Microbiota from Specific Pathogen-Free Mice Reduces Campylobacter jejuni Chicken Colonization

Campylobacter jejuni, a prevalent foodborne bacterial pathogen, is mainly transmitted from poultry with few effective prevention approaches. In this study, we aimed to investigate the role of microbiota on C. jejuni chicken colonization. Microbiota from specific pathogen-free (SPF) mouse stools were collected as SPF-Aerobe and SPF-Anaerobe. Birds were colonized with SPF-Aerobe or SPF-Anaerobe at day 0 and infected with C. jejuni AR101 at day 12. Notably, C. jejuni AR101 colonized at 5.3 and 5.6 log10 C. jejuni CFU/g chicken cecal digesta at days 21 and 28, respectively, while both SPF-Aerobe and SPF-Anaerobe microbiota reduced pathogen colonization. Notably, SPF-Aerobe and SPF-Anaerobe increased cecal phylum Bacteroidetes and reduced phylum Firmicutes compared to those in the nontransplanted birds. Interestingly, microbiota from noninfected chickens, SPF-Aerobe, or SPF-Anaerobe inhibited AR101 in vitro growth, whereas microbiota from infected birds alone failed to reduce pathogen growth. The bacterium Enterobacter102 isolated from infected birds transplanted with SPF-Aerobe inhibited AR101 in vitro growth and reduced pathogen gut colonization in chickens. Together, SPF mouse microbiota was able to colonize chicken gut and reduce C. jejuni chicken colonization. The findings may help the development of effective strategies to reduce C. jejuni chicken contamination and campylobacteriosis.

To reduce campylobacteriosis, different measures have been implemented to reduce enteritis by reducing C. jejuni contamination in animal food, particularly pre-and post-harvest poultry. The intervention methods include strict biosecurity on farms [17], vaccines [18], probiotics [19], phages [20], decontamination of poultry carcasses in the post-slaughter process [21], facility design and management, reducing contamination in feed, transportation, and other sources, and other strategies [2]. It is estimated that decreasing Campylobacter

SPF-Aerobe and SPF-Anaerobe Modulated the Chicken Microbiota
We reasoned that the colonization reduction of AR101 in chickens might come from chicken microbiota alteration by the mouse microbiota transplantation. To assess this hypothesis, we used phylum-specific primers to analyze the microbiota composition change. Notably, SPF-Aerobe and SPF-Anaerobe reduced the relative abundance of the phylum Firmicutes compared to that of uninfected birds (61.3 and 52.9 vs. 97.5%) and infected birds (51.9 and 50.9 vs. 86.7%, respectively), while the relative abundance of Bacteroidetes was increased compared to that of uninfected birds (38.4 and 44.7 vs. 2.3%) and infected birds (42.0 and 47.6 vs. 12.4%, respectively) ( Figure 2). Interestingly, C. jejuni colonization modulated chicken cecal microbiota of the phyla Bacteroidetes and Firmicutes. Importantly, most of the relative abundance was significant (Table 1). These results indicate that SPF-Aerobe, SPF-Anaerobe, and C. jejuni were able to colonize and change the microbiota in the chicken gut.  Murine microbiota reduced C. jejuni AR101 chicken colonization. Zero-day-old broiler chickens precolonized with SPF-Aerobe and SPF-Anaerobe were infected with C. jejuni AR101 at 12 days of age. The birds were euthanized at days 21 and 28. The bird cecal digesta was collected, serially diluted, and cultured on Campylobacter-selective agar plates prepared in-house at 42 • C under microaerobic atmosphere. (A) C. jejuni chicken colonization in the ceca of the birds at day 21. The bird number for each group was: noninfected (n = 10), Cj AR101 (n = 10), SPF-Aerobe (n = 5), and SPF-Anaerobe (n = 10). (B) C. jejuni chicken colonization in the ceca of the birds at day 28. The bird number for each group was: noninfected (n = 20), Cj AR101 (n = 20), SPF-Aerobe (n = 10), and SPF-Anaerobe (n = 20). All graphs depict the mean + SEM. Significant if p < 0.05. The results are representative of three independent experiments.

