella 56% (n=10) Number of companies with validated biosecurity management for Campylobacter 11% (n=2) Number of companies with validated biosecurity management for Salmonella 39% (n=7) Poultry veterinarians (n=18) Median number of farms under supervision 500 (Range: 200 ~ 2000) Number of companies conducting on-farm microbiological testing for Campylobacter 11% (n=2) Number of companies conducting on-farm microbiological testing for Salmonella 33% (n=6) Number of companies with validated biosecurity management for Campylobacter 0% (n=0) Number of companies with validated biosecurity management for Salmonella 11% (n=2)

Table 2.2. List of pre-harvest interventions and management practices implemented on-farm to reduce foodborne pathogen colonization in broiler chickens. (n* = total number of responses from broiler farm mangers, n† = total number of responses from poultry veterinarians) Broiler farm managers Poultry veterinarians Drinking water management (n*=15, n†=14) Sanitation using chlorination or hypochlorination 87% (13) 100% (14) Sanitation using oxidizers 73% (11) 93% (13) Administration of acidifier 93% (14) 93% (13) Other 7% (1) 36% (5) Litter management (n*=15, n†=14) Chemical treatment of litter (litter acidifier) 93% (14) 100% (14) Biological treatment of litter (microbiological or enzymatic) 20% (3) 43% (6) Other 7% (1) 21% (3) Feed practice (n*=14, n†=13) Administration of feed acidifier 14% (2) 38% (5) Administration of prebiotics 21% (3) 69% (9) Administration of probiotics 43% (6) 77% (10) Other 0% (0) 15% (2)

Table 2.3. Opinions on the effectiveness of pre- and post-harvest interventions and management practices implemented on farm and in the processing plant to reduce Campylobacter contamination in broiler chickens and carcasses. Processing plant manager Yes No Don’t know In your opinion, does cleaning transportation vehicles, transportation modules 56% (10) 28% (5) 17% (3) and/or catching equipment help to reduce cross-contamination of Campylobacter between flocks? (n=18) In your opinion, are the current scalding practices effective at reducing 56% (10) 28% (5) 17% (3) Campylobacter contamination on carcasses? (n=18) In your opinion, are the current picking practices effective at reducing 59% (10) 24% (4) 18% (3) Campylobacter contamination on carcasses? (n=17) In your opinion, are the current evisceration practices effective at reducing 88% (15) 12% (2) 0% (0) Campylobacter contamination on carcasses? (n=17) In your opinion, are the current chilling practices effective at reducing 88% (15) 6% (1) 6% (1) Campylobacter contamination on carcasses? (n=17) In your opinion, are the current second process practices effective at reducing 82% (14) 12% (2) 6% (1) Campylobacter contamination on carcasses? (n=17) Broiler farm manager Yes No Don’t know In your opinion, does chlorination or hypochlorination reduce Campylobacter 40% (6) 7% (1) 53% (8) colonization in birds? (n=15) In your opinion, does sanitizing drinking water using an oxidizer reduce 40% (6) 0% (0) 60% (9) Campylobacter colonization in birds? (n=15) In your opinion, does administering a water acidifier reduce Campylobacter 27% (4) 7% (1) 67% (10) colonization in birds? (n=15) In your opinion, does chemical treatment of litter reduce Campylobacter 27% (4) 7% (1) 67% (10) colonization in birds? (n=15) In your opinion, does biological treatment of litter reduce Campylobacter 20% (3) 13% (2) 67% (10) colonization in birds? (n=15)

In your opinion, does feed acidifier reduce Campylobacter colonization in birds? 7% (1) 0% (0) 93% (13) (n=14) In your opinion, does administering prebiotics reduce Campylobacter colonization 0% (0) 7% (1) 93% (13) in birds? (n=14) In your opinion, does administering probiotics reduce Campylobacter colonization 0% (0) 7% (1) 93% (13) in birds? (n=14) In your opinion, does improvement in biosecurity of farms reduce Campylobacter 43% (6) 21% (3) 36% (5) colonization in birds? (n=14) Poultry veterinarian Yes No Don’t know In your opinion, does chlorination or hypochlorination reduce Campylobacter 29% (4) 29% (4) 43% (6) colonization in birds? (n=14) In your opinion, does sanitizing drinking water using an oxidizer reduce 21% (3) 21% (3) 57% (8) Campylobacter colonization in birds? (n=14) In your opinion, does administering a water acidifier reduce Campylobacter 36% (5) 14% (2) 50% (7) colonization in birds? (n=14) In your opinion, does chemical treatment of litter reduce Campylobacter 29% (4) 50% (7) 21% (3) colonization in birds? (n=14) In your opinion, does biological treatment of litter reduce Campylobacter 22% (2) 78% (7) 0% (0) colonization in birds? (n=9) In your opinion, does feed acidifier reduce Campylobacter colonization in birds? 15% (2) 15% (2) 69% (9) (n=13) In your opinion, does administering prebiotics reduce Campylobacter colonization 15% (2) 15% (2) 69% (9) in birds? (n=13) In your opinion, does administering probiotics reduce Campylobacter colonization 23% (3) 23% (3) 54% (7) in birds? (n=13) In your opinion, does improvement in biosecurity of farms reduce Campylobacter 31% (4) 46% (6) 23% (3) colonization in birds? (n=13)

Table 2.4. List of post-harvest interventions implemented in the processing plant to minimize microbial contamination on broiler chicken products. Processing plant manager Are transportation modules cleaned between flocks? (n=18) Yes 5.6% (1) No 88.9% (16) Don’t know 5.6 % (1) Immersion scalding method (n=18) Use of pre-scald brush 39% (7) Use of counter flow of process water 89% (16) Use of multiple stage tanks 94% (17) Administration of interventions to scald water (e.g., PAA, salt, adjust pH) 33% (6) Use of post-scald wash 39% (7) Administration of intervention to post-scald wash (e.g., PAA, salt, adjust pH) 50% (9) Steam immersion scalding method (n=2) Use pre-scald brush system 50% (1) Use post-scald wash 50% (1) Administration of intervention to post-scald wash (e.g., PAA, salt, adjust pH) 50% (1) Picking method (n=17) Use of continuous rinsing of carcasses during picking 59% (10) Use of continuous rinsing of equipment during picking 82% (14) Use of post-pick application of chemicals or antimicrobial agents (in wash, dip or other 82% (14) manner) Evisceration (n=17) Use of automated re-hang 94% (16) Use of daily replacement of opener blades 65% (11) Use of whole carcass rinse during evisceration 94% (16) Use of post-evisceration wash 100% (17) Application of chemicals or antimicrobials in post-evisceration wash 100% (17)

Chilling method – Immersion chilling (n=17) Use of pre-chill tanks 94% (16) Use of antimicrobial treatment in pre-chill tanks 94% (16) Use of counter-current flow during chilling 100% (17) Addition of chlorine in chill water 12% (2) Use of other antimicrobial agents in chill water 100% (17) Use of oxidation reduction potential meter (ORP) 12% (2) Chilling method – Air chilling (n=1) Use of pre-chill stage 100% (1) Use of antimicrobial treatment in pre-chill stage 100%(1) Addition of chlorine in chill air 100% (1) Regular maintenance of chains and shackles 100% (1) Second process (n=17) Application of chemicals or antimicrobial agents (in wash, dip or other manner) 76% (13) Belt sanitation 94% (16)

Table 2.5. New recommendations or changes that have been implemented within the past year to meet the USDA-FSIS performance standards for Campylobacter and Salmonella on raw poultry products by the U.S. broiler industry. Processing Broiler farm Poultry plant managers managers veterinarians (n=20) (n=18) (n=18) Have any changes been made specifically for Campylobacter contamination mitigation strategies? Yes 55% (11) 27.8%( 5) 16.7% (3) Implementation of new on-farm interventions (vaccines, 5% (1) 14.3% (2) 8.3% (1) chemicals, probiotics, prebiotics or antibiotics) Implementation or improvement of biosecurity and other 5% (1) 35.7% (5) 16.7% (2) management strategies Implementation of on-farm microbiological testing 0% (0) 14.3% (2) 16.7% (2) Implementation of new post-harvest interventions in the 50% (10) 21.4% (3) 25% (3) processing plant Implementation of in-plant microbiological testing 30% (6) 14.3% (2) 25% (3) Others 10% (2) 0 8.3% (1) No 40% (8) 35% (7) 83.3% (15) I don’t know if any changes have been made 5% (1) 30% (6) 0% (0) Have any changes been made specifically for Salmonella contamination mitigation strategies? Yes 70% (14) 88.9% (16) 61.1% (11) Implementation of new on-farm interventions (vaccine, chemicals, 17.6% (6) 29.5% (13) 25.6% (11) probiotics, prebiotics or antibiotics) Implementation or improvement of biosecurity and other 11.8% (4) 25% (11) 18.6% (8) management strategies Implementation of on-farm microbiological testing 5.9% (2) 22.7% (10) 14% (6) Implementation of new post-harvest intervention in the processing 35.2% (12) 15.9% (7) 20.9% (9) plant Implementation of in-plant microbiological testing 20.6% (7) 6.8% (3) 16.3% (7)

Others 8.8% (3) 0% (0) 4.7% (2) No 25% (5) 11.1% (2) 38.9% (7) I don’t know if any changes have been made 5% (1) 0% (0) 0% (0)

Figure 2.1. The approximate incidence of Campylobacter and Salmonella from internal microbiological testing using whole carcasses at processing plants.

Figure 2.2. The number of broiler farms that are regularly tested positive for Campylobacter and Salmonella in the processing plant.

Figure 2.3. Scores from a 5-point Likert scale showing the survey participants’ understanding of the changes made to the United States Department of Agriculture– Food Safety and Inspection Service performance standards for Campylobacter and Salmonella on raw poultry products.

Figure 2.4. Comparison of understanding of on-farm ecology and epidemiology of Campylobacter in broiler chickens.

Figure 2.5. Ranking of on-farm management strategies targeted to reduce Campylobacter colonization in broiler chickens based on their effectiveness.

Figure 2.6. Scores from a 5-point Likert scale showing the survey participants’ familiarity with the risk factors associated with the increased Campylobacter contamination in broiler chickens and on carcasses.

Chapter Three: Investigation of indirect selective pressures conferring fitness advantages in fluoroquinolone-resistant Campylobacter jejuni in vitro

Summary

Campylobacteriosis caused by C. jejuni is a serious, yet common foodborne disease in the U.S. The prevalence of fluoroquinolone-resistant C. jejuni from poultry has continued to increase despite the withdrawal of fluoroquinolone use in the U.S. poultry industry since 2005. To date no clear selective pressures that could explain this effect have been documented. In this study we investigated two factors, limited bioavailability of iron in poultry and activity of microcin B17 produced by E. coli commensal in poultry as indirect selective pressures conferring fitness advantages in fluoroquinolone-resistant C. jejuni compared to its susceptible wild-type counterpart. Five fluoroquinolone- susceptible C. jejuni isolates were selected from litter collected from commercial broiler farms. Using antibiotic selection, five fluoroquinolone-resistant strains were created.

Relative expression of six genes involved in iron acquisition and regulation was compared between the resistant and susceptible strains using quantitative PCR under normal and iron-limiting conditions. High variability in the relative gene expression was observed among the strains, with only one resistant strain showing the consistent upregulation of the measured genes compared to the susceptible wild-type strain.

Application of cell lysates containing active microcin B17 to C. jejuni culture showed that microcin B17 inhibits growth of C. jejuni regardless of the fluoroquinolone resistance phenotype. Our results suggest that two hypotheses tested in this study are not adequate explanations of the molecular mechanisms behind the enhanced fitness of FQ-R

C. jejuni compared to FQ-S C. jejuni. This study highlights the needs for better insight into complex situation of fluoroquinolone resistant C. jejuni ecology in poultry environment.

Introduction

Campylobacter jejuni (C. jejuni) is the most common cause of campylobacteriosis globally and is generally considered a commensal organism of poultry 66. Chicken products are a major source of C. jejuni in humans 170,209. Risk factors for campylobacteriosis include consumption of undercooked chicken, handling of poultry products and cross-contamination of other foods, especially those consumed raw 210.

Acute campylobacteriosis in humans is generally self-limiting, and antibiotic treatment is rarely indicated. However, at-risk subpopulations, such as the elderly, children and the immunocompromised, may require antibiotic treatment for campylobacteriosis 51. While macrolides are a common choice of antibiotic for treating campylobacteriosis, especially in children, fluoroquinolone (FQ) antimicrobials are considered the primary choice because of their efficacy and broad-spectrum of activity in adult patients 50,52.

Enrofloxacin, a FQ antibiotic approved in veterinary medicine, was used in the U.S. poultry industry for the treatment and control of disease related to Escherichia coli (E. coli) and Pasterella multocida infections 211. However, due to concerns over the development of FQ-resistant (FQ-R) C. jejuni, the US Food and Drug Administration issued the withdrawal of enrofloxacin from the use in U.S. poultry production in 2005 212.

Following the withdrawal, the prevalence of FQ-R Campylobacter in poultry, poultry products and human infections has continued to rise in the absence of the direct selective

pressure in poultry as evidenced by National Antimicrobial Resistance Monitoring

System reports 213. In fact, the increase in FQ resistance in human C. jejuni infections has continued at the same rate as it did prior to the withdrawal. Reasons for this stability of

FQ-R C. jejuni subpopulations in chickens remain elusive. While the molecular basis of

FQ resistance in Campylobacter has been well studied, the ecological drivers of FQ resistance dynamics within Campylobacter subpopulations are less well understood.

Some evidence suggests that FQ resistance confers a fitness advantage to Campylobacter possessing this trait, even in the absence of selective pressures from FQs 156. However, this result has not been uniformly replicated in all settings 214, indicating that this effect may be strain- or context-dependent.

FQ resistance in C. jejuni most frequently derives from a point mutation in the DNA gyrase A gene, gyrA, with a substitution of isoleucine for threonine at position 86 162,164.

This point mutation has been shown to reduce negative supercoiling of DNA, which alters DNA topology during transcription, replication and recombination 157. A recent study that compared transcriptomic expressions between isogenically-paired FQ-R and fluoroquinolone-susceptible (FQ-S) C. jejuni showed differential gene expressions in vitro 215. Specifically, genes involved in iron uptake and regulation were up-regulated in

FQ-R compared to its isogenic wild-type strain 215. These results suggest that reduced negative supercoiling due to the acquisition of FQ resistance may enhance iron-uptake and regulation systems in C. jejuni in the absence of FQ, allowing FQ-R subpopulations to persist in the broiler environment.

Similar to the iron effect, microcin B17 (mccB17), an antibacterial peptide produced by poultry commensal E. coli may exert differential selective pressures on FQ-R and FQ-S

C. jejuni. The mode of inhibition of mccB17 is speculated to be very similar to that of

FQs. MccB17 is known to interact with DNA gyrase A and stabilizes a cleavage complex formed between the enzyme and double-stranded DNA during DNA replication, resulting in chromosome fragmentation which leads to cell death 216. While mccB17 has been shown to be active against several enterobacteria, including E. coli, Salmonella spp. and

Klebsiella spp., its effect on Campylobacter spp. has never been demonstrated 217,218.

Additionally, quinolone resistant E. coli is resistant to the activity of mccB17 219, suggesting that FQ resistance in C. jejuni may confer protection against the activity of mccB17, thereby enhancing the fitness of FQ-R C. jejuni in the broiler environment.

The purpose of this study was to investigate two putative indirect selective pressures conferring fitness advantages to FQ-R C. jejuni compared to its wild-type C. jejuni in vitro, and improve our understanding of the molecular mechanisms that enable the persistence of FQ-R C. jejuni in broiler chickens despite the lack of direct selective pressure. The current study aimed to test two hypotheses: 1) the FQ resistance conferred by a gyrA point mutation enhances iron acquisition and regulation in FQ-R C. jejuni, increasing its fitness as compared to its isogenic wild-type C. jejuni in vitro, and 2) the

FQ resistance conferred by a gyrA point mutation inhibits the activity of mccB17 produced by poultry commensal E. coli, allowing survival of FQ-R C. jejuni as compared to its wild-type C. jejuni in vitro.

Materials and Methods Bacterial Strains and Culture

Five FQ-S C. jejuni strains were isolated from litter samples collected from five different broiler farms located in the U.S in 2014. Bolton enrichment broth (Oxoid Ltd.,

Basingstoke, England) supplemented with Oxoid Preston Campylobacter selective supplement (2,500 IU polymyxin B, 1µg/mL rifampicin, 1 µg/mL trimethoprim, 100

µg/mL cycloheximide and 5% lysed horse blood) and Campy Cefex agar (16.5 mg/mL cefoperazone, 0.1 g/mL cycloheximide, 5% lysed horse blood) were used to isolate

Campylobacter-like colonies. Multiplex PCR using a lipid A gene, lpxA, was performed to confirm and speciate the suspected Campylobacter isolates 220. To confirm the susceptibility to ciprofloxacin, broth microdilution susceptibility test was performed using the Sensititre® (Trek Diagnostics Systems, Cleveland, OH) according to the manufacturer’s instructions. Briefly, several colonies of each isolate grown from blood agar were suspended in 5 mL of Sensititre® cation adjusted Mueller-Hinton broth with

TES buffer (CAMHBT). The suspension was adjusted to 0.5 McFarland turbidity standard using a Sensititre® nephelometer. Hundred µL of the adjusted suspension was transferred to 11 mL Sensititre cation adjusted Mueller-Hinton broth with TES buffer and lysed horse blood (CAMHBT + LHB). Each well of a Sensititre® Campylobacter

CAMPY AST plate was inoculated with 100 µL of the new suspension using the

Sensititre Trek AutoInoculator®. C. jejuni ATCC 33560 was used as a quality control strain. The plate was incubated at 42ºC for 24hrs under microaerophilic conditions (85%

N2, 10% CO2 and 5% O2). Following incubation, the plate was read by the Sensititre

Vizion® Digital MIC viewing system. According to the Clinical & Laboratory Standards

Institute (CLSI) guidelines, strains with a MIC value below or equal to 1 µg/mL was considered susceptible to ciprofloxacin.

To create complementary FQ-R C. jejuni strains, the five selected susceptible strains underwent ciprofloxacin selection on Mueller-Hinton agar. FQ-S C. jejuni were grown on

Mueller-Hinton agar supplemented with a low concentration of ciprofloxacin (0.5

µg/mL) and subsequently passaged to reach a minimum ciprofloxacin MIC value of 16

µg/mL. To confirm complementarity between FQ-R and FQ-S C. jejuni strains, 1) gyrA genes were fully sequenced using Sanger sequencing to confirm the point mutation (Thr-

86 → Ile), and 2) the growth kinetics of FQ-R and FQ-S C. jejuni strains were measured and compared using growth curves assays, as described below.

Sanger sequencing of gyrA and gyrB genes

The genomic DNA of each FQ-R and FQ-S C. jejuni strain was extracted using a

PureLink Genomic DNA Mini Kit (Invitrogen, Carlsbad, CA) following the manufacturer’s instructions. DNA was eluted in 50 µL nuclease-free water, and its quality and quantity were analyzed using the NanoDrop. The gyrA gene was PCR- amplified before submission for Sanger sequencing. The PCR master mix was prepared using a Q5 High-Fidelity Master Mix kit (New England Biolabs, Ipswich, MA) to a final volume of 50µL (25 µL of Q5 High-Fidelity 2X Master mix, 10 µM of GyrA-1 forward and GyrA-3 reverse primers and 5 µL genomic DNA template). The thermocycler conditions used were initial denaturation at 98ºC for 30 seconds followed by 98ºC for 10 seconds, 63ºC for 30 seconds, 72ºC for 90 seconds with a final extension at 72ºC for 2 minutes for 25 cycles. Additionally, gyrB was sequenced to check if the acquisition of FQ resistance resulted in additional mutations in the gene. The PCR mastermix was prepared to amplify the gyrB gene using gyrB-1 forward and gyrB-2 reverse primers. The thermocycler conditions used to amplify gyrB gene were initial denaturation at 98ºC for

30 seconds, followed by 98ºC for 10 seconds, 62ºC for 30 seconds, 72ºC for 90 seconds with a final extension at 72ºC for 2 minutes for 25 cycles. The PCR amplicons were

visualized on 1.0 % agarose gel under UV light after staining with ethidium bromide. The expected PCR amplicon sizes were 2,755bp and 2,811bp for gyrA and gyrB, respectively.

The primers used for gyrA and gyrB amplification are listed in Table 3.1.

The PCR amplicons were purified using a QIAquick PCR purification kit (QIAGEN,

Hilden, Germany) according to the manufacturer’s instructions, and DNA was eluted in

25 µL of nuclease-free water. To sequence the gyrA and gyrB gene, 1µL of each of the gyrA-1, 2, 3 and gyrB-1, 2, 3 forward and reverse primers at 3.2 µM concentration was mixed with 5 µL of the purified DNA, and submitted for Sanger sequencing on the

Applied Biosystems 3730xl DNA Analyzer at the University of Minnesota Genomic

Center. The sequence of gyrA of C. jejuni NCTC11168 was used as a reference to check the point mutation responsible for FQ resistance in the FQ-R strains. The chromatogram of the aligned sequences was visualized in MEGA ver. 6.0 221. Additionally, gyrA and gyrB sequences were compared between FQ-R and FQ-S strains to check for any additional mutations caused by the acquisition of FQ resistance.

Growth curve experiments

Growth curve experiments were conducted to compare fitness kinetics of FQ-R and FQ-S

C. jejuni strains. Each C. jejuni strain was grown in Mueller-Hinton broth under microaerophilic conditions at 42°C for 24hr. The 24-hr culture was transferred to 10 mL of fresh Mueller-Hinton broth with an optical density (OD) at 600 nm of 0.02 and incubated under microaerophilic conditions at 42°C without shaking. During incubation,

OD was measured by taking 40 µL of the culture at 0, 6, 24, 30 and 48hr. An additional

100 µL of the culture was serially diluted using Mueller-Hinton broth and plated at 10-6,

10-8, 10-12 dilution triplicates. The plates were incubated under microaerophilic conditions

at 42°C for 48hr. The number of colonies was counted for each dilution to calculate

CFU/mL. The growth curve experiment was conducted for each strain with three biological replicates and three technical replicates. To statistically compare the growth kinetics between FQ-R and FQ-S C. jejuni, pairwise permutation tests (n=10,000) implemented in the statmod R package were used to compare the overall OD600 measurements between strains 222,223.

RNA Extraction

Approximately 107 CFU of FQ-R and FQ-S C. jejuni strains were grown in Mueller-

Hinton broth in microaerophilic conditions at 42°C for overnight. The overnight culture was transferred to fresh Mueller-Hinton broth at an OD of 0.05 and incubated under microaerophilic conditions at 42°C for 24hr. Following incubation, 2, 2’- dipyridyl, a high-affinity iron chelator dissolved in 50% molecular-grade ethanol and 50% nuclease- free water, was added to each Campylobacter culture at 0 and 100 µM. At 0, 3, 5, 10, 30 and 60 minutes post-addition of 2, 2’-dipyridyl, 1 mL of the culture was collected and centrifuged (13,000 rpm for 5 minutes at room temperature) to collect a cell pellet. The cell pellet was suspended in 750 µL of TRIzol (Invitrogen, Carlsbad, CA) and stored at -

80°C until RNA extraction. RNA was extracted following the TRIzol manufacturer’s instructions. Total RNA was eluted in DEPC-treated water in a final volume of 50 µL.

Eluted RNA was treated with RNase-free DNase I (New England Biolabs, Ipswich, MA) following the manufacturer’s instructions to remove DNA contamination. The quality and concentration of the DNase-treated RNA was measured using the NanoDrop. When

RNA concentration or quality was poor (concentration below 10 ng/µL or 260/280 ratio below 1.3), RNA was re-precipitated using isopropanol and 1 µL of GlycoBlue

(Invitrogen, Carlsbad, CA). RNAs were standardized to 10 ng/µL in DEPC-treated water.

The standardized RNA was reverse transcribed to cDNA using a high-capacity cDNA reverse transcription kit with RNase inhibitor (Applied Biosystems, Foster City, CA) following the manufacturer’s instructions.

Quantitative real-time polymerase chain reaction

Quantitative real-time polymerase chain reaction (RT-qPCR) was performed to measure relative expression levels of six genes involved in iron acquisition and regulation. These six genes were chosen from a previous microarray study, as they were shown to be upregulated in FQ-R C. jejuni compared to the susceptible strain when iron concentration in the growth media was not altered 215 (Table 3.2). RNA polymerase subunit A gene, rpoA, was selected as a reference gene for normalization 224. The RT-qPCR master mix was prepared using a PerfeCTa SYBR Green SuperMix with low ROX reagent kit

(QuantaBio, Beverly, MA) to a final volume of 20 µL (10 uL PerfeCTa SYBR Green

SuperMix low ROX (2x), 300 nM of forward and reverse primers each and 3 µL reverse- transcribed cDNA template). Strategene Mx3005P was used to run RT-qPCR with an initial denaturation at 95°C for 3 minutes, 40 PCR cycling at 95°C for 15 seconds and

56°C for 30 seconds, and a final extension at 95°C for 3 minutes followed by melting curve analysis. PCR reactions were run in three biological replicates of each strain with technical duplicates. Each reaction plate included baseline samples (time point zero) to measure the relative changes in gene expression over time within each strain. In addition, each plate included a negative control and a no-transcriptase control. Primers used in the

RT-qPCR are listed in Table 3.1.

To determine PCR efficiency of a primer, a standard curve was generated for each primer. Genomic DNA was extracted from two FQ-R and FQ-S C. jejuni strains using a

PureLink Genomic DNA Mini Kit (Invitrogen, Carlsbad, CA) following the manufacturer’s instructions. Total DNA was eluted in 50 µL of nuclease-free water, and its quality and concentration were measured using the NanoDrop. Each DNA aliquot was

2-fold diluted in nuclease-free water. The RT-qPCR mastermix and thermocycler conditions used to generate standard curves were the same as the above. The standard curves were run in three biological replicates of CJ-A and CJ-D FQ-R and FQ-S strains with technical triplicates. Each reaction plate included negative and no-transcriptase controls.

Production of mccB17 and its effect on the survival of FQ-R and FQ-S C. jejuni

An E. coli DH5α strain containing a pUC19 plasmid with a mcb operon responsible for production and transportation of mccB17 was provided by Dr. Tony Maxwell (Jones

Innes Center, Norwich, UK). The strain was cultured overnight in Luria-Bertani (LB) broth supplemented with 100 µg/mL ampicillin at 37°C with shaking. One mL of the overnight culture was transferred to M63 minimal media (2 µg/µL (NH4)2SO4, 13.6

µg/µL KH2PO4, 0.5 µg/mL FeSO4·7H2O, 1 mM MgSO4, 1 µg/mL thiamine, 0.2% glucose and 100 µg/mL ampicillin) and cultured at 37°C for 40hr with shaking. After incubation, cell pellets were collected by centrifugation (4300 rpm for 10 minutes at room temperature). The cell pellets were suspended in 500 uL of 100 mM acetic acid and

1 mM EDTA and boiled for 10 minutes. The boiled mixture was centrifuged at 4300 rpm for 10 minutes at room temperature and filtered through a 0.2 mM membrane filter to

collect cell lysates containing active mccB17. Cell lysates from wild-type E. coli DH5α were extracted using the above method and used as a negative control.

To test if the extracted cell lysate contained active mccB17, a spot-plating assay was performed on wild-type E. coli DH5α. A top-agar layer was prepared by spreading 3 mL of LB agar containing 0.7 % agarose and 100 µL of mid-log phase wild-type E. coli

DH5α culture on top of LB agar (1.7 % agarose). After the top-agar layer was solidified,

5 µL of the filtered cell lysate extracted from mccB17 producing E. coli DH5α was applied. In addition, to detect the concentration at which the growth inhibition effect of mccB17 was prevented, 1 mL of the cell lysates was serially diluted two-fold in LB broth, and 5 µL of each dilution was spot-plated on the top-agar. Spot-plated agar plates were incubated at 37°C overnight. After incubation, any zone of growth inhibition was recorded.

To confirm that the growth inhibition observed in the wild-type E. coli DH5α was due to the activity of mccB17, a similar spot-plating assay was performed using ciprofloxacin resistant E. coli DH5α. FQ-S E. coli DH5α was passaged through various concentrations

(0.2 - 4 µg/mL) of ciprofloxacin to create resistant strains. A top-agar layer containing

FQ-R E. coli DH5α (ciprofloxacin MIC ≥ 4 µg/mL) was created. Five µL of the filtered cell lysates from mccB17 producing E. coli DH5α was applied. Additionally, 5 µL of 2- fold serially diluted cell lysates was applied onto the top agar layer. All plates were incubated at 37ºC overnight.

Similar spot-plating assays were performed on FQ-R and FQ-S C. jejuni to measure the effect of mccB17 on the growth of C. jejuni strains. A top-agar layer containing 0.7 %

agarose and 200 µL of mid-log phase FQ-R or FQ-S C. jejuni culture was created on top of Mueller-Hinton agar containing 1.7 % agarose. After the top-agar layer was solidified,

5 µL of cell lysates extracted from wild-type and mccB17 producing E. coli DH5α were spot-plated. To determine the relative concentration at which the growth inhibition effects of mccB17 was prevented, 1 mL of the cell lysates containing active mccB17 was 2-fold serially diluted in Mueller-Hinton broth. Five µL of each dilution was spot-plated on the top-agar layer. Spot-plated plates were incubated for 24hr at 42ºC under microaerophilic conditions. After incubation, any zone of inhibition was recorded.

To measure the prevalence of E. coli containing a mcb operon in the poultry environment, a total of 391 E. coli isolates obtained from broiler litter samples was screened. These isolates were first grown on MacConkey II (Becton Dickinson GmbH, Heidelberg,

Germany) agar to confirm as E. coli. Each isolate was checked for the presence of a 5.1kb mcb operon using PCR amplification. Briefly, genomic DNA of each isolate was extracted using 5 mg/mL Proteinase K and 1x TE buffer. The PCR master mix was prepared using a

Q5 High-fidelity 2X Master Mix kit (New England Biolabs, Ipswich, MA) in a final volume of 25 µL (12.5 µL Q5 master mix, 100 mM of forward and reverse primers and 3

µL of DNA template) (Table 3.1). The cycling conditions used were 98°C for 30 seconds,

98°C for 10 seconds, 56°C for 30 seconds, and 72°C for 90 seconds with a final extension time of 2 minutes at 72°C (30 cycles total). Each PCR reaction was visualized on 1.0 % agarose gel under UV light after staining with ethidium bromide.

Relative quantification and statistical analysis

The relative expression ratio between FQ-R and FQ-S C. jejuni was calculated using the

Pfaffl ΔΔCt method (Equation 1) by normalizing expression levels of target genes to that

of the rpoA reference gene 225. To compare the expression levels of the measured genes over time, the Ct values at baseline (time point = 0) were considered the untreated samples, while the Ct values of other time points (time point = 3, 5, 10, 30 and 60min) were considered treated samples. To compare the expression levels between FQ-R and

FQ-S C. jejuni, the Ct values of FQ-R C. jejuni were considered treated samples and FQ-

S C. jejuni as untreated, control samples. The Pfaffl method does not assume equal PCR amplification efficiency across all samples. Thus, the PCR efficiency of reference and target gene primers was assessed individually by generating standard curves. REST 2009 developed by Pfaffl et al. was used to calculate an expression ratio assuming unequal

PCR efficiencies for each primer set, as well as to run a pairwise reallocation randomization statistical test to compare the mean Ct values between control and treated samples using 10,000 randomization 226. Data visualization was performed in RStudio

(ver. 1.1.419)182 using R version 3.5.0227 and R package ggplot2 (ver. 3.1.0) 185.

