The Effect of Clostridium butyricum on Gut Microbiota, Immune Response and Intestinal Barrier Function During the Development of Necrotic Enteritis in Chickens

Necrotic enteritis (NE) causes huge economic losses to the poultry industry. Probiotics are used as potential alternatives to antibiotics to prevent NE. It is known that Clostridium butyricum can act as a probiotic that can prevent infection. However, whether or not it exerts a beneficial effect on NE in chickens remains elusive. Therefore, we investigated the impact of C. butyricum on immune response and intestinal microbiota during the development of NE in chickens, including experimental stages with basal diets, high-fishmeal-supplementation diets, and Clostridium perfringens challenge. Chickens were divided into two groups from day 1 to day 20: one group had its diet supplemented with C. butyricum supplementation and one did not. At day 20, the chickens were divided into four groups: C. perfringens challenged and unchallenged chickens with and without C. butyricum supplementation. All groups were fed a basal diet for 13 days and thereafter a basal diet with 50% fishmeal from day 14 to 24. Chickens were infected with C. perfringens from day 21 to 23. At days 13, 20 and 24, samples were collected for analysis of the relative expression of immune response and intestinal mucosa barrier-related genes and intestinal microbes. The results show that C. butyricum can inhibit the increase in IL-17A gene expression and the reduction in Claudin-1 gene induced-expression caused by C. perfringens challenge. Moreover, C. butyricum was found to increase the expression of anti-inflammatory IL-10 in infected chickens. Although C. butyricum was found to have a significant beneficial effect on the structure of intestinal bacteria in the basal diet groups and decrease the abundance of C. perfringens in the gut, it did not significantly affect the occurrence of intestinal lesions and did not significantly correct the shift in gut bacterial composition post C. perfringens infection. In conclusion, although C. butyricum promotes the expression of anti-inflammatory and tight junction protein genes and inhibits pro-inflammatory genes in C. perfringens-challenged chickens, it is not adequate to improve the structure of intestinal microbiota in NE chickens. Therefore, more effective schemes of C. butyricum supplementation to prevent and treat NE in chickens need to be identified.

Necrotic enteritis (NE) causes huge economic losses to the poultry industry. Probiotics are used as potential alternatives to antibiotics to prevent NE. It is known that Clostridium butyricum can act as a probiotic that can prevent infection. However, whether or not it exerts a beneficial effect on NE in chickens remains elusive. Therefore, we investigated the impact of C. butyricum on immune response and intestinal microbiota during the development of NE in chickens, including experimental stages with basal diets, highfishmeal-supplementation diets, and Clostridium perfringens challenge. Chickens were divided into two groups from day 1 to day 20: one group had its diet supplemented with C. butyricum supplementation and one did not. At day 20, the chickens were divided into four groups: C. perfringens challenged and unchallenged chickens with and without C. butyricum supplementation. All groups were fed a basal diet for 13 days and thereafter a basal diet with 50% fishmeal from day 14 to 24. Chickens were infected with C. perfringens from day 21 to 23. At days 13, 20 and 24, samples were collected for analysis of the relative expression of immune response and intestinal mucosa barrierrelated genes and intestinal microbes. The results show that C. butyricum can inhibit the increase in IL-17A gene expression and the reduction in Claudin-1 gene inducedexpression caused by C. perfringens challenge. Moreover, C. butyricum was found to increase the expression of anti-inflammatory IL-10 in infected chickens. Although C. butyricum was found to have a significant beneficial effect on the structure of intestinal bacteria in the basal diet groups and decrease the abundance of C. perfringens in the gut, it did not significantly affect the occurrence of intestinal lesions and did not significantly correct the shift in gut bacterial composition post C. perfringens infection. In conclusion, although C. butyricum promotes the expression of anti-inflammatory