SPF-Aerobe and SPF-Anaerobe Modulated the Chicken Microbiota
We reasoned that the colonization reduction of AR101 in chickens might come from chicken microbiota alteration by the mouse microbiota transplantation. To assess this hypothesis, we used phylum-specific primers to analyze the microbiota composition change. Notably, SPF-Aerobe and SPF-Anaerobe reduced the relative abundance of the phylum Firmicutes compared to that of uninfected birds (61.3 and 52.9 vs. 97.5%) and infected birds (51.9 and 50.9 vs. 86.7%, respectively), while the relative abundance of Bacteroidetes was increased compared to that of uninfected birds (38.4 and 44.7 vs. 2.3%) and infected birds (42.0 and 47.6 vs. 12.4%, respectively) ( Figure 2). Interestingly, C. jejuni colonization modulated chicken cecal microbiota of the phyla Bacteroidetes and Firmicutes. Importantly, most of the relative abundance was significant (Table 1). These results indicate that SPF-Aerobe, SPF-Anaerobe, and C. jejuni were able to colonize and change the microbiota in the chicken gut.

Chicken Noninfected Microbiota and Mouse SPF Microbiota Reduced C. jejuni Growth
Upon validation of transplanted mouse SPF microbiota reducing C. jejuni chicken colonization, we reasoned that the mouse SPF microbiota would directly inhibit C. jejuni AR101 growth, while the chicken microbiota would not. To examine this hypothesis, C. jejuni AR101 inoculum was co-cultured with microbiota from noninfected, SPF-Aerobe, and SPF-Anaerobe chickens in the Campylobacter Enrichment (CE) Broth at 42 • C for 24 h under anaerobic conditions. Notably, both SPF-Aerobe and SPF-Anaerobe reduced C. jejuni AR101 by more than 1-log compared to the C. jejuni AR101 culture-alone group (6.8 ± 0.2 and 6.3 ± 0.1 vs. 8.6 ± 0.3 log 10 C. jejuni CFU/mL, respectively) ( Figure 3). Interestingly, the microbiota from noninfected chickens also reduced C. jejuni AR101 growth by more than 2-log compared to that in the C. jejuni AR101 culture alone (6.4 ± 0.4 vs. 8.6 ± 0.3 log 10 C. jejuni CFU/mL, respectively). Notably, each microbiota at 24 h increased the number of CFU compared to that at 0 h ( Figure S2). serially diluted, and cultured on Campylobacter-selective agar plates prepared in-house at 42 °C under microaerobic atmosphere. (A) C. jejuni chicken colonization in the ceca of the birds at day 21. The bird number for each group was: noninfected (n = 10), Cj AR101 (n = 10), SPF-Aerobe (n = 5), and SPF-Anaerobe (n = 10). (B) C. jejuni chicken colonization in the ceca of the birds at day 28. The bird number for each group was: noninfected (n = 20), Cj AR101 (n = 20), SPF-Aerobe (n = 10), and SPF-Anaerobe (n = 20). All graphs depict the mean + SEM. Significant if p < 0.05. The results are representative of three independent experiments.

SPF-Aerobe and SPF-Anaerobe Modulated the Chicken Microbiota
We reasoned that the colonization reduction of AR101 in chickens might come from chicken microbiota alteration by the mouse microbiota transplantation. To assess this hypothesis, we used phylum-specific primers to analyze the microbiota composition change. Notably, SPF-Aerobe and SPF-Anaerobe reduced the relative abundance of the phylum Firmicutes compared to that of uninfected birds (61.3 and 52.9 vs. 97.5%) and infected birds (51.9 and 50.9 vs. 86.7%, respectively), while the relative abundance of Bacteroidetes was increased compared to that of uninfected birds (38.4 and 44.7 vs. 2.3%) and infected birds (42.0 and 47.6 vs. 12.4%, respectively) ( Figure 2). Interestingly, C. jejuni colonization modulated chicken cecal microbiota of the phyla Bacteroidetes and Firmicutes. Importantly, most of the relative abundance was significant (Table 1). These results indicate that SPF-Aerobe, SPF-Anaerobe, and C. jejuni were able to colonize and change the microbiota in the chicken gut.