Equation 1:

∆Ct(target) = 퐶푡(푡푎푟푔푒푡 푔푒푛푒 𝑖푛 푐표푛푡푟표푙) − 퐶푡(푡푎푟푔푒푡 푔푒푛푒 𝑖푛 푡푟푒푎푡푒푑 푠푎푚푝푙푒)

∆퐶푡(푟푒푓) = 퐶푡(푟푒푓 푔푒푛푒 𝑖푛 푐표푛푡푟표푙) − 퐶푡(푟푒푓 푔푒푛푒 𝑖푛 푡푟푒푎푡푒푑 푠푎푚푝푙푒)

퐸(푡푎푟푔푒푡)∆퐶푡(푡푎푟푔푒푡) 푁표푟푚푎푙𝑖푧푒푑 푟푒푙푎푡𝑖푣푒 푒푥푝푟푒푠푠𝑖표푛 푟푎푡𝑖표 = 퐸(푟푒푓)∆퐶푡(푟푒푓)

푤ℎ푒푟푒 퐸 = 푃퐶푅 푎푚푝푙𝑖푓𝑖푐푎푡𝑖표푛 푒푓푓𝑖푐𝑖푒푛푐푦 푎푛푑

1 − −1 E = 10 slope of standard curve

Results

Five C. jejuni strains isolated from broiler farms and confirmed to be ciprofloxacin susceptible (MIC ≤ 1 µg/mL) underwent antibiotic selection to create matching FQ resistant strains. These FQ-R strains had ciprofloxacin MICs ranging from 16 – 64

µg/mL (Table 3.3). Comparison of the gyrA gene by Sanger sequencing showed that all

FQ-R strains had a nonsynonymous point mutation at position 86 which changed threonine to isoleucine, the most commonly reported point mutation responsible for FQ resistance in C. jejuni. Examination of the gyrB gene showed that two of the FQ-R strains

(CJ-D-R and CJ-E-R) had an additional nonsynonymous point mutation at position 460 which changed serine to phenylalanine. Growth curve experiments were conducted to compare growth characteristics of FQ-R and FQ-S C. jejuni strains to confirm that the acquisition of FQ resistance did not affect growth kinetics of FQ-R strains compared to the wild-type strains. A pairwise permutation (n=10,000) test was run to compare the difference in OD600 readings over time. There was no significant difference in growth kinetics between each FQ-R strain and its matching FQ-S strain (Table 3.4).

Analysis of differential expression of iron acquisition and regulation genes between FQ-

R and FQ-S C. jejuni

RT-qPCR was performed to measure the expressions of the six genes involved in iron acquisition and regulation under normal and iron-limiting conditions over time, and how these expressions differ between FQ-R and FQ-S C. jejuni. Figure 3.1 shows the changes in the expression ratios of the six genes within each strain over time relative to its baseline (time point = 0) in normal and iron-limiting conditions. Overall, the expression of cj0179 was significantly upregulated (p-value < 0.05) over time, with the highest

expression ratio observed in CJ-B-R strain at 60min post-addition of 2, 2’-dipyridyl. The expression of cj1664 and cj1661 were consistently upregulated as well, especially in the later time points. However, the expression of cj0426 and cj0427 showed no statistical differences in the gene expression compared to the baseline level. Notably, similar patterns of expression could be observed within pairs. For instance, CJ-C-S and CJ-C-R showed upregulations of the expression of cj0179 at 30min followed by a sudden drop in the expression at 60min in iron-limiting conditions.

The expression ratios of the six genes in FQ-R C. jejuni relative to its matching wild-type strain are presented in Figure 3.2. Overall, no general trends could be observed, and the direction of differential expression was not in agreement for all six genes across the time points. Instead, great variability in the expression levels was observed among strains.

Particularly, CJ-A-R and CJ-D-R strains showed consistent upregulation of the genes relative to their wild-type under the normal and iron-limiting conditions. The upregulation of cj0179 observed in CJ-D-R was statistically significant (p-value < 0.05) compared to the wild-type in iron-limiting conditions. CJ-B-R, however, showed consistent downregulation of the genes relative to the wild-type at each time point.

Statistically significant downregulations of cj0179 and cj0334 were observed in the early time points in CJ-B-R compared to the susceptible strain under the normal condition. No difference in the expression levels of the genes was observed in CJ-B-R and CJ-E-R relative to their matching wild-type strains over time.

Effects of mccB17 on the growth of FQ-R and FQ-S C. jejuni

The activity of mccB17 was confirmed by conducting a spot-plating assay of E. coli

DH5α strain with and without pUC19 plasmid containing a mcb operon. The assay results

show that the growth of wild-type E. coli DH5α was inhibited by the cell lysates obtained from the mccB17 producing strain while the growth of the mccB17 producing strain was not affected (Figure 3.3A and B). As previous studies have shown, the growth of ciprofloxacin resistant E. coli DH5α (MIC ≥ 4 µg/mL) was not inhibited by mccB17

(Figure 3.3C). When the cell lysate was two-fold diluted in LB broth and applied on the lawn of wild-type E. coli DH5α strain that was susceptible to ciprofloxacin, growth was completely inhibited through the 1:16 dilution.

A similar spot-plating assay was performed to test the effect of mccB17 on the growth of

FQ-R and FQ-S C. jejuni in vitro. The assay results showed that all Campylobacter strains tested were susceptible to the activity of mccB17 regardless of fluoroquinolone resistance phenotype (Figure 3.4). Interestingly, when the mccB17 cell lysate was two- fold diluted in Mueller-Hinton broth and applied to the top agar layer containing C. jejuni strains, complete inhibition of growth was only observed at the original, undiluted concentration.

To estimate the prevalence of E. coli harboring a mcb operon and potentially capable of producing mccB17 in the poultry environment, a total of 391 E. coli isolates obtained from broiler litter was screened for the presence of 5.1kb mcb operon. Of all the isolates screened, one isolate was confirmed to harbor the operon confirmed by PCR amplification. However, when the lysate extracted from this field isolate was applied to the top agar layer of wild-type E. coli DH5α, no zone of inhibition was observed.

Discussion

In the present study, we investigated putative molecular mechanisms behind the persistence of FQ resistant C. jejuni in the poultry environment in the absence of direct selective pressure. The putative molecular indirect selective pressures investigated in this study were: 1) limited bioavailability of iron in the poultry intestine and 2) activity of mccb17 produced by poultry commensal E. coli. Based on the results of this study, the enhanced fitness of certain FQ-R C. jejuni strains did not appear to be attributed to increased expression of iron acquisition and regulation genes or presence and activity of mccB17 produced from poultry commensal E. coli.

Iron is an important micronutrient for bacteria. The bioavailability of iron is often limited in vivo, especially in avian hosts, as iron is tightly bound in tissue or by iron-binding proteins and is, thus inaccessible to bacteria 228. The low bioavailability of iron in the environment is considered a cue for enteric bacteria to express virulence genes 229.

Because the acquisition of iron and regulation of intracellular iron to prevent the generation of oxidative stress are essential for bacterial survival, increased expression of these process was hypothesized to be one of the putative molecular mechanisms behind the enhanced fitness of FQ-R C. jejuni in poultry. The RT-qPCR results show that one out of the five FQ-R strains (CJ-D-R) showed consistent upregulation of the gene expression over multiple time points. The variability in relative gene expression, especially when comparing the isogenic FQ-S and FQ-R strains potentially, reflects the high genetic variability observed among Campylobacter strains. The Campylobacter genome is known to exhibit high genetic variation due to frequent DNA recombination

230-232 and a high number of homopolymeric DNA repeats that are subject to frequent mutations 233,234. As suggested by Wilson et al. 235, intraspecific genetic diversity may

translate into high phenotypic variation which may give advantages to C. jejuni by enhancing its ability to colonize multiple species. Because the strains used in the present study were isolated from poultry farms, it can be speculated that the different genetic backgrounds of the strains used in the study may have resulted in varied patterns of the gene expressions. Previous studies seem to suggest that FQ resistance, depending on the strain’s genetic background, could either impose or remove a fitness cost in C. jejuni. A study by Luo et al. 156 showed that S3B FQ-R C. jejuni strain isolated from chicken outcompeted its isogenic wild-type in vivo, while the other resistant strain of a different genetic background was instead outcompeted by its wild-type. Similarly, a study by

Zeitouni et al. 214 showed that FQ-R C. jejuni was outcompeted by its isogenic wild-type in vivo. Currently, it is unclear whether the increased expression of the genes related to iron acquisition and regulation, as seen in CJ-D-R, leads to enhanced fitness in vitro and in vivo. Future studies comparing growth of CJ-D-R and CJ-D-S in vitro under iron- limiting conditions and also in vivo to assess if the differential gene expression leads to differential growth kinetics are needed.

When gene expression was compared within strain at each time point relative to the baseline, the expression of cj0179 was generally upregulated, especially in iron-limiting conditions. The cj0179 gene transcribes exbB1, a component of an energy transduction system that is involved in iron transport 229. The upregulation of this gene in iron-limiting conditions was previously reported in a study that compared relative transcription levels in iron-rich and iron-limiting growth conditions using C. jejuni NCTC 11168 228.

Likewise, the expression of cj1661, a gene that encodes a ferric-binding component of a putative ABC transport system 228, was upregulated over time compared to baseline in

most of the strains. It was expected that C. jejuni would increase transcription of genes involved with iron acquisition when iron concentration was low. Increased uptake of iron, in turn, increases the concentration of intracellular iron, causing oxidative stress.

Therefore, the transcription of genes involved in oxidative stress regulation, such as cj1664 and cj0334, were expected to be upregulated. Only the expression of cj1664, which encodes thioredoxin was upregulated in the later time points. It is unclear why upregulation of cj0334 was not observed in the present study. With cj1664 and cj0334 regulated by two different regulators, Fur and PerR, respectively, it may require a longer time to see the upregulation of cj0334, as shown by Holmes et al. 228, where RNA was extracted in iron-rich and iron-limiting conditions after mid-late exponential phase was reached. The expression of cj0427 was previously shown to be upregulated in iron- limiting conditions 228. In the present study, the expression levels of cj0427 and cj0426, a gene thought to be co-transcribed with cj0427 to form a putative ABC transport system involved in iron transportation 215, did not change over time. It is possible that these two genes are not involved with iron acquisition in Campylobacter.

The current study investigated relative expression of six genes involved in iron acquisition and oxidative stress regulation in Campylobacter. Iron acquisition and oxidative stress regulation in Campylobacter has been studied extensively, and these studies report a large number of genes that are known to or putatively related to regulation of iron uptake 229, oxidative stress response (perR, sodB, katA, cj0169, etc.) 236-

238 and outer membrane iron-transportation proteins 229, as well as proteins required to generate energy for iron transport 229,239. It is possible that genes conferring the fitness advantage to the FQ-R C. jejuni subpopulation were not assessed in this study. It is also

plausible that the specific genes proving a fitness advantage to the FQ-R C. jejuni subpopulation are strain-specific and vary across the C. jejuni population. Additional work comparing transcriptional profiles between FQ-R and FQ-S C. jejuni from multiple sources using RNAseq would be helpful for revealing additional genes and molecular mechanisms that may be involved with the enhanced fitness of FQ-R C. jejuni.

Microcin, mainly produced by enterobacteria of fecal origin, is a naturally produced antibacterial peptide 240. MccB17, produced by E. coli commensal to poultry, targets

DNA gyrase, an essential enzyme required for DNA replication 241. It is thought that mccB17 stabilizes a temporary cleavage formed between DNA and DNA gyrase enzyme during replication, making it irreversible and leading to breakage of double-stranded

DNA 242. Accumulation of DNA breakages leads to an SOS response in E. coli which ultimately results in cell death 219. This mode of inhibition is considered to be similar to that of FQs, although mccB17 is considered a weaker inhibitor of DNA gyrase compared to FQs 243. The gyrA or gyrB mutation conferring quinolone resistance in E. coli was shown to provide limited resistance to mccB17 in vitro 219. Additionally, E. coli carrying a pPY113 plasmid harboring a mcb operon for mccB17 production, transportation and immunity showed strong resistant to FQ antibiotics such as sparfloxacin 244. It was therefore hypothesized that FQ resistance in Campylobacter can also confer protection against mccB17.

All C. jejuni strains used in this study were susceptible to the cell lysate containing mccB17, regardless of their sensitivity to FQs. Unlike in E. coli, it appears that the gyrA point mutation conferring FQ resistance in Campylobacter does not protect

Campylobacter from the activity of mccB17. In addition, while E. coli lacking an mcb

operon was sensitive to mccB17 until the 1:16 dilution, C. jejuni was only sensitive to the undiluted cell lysate. The results indicate that the inhibition mechanism of mccB17 works differently in C. jejuni than in E. coli, and a greater concentration of mccB17 is required to inhibit the growth of C. jejuni. Because the current study did not purify and measure the exact concentration of mccB17 that was present in the cell lysate, we were unable to estimate the growth inhibition threshold of mccB17 on Campylobacter.

In this study, we screened broiler litter samples to measure the prevalence of E. coli carrying a 5.1kb mcb operon. One out of 391 isolates was confirmed to carry the operon, suggesting that mccB17 producing E. coli can be detected in the poultry environment; given that the typical load of E. coli found in broiler litter is estimated to be around 108

CFU/g, it is unclear how commonly E. coli in the broiler environment carry this operon.

The spot-playing assay showed that the cell lysate of this field isolate did not inhibit the growth of sensitive E. coli. The field isolate was either unable to produce active mccB17 or the concentration of mccB17 produced was too low, thereby producing a minimal inhibition effect. The latter explanation is more plausible in this case, as E. coli harboring a mcb operon on a pUC19 plasmid is known to overproduce mccB17. Further work on purification and quantification of active mccB17 using immunoprecipitation or reverse- phase high performance liquid chromatography from the field isolate, as well as from the overproducing E. coli strain, is needed to measure the growth inhibition threshold on

Campylobacter. In addition, understanding the molecular mechanisms by which mccB17 inhibits the growth of Campylobacter, as well as its efficacy on Campylobacter in the chicken host, will be crucial for the utilization of mccB17 as a potential on-farm intervention to reduce Campylobacter colonization in poultry.

Another notable result from this study is that two of the FQ-R strains (CJ-D-R and CJ-E-

R) carried an additional missense point mutation in gyrB gene which changed serine at position 460 to phenylalanine. Several studies that screened gyrA and gyrB genes of FQ-

R C. jejuni isolates from different sources have identified various point mutations in gyrB genes in addition to the gyrA point mutation conferring FQ resistance. However, none of these gyrB point mutations have been reported as missense mutations involved with the acquisition of FQ resistance in C. jejuni 245-248. A study conducted by Chatur et al. 249 reported nonsense mutations in gyrB of FQ-R C. jejuni isolated from human and animal sources. It appears that the occurrence of point mutations in gyrB reflects common polymorphism in Campylobacter, and is expected to be observed frequently 249. The potential effects of the Ser-460-Phe gyrB point mutation in C. jejuni is unknown. Given the results of the present study, this particular point mutation does not appear to affect the growth characteristics of the strain compared to the wild-type in vitro. Additionally, this point mutation does not affect the sensitivity to mccB17 antibacterial activity in vitro. It is unclear if the point mutation had any effects on the expression levels of the six genes as the RT-qPCR comparison of the relative gene expression levels was made only within each pair, not between pairs. Further work is needed to better understand the effect of the

Ser-460-Phe point mutation in gyrB on the fitness of C. jejuni as well as its effect on the expressions of the iron acquisition and regulation genes.

Table 3.1. Oligonucleotide primers used in this study.

Gene Amplicon Forward (5’-3’) Reverse (5’-3’) Reference size (bp) GyrA-1 1081 CAACTGGTTCTAGCCTTT AATTCACTCATAGCCTC 156 TG ACG GyrA-2 1116 TAGAAGCTTATCGCACA GTGGTTGGAATAATAGC 156 GGG CATG GyrA-3 1014 AGCCGTTACGACTTATG AAAAGCAATAAGCGTTC 156 ATG CAACT GyrB-1 2544 AGACAATATGGAAGCAA TTACACATCCAAATGCTT AAACAG TAC GyrB-2 2580 ATGCAAGAAAATTACGG TCACCTTTAGTGGAAGC 156 TGC AAAT GyrB-3 1576 GTGAATGTTGAAGTAGC TTACACATCCAAATGCTT TTTGC TAC Cj0179 66 ATGGCTTGAGTTTGCTTG GCCAAAGAAGCATTGCC CG AAG Cj0426 71 GCACCTTACAAAGTGCG TCAAGCCAAGAAATCGC GAT CTC Cj0427 67 CGTAGAGTTTGATAATG TGCGAATGGTGATTTCTT GCAAGG TATCG Cj0334 67 TCAAGGTGGTATTGGTC CATGGCGAACTGTTCCA AGGT TCA Cj1661 61 GCACAGCACTTTCTCAA AGCAACCAAGTAAGGCA ATCA ATAAGT Cj1664 95 GGCTGCCCTTCTTGCTTA TCGGCTAAAACCTTAAC AA CACA rpoA 109 CGAGCTTGCTTTGATGA AGTTCCCACAGGAAAAC 224 GTG CTA E. coli 5100 TCCATGGAATTAAAAGC TAATATCTCGAGTTATCC 250 mcb GAGTGAATTTG CCCTACAACCACT operon

Table 3.2. List of genes and their potential functions measured using RT-qPCR. Gene ID Description Iron-uptake Cj0179 Encodes exbB1, a component of TonB which generates energy for iron transportation 237 Cj1661 Encodes a ferric-binding protein of a putative ABC transport system 228 Cj0426 & Co-transcribed genes which are likely to form ABC Cj0427 transport system involved in iron-uptake 228 Iron Cj0334 Encodes alkyl hydroperoxide reductase (AhpC) which is regulation involved in oxidative stress response and known to be (oxidative regulated by intracellular iron concentration by PerR 251 stress defense) Cj1664 Encodes thioredoxins which is involved in oxidative stress response by facilitating reduction 228

Table 3.3. Characteristics of the FQ-R and FQ-S C. jejuni strains used in the study. Strain Ciprofloxacin MIC Nalidixic acid MIC Point mutation in Point mutation in (µg/mL) (µg/mL) gyrA gyrB CJ-A-S 0.12 4 None None CJ-A-R 16 64 Thr-86 → Ile None CJ-B-S 0.12 4 None None CJ-B-R 32 64 Thr-86 → Ile None CJ-C-S 0.12 8 None None CJ-C-R 64 64 Thr-86 → Ile None CJ-D-S 0.12 4 None None CJ-D-R 32 64 Thr-86 → Ile Ser-460 → Phe CJ-E-S 0.06 4 None None CJ-E-R 16 64 Thr-86 → Ile Ser-460 → Phe

Table 3.4. Comparison of growth kinetics between FQ-S and FQ-R C. jejuni using pairwise permutation statistical tests with 10,000 simulations on OD600. Test statistics P-value CJ-A-S vs. CJ-A-R -0.336 0.667 CJ-B-S vs. CJ-B-R -0.0232 0.982 CJ-C-S vs. CJ-C.R 1.46 0.0874 CJ-D-S vs. CJ-D-R 1.04 0.187 CJ-E-S vs. CJ-E-R -0.643 0.514

Figure 3.1. Relative expression ratio of iron acquisition and oxidative stress regulation genes in FQ-R and FQ-S C. jejuni strains over time. The solid line represents relative expression ratio over time relative to the baseline (time point = 0) when 0 µM of 2, 2-dipyridyl was added to the culture. The dotted line represents the relative expression ratio over time relative to the baseline (time point = 0) when 100 µM 2, 2-dipyrdyl was added to the culture to create iron-limiting conditions. . * p < 0.05 under a pairwise reallocation randomization test using 10,000 randomizations.

Figure 3.2. Relative expression ratios of iron acquisition and oxidative stress regulation genes in FQ-R C. jejuni relative to FQ-S C. jejuni under normal and iron-limiting conditions. Each gene is represented in one row and each FQ-R strain is represented in one column. (A) Relative expression ratio between FQ-R and FQ-S C. jejuni under normal conditions. Error bars represent 95% confidence intervals. (B) Relative expression ratio between FQ-R and FQ-S C. jejuni under iron-limiting conditions. Error bars represent 95% confidence intervals. * p < 0.05 under a pairwise reallocation randomization test using 10,000 randomizations.

Figure 3.3. The effect of mccB17 on the growth of E. coli DH5α strains susceptible or resistant to mccB17 and ciprofloxacin. Cell lysates extracted from three different E. coli strains were applied directly to the top-agar layers. ‘Puc19’ represents cell lysates extracted from mcb17 producing E. coli. ‘DH5α’ represents cell lysates extracted from wild-type E. coli DH5α strain. ‘I’ represents cell lysates extracted from a field E. coli isolate obtained from broiler litter and confirmed to harbor a mcb operon by PCR amplification. (A) E. coli DH5α strain harboring a pUC19 plasmid with a mcb operon (positive control). (B) Wild-type E. coli DH5α susceptible to ciprofloxacin (negative control). (C) Wild-type E. coli DH5α resistant to ciprofloxacin (MIC ≥ 4 µg/mL). (D) Wild-type E. coli DH5α strain susceptible to ciprofloxacin. Cell lysates applied to this plate was extracted from mccb17 producing E. coli DH5α, and the cell lysate was two- th fold diluted in LB broth to create ½ to 1/64 dilutions.

Figure 3.4. The effect of mccB17 on the growth of FQ-R and FQ-S C. jejuni. Cell lysates applied on top-agar layers were extracted from mccB17 producing E. coli DH5α strain. Undiluted cell lysate (represented as ‘Ori’) and two-fold diluted lysates (in Mueller-Hinton broth) were directly applied to top-agar layers containing FQ-R and FQ- S C. jejuni strains.

Chapter Four: Characterization of temporal changes in the bacterial microbiota of conventional broiler litter

Summary

Broiler litter, which is comprised of a mixture of bedding materials, spilled feed and bird excreta, harbors a complex microbiota that plays an essential role in the development of broiler chickens, ultimately affecting health and growth performance of birds. The purpose of this study was to characterize temporal changes in the bacterial microbiota of broiler litter obtained from commercial broiler farms and to compare litter microbiota among and within farms with different characteristics. Seven commercial broiler farms were enrolled in the study. From these farms, one house was selected and sampled weekly for two consecutive flock cycles from two sections of the broiler house (brooding and non-brooding areas). DNA was extracted from the composite litter samples (n=178) for 16s rRNA gene sequencing using the V4 hypervariable region. Additionally, an MPN method was used to enumerate total and fluoroquinolone-resistant Campylobacter in the litter. The prevalence and average logCFU/gram of total and fluoroquinolone-resistant

Campylobacter were 15.7% (5.6 logCFU/gram) and 5.6% (5.16 logCFU/gram), respectively. The effects of flock cycle, age of bird, antibiotic use and Campylobacter status on litter microbiota were assessed. Age of bird and flock cycle were significant factors driving the changes in the bacterial diversity and composition of the litter microbiota. Additionally, bacterial genera responsible for decomposing organic materials or halotolerant were observed in a higher abundance in the second flock compared to the first flock. No differences in the bacterial diversity and composition were observed between Campylobacter-positive and Campylobacter-negative litter samples. However,

several genera were detected in a lower abundance in the positive samples compared to the negative samples, including Lactobacillus spp. and Acinetobacter spp. Future studies are warranted to further investigate how litter bacterial microbiota is affected by different litter and husbandry management practices. Additionally, investigating the interactions between Campylobacter and bacterial genera that were differentially abundant in the litter, especially Lactobacillus spp., may provide new insight into the development of on- farm mitigation tools to reduce Campylobacter colonization in broiler chickens.

Introduction The microbiome of gastrointestinal (GI) tract of chickens plays an essential role in the development of immune system 252, nutrition digestion and absorption 253, and pathogen exclusion 254, ultimately affecting the health and growth performance of chickens. During the grow-out period, chicks are exposed to a wide-variety of sources of bacteria that colonizes their GI tract, transforming simple, transient GI microbiome into a more complex and stable community over time 255,256. Litter, which is a mixture of bedding materials, spilled feed and bird excreta, is an important exogenous source of bacteria in chickens 255,257. From day 1 of the hatch, chicks consume litter regularly 258, becoming exposed to the bacteria present in the litter. As such, litter bacterial community heavily influences the chicken’s GI microbiome, especially during the early stages of growth

255,259-261.

In the U.S. broiler industry, the reuse of litter for multiple cycles of flocks is a common practice due to the cost of disposing and replacing litter after every flock 262. Litter reuse practice often involves removing caked litter, windrowing or composting of litter to reduce bacterial loads and spreading of new bedding materials prior to the placement of a

new flock 263. While reusing the litter is economically efficient, it also increases the risk of human and animal pathogen transmission between flocks through the litter 264,265. A study by Kassem et al. indeed showed that Campylobacter spp. (Campylobacter) was able to survive a longer period of time in reused litter compared to fresh litter 265.

Similarly, Chen et al. and Soliman et al. showed that the survival of fecal coliforms and parasite oocyst was favored in reused litter unless additional chemical treatment was applied to the litter to reduce moisture and ammonia levels, indicating that litter management practices have an important role in the poultry gut health as well as for the food safety and public health 266-268.

A number of factors can influence diversity and composition of bacterial community of the broiler litter. Previous studies have shown that the age of litter 255,259,261, husbandry practices 261,269,270, and the use of chemical litter treatment 271,272 can affect the physiochemical conditions of the litter, which ultimately influence the types and levels of bacteria found on the litter. However, these studies have focused on detecting culturable bacteria 273,274 or employing PCR to detect specific pathogens of human importance 275.

More recent studies that have utilized the direct 16s rRNA gene sequencing technology were, however, conducted using a cross-sectional study design or as pen trials using a small number of birds in a controlled condition 255,259,261,269,276. In a commercial broiler farm, a number of factors, such as on-farm biosecurity measures, litter management practices, husbandry practices, high stocking density, as well as the environment where the farm is located can influence which bacteria are introduced to and persist in the litter.

More importantly, effects of age of birds, an important factor that seem to predict the GI

bacterial microbiota of broiler chickens 277, on broiler litter microbiota has never been investigated.

The overall aim of the study was to use high-throughput sequencing techniques to better define broiler litter microbiota of commercial broiler farms and characterize temporal changes in the bacterial diversity and composition associated with bird age and flock cycle. In addition, the current study explored the effects of farm characteristics, such as use of antibiotics and presence of Campylobacter on the bacterial litter microbiota.

Methods Study design and sample collection

Seven commercial broiler farms located in the U.S. were enrolled in the study. One house was selected from each farm and sampled weekly for two consecutive flock cycles. For each house, sampling began at the placement of chicks into the house for flock 1 and continued weekly until the birds were removed at the conclusion of flock 2 for slaughter.

No litter sample was collected during the down time. Approximately 5 random grab litter samples were collected from three segments of the house: front, middle and back (Figure

4.1). Grab samples from each location were pooled into a composite sample and placed in

Whirl-Pak bags, resulting in three composite litter samples per house per sampling event.

Composite litter samples were delivered to the laboratory overnight in a dry-ice pack. All samples were processed the following day upon delivery to the laboratory.

Litter processing and DNA extraction

Phosphate-Buffered saline was added to the composite litter sample and vortexed for 1 minute to homogenize the litter. One mL of the litter wash was saved for DNA extraction. DNA extraction was performed using PowerSoil DNA Isolation Kit (Mo Bio

Laboratories, Carlsbad, CA) following the manufacturer’s instructions. Litter wash from the middle and back section of the house was pooled before DNA extraction, resulting two DNA samples per house per sampling event (Figure 4.1).

Campylobacter enumeration using Most Probable Number

Thirty mL of litter wash liquid was centrifuged at 2,500g for 15 minutes at room temperature. Supernatant was collected and centrifuged again at 3,000g for 15 minutes to collect a cell pellet. The cell pellet was resuspended in 9 mL of Bolton enrichment broth

(20 µg/mL cefoperazone, 20 µg/mL vancomycin, 20 µg/mL trimethoprim, 50 µg/mL cycloheximide and 5% lysed horse blood) and vortexed for a minute. A miniature three- tube Most Probable Number (MPN) enumeration method was used to estimate CFU/g of total and fluoroquinolone-resistant (FQ-R) Campylobacter in the broiler chicken litter.

Briefly, 200 µL of the cell pellet inoculum was added to wells in the 1st row of a 96-well

2-mL deep-well plate containing 800 µL of Bolton enrichment broth. Five-fold serial dilutions were performed sequentially. Each sample was run in triplicate. The 96-well plates were incubated for 24hrs at 42ºC under microaerophilic conditions (5% O2, 10%

CO2 and 85% N2). After incubation, 10 µL of each dilution was spot-plated on Campy

Cefex agar with Preston supplement (16.5 mg/mL cefoperazone, 1 µg/mL rifampicin, 1

µg/mL trimethoprim, 100 µg/mL cycloheximide, 2500 IU polymyxin and 5% lysed horse blood) to enumerate the total Campylobacter. Campy Cefex agar with Preston supplement and ciprofloxacin (4 µg/mL) was used to enumerate FQ-R Campylobacter.

Plates were incubated for 48hrs at 42ºC under microaerophilic conditions. Following incubation, plates were examined for growth of Campylobacter-like colonies. MPN wells

from the plates found to contain Campylobacter-like colonies were used to calculate

MPNs as previously described 278.

16s rRNA gene sequencing and processing

Dual-index amplification was performed to prepare the sequencing library using the

KAPA Hifidelity HotStart Polymerase (Kapa Biosystems, Inc., Wilmington, MA). The

V4 hypervariable region of the 16s rRNA gene was initially amplified using the

Meta_V4_515F (5’-

TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGGTGCCAGCMGCCGCGGTA

A-3’) and Meta_V4_806R (5’-

GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGGACTACHVGGGTWTCTA

AT-3’) using the following PCR cycling parameters: one cycle of 95ºC for 5 minutes, followed by 25 cycles of 98ºC for 20 seconds, 55ºC for 15 seconds and 72ºC for 1 minute and one cycle of 72ºC for 5 minutes. The amplicons were diluted 1:100, and 5µL was used in the second PCR reaction. The second PCR amplification was performed to add different combinations of forward and reverse indexing primers using the following cycling parameters: one cycle of 95ºC for 5 minutes, followed by 10 cycles of 98ºC for

20 seconds, 55ºC for 15 seconds and 72ºC for 1 minutes and one cycle of 72ºC for 5 minutes. Pooled amplicon was denatured with NaOH to 8 pM in Illumina’s HT1 buffer, spiked with 15 % PhiX, and heat denatured at 96ºC for 2 minutes. Sequencing was performed using the Illumina Miseq 300-bp paired-end sequencing (Illumina, San Diego,

CA) at the University of Minnesota Genomics Center.

After sequencing, demultiplexed paired-end reads were quality filtered and processed using the Divisive Amplicon Denoising Algorithm 2 (DADA2) (ver.1.10.1) pipeline 279 in RStudio (ver. 1.1.419) 280 using R version 3.5.2 227. Amplicon sequence variants

(ASV) were filtered to remove bimeras, and taxonomy was assigned to the genus taxonomic level using the SILVA 16s rRNA sequence database (ver. 132) 281,282. Any sequence reads not assigned to the kingdom Bacteria or assigned to the family

Mitochondria, the class Chloroplast or the phylum Cyanobacteria or Chloroplast were removed.