INTRODUCTION
Necrotic enteritis (NE), which is caused by Clostridium perfringens in broiler chickens, is a common disease in the poultry industry, resulting in huge financial losses every year. It is estimated that NE causes losses of more than $US 6 billion per year and poses a substantial risk to public health (Wade Ben, 2015). However, C. perfringens infection alone does not cause clinical NE (Olkowski et al., 2006;Shojadoost et al., 2012). Predisposing factors, such as a high-fishmeal-supplementation diet or/and Eimeria coinfection, combined with C. perfringens infection, are required to induce NE in chickens (Wu et al., 2010(Wu et al., , 2011Fernandes da Costa et al., 2013). It has been reported that high-fishmeal diets that include high available nutrition cause an increase in C. perfringens-associated lesions in the chicken gut (Drew et al., 2004). We used a high-fishmeal-supplementation diet and C. perfringens challenge to induce NE in chickens. According to the findings of previous studies (Wu et al., 2010;Fernandes da Costa et al., 2013), the induction of NE in this study was divided into three main stages: (i) 1-13-day-old chickens on a basal diet, (ii) 14-20-day-old chickens on a high-fishmealsupplementation diet, (iii) 21-24-day-old chickens on a highfishmeal-supplementation diet and C. perfringens challenge.
Usually, NE is prevented through the use of antibiotics. However, many countries have restricted the use of antibiotics owing to the emergence of resistant strains, disruption of the host gut microbiota, and residual contamination of products and the environment. Probiotics, essential oils and organic acids are among the potential alternatives to antibiotics (Allaart et al., 2013). Clostridium butyricum is a butyric acid-producing grampositive anaerobe that commonly exists in the intestines of humans and animals. In chickens, butyric acid inhibits pathogens and improves intestine barrier function (Meimandipour et al., 2010;Zhang et al., 2011), benefits intestinal microbiota, and stimulates immune factors (Kong et al., 2011;Zhang et al., 2011;Gao et al., 2012b;Yang et al., 2012). C. butyricum has been used as a probiotic that decreases the clinical signs of several diseases such as inflammatory bowel disease (IBD) and antibiotic-associated diarrhea (Seki et al., 2003;Yasueda et al., 2016). Moreover, previous studies have shown that C. butyricum could be a suitable alternative for reducing colonization by pathogens such as Salmonella and Escherichia coli (Zhang et al., 2016;Zhao et al., 2017).
However, the effect of C. butyricum on NE in chickens remains to be elucidated. Therefore, we aimed to explore the effects of C. butyricum on gut microbiota, immune response, and intestinal barrier function-related genes during the development of NE in chickens.

Ethics Statement
This study was carried out in accordance with the principles of the Basel Declaration and the recommendations of the Institutional Animal Care and Use Committee guidelines, Guangdong Academy of Agricultural Sciences. The study protocol was approved by the Guangdong Academy of Agricultural Sciences (No. SYXK (yue)-20180108).

Bacterial Strains
The pathogenic C. perfringens cpnetBF carrying the netB gene was isolated from the intestines of broiler chickens with NE in Guangdong Province, China (Huang et al., 2018). The strain was stored at −80 • C in fluid thioglycollate broth (FT, Beckson, Dickinson, and Company) supplemented with 30% glycerol. C. butyricum YH 018 with a count of 1 × 10 9 cfu/g was obtained from Yihao Biotic Co., Ltd., China. The phylogenetic relationships based on 16S rRNA gene sequences are shown in Supplementary Figure S1. The concatenated sequences were aligned by MEGA 6 to infer a maximum-likelihood tree. C. butyricum YH 018 is clustered with type strains of species of the C. butyricum phylogroup including C. butyricum strain JCM 1391T and ATCC 19398T.