Chicken Noninfected Microbiota and Mouse SPF Microbiota Reduced C. jejuni Growth
Upon validation of transplanted mouse SPF microbiota reducing C. jejuni chicken colonization, we reasoned that the mouse SPF microbiota would directly inhibit C. jejuni AR101 growth, while the chicken microbiota would not. To examine this hypothesis, C. jejuni AR101 inoculum was co-cultured with microbiota from noninfected, SPF-Aerobe, and SPF-Anaerobe chickens in the Campylobacter Enrichment (CE) Broth at 42 ˚C for 24 h under anaerobic conditions. Notably, both SPF-Aerobe and SPF-Anaerobe reduced C. jejuni AR101 by more than 1-log compared to the C. jejuni AR101 culture-alone group (6.8 ± 0.2 and 6.3 ± 0.1 vs. 8.6 ± 0.3 log10 C. jejuni CFU/ml, respectively) ( Figure 3). Interestingly, the microbiota from noninfected chickens also reduced C. jejuni AR101 growth by more than 2-log compared to that in the C. jejuni AR101 culture alone (6.4 ± 0.4 vs. 8.6 ± 0.3 log10 C. jejuni CFU/ml, respectively). Notably, each microbiota at 24 h increased the number of CFU compared to that at 0 h ( Figure S2).  Because of the unexpected result of noninfected chicken microbiota reducing C. jejuni AR 101 in vitro growth, we then modulated our hypothesis that C. jejuni possibly modulated chicken microbiota for its growth and colonization. To address this reasoning, we co-cultured C. jejuni AR101 with chicken microbiota from only infected birds (Cj-MB), transplanted with SPF-Aerobe and infected birds (Cj-SPF-Aerobe), and transplanted with SPF-Anaerobe and infected birds (Cj-SPF-Anaerobe). Interestingly, the three-chicken microbiota themselves could not grow a single colony on the Campylobacter-selective plates (Figure 4), suggesting that C. jejuni lost culturability after storing with microbiota. Notably, Cj-MB did not reduce C. jejuni AR101 in vitro growth compared to that in positive control of Cj AR101 culture alone. Consistently, Cj-SPF-Aerobe and Cj-SPF-Anaerobe reduced C. jejuni AR101 in vitro growth by more than 3-log compared to that in the Cj AR101 culture-alone group (3.7 ± 0.6 and 0.8 ± 0.5 vs. 7.3 ± 0.1 log 10 C. jejuni CFU/mL, respectively). Consistently, each microbiota at 24 h increased number of CFU compared to that at 0 h ( Figure S3). These results suggest that C. jejuni modulated the chicken microbiota for its growth and colonization, while the transplanted mouse SPF microbiota resisted against pathogen growth.
Because of the unexpected result of noninfected chicken microbiota reducing AR 101 in vitro growth, we then modulated our hypothesis that C. jejuni possibl lated chicken microbiota for its growth and colonization. To address this reason co-cultured C. jejuni AR101 with chicken microbiota from only infected birds transplanted with SPF-Aerobe and infected birds (Cj-SPF-Aerobe), and transplan SPF-Anaerobe and infected birds (Cj-SPF-Anaerobe). Interestingly, the three-chic crobiota themselves could not grow a single colony on the Campylobacter-selectiv (Figure 4), suggesting that C. jejuni lost culturability after storing with microbiot bly, Cj-MB did not reduce C. jejuni AR101 in vitro growth compared to that in control of Cj AR101 culture alone. Consistently, Cj-SPF-Aerobe and Cj-SPF-Anae duced C. jejuni AR101 in vitro growth by more than 3-log compared to that in the C culture-alone group (3.7 ± 0.6 and 0.8 ± 0.5 vs. 7.3 ± 0.1 log10 C. jejuni CFU/ml, respe Consistently, each microbiota at 24 h increased number of CFU compared to th ( Figure S3). These results suggest that C. jejuni modulated the chicken microbio growth and colonization, while the transplanted mouse SPF microbiota resisted pathogen growth. In vitro co-culture of C. jejuni-modulated microbiota and C. jejuni AR101. C. jeju was co-cultured with microbiota from infected-alone birds (Cj-MB), transplanted with SP and infected birds (Cj-SPF-Aerobe), and transplanted with SPF-Anaerobe and infected SPF-Anaerobe). AR101 growth was quantified by serially diluting and plating on the Camp selective agar plates. All graphs depict the mean + SEM. Significant if p < 0.05. The result resentative of three independent experiments.