All subsequent bioinformatics analyses were performed using the R packages Phyloseq

(ver. 1.24.2) 283 and Vegan (ver. 2.5-4) 284. Differences in alpha diversity (observed richness and Shannon diversity index) were compared within and among farms using the

Wilcoxon Rank Sum test and Kruskal-Wallis Rank Sum Test. Prior to assessing beta diversity, the read counts were normalized using Cumulative Sum Scaling (CSS) implemented in the R package metagenomeSeq (ver. 1.24.1) 285. The default value of 0.5 was used for the normalization. The differences in the community membership (absence or presence of ASVs via Jaccard index of similarity) and composition (relative abundance of ASVs via Bray-Curtis dissimilarity index) were investigated using the Principal

Coordinate Analysis (PCoA) ordination method. After the preliminary assessment of beta diversity within farms, the reads from the front and middle-back section of the farm were combined, resulting in 89 litter sample for further analysis. Redundancy analysis (RDA), an equivalent to constrained principal component analysis (PCA), was performed to identify a set of farm variables that best describes the patterns of variation in the dissimilarities observed in the composition of ASVs as measured by Bray-Curtis

dissimilarity index among litter samples. Forward model selection was performed using the ordistep function in Vegan with 999 permutations. The initial model included intercept and farm ID as a conditional term. The available farm variables were flock cycles, age of birds (categorized into weeks; week 1- 7), Campylobacter status and antibiotic use. Total number of reads was included in the model to control for the different sequencing depth among the samples. Due to the wide range of the sequencing depth, the total number of reads was studentized before it was added to the model.

Analysis of variance (ANOVA) was performed on the final model to test the overall model significance and to assess significance of each variable included in the model using 999 permutations.

Permutational Multivariate Analysis of Variance (PERMANOVA) was performed to identify farm variables that were statistically associated with the shifts in the bacterial composition in the broiler litter using the adonis2 function in Vegan. To test the homoscedasticity assumption of PERMANOVA, multivariate homogeneity of groups dispersion test was performed using betadisper function in Vegan. Predictor variables included in the PERMANOVA analysis were age of birds in weeks (nested within flock), flock cycles (nested within farms), farm ID as a strata term which limits permutations to be constrained by farm, Campylobacter status, antibiotic use and the studentized total number of reads. Marginal effects of each term was assessed. Akaike information criterion (AIC) was calculated for each model. The model with the lowest AIC and residual R2 was chosen as the best model.

To identify ASVs that were differentially abundant within and among farms, differential abundance testing was performed using the R package DESeq2 (ver. 1.22.2) 286 at the

genus level using the raw ASV table as suggested by McMurdie and Holmes 287. The top

300 most abundant ASVs were included in the analysis. First, a comparison was made between conventional and NAE farms. Second, the data was split by the farm type, and within each farm type effect of flock cycle (flock 2 vs. flock 1) was tested. Although

Campylobacter status was not identified as a significant variable that could explain the shift in the composition of the litter microbiota, DESeq2 was run by Campylobacter status. Because Campylobacter was not detected until week 3 of bird age, litter samples from week 1 and 2 were removed prior to performing DESeq2 analysis. Pairwise comparison was made using the Wald test. P-values were adjusted for multiple comparisons using the Benjamin-Hochberg adjustment. All data visualization was performed in R using ggplot2 (ver. 3.1.0) 185.

Results Litter samples collected from the seven commercial broiler farms enrolled in the study were sampled between February and September 2017. Each farm was sampled weekly for two consecutive flock cycles, from week 1 to week 6-7 of bird age, resulting in a total of 178 broiler litter samples. Characteristics of the each farm enrolled in the study is listed in Table 4.1. The major differences in the husbandry management practices among the farms were a complete cleanout of broiler house prior to the placement of flock 1 in farms 1 – 5 and the use of antibiotics in farms 6 and 7. The information about types of antibiotics and doses administered to birds reared in farms 6 and 7 could not be collected.

Prevalence of total and FQ-R Campylobacter in broiler litter

MPN was performed on each sample to detect and enumerate total and FQ-R

Campylobacter. The prevalence and average level of total and FQ-R Campylobacter was

15.7% (5.6 logCFU/g) and 5.6% (5.2 logCFU/g), respectively. Farms 1, 5 and 6 remained

Campylobacter-positive throughout the two flock cycles, while farms 2 and 3 were only positive during the second flock (Figure 4.2). Farms 4 and 7 stayed Campylobacter- negative throughout the study period. FQ-R Campylobacter was only detected in the second flock of farms 5 and 6. Among the Campylobacter-positive farms, Campylobacter was not detected until week 3 of bird age.

Comparison of broiler litter bacterial community diversity and composition within and among farms

Longitudinal weekly sampling of seven conventional broiler farms over two consecutive flock cycles resulted in 178 broiler litter samples, which were processed and sequenced for the 16s rRNA gene’s V4 hypervariable region. A total of 1,986,812 reads was generated with an average sequencing depth of 11,161 reads per sample. After removing bimeras and filtering out ASVs not assigned to Bacteria, 3,155 ASVs were retained for downstream analysis. ASVs were classified to 17 bacterial phyla and 509 genera, with

Actinobacteria (24.1%) Firmicutes (26%) and Proteobacteria (26%) detected in all samples.

Alpha diversity within and among farms was assessed using the observed richness

(number of ASVs) and Shannon’s diversity index (Figure 4.3). Higher alpha diversity was observed in the middle-back (non-brooding) section of the broiler house compared to the front (brooding) section, with statistically different alpha diversity observed at week

2, 3 and 5 of bird age. Higher alpha diversity was observed in the second flock compared to the first flock in most of the farms. Conventional farms, however, showed the opposite results; both farms 6 and 7 showed higher observed richness and Shannon’s diversity

index in the first flock compared to the second flock. Additionally, conventional farms showed overall higher alpha diversity compared to No Antibiotics Ever (NAE) farms. No difference in alpha diversity was observed between Campylobacter-positive and

Campylobacter-negative litter samples.

Alpha diversity was also compared between two flock cycles across age of bird within each farm type (Figure 4.4). The general trends of increasing richness over time in the first flock followed by the stabilization of the richness during the second flock was observed among the NAE farms. Examining the Shannon’s diversity index, a fluctuation was observed especially in the second flock. The observed number of ASVs was statistically different weekly in the first flock, while the Shannon’s diversity index was statistically different weekly in the second flock in the NAE farms. In the conventional farms, overall, a great variation in alpha diversity was observed. Only the observed number of ASVs was statistically different weekly in the second flock.

The differences in the community membership and composition were visualized using

PCoA ordination methods using Jaccard and Bray-Curtis indices (Figure 4.5). The middle and back sections of the farm were combined as the samples from these two sections of the farm seem to resemble each other (Figure S4.1) and to reduce the hierarchical structure of the data. Conventional farms (farms 6 and 7) were clearly distinct from NAE farms (farms 1 – 5) during both flock cycles in terms of ASV membership and compositions (Figure 4.5A and B). When examining the samples by age of bird and flock cycle, samples from the first flock, especially from the early time points (weeks 1-3), were not grouped together with the rest of the samples. In fact, the samples from the end of the first flock cycle resembled those from the second flock, and the samples form the

second flock were more similar to each other regardless of the age of bird (Figure 4.5C and 4D). No differentiation between Campylobacter-positive and –negative samples was observed (data not shown).

Constrained RDA was performed to identify a set of variables that best describes the variation observed in the bacterial composition among litter samples. Using the forward model selection, three variables were included in the final model: flock cycles, age of bird and studentized total number of reads. The farm ID was included in the model as a condition term (i.e. the farm effect was partialled-out before permutation to control for farm differences). The variables (conditional and constrained terms) included in the final model explained 50% of the variation (Table 4.2). Both the final model and the variables included in the model were statistically significant (Table 4.2). The dissimilarity observed in the bacterial composition was visualized using RDA ordination method

(Figure 4.6). After controlling for the farm effects, the differentiation that was observed between NAE and conventional farms from the PCoA plot was no longer observable

(Figure 4.6A). A clear distinct between the litter samples collected during the first flock and the second flock was still evident after controlling for the farm effect (Figure 4.6B).

The samples from the first flock was more dispersed than the samples from the second flock. In addition, samples from the early time points, especially the samples from week

2 and 3, of the first flock were not grouped together with the rest of the samples (Figure

4.6C).

To test the homoscedasticity within variable assumption of PERMANOVA, analysis of multivariate homogeneity of groups dispersion test was performed. Farm variables tested were: farm ID, flock cycle, age of bird in week, antibiotic use and Campylobacter status

of litter samples. Statistically different variance was observed within each variable tested

(Table 4.3 and Figure S4.2). Although the homoscedasticity assumption of

PERMANOVA was not met, PERMANOVA was performed to identify farm variables that were associated with the shifts in the bacterial composition among broiler litter samples. Several different models were considered, and the model with the lowest residual R2 and AIC value was selected as the best model (Table S4.1). The variables included in the best model and the relevant PERMANOVA results can be found in Table

4.4. While controlling for the farm effects and the hierarchical structure of the data, flock cycles, age of bird and studentized total number of reads were statistically significant variables that were associated with the changes in the bacterial composition in the broiler litter samples. It appears that Campylobacter status of litter and antibiotic use do not contribute significantly to the differences observed in the litter bacterial composition.

PERMANOVA was performed again after splitting the data by antibiotic use

(conventional vs. NAE) and similar results were observed. The studentized total number of reads and age of birds were statistically significant in both conventional and NAE farms, while flock cycle was additionally significant in the NAE farms (data not shown).

Differentially abundant bacterial genera by farm type and flock cycle

To better understand how microbial composition of broiler litter changes over time, relative abundances of major phyla (relative abundance > 1%) were plotted by flock cycle and age of bird within each farm type (Figure 4.7). Both farm types showed that relative proportion of major phyla changed according to the age of bird. Initially,

Firmicutes was a highly abundant phylum during the early time points but the relative proportion of Actinobacteria and Proteobacteria increased over time. However,

Firmicutes remained the dominant phylum in all samples over all time points. Because litter microbiota is dominated by Firmicutes, core families belong to Firmicutes were further investigated by plotting the relative abundance of families that were present in all litter samples (Figure 4.8). The core Firmicutes families present in the litter samples differed by farm type. Among the NAE farms, Ruminococcaceae was the major family in the early time points, particularly during the first flock. After week 4 of the first flock cycle, the core Firmicutes family predominance shifted from Ruminococcaceae to

Staphylococcaceae and Lactobacillaceae. Among the conventional farms,

Lactobacillaceae was the major family present in all time points, followed by

Staphylococcaceae, Ruminococcaceae and Carnobacteriaceae.

Additionally, a heatmap showing the top 50 most abundant ASVs was plotted to identify core bacterial genera by farm type across time (Figure 4.9). A very clear distinction between NAE and conventional farms was observed. Several ASVs classified to Yaniella,

Salinicoccus, Sphingobacterium, Staphylococcus, Acinetobacter and Pseudomonas were absent in the first two or three weeks of bird age in the first flock of the NAE farms, yet these ASVs were abundant in the conventional farms. In contrast, ASVs classified to

Klebsiella, Acinetobacter and Staphylococcus were absent in the conventional farms but present in the NAE farms. Because ASVs were assigned to the genus taxonomic level, specific species of Lactobacillus, Acinetobacter and Staphylococcus that were either absent or present in each farm type could not be identified. There were 5 ASVs that could not be assigned to the genus level. Looking at the family taxonomic level of these 5

ASVs, they belong to Sphingobacteriaceae, Bacillaceae, Xanthomonadaceae,

Enterobacteriaceae and Lachnospiraceae.

To identify ASVs that were differentially abundant within and among farms, differential abundance testing was performed using DESeq2. First, a comparison was made between two flocks within each farm type (Figure 4.10, Table S4.2). Of the 135 ASVs that were differentially abundant in the second flock relative to the first flock in the NAE farms, 73

ASVs were more abundant and 62 ASVs were less abundant. Overall, ASVs classified to

Firmicutes and Proteobactera were less abundant in the second flock. Several ASVs classified to Acinetobacter were less abundant in the second flock. Among the conventional farms, only 34 ASVs were differentially abundant in the second flock relative to the first flock: 16 ASVs were more abundant in the second flock and 18 ASVs were less abundant.

Similar analysis was performed by farm type (conventional vs. NAE) over all time points

(Figure 4.11, Table S4.3). Overall, 70 bacterial genera were more abundant in the NAE farms compared to the conventional farms. Of these, 7 ASVs classified to Lactobacillus were less abundant in the NAE farms compared to the conventional farms. Sixty three

ASVs were less abundant in the conventional farms compared to the NAE farms: ASVs classified to Acinetobacter, Comamonas and Staphylococcus were more abundant in the

NAE farms compared to the conventional farms.

Relationship between Campylobacter status and broiler litter microbiome

The difference in the litter bacterial composition was further investigated by comparing relative abundance of major phyla (relative abundance > 1%) and differentially abundant

ASVs between Campylobacter-positive and Campylobacter–negative litter samples

(Figure 4.12). Four major phyla, Actinobacteria, Bacteroidetes, Firmicutes and

Proteobacteria, were identified in both groups (Figure 4.12A). Higher relative

abundances of Actinobacteria (28.0% vs. 25.3%) and Bacteroidetes (8.1% vs. 6.4%) and lower relative abundances of Firmicutes (47.9% vs. 48.4%) and Proteobacteria (15.8% vs. 19.7%) were observed in the positive samples compared to the negative samples.

However, the differences in the relative abundance of the four major phyla were not statistically significant between the positive and negative samples. DESeq2 was performed to identify ASVs that were differentially abundant in the positive samples compared to the negative samples (Figure 4.12B). Compared to the negative samples, 24

ASVs were differentially abundant in the positive samples. Lactobacillus and

Acinetobacter were two bacterial genera less abundant and Staphylococcus was more abundant in the positive samples compared to the negative samples. There were 5 ASVs that could not be classified to the genus level: 1 ASV that was less abundant was classified to Burkholdeeriaceae and 4 ASVs that were more abundant were classified to

Sphingobacteriaceae and Weeksellaceae. A complete list of the differentially abundant

ASVs and log2 fold changes can be found in Table S4.4.

Discussion Understanding and defining the broiler litter microbiome is important for proper litter management and improvement of flock health, as the litter microbiome and the chicken

GI microbiome are interconnected and influence each other heavily. In the current study, litter samples from commercial broiler farms belonging to the same vertically integrated broiler production company were collected to characterize the temporal changes in the litter microbiota over two consecutive flock cycles. Additionally, the study examined the differences in litter microbiota between two farm types (conventional vs. NAE) and sample types (Campylobacter-positive vs. Campylobacter-negative).

Prevalence and estimated loads of total and FQ-resistant Campylobacter were measured in each litter sample using the MPN method. Of the 7 farms, two farms (farms 4 and 7) remained Campylobacter-free during the study period. The broiler farms enrolled in the study were under contract with the same broiler production company and located in the same region. Thus, it was assumed that they shared similar husbandry and farm management practices. However, because details about the broiler house conditions (e.g., age of the broiler house, ventilation types used, environmental conditions around the farm, etc.), which are important on-farm risk factors associated with increased presence of Campylobacter in broiler chickens, were not collected, specific farm factors that could have contributed to the introduction of Campylobacter could not be compared between

Campylobacter-positive and Campylobacter-negative farms.

Among the Campylobacter-positive farms, Campylobacter were not detected in the litter until the flock reached at least 3 weeks of age, which concurred with the previous research 81,95,288. Within the Campylobacter-positive broiler houses, Campylobacter was not detected in all sections of the house. This result suggests that there may be areas within the broiler house that may preferentially favor the survival of Campylobacter in broiler litter. It has been shown that wet litter found under water lines and other areas where high moisture is found on litter favors Campylobacter growth 191. It is likely that upon exposure to hostile environmental conditions, such as dry litter, Campylobacter can undergo viable-but-non-culturable (VBNC) state as suggested by previous studies 289,290.

As such, using only one or two litter samples to measure occurrence and/or levels of

Campylobacter in broiler litter may give false negative results. Future studies looking at the prevalence or levels of Campylobacter on the broiler litter should consider sampling

more than one composite litter sample and using a combination of enrichment culturing methods and quantitative approaches to better understand the spatial heterogeneity of

Campylobacter distribution in the broiler house.

Just as the GI microbiota of broiler chickens is heavily influenced by flock cycle and age of bird 277,291-294, our finding suggests that the overall structure of litter microbiota also changed by flock cycle and age of bird. Differences in microbial diversity and composition could be observed between the two consecutive flock cycles. These differences were more noticeable among the NAE farms compared to the conventional farms. The number of ASVs observed in the litter samples collected from the NAE samples started out low at the beginning of the first flock then gradually increased over time, eventually stabilizing during the second flock. Shifts of the bacterial composition observed by the flock cycle also suggest that the effect of flock cycle was also stronger in the NAE farms. The results of PERMANOVA reinforced the patterns observed in the

PCoA and RDA ordination plots. Flock cycle and age of bird were two statistically significant factors driving the shifts in the litter bacterial composition. However, when

PERMANOVA was performed within each farm type, flock cycle was no longer statistically significant in the conventional farms, suggesting that perhaps there were other confounding factors responsible for the pronounced flock cycle effects observed in the NAE farms.

One possible factor was the use of antibiotics. In our study, Lactobacillus was one of the genera that was less abundant in the NAE farms compared to the conventional farms.

This result was in contrast with several other studies that have investigated the modulatory effects of antibiotics on the GI microbiota of broiler chickens. The previous

studies have found that the use of antibiotics, specifically tylosin and virginiamycin in broiler chickens led to a reduction of abundance of Lactobacillus 295-297. However, a study by Dumonceaux et al. showed results similar to ours in which an increased abundance of Lactobacilus was observed in day-49 old broiler chickens fed subtherapeutic doses of virginiamycin 298. Likewise, a study by Collier et al. showed that administration of tylosin as growth-promoters in swine led to an increased abundance of

Lactobacillus 299. It is clear that the use of antibiotics alters bacterial composition of GI tract in broiler chickens. However, it seems that the types and doses of antibiotics and age of bird at which antibiotics are administered could modify the effects of antibiotics have on the GI microbiota. We could not collect the information regarding the antibiotics usage in the conventional farms. Furthermore, there were other differences between the

NAE and conventional farms, such as how the litter was managed prior to the sampling and season of grow-out. Thus, it could not be concluded that the differences seen in the litter microbiota between the two farm types was due to the use of antibiotics.

As mentioned previously, another possible confounding factor was the complete replacement of old litter with fresh litter which took place prior to the first flock placement in the NAE farms. Dry conditions of fresh litter is considered unfavorable for bacterial growth 269,300 and a complete replacement of litter prevents carry-over of bacteria between flocks. As such, alpha diversity of fresh litter starts out low, as shown by our results. As litter becomes saturated with bird excreta, water, and feed, the litter conditions become more suitable for bacterial growth, allowing the litter microbiome to stabilize over time. In fact, the results from DESeq2 comparing the differentially abundant bacterial genera between the NAE and conventional farms further suggest that

the cleanout of the used litter could be one of the influential factors driving the changes observed in the litter microbial composition. Previous studies have found that reused litter had higher bacterial diversity and was dominated by halotolerant or alkaliphilic bacteria and butyrate-producing bacteria 255. Among the bacterial genera less abundant in the NAE farms relative to the conventional farms, several ASVs classified as halotolerant bacteria, including Brevibacterium 256,301, Brachybacterium 302 and Halomonas 303,304 were detected. Additionally, alkalitolerant bacteria, Dietzia 305, and butyrate-producing bacteria, Butyricicoccus 306, were less abundant in the NAE farms compared to the conventional farms.

Lastly, the season at which samples were collected was also different between the NAE and conventional farms. Litters were collected during colder months for the NAE farms while samples were collected during warmer months for the conventional farms. A study by Oakley et al. showed that seasonality was the most influential factor driving the changes in the cecal microbiome of broiler chickens 307. Notably, samples collected in winter had lower alpha diversity and several bacterial taxa, including Lachnospiraceae and Ruminococcaceae that were more abundant compared to the samples collected in summer. Our results corresponded with that of Oakley et al. Of interest, higher alpha diversity was observed in the litter samples collected from warmer months (conventional farms) compared to the colder months (NAE farms). Ruminococcaceae were among the

ASVs that were less abundant in the NAE farms relative to the conventional farms.

While the flock cycle and age of bird were statistically significant variables driving the shifts of bacteria composition in broiler litter according to the PERMANOVA analyses, it should be noted that homoscedasticity assumption of PERMANOVA was not met for

both variables (Table 4.3). PERMANOVA is considered to be more robust to heterogeneity of group dispersions than Analysis of Similarity (ANOSIM), especially for balanced designs 308. However, it is possible that the significant results of the

PERMANOVA analyses may be due to the heterogeneous dispersions observed within variables.

Despite the differences observed in the bacterial diversity and composition of litter samples from two farm types, Firmicutes was identified as the dominant phylum in all litter samples (Figure 4.7), which corroborated with previous studies 255,259,309.

Firmicutes, being the core phylum of the GI tract of broiler chickens 292,294,310-312, it was expected that Firmicutes to be the dominant phylum of litter microbiota as well. The dynamics of litter microbial composition was further investigated by examining the core families belong to Firmicutes (Figure 4.8) and the top 50 most abundant ASVs among all litter samples over all time points (Figure 4.9). The relative abundance of the core

Firmicutes families varied by the farm type across age of bird again, yet

Lactobacillaceae, Ruminococcaceae, and Staphylococcaceae were three most predominant families detected in both farm types. Several ASVs classified to

Lactobacillus, Escherichia/Shigella, Faecalibacterium, and Streptococcus were present in all samples across age of bird. These bacterial genera are common genera found in the

GI tract of broiler chickens 277,294,307,312,313, once again indicating that broiler litter microbiota is strongly influenced by the GI microbiota of broiler chickens.

In addition to the taxa commonly found in the GI tract of broiler chickens, bacterial genera associated with decomposing organic materials were detected in a higher abundance in the litter samples from the second flock relative to the samples from the

100

first flock. Bracybacterium, which was previously isolated from poultry litter and shown to decompose uric acid 302, was more abundant in the second flock compared to the first flock. Likewise, Brevibacterium, bacterial genus known to decompose wood 256,301, were detected in a higher abundance in the second flock compared to the first flock.

It is well known that a broiler house is composed of distinct areas (e.g., under the waterline, brooding area, near the ventilation fans, etc.) where heterogeneous bacterial populations can be found 275,309. A study by Lovanh showed that distinct bacterial structures could be observed in different parts of the broiler house, and this distinct bacterial structures corresponded to litter physiochemical conditions 275. Likewise, a study by Locatelli showed that significant different beta diversity could be observed by sampling different areas within a broiler house 309. In this study, we tried to take account for the spatial heterogeneity of bacterial populations by sampling two sections within the broiler house, mainly brooding and non-brooding areas.

Brooding period allows broiler chicks to maintain proper body temperature and consume energy for development, rather than generating body heat, allowing chicks to gain more body weights 314. A section of the broiler house is typically set up as a brooding area where a local heat source, food and water are provided to achieve optimal growth conditions for chicks to grow until they reach 10-14 days of age. The broiler houses enrolled in the study used the front section of the house as brooding areas at the beginning of each flock cycle. As such, each broiler house was divided into brooding

(front section) and non-brooding (middle-back) sections and sampled separately.

Difference in the litter microbial composition was expected between these two sections, especially during the brooding period (week 1 and 2) as chicks are restricted to the

101

brooding area during that time period. However, litter microbial composition, in terms of

ASV membership and relative abundances, was not distinct between the two sections of the broiler house. PERMANOVA results also showed that house section was not a significant factor driving the shifts in the litter microbiome. This may be due to the fact that a brooding period is short and the changes in the microbial composition observed during the brooding period is rather transient such that once the brooding period is over, microbial composition of the litter of the two sections become similar to each other as the birds are allowed to freely roam around the broiler house. Additionally, random grab samples from each section of the house was pooled to create composite samples, a common sampling strategy adopted in poultry studies. As demonstrated by Locatelli et al., pooling litter samples to create a single composite sample could affect the bacterial profiles 309. It is possible that pooling of samples might have homogenized the differences otherwise could have been observed. Future studies should take into account varying litter conditions within the poultry house in the sampling scheme and study design to capture heterogeneous litter microbiota within the broiler house.

Another objective of the current study was to explore whether an occurrence of

Campylobacter in broiler litter affects the litter microbiota. Because Campylobacter is an important human foodborne pathogen, several studies have investigated whether the absence or presence of Campylobacter affects the bacterial composition of the gastrointestinal microbiome in broiler chickens 307,315-317. In this study, the

Campylobacter status of the litter sample had no effect on the overall alpha diversity and beta diversity of the litter microbiome. No statistically significant difference was observed in the abundance of three major phyla (Actinobacteria, Firmicutes and

102

Proteobacteria) between the positive and negative samples, which corroborated the previous studies. However, several ASVs were differentially abundant in the positive samples compared to the negative samples. One ASV classified to Staphylococcus was more abundant in the positive samples compared to the negative samples, similar to the results observed in Sofka et al 316 and Saklaridis et al 315. Interestingly, several ASVs classified to Sphingobacteriaceae, a family previously found in aerobic 318 and fresh litter

255, was more abundant in the positive samples. Because Campylobacter is susceptible to desiccation 300, dry condition of fresh litter is thought to be lethal to Campylobacter. As such, the co-occurrence of Sphingobacterium and Campylobacter in the broiler litter seems unlikely. Interaction between Sphingobacterium and Campylobacter should be further investigated.

Two ASVs classified to Lactobacillus spp. were less abundant in the positive samples compared to the negative samples. Similar results have been reported in which higher abundance of Lactobacillus was detected in the cecal and fecal microbiome of broiler chickens 315,317. It has been suggested that bacteriocins produced by lactic acid bacteria, such as Lactobacillus and Enterococcus could inhibit the growth of Campylobacter in broiler chickens, and as the result, a number of novel bacteriocins isolated from lactic acid bacteria has been discovered 175,300,319,320. This finding suggests that Lactobacillus and Campylobacter may be competing for the same ecological niche in the intestinal track of broiler chickens. Likewise, Parabacteroides, another genus that was found to be less abundant in the positive samples compared to the negative sample, was found to be in a high abundance in broiler chickens fed with prebiotics together with Lactobacillus

321. Additionally, Acinetobacter, another ASV that was less abundant in the positive

103

sample compared to the negative samples, was found to be capable of growing in conditions that are suitable for Campylobacter 322. A study by Fernando et al. was able to recover Acinetobacter baumanii from agricultural niches known to be suitable for

Campylobacter 322. Previously, Acinetobacter spp. was thought to be strictly aerobe and rarely found outside hospital settings, however, Fernando et al. showed that A. baumanii can grow under microaerophilic conditions at 42ºC on Campylobacter-specific selective growth media, which typically contains several antibiotics Campylobacter is intrinsically resistant to 322. It appears that because these bacterial taxa are competing for the shared ecological niche in broiler chickens, a decrease in the abundance of one taxon leads to the others to occupy the available niche.

Based on our results and results from the previous studies, it is plausible that decreased abundance of Lactobacillus is associated with the presence of Campylobacter in broiler chickens. However, this conclusion should be approached with caution as the temporality of this observation is unclear. It is unclear whether the colonization of Campylobacter in broiler chickens leads to a decrease in the abundance of Lactobacillus or a decrease in the abundance in Lactobacillus for unknown reasons leads to an increased chance of colonization with Campylobacter in broiler chickens. Future studies should consider investigating interactions between Campylobacter and other commensal organisms found in broiler chickens, such as Lactobacillus and Parabacteroides, to better understanding of how Campylobacter colonization in the broiler chicken affects both the GI and litter microbiome.

104

Table 4.1. Descriptions of the commercial broiler chicken farms enrolled in the study. Sampling Antibiotic Litter condition Campylobacter Campylobacter FQ-R FQ-R season use status (Flock 1) status (Flock 2) Campylobacter Campylobacter status (Flock 1) status (Flock 2) Farm 1 Winter No Complete clean out Yes Yes No No (February – before the placement May) of flock 1 Farm 2 Winter No Complete clean out No Yes No No (February - before the placement June) of flock 1 Farm 3 Winter No Complete clean out No Yes No No (February – before the placement June) of flock 1 Farm 4 Winter No Complete clean out No No No No (February – before the placement Mary) of flock 1 Farm 5 Winter No Complete clean out Yes Yes No Yes (February – before the placement June) of flock 1 Farm 6 Summer Yes No No No No (June – September) Farm 7 Summer Yes Yes Yes No Yes (June – September)

105

Table 4.2. Summary of the final constrained model that best describes the variation observed in the bacterial composition of the broiler litter samples.

Final Model: Flock cycles + Age of birds + Studentized total number of reads | Farm ID Inertia Proportion (variance) Total 30810584 1 Conditioned 3695156 0.1199 Constrained 11766361 0.3819 Unconstrained 15349067 0.4982 Overall model significance DF Variance F-statistics P-value Model 8 11766361 7.0909 0.001 Residual 74 15349067 Significance of constraints included in the final model DF Variance F-statistics P-value Total number 1 2891500 13.9403 0.001 of reads Flock 1 3026983 14.5935 0.001 Sampling 6 4610579 3.7047 0.001 weeks Residual 74 15349067

106

Table 4.3. Results of the analysis of multivariate homogeneity of groups dispersions among farm variables. Farm DF Sum of Squares Mean Squares F-statistic P-value Groups 6 0.25567 0.042612 2.7671 0.01684 Residuals 82 1.26279 0.015400 Flock DF Sum of Squares Mean Squares F-statistic P-value Groups 1 0.34829 0.34829 75.199 2.11*10-13 Residuals 87 0.40294 0.00463 Age of birds (in week) DF Sum of Squares Mean Squares F-statistic P-value Groups 6 0.14745 0.0245756 4.1595 0.001067 Residuals 82 0.48448 0.0059083 Antibiotic use DF Sum of Squares Mean Squares F-statistic P-value Groups 1 0.22004 0.220044 18.623 4.197*10-5 Residuals 87 1.02795 0.011815 Campylobacter status DF Sum of Squares Mean Squares F-statistic P-value Groups 1 0.06608 0.066083 6.1856 0.01479 Residuals 87 0.92945 0.010683

107

Table 4.4. Summary of PERMANOVA based on Bray-Curtis dissimilarities showing the farm variables associated with the shifts in the bacterial composition in the broiler litter samples. Final Model DF Sum of Squares R2 Pseudo F-statistic P-value Studentized total 1 0.8512 0.05147 10.3522 0.001 number of reads Campylobacter 1 0.0598 0.00361 0.7268 0.697 status Farm:Flock 6 1.2697 0.07678 2.5736 0.001 Flock:Age of 12 3.7973 0.22962 3.8485 0.001 birds Residuals 61 5.0156 0.30330 Total 88 16.5371 1.0 AIC 199.5179

108

Figure 4.1. Sampling scheme. Seven broiler chicken farms carrying two consecutive flocks were sampled weekly. Within each broiler chicken house, the house was divided into three sections: front (brooding area), middle and back (non-brooding area). Each week, approximately five random grab litter samples were collected from each section of the house and pooled, resulting in three composite litter samples per sampling event. Prior to DNA extraction, litter wash from middle and back was pooled, resulting in two DNA samples (front – brooding; middle-back – non-brooding) per house per week.