Animal Trial Design and Sample Collection
One-day-old Ross 308 commercial chickens (negative for C. perfringens) were purchased from a hatchery in Guangzhou City, Guangdong Province, China. The chickens were given food and water ad libitum and maintained in a 12 h light/12 h dark cycle. The animal trial was conducted at a poultry house located at the Institute of Animal Health, Guangdong Academy of Agricultural Sciences. A total of 120 chickens were randomly divided into two groups. All groups were fed an antibioticfree basic chicken diet between the ages of 1 and 13 days (Supplementary Table S1) and then switched to a wheatbased diet containing 50% fishmeal from day 14 to 24. One group was supplemented with 1 × 10 9 cfu of C. butyricum/kg between the ages of 1 and 24 days. At an age of 20 days, each of the above-mentioned groups of chickens was further divided into two groups. From the ages of 21 days to 23 days, the C. butyricum supplemented and non-supplemented groups were challenged with C. perfringens cpnetBF-inoculated feed, as previously described (Stanley et al., 2012).
Chickens from each group were euthanized using cervical dislocation at an age of either 13, 20, or 24 days. Twelve replicates each of the jejunum-ileum contents of groups CB_B, CK_B, CB_FM, and CK_FM and six replicates each of the jejunumileum contents of groups CB/CP, CB, CP, and CK groups were collected for DNA extraction. The 12 ileal tissue replicates in each of the CB_B, CK_B, CB_FM and CK_FM groups and the six ileal tissue replicates in each of the CB/CP, CB, CP and CK groups were collected from a distance of 3 cm distal to the Meckel's diverticulum and were immediately frozen in liquid nitrogen. The C. butyricum supplemented group was named CB_B (n = 60) and CB_FM (n = 48) at ages of 13 and 20 days, respectively, while the C. butyricum non-supplemented group was named CK_B (n = 60) and CK_FM (n = 48) at ages of 13 and 20, respectively. At an age of 24 days, the names of the four groups were as follows: CB/CP (n = 18): C. perfringenschallenged chickens with C. butyricum supplementation, CB (n = 18): C. perfringens-unchallenged chickens with C. butyricum supplementation, CP (n = 18): C. perfringens-challenged chickens without C. butyricum supplementation, and CK (n = 18): C. perfringens-unchallenged chickens without C. butyricum supplementation, as shown in Figure 1.

Intestinal Lesion Score
At an age of 24 days, the lesions of the small intestine (duodenum to ileum) were scored on the basis of 10 replicates each from the four groups at the challenge stage, as previously described (Keyburn et al., 2006): 0, no gross lesions; 1, thin or friable walls; 2, focal necrosis or ulceration (one to five foci); 3, focal necrosis or ulceration (six to 15 foci); 4, focal necrosis or ulceration (16 or more foci); 5, patches of necrosis 2-3 cm long; 6, diffused necrosis typical of field cases.
DNA Extraction and the Abundance of C. perfringens DNA from the jejunum-ileum contents was extracted using a QIAamp Fast DNA Stool Mini Kit (Qiagen, Valencia, CA, United States), according to the manufacturer's instructions. Total DNA was quantified on a NanoDrop R ND-2000 UV spectrophotometer (NanoDrop Technologies, Wilmington, DE). Only DNA samples with an A260/A280 value >1.7 and A260/A230 value >1.8 were used for further analysis (Li et al., 2014). The extracts were stored at −20 • C until used (Singh et al., 2015).
Quantitative PCR reaction was performed on the intestinal DNA template, in triplicate, using an SYBR Premix Ex Taq (Takara Bio, Otsushi, Japan). The expression of the C. perfringens genes was normalized using the 16S rRNA gene (Buffie et al., 2015). The relative abundance of C. perfringens in the intestine was calculated based on the value of the 16S rRNA gene using the 2 − Ct method. The primers used for the qPCR of C. perfringens and 16S rRNA gene are shown in Supplementary Table S2. 16s rRNA Gene Sequencing The gene-specific sequences for the 16S V3-V4 region were amplified using primers 338F (5 -ACTCCTACGGGAGGCA GCAG-3 ) and 806R (5 -GGACTACHVGGGTWTCTAAT-3 ) using a thermocycler PCR system (GeneAmp 9700, ABI, United States). The PCR conditions were as follows: 3 min of denaturation at 95 • C, 27 cycles of 30 s at 95 • C, 30s for annealing at 55 • C, 45s for elongation at 72 • C, and a final extension step at 72 • C for 10 min. PCR reactions were performed in triplicate using a 20-µL mixture containing 4 µL of 5 × FastPfu Buffer, 2 µL of 2.5 mM dNTPs, 0.8 µL of each primer (5 µM), 0.4 µL of FastPfu Polymerase, and 10 ng of template DNA. The resulting PCR products were extracted from the 2% agarose gel and further purified using the AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, United States) and quantified using a QuantiFluor TM -ST (Promega, United States) according to the manufacturer's protocol. Bar-coded V3-V4 amplicons were sequenced using the 2 × 300 paired-end method by Illumina MiSeq with a 7-cycle index read. Sequence processing was performed using QIIME 2 to process the data. Operational taxonomic units (OTUs) were clustered using a 97% similarity cutoff in UPARSE (version 7.1 1 ), and chimeric sequences were identified and removed using UCHIME. The taxonomy of each 16S rRNA gene sequence was analyzed using an RDP Classifier algorithm 2 against the Silva (SSU123) 16S rRNA database using a confidence threshold of 70%. Simpson and abundance-based coverage (ACE) indices were calculated for each time point to represent taxonomic alpha diversity. The Venn diagrams were calculated using R software. The microbial community structures of the different samples were compared using Bray-Curtis dissimilarity. Principal coordinate analysis (PCoA) was conducted to assess the relationships among the different groups. The linear discriminate analysis (LDA) effect size (LEfse) using the non-parametric factorial Kruskal-Wallis (KW) sum-rank test was used to identify significant differences in the relative abundance of bacteria between two groups. The functional potentials of the intestinal microbiotas were predicted using the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt).