An Aerobic Bacterial Isolate Reduced C. jejuni AR101 in vitro Growth
Next, we wanted to identify and isolate the individual bacteria from the pr SPF microbiota. About 100 bacterial colonies were isolated using BHI plates at 42 ° anaerobic conditions for 48 h. The colonies were individually co-cultured with for 24 h, and then C. jejuni was enumerated on the Campylobacter-selective plates p in-house. Unfortunately, none of the bacteria were able to inhibit C. jejuni grow the co-culture method. By accident, during one chicken trial, a bacterial colony fro gavaged with mouse SPF-Aerobe was able to grow with pink color on the Campy selective plate compared to the dark red color of C. jejuni. The bacterial colony lected and later named Enterobacter102. We reasoned that this bacterium might jejuni infection. Enterobacter102 was rod-shaped, stained Gram-negative, and had t size as E. coli. Enterobacter102 also grew in pink colonies on a MacConkey agar p DNA from Enterobacter102 was isolated, and 16S rDNA was PCR-amplified, Sa In vitro co-culture of C. jejuni-modulated microbiota and C. jejuni AR101. C. jejuni AR101 was co-cultured with microbiota from infected-alone birds (Cj-MB), transplanted with SPF-Aerobe and infected birds (Cj-SPF-Aerobe), and transplanted with SPF-Anaerobe and infected birds (Cj-SPF-Anaerobe). AR101 growth was quantified by serially diluting and plating on the Campylobacterselective agar plates. All graphs depict the mean + SEM. Significant if p < 0.05. The results are representative of three independent experiments.

An Aerobic Bacterial Isolate Reduced C. jejuni AR101 In Vitro Growth
Next, we wanted to identify and isolate the individual bacteria from the protective SPF microbiota. About 100 bacterial colonies were isolated using BHI plates at 42 • C under anaerobic conditions for 48 h. The colonies were individually co-cultured with C. jejuni for 24 h, and then C. jejuni was enumerated on the Campylobacter-selective plates prepared in-house. Unfortunately, none of the bacteria were able to inhibit C. jejuni growth using the co-culture method. By accident, during one chicken trial, a bacterial colony from birds gavaged with mouse SPF-Aerobe was able to grow with pink color on the Campylobacterselective plate compared to the dark red color of C. jejuni. The bacterial colony was selected and later named Enterobacter102. We reasoned that this bacterium might resist C. jejuni infection. Enterobacter102 was rod-shaped, stained Gram-negative, and had the same size as E. coli. Enterobacter102 also grew in pink colonies on a MacConkey agar plate. The DNA from Enterobacter102 was isolated, and 16S rDNA was PCR-amplified, Sangersequenced, and confirmed to be in 95% alignment with Enterobacter sp. To functionally dissect the interaction between Enterobacter102 and C. jejuni AR101, in vitro co-culture was performed. Interestingly, Enterobacter102 showed the ability to reduce C. jejuni AR101 colonization by more that 2-log compared to that in the Cj AR101 culture-alone group (4.6 ± 0.1 vs. 7.3 ± 0.1 log 10 C. jejuni CFU/mL) ( Figure 5). Not surprisingly, the number of Enterobacter102 increased at 24 h compared to that at 0 h ( Figure S4). Although the reduced C. jejuni could result from depleted nutrients in the presence of a microbiota (SPF microbiota or Enterobacter102), the comparable growth between Cj AR101 and Cj-MB AR101 in Figure 4 suggested that the microbiota was an important factor influencing C. jejuni growth. These results suggested that Enterobacter102 has potential to reduce C. jejuni AR101 chicken colonization.
quenced, and confirmed to be in 95% alignment with Enterobacter sp. sect the interaction between Enterobacter102 and C. jejuni AR101, in v performed. Interestingly, Enterobacter102 showed the ability to reduce onization by more that 2-log compared to that in the Cj AR101 cultur 0.1 vs. 7.3 ± 0.1 log10 C. jejuni CFU/ml) ( Figure 5). Not surprisingly, th bacter102 increased at 24 h compared to that at 0 h ( Figure S4). Altho jejuni could result from depleted nutrients in the presence of a microb or Enterobacter102), the comparable growth between Cj AR101 and Cj-4 suggested that the microbiota was an important factor influencin These results suggested that Enterobacter102 has potential to redu chicken colonization. Figure 5. In vitro co-culture of Enterobacter102 and C. jejuni AR101. C. jejuni A with Enterobacter102. AR101 growth was quantified by serially diluting and p ter-selective agar plates. All graphs depict the mean + SEM. Significant if p representative of three independent experiments.

Enterobacter102 Reduced C. jejuni AR101 Chicken Colonization
Encouraged by the result of Enterobacter102 reducing C. jejuni AR we then performed chicken experiments. The birds were colonized w Enterobacter102 at day 0, infected with C. jejuni at day 12, and euthan 28. Consistently with the in vitro experiments, Enterobacter102 redu chicken colonization by more than 1-log at day 21 in comparison to t birds of the Cj AR101 group (4.0 ± 0.0vs. 5.3 ± 0.4 log10 C. jejuni CFU/g c 6A). Notably, Enterobacter102 continued to reduce C. jejuni AR101 chic more than 2-log at day 28 (2.4 ± 0.9vs. 5.7 ± 0.2 log10 C. jejuni CFU/g ce 6B). These results suggest that the bacterial isolate of Enterobacter10 AR101 growth and reduced the pathogen's chicken colonization.