Broiler chicken house Entrance

109

Figure 4.2. Estimated levels (logCUF/gram) of total and fluoroquinolone-resistant Campylobacter spp. in the broiler litter. The estimated logCFU/gram of (A) total and (B) fluoroquinolone-resistant Campylobacter spp. in broiler litter was measured by the MPN method. The prevalence of total and fluoroquinolone-resistant Campylobacter spp. was 15.7% (n=28) and 5.6% (n=10), respectively.

110

Figure 4.3. Comparison of alpha diversity of the broiler litter within and among farms. Alpha diversity was represented using the observed richness (number of ASVs) and Shannon’s diversity index. Multiple comparisons of alpha diversity was made using the Wilcoxon Rank Sum test. Pairwise comparisons of front (brooding) and middle-back(non-brooding) sections of the broiler farms over time using observed richness (A) and Shannon’s diversity index (B). Comparisons of observed richness (C) and Shannon’s diversity index (D) between flock 1 and flock within each farms. Comparisons of observed richness (E) and Shannon’s diversity index (F) between conventional and No-Antibiotic-Ever farms. Comparisons of observed richness (G) and Shannon’s diversity index (H) between Campylobacter-negative and Campylobacter-positive litter samples. * P-value ≤ 0.05, ** P-value ≤ 0.01, *** P-value ≤ .001.

111

Figure 4.4. Comparison of alpha diversity of the broiler litter by age of bird and farm type over two consecutive flock cycles. Alpha diversity measured by (A) observed number of ASVs and (B) Shannon’s diversity index was compared over time within each farm type (conventional and No-Antibiotic-Ever farms). Kruskal-Wallis Rank Sum test was performed to compare the weekly alpha diversity within each flock cycle and farm type.

112

Figure 4.5. Principal coordinate analysis plots of beta diversity measured by Jaccard and Bray-Curtis indices. The relative abundance of ASVs was normalized using cumulative sum scaling method with the scaling factor of 0.5. Jaccard similarity index was calculated to assess ASV membership among the litter samples. Bray-Curtis dissimilarity index was calculated to assess ASV composition among the samples. Plots depict (A) ASV membership by farms and flock cycles, (B) ASV composition by farms and flock cycles, (C) ASV membership by age of birds and flock cycles and (D) ASV composition by age of birds in weeks and flock cycles. Each sample was labeled by flock cycle (1 – first flock, 2 – second flock).

113

Figure 4.6. Constrained redundancy analysis ordination plots showing the effect of flock cycles and age of bird after controlling for the farm effects. The plots depict redundancy analysis ordination plots of the broiler litter samples based on Bray- Curtis dissimilarities on the relative abundance of ASV colored by (A) farm ID, (B) flock cycle and (C) age of bird. Forward model selection with 999 permutations was performed to identify the variables that best describes the variation observed in the bacterial composition in the samples. The variables included in the model that was used to generate ordination plots were flock cycles, sampling weeks and studentized total number of reads conditioned by farm IDs.

114

Figure 4.7. Changes in the microbial composition of the broiler litter at the phylum taxonomic level over time. The relative abundance of major phyla (relative abundance > 1%) detected in the litter samples is presented in stacked bar plots across age of bird in weeks by flock cycle within each farm type.

115

Figure 4.8. Changes in the relative abundance of families belong to Firmicutes over time. The relative abundance of Firmicutes families that were detected among all samples (prevalence = 100%) is presented in stacked bar plots across age of bird in weeks by flock cycle within each farm type.

116

Figure 4.9. Top 50 most abundant ASVs among farms across time. A heatmap of the log2 transformed relative abundance of the top 50 most abundant ASVs classified to the genus taxonomic level is shown. Darker shades of red represent a higher relative abundance and white color represents absence of the particular ASV. Samples are ordered by farm ID, flock cycle and age of bird in week. Names of genus listed more than once represent multiple ASVs classified to that genus. NA on the y-axis indicates ASVs that could not be assigned to the genus taxonomic level.

117

Figure 4.10. Bacterial genera differentially abundant between two flock cycles within each farm type. DESeq2 differential abundance testing was performed on the top 300 most abundant ASVs to identify bacterial genera that were statistically differentially abundant (alpha < 0.01) between the first and the second flocks within (A) NAE farms and (B) conventional farms. Each circle represents ASV at the genus level with color denoting the phylum to which the ASV is classified to. Multiple ASVs belonging to

118

Figure 4.11. Bacterial genera differentially abundant in NAE farms relative to conventional farms. DESeq2 differential abundance testing was performed on the top 300 most abundant ASVs to identify bacterial genera that were statistically differentially abundant (alpha < 0.01) among the litter samples from the NAE farms relative to the litter samples from the conventional farms over all time points. Each circle represents ASV at the genus level with color denoting the phylum to which the ASV is classified to.

119

Figure 4.12. Plot summarizing the differences observed in the bacterial composition between Campylobacter-positive and Campylobacter-negative litter samples. Because Campylobacter was not detected until week 3 of bird age, litter samples collected in week 1 and 2 of bird age were excluded from the analysis. Plots depict (A) the differences in the relative abundance of phyla (relative abundance > 1%) compared using the Kruskalis-Wallis Rank Sum test between Campylobacter-positive and Campylobacter-negative litter samples and (B) differentially abundant ASVs (alpha < 0.01) in Campylobacter-positive samples compared to the negative samples using DESeq2. Each circle represents ASV at the genus level with color denoting the phylum to which the ASV is classified to.

120

Supplementary Table 4.1. Results of PERMANOVA analyses showing all models tested prior to picking the best model. Age represents age of bird categorized into weeks (week 1 – 7). AB represents antibiotic use (Yes or No). Farm represents farm ID (1-7). Flock represents flock cycles (1 or 2). CampyStatus represents Campylobacter status of a sample (Yes or No). TotalReadsZ represents studentized total number of reads. Model 1 ~Age + AB + Farm + Flock + CampyStatus + TotalReadsZ DF Sum of Squares R2 Pseudo F- P-value statistic Age 6 1.9678 0.119 2.9338 0.000999 AB 0 0 0 Farm 5 1.1020 0.06664 1.8715 0.001998 Flock 1 1.6464 0.09956 14.7270 0.000999 CampyStatus 1 0.0623 0.00377 0.5571 0.865135 TotalReadsZ 1 1.0108 0.06112 9.0418 0.000999 Residuals 73 7.3785 0.44618 Total 88 16.5371 1.0 AIC 218.8424 Model 2 ~Age + AB + Flock + CampyStatus + TotalReadsZ, strata=Farm DF Sum of Squares R2 Pseudo F- P-value statistic Age 6 2.0046 0.12122 2.8134 0.001 AB 1 1.7056 0.10314 14.3625 0.001 Flock 1 1.6711 0.10105 14.0718 0.001 CampyStatus 1 0.1221 0.00738 1.0280 0.346 TotalReadsZ 1 1.1320 0.06845 9.5323 0.001 Residuals 78 9.2629 0.56013 Total 88 16.5371 1.0

121

AIC 220.1152 Model 3 ~ AB + Flock/Farm + Flock/Age + CampyStatus + TotalReadsZ, strata=Farm DF Sum of Squares R2 Pseudo F- P-value statistic AB 0 0.0 0.0 Farm:Flock 6 1.2697 0.07678 2.5736 0.001 Flock:Age 12 3.7973 0.22962 3.8485 0.001 CampyStatus 1 0.0598 0.00361 0.7268 0.676 TotalReadsZ 1 0.8512 0.05147 10.3522 0.001 Residuals 61 5.0156 0.30330 Total 88 16.5371 1.0 AIC 199.5179 Model 4 ~ Flock/Farm + Age/Flock + CampyStatus + TotalReadsZ, strata=Farm + Flock DF Sum of Squares R2 F-statistic P-value Farm:Flock 6 1.2697 0.07678 2.5736 0.001 Flock:Age 12 3.7973 0.22962 3.8485 0.001 CampyStatus 1 0.0598 0.00361 0.7268 0.676 TotalReadsZ 1 0.8512 0.05147 10.3522 0.001 Residuals 61 5.0156 0.30330 Total 88 16.5371 1.0 AIC 199.5179 Model 5 ~ Flock/Farm + Age/Flock + TotalReadsZ, strata=Farm + Flock DF Sum of Squares R2 Pseudo F- P-value statistic Farm:Flock 6 1.2744 0.24410 4.1093 0.001

122

Flock:Age 12 4.0367 0.24410 4.1093 0.001 TotalReadsZ 1 0.8555 0.05173 10.4503 0.001 Residuals 62 5.0754 0.30691 Total 88 16.5371 1.0 AIC 198.572 Model 6 ~ CampyStatus + Age/Flock + Flock/Farm + TotalReadsZ, strata=Farm + Flock DF Sum of Squares R2 F-statistic P-value CampyStatus 1 0.0598 0.00361 0.7268 0.681 Farm:Flock 6 1.2697 0.07678 .5736 0.001 Flock:Age 12 3.7973 0.22962 3.8485 0.001 TotalReadsZ 1 0.8512 0.05147 10.3522 0.001 Residuals 61 5.0156 0.30330 Total 88 16.5371 1.0 AIC 199.5179

123

Supplementary Table 4.2. Results of DESeq2 differential abundance testing showing ASVs that were differentially abundant in flock 2 relative to flock 1 within No-Antibiotic-Ever and conventional farms.

No Antibiotic Ever farms (Flock 2 vs. Flock 1) Log 2 fold Standard Adjusted Change Phylum Class Order Family Genus change error P-value SV145 -30 3.003303 2.55E-21 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV373 -9.35458 2.341941 0.00024 Decrease Proteobacteria Gammaproteobacteria Betaproteobacteriales Burkholderiaceae NA SV348 -8.33669 1.409616 3.23E-08 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Pseudomonadaceae Pseudomonas SV256 -6.78687 1.290437 9.23E-07 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Pseudomonadaceae Pseudomonas SV162 -6.55932 1.006476 9.83E-10 Decrease Proteobacteria Gammaproteobacteria Betaproteobacteriales Burkholderiaceae Comamonas SV238 -6.42329 1.119958 8.34E-08 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV100 -6.28544 0.914291 1.17E-10 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Pseudomonadaceae Pseudomonas SV146 -6.06405 1.229953 4.48E-06 Decrease Proteobacteria Gammaproteobacteria Xanthomonadales Xanthomonadaceae Stenotrophomonas SV32 -5.13265 0.482366 5.79E-24 Decrease Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae Klebsiella SV246 -5.09142 0.598109 7.29E-16 Decrease Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae Klebsiella SV95 -4.69327 0.906515 1.41E-06 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV245 -4.37983 0.897486 5.48E-06 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV122 -4.26828 1.450791 0.007702 Decrease Actinobacteria Actinobacteria Bifidobacteriales Bifidobacteriaceae Bifidobacterium SV258 -4.00753 0.816373 4.91E-06 Decrease Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae Sphingobacterium SV273 -3.9815 0.753555 8.26E-07 Decrease Proteobacteria Gammaproteobacteria Enterobacteriales Enterobacteriaceae Escherichia/Shigella SV123 -3.00659 0.598545 2.93E-06 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV268 -2.93769 0.863121 0.001865 Decrease Actinobacteria Actinobacteria Micrococcales Micrococcaceae Glutamicibacter SV210 -2.22455 0.777729 0.009425 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Psychrobacter SV48 -2.21779 0.742147 0.006786 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV266 -2.20654 0.373523 3.26E-08 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Flavonifractor SV105 -2.13522 0.336752 2.75E-09 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Pygmaiobacter SV382 -2.03642 0.515626 0.000287 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV41 -2.0223 0.599914 0.002081 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter

124

SV216 -1.84206 0.335021 2.95E-07 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV76 -1.61426 0.26871 2.09E-08 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV141 -1.56014 0.260642 2.31E-08 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV144 -1.50434 0.34718 6.21E-05 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV310 -1.47714 0.385966 0.000442 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Pygmaiobacter SV150 -1.42856 0.343729 0.000123 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV323 -1.39614 0.481204 0.008509 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium SV177 -1.37614 0.324233 9.01E-05 Decrease Actinobacteria Coriobacteriia Coriobacteriales Eggerthellaceae CHKCI002 SV116 -1.35038 0.266327 2.43E-06 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Flavonifractor SV260 -1.29195 0.355162 0.000851 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae UBA1819 SV138 -1.23724 0.261644 1.11E-05 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae DTU089 SV104 -1.23529 0.180676 1.43E-10 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Butyricicoccus SV244 -1.21475 0.398514 0.005804 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Caproiciproducens SV117 -1.16985 0.183095 2.08E-09 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Butyricicoccus SV383 -1.15009 0.393744 0.008117 Decrease Firmicutes Clostridia Clostridiales Lachnospiraceae NA SV322 -1.11771 0.285024 0.000314 Decrease Firmicutes Clostridia Clostridiales Lachnospiraceae NA SV102 -1.11689 0.188679 3.23E-08 Decrease Firmicutes Erysipelotrichia Erysipelotrichales Erysipelotrichaceae Erysipelatoclostridiu m SV103 -1.11476 0.263128 9.08E-05 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV221 -1.09841 0.370269 0.007229 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Pseudoflavonifracto r SV278 -1.03457 0.35454 0.008128 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV285 -1.03338 0.2854 0.000899 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_ UCG-014 SV265 -1.02555 0.268146 0.000442 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV74 -0.96843 0.146081 5.62E-10 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV173 -0.95425 0.264518 0.000937 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV113 -0.94301 0.214461 4.70E-05 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV20 -0.93022 0.167794 2.34E-07 Decrease Firmicutes Clostridia Clostridiales Lachnospiraceae CHKCI001

125

SV253 -0.90321 0.311575 0.008513 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Oscillibacter SV277 -0.89908 0.268613 0.002227 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Oscillibacter SV185 -0.84524 0.236854 0.001077 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Butyricicoccus SV299 -0.81833 0.276912 0.00744 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminiclostridium_ 5 SV101 -0.80683 0.220379 0.000785 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Butyricicoccus SV199 -0.79343 0.229858 0.001603 Decrease Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia SV143 -0.78144 0.212194 0.000729 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV263 -0.72104 0.238554 0.006266 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Oscillibacter SV203 -0.72044 0.227664 0.004125 Decrease Firmicutes Erysipelotrichia Erysipelotrichales Erysipelotrichaceae Erysipelatoclostridiu m SV152 -0.67226 0.175266 0.000437 Decrease Firmicutes Erysipelotrichia Erysipelotrichales Erysipelotrichaceae Erysipelatoclostridiu m SV142 -0.65232 0.183239 0.001102 Decrease Firmicutes Erysipelotrichia Erysipelotrichales Erysipelotrichaceae Merdibacter SV68 -0.59526 0.135144 4.61E-05 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV60 -0.59144 0.187352 0.004197 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV25 0.7573 0.229397 0.002601 Increase Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV158 0.76668 0.256326 0.006781 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae NA SV119 0.820464 0.164304 3.35E-06 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae NA SV39 0.834415 0.285039 0.008012 Increase Firmicutes Bacilli Lactobacillales Streptococcaceae Streptococcus SV118 0.842729 0.159156 8.05E-07 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae Sellimonas SV4 1.01221 0.214585 1.16E-05 Increase Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV129 1.261707 0.368914 0.001772 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium SV286 1.269117 0.438738 0.008617 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_ UCG-014 SV11 1.549842 0.307664 2.78E-06 Increase Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV5 1.707004 0.434101 0.000304 Increase Actinobacteria Actinobacteria Micrococcales Dermabacteraceae Brachybacterium SV251 1.726801 0.564514 0.005648 Increase Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV14 1.798492 0.420753 7.98E-05 Increase Actinobacteria Actinobacteria Corynebacteriales Corynebacteriaceae Corynebacterium_1 SV261 1.812437 0.539241 0.002137 Increase Firmicutes Bacilli Lactobacillales Carnobacteriaceae Jeotgalibaca

126

SV90 1.896161 0.601608 0.004233 Increase Bacteroidetes Bacteroidia Bacteroidales Rikenellaceae Alistipes SV182 1.978234 0.632486 0.004556 Increase Firmicutes Bacilli Lactobacillales Carnobacteriaceae Atopostipes SV15 2.037776 0.590684 0.001603 Increase Firmicutes Bacilli Lactobacillales Aerococcaceae Facklamia SV34 2.058534 0.71986 0.009425 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae Sphingobacterium SV288 2.175423 0.629208 0.001589 Increase Firmicutes Bacilli Bacillales Bacillaceae NA SV57 2.205193 0.712877 0.005074 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae NA SV12 2.24438 0.536267 0.000112 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Jeotgalicoccus SV35 2.257309 0.603882 0.000592 Increase Firmicutes Bacilli Bacillales Bacillaceae NA SV92 2.324792 0.710539 0.002862 Increase Proteobacteria Gammaproteobacteria Xanthomonadales Xanthomonadaceae Lysobacter SV271 2.351185 0.624124 0.000538 Increase Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV16 2.37417 0.50639 1.31E-05 Increase Actinobacteria Actinobacteria Micrococcales Dermabacteraceae Brachybacterium SV108 2.377707 0.572028 0.000123 Increase Firmicutes Bacilli Bacillales Bacillaceae NA SV255 2.387224 0.792357 0.006365 Increase Actinobacteria Actinobacteria Streptosporangiales Nocardiopsaceae Nocardiopsis SV33 2.387893 0.488561 5.37E-06 Increase Actinobacteria Actinobacteria Micrococcales Dermabacteraceae Brachybacterium SV9 2.444318 0.506503 7.09E-06 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Jeotgalicoccus SV287 2.446893 0.643485 0.000477 Increase Firmicutes Bacilli Bacillales Bacillaceae Oceanobacillus SV342 2.500109 0.590107 9.08E-05 Increase Firmicutes Bacilli Lactobacillales Carnobacteriaceae Alkalibacterium SV3 2.512148 0.459851 3.43E-07 Increase Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV130 2.525266 0.667747 0.000513 Increase Firmicutes Bacilli Bacillales Planococcaceae Sporosarcina SV72 2.637485 0.57756 2.25E-05 Increase Firmicutes Bacilli Bacillales Bacillaceae NA SV81 2.652819 0.646757 0.000154 Increase Actinobacteria Actinobacteria Corynebacteriales Corynebacteriaceae Corynebacterium_1 SV13 2.660449 0.752057 0.001188 Increase Proteobacteria Gammaproteobacteria Xanthomonadales Xanthomonadaceae NA SV69 2.700557 0.705859 0.000442 Increase Firmicutes Bacilli Bacillales Bacillaceae Pseudogracilibacillu s SV61 2.778637 0.577319 7.43E-06 Increase Actinobacteria Actinobacteria Streptosporangiales Nocardiopsaceae Nocardiopsis SV190 2.933901 0.973042 0.006365 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae NA SV21 3.044448 0.509976 2.46E-08 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus SV67 3.211492 0.688239 1.42E-05 Increase Actinobacteria Actinobacteria Corynebacteriales Dietziaceae Dietzia

127

SV166 3.211838 0.561361 8.80E-08 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV19 3.272691 0.441333 3.31E-12 Increase Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV37 3.375121 0.517956 9.83E-10 Increase Actinobacteria Actinobacteria Corynebacteriales Dietziaceae Dietzia SV17 3.376771 0.638059 8.05E-07 Increase Firmicutes Bacilli Lactobacillales Carnobacteriaceae Atopostipes SV38 3.456149 0.548833 3.50E-09 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV59 3.479224 0.606519 8.34E-08 Increase Firmicutes Bacilli Bacillales Bacillaceae Virgibacillus SV211 3.546027 0.649112 3.43E-07 Increase Firmicutes Bacilli Bacillales Bacillaceae Oceanobacillus SV223 3.571585 0.627925 1.04E-07 Increase Proteobacteria Alphaproteobacteria Rhodobacterales Rhodobacteraceae Oceaniovalibus SV242 3.588601 0.548919 9.38E-10 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae NA SV234 3.599856 0.792268 2.47E-05 Increase Proteobacteria Gammaproteobacteria Xanthomonadales Xanthomonadaceae NA SV131 3.631751 0.931167 0.000339 Increase Firmicutes Bacilli Bacillales Staphylococcaceae NA SV204 3.844688 0.919223 0.000112 Increase Deinococcus- Deinococci Deinococcales Trueperaceae Truepera Thermus SV319 3.852369 1.023721 0.000541 Increase Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Acinetobacter SV233 3.891474 0.715862 3.89E-07 Increase Firmicutes Clostridia Clostridiales Family_XI Gallicola SV178 3.954473 0.74311 7.18E-07 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV133 3.956822 0.552077 1.64E-11 Increase Actinobacteria Actinobacteria Micrococcales Ruaniaceae Haloactinobacteriu m SV6 4.007883 0.531155 1.35E-12 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus SV179 4.166972 0.941697 4.26E-05 Increase Proteobacteria Alphaproteobacteria Rickettsiales Mitochondria NA SV180 4.204019 0.715339 3.80E-08 Increase Firmicutes Bacilli Bacillales Bacillaceae NA SV93 4.403741 0.623904 3.37E-11 Increase Firmicutes Bacilli Bacillales Bacillaceae Pseudogracilibacillu s SV22 4.405053 0.609998 1.19E-11 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Salinicoccus SV132 4.428052 0.607089 7.53E-12 Increase Firmicutes Bacilli Bacillales Bacillaceae NA SV237 4.521312 0.519668 1.67E-16 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV326 4.626632 0.914376 2.52E-06 Increase Actinobacteria Actinobacteria Streptosporangiales Nocardiopsaceae Nocardiopsis SV157 4.737779 0.721132 7.95E-10 Increase Firmicutes Bacilli Lactobacillales Carnobacteriaceae Atopostipes SV84 4.748582 1.013417 1.31E-05 Increase Bacteroidetes Bacteroidia Flavobacteriales Flavobacteriaceae Salinimicrobium

128

SV58 4.794038 0.740888 1.27E-09 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae NA SV47 4.857284 0.639816 1.05E-12 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV27 5.18457 0.595974 1.67E-16 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Salinicoccus SV112 5.275281 0.629433 1.97E-15 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Enteractinococcus SV109 5.368887 1.083465 4.01E-06 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus SV289 5.812203 0.638279 6.41E-18 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Enteractinococcus SV44 6.505986 0.663016 9.93E-21 Increase Firmicutes Bacilli Lactobacillales Leuconostocaceae Weissella

Conventional Farms (Flock 2 vs. Flock 1) Log 2 fold Standard Adjusted Change Phylum Class Order Family Genus change error P-value SV367 -4.94502 1.331042 0.002343 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV236 -4.81104 1.184068 0.000845 Decrease Proteobacteria Gammaproteobacteria Betaproteobacteriale Burkholderiaceae Oligella s SV228 -4.54821 1.089895 0.000605 Decrease Proteobacteria Gammaproteobacteria Betaproteobacteriale Burkholderiaceae Oligella s SV56 -4.30654 1.062743 0.000845 Decrease Actinobacteria Actinobacteria Corynebacteriales Corynebacteriace Corynebacterium ae _1 SV79 -3.93652 0.791737 2.84E-05 Decrease Proteobacteria Gammaproteobacteria Betaproteobacteriale Burkholderiaceae Paenalcaligenes s SV148 -3.88861 0.855015 0.000176 Decrease Proteobacteria Gammaproteobacteria Pseudomonadales Moraxellaceae Psychrobacter SV485 -3.85507 1.012501 0.001829 Decrease Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriac Sphingobacteriu eae m SV188 -3.45962 0.865762 0.000966 Decrease Bacteroidetes Bacteroidia Flavobacteriales Flavobacteriaceae NA SV70 -2.98536 0.788167 0.001829 Decrease Proteobacteria Gammaproteobacteria Aeromonadales Aeromonadaceae Oceanimonas SV53 -2.75649 0.660779 0.000605 Decrease Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV13 -2.65329 0.674627 0.001144 Decrease Proteobacteria Gammaproteobacteria Xanthomonadales Xanthomonadace NA ae SV131 -2.59017 0.546091 7.89E-05 Decrease Firmicutes Bacilli Bacillales Staphylococcacea NA e SV57 -2.27362 0.566432 0.000943 Decrease Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriac NA eae

129

SV355 -1.46669 0.415212 0.00426 Decrease Firmicutes Bacilli Bacillales Bacillaceae NA SV35 -1.21403 0.235275 1.23E-05 Decrease Firmicutes Bacilli Bacillales Bacillaceae NA SV202 -1.16175 0.331868 0.004642 Decrease Firmicutes Bacilli Bacillales Bacillaceae Virgibacillus SV33 -1.06836 0.308684 0.005208 Decrease Actinobacteria Actinobacteria Micrococcales Dermabacteracea Brachybacterium e SV14 -0.94017 0.26217 0.003729 Decrease Actinobacteria Actinobacteria Corynebacteriales Corynebacteriace Corynebacterium ae _1 SV22 1.674794 0.27191 2.19E-07 Increase Firmicutes Bacilli Bacillales Staphylococcacea Salinicoccus e SV27 1.674902 0.298943 2.11E-06 Increase Firmicutes Bacilli Bacillales Staphylococcacea Salinicoccus e SV38 1.262642 0.278691 0.000176 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV84 2.167977 0.544404 0.000975 Increase Bacteroidetes Bacteroidia Flavobacteriales Flavobacteriaceae Salinimicrobium SV97 2.176248 0.668064 0.009916 Increase Actinobacteria Actinobacteria Streptosporangiales Nocardiopsaceae Nocardiopsis SV153 4.318825 0.974242 0.000232 Increase Proteobacteria Alphaproteobacteria Rickettsiales Mitochondria NA SV157 2.665891 0.514948 1.23E-05 Increase Firmicutes Bacilli Lactobacillales Carnobacteriacea Atopostipes e SV226 1.316809 0.403236 0.009916 Increase Actinobacteria Actinobacteria Micrococcales Ruaniaceae Ruania SV251 3.160272 0.60594 1.23E-05 Increase Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV267 2.264688 0.513298 0.000236 Increase Firmicutes Bacilli Lactobacillales Carnobacteriacea Alkalibacterium e SV279 2.944419 0.830543 0.004204 Increase Firmicutes Bacilli Lactobacillales Carnobacteriacea Alloiococcus e SV311 2.973822 0.660431 0.000183 Increase Firmicutes Bacilli Bacillales Staphylococcacea Salinicoccus e SV345 2.970573 0.527144 2.11E-06 Increase Actinobacteria Actinobacteria Micrococcales Intrasporangiacea NA e SV387 3.261507 0.861201 0.001829 Increase Firmicutes Bacilli Lactobacillales Carnobacteriacea Alloiococcus e SV431 3.28951 0.969944 0.006518 Increase Firmicutes Clostridia Clostridiales Family_XI Anaerococcus SV459 1.99927 0.481739 0.000623 Increase Actinobacteria Actinobacteria Micrococcales Bogoriellaceae Georgenia

130

Supplementary Table 4.3. Results of DESeq2 differential abundance testing showing ASVs that were differentially abundant in No-Antibiotic-Ever farms relative to the conventional farms over all time points.