RNA Isolation and RT-qPCR of the Immune Genes
Total RNA was isolated from ileal tissue using the EZNA R Total RNA Isolation Kit (Omega Bio-Tek) according to the manufacturer's instructions. RNA was eluted in DEPCtreated water, quantified using a NanoDrop ND-2000 UV spectrophotometer (NanoDrop R Technologies, Wilmington, DE, United States), and stored at −80 • C until further use.
Reverse transcription of the isolated RNA was performed using an M-MLV First Strand cDNA Synthesis Kit (Omega Bio-Tek) following the manufacturer's instructions. RT-qPCR was performed with cDNA temple in triplicate using SYBR Premix Ex Taq (Takara Bio, Otsushi, Japan). The expressions of TLR2, IL-10, TNF, IL-1β, NF-κB, IL-17A, Claudin-1, Claudin-2, Occduin1, and IgA genes were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Zhang et al., 2017). The relative expression was calculated based on the expression of GAPDH, using the 2 − Ct method. The primers used for RT-qPCR are shown in Supplementary Table S2.

Statistical Analysis
Statistical analysis of the relative expression of mRNA, the relative abundance of C. perfringens and the alpha diversity indices of Simpson and ACE were performed using ANOVA. A p-value of 1 http://drive5.com/uparse/ 2 http://rdp.cme.msu.edu/classifier/classifier.jsp <0.05 was considered to be statistically significant. For species abundance, comparisons were made using the Kruskal-Wallis test with Benjamini-Hochberg p-value correction.

Clostridium perfringens Colonization
All chickens were negative for C. perfringens until the challenge, and the unchallenged groups remained negative throughout this study. In the challenged groups, the relative abundance of C. perfringens in chickens with C. butyricum supplementation was significantly lower than that of C. butyricum nonsupplemented chickens (p < 0.05, Figure 2).

Intestinal Lesion Score
No intestinal lesions were observed in unchallenged chickens. The infection-induced intestinal lesion score was not influenced by the presence or absence of C. butyricum supplementation (p > 0.05, Figure 3).

Expression of Immune Response Genes
The expressions of the immune and intestinal barrier-related genes detected in the different groups are shown in Figure 4. In the basal diet chickens, only IL-17A expression was significantly different between the C. butyricum supplemented and nonsupplemented groups. The relative expressions of TLR2, IL-10, Claudin-1, and Claudin-2 were significantly higher in the CB_FM group than in the CK_FM group (p < 0.05). Compared with the other three groups, C. perfringens infection induced the highest expression of TNF in the CB/CP group. The expression of IL-10 showed no significant difference between the CP and CK group; however, it was significantly higher in the CB/CP group than FIGURE 2 | Relative abundance of C. perfringens in the jejunum-ileum lumen post 1-day challenge. All chickens tested negative for C. perfringens until experimentally challenged, and the unchallenged groups tested negative for C. perfringens throughout the study. * p < 0.05, measured using ANOVA; n = 6.
Frontiers in Microbiology | www.frontiersin.org in the CP group. The relative expression of IL-17A increased significantly in the CP group compared in with the CK group and significantly decreased in the CB/CP and CB groups compared in with the CP and CK groups. Conversely, the expression of Claudin-1 significantly decreased in the CP group compared with in the CK group and increased significantly in the CB/CP group compared in with the CP group (p < 0.05). The C. perfringens challenge significantly increased the relative expression of IgA, but C. butyricum had no effect on the level of IgA. There were no significant differences between the relative expressions of the IL-1β, NF-κB, and occludin-1 genes of the groups at different stages.