Enterobacter102 Reduced C. jejuni AR101 Chicken Colonization
Encouraged by the result of Enterobacter102 reducing C. jejuni AR101 in vitro growth, we then performed chicken experiments. The birds were colonized with 10 8 CFU/chick of Enterobacter102 at day 0, infected with C. jejuni at day 12, and euthanized at days 21 and 28. Consistently with the in vitro experiments, Enterobacter102 reduced C. jejuni AR101 chicken colonization by more than 1-log at day 21 in comparison to that in only infected birds of the Cj AR101 group (4.0 ± 0.0 vs. 5.3 ± 0.4 log 10 C. jejuni CFU/g cecal digesta) ( Figure 6A). Notably, Enterobacter102 continued to reduce C. jejuni AR101 chicken colonization by more than 2-log at day 28 (2.4 ± 0.9 vs. 5.7 ± 0.2 log 10 C. jejuni CFU/g cecal digesta) ( Figure 6B). These results suggest that the bacterial isolate of Enterobacter102 inhibited C. jejuni AR101 growth and reduced the pathogen's chicken colonization.

Discussion
Poultry is the main reservoir of the prevalent foodborne bacterial pathogen C. jejuni which asymptomatically colonizes the birds [39]. However, the pathogen fails to colonize SPF or conventionally raised mice [30,38]. We then hypothesized that the microbiota from SPF mice might resist against C. jejuni infection, while the chicken microbiota might be susceptible to the pathogen. Here we report that mouse SPF-Aerobe and SPF-Anaero microbiota reduced C. jejuni chicken colonization at days 21 and 28. Notably, SPF-Aerobe and SPF-Anaerobe increased chicken cecal phylum Bacteroidetes and reduced phylum Firmicutes compared to those in the infected-alone birds. Interestingly, the uninfected chicken microbiota, SPF-Aerobe, or SPF-Anaerobe inhibited AR101 in vitro growth. Microbiota from birds transplanted with SPF-Aerobe or SPF-Anaerobe and infected inhibited AR101 in vitro growth, whereas microbiota from C. jejuni-infected-alone birds did not. Enterobacter102 isolated from infected birds transplanted with SPF-Aerobe reduced AR101 in vitro growth and chicken colonization. Altogether, these findings revealed that mouse SPF microbiota is able to colonize the chicken gut and resists against C. jejuni colonization in chickens, suggesting a potential strategy to reduce C. jejuni chicken contamination.
A notable observation from this study is that the mouse microbiota was able to be successfully transplanted into chickens and to reduce C. jejuni chicken colonization. It is a well-known medical practice to transplant a healthy donor's microbiota to treat a human Clostridium perfringens infection [40]. The microbiota compositions of human recipients are comparable to those of the human donor's, and the C. difficile infection is reduced. Consistently, microbiota composition in recipient piglets is similar to that of human donors in an intermammalian microbiota transplantation [41], suggesting that it is feasible to transplant microbiota between animals within the class level of Mammalia. In the current study, we successfully transplanted mouse (class Mammalia) microbiota to chickens (class Aves), suggesting it is possible to transplant microbiota between animals within the phy-