Log 2 fold Standard Adjusted Change Phylum Class Order Family Genus change error p-value SV58 -8.107 0.946474 5.38E-16 Decrease Bacteroidetes Bacteroidia Sphingobacterial Sphingobacteriace NA es ae SV186 -8.00948 0.937551 5.61E-16 Decrease Bacteroidetes Bacteroidia Sphingobacterial Sphingobacteriace NA es ae SV249 -7.61816 1.012679 1.24E-12 Decrease Actinobacteria Actinobacteria Corynebacteriale Corynebacteriacea Corynebacterium_1 s e SV207 -7.48884 1.059908 2.53E-11 Decrease Firmicutes Clostridia Clostridiales Family_XI Anaerococcus SV292 -7.254 0.88837 9.60E-15 Decrease Actinobacteria Actinobacteria Corynebacteriale Corynebacteriacea Corynebacterium_1 s e SV84 -6.90997 0.820237 1.36E-15 Decrease Bacteroidetes Bacteroidia Flavobacteriales Flavobacteriaceae Salinimicrobium SV78 -6.60036 0.413374 6.50E-55 Decrease Firmicutes Clostridia Clostridiales Peptostreptococcac Romboutsia eae SV151 -6.53983 0.8379 1.49E-13 Decrease Firmicutes Bacilli Bacillales Staphylococcaceae Nosocomiicoccus SV70 -6.42795 1.373256 1.62E-05 Decrease Proteobacteria Gammaproteobact Aeromonadales Aeromonadaceae Oceanimonas eria SV77 -6.38208 0.887123 1.05E-11 Decrease Proteobacteria Gammaproteobact Xanthomonadale Xanthomonadacea NA eria s e SV231 -6.19139 0.91003 1.46E-10 Decrease Actinobacteria Actinobacteria Micrococcales Micrococcaceae Enteractinococcus SV293 -6.09586 0.944717 1.43E-09 Decrease Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides SV131 -5.9376 1.178873 3.23E-06 Decrease Firmicutes Bacilli Bacillales Staphylococcaceae NA SV73 -5.56328 0.63106 1.19E-16 Decrease Actinobacteria Actinobacteria Streptosporangia Nocardiopsaceae Nocardiopsis les SV61 -5.48312 0.624555 1.23E-16 Decrease Actinobacteria Actinobacteria Streptosporangia Nocardiopsaceae Nocardiopsis les SV301 -5.34231 0.583915 8.61E-18 Decrease Actinobacteria Actinobacteria Micrococcales Ruaniaceae Ruania SV133 -5.31407 0.723676 3.78E-12 Decrease Actinobacteria Actinobacteria Micrococcales Ruaniaceae Haloactinobacterium SV279 -5.0599 0.912618 2.53E-07 Decrease Firmicutes Bacilli Lactobacillales Carnobacteriaceae Alloiococcus SV227 -4.84938 0.89519 5.05E-07 Decrease Firmicutes Bacilli Bacillales Bacillaceae Pseudogracilibacillus SV153 -4.71635 1.038877 2.91E-05 Decrease Proteobacteria Alphaproteobacter Rickettsiales Mitochondria NA ia

131

SV267 -4.65147 0.83565 2.30E-07 Decrease Firmicutes Bacilli Lactobacillales Carnobacteriaceae Alkalibacterium SV22 -4.6227 0.804947 9.01E-08 Decrease Firmicutes Bacilli Bacillales Staphylococcaceae Salinicoccus SV269 -4.58807 0.527 1.89E-16 Decrease Firmicutes Bacilli Bacillales Bacillaceae Gracilibacillus SV92 -4.38651 0.852217 1.89E-06 Decrease Proteobacteria Gammaproteobact Xanthomonadale Xanthomonadacea Lysobacter eria s e SV180 -4.12221 1.191228 0.001668 Decrease Firmicutes Bacilli Bacillales Bacillaceae NA SV226 -3.96049 0.473108 1.90E-15 Decrease Actinobacteria Actinobacteria Micrococcales Ruaniaceae Ruania SV82 -3.93717 0.525625 1.37E-12 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV26 -3.85866 1.186349 0.003299 Decrease Bacteroidetes Bacteroidia Sphingobacterial Sphingobacteriace NA es ae SV255 -3.84486 1.066363 0.001074 Decrease Actinobacteria Actinobacteria Streptosporangia Nocardiopsaceae Nocardiopsis les SV17 -3.83996 0.973851 0.000316 Decrease Firmicutes Bacilli Lactobacillales Carnobacteriaceae Atopostipes SV211 -3.61246 0.901469 0.00026 Decrease Firmicutes Bacilli Bacillales Bacillaceae Oceanobacillus SV183 -3.57883 0.76872 1.79E-05 Decrease Actinobacteria Actinobacteria Corynebacteriale Corynebacteriacea Corynebacterium_1 s e SV67 -3.53232 1.094191 0.003525 Decrease Actinobacteria Actinobacteria Corynebacteriale Dietziaceae Dietzia s SV202 -3.48958 0.612325 1.13E-07 Decrease Firmicutes Bacilli Bacillales Bacillaceae Virgibacillus SV200 -3.45717 1.016394 0.002031 Decrease Firmicutes Bacilli Bacillales Staphylococcaceae Aliicoccus SV190 -3.43943 1.075122 0.003865 Decrease Bacteroidetes Bacteroidia Sphingobacterial Sphingobacteriace NA es ae SV16 -3.39479 0.648706 1.25E-06 Decrease Actinobacteria Actinobacteria Micrococcales Dermabacteraceae Brachybacterium SV93 -3.38119 0.829472 0.000199 Decrease Firmicutes Bacilli Bacillales Bacillaceae Pseudogracilibacillus SV147 -3.36322 0.717131 1.58E-05 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV97 -3.35084 0.620785 5.33E-07 Decrease Actinobacteria Actinobacteria Streptosporangia Nocardiopsaceae Nocardiopsis les SV72 -3.34123 0.732033 2.64E-05 Decrease Firmicutes Bacilli Bacillales Bacillaceae NA SV27 -3.31305 0.891296 0.000737 Decrease Firmicutes Bacilli Bacillales Staphylococcaceae Salinicoccus SV272 -3.31199 0.770925 8.41E-05 Decrease Firmicutes Bacilli Bacillales Bacillaceae Virgibacillus SV38 -3.28671 0.655486 3.55E-06 Decrease Actinobacteria Actinobacteria Micrococcales Micrococcaceae Yaniella SV132 -3.24946 0.890324 0.000926 Decrease Firmicutes Bacilli Bacillales Bacillaceae NA

132

SV10 -2.99468 0.945726 0.004114 Decrease Proteobacteria Gammaproteobact Oceanospirillales Halomonadaceae Halomonas eria SV59 -2.93699 1.01228 0.008509 Decrease Firmicutes Bacilli Bacillales Bacillaceae Virgibacillus SV137 -2.90857 0.815693 0.001196 Decrease Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides SV37 -2.70647 0.858219 0.004244 Decrease Actinobacteria Actinobacteria Corynebacteriale Dietziaceae Dietzia s SV157 -2.70272 0.848158 0.003986 Decrease Firmicutes Bacilli Lactobacillales Carnobacteriaceae Atopostipes SV223 -2.60547 0.859294 0.005829 Decrease Proteobacteria Alphaproteobacter Rhodobacterales Rhodobacteraceae Oceaniovalibus ia SV91 -2.55603 0.340563 1.31E-12 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV19 -2.25522 0.710969 0.004094 Decrease Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV294 -2.09923 0.571732 0.00086 Decrease Firmicutes Clostridia Clostridiales Clostridiaceae_1 Clostridium_sensu_stricto _1 SV3 -1.97543 0.610967 0.003496 Decrease Actinobacteria Actinobacteria Micrococcales Brevibacteriaceae Brevibacterium SV52 -1.89936 0.628437 0.005972 Decrease Bacteroidetes Bacteroidia Bacteroidales Bacteroidaceae Bacteroides SV11 -1.65953 0.463281 0.001149 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV8 -1.41775 0.291426 7.01E-06 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV64 -1.2763 0.376708 0.002112 Decrease Firmicutes Bacilli Lactobacillales Enterococcaceae Enterococcus SV170 -1.19335 0.382961 0.004731 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_UCG- 007 SV129 -1.07654 0.345698 0.004731 Decrease Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium SV25 -1.00643 0.277642 0.001008 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV118 -0.77412 0.186932 0.000152 Decrease Firmicutes Clostridia Clostridiales Lachnospiraceae Sellimonas SV155 0.72878 0.203063 0.001132 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae Shuttleworthia SV117 0.774305 0.25441 0.005657 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Butyricicoccus SV192 0.874029 0.236059 0.000771 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminiclostridium_5 SV60 0.952292 0.228172 0.000138 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV201 0.965828 0.316857 0.005616 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_UCG- 014 SV171 1.013195 0.302738 0.002405 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae Lachnospiraceae_NC2004 _group SV143 1.035169 0.351209 0.00751 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella

133

SV173 1.051667 0.342933 0.005411 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV55 1.058144 0.254541 0.000147 Increase Firmicutes Bacilli Lactobacillales Enterococcaceae Enterococcus SV20 1.066783 0.282239 0.000582 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae CHKCI001 SV217 1.088419 0.323125 0.002246 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae NA SV142 1.110244 0.33351 0.002539 Increase Firmicutes Erysipelotrichia Erysipelotrichale Erysipelotrichacea Merdibacter s e SV113 1.120338 0.351771 0.003986 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV140 1.178676 0.400906 0.007573 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_UCG- 014 SV203 1.199711 0.381874 0.004383 Increase Firmicutes Erysipelotrichia Erysipelotrichale Erysipelotrichacea Erysipelatoclostridium s e SV49 1.216697 0.243942 3.99E-06 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV176 1.392454 0.454955 0.005476 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_UCG- 014 SV175 1.472836 0.372928 0.000314 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_UCG- 014 SV144 1.490125 0.421478 0.001299 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV177 1.505421 0.381204 0.000314 Increase Actinobacteria Coriobacteriia Coriobacteriales Eggerthellaceae CHKCI002 SV265 1.548 0.392818 0.000316 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV209 1.614455 0.350113 2.14E-05 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae Lachnospiraceae_NC2004 _group SV285 1.64205 0.431874 0.000545 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Ruminococcaceae_UCG- 014 SV260 1.761185 0.492937 0.001177 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae UBA1819 SV105 1.785907 0.603573 0.007293 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Pygmaiobacter SV74 1.813244 0.335607 5.32E-07 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV103 1.86371 0.403466 2.10E-05 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae NA SV50 1.915197 0.370021 1.66E-06 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium SV278 2.503835 0.731754 0.001905 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Fournierella SV139 2.521477 0.416132 1.52E-08 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV266 2.588025 0.72646 0.001198 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Flavonifractor SV136 2.603443 0.55078 1.34E-05 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium

134

SV125 2.653045 0.361463 3.78E-12 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV88 2.76171 0.939226 0.007573 Increase Firmicutes Bacilli Lactobacillales Carnobacteriaceae Jeotgalibaca SV300 2.933812 0.721568 0.000205 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV262 3.063057 0.691696 4.83E-05 Increase Firmicutes Clostridia Clostridiales Lachnospiraceae Blautia SV43 3.119665 1.01196 0.00517 Increase Proteobacteria Gammaproteobact Pseudomonadale Pseudomonadacea Pseudomonas eria s e SV87 3.185173 0.399872 4.49E-14 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Subdoligranulum SV107 3.271753 0.575129 1.16E-07 Increase Proteobacteria Alphaproteobacter Rickettsiales Mitochondria NA ia SV109 3.488966 0.983725 0.001258 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus SV241 3.550536 1.121722 0.004114 Increase Proteobacteria Gammaproteobact Pseudomonadale Pseudomonadacea Pseudomonas eria s e SV283 3.593649 1.24204 0.008663 Increase Actinobacteria Actinobacteria Micrococcales Dermabacteraceae Brachybacterium SV56 3.692604 0.878308 0.000123 Increase Actinobacteria Actinobacteria Corynebacteriale Corynebacteriacea Corynebacterium_1 s e SV160 3.705896 1.168382 0.004094 Increase Firmicutes Bacilli Lactobacillales Enterococcaceae Enterococcus SV235 3.781162 0.995616 0.000547 Increase Proteobacteria Gammaproteobact Enterobacteriales Enterobacteriaceae Providencia eria SV121 3.836292 1.105011 0.001616 Increase Proteobacteria Gammaproteobact Betaproteobacter Burkholderiaceae Alcaligenes eria iales SV95 4.249251 1.080944 0.000325 Increase Proteobacteria Gammaproteobact Pseudomonadale Moraxellaceae Acinetobacter eria s SV32 4.278504 0.863287 4.59E-06 Increase Proteobacteria Gammaproteobact Enterobacteriales Enterobacteriaceae Klebsiella eria SV134 4.449829 1.458493 0.005609 Increase Proteobacteria Gammaproteobact Cardiobacteriales Wohlfahrtiimonad Wohlfahrtiimonas eria aceae SV123 4.619622 1.069355 7.67E-05 Increase Proteobacteria Gammaproteobact Pseudomonadale Moraxellaceae Acinetobacter eria s SV246 4.632869 1.316316 0.001365 Increase Proteobacteria Gammaproteobact Enterobacteriales Enterobacteriaceae Klebsiella eria SV259 4.695043 1.513246 0.004877 Increase Proteobacteria Gammaproteobact Cardiobacteriales Wohlfahrtiimonad Ignatzschineria eria aceae SV24 4.741467 1.192363 0.000291 Increase Bacteroidetes Bacteroidia Sphingobacterial Sphingobacteriace Sphingobacterium es ae SV205 4.895341 0.849743 8.36E-08 Increase Proteobacteria Alphaproteobacter Rickettsiales Mitochondria NA ia

135

SV48 5.246157 1.326621 0.000314 Increase Proteobacteria Gammaproteobact Pseudomonadale Moraxellaceae Acinetobacter eria s SV45 5.333593 1.128074 1.34E-05 Increase Proteobacteria Gammaproteobact Enterobacteriales Enterobacteriaceae NA eria SV30 5.3387 1.288463 0.000152 Increase Proteobacteria Gammaproteobact Aeromonadales Aeromonadaceae Oceanisphaera eria SV257 5.41384 0.9302 6.08E-08 Increase Firmicutes Clostridia Clostridiales Ruminococcaceae Faecalibacterium SV110 5.810629 1.377545 0.000117 Increase Bacteroidetes Bacteroidia Bacteroidales Rikenellaceae Alistipes SV187 5.881209 0.989867 3.03E-08 Increase Firmicutes Bacilli Lactobacillales Leuconostocaceae Weissella SV268 5.896778 1.200777 5.67E-06 Increase Actinobacteria Actinobacteria Micrococcales Micrococcaceae Glutamicibacter SV254 6.153797 1.408055 6.20E-05 Increase Proteobacteria Gammaproteobact Cardiobacteriales Wohlfahrtiimonad Ignatzschineria eria aceae SV169 6.263384 0.89914 4.89E-11 Increase Actinobacteria Actinobacteria Bifidobacteriales Bifidobacteriaceae Bifidobacterium SV127 6.408976 1.011704 2.97E-09 Increase Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV122 6.711381 1.328763 3.07E-06 Increase Actinobacteria Actinobacteria Bifidobacteriales Bifidobacteriaceae Bifidobacterium SV41 6.944896 1.125618 7.89E-09 Increase Proteobacteria Gammaproteobact Pseudomonadale Moraxellaceae Acinetobacter eria s SV28 7.445912 1.096443 1.52E-10 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus SV162 12.29603 1.952403 3.62E-09 Increase Proteobacteria Gammaproteobact Betaproteobacter Burkholderiaceae Comamonas eria iales SV145 19.60809 3.644646 5.73E-07 Increase Proteobacteria Gammaproteobact Pseudomonadale Moraxellaceae Acinetobacter eria s

136

Log 2 fold Standar Adjusted Change Phylum Class Order Family Genus change d error p-value SV1580 -29.9985 2.844431 7.93E-23 Decrease Proteobacteria Gammaproteo Pseudomonadales Moraxellaceae Acinetobacter bacteria SV456 -29.9954 2.550055 1.83E-28 Decrease Bacteroidetes Bacteroidia Flavobacteriales Flavobacteriaceae Myroides SV2085 -29.9927 3.502938 3.70E-15 Decrease Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae Parapedobacter SV1897 -29.9886 3.500121 3.70E-15 Decrease Proteobacteria Gammaproteo Gammaproteobacteria Unknown_Family Candidatus_Berkiell bacteria _Incertae_Sedis a SV361 -25.1421 2.868213 1.39E-15 Decrease Bacteroidetes Bacteroidia Bacteroidales Barnesiellaceae Barnesiella SV579 -22.8964 3.502624 1.57E-08 Decrease Bacteroidetes Bacteroidia Flavobacteriales Weeksellaceae Empedobacter SV373 -21.9738 2.561385 3.70E-15 Decrease Proteobacteria Gammaproteo Betaproteobacteriales Burkholderiaceae NA bacteria SV314 -20.6083 3.372142 2.12E-07 Decrease Actinobacteri Coriobacteriia Coriobacteriales Atopobiaceae Olsenella a SV1858 -20.2835 2.961158 2.02E-09 Decrease Proteobacteria Gammaproteo Betaproteobacteriales Burkholderiaceae Parapusillimonas bacteria SV1466 -18.4679 2.646597 8.99E-10 Decrease Bacteroidetes Bacteroidia Bacteroidales Dysgonomonadaceae Fermentimonas SV1644 -18.0035 3.502448 4.58E-05 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV1650 -17.4018 1.959991 6.79E-16 Decrease Proteobacteria Gammaproteo Cellvibrionales Cellvibrionaceae Cellvibrio bacteria SV145 -16.8472 3.503736 0.000241 Decrease Proteobacteria Gammaproteo Pseudomonadales Moraxellaceae Acinetobacter bacteria SV1879 -16.2764 2.537539 3.27E-08 Decrease Proteobacteria Gammaproteo Betaproteobacteriales Burkholderiaceae Lampropedia bacteria SV904 -14.3866 3.501567 0.004982 Decrease Bacteroidetes Bacteroidia Bacteroidales Tannerellaceae Parabacteroides SV989 -14.0145 3.177415 0.001405 Decrease Bacteroidetes Bacteroidia Flavobacteriales Weeksellaceae Empedobacter SV357 -4.42164 0.945304 0.000436 Decrease Firmicutes Bacilli Lactobacillales Lactobacillaceae Lactobacillus SV109 5.125339 1.140506 0.001 Increase Firmicutes Bacilli Bacillales Staphylococcaceae Staphylococcus SV2087 15.4175 3.502872 0.001405 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae Sphingobacterium SV1821 18.91899 3.499101 1.13E-05 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae NA SV1402 19.93146 3.501345 2.35E-06 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae NA

137

SV1500 21.2829 3.496338 2.30E-07 Increase Bacteroidetes Bacteroidia Sphingobacteriales Sphingobacteriaceae NA SV1898 30 3.503229 3.70E-15 Increase Proteobacteria Gammaproteo Xanthomonadales Rhodanobacteraceae Pseudofulvimonas bacteria SV2158 30 3.503284 3.70E-15 Increase Bacteroidetes Bacteroidia Flavobacteriales Weeksellaceae NA

138

Supplementary Figure 4.1. Principal coordinate analysis ordination plots based on Jaccard similarities and Bray-Curtis dissimilarities showing the ASV membership and composition between front and middle-back sections of the broiler houses. The relative abundance of ASVs was normalized using cumulative sum scaling method with scaling factor of 0.5. Plot depicts (A) ASV membership measured by Jaccard and (B) ASV composition measured by Bray-Curtis with colors denoting the two farm sections.

139

Supplementary Figure 4.2. Principal coordinate analysis ordination plots showing dispersions of farm variables tested in PERMANOVA analysis. Plots depict principal coordinate analysis ordination plots by (A) farm ID, (B) age of bird, (C) flock cycle, (D) antibiotic use and (E) Campylobacter status with respect to group centroids. Ellipses drawn around each group centroids represent on standard deviation of the data.

140

Chapter Five: Conclusion

Campylobacteriosis, primarily caused by C. jejuni, is one of the most commonly reported foodborne diseases in the U.S. With campylobacteriosis as a leading foodborne illness, even surpassing salmonellosis, there is a need to improve our understanding of

Campylobacter in broiler chickens, the major source of C. jejuni in humans, to control and prevent campylobacteriosis in humans. The four chapters of this dissertation focused on different aspects of ecology and epidemiology of C. jejuni in broiler chicken production system in the U.S. and to fill knowledge gaps exist in the field. In Chapter one, we reviewed public health burden of campylobacteriosis in humans and epidemiology of Campylobacter in broiler production chain with an emphasis on the emergence of antimicrobial resistant Campylobacter.

In Chapter two, we assessed the current Campylobacter interventions and control measures practiced by the U.S. broiler industry to reduce Campylobacter contamination on broiler chicken products as well as perceptions of Campylobacter in broiler chickens through an on-line survey of key industry stakeholders. We found that while the USDA-

FSIS recommended guidelines to reduce Campylobacter on broiler products are practiced widely by the establishments included in the survey, many voiced the lack of interventions to effectively reduce Campylobacter colonization on farms, demonstrating the need for the development of on-farm Campylobacter interventions. In addition, we found gaps in the understanding of Campylobacter epidemiology in broiler chickens among the participants, specifically regarding the mode of transmission and risk factors associated with the increased risk of Campylobacter contamination during the production

141

stages. These findings suggest that continued food safety training and education programs about Campylobacter in broiler production targeted to the industry stakeholders can lead to improvement in Campylobacter control and ultimately, to improvement in food safety.

In Chapter three, we explored putative indirect selective pressures favoring the persistence of fluoroquinolone-resistant (FQ-R) C. jejuni in poultry environment in the absence of the direct selective pressure of fluoroquinolone use. We tested two hypotheses, enhanced iron acquisition and regulation of FQ-R C. jejuni compared to susceptible C. jejuni and activity of microcin B17 produced from E. coli commensal to poultry. We observed that while there was a high variation in the relative expression levels of the genes that are involved in iron acquisition and regulation among the strains tested, no difference in the expression levels was observed between FQ-R and susceptible

C. jejuni strains. This high variation in the gene expression level may be due to a high frequency of DNA recombination observed in the Campylobacter genome, which may aid Campylobacter to survive both inside and outside the host species. In addition, we observed that regardless of the fluoroquinolone resistance phenotype, C. jejuni was sensitive to microcin B17. Nonetheless, we found that neither of the hypotheses explored in this chapter was likely to be responsible for the persistence of FQ-R C. jejuni in the poultry environment.

In Chapter four, we described the temporal changes in the bacterial microbiota of broiler litter and how the presence of Campylobacter in broiler farms affects the litter microbiota composition using composite litter samples collected from seven commercial broiler farms in the U.S. We looked at the effects of antibiotic use (conventional farms vs. no- antibiotic-ever farms), flock cycles, age of bird and presence of Campylobacter on farm

142

on the bacterial diversity and composition of broiler litter microbiota. We found that bacterial composition of broiler litter is significantly influenced by flock cycle and age of bird. Specifically, litter samples from the second flock cycle had a higher abundance of organic material decomposing bacteria, halotolerant and alkaliphilic bacteria, which are commonly found in reused and old litter, compared to the litter samples from the first flock cycle. While no significant difference in bacterial diversity and overall composition was observed between Campylobacter-positive and Campylobacter-negative samples, a lower abundance of Lactobacillus spp. was observed in the positive samples compared to the negative samples. This result, which corroborates previous studies, suggests a potential interaction between Campylobacter spp., and Lactobacillus spp. can be a useful tool in controlling Campylobacter in broiler chickens on farm.

Although ecology and epidemiology of Campylobacter in broiler chickens have been studied extensively in the last few decades, there are several key areas that need further research. First, selection pressures driving the emergence and persistence of FQ-R C. jejuni in broiler chickens need to be investigated. Previous work has raised the possibility of the existence of C. jejuni isolates that have a higher frequency of a spontaneous point mutation responsible for fluoroquinolone resistance in Campylobacter 323, which can explain a consistent 20-30% prevalence of FQ-R C. jejuni detected in retail products and chicken carcass samples. However, the molecular mechanisms that enable these mutator strains to persist in the absence of the direct selective pressure, fluoroquinolone use in broiler chickens, need to be identified. Comparison of transcriptomic profiles between

FQ-R and susceptible C. jejuni isolates from multiple stages of the broiler production chain can shed insights into mechanisms behind the increased fitness of the resistant

143

strains. Additionally, future research can expand our work to investigate the relationship between Campylobacter and Lactobacillus spp. in both broiler litter and broiler gut, especially focusing on the temporality and interactions between these two bacterial genera at the molecular level. These future studies may provide additional tools to combat Campylobacter colonization in broiler chickens and improve broiler health and performance. Taken together, this body of work has utilized multidisciplinary approaches, from qualitative survey to bioinformatics, to shed light on various aspects of ecology and epidemiology of Campylobacter in broiler chickens. This body of work will contribute to the collective efforts to mitigate Campylobacter contamination in broiler chicken products and to improve food safety and public health.

144

Bibliography 1. Skirrow MB. John McFadyean and the centenary of the first isolation of Campylobacter species. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2006;43(9):1213-1217. 2. Jones FS, Orcutt M, Little RB. Vibrios (VIBRIO JEJUNI, N.SP.) associated with intestinal disorders of corws and calves. The Journal of experimental medicine. 1931;53(6):853-863. 3. Doyle LP. A vibrio associated with swine dysentery. American Journal of Veterinary Research. 1944;5:3 - 5. 4. Acheson D, Allos BM. Campylobacter jejuni Infections: Update on Emerging Issues and Trends. Clinical Infectious Diseases. 2001;32(8):1201-1206. 5. Dekeyser P, Gossuin-Detrain M, Butzler JP, Sternon J. Acute enteritis due to related vibrio: first positive stool cultures. The Journal of infectious diseases. 1972;125(4):390-392. 6. Sebald M, Veron M. Base DNA Content and Classification of Vibrios. Annales de l'Institut Pasteur. 1963;105:897-910. 7. Véron M, Chatelain R. Taxonomic Study of the Genus Campylobacter Sebald and Véron and Designation of the Neotype Strain for the Type Species, Campylobacter fetus (Smith and Taylor) Sebald and Véron. International Journal of Systematic and Evolutionary Microbiology. 1973;23(2):122-134. 8. DOYLE MP, ROMAN DJ. Growth and Survival of Campylobacter fetus subsp. jejuni as a Function of Temperature and pH. Journal of Food Protection. 1981;44(8):596-601. 9. Hazeleger WC, Wouters JA, Rombouts FM, Abee T. Physiological activity of Campylobacter jejuni far below the minimal growth temperature. Applied and environmental microbiology. 1998;64(10):3917-3922. 10. Bolton FJ, Wareing DR, Sails AD. Comparison of a novel microaerobic system with three other gas-generating systems for the recovery of Campylobacter species from human faecal samples. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology. 1997;16(11):839-842. 11. Hofreuter D. Defining the metabolic requirements for the growth and colonization capacity of Campylobacter jejuni. Frontiers in cellular and infection microbiology. 2014;4:137-137. 12. Guccione E, Leon-Kempis Mdel R, Pearson BM, et al. Amino acid-dependent growth of Campylobacter jejuni: key roles for aspartase (AspA) under microaerobic and oxygen-limited conditions and identification of AspB (Cj0762), essential for growth on glutamate. Molecular microbiology. 2008;69(1):77-93. 13. Sheppard SK, Maiden MCJ. The evolution of Campylobacter jejuni and Campylobacter coli. Cold Spring Harbor perspectives in biology.7(8):a018119- a018119. 14. Wagenaar JA, Newell DG, Kalupahana RS, Mughini-Gras L. Campylobacter: Animal Reservoirs, Human Infections, and Options for Control. In: Sing A, ed. Zoonoses - Infections Affecting Humans and Animals: Focus on Public Health Aspects. Dordrecht: Springer Netherlands; 2015:159-177.

145

15. Smith S, Meade J, Gibbons J, McGill K, Bolton D, Whyte P. The impact of environmental conditions on Campylobacter jejuni survival in broiler faeces and litter. Infection ecology & epidemiology. 2016;6:31685-31685. 16. Rosenquist H, Boysen L, Knøchel S. Survival of Campylobacter jejuni in different gas mixtures. FEMS Microbiology Letters. 2007;266(2):152-157. 17. Line JE. Influence of relative humidity on transmission of Campylobacter jejuni in broiler chickens. Poultry Science. 2006;85(7):1145-1150. 18. Doyle MP, Roman DJ. Sensitivity of Campylobacter jejuni to Drying. Journal of Food Protection. 1982;45(6):507-510. 19. Waterman SC. The heat-sensitivity of Campylobacter jejuni in milk. The Journal of hygiene. 1982;88(3):529-533. 20. Klančnik A, Vučković D, Jamnik P, Abram M, Možina SS. Stress response and virulence of heat-stressed Campylobacter jejuni. Microbes and environments. 2014;29(4):338-345. 21. Hughes R-A, Hallett K, Cogan T, Enser M, Humphrey T. The Response of Campylobacter jejuni to Low Temperature Differs from That of Escherichia coli. Applied and Environmental Microbiology. 2009;75(19):6292-6298. 22. Rollins DM, Colwell RR. Viable but nonculturable stage of Campylobacter jejuni and its role in survival in the natural aquatic environment. Applied and environmental microbiology. 1986;52(3):531-538. 23. Hazeleger W, Arkesteijn C, Toorop-Bouma A, Beumer R. Detection of the coccoid form of Campylobacter jejuni in chicken products with the use of the polymerase chain reaction. International journal of food microbiology. 1994;24(1-2):273-281. 24. Tholozan JL, Cappelier JM, Tissier JP, Delattre G, Federighi M. Physiological characterization of viable-but-nonculturable Campylobacter jejuni cells. Applied and environmental microbiology. 1999;65(3):1110-1116. 25. Lázaro B, Cárcamo J, Audícana A, Perales I, Fernández-Astorga A. Viability and DNA maintenance in nonculturable spiral Campylobacter jejuni cells after long- term exposure to low temperatures. Applied and environmental microbiology. 1999;65(10):4677-4681. 26. Ziprin RL, Droleskey RE, Hume ME, Harvey RB. Failure of viable nonculturable Campylobacter jejuni to colonize the cecum of newly hatched leghorn chicks. Avian diseases. 2003;47(3):753-758. 27. Stern NJ, Jones DM, Wesley IV, Rollins DM. Colonization of chicks by nonculturable Campylobacter spp. Letters in Applied Microbiology. 1994;18(6):333- 336. 28. Medema GJ, Schets FM, van de Giessen AW, Havelaar AH. Lack of colonization of 1 day old chicks by viable, non-culturable Campylobacter jejuni. The Journal of applied bacteriology. 1992;72(6):512-516. 29. Cappelier JM, Magras C, Jouve JL, Federighi M. Recovery of viable but nonculturable Campylobacter jejuni cells in two animal models. Food Microbiology. 1999;16(4):375-383. 30. Reuter M, Mallett A, Pearson BM, van Vliet AHM. Biofilm Formation by Campylobacter jejuni Is Increased under Aerobic Conditions. Applied and Environmental Microbiology. 2010;76(7):2122-2128.

146

31. Brown HL, Reuter M, Salt LJ, Cross KL, Betts RP, van Vliet AHM. Chicken Juice Enhances Surface Attachment and Biofilm Formation of Campylobacter jejuni. Applied and Environmental Microbiology. 2014;80(22):7053-7060. 32. Ica T, Caner V, Istanbullu O, et al. Characterization of Mono- and Mixed-Culture Campylobacter jejuni Biofilms. Applied and Environmental Microbiology. 2012;78(4):1033-1038. 33. Teh AHT, Lee SM, Dykes GA. Does Campylobacter jejuni form biofilms in food- related environments? Applied and environmental microbiology. 2014;80(17):5154-5160. 34. Mead PS, Slutsker L, Dietz V, et al. Food-related illness and death in the United States. Emerging infectious diseases. 1999;5(5):607-625. 35. Boysen L, Rosenquist H, Larsson JT, et al. Source attribution of human campylobacteriosis in Denmark. Epidemiology and Infection. 2013;142(8):1599- 1608. 36. Silva J, Leite D, Fernandes M, Mena C, Gibbs PA, Teixeira P. Campylobacter spp. as a Foodborne Pathogen: A Review. Frontiers in microbiology. 2011;2:200- 200. 37. Montgomery MP, Robertson S, Koski L, et al. Multidrug-Resistant Campylobacter jejuni Outbreak Linked to Puppy Exposure - United States, 2016- 2018. MMWR Morbidity and mortality weekly report. 2018;67(37):1032-1035. 38. Evans MR, Roberts RJ, Ribeiro CD, Gardner D, Kembrey D. A milk-borne campylobacter outbreak following an educational farm visit. Epidemiol Infect. 1996;117(3):457-462. 39. Lahti E, Rehn M, Ockborn G, et al. Outbreak of Campylobacteriosis Following a Dairy Farm Visit: Confirmation by Genotyping. Foodborne pathogens and disease. 2017;14(6):326-332. 40. Helms M, Simonsen J, Mølbak K. Foodborne Bacterial Infection and Hospitalization: A Registry-Based Study. Clinical Infectious Diseases. 2006;42(4):498-506. 41. CDC. Questions and Answers: Campylobacter. 2017; https://www.cdc.gov/campylobacter/faq.html. Accessed 02/28, 2019. 42. CDC. Foodborne Diseases Active Surveillance Network (FoodNet): FoodNet 2015 Surveillance Report (Final Data). 2017; https://www.cdc.gov/foodnet/reports/annual-reports-2015.html. Accessed 02/28, 2019. 43. Facciolà A, Riso R, Avventuroso E, Visalli G, Delia SA, Laganà P. Campylobacter: from microbiology to prevention. Journal of preventive medicine and hygiene. 2017;58(2):E79-E92. 44. Sibanda N, McKenna A, Richmond A, et al. A Review of the Effect of Management Practices on Campylobacter Prevalence in Poultry Farms. Frontiers in microbiology. 2018;9:2002-2002. 45. Platts-Mills JA, Kosek M. Update on the burden of Campylobacter in developing countries. Current opinion in infectious diseases. 2014;27(5):444-450. 46. Kaakoush NO, Castaño-Rodríguez N, Mitchell HM, Man SM. Global Epidemiology of Campylobacter Infection. Clinical microbiology reviews. 2015;28(3):687-720.

147

47. Robinson DA. Infective dose of Campylobacter jejuni in milk. British medical journal (Clinical research ed). 1981;282(6276):1584-1584. 48. Blaser MJ, Clements ML, Levine MM, Black RE, Hughes TP. Experimental Campylobacter jejuni Infection in Humans. The Journal of infectious diseases. 1988;157(3):472-479. 49. Tribble DR, Baqar S, Scott DA, et al. Assessment of the duration of protection in Campylobacter jejuni experimental infection in humans. Infection and immunity. 2010;78(4):1750-1759. 50. Eiland LS, Jenkins LS. Optimal treatment of campylobacter dysentery. The journal of pediatric pharmacology and therapeutics : JPPT : the official journal of PPAG. 2008;13(3):170-174. 51. Pacanowski J, Lalande V, Lacombe K, et al. Campylobacter bacteremia: clinical features and factors associated with fatal outcome. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2008;47(6):790-796. 52. Allos BM. Campylobacter jejuni Infections: update on emerging issues and trends. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2001;32(8):1201-1206. 53. Domingues AR, Pires SM, Halasa T, Hald T. Source attribution of human campylobacteriosis using a meta-analysis of case-control studies of sporadic infections. Epidemiology and Infection. 2012;140(6):970-981. 54. Rodrigues LC, Cowden JM, Wheeler JG, et al. The study of infectious intestinal disease in England: risk factors for cases of infectious intestinal disease with Campylobacter jejuni infection. Epidemiol Infect. 2001;127(2):185-193. 55. Group ftEIPFW, Smith KE, Kassenborg HD, et al. Fluoroquinolone-Resistant Campylobacter Infections: Eating Poultry Outside of the Home and Foreign Travel Are Risk Factors. Clinical Infectious Diseases. 2004;38(Supplement_3):S279-S284. 56. Orr KE, Lightfoot NF, Sisson PR, et al. Direct milk excretion of Campylobacter jejuni in a dairy cow causing cases of human enteritis. Epidemiology and infection. 1995;114(1):15-24. 57. Campagnolo ER, Philipp LM, Long JM, Hanshaw NL. Pet-associated Campylobacteriosis: A persisting public health concern. Zoonoses and public health. 2018;65(3):304-311. 58. Mughini Gras L, Smid JH, Wagenaar JA, et al. Risk factors for campylobacteriosis of chicken, ruminant, and environmental origin: a combined case-control and source attribution analysis. PloS one. 2012;7(8):e42599-e42599. 59. Ravel A, Pintar K, Nesbitt A, Pollari F. Non food-related risk factors of campylobacteriosis in Canada: a matched case-control study. BMC public health. 2016;16(1):1016-1016. 60. Ricotta EE, Palmer A, Wymore K, et al. Epidemiology and Antimicrobial Resistance of International Travel-Associated Campylobacter Infections in the United States, 2005–2011. American Journal of Public Health. 2014;104(7):e108- e114. 61. Tustin J, Laberge K, Michel P, et al. A national epidemic of campylobacteriosis in Iceland, lessons learned. Zoonoses and public health. 2011;58(6):440-447.