Intestinal Bacterial Diversity Analysis
A total of 856,602, 1,034,832, and 916,190 high-quality 16S rRNA sequences were obtained from 13-, 20-, and 24-day-old intestinal content samples, respectively. The Simpson and ACE indices were used as measures of the alpha diversity (evenness and richness) of the gut microbiota ( Figure 5A). In the basal diet groups, no significant differences were found between the ACE and Simpson index values of the CB_B and CK_B groups. During the stage of high-fishmeal supplementation, C. butyricum significantly increased the Simpson indices in the CB_FM group compared with in the non-supplemented CK_FM group (p < 0.05). Infection by C. perfringens significantly decreased the ACE indices compared with the unchallenged groups, CB and CK (p < 0.05). However, there was no significant difference between the CB/CP and CP groups.
A Venn diagram reveals the shared and specific OTUs of the different groups ( Figure 5B). Dietary C. butyricum decreased the number of unique OTUs in the CB_B (58) and CB_FM (92)  groups compared with the CK_B (129) and CK_FM (155) groups, respectively. At the stage of C. perfringens challenge, the four groups shared 126 OTUs. In addition, compared with the CB/CP group, the CP group had a less unique OTU and shared less OTUs with the CK group.
Principal coordinate analysis (PCoA) of Bray-Curtis indices demonstrates the diversity of bacteria among the different groups ( Figure 5C). Before C. perfringens challenge, the gut bacterial structure of the chickens with C. butyricum supplementation (CB_B and CB_FM) was found to be distinct from that of the groups without C. butyricum (CK_B and CK_FM). After C. perfringens challenge, the samples from the CB/CP and CP groups were found to be distinct from those of the CB and CK groups. However, the microbial structure of the CB/CP group was not significantly different from that of the CP group (p < 0.05).

Structure and Composition of Intestinal Bacteria
At the order level, the predominant microorganisms in the chicken intestine were Lactobacillales and Clostridiales (Figure 6 and Supplementary Figure S2). Following highfishmeal supplementation, Corynebacteriales decreased dramatically in the CB_FM group compared with in the CK_FM group (Supplementary Figure S2). After C. perfringens challenge, Clostridiales significantly increased in the CP group compared with in the CK group, whereas Bacillales decreased. Moreover, these microbiotic changes were significantly alleviated in the CB/CP group. In addition, LEfSe showed that high-fishmeal supplementation specifically increased the overrepresented bacteria, Clostridiales, in the CB_FM group (Figure 6), while similar results were also noted in the CB/CP group. These observations reveal that dietary C. butyricum increases the relative proportion of Clostridiales at different stages of NE.
For further investigation of the composition of bacteria at different stages, the significance of differences in microbial species abundance was analyzed (Supplementary Table S3). C. butyricum increased significantly in the C. butyricum-supplemented groups, especially at the stage of high-fishmeal supplementation. Moreover, the abundance of Lactobacillus_sp._KC45b significantly increased and Candidatus Arthromitus and unclassified_g__Brachybacterium significantly decreased in the CB_B group compared with in the CK_B group. In the C. perfringens-challenged groups, CB/CP and CP, the abundance of Lactobacillus salivarius decreased significantly compared with in the CK group. Notably, the abundances of Weissella thailandensis, unclassified_g__Weissella, and Pediococcus acidilactici in the CB/CP group were higher than in the CK group. However, no significant differences between the CB/CP and CP groups or between the CB and CK groups were found using cutoff values of p < 0.05 and FDR < 0.1.