Discussion
Poultry is the main reservoir of the prevalent foodborne bacterial pathogen C. jejuni which asymptomatically colonizes the birds [39]. However, the pathogen fails to colonize SPF or conventionally raised mice [30,38]. We then hypothesized that the microbiota from SPF mice might resist against C. jejuni infection, while the chicken microbiota might be susceptible to the pathogen. Here we report that mouse SPF-Aerobe and SPF-Anaero microbiota reduced C. jejuni chicken colonization at days 21 and 28. Notably, SPF-Aerobe and SPF-Anaerobe increased chicken cecal phylum Bacteroidetes and reduced phylum Firmicutes compared to those in the infected-alone birds. Interestingly, the uninfected chicken microbiota, SPF-Aerobe, or SPF-Anaerobe inhibited AR101 in vitro growth. Microbiota from birds transplanted with SPF-Aerobe or SPF-Anaerobe and infected inhibited AR101 in vitro growth, whereas microbiota from C. jejuni-infected-alone birds did not. Enterobacter102 isolated from infected birds transplanted with SPF-Aerobe reduced AR101 in vitro growth and chicken colonization. Altogether, these findings revealed that mouse SPF microbiota is able to colonize the chicken gut and resists against C. jejuni colonization in chickens, suggesting a potential strategy to reduce C. jejuni chicken contamination.
A notable observation from this study is that the mouse microbiota was able to be successfully transplanted into chickens and to reduce C. jejuni chicken colonization. It is a well-known medical practice to transplant a healthy donor's microbiota to treat a human Clostridium perfringens infection [40]. The microbiota compositions of human recipients are comparable to those of the human donor's, and the C. difficile infection is reduced. Consistently, microbiota composition in recipient piglets is similar to that of human donors in an intermammalian microbiota transplantation [41], suggesting that it is feasible to transplant microbiota between animals within the class level of Mammalia. In the current study, we successfully transplanted mouse (class Mammalia) microbiota to chickens (class Aves), suggesting it is possible to transplant microbiota between animals within the phylum level of Chordata. Apparently, the difference of body temperature (42 • C in chickens and 37 • C in mice) and intestinal anatomy between the animals did not reduce the donor mouse microbiota colonization in the recipient chickens. A meta-data analysis study showed that chicken microbiota at the phylum level is mainly comprised of 13 phyla, including Firmicutes (70.0%), Bacteroidetes (12.3%), Proteobacteria (9.3%), and other small proportions of Actinobacteria, Cyanobacteria, Spirochaetes, Synergisteles, Fusobacteria, Tenericutes, and Verrucomicrobia [42]. Consistent with this finding, we found that birds without a mouse microbiota transplantation had the phylum Firmicutes majority, while microbiota-transplanted birds dramatically reduced Firmicutes and increased Bacteroidetes, independently of C. jejuni infection. Interestingly, the microbiota in mice is composed of the phyla Firmicutes at 54% and Bacteroidetes at 30% [43], which is close to the composition of our transplanted chicken microbiota. A field survey study reported that birds from the farms with the highest Campylobacter counts show the highest percentage of Firmicutes and the lowest percentage of Bacteroidetes in their microbiota, although microbiota composition is highly variable between or within farms [44]. In addition, the significant reduction of C. jejuni colonization by SPF-Aerobe and SPF-Anaerobe microbiota in both days 21 and 28 suggested that the microbiota may continue to reduce pathogen colonization for a longer period of time. This experiment was cut short because of the pen size constrain. It would be interesting in the future to conduct follow-up experiments to reduce C. jejuni colonization by SPF microbiota for birds at the market age of days 35-45. Together, these data showed that mouse SPF microbiota is transplantable to reduce C. jejuni chicken colonization.
After the evaluation of the protective effect of the mouse SPF microbiota, it is imperative to isolate and identify individual bacteria in the microbiota against C. jejuni chicken colonization for further functional evaluation. In a human longevity study, Sato and colleagues have plate-cultured, isolated, and evaluated a group of 68 bile acid metabolizing bacteria [45]. They found that Parabacteroides merdae and Odoribacteraceae strains produced isoalloLCA and reduced Gram-positive multidrug-resistant pathogens, such as C. difficile and vancomycin-resistant Enterococcus faecium [45]. A microbiota with higher level of genera Clostridium XI, Bifidobacterium, and Lactobacillus is associated with resistance to C. jejuni-induced colitis in mice [30]. Interestingly, probiotics Bifidobacterium longum PCB133 and a xylo-oligosaccharide do not decrease C. jejuni chicken colonization [46]. We have cocultured C. jejuni with various ATCC or lab-isolated bacteria, such as Bifidobacterium longum and Clostridium scindens, and we did not find the bacteria to reduce C. jejuni in vitro growth (data not shown). During our search for individual microbiota against C. jejuni, we found that the Enterobacter102 from microbiota of SPF-Aerobe grew as pink colonies on the Campylobacter-selective plates. Later, we found that Enterobacter102 reduced C. jejuni in vitro growth and chicken colonization. Probiotic application of Enterobacter sp. improves both Mediterranean fruit fly (medfly) pupal and adult productivity and reduces rearing duration [47]. Most other reports showed that Enterobacter sp. is a pathogen and induces intesitnal inflammation [48,49]. Future research on how Enterobacter102 reduces C. jejuni growth and chicken colonization is much needed. We are working on identifying Enterobacter102 and other bacterial candidates by culture-isolation and 16S rDNA Sanger sequencing. Together, these data suggest that individual bacteria in the SPF microbiota might be able to be isolated and used to reduce C. jejuni growth and chicken colonization.