148

62. Stern NJ, Hiett KL, Alfredsson GA, et al. Campylobacter spp. in Icelandic poultry operations and human disease. Epidemiol Infect. 2003;130(1):23-32. 63. Sears A, Baker MG, Wilson N, et al. Marked campylobacteriosis decline after interventions aimed at poultry, New Zealand. Emerging infectious diseases. 2011;17(6):1007-1015. 64. Wagenaar JA, French NP, Havelaar AH. Preventing Campylobacter at the Source: Why Is It So Difficult? Clinical Infectious Diseases. 2013;57(11):1600-1606. 65. National Chicken Council. Vertical Integration - The National Chicken Council. 2019; https://www.nationalchickencouncil.org/industry-issues/vertical- integration. Accessed 03/03/2019, 2019. 66. Van Deun K, Pasmans F, Ducatelle R, et al. Colonization strategy of Campylobacter jejuni results in persistent infection of the chicken gut. Veterinary microbiology. 2008;130(3-4):285-297. 67. Beery JT, Hugdahl MB, Doyle MP. Colonization of gastrointestinal tracts of chicks by Campylobacter jejuni. Applied and environmental microbiology. 1988;54(10):2365-2370. 68. Newell DG, Fearnley C. Sources of Campylobacter colonization in broiler chickens. Appl Environ Microbiol. 2003;69(8):4343-4351. 69. Hansson I, Pudas N, Harbom B, Engvall EO. Within-flock variations of Campylobacter loads in caeca and on carcasses from broilers. International journal of food microbiology. 2010;141(1):51-55. 70. Hansson I, Engvall EO, Lindblad J, Gunnarsson A, Vagsholm I. Surveillance programme for Campylobacter species in Swedish broilers, July 2001 to June 2002. The Veterinary record. 2004;155(7):193-196. 71. Hansson I, Forshell LP, Gustafsson P, et al. Summary of the Swedish Campylobacter program in broilers, 2001 through 2005. J Food Prot. 2007;70(9):2008-2014. 72. Gregory E, Barnhart H, Dreesen DW, Stern NJ, Corn JL. Epidemiological study of Campylobacter spp. in broilers: source, time of colonization, and prevalence. Avian diseases. 1997;41(4):890-898. 73. Sahin O, Zhang Q, Meitzler JC, Harr BS, Morishita TY, Mohan R. Prevalence, antigenic specificity, and bactericidal activity of poultry anti-Campylobacter maternal antibodies. Appl Environ Microbiol. 2001;67(9):3951-3957. 74. Kaino K, Hayashidani H, Kaneko K, Ogawa M. Intestinal colonization of Campylobacter jejuni in chickens. Nihon juigaku zasshi The Japanese journal of veterinary science. 1988;50(2):489-494. 75. Achen M, Morishita TY, Ley EC. Shedding and colonization of Campylobacter jejuni in broilers from day-of-hatch to slaughter age. Avian diseases. 1998;42(4):732-737. 76. Dipineto L, Gargiulo A, Russo TP, et al. Campylobacter jejuni, Campylobacter coli, and cytolethal distending toxin genes in laying hens. Avian diseases. 2011;55(1):103-105. 77. Callicott KA, Friethriksdottir V, Reiersen J, et al. Lack of evidence for vertical transmission of Campylobacter spp. in chickens. Applied and environmental microbiology. 2006;72(9):5794-5798.

149

78. Shanker S, Lee A, Sorrell TC. Campylobacter jejuni in broilers: the role of vertical transmission. The Journal of hygiene. 1986;96(2):153-159. 79. Petersen L, Nielsen EM, On SL. Serotype and genotype diversity and hatchery transmission of Campylobacter jejuni in commercial poultry flocks. Veterinary microbiology. 2001;82(2):141-154. 80. Jacobs-Reitsma WF. Campylobacter bacteria in breeder flocks. Avian diseases. 1995;39(2):355-359. 81. Jacobs-Reitsma WF, van de Giessen AW, Bolder NM, Mulder RW. Epidemiology of Campylobacter spp. at two Dutch broiler farms. Epidemiol Infect. 1995;114(3):413-421. 82. Pearson AD, Greenwood M, Healing TD, et al. Colonization of broiler chickens by waterborne Campylobacter jejuni. Appl Environ Microbiol. 1993;59(4):987- 996. 83. Hald B, Skovgård H, Bang DD, et al. Flies and Campylobacter infection of broiler flocks. Emerging infectious diseases. 2004;10(8):1490-1492. 84. Dale EL, Nolan SP, Berghaus RD, Hofacre CL. On farm prevention of Campylobacter and Salmonella: lessons learned from basic biosecurity interventions. The Journal of Applied Poultry Research. 2015;24(2):222-232. 85. Hiett KL, Stern NJ, Fedorka-Cray P, Cox NA, Musgrove MT, Ladely S. Molecular subtype analyses of Campylobacter spp. from Arkansas and California poultry operations. Applied and environmental microbiology. 2002;68(12):6220- 6236. 86. Hald B, Skov MN, Nielsen EM, et al. Campylobacter jejuni and Campylobacter coli in wild birds on Danish livestock farms. Acta veterinaria Scandinavica. 2016;58:11-11. 87. Zweifel C, Scheu KD, Keel M, Renggli F, Stephan R. Occurrence and genotypes of Campylobacter in broiler flocks, other farm animals, and the environment during several rearing periods on selected poultry farms. International journal of food microbiology. 2008;125(2):182-187. 88. Bull SA, Allen VM, Domingue G, et al. Sources of Campylobacter spp. colonizing housed broiler flocks during rearing. Applied and environmental microbiology. 2006;72(1):645-652. 89. O'Mahony E, Buckley JF, Bolton D, Whyte P, Fanning S. Molecular Epidemiology of Campylobacter Isolates from Poultry Production Units in Southern Ireland. PLOS ONE. 2011;6(12):e28490. 90. Hansson I, Vågsholm I, Svensson L, Olsson Engvall E. Correlations between Campylobacter spp. prevalence in the environment and broiler flocks. Journal of Applied Microbiology. 2007;103(3):640-649. 91. Messens W, Herman L, De Zutter L, Heyndrickx M. Multiple typing for the epidemiological study of contamination of broilers with thermotolerant Campylobacter. Veterinary microbiology. 2009;138(1-2):120-131. 92. Hald B, Wedderkopp A, Madsen M. Thermophilic Campylobacter spp. in Danish broiler production: a cross-sectional survey and a retrospective analysis of risk factors for occurrence in broiler flocks. Avian pathology : journal of the WVPA. 2000;29(2):123-131.

150

93. McCarthy ND, Colles FM, Dingle KE, et al. Host-associated genetic import in Campylobacter jejuni. Emerging infectious diseases. 2007;13(2):267-272. 94. Agunos A, Waddell L, Léger D, Taboada E. A Systematic Review Characterizing On-Farm Sources of Campylobacter spp. for Broiler Chickens. PLOS ONE. 2014;9(8):e104905. 95. Evans SJ, Sayers AR. A longitudinal study of campylobacter infection of broiler flocks in Great Britain. Preventive veterinary medicine. 2000;46(3):209-223. 96. Kapperud G, Skjerve E, Vik L, et al. Epidemiological investigation of risk factors for campylobacter colonization in Norwegian broiler flocks. Epidemiol Infect. 1993;111(2):245-255. 97. McDowell SW, Menzies FD, McBride SH, et al. Campylobacter spp. in conventional broiler flocks in Northern Ireland: epidemiology and risk factors. Preventive veterinary medicine. 2008;84(3-4):261-276. 98. Chowdhury S, Sandberg M, Themudo GE, Ersboll AK. Risk factors for Campylobacter infection in Danish broiler chickens. Poult Sci. 2012;91(10):2701- 2709. 99. Ellis-Iversen J, Ridley A, Morris V, et al. Persistent environmental reservoirs on farms as risk factors for Campylobacter in commercial poultry. Epidemiology and infection. 2012;140(5):916-924. 100. Guerin MT, Martin SW, Reiersen J, et al. Temperature-related risk factors associated with the colonization of broiler-chicken flocks with Campylobacter spp. in Iceland, 2001-2004. Preventive veterinary medicine. 2008;86(1-2):14-29. 101. Sandberg M, Sorensen LL, Steenberg B, Chowdhury S, Ersboll AK, Alban L. Risk factors for Campylobacter colonization in Danish broiler flocks, 2010 to 2011. Poult Sci. 2015;94(3):447-453. 102. Kuana SL, Santos LR, Rodrigues LB, et al. Risk factors and likelihood of Campylobacter colonization in broiler flocks. Brazilian Journal of Poultry Science. 2007;9:201-204. 103. Sommer HM, Høg BB, Larsen LS, et al. Analysis of farm specific risk factors for Campylobacter colonization of broilers in six European countries. Microbial Risk Analysis. 2016;2-3:16-26. 104. Näther G, Alter T, Martin A, Ellerbroek L. Analysis of risk factors for Campylobacter species infection in broiler flocks. Poultry Science. 2009;88(6):1299-1305. 105. Allen VM, Weaver H, Ridley AM, et al. Sources and spread of thermophilic Campylobacter spp. during partial depopulation of broiler chicken flocks. Journal of food protection. 2008;71(2):264-270. 106. Humphrey T, O'Brien S, Madsen M. Campylobacters as zoonotic pathogens: a food production perspective. International journal of food microbiology. 2007;117(3):237-257. 107. Hermans D, Van Deun K, Messens W, et al. Campylobacter control in poultry by current intervention measures ineffective: urgent need for intensified fundamental research. Vet Microbiol. 2011;152(3-4):219-228. 108. Line JE, Bailey JS. Effect of on-farm litter acidification treatments on Campylobacter and Salmonella populations in commercial broiler houses in northeast Georgia. Poultry science. 2006;85(9):1529-1534.

151

109. Onysk A, Weyand C, Mayer DG, et al. On-farm Campylobacter and Escherichia coli in commercial broiler chickens: Re-used bedding does not influence Campylobacter emergence and levels across sequential farming cycles. Poultry Science. 2016;95(5):1105-1115. 110. Svetoch EA, Stern NJ, Eruslanov BV, et al. Isolation of Bacillus circulans and Paenibacillus polymyxa strains inhibitory to Campylobacter jejuni and characterization of associated bacteriocins. J Food Prot. 2005;68(1):11-17. 111. Nazef L, Belguesmia Y, Tani A, Prevost H, Drider D. Identification of lactic acid bacteria from poultry feces: evidence on anti-campylobacter and anti-listeria activities. Poult Sci. 2008;87(2):329-334. 112. Stern NJ, Svetoch EA, Eruslanov BV, et al. Isolation of a Lactobacillus salivarius strain and purification of its bacteriocin, which is inhibitory to Campylobacter jejuni in the chicken gastrointestinal system. Antimicrobial agents and chemotherapy. 2006;50(9):3111-3116. 113. Line JE, Svetoch EA, Eruslanov BV, et al. Isolation and Purification of Enterocin E-760 with Broad Antimicrobial Activity against Gram-Positive and Gram- Negative Bacteria. Antimicrobial agents and chemotherapy. 2008;52(3):1094- 1100. 114. Neal-McKinney JM, Samuelson DR, Eucker TP, Nissen MS, Crespo R, Konkel ME. Reducing Campylobacter jejuni colonization of poultry via vaccination. PloS one. 2014;9(12):e114254. 115. Wyszynska A, Raczko A, Lis M, Jagusztyn-Krynicka EK. Oral immunization of chickens with avirulent Salmonella vaccine strain carrying C. jejuni 72Dz/92 cjaA gene elicits specific humoral immune response associated with protection against challenge with wild-type Campylobacter. Vaccine. 2004;22(11-12):1379-1389. 116. Meunier M, Guyard-Nicodème M, Vigouroux E, et al. Promising new vaccine candidates against Campylobacter in broilers. PloS one. 2017;12(11):e0188472- e0188472. 117. Ghareeb K, Awad WA, Mohnl M, et al. Evaluating the efficacy of an avian- specific probiotic to reduce the colonization of Campylobacter jejuni in broiler chickens. Poult Sci. 2012;91(8):1825-1832. 118. Smialek M, Burchardt S, Koncicki A. The influence of probiotic supplementation in broiler chickens on population and carcass contamination with Campylobacter spp. - Field study. Research in veterinary science. 2018;118:312-316. 119. Neal-McKinney JM, Lu X, Duong T, et al. Production of organic acids by probiotic lactobacilli can be used to reduce pathogen load in poultry. PLoS One. 2012;7(9):e43928. 120. Nishiyama K, Seto Y, Yoshioka K, et al. Lactobacillus gasseri SBT2055 reduces infection by and colonization of Campylobacter jejuni. PLoS One. 2014;9(9):e108827. 121. Stern NJ, Clavero MR, Bailey JS, Cox NA, Robach MC. Campylobacter spp. in broilers on the farm and after transport. Poultry science. 1995;74(6):937-941. 122. Slader J, Domingue G, Jorgensen F, et al. Impact of transport crate reuse and of catching and processing on Campylobacter and Salmonella contamination of broiler chickens. Applied and environmental microbiology. 2002;68(2):713-719.

152

123. Hansson I, Ederoth M, Andersson L, Vagsholm I, Olsson Engvall E. Transmission of Campylobacter spp. to chickens during transport to slaughter. J Appl Microbiol. 2005;99(5):1149-1157. 124. Rasschaert G, Houf K, De Zutter L. External contamination of Campylobacter- free flocks after transport in cleaned and disinfected containers. Journal of food protection. 2007;70(1):40-46. 125. Newell DG, Elvers KT, Dopfer D, et al. Biosecurity-based interventions and strategies to reduce Campylobacter spp. on poultry farms. Applied and environmental microbiology. 2011;77(24):8605-8614. 126. Mulder RWAW, Dorresteijn LWJ, Van Der Broek J. Cross‐contamination during the scalding and plucking of broilers. British Poultry Science. 1978;19(1):61-70. 127. Berrang M, Smith D, Meinersmann R. Variations on standard broiler processing in an effort to reduce Campylobacter numbers on postpick carcasses. Journal of Applied Poultry Research. 2011;20(2):197-202. 128. Rosenquist H, Sommer HM, Nielsen NL, Christensen BB. The effect of slaughter operations on the contamination of chicken carcasses with thermotolerant Campylobacter. International Journal of Food Microbiology. 2006;108(2):226- 232. 129. Wempe JM, Genigeorgis CA, Farver TB, Yusufu HI. Prevalence of Campylobacter jejuni in two California chicken processing plants. Applied and environmental microbiology. 1983;45(2):355-359. 130. U.S Department of Agriculture - Food Safety and Inspection Service. Compliance Guideline for Controlling Salmonella and Campylobacter in poultry - Second Edition. Washington D.C. 2008. 131. Berrang ME, Buhr RJ, Cason JA, Dickens JA. Broiler carcass contamination with Campylobacter from feces during defeathering. J Food Prot. 2001;64(12):2063- 2066. 132. Izat AL, Gardner FA, Denton JH, Golan FA. Incidence and level of Campylobacter jejuni in broiler processing. Poult Sci. 1988;67(11):1568-1572. 133. Fletcher DL, Craig EW, Arnold JW. An Evaluation of On-Line “Reprocessing” on Visual Contamination and Microbiological Quality of Broilers. The Journal of Applied Poultry Research. 1997;6(4):436-442. 134. U.S Department of Agriculture - Food Safety and Inspection Service. DRAFT FSIS Compliance Guideline For Controlling Salmonella and Campylobacter in Raw Poultry. Washington D.C. 2015. 135. Smith J, Bilgili SF, McKee SR, Corkran S, Singh M. Evaluation of post-chill applications of antimicrobials against Campylobacter jejuni on poultry carcasses. The Journal of Applied Poultry Research. 2015;24(4):451-456. 136. Buhr RJ, Walker JM, Bourassa DV, Caudill AB, Kiepper BH, Zhuang H. Impact of broiler processing scalding and chilling profiles on carcass and breast meat yield1. Poultry Science. 2014;93(6):1534-1541. 137. Tan SY, Tatsumura Y. Alexander Fleming (1881-1955): Discoverer of penicillin. Singapore medical journal. 2015;56(7):366-367. 138. Gustafson RH, Bowen RE. Antibiotic use in animal agriculture. J Appl Microbiol. 1997;83(5):531-541.

153

139. Landers TF, Cohen B, Wittum TE, Larson EL. A review of antibiotic use in food animals: perspective, policy, and potential. Public health reports (Washington, DC : 1974). 2012;127(1):4-22. 140. Van Boeckel TP, Brower C, Gilbert M, et al. Global trends in antimicrobial use in food animals. Proceedings of the National Academy of Sciences. 2015;112(18):5649-5654. 141. USDA. FACT SHEET: Veterinary Feed Directive Final Rule and Next Steps. Silver Spring, MD: USDA; 10/13/2017 2017. 142. Corry JE, Post DE, Colin P, Laisney MJ. Culture media for the isolation of campylobacters. International journal of food microbiology. 1995;26(1):43-76. 143. Luangtongkum T, Jeon B, Han J, Plummer P, Logue CM, Zhang Q. Antibiotic resistance in Campylobacter: emergence, transmission and persistence. Future microbiology. 2009;4(2):189-200. 144. Tang Y, Fang L, Xu C, Zhang Q. Antibiotic resistance trends and mechanisms in the foodborne pathogen, Campylobacter. Animal health research reviews. 2017;18(2):87-98. 145. Engberg J, Aarestrup FM, Taylor DE, Gerner-Smidt P, Nachamkin I. Quinolone and macrolide resistance in Campylobacter jejuni and C. coli: resistance mechanisms and trends in human isolates. Emerging infectious diseases. 2001;7(1):24-34. 146. Cha W, Mosci R, Wengert SL, et al. Antimicrobial Susceptibility Profiles of Human Campylobacter jejuni Isolates and Association with Phylogenetic Lineages. Front Microbiol. 2016;7:589. 147. Okeke IN, Laxminarayan R, Bhutta ZA, et al. Antimicrobial resistance in developing countries. Part I: recent trends and current status. The Lancet Infectious diseases. 2005;5(8):481-493. 148. Bywater R, Deluyker H, Deroover E, et al. A European survey of antimicrobial susceptibility among zoonotic and commensal bacteria isolated from food- producing animals. The Journal of antimicrobial chemotherapy. 2004;54(4):744- 754. 149. Caldwell DB, Wang Y, Lin J. Development, stability, and molecular mechanisms of macrolide resistance in Campylobacter jejuni. Antimicrobial agents and chemotherapy. 2008;52(11):3947-3954. 150. Sproston EL, Wimalarathna HML, Sheppard SK. Trends in fluoroquinolone resistance in Campylobacter. Microbial genomics. 2018;4(8):e000198. 151. Martinez M, McDermott P, Walker R. Pharmacology of the fluoroquinolones: a perspective for the use in domestic animals. Veterinary journal (London, England : 1997). 2006;172(1):10-28. 152. Nachamkin I, Ung H, Li M. Increasing fluoroquinolone resistance in Campylobacter jejuni, Pennsylvania, USA,1982-2001. Emerging infectious diseases. 2002;8(12):1501-1503. 153. Wu C, Liu D, Yao H, et al. Species shift and multidrug resistance of Campylobacter from chicken and swine, China, 2008–14. Journal of Antimicrobial Chemotherapy. 2015;71(3):666-669. 154. Andersson MI, MacGowan AP. Development of the quinolones. The Journal of antimicrobial chemotherapy. 2003;51 Suppl 1:1-11.

154

155. Kassenborg HD, Smith KE, Vugia DJ, et al. Fluoroquinolone-resistant Campylobacter infections: eating poultry outside of the home and foreign travel are risk factors. Clin Infect Dis. 2004;38 Suppl 3:S279-284. 156. Luo N, Pereira S, Sahin O, et al. Enhanced in vivo fitness of fluoroquinolone- resistant Campylobacter jejuni in the absence of antibiotic selection pressure. Proceedings of the National Academy of Sciences of the United States of America. 2005;102(3):541-546. 157. Han J, Wang Y, Sahin O, et al. A fluoroquinolone resistance associated mutation in gyrA Affects DNA supercoiling in Campylobacter jejuni. Front Cell Infect Microbiol. 2012;2:21. 158. Luo N, Sahin O, Lin J, Michel LO, Zhang Q. In vivo selection of Campylobacter isolates with high levels of fluoroquinolone resistance associated with gyrA mutations and the function of the CmeABC efflux pump. Antimicrobial agents and chemotherapy. 2003;47(1):390-394. 159. McDermott PF, Bodeis SM, English LL, et al. Ciprofloxacin resistance in Campylobacter jejuni evolves rapidly in chickens treated with fluoroquinolones. The Journal of infectious diseases. 2002;185(6):837-840. 160. Piddock LJ, Ricci V, McLaren I, Griggs DJ. Role of mutation in the gyrA and parC genes of nalidixic-acid-resistant salmonella serotypes isolated from animals in the United Kingdom. The Journal of antimicrobial chemotherapy. 1998;41(6):635-641. 161. Morgan-Linnell SK, Becnel Boyd L, Steffen D, Zechiedrich L. Mechanisms Accounting for Fluoroquinolone Resistance in Escherichia coli Clinical Isolates. Antimicrobial agents and chemotherapy. 2009;53(1):235-241. 162. Ge B, McDermott PF, White DG, Meng J. Role of efflux pumps and topoisomerase mutations in fluoroquinolone resistance in Campylobacter jejuni and Campylobacter coli. Antimicrobial agents and chemotherapy. 2005;49(8):3347-3354. 163. Yan M, Sahin O, Lin J, Zhang Q. Role of the CmeABC efflux pump in the emergence of fluoroquinolone-resistant Campylobacter under selection pressure. J Antimicrob Chemother. 2006;58(6):1154-1159. 164. Zirnstein G, Li Y, Swaminathan B, Angulo F. Ciprofloxacin resistance in Campylobacter jejuni isolates: detection of gyrA resistance mutations by mismatch amplification mutation assay PCR and DNA sequence analysis. Journal of clinical microbiology. 1999;37(10):3276-3280. 165. Ni Bhriain N, Dorman CJ, Higgins CF. An overlap between osmotic and anaerobic stress responses: a potential role for DNA supercoiling in the coordinate regulation of gene expression. Mol Microbiol. 1989;3(7):933-942. 166. Rui S, Tse-Dinh YC. Topoisomerase function during bacterial responses to environmental challenge. Front Biosci. 2003;8:d256-263. 167. Han J. Molecular mechanisms involved in the emergence and fitness of fluoroquinolone-resistant Campylobacter jejuni.Ames, Iowa, Iowa State University; 2009. 168. Bolton DJ. Campylobacter virulence and survival factors. Food microbiology. 2015;48:99-108.

155

169. Scallan E, Hoekstra RM, Angulo FJ, et al. Foodborne illness acquired in the United States--major pathogens. Emerging infectious diseases. 2011;17(1):7-15. 170. Domingues AR, Pires SM, Halasa T, Hald T. Source attribution of human campylobacteriosis using a meta-analysis of case-control studies of sporadic infections. Epidemiology and infection. 2012;140(6):970-981. 171. Helms M, Simonsen J, Molbak K. Foodborne bacterial infection and hospitalization: a registry-based study. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2006;42(4):498-506. 172. U.S. Department of Agriculture - Food Safety and Inspection Service. Changes to the Salmonella and Campylobacter Verification Testing Program: Revised Categorization and Follow-Up Sampling Procedures. 2018. 83:56046 - 56049. 173. U.S. Department of Agriculture - Food Safety and Inspection Service. New Performance Standards for Salmonella and Campylobacter in Not-Ready-to-Eat comminuted Chicken and Turkey Products and Raw Chicken Parts and Changes to Related Agency Verification Procedures: Response to Comments and Announcement of Implementation Schedule. 2016. 81:7285 - 7300. 174. U.S. Department of Agriculture - Food Safety and Inspection Service. Action to take in raw poultry establishments exceeding Salmonella performance standards. Washington, D.C. 2018. 175. Lin J. Novel approaches for Campylobacter control in poultry. Foodborne pathogens and disease. 2009;6(7):755-765. 176. Widders PR, Perry R, Muir WI, Husband AJ, Long KA. Immunisation of chickens to reduce intestinal colonisation with Campylobacter jejuni. British poultry science. 1996;37(4):765-778. 177. Skanseng B, Kaldhusdal M, Moen B, et al. Prevention of intestinal Campylobacter jejuni colonization in broilers by combinations of in-feed organic acids. Journal of applied microbiology. 2010;109(4):1265-1273. 178. Chaveerach P, Keuzenkamp DA, Urlings HA, Lipman LJ, van Knapen F. In vitro study on the effect of organic acids on Campylobacter jejuni/coli populations in mixtures of water and feed. Poultry science. 2002;81(5):621-628. 179. Magana M, Chatzipanagiotou S, Burriel AR, Ioannidis A. Inquiring into the Gaps of Campylobacter Surveillance Methods. Veterinary sciences. 2017;4(3):36. 180. Rosenquist H, Nielsen NL, Sommer HM, Norrung B, Christensen BB. Quantitative risk assessment of human campylobacteriosis associated with thermophilic Campylobacter species in chickens. International journal of food microbiology. 2003;83(1):87-103. 181. Food and Agriculture Organization/The United Nations/World Health Organization. Risk Assessment of Campylobacter Spp. in Broiler Chickens: Interpretative Summary. Geneva, Switzerland; 2009. 182. RStudio Team. RStudio: integrated development environment for R. Boston, MA. 2015. http://www.rstudio.com. 183. R Core Team. R: A language and environment for statistical computing. Vienne, Austria. 2013. http://www.R-project.org. 184. Smeets B. Licorice: Plot Likert-Like Data. R Package. Version 0.12015. 185. Wickham H. ggplot2: elegant graphics for data analysis. Springer; 2016.

156

186. U.S. Department of Agriculture - Food Safety and Inspection Service. Compliance Guidelines for Controlling Salmonella and Campylobacter in Raw Poultry. Washington, D.C. 2015. 80:78166 - 78167. 187. Hofacre CL, Berghaus RD, Dale EL, Nolan SP. On farm prevention of Campylobacter and Salmonella: lessons learned from basic biosecurity interventions. The Journal of Applied Poultry Research. 2015;24(2):222-232. 188. Hald B, Rattenborg E, Madsen M. Role of batch depletion of broiler houses on the occurrence of Campylobacter spp. in chicken flocks. Letters in applied microbiology. 2001;32(4):253-256. 189. Hansson I, Vagsholm I, Svensson L, Olsson Engvall E. Correlations between Campylobacter spp. prevalence in the environment and broiler flocks. Journal of applied microbiology. 2007;103(3):640-649. 190. Millman C, Christley R, Rigby D, Dennis D, O'Brien SJ, Williams N. "Catch 22": Biosecurity awareness, interpretation and practice amongst poultry catchers. Preventive veterinary medicine. 2017;141:22-32. 191. Sahin O, Kassem, II, Shen Z, Lin J, Rajashekara G, Zhang Q. Campylobacter in Poultry: Ecology and Potential Interventions. Avian diseases. 2015;59(2):185- 200. 192. van Wagenberg CPA, van Horne PLM, Sommer HM, Nauta MJ. Cost- effectiveness of Campylobacter interventions on broiler farms in six European countries. Microbial Risk Analysis. 2016;2-3:53-62. 193. Hastings R, Colles FM, McCarthy ND, Maiden MC, Sheppard SK. Campylobacter genotypes from poultry transportation crates indicate a source of contamination and transmission. Journal of applied microbiology. 2011;110(1):266-276. 194. Sahin O, Kobalka P, Zhang Q. Detection and survival of Campylobacter in chicken eggs. Journal of applied microbiology. 2003;95(5):1070-1079. 195. U.S. Department of Agriculture - Food Safety and Inspection Service. FSIS Constituent Update - FSIS to Implement Enrichment Method to Detect Campylobacter in All Raw Poultry Samples. 2018. 196. Van Immerseel F, Methner U, Rychlik I, et al. Vaccination and early protection against non-host-specific Salmonella serotypes in poultry: exploitation of innate immunity and microbial activity. Epidemiology and infection. 2005;133(6):959- 978. 197. Filho RACP, de Paiva JB, Argüello YMS, et al. Efficacy of several vaccination programmes in commercial layer and broiler breeder hens against experimental challenge with Salmonella enterica serovar Enteritidis. Avian Pathology. 2009;38(5):367-375. 198. Desin TS, Koster W, Potter AA. Salmonella vaccines in poultry: past, present and future. Expert Review of Vaccines. 2013;12:87+. 199. Dórea FC, Cole DJ, Hofacre C, et al. Effect of Salmonella Vaccination of Breeder Chickens on Contamination of Broiler Chicken Carcasses in Integrated Poultry Operations. Applied and environmental microbiology. 2010;76(23):7820-7825. 200. Berghaus RD, Thayer SG, Maurer JJ, Hofacre CL. Effect of vaccinating breeder chickens with a killed Salmonella vaccine on Salmonella prevalences and loads in

157

breeder and broiler chicken flocks. Journal of food protection. 2011;74(5):727- 734. 201. Nakamura M, Nagata T, Okamura S, Takehara K, Holt PS. The effect of killed Salmonella enteritidis vaccine prior to induced molting on the shedding of s. enteritidis in laying hens. Avian diseases. 2004;48(1):183-188. 202. Woodward MJ, Gettinby G, Breslin MF, Corkish JD, Houghton S. The efficacy of Salenvac, a Salmonella enterica subsp. Enterica serotype Enteritidis iron- restricted bacterin vaccine, in laying chickens. Avian pathology : journal of the WVPA. 2002;31(4):383-392. 203. Inoue AY, Berchieri A, Jr., Bernardino A, Paiva JB, Sterzo EV. Passive immunity of progeny from broiler breeders vaccinated with oil-emulsion bacterin against Salmonella enteritidis. Avian diseases. 2008;52(4):567-571. 204. Chaisowwong W, Kusumoto A, Hashimoto M, Harada T, Maklon K, Kawamoto K. Physiological characterization of Campylobacter jejuni under cold stresses conditions: its potential for public threat. The Journal of veterinary medical science. 2012;74(1):43-50. 205. Oh E, McMullen L, Jeon B. Impact of oxidative stress defense on bacterial survival and morphological change in Campylobacter jejuni under aerobic conditions. Front Microbiol. 2015;6:295. 206. Musgrove MT, Berrang ME, Byrd JA, Stern NJ, Cox NA. Detection of Campylobacter spp. in ceca and crops with and without enrichment. Poultry science. 2001;80(6):825-828. 207. Colles FM, McCarthy ND, Layton R, Maiden MC. The prevalence of Campylobacter amongst a free-range broiler breeder flock was primarily affected by flock age. PloS one. 2011;6(12):e22825. 208. Newell DG, Fearnley C. Sources of Campylobacter Colonization in Broiler Chickens. Applied and environmental microbiology. 2003;69(8):4343-4351. 209. Sheppard SK, Dallas JF, Strachan NJ, et al. Campylobacter genotyping to determine the source of human infection. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2009;48(8):1072-1078. 210. Friedman CR, Hoekstra RM, Samuel M, et al. Risk factors for sporadic Campylobacter infection in the United States: A case-control study in FoodNet sites. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America. 2004;38 Suppl 3:S285-296. 211. Martinez M, McDermott P, Walker R. Pharmacology of the fluoroquinolones: A perspective for the use in domestic animals. The Veterinary Journal. 2006;172(1):10-28. 212. U.S. Food and Drug Administration. Food and Drug Administration. Enrofloxacin for Poultry; Final Decision on Withdrawal of New Animal Drug Application Following Formal Evidentiary Public Hearing; Availability. Federal Register. Vol 702005:44105. 213. U.S Food and Drug Administration - National Antimicrobial Resistance Monitoring System. National Antimicrobial Resistance Monitoring System Integrated Report - 2015.2015.