Predicting the Function of Intestinal Bacteria
The functions of the intestinal microbiota metagenome were predicted using PICRUSt from the Kyoto Encyclopedia of Genes and Genomes pathways (Figure 7). A total of 281 functions were detected in all of the groups. No significant differences were found between the CB_B and CK_B groups and the CB/CP and CP groups. However, there were significant differences (Mann-Whitney U test, p < 0.05, FDR < 0.1) between the CB_FM and CK_FM groups in 96 functions, comprising 34.16% of the total functions, including purine metabolism and pyrimidine metabolism. Moreover, there were 16 and 14 significant differences between the CB/CP and CK groups and the CP and CK groups, respectively, for example, fatty acid biosynthesis, pentose and glucuronate interconversions, and glutathione metabolism.
All raw sequences were submitted to the Sequence Read Archive database at NCBI (accession no. SRP159781).

DISCUSSION
Previous reports have shown that NE is caused by factors including a fishmeal diet, Eimeria, and C. perfringens and is associated with a shift in gut microbiota structures and inflammation (Stanley et al., 2012(Stanley et al., , 2014. C. butyricum is a probiotic that has been proven to alleviate certain infections. Hence, we investigated the effect of C. butyricum on gut bacterial community composition and immune factors during the development of the NE caused by a high-fishmeal diet and C. perfringens infection. The immune system plays a critical role in fighting infections. It has been reported that dietary C. butyricum can promote the expression of Claudin and Occludin to maintain intestinal barrier function, activate the expression of NF-κB, TLR2, and IL-10 and decrease pro-inflammatory cytokines, IL-1β, and TNF expression to modulate inflammatory response (Gao et al., 2012b;Hayashi et al., 2013;Zhang et al., 2016;Chen et al., 2018;Li H. et al., 2018;Liu et al., 2018). However, other studies have reported that C. butyricum has no influence on the immune system of neonatal mice and pigs (Miao et al., 2018;Peeters et al., 2019). In line with this finding, in this study, during the basal diet stage, no significant differences were found between C. butyricum supplemented and nonsupplemented groups in the expression of every intestinal barrier function and inflammatory gene apart from IL-17A. A previous study reported that the composition of the diet is associated with the efficacy of C. butyricum. A C. butyricum supplemented diet improves the immune response and promotes intestinal barrier function with a low digestibility diet, whereas C. butyricum has no effect with a highly digestible diet . Based on these findings, we speculate that intestinal barrier function and inflammatory genes are probably not affected by C. butyricum in the findings of this study due to the type of diet.
Intestinal dysbiosis can elevate the production of proinflammatory cytokines, IL-1β, TNF, and NF-κB, and C. butyricum can decrease the expression of these factors in piglets, suggesting that C. butyricum can alleviate the inflammatory response in piglets Cao et al., 2019;Wang et al., 2019). Zhang et al. reported that C. butyricum promotes the expression of TNF in chickens with E. coli K88 challenge (Zhang et al., 2016). Similar results were obtained in the current study, where C. butyricum was found to increase the expression of TNF in response to C. perfringens infection. A TNF-induced inflammatory response is common in chickens infected with pathogens such as Salmonella (Feng et al., 2016). Similarly, the NetB toxin of C. perfringens increases the TNF mRNA level in the intestinal lymphocytes of chickens (Jang et al., 2012). However, in this study, the expressions of TNF and TRL2 were not affected by C. perfringens challenge, which is FIGURE 6 | Differential enrichment of intestinal microbiota between groups measured using LEfSe (LDA score = 3). Bars represent significant abundance from phylum to genus between (A) the CB_B and CK_B groups, (B) the CB_FM and CK_FM groups, (C) and the CB/CP and CP groups. CB_B, CK_B, CB_FM and CK_FM: n = 12; CB/CP, CB, CP and CK: n = 6.