Another interesting finding from the current study is that the microbiota from noninfected birds at day 28 was able to reduce C. jejuni in vitro growth, while microbiota from infected-alone birds failed to reduce pathogen growth. The results suggest that C. jejuni might have modulated the chicken microbiota for facilitating pathogen colonizing and thriving in the gut. It is a consensus that intestinal microbiota influences C. jejuni colonization and induction of enteritis [24,50], as also discussed in the paragraphs above. However, few reports showed that C. jejuni modulates the microbiota to benefit its own colonization. Salmonella Enteritidis infection reduces the overall diversity of the chicken microbiota population with an expansion of the Enterobacteriaceae family for promoting pathogen colonization [51]. In the current study, we found that a C. jejuni infection increased the phylum Bacteroidetes compared to that in noninfected birds. Future research is needed to identify which specific bacteria are increased to facilitate C. jejuni colonization.
In conclusion, the mouse SPF microbiota was able to colonize chicken ceca and reduced C. jejuni chicken colonization. The reduction of C. jejuni chicken colonization might come from reduced bacteria in the phylum Firmicutes and/or increased bacteria in the phylum Bacteroidetes. Notably, Enterobacter102 reduced C. jejuni in vitro growth and chicken colonization. Altogether, these findings provide a feasible strategy to reduce C. jejuni chicken contamination and human campylobacteriosis.  Table S1. The birds obtained from Cobb-Vantress (Siloam Springs, AR, USA) were neck-tagged and randomly assigned to floor pens with a controlled age-appropriate environment. The birds were fed a corn-soybean meal-based starter diet during days 0-10 and a grower diet during days 11-28. The basal diet was formulated as described earlier [37,52]. Stool from eight-week-old SPF BL6 Il10 −/− mice fed a chew diet was freshly collected and immediately suspended in 30% glycerol PBS stock and stored at −80 • C. The stool samples were cultured on brain heart infusion (BHI, BD Biosciences, Franklin Lakes, NJ, USA) agar plates at 42 • C for 48 h under aerobic or anaerobic conditions using the GasPak system (BD Biosciences, Franklin Lakes, NJ, USA) and collected as SPF-Aerobe and SPF-Anaerobe microbiota. The microbiota was added glycerol at final 30% and stored at −80 • C. Before the chicken colonization experiment, the SPF-Aerobe and SPF-Anaerobe microbiota were cultured on a BHI plate for 48 h, collected in PBS, and enumerated by OD 600 and plating. OD 600 of 1 was estimated at about 10 8 CFU/mL. At chicken experiments, chicks at day 0 were orally gavaged once with PBS or 10 8 CFU/bird SPF-Aerobe or 10 8 CFU/bird SPF-Anaerobe. For the chicken experiment of Enterobacter102, a total of 90 zero-day-old Cobb 500 broiler chicks were randomly allocated into cohorts of 30 birds per group. The birds were fed and raised similarly to those in the SPF microbiota experiment. The chickens were orally gavaged once with PBS or 10 8 CFU/bird of Enterobacter102 in 0.5 mL/bird on day 0. Two days before infection, frozen stock of C. jejuni AR101 (isolated at Dr. Billy Hargis's lab at University of Arkansas at Fayetteville) were cultured microaerobically at 42 • C for 48 h on C. jejuni-selective blood plates. The motility of C. jejuni was ensured under a microscope as described before [53] and routinely examined on semisolid MH (0.4% agar) plates. C. jejuni AR101 in PBS was estimated as that OD 600 of 0.468 was 10 10 CFU/mL. The bacterium was also serially diluted, cultured on the Campylobacter-selective plates, and enumerated 48 h later. The Campylobacter-selective plate was prepared in-house and it consisted of Bolton's Campylobacter Enrichment (CE) Broth (Neogen Food Safety, Lasing, MI, USA), 1.5% agar (VWR, USA, OH), 5% lysed horse blood (VWR, Radner Township, PA, USA), five antibiotics (20 mg/L cefoperazone, 50 mg/L cycloheximide, 20 gm/L trimethoprim, 20 mg/L vancomycin, and 0.35 mg/L polymyxin B), 500 mg/L ferrous sulfate, and 200 mg/L triphenyl-tetrazolium chloride (TTC) (all from Sigma-Aldrich, St. Louis, MO, USA). The ferrous sulfate and TTC were used to make C. jejuni colonies dark red. The birds were gavaged with 1 mL PBS or 10 9 CFU/bird C. jejuni AR101 at day 12 [37]. Chicken body weight was measured at days 0 and 28. Because of the pen size constrains, the birds were randomly euthanized at days 21 and 28 to collect cecal samples for enumerating C. jejuni, and the exact bird numbers are listed in figure legends.