158

214. Zeitouni S, Kempf I. Fitness cost of fluoroquinolone resistance in Campylobacter coli and Campylobacter jejuni. Microbial drug resistance (Larchmont, NY). 2011;17(2):171-179. 215. Han J. Molecular mechanisms involved in the emergence and fitness of fluoroquinolone-resistant Campylobacter jejuni, Iowa State University; 2009. 216. Garrido MC, Herrero M, Kolter R, Moreno F. The export of the DNA replication inhibitor Microcin B17 provides immunity for the host cell. The EMBO journal. 1988;7(6):1853-1862. 217. Baquero F, Moreno F. The microcins. FEMS Microbiology Letters. 1984;23(2):117-124. 218. Destoumieux-Garzon D, Peduzzi J, Rebuffat S. Focus on modified microcins: structural features and mechanisms of action. Biochimie. 2002;84(5-6):511-519. 219. Heddle JG, Blance SJ, Zamble DB, et al. The antibiotic microcin B17 is a DNA gyrase poison: characterisation of the mode of inhibition. Journal of molecular biology. 2001;307(5):1223-1234. 220. Klena JD, Parker CT, Knibb K, et al. Differentiation of Campylobacter coli, Campylobacter jejuni, Campylobacter lari, and Campylobacter upsaliensis by a multiplex PCR developed from the nucleotide sequence of the lipid A gene lpxA. Journal of clinical microbiology. 2004;42(12):5549-5557. 221. Tamura K, Stecher G, Peterson D, Filipski A, Kumar S. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Molecular biology and evolution. 2013;30(12):2725-2729. 222. Elso CM, Roberts LJ, Smyth GK, et al. Leishmaniasis host response loci (lmr1-3) modify disease severity through a Th1/Th2-independent pathway. Genes and immunity. 2004;5(2):93-100. 223. Baldwin T, Sakthianandeswaren A, Curtis JM, et al. Wound healing response is a major contributor to the severity of cutaneous leishmaniasis in the ear model of infection. Parasite immunology. 2007;29(10):501-513. 224. Ritz M, Garenaux A, Berge M, Federighi M. Determination of rpoA as the most suitable internal control to study stress response in C. jejuni by RT-qPCR and application to oxidative stress. Journal of microbiological methods. 2009;76(2):196-200. 225. Pfaffl MW. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic acids research. 2001;29(9):e45. 226. Pfaffl MW, Horgan GW, Dempfle L. Relative expression software tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic acids research. 2002;30(9):e36-e36. 227. R Core Team. R: A language and environment for statistical computing. 2013. http://www.R-project.org 228. Holmes K, Mulholland F, Pearson BM, et al. Campylobacter jejuni gene expression in response to iron limitation and the role of Fur. Microbiology. 2005;151(Pt 1):243-257. 229. van Vliet AH, Ketley JM, Park SF, Penn CW. The role of iron in Campylobacter gene regulation, metabolism and oxidative stress defense. FEMS microbiology reviews. 2002;26(2):173-186.

159

230. Fearnhead P, Smith NG, Barrigas M, Fox A, French N. Analysis of recombination in Campylobacter jejuni from MLST population data. Journal of molecular evolution. 2005;61(3):333-340. 231. Wilson DJ, Gabriel E, Leatherbarrow AJ, et al. Rapid evolution and the importance of recombination to the gastroenteric pathogen Campylobacter jejuni. Molecular biology and evolution. 2009;26(2):385-397. 232. Suerbaum S, Lohrengel M, Sonnevend A, Ruberg F, Kist M. Allelic diversity and recombination in Campylobacter jejuni. J Bacteriol. 2001;183(8):2553-2559. 233. Wassenaar TM, Wagenaar JA, Rigter A, Fearnley C, Newell DG, Duim B. Homonucleotide stretches in chromosomal DNA of Campylobacter jejuni display high frequency polymorphism as detected by direct PCR analysis. FEMS Microbiol Lett. 2002;212(1):77-85. 234. van Putten JP, van Alphen LB, Wosten MM, de Zoete MR. Molecular mechanisms of campylobacter infection. Current topics in microbiology and immunology. 2009;337:197-229. 235. Wilson DL, Rathinam VA, Qi W, et al. Genetic diversity in Campylobacter jejuni is associated with differential colonization of broiler chickens and C57BL/6J IL10-deficient mice. Microbiology. 2010;156(Pt 7):2046-2057. 236. Pesci EC, Cottle DL, Pickett CL. Genetic, enzymatic, and pathogenic studies of the iron superoxide dismutase of Campylobacter jejuni. Infection and immunity. 1994;62(7):2687-2694. 237. van Vliet AHM, Baillon M-LA, Penn CW, Ketley JM. Campylobacter jejuni Contains Two Fur Homologs: Characterization of Iron-Responsive Regulation of Peroxide Stress Defense Genes by the PerR Repressor. Journal of Bacteriology. 1999;181(20):6371-6376. 238. Grant KA, Park SF. Molecular characterization of katA from Campylobacter jejuni and generation of a catalase-deficient mutant of Campylobacter coli by interspecific allelic exchange. Microbiology. 1995;141 ( Pt 6):1369-1376. 239. Guerry P, Perez-Casal J, Yao R, McVeigh A, Trust TJ. A genetic locus involved in iron utilization unique to some Campylobacter strains. Journal of bacteriology. 1997;179(12):3997-4002. 240. Herrero M, Moreno F. Microcin B17 blocks DNA replication and induces the SOS system in Escherichia coli. Journal of general microbiology. 1986;132(2):393-402. 241. Pierrat OA, Maxwell A. Evidence for the role of DNA strand passage in the mechanism of action of microcin B17 on DNA gyrase. Biochemistry. 2005;44(11):4204-4215. 242. Vizan JL, Hernandez-Chico C, del Castillo I, Moreno F. The peptide antibiotic microcin B17 induces double-strand cleavage of DNA mediated by E. coli DNA gyrase. The EMBO journal. 1991;10(2):467-476. 243. Collin F, Karkare S, Maxwell A. Exploiting bacterial DNA gyrase as a drug target: current state and perspectives. Appl Microbiol Biotechnol. 2011;92(3):479- 497. 244. Lomovskaya O, Kawai F, Matin A. Differential regulation of the mcb and emr operons of Escherichia coli: role of mcb in multidrug resistance. Antimicrobial agents and chemotherapy. 1996;40(4):1050-1052.

160

245. Piddock LJ, Ricci V, Pumbwe L, Everett MJ, Griggs DJ. Fluoroquinolone resistance in Campylobacter species from man and animals: detection of mutations in topoisomerase genes. The Journal of antimicrobial chemotherapy. 2003;51(1):19-26. 246. Yang W, Zhang M, Zhou J, Pang L, Wang G, Hou F. The Molecular Mechanisms of Ciprofloxacin Resistance in Clinical Campylobacter jejuni and Their Genotyping Characteristics in Beijing, China. Foodborne Pathog Dis. 2017;14(7):386-392. 247. Bakeli G, Sato K, Kumita W, et al. Antimicrobial susceptibility and mechanism of quinolone resistance in Campylobacter jejuni strains isolated from diarrheal patients in a hospital in Tokyo. Journal of infection and chemotherapy : official journal of the Japan Society of Chemotherapy. 2008;14(5):342-348. 248. Payot S, Cloeckaert A, Chaslus-Dancla E. Selection and characterization of fluoroquinolone-resistant mutants of Campylobacter jejuni using enrofloxacin. Microbial drug resistance (Larchmont, NY). 2002;8(4):335-343. 249. Yogesh A. Chatur MNB, Shivani Modi, Jitendrakumar B. Nayak. Fluoroquinolone resistance and detection of topoisomerase gene mutation in Campylobacter jejuni isolated from animal and human sources. International Journal of Current Microbiology and Applied Sciences. 2014;3(6):773 - 783. 250. Metelev M, Serebryakova M, Ghilarov D, Zhao Y, Severinov K. Structure of Microcin B-Like Compounds Produced by Pseudomonas syringae and Species Specificity of Their Antibacterial Action. Journal of Bacteriology. 2013;195(18):4129-4137. 251. Baillon M-LA, van Vliet AHM, Ketley JM, Constantinidou C, Penn CW. An Iron-Regulated Alkyl Hydroperoxide Reductase (AhpC) Confers Aerotolerance and Oxidative Stress Resistance to the Microaerophilic Pathogen Campylobacter jejuni. Journal of Bacteriology. 1999;181(16):4798-4804. 252. Jeurissen SH, Lewis F, van der Klis JD, Mroz Z, Rebel JM, ter Huurne AA. Parameters and techniques to determine intestinal health of poultry as constituted by immunity, integrity, and functionality. Current issues in intestinal microbiology. 2002;3(1):1-14. 253. Clavijo V, Flórez MJV. The gastrointestinal microbiome and its association with the control of pathogens in broiler chicken production: A review. Poultry science. 2017;97(3):1006-1021. 254. Borda-Molina D, Seifert J, Camarinha-Silva A. Current Perspectives of the Chicken Gastrointestinal Tract and Its Microbiome. Computational and structural biotechnology journal. 2018;16:131-139. 255. Wang L, Lilburn M, Yu Z. Intestinal Microbiota of Broiler Chickens As Affected by Litter Management Regimens. Frontiers in microbiology. 2016;7:593. 256. Lu J, Idris U, Harmon B, Hofacre C, Maurer JJ, Lee MD. Diversity and succession of the intestinal bacterial community of the maturing broiler chicken. Applied and environmental microbiology. 2003;69(11):6816-6824. 257. Pan D, Yu Z. Intestinal microbiome of poultry and its interaction with host and diet. Gut microbes. 2014;5(1):108-119.

161

258. Malone GW, Chaloupka GW, Saylor WW. Influence of Litter Type and Size on Broiler Performance.: 1. Factors Affecting Litter Consumption1. Poultry science. 1983;62(9):1741-1746. 259. Cressman MD, Yu Z, Nelson MC, Moeller SJ, Lilburn MS, Zerby HN. Interrelations between the microbiotas in the litter and in the intestines of commercial broiler chickens. Applied and environmental microbiology. 2010;76(19):6572-6582. 260. Torok VA, Hughes RJ, Ophel-Keller K, Ali M, MacAlpine R. Influence of different litter materials on cecal microbiota colonization in broiler chickens. Poultry science. 2009;88(12):2474-2481. 261. Wei S, Gutek A, Lilburn M, Yu Z. Abundance of pathogens in the gut and litter of broiler chickens as affected by bacitracin and litter management. Veterinary Microbiology. 2013;166(3):595-601. 262. Nagaraj M, Wilson CAP, Saenmahayak B, Hess JB, Bilgili SF. Efficacy of a Litter Amendment to Reduce Pododermatitis in Broiler Chickens1. The Journal of Applied Poultry Research. 2007;16(2):255-261. 263. Thaxton Y, L. Balzli C, D. Tankson J. Relationship of Broiler Flock Numbers to Litter Microflora 1. Vol 122003. 264. Roll VF, Dai Pra MA, Roll AP. Research on Salmonella in broiler litter reused for up to 14 consecutive flocks. Poultry science. 2011;90(10):2257-2262. 265. Kassem, II, Sanad Y, Gangaiah D, Lilburn M, Lejeune J, Rajashekara G. Use of bioluminescence imaging to monitor Campylobacter survival in chicken litter. Journal of applied microbiology. 2010;109(6):1988-1997. 266. Chen Z, Wang H, Ionita C, Luo F, Jiang X. Effects of Chicken Litter Storage Time and Ammonia Content on Thermal Resistance of Desiccation-Adapted Salmonella spp. Applied and environmental microbiology. 2015;81(19):6883- 6889. 267. Soliman ES, Sallam NH, Abouelhassan EM. Effectiveness of poultry litter amendments on bacterial survival and Eimeria oocyst sporulation. Veterinary world. 2018;11(8):1064-1073. 268. Kim J, Diao J, Shepherd MW, Jr., et al. Validating thermal inactivation of Salmonella spp. in fresh and aged chicken litter. Applied and environmental microbiology. 2012;78(4):1302-1307. 269. Dumas MD, Polson SW, Ritter D, et al. Impacts of Poultry House Environment on Poultry Litter Bacterial Community Composition. PLOS ONE. 2011;6(9):e24785. 270. Lopes M, Roll VFB, Leite FL, et al. Quicklime treatment and stirring of different poultry litter substrates for reducing pathogenic bacteria counts. Poultry science. 2013;92(3):638-644. 271. Huff WE, Malone GW, Chaloupka GW. Effect of litter treatment on broiler performance and certain litter quality parameters. Poultry science. 1984;63(11):2167-2171. 272. Pope MJ, Cherry TE. An evaluation of the presence of pathogens on broilers raised on poultry litter treatment-treated litter. Poultry science. 2000;79(9):1351- 1355.

162

273. Martin SA, McCann MA, Waltman WD, II. Microbiological Survey of Georgia Poultry Litter. The Journal of Applied Poultry Research. 1998;7(1):90-98. 274. Terzich M, Pope MJ, Cherry TE, Hollinger J. Survey of Pathogens in Poultry Litter in the United States. The Journal of Applied Poultry Research. 2000;9(3):287-291. 275. Lovanh N, Cook KL, Rothrock MJ, Miles DM, Sistani K. Spatial Shifts in Microbial Population Structure Within Poultry Litter Associated with Physicochemical Properties. Poultry science. 2007;86(9):1840-1849. 276. Lu J, Sanchez S, Hofacre C, Maurer JJ, Harmon BG, Lee MD. Evaluation of Broiler Litter with Reference to the Microbial Composition as Assessed by Using 16S rRNA and Functional Gene Markers. Applied and environmental microbiology. 2003;69(2):901-908. 277. Johnson TJ, Youmans BP, Noll S, et al. A Consistent and Predictable Commercial Broiler Chicken Bacterial Microbiota in Antibiotic-Free Production Displays Strong Correlations with Performance. Applied and environmental microbiology. 2018;84(12):e00362-00318. 278. Blodgett R. Appendix 2: most probable number from serial dilutions. Bacteriological analytical manual 2010; https://www.fda.gov/Food/FoodScienceResearch/LaboratoryMethods/ucm109656 .htm#tab1. Accessed March 14, 2017. 279. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: High-resolution sample inference from Illumina amplicon data. Nature Methods. 2016;13:581. 280. Team R. RStudio: integrated development for R. RStudio, Inc, Boston, MA. http://www.rstudio.com. 2015. 281. Pruesse E, Quast C, Knittel K, et al. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic acids research. 2007;35(21):7188-7196. 282. Quast C, Pruesse E, Yilmaz P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic acids research. 2013;41(Database issue):D590-596. 283. McMurdie PJ, Holmes S. phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLOS ONE. 2013;8(4):e61217. 284. Oksanen J, Kindt R, Legendre P, O'Hara B, Stevens MHH. vegan: Community Ecology Package. 2007. 285. Paulson JN, Stine OC, Bravo HC, Pop M. Differential abundance analysis for microbial marker-gene surveys. Nat Methods. 2013;10(12):1200-1202. 286. Love M, Anders S, Huber W. Differential analysis of count data–the deseq2 package. Genome Biol. 2014;15:550. 287. McMurdie PJ, Holmes S. Waste Not, Want Not: Why Rarefying Microbiome Data Is Inadmissible. PLOS Computational Biology. 2014;10(4):e1003531. 288. Neill SD, Campbell JN, Greene JA. Campylobacter species in broiler chickens. Avian pathology : journal of the WVPA. 1984;13(4):777-785.

163

289. I. KI, A. HY, Dipak K, et al. Nonculturability Might Underestimate the Occurrence of Campylobacter in Broiler Litter. Foodborne Pathogens and Disease. 2017;14(8):472-477. 290. Rothrock MJ, Jr., Cook KL, Bolster CH. Comparative quantification of Campylobacter jejuni from environmental samples using traditional and molecular biological techniques. Canadian journal of microbiology. 2009;55(6):633-641. 291. Donaldson EE, Stanley D, Hughes RJ, Moore RJ. The time-course of broiler intestinal microbiota development after administration of cecal contents to incubating eggs. PeerJ. 2017;5:e3587. 292. Oakley BB, Kogut MH. Spatial and Temporal Changes in the Broiler Chicken Cecal and Fecal Microbiomes and Correlations of Bacterial Taxa with Cytokine Gene Expression. Frontiers in Veterinary Science. 2016;3(11). 293. Ranjitkar S, Lawley B, Tannock G, Engberg RM. Bacterial Succession in the Broiler Gastrointestinal Tract. Applied and environmental microbiology. 2016;82(8):2399-2410. 294. Ocejo M, Oporto B, Hurtado A. 16S rRNA amplicon sequencing characterization of caecal microbiome composition of broilers and free-range slow-growing chickens throughout their productive lifespan. Scientific Reports. 2019;9(1):2506. 295. Knarreborg A, Simon MA, Engberg RM, Jensen BB, Tannock GW. Effects of dietary fat source and subtherapeutic levels of antibiotic on the bacterial community in the ileum of broiler chickens at various ages. Applied and environmental microbiology. 2002;68(12):5918-5924. 296. Lin J, Hunkapiller AA, Layton AC, Chang Y-J, Robbins KR. Response of Intestinal Microbiota to Antibiotic Growth Promoters in Chickens. Foodborne Pathogens and Disease. 2013;10(4):331-337. 297. Danzeisen JL, Kim HB, Isaacson RE, Tu ZJ, Johnson TJ. Modulations of the Chicken Cecal Microbiome and Metagenome in Response to Anticoccidial and Growth Promoter Treatment. PLOS ONE. 2011;6(11):e27949. 298. Dumonceaux TJ, Hill JE, Hemmingsen SM, Van Kessel AG. Characterization of Intestinal Microbiota and Response to Dietary Virginiamycin Supplementation in the Broiler Chicken. Applied and environmental microbiology. 2006;72(4):2815- 2823. 299. Collier CT, Smiricky-Tjardes MR, Albin DM, et al. Molecular ecological analysis of porcine ileal microbiota responses to antimicrobial growth promoters. Journal of animal science. 2003;81(12):3035-3045. 300. Burgess CM, Gianotti A, Gruzdev N, et al. The response of foodborne pathogens to osmotic and desiccation stresses in the food chain. International Journal of Food Microbiology. 2016;221:37-53. 301. Fu TT, Thayer DW. Comparison of batch and semicontinuous cultures for production of protein from mesquite wood by Brevibacterium sp. JM98A. Biotechnology and bioengineering. 1975;17(12):1749-1760. 302. Collins MD, Brown J, Jones D. Brachybacterium faecium gen. nov., sp. nov., a coryneform bacterium from poultry deep litter. International Journal of Systematic and Evolutionary Microbiology. 1988;38(1):45-48.

164

303. Kim KK, Lee KC, Jeong H, Stevens DA, Lee J-S. Draft Genome Sequence of the Human Pathogen Halomonas stevensii S18214T. Journal of Bacteriology. 2012;194(18):5143-5143. 304. Vreeland RH, Martin EL. Growth characteristics, effects of temperature, and ion specificity of the halotolerant bacterium Halomonas elongata. Canadian journal of microbiology. 1980;26(7):746-752. 305. Koerner RJ, Jones AL, Goodfellow M. The genus Dietzia: a new home for some known and emerging opportunist pathogens. FEMS Immunology & Medical Microbiology. 2009;55(3):296-305. 306. Eeckhaut V, Wang J, Van Parys A, et al. The Probiotic Butyricicoccus pullicaecorum Reduces Feed Conversion and Protects from Potentially Harmful Intestinal Microorganisms and Necrotic Enteritis in Broilers. Frontiers in microbiology. 2016;7:1416-1416. 307. Richardson E, Vasconcelos EJR, Calloway KN, et al. The cecal microbiome of commercial broiler chickens varies significantly by season. Poultry science. 2018;97(10):3635-3644. 308. Anderson M, Walsh D. What null hypothesis are you testing? PERMANOVA, ANOSIM and the Mantel test in the face of heterogeneous dispersions. Ecol Monogr. 2013;83:557-574. 309. Locatelli A, Rothrock MJ, Jr, Hiett KL, Caudill AC. Do fecal and litter microbiomes vary within the major areas of a commercial poultry house, and does this affect sampling strategies for whole-house microbiomic studies? The Journal of Applied Poultry Research. 2017;26(3):325-336. 310. Huang P, Zhang Y, Xiao K, et al. The chicken gut metagenome and the modulatory effects of plant-derived benzylisoquinoline alkaloids. Microbiome. 2018;6(1):211-211. 311. Hou Q, Kwok L-Y, Zheng Y, et al. Differential fecal microbiota are retained in broiler chicken lines divergently selected for fatness traits. Scientific Reports. 2016;6:37376. 312. Choi JH, Kim GB, Cha CJ. Spatial heterogeneity and stability of bacterial community in the gastrointestinal tracts of broiler chickens. Poultry science. 2014;93(8):1942-1950. 313. Mohd Shaufi MA, Sieo CC, Chong CW, Gan HM, Ho YW. Deciphering chicken gut microbial dynamics based on high-throughput 16S rRNA metagenomics analyses. Gut pathogens. 2015;7:4-4. 314. Akşit M, Yalçin S, Yenisey Ç, Özdemir D. Brooding temperatures for chicks acclimated to heat during incubation: effects on post-hatch intestinal development and body weight under heat stress. British Poultry Science. 2010;51(3):444-452. 315. Sakaridis I, Ellis RJ, Cawthraw SA, et al. Investigating the Association Between the Caecal Microbiomes of Broilers and Campylobacter Burden. Frontiers in microbiology. 2018;9:927-927. 316. Sofka D, Pfeifer A, Gleiss B, Paulsen P, Hilbert F. Changes within the intestinal flora of broilers by colonisation with Campylobacter jejuni. Berliner und Munchener tierarztliche Wochenschrift. 2015;128(3-4):104-110. 317. Kaakoush NO, Sodhi N, Chenu JW, Cox JM, Riordan SM, Mitchell HM. The interplay between Campylobacter and Helicobacter species and other

165

gastrointestinal microbiota of commercial broiler chickens. Gut pathogens. 2014;6:18-18. 318. Liu Y, Zhang Z, Li Y, Fei Y. Response of microbial communities to roxarsone under different culture conditions. Canadian journal of microbiology. 2017;63(8):661-670. 319. Robyn J, Rasschaert G, Messens W, Pasmans F, Heyndrickx M. Screening for lactic acid bacteria capable of inhibiting Campylobacter jejuni in in vitro simulations of the broiler chicken caecal environment. Beneficial microbes. 2012;3(4):299-308. 320. Stern NJ, Svetoch EA, Eruslanov BV, et al. Isolation of a Lactobacillus salivarius Strain and Purification of Its Bacteriocin, Which Is Inhibitory to Campylobacter jejuni in the Chicken Gastrointestinal System. Antimicrobial Agents and Chemotherapy. 2006;50(9):3111-3116. 321. Mesa D, Lammel DR, Balsanelli E, et al. Cecal Microbiota in Broilers Fed with Prebiotics. Frontiers in genetics. 2017;8:153-153. 322. Fernando DM, Khan IUH, Patidar R, et al. Isolation and Characterization of Acinetobacter baumannii Recovered from Campylobacter Selective Medium. Frontiers in microbiology. 2016;7:1871.

Appendices Appendix A. Survey Questionnaire.

166

Q1.1 Thank you for participating in this survey regarding Campylobacter in broiler production. This survey is intended to better understand pre- and post-harvest interventions and management practices that have been used in the past or are currently being used by the U.S. broiler industry to mitigate Campylobacter contamination of chicken products and to meet USDA Food Safety and Inspection Service (FSIS) Salmonella and Campylobacter performance standards.

ALL RESPONSES ARE CONFIDENTIAL AND ANONYMOUS.

This survey should take a maximum of 30 minutes to complete.

If you have any questions about this survey, please contact Dr. Randall Singer at [email protected].

Q1.2 What is your current job responsibility? o Broiler manager / Farm supervisor o Plant manager / Food safety or QA team member o Poultry veterinarian

Processing plant manager/QA manager

Q2.1 The following questions will ask about current USDA Food Safety and Inspection Service (FSIS) performance standards for Campylobacter and Salmonella in poultry

167

products.

For each question below, please select the response that best describes your understanding of the topic.

Q2.2 Please indicate how familiar you are with the FSIS performance standards for Campylobacter and Salmonella that were implemented for whole birds in 2011 and for chicken parts in July 2016. o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q2.3 Are you currently meeting the performance standards for Campylobacter and Salmonella as determined by one or both of the following approaches? Yes No Don't know N/A

Internal testing o o o o

USDA testing o o o o

Q2.4 Are there any changes (i.e., new control strategy or intervention applied pre- or post- harvest) that have been implemented in your processing plant within the past year to control Campylobacter to help meet the new performance standards? o Yes o No, no changes have been made regarding Campylobacter control in my processing plant o No, my establishment is already in compliance and no changes are needed o No, my establishment is exempted from meeting the performance standards due to establishment type or size o I don't know if any changes have been made regarding Campylobacter control in my processing plant

168

Q2.5 What changes or recommendations have you made specifically to meet the new performance standards for Campylobacter? Please check all that apply. ▢ Implement new on-farm interventions (such as vaccines, chemicals, prebiotics and probiotics, and antibiotics) to reduce Campylobacter colonization in birds ▢ Implement or improve biosecurity and other management strategies on-farm to reduce Campylobacter colonization in birds ▢ Implement on-farm microbiological testing to check Campylobacter prevalence or contamination levels ▢ Implement new post-harvest interventions in the processing facility to reduce Campylobacter contamination on the final product ▢ Implement in-plant microbiological testing to check Campylobacter contamination levels during broiler processing ▢ Other - Please specify your answer below

Q2.6 Are there any changes (i.e. new control strategy or intervention applied pre- or post-harvest) that have been implemented in your processing plant within the past year to control Salmonella to help meet the new performance standards? o Yes o No, no changes have been made regarding Salmonella control in my processing plant o No, my establishment is already in compliance and no changes are needed o No, my establishment is exempted from meeting the performance standards due to establishment type or size o I don't know if any changes have been made regarding Salmonella control in my processing plant

Q2.7 What changes or recommendations have you made specifically to meet the new performance standards for Salmonella? Please check all that apply. ▢ Implement new on-farm interventions (such as vaccines, chemicals, prebiotics and probiotics, antibiotics) to reduce Salmonella colonization in birds ▢ Implement or improve biosecurity and other management strategies on-farm to reduce Salmonella colonization in birds ▢ Implement on-farm microbiological testing to check Salmonella prevalence or contamination levels ▢ Implement new post-harvest interventions in the processing facility to reduce Salmonella contamination in final products ▢ Implement in-plant microbiological testing to check Salmonella contamination levels during broiler processing ▢ Other - Please specify your answer below

169

Q2.8 The following questions will ask about your understanding of Campylobacter in poultry. For each question below, please select the response that best describes your understanding.

Broiler chicks can be colonized with Campylobacter at hatch if the breeding o True o False o Don't know hens are colonized with Campylobacter. Campylobacter- negative broilers can acquire Campylobacter from the environment o True o False o Don't know (soil, water, insects, etc.). Campylobacter-positive broilers shed Campylobacter in their o True o False o Don't know feces. Effective on-farm biosecurity can completely eliminate o True o False o Don't know Campylobacter from the broiler farm. Farm equipment such as transportation vehicles and modules can be o True o False o Don't know contaminated with Campylobacter. The incidence of Campylobacter decreases during second o True o False o Don't know processing.

170

Q2.9 Are you familiar with factors on the farm that increase the risk of Campylobacter colonization of the broiler farm environment? o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q2.10 Are you familiar with factors in the processing plant that increase the risk of Campylobacter contamination on the final product? o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q2.11 In your opinion, which of the following on-farm interventions or management practices are the most effective at reducing Campylobacter colonization in broiler chickens? Please rank the following on-farm interventions or management practices from MOST EFFECTIVE to LEAST EFFECTIVE for controlling Campylobacter.

To rank these choices, simply click and drag the choice to its appropriate place. ______Improvement in biosecurity ______Improvement in drinking water quality ______Improvement in litter quality ______Administration of antibiotics ______Administration of feed supplements such as prebiotics and/or probiotics

171

Q2.12 The following questions relate to basic information about the processing plant at which you are currently employed.

Q2.13 How many chickens are slaughtered weekly at your establishment?