Frontiers in Microbiology | www.frontiersin.org FIGURE 7 | Heatmap of predicted KEGG pathways of the gut bacteria in each group. The gene copy numbers of each group are shown as mean values. The significant differences between two groups at different stages were measured using the Mann-Whitney U test with a cutoff of p < 0.05 and FDR < 0.1. No significant difference was found between the CB_B and CK_B groups, the CB/CP and CP groups, and the CB and CK groups. Comparisons between the CB/CP and CB groups and the CB and CP groups were not performed. CB_B, CK_B, CB_FM, and CK_FM: n = 12; CB/CP, CB, CP, and CK: n = 6. Comparisons between CB_FM and CK_FM groups, CB/CP and CK groups and CP and CK groups are presented in panel (A), (B), and (C) respectively. consistent with the results of previous studies (Lu et al., 2009;Du et al., 2016). The high expression of IL-17A is associated with the initiation of several inflammatory diseases (Kolls and Linden, 2004). As previously mentioned, the results of this study demonstrate that C. perfringens challenge can upregulate the expression of IL-17 (Fasina and Lillehoj, 2019). Previous research has shown that a decrease in IL-17A expression can ameliorate inflammation (Kanai et al., 2015). Various mechanisms have been proposed to explain the anti-inflammatory effects of probiotics on intestinal inflammation, one of which is suppression of the expression of IL-17 (Tanabe, 2013). At the same time, it has been proven that the intake of Bacillus subtilis by mice with inflammatory bowel disease can downregulate the expression of Il-17 in order to decrease intestinal damage (Gong et al., 2016). We observed that C. butyricum could inhibit the upregulation of IL-17A expression induced by C. perfringens. Moreover, C. butyricum increases the expression of anti-inflammatory IL-10 mRNA, which plays a vital role in preventing many inflammatory diseases (Gao et al., 2012a). For instance, Bifidobacterium-induced IL-10 production benefits patients with colitis (Imaoka et al., 2008). In the same manner, we found that C. butyricum increases the expression of IL-10 mRNA levels in C. perfringens-challenged chickens, a situation that may indicate the prevention of NE.
IgA is the predominant intestinal immunoglobulin that contributes to intestinal barrier function (Pabst, 2012). Previous studies have reported that dietary C. butyricum supplementation can promote the expression of IgA in rabbits, broiler chickens, and mice (Murayama et al., 1995;Yang et al., 2012;Liu et al., 2018). However, we observed that dietary C. butyricum has no significant effect on the expression of IgA in the intestine. This is similar to the results of other studies, which demonstrate that C. butyricum supplementation does not affect the levels of IgA in neonatal mice, broilers, and laying hens (Liao et al., 2015;Miao et al., 2018;Zhan et al., 2019). Perdigon et al. (1995) show that Lactobacillus casei and Lactobacillus acidophilus improve the level of IgA in a dose-dependent manner. Therefore, we assume that the different effects of C. butyricum on the level of IgA may be related to the dose of C. butyricum.
Occludin and Claudin are tight junction proteins that play critical roles in regulating intestinal epithelial barrier function and preventing macromolecular transmission (Ballard et al., 1995;Fanning et al., 1998). We found that C. perfringens and a high-fishmeal diet decrease levels of Claudin-1 without influencing the expression of Occludin and Claudin-2, which is contrary to the report by Liu et al. (2012). However, C. butyricum supplementation was found to elevate the expression of Claudin-1 in C. perfringens-infected chickens. This finding suggests that C. butyricum may ameliorate damage to the intestinal barrier induced by C. perfringens.
In addition, we found that dietary C. butyricum could decrease the abundance of C. perfringens in the intestine of challenged chickens, consistent with previous reports that probiotics can eradicate pathogens (Gaggia et al., 2010;Yang et al., 2012). However, C. butyricum did not have any effect on intestinal lesion.
With regard to the alpha diversity of the gut microbiota, we observed that during the basal diet stage, C. butyricum had no significant effect on the ACE and Simpson indices, which is consistent with previous reports that the alpha diversity of intestinal bacteria is not affected by C. butyricum (Zhang et al., 2018). However, in the high-fishmeal-supplementation groups, C. butyricum supplementation was found to increase the Simpson indices, indicating that C. butyricum could reduce the diversity of intestinal bacteria. We speculate that high-fishmeal diets provide an opportunity for the proliferation of C. butyricum, which then competes with other bacteria for nutrients and niches and hence reduces the microbial diversity of the chicken intestine under a high-fishmeal diet. In addition, C. perfringens causes a decrease in the ACE index, which cannot be improved with C. butyricum supplementation.
The results of the PCoA indicate that in the uninfected groups, C. butyricum supplementation significantly alters the structure of intestinal microbiota, independent of a high-fishmeal diet, which is consistent with the results of a previous study (Kong et al., 2011). Nevertheless, no significant difference was found between the PCoA results of the C. butyricum supplemented and non-supplemented groups after infection. This indicates that C. butyricum is unable to alter the disruption of the gut bacterial structure caused by C. perfringens infection.
In the basal diet groups, dietary C. butyricum could decrease the abundance of harmful bacteria, such as Candidatus Arthromitus and unclassified_g__Brachybacterium, and increase the abundance of beneficial bacteria, such as Lactobacillus_sp._KC45b. Candidatus Arthromitus, one group of segmented filamentous bacteria (SFB), has been found in the intestinal bacteria of vertebrates such as chickens, rodents, and fish. It has been reported that the presence of SFB is correlated with diarrhea in poultry (Goodwin and Krabill, 1989;Angel et al., 1990;Morishita et al., 1992). Brachybacterium spp. were first isolated from poultry deep litter (Collins et al., 1988) and have previously been considered non-harmful bacteria. However, a recent study shows that Brachybacterium spp. are associated with blood infections in humans (Tamai et al., 2018). Lactobacillus spp. exerts a beneficial effect of resistance to pathogens such as Campylobacter (Stern et al., 2001) and Clostridium populations (Decroos et al., 2004). This result is consistent with previous studies that indicate that C. butyricum benefits the composition of intestinal bacteria (Yang et al., 2012). However, similar results were not found in the high-fishmeal-supplemented groups, indicating that although a high-fishmeal diet could facilitate the abundance of C. butyricum compared with that of basal diet groups, the structure and composition of intestinal bacteria in high-fishmeal-supplemented chickens is not better than that of basal diet chickens.
C. perfringens can disrupt the balance of chicken intestinal bacteria. For example, we observed that C. perfringens challenge decreases the abundance of the beneficial bacterium Lactobacillus salivarius, and this finding is consistent with those of previous studies Huang et al., 2018). The results of the PCoA and species composition show that there are no significant differences between the intestinal microbiota structures in the CB/CP and CP groups, indicating that the beneficial effect of C. butyricum is less than the harmful effect of C. perfringens on the structure of the intestinal flora. However, compared with the CK group, the abundance of beneficial bacteria Weissella thailandensis, unclassified_g__Weissella, and Pediococcus acidilactici increased in the CB/CP group. Weissella thailandensis, a class of lactic acid bacteria (LAB), are most commonly used as probiotics and play key roles in disease resistance (Rolfe, 2000). Previous studies have shown that Pediococcus acidilactici can act as a potential probiotic as it produces lactic acid and bacteriocins against other enteric pathogens (Daeschel and Klaenhammer, 1985;Klaenhammer, 1993). These results suggest that C. butyricum has a certain beneficial effect on the intestinal microbiota of NE chickens.
It has been reported that the probiotics Lactobacillus acidophilus and Bacillus coagulans Wu et al., 2018) significantly decrease the lesion score and have a significant beneficial influence on the intestinal bacteria of NE chickens. It has also been demonstrated that C. butyricum exerts preventive and therapeutic effects on bacterial infections such as enterohemorrhagic Escherichia coli, Clostridium difficile, and Salmonella enteritidis (Takahashi et al., 2004;Zhao et al., 2017;Oka et al., 2018). However, another study reported that C. butyricum could not prevent Salmonella typhimurium infection (Peeters et al., 2019). In line with this finding, we observed that C. butyricum has no significant effect on the lesion score and the structure of intestinal microbiota after C. perfringens challenge. The reason we speculate is that C. butyricum could not improve the gut microbiota disturbance induced by NE. In other words, the beneficial effect of C. butyricum on the structure of intestinal bacteria is not as strong as the harmful effect of NE. In addition, the dose of C. butyricum also has an effect. Different doses of