Materials and Methods
This experiment was conducted until 28 days of age because of the pen size constrain. Cecal digesta samples of all the groups were collected for DNA isolation. Another set of cecal digesta were serially diluted ten-fold with sterile PBS and cultured on the Campylobacterselective plates at 42 • C for 48 h under a microaerophilic atmosphere. Emerged colonies were positively determined as C. jejuni only when they were dark red and shining, round, and with a smooth surface. The colonies were also examined under a microscope for size and motility evaluation [53]. The CFU per gram digesta was then calculated.

Isolation of Enterobacter102
When chicken cecal digesta were cultured on C. jejuni selective plates, pink colonies were grown on the plate, compared to dark red C. jejuni colonies. The pink colony was named Enterobacter102. Under a light microscope, Enterobacter102 was rod-shaped and larger than C. jejuni. Enterobacter102 was able to grow aerobically, stained Gram-negative, and showed pink colonies on a MacConkey plate.

Identification of Bacterial Species Using 16s rDNA and Sanger Sequencing
Either C. jejuni AR101 or Enterobacter102 was derived from a single colony. To isolate DNA for Sanger sequencing, the bacteria were spread on the respective agar plates. The bacteria were collected, and DNA was extracted. Genomic DNA from C. jejuni AR101 or Enterobacter102 was amplified by PCR of the 16S rDNA gene region with universal primers (27Fw1: 5 -AGAGTTTGATCMTGGCTCAG-3 , 1492R: 5 -CGGTTACCTTGTTACGACTT-using the NCBI genome database. The bacteria were given species names with >98.0% and 95.0% of 16S rDNA sequence homology for Campylobacter jejuni and Enterobacter sp., respectively. The 16s rDNA sequences were uploaded at NCBI with submission numbers of SUB10285129 and SUB10285090.

In Vitro Co-Culturing of C. jejuni with Various Microbiota
The impact of various microbiota on C. jejuni growth was evaluated. Briefly, C. jejuni AR101 from frozen stocks was cultured and grown on the Campylobacter-selective plates in a microaerophilic atmosphere for 48 h using the GasPak system (BD Biosciences, Franklin Lakes, NJ, USA). C. jejuni at 6.3 × 10 8 CFU was co-cultured with noninfected microbiota at 2.0 × 10 8 CFU, SPF-Aerobe at 1.8 × 10 8 CFU, or SPF-Anaerobe at 8.4 × 10 8 CFU in 1 mL of CE broth. C. jejuni at 7.7 × 10 8 CFU was co-cultured with Cj-MB at 6.7 × 10 7 CFU, Cj-SPF-Aerobe at 1.6 × 10 9 CFU, or Cj-SPF-Anaerobe at 1.5 × 10 9 CFU in 1 mL of CE broth. C. jejuni at 1.8 × 10 8 CFU and 4.4 × 10 7 CFU Enterobacter102 were co-cultured in 1 mL of CE broth. The experiments were carried out in triplicate. Because C. jejuni growth would be reduced within 24 h in anaerobic conditions [54], the co-culture bacteria were incubated for 24 h at 42 • C under anaerobic conditions using the GasPak system (BD Biosciences, Franklin Lakes, NJ, USA) to mimic cecal air conditions. C. jejuni growth was measured by serial dilution and plating on the Campylobacter-selective plates for enumeration. When co-culture with C. jejuni, Enterobacter102 was counted with pink colonies compared to dark red colonies of C. jejuni. Cj-MB, Cj-SPF-Aerobe, and Cj-SPF-Anaerobe themselves could not grow a single colony on the Campylobacter-selective plates, suggesting that C. jejuni lost culturability after storing with microbiota.

Statistical Analysis
All values are shown as mean ± standard error of the mean as indicated. Differences between groups were analyzed using the nonparametric Mann-Whitney U test performed using GraphPad Prism 7.0 software. C. jejuni CFU was transformed with a formula of log10 (CFU + 1). The results were considered statistically significant if p-values were <0.05.