______

Q2.14 Does your processing establishment slaughter chickens and/or further process the chicken? Please check all that apply. ▢ Slaughter ▢ Second processing

Q2.15 For second processing, are any primary materials sourced from another plant? Please check all that apply. ▢ Yes ▢ No ▢ Don't know

Q2.16 In which of the following categories does the external supplier typically fall into with respect to the FSIS performance standards for Campylobacter? o Category 1 o Category 2 o Category 3 o Prefer not to answer o Don't know

Q2.17 In which of the following categories does the external supplier typically fall into with respect to the FSIS performance standards forSalmonella? o Category 1 o Category 2 o Category 3 o Prefer not to answer o Don't know

Q2.18 Does your processing establishment test incoming source materials for Campylobacter and/or Salmonella? o Yes o No o Don't know

172

Q2.19 Does your processing establishment require external suppliers to test source materials for Campylobacter and/or Salmonella? o Yes o No o Don't know

Q2.20 Does your processing establishment use antimicrobial interventions (such as antimicrobial dips or sprays, UV light systems, etc.) to reduce microbial contamination (Campylobacter and/or Salmonella) in second processing? o Yes o No o Don't know

Q2.21 Does your processing establishment do its own microbial testing for Campylobacter and/or Salmonella contamination? o Yes o No o Don't know

Q2.22 Where do you take samples for Campylobacter and/or Salmonella? Please check all that apply. Live haul Pre- Post- Second Don't Re-hang Evisceration (outside) chill chill processing know

Campylobacter ▢ ▢ ▢ ▢ ▢ ▢ ▢

Salmonella ▢ ▢ ▢ ▢ ▢ ▢ ▢

Q2.23 The approximate incidence of Campylobacter in this plant is: Prefer Don't 0% 0-10% 10-25% 25-50% >50% not to N/A know answer Internal testing - Whole o o o o o o o o carcass Internal testing - Second o o o o o o o o processing

173

Q2.24 The approximate incidence of Salmonella in this plant is: Prefer Don't 0% 0-10% 10-25% 25-50% >50% not to N/A know answer Internal testing - Whole o o o o o o o o carcass Internal testing - Second o o o o o o o o processing

Q2.25 Based on the USDA FSIS sampling, what category is this plant with respect to: Prefer no to Category 1 Category 2 Category 3 Don't know answer

Campylobacter o o o o o

Salmonella o o o o o

Q2.26 Does your processing establishment have validated interventions throughout processing to specifically reduce Campylobacter contamination on product? o Yes o No o Don't know

Q2.27 Does your processing establishment have validated interventions throughout processing to specifically reduce Salmonella contamination on product? o Yes o No o Don't know

Q2.28 Is the Campylobacter contamination status of incoming flocks known prior to slaughter? o Yes, microbiological test results are available for every incoming flock o Yes, microbiological test results are available for a subset of incoming flocks o No o Don't know

174

Q2.29 Do you use these data for logistic slaughter (i.e. are Campylobacter-free flocks processed before Campylobacter-positive flocks)? o Yes o No o Don't know

Q2.30 Do you use these data to inform product disposition (i.e. send meat from Campylobacter- positive flocks to a cooked product line)? o Yes o No o Don't know

Q2.31 Is the Salmonella contamination status of incoming flocks known prior to slaughter? o Yes, microbiological test results are available for every incoming flock o Yes, microbiological test results are available for a subset of incoming flocks o No o Don't know

Q2.32 Do you use these data for logistic slaughter (i.e. are Salmonella-free flocks processed before Salmonella-positive flocks)? o Yes o No o Don't know

Q2.33 Do you use these data to inform product disposition (i.e. send meat from Salmonella- positive flocks to a cooked product line)? o Yes o No o Don't know

Q2.34 In your opinion, is a Campylobacter-positive flock more likely to be contaminated with Salmonella?

o Yes o No o Don't know

175

Q2.35 In your opinion, is a Salmonella-positive flock more likely to be contaminated with Campylobacter? o Yes o No o Don't know

Q2.36 Please rank the following processing steps in order of importance for minimizing Campylobacter contamination on product. 1 = most important 7 = least important

To rank these choices, simply click and drag the choice to its appropriate place. ______Live haul ______Scalding ______Picking ______Evisceration ______Chilling ______Post-chill ______Further processing into parts

Q2.37 Are transportation modules cleaned between flocks? o Yes o No o I do not know because contract live haul crews are responsible for cleaning modules o Don't know

Q2.38 In your opinion, does cleaning transportation vehicles, transportation modules and/or catching equipment help to reduce cross-contamination of Campylobacter and/or Salmonella between flocks? Yes No Don't know

Campylobacter o o o

Salmonella o o o

176

Q2.39 The following questions will ask about current procedures during scalding that are implemented in your processing establishment to reduce Campylobacter and/or Salmonella contamination on carcasses.

Q2.40 Which of the following scalding methods is currently being used and how long has it been used in your plant? Less than 1 year 1 year or more Don't know Not used Immersion scalding o o o o Steam spraying scalding o o o o

Q2.41 Why did your establishment switch to a different scalding method? Please check all that apply. ▢ The previous scalding method was difficult to maintain ▢ The previous scalding method was difficult to clean ▢ The previous scalding method was damaging carcasses ▢ The previous scalding method was causing microbial cross-contamination (Campylobacter or Salmonella) between carcasses ▢ Other - please specify your answer below

Q2.42 Which of the following procedures are currently being used during immersion scalding and how long have they been used? Less than 1 year 1 year or more Don't know Not used Pre-scald brush system o o o o Counter flow of process water o o o o Multiple stage tanks o o o o Administration of interventions to scald water (e.g. o o o o PAA, salt, adjust pH)

Post-scald wash o o o o Administration of interventions to post-scald wash o o o o (e.g. PAA, salt, adjust pH)

177

Q2.43 Which of the following procedures are currently being used during steam scalding and how long have they been used? Less than 1 year 1 year or more Don't know Not used Pre-scald brush system o o o o

Post-scald wash o o o o Administration of interventions to post-scald wash o o o o (e.g. PAA, salt, adjust pH)

Q2.44 Were any of the recent changes in scalding procedures made to help meet the FSIS performance standards for Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q2.45 In your opinion, are the current scalding practices effective at reducing Campylobacter contamination on carcasses? o Yes o No o Don't know

178

Q2.46 The following questions will ask about current procedures during picking that are implemented in your processing establishment to reduce Campylobacter and/or Salmonella contamination on carcasses.

Q2.47 Which of the following procedures are currently being used during picking and how long have they been used? Less than 1 year 1 year or more Don't know Not used Continuous rinsing of carcasses during o o o o picking Continuous rinsing of equipment o o o o during picking Post-pick application of chemicals or antimicrobial o o o o agents (in wash, dip or other manner)

Q2.48 Were any of the recent changes in picking procedures made to help meet the FSIS performance standards for Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q2.49 In your opinion, are the current picking practices effective at reducing Campylobacter contamination on carcasses? o Yes o No o Don't know

179

Q2.50 The following questions will ask about current procedures during evisceration that are implemented in your processing establishment to reduce Campylobacter and/or Salmonella contamination on carcasses.

Q2.51 Which of the following procedures are currently being used during evisceration and how long have they been used? Less than 1 year 1 year or more Don't know Not used

Automated re-hang o o o o Daily replacement of opener blades o o o o Whole carcass rinse during o o o o evisceration Post-evisceration wash o o o o Application of chemicals or antimicrobial o o o o agents in post- evisceration wash

Q2.52 Were any of the recent changes in evisceration procedures made to help meet the FSIS performance standards for Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q2.53 In your opinion, are the current evisceration practices effective at reducing Campylobacter contamination on carcasses? o Yes o No o Don't know

180

Q2.54 The following questions will ask about current procedures during chilling implemented in your processing establishment to reduce Campylobacter and/or Salmonella contamination on carcasses.

Q2.55 Which chilling method are you currently using and how long has it been used? Less than 1 year 1 year or more Don't know Not used

Air chilling o o o o

Immersion chilling o o o o

Q2.56 Why did your processing establishment switch to a different chilling method? Please check all that apply. ▢ The previous chilling method was difficult to operate and/or maintain ▢ The previous chilling method was causing microbial cross-contamination (Campylobacter and/or Salmonella) between carcasses ▢ The previous chilling method was not effective at meeting regulation requirements (e.g. carcass moisture retention requirement) ▢ Other - please specify your answer below

Q2.57 Which of the following procedures are currently being used during immersion chilling and how long have they been used?

Less than 1 year 1 year or more Don't know Not used Use of pre-chill tanks o o o o Antimicrobial treatment in pre- o o o o chill tanks Use of counter- current flow during o o o o chilling Addition of chlorine in chill o o o o water Addition of other antimicrobial agents in chill o o o o water Use of Oxidation Reduction Potential o o o o (ORP) meter

181

Q2.58 Which of the following procedures are currently being used during air chilling and how long have they been used? Less than 1 year 1 year or more Don't know Not used Use of a pre-chill stage o o o o Antimicrobial treatment in a pre- o o o o chill stage Addition of chlorine in chill air o o o o Regular maintenance of o o o o chains and shackles

Q2.59 Were any of the recent changes in chilling procedures made to help meet the FSIS performance standards for Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q2.60 In your opinion, are the current chilling practices effective at reducing Campylobacter contamination on carcasses? o Yes o No o Don't know

Q2.61 The following questions will ask about current procedures during second process that are implemented in your processing establishment to reduce Campylobacter and/or Salmonella contamination on carcasses.

Q2.62 Which of the following procedures are currently being used during second process and how long have they been used? Less than 1 year 1 year or more Don't know Not used Application of chemicals or antimicrobial agents (in wash, o o o o dip or other manner)

Belt sanitation o o o o

182

Q2.63 Were any of the recent changes in second process made to help meet the FSIS performance standards for Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q2.64 In your opinion, are the current second process practices effective at reducing Campylobacter contamination on carcasses? o Yes o No o Don't know

183

Poultry company farm supervisor/operations manager

Q4.1 The following questions will ask about current USDA Food Safety and Inspection Service (FSIS) performance standards for Campylobacter and Salmonella in poultry products.

For each question below, please select the response that best describes your understanding of the topic.

Q4.2 Please indicate how familiar you are with the FSIS performance standards for Campylobacter and Salmonella that were implemented for whole birds in 2011 and for chicken parts in July 2016. o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q4.3 Are you currently meeting the performance standards for Campylobacter and Salmonella as determined by one or both of the following approaches? Yes No Don't know N/A

Internal testing o o o o

USDA testing o o o o

Q4.4 Are there any changes (i.e. new control strategy or intervention applied pre- or post-harvest) that have been implemented on farms under your supervision to control Campylobacter to help meet the new performance standards? o Yes o No, no changes have been made regarding Campylobacter control on farm(s) under my supervision o I don't know if any changes have been made regarding Campylobacter control on farm(s) under my supervision

184

Q4.5 What changes or recommendations have you made specifically to meet the new performance standards for Campylobacter? Please check all that apply. ▢ Implement new on-farm interventions (such as vaccines, chemicals, prebiotics and probiotics, antibiotics) to reduce Campylobacter colonization in birds ▢ Implement or improve biosecurity and other management strategies on-farm to reduce Campylobacter colonization in birds ▢ Implement on-farm microbiological testing to check Campylobacter prevalence or contamination levels ▢ Implement new post-harvest interventions in the processing facility to reduce Campylobacter contamination on the final product ▢ Implement in-plant microbiological testing to check Campylobacter contamination levels during broiler processing ▢ Other - Please specify your answer below

Q4.6 Are there any changes (i.e. new control strategy or intervention applied pre- or post-harvest) that have been implemented on farms under your supervision to control Salmonella to help meet the new performance standards? o Yes o No, no changes have been made regarding Salmonella control on farm(s) under my supervision o I don't know if any changes have been made regarding Salmonella control on farm(s) under my supervision

Q4.7 What changes or recommendations have you made specifically to meet the new performance standards for Salmonella? Please check all that apply. ▢ Implement new on-farm interventions (such as vaccines, chemicals, prebiotics and probiotics, antibiotics) to reduce Salmonella colonization in birds ▢ Implement or improve biosecurity and other management strategies on-farm to reduce Salmonella colonization in birds ▢ Implement on-farm microbiological testing to check Salmonella prevalence or contamination levels ▢ Implement new post-harvest interventions in the processing facility to reduce Salmonella contamination in final products ▢ Implement in-plant microbiological testing to check Salmonella contamination levels during broiler processing ▢ Other - Please specify your answer below

185

Q4.8 The following questions will ask about your understanding of Campylobacter in poultry. For each question below, please select the response that best describes your understanding.

Broiler chicks can be colonized with Campylobacter at hatch if the breeding o True o False o Don't know hens are colonized with Campylobacter. Campylobacter- negative broilers can acquire Campylobacter from the environment o True o False o Don't know (soil, water, insects, etc.). Campylobacter-positive broilers shed Campylobacter in their o True o False o Don't know feces. Effective on-farm biosecurity can completely eliminate o True o False o Don't know Campylobacter from the broiler farm. Farm equipment such as transportation trucks and modules can be o True o False o Don't know contaminated with Campylobacter. The incidence of Campylobacter decreases during second o True o False o Don't know processing.

Q4.9 Are you familiar with factors on the farm that increase the risk of Campylobacter colonization of the broiler farm environment? o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

186

Q4.10 Are you familiar with factors in the processing plant that increase the risk of Campylobacter contamination of the final product? o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q4.11 In your opinion, which of the following on-farm interventions or management practices are the most effective at reducing Campylobacter colonization in broiler chickens? Please rank the following on-farm interventions or management practices from MOST EFFECTIVE to LEAST EFFECTIVE for controlling Campylobacter. To rank these choices, simply click and drag the choice to its appropriate place. ______Improvement in biosecurity ______Improvement in drinking water quality ______Improvement in litter quality ______Administration of antibiotics ______Administration of feed supplements such as prebiotics and/or probiotics

Q4.12 If a Campylobacter vaccine that can reduce Campylobacter (but cannot eliminate it) were to be commercially available, how much would you be willing to pay per 1000 chickens to vaccinate flocks that are under your supervision? o Less than $8.00 per 1000 chickens o $8.01 ~ $12.00 per 1000 chickens o $12.01 ~ $16.00 per 1000 chickens o Whatever the asking price o Would not pay for a Campylobacter vaccine

Q4.13 If a Campylobacter vaccine that can reduce and potentially eliminate Campylobacter from chickens were to be commercially available, how much would you be willing to pay per 1000 chickens to vaccinate flocks that are under your supervision? o Less than $8.00 per 1000 chickens o $8.01 ~ $12.00 per 1000 chickens o $12.01 ~ $16.00 per 1000 chickens o Whatever the asking price o Would not pay for a Campylobacter vaccine

187

Q4.14 If a Campylobacter and Salmonella combination vaccine that can reduce Campylobacter and Salmonella from chickens were to be commercially available, how much would you be willing to pay per 1000 chickens to vaccinate flocks that are under your supervision? o Less than $12.00 per 1000 chickens o $12.01 ~ $16.00 per 1000 chickens o $16.01 ~ $20.00 per 1000 chickens o Whatever the asking price o Would not pay for a combination vaccine

Q4.15 The following questions will ask about the broiler company at which you are currently employed and the broiler farms you are supervising.

Q4.16 How many broiler farms are under your supervision?

______

Q4.17 Do farms under your supervision regularly test positive in the slaughter plant for: Prefer not to Yes No Don't know answer

Campylobacter o o o o

Salmonella o o o o

Q4.18 Has your company had trouble meeting the USDA-FSIS performance standards for Campylobacter or Salmonella on the final product? Prefer not to Yes No Don't know answer

Campylobacter o o o o

Salmonella o o o o

188

Q4.19 Does your company require farms to practice feed withdrawal before birds are sent to slaughter? o Yes o No o Don't know

Q4.20 What is the recommended feed withdrawal period prior to load out? o At least 4 hours o At least 6 hours o At least 8 hours o At least 10 hours o At least 12 hours o Other - please specify your answer below

Q4.21 Does your company conduct on-farm microbiological testing to detect and/or measure Campylobacter in flocks? o Yes o No o Don't know

Q4.22 Please specify when farms are sampled for Campylobacter contamination. Please check all that apply. ▢ During brooding ▢ During growout ▢ During the feed withdrawal period ▢ Other - Please specify your answer below

Q4.23 Which of the following test results are available from the on-farm microbiological testing? Please check all that apply. *CFU = colony forming unit

▢ Detection (presence/absence) of Campylobacter ▢ Enumeration/quantification (logCFU or CFU) of Campylobacter ▢ Other - please specify your answer below ______

189

Q4.24 Does your company conduct on-farm microbiological testing to detect and/or measure Salmonella in flocks? o Yes o No o Don't know

Q4.25 Please specify when farms are sampled for Salmonella contamination. Please check all that apply. ▢ During brooding ▢ During growout ▢ During the feed withdrawal period ▢ Other - Please specify your answer below ______

Q4.26 Which of the following test results are available from the on-farm microbiological testing? Please check all that apply. *CFU = colony forming unit

▢ Detection (presence/absence) of Salmonella ▢ Enumeration/quantification (logCFU or CFU) of Salmonella ▢ Other - please specify your answer below ______

Q4.27 In your opinion, is a flock that is colonized with Salmonella more likely to be colonized with Campylobacter compared to a flock that is Salmonella negative? o Yes o No o Don't know

190

Q4.28 The following questions will ask about current drinking water interventions to reduce Campylobacter and/or Salmonella colonization in birds. Q4.29 What percentage of farms under your supervision use the following drinking water interventions? None 1 - 25% 26-50% 51- 99% All farms Sanitation using chlorination or o o o o o hypochlorination Sanitation using oxidizers (chlorine dioxide, hydrogen o o o o o peroxide, peracetic acid, etc.) Administration of acidifier o o o o o

Other o o o o o

Q4.30 Please specify the chlorination or hypochlorination product or chemical that is used for drinking water sanitation and when the product or chemical is used. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢

Q4.31 Please specify the oxidizer product or chemical that is used for drinking water and when the product or chemical is used. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

191

Q4.32 Please specify the water acidifier product or chemical that is used for drinking water and when the product or chemical is used. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q4.33 Were any of the drinking water interventions listed above implemented specifically to help meet the FSIS performance standards on Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q4.34 In your opinion, does chlorination or hypochlorination reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q4.35 In your opinion, does sanitizing drinking water using an oxidizer reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o Q4.36 In your opinion, does administering a water acidifier reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

192

Q4.37 The following questions will ask about current litter management practices to reduce Campylobacter and/or Salmonella colonization in birds.

Q4.38 What percentage of farms under your supervision use the following litter management practices? None 1 - 25% 26 - 50% 51 - 99% All farms Chemical treatment of litter (such as o o o o o litter acidifier) Biological treatment of litter o o o o o (microbiological or enzymatic)

Other o o o o o

Q4.39 Please specify the chemical litter treatment that is used and when the product or chemical is applied to the litter. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q4.40 Please specify the biological litter treatment that is used and when the product is applied to the litter. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

193

Q4.41 Were any of the litter management practices listed above implemented specifically to help meet the FSIS performance standards on Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q4.42 In your opinion, does chemical treatment of litter reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q4.43 In your opinion, does biological treatment of litter reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q4.44 The following questions will ask about current feed practices to reduce Campylobacter and/or Salmonella colonization in birds.

Q4.45 What percentage of farms under your supervision use the following feed practices? None 1 - 25% 26 - 50% 51 - 99% All farms Administration of feed acidifier o o o o o Administration of prebiotics o o o o o Administration of probiotics o o o o o

Other o o o o o

194

Q4.46 Please specify the feed acidifier product or chemical that is used and when the product or chemical is administered. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q4.47 Please specify the prebiotics product that is used and when the product is administered. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q4.48 Please specify the probiotics product that is used and when the product is administered. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

195

Q4.49 Were any of the feed practices listed above implemented specifically to help meet the FSIS performance standards on Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q4.50 In your opinion, do feed acidifiers reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q4.51 In your opinion, does administering prebiotics reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q4.52 In your opinion, does administering probiotics including competitive exclusion treatment and direct-fed microbials reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q4.53 The following questions will ask about current biosecurity measures adopted on farms to reduce Campylobacter and/or Salmonella colonization in birds.

Q4.54 Does your company require farms to have a farm-specific biosecurity management plan for Campylobacter? o Yes o No o Don't know

196

Q4.55 Does your company require farms to have a farm-specific biosecurity management plan for Salmonella? o Yes o No o Don't know

Q4.56 Please select biosecurity measures currently adopted on the farms under your supervision. Please check all that apply. ▢ Separate anteroom for each poultry house ▢ Use of bootdip ▢ Visitor control ▢ Livestock or pet control ▢ Rodent management plan ▢ Darkling beetle management plan ▢ Other pest management plan ▢ Cleaning and disinfection of poultry houses and equipment ▢ Annual (or more frequent) worker training on hygiene and biosecurity practices ▢ Veterinary consultation ▢ Other - Please specify your answer below ▢ Don't know

Q4.57 In your opinion, does improvement in biosecurity of farms reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

197

Poultry veterinarian

Q5.1 The following questions will ask about current USDA Food Safety and Inspection Service (FSIS) performance standards for Campylobacter and Salmonella in poultry products.

For each question below, please select the response that best describes your understanding of the topic.

Q5.2 Please indicate how familiar you are with the FSIS performance standards for Campylobacter and Salmonella that were implemented for whole birds in 2011 and for chicken parts in July 2016. o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q5.3 Are you currently meeting the performance standards for Campylobacter and Salmonella as determined by one or both of the following approaches? Yes No Don't know N/A

Internal testing o o o o

USDA testing o o o o

Q5.4 Are there any changes (i.e. new control strategy or intervention applied pre- or post-harvest) that have been implemented on farms under your veterinary supervision to control Campylobacter to help meet the new performance standards? o Yes o No, no changes have been made regarding Campylobacter control on farm(s) under my supervision o I don't know if any changes have been made regarding Campylobacter control on farm(s) under my supervision

198

Q5.5 What changes or recommendations have you made specifically to meet the new performance standards for Campylobacter? Please check all that apply. ▢ Implement new on-farm interventions (such as vaccines, chemicals, prebiotics and probiotics, antibiotics) to reduce Campylobacter colonization in birds ▢ Implement or improve biosecurity and other management strategies on-farm to reduce Campylobacter colonization in birds ▢ Implement on-farm microbiological testing to check Campylobacter prevalence or contamination levels ▢ Implement new post-harvest interventions in the processing facility to reduce Campylobacter contamination on the final product ▢ Implement in-plant microbiological testing to check Campylobacter contamination levels during broiler processing ▢ Other - Please specify your answer below ______

Q5.6 Are there any changes (i.e. new control strategy or intervention applied pre- or post-harvest) that have been implemented on farms under your veterinary supervision to control Salmonella to help meet the new performance standards? o Yes o No, no changes have been made regarding Salmonella control in farm(s) under my supervision o I don't know if any changes have been made regarding Salmonella control in farm(s) under my supervision

Q5.7 What changes or recommendations have you made specifically to meet the new performance standards for Salmonella? Please check all that apply. ▢ Implement new on-farm interventions (such as vaccines, chemicals, prebiotics and probiotics, antibiotics) to reduce Salmonella colonization in birds ▢ Implement or improve biosecurity and other management strategies on-farm to reduce Salmonella colonization in birds ▢ Implement on-farm microbiological testing to check Salmonella prevalence or contamination levels ▢ Implement new post-harvest interventions in the processing facility to reduce Salmonella contamination in final products ▢ Implement in-plant microbiological testing to check Salmonella contamination levels during broiler processing ▢ Other - Please specify your answer below ______

199

Q5.8 The following questions will ask about your understanding of Campylobacter in poultry. For each question below, please select the response that best describes your understanding.

Broiler chicks can be colonized with Campylobacter at hatch if the breeding o True o False o Don't know hens are colonized with Campylobacter. Campylobacter- negative broilers can acquire Campylobacter from the environment o True o False o Don't know (soil, water, insects, etc.). Campylobacter-positive broilers shed Campylobacter in their o True o False o Don't know feces. Effective on-farm biosecurity can completely eliminate o True o False o Don't know Campylobacter from the broiler farm. Farm equipment such as transportation trucks and modules can be o True o False o Don't know contaminated with Campylobacter. The incidence of Campylobacter decreases during second o True o False o Don't know processing.

Q5.9 Are you familiar with factors on the farm that increase the risk of Campylobacter colonization of the broiler farm environment? o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

200

Q5.10 Are you familiar with factors in the processing plant that increase the risk of Campylobacter contamination of the final product? o Not familiar at all o Slightly familiar o Moderately familiar o Very familiar o Extremely familiar

Q5.11 In your opinion, which of the following on-farm interventions or management practices are the most effective at reducing Campylobacter colonization in broiler chickens? Please rank the following on-farm interventions or management practices from MOST EFFECTIVE to LEAST EFFECTIVE for controlling Campylobacter.

Q5.12 If a Campylobacter vaccine that can reduce Campylobacter (but cannot eliminate it) were to be commercially available, how much would you be willing to pay per 1000 chickens to vaccinate flocks that are under your supervision? o Less than $8.00 per 1000 chickens o $8.01 ~ $12.00 per 1000 chickens o $12.01 ~ $16.00 per 1000 chickens o Whatever the asking price o Would not pay for a Campylobacter vaccine

Q5.13 If a Campylobacter vaccine that can reduce and potentially eliminate Campylobacter from chickens were to be commercially available, how much would you be willing to pay per 1000 chickens to vaccinate flocks that are under your supervision? o Less than $8.00 per 1000 chickens o $8.01 ~ $12.00 per 1000 chickens o $12.01 ~ $16.00 per 1000 chickens o Whatever the asking price o Would not pay for a Campylobacter vaccine

201

Q5.14 If a Campylobacter and Salmonella combination vaccine that can reduce Campylobacter and Salmonella from chickens were to be commercially available, how much would you be willing to pay per 1000 chickens to vaccinate flocks that are under your supervision? o Less than $12.00 per 1000 chickens o $12.01 ~ $16.00 per 1000 chickens o $16.01 ~ $20.00 per 1000 chickens o Whatever the asking price o Would not pay for a combination vaccine

Q5.15 The following questions are related to basic information about the broiler farms under your veterinary supervision.

Q5.16 Approximately how many farms are under your veterinary supervision?

______

Q5.17 Do farms under your veterinary supervision regularly test positive in the slaughter plant for: Yes No Don't know

Campylobacter o o o

Salmonella o o o Q5.18 Has your company had trouble meeting the USDA-FSIS performance standards for Campylobacter or Salmonella on the final product? Prefer not to Yes No Don't know answer

Campylobacter o o o o

Salmonella o o o o

Q5.19 Do farms under your veterinary supervision practice feed withdrawal before birds are sent to slaughter? o Yes o No o Don't know

202

Q5.20 What is your recommended feed withdrawal period prior to load out? o At least 4 hours o At least 6 hours o At least 8 hours o At least 10 hours o At least 12 hours o Other - please specify your answer below ______

Q5.21 Does your company conduct on-farm microbiological testing to detect and/or measure Campylobacter in flocks? o Yes o No o Don't know

Q5.22 Please specify when farms are sampled for Campylobacter contamination. Please check all that apply. ▢ During brooding ▢ During growout ▢ During the feed withdrawal period ▢ Other - Please specify your answer below ______

Q5.23 Which of the following test results are available from the on-farm microbiological testing? Please check all that apply. *CFU = colony forming unit

▢ Detection (presence/absence) of Campylobacter ▢ Enumeration/quantification (logCFU or CFU) of Campylobacter ▢ Other - please specify your answer below ______

Q5.24 Does your company conduct on-farm microbiological testing to detect and/or measure Salmonella in flocks? o Yes o No o Don't know

203

Q5.25 Please specify when farms are sampled for Salmonella contamination. Please check all that apply. ▢ During brooding ▢ During growout ▢ During the feed withdrawal period ▢ Other - Please specify your answer below ______

Q5.26 Which of the following test results are available from the on-farm microbiological testing? Please check all that apply. *CFU = colony forming unit

▢ Detection (presence/absence) of Salmonella ▢ Enumeration/quantification (logCFU or CFU) of Salmonella ▢ Other - please specify your answer below ______

Q5.27 In your opinion, is a flock that is colonized with Salmonella more likely to be colonized with Campylobacter compared to a flock that is Salmonella negative? o Yes o No o Don't know

Q5.28 The following questions will ask about current drinking water interventions to reduce Campylobacter and/or Salmonella colonization in birds. Q5.29 What percentage of farms under your veterinary supervision use the following drinking water interventions? None 1 - 25% 26-50% 51- 99% All farms Sanitation using chlorination or o o o o o hypochlorination Sanitation using oxidizers (chlorine dioxide, hydrogen o o o o o peroxide, peracetic acid, etc.) Administration of acidifier o o o o o

Other o o o o o

204

Q5.30 Please specify the chlorination/hypochlorination product or chemical that is used in the drinking water and when the product or chemical is used. Please check all that apply.

Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢

Q5.31 Please specify the oxidizer product or chemical that is used in the drinking water and when the product or chemical is used. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q5.32 Please specify the water acidifier product or chemical that is used in the drinking water and when the product or chemical is used. Please check all that apply.

Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

205

Q5.33 Were any of the drinking water interventions listed above implemented specifically to help meet the FSIS performance standards on Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q5.34 In your opinion, does chlorination or hypochlorination reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q5.35 In your opinion, does sanitizing drinking water using an oxidizer reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q5.36 In your opinion, does administering a water acidifier reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

206

Q5.37 The following questions will ask about current litter management practices to reduce Campylobacter and/or Salmonella colonization in birds

Q5.38 What percentage of farms under your veterinary supervision use the following litter management practices? None 1 - 25% 26 - 50% 51 - 99% All farms Chemical treatment of litter (such as o o o o o litter acidifier) Biological treatment of litter o o o o o (microbiological or enzymatic)

Other o o o o o

Q5.39 Please specify the chemical litter treatment that is used and when the product or chemical is applied to the litter. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q5.40 Please specify the biological litter treatment that is used and when the product is applied to the litter. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

207

Q5.41 Were any of the litter management practices listed above implemented specifically to help meet the FSIS performance standards for Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q5.42 In your opinion, does chemical treatment of litter reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacer o o o

Salmonella o o o

Q5.43 In your opinion, does biological treatment of litter reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q5.44 The following questions will ask about current feed practices to reduce Campylobacter and/or Salmonella colonization in birds.

Q5.45 What percentage of farms under your veterinary supervision use the following feed practices? None 1 - 25% 26 - 50% 51 - 99% All farms Administration of feed acidifier o o o o o Administration of prebiotics o o o o o Administration of probiotics o o o o o

Other o o o o o

208

Q5.46 Please specify the feed acidifier product or chemical that is used and when the product or chemical is administered. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q5.47 Please specify the prebiotics that are used and when they are administered. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

Q5.48 Please specify the probiotics that are used and when they are administered. Please check all that apply. Feed Cleanout Brooding Growout withdrawal Don't know period Product/chemical 1 ▢ ▢ ▢ ▢ ▢ Product/chemical 2 ▢ ▢ ▢ ▢ ▢ Product/chemical 3 ▢ ▢ ▢ ▢ ▢

209

Q5.49 Were any of the feed practices listed above implemented specifically to help meet the FSIS performance standards on Campylobacter and/or Salmonella? o Yes o No o Don't know o N/A

Q5.50 In your opinion, do feed acidifiers reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q5.51 In your opinion, does administering prebiotics reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q5.52 In your opinion, does administering probiotics including competitive exclusion treatment and direct-fed microbials reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

Q5.53 The following questions will ask about current biosecurity measures adopted on farms to reduce Campylobacter and/or Salmonella colonization in birds.

Q5.54 Does your company require farms to have a farm-specific biosecurity management plan for Campylobacter? o Yes o No o Don't know

210

Q5.55 Does your company require farms to have a farm-specific biosecurity management plan for Salmonella? o Yes o No o Don't know

Q5.56 Please select biosecurity measures currently adopted on the farms under your veterinary supervision. Please check all that apply. ▢ Separate anteroom for each poultry house ▢ Use of bootdip ▢ Visitor control ▢ Livestock or pet control ▢ Rodent management plan ▢ Darkling beetle management plan ▢ Other pest management plan ▢ Cleaning and disinfection of poultry houses and equipment (9) ▢ Annual (or more frequent) worker training on hygiene and biosecurity practices (10) ▢ Veterinary consultation ▢ Other - Please specify your answer below ▢ Don't know

Q5.57 In your opinion, does improvement in biosecurity of farms reduce Campylobacter and/or Salmonella colonization in birds? Yes No Don't know

Campylobacter o o o

Salmonella o o o

211

Q3.1 Enter any additional comments about this production unit in the space below.

______

Q3.2 Enter any comments about this survey in the space below.

______

Q3.3 If you have any questions about this survey, please contact Dr. Randall Singer at [email protected]. Click the button below to complete the survey

o Complete and Submit survey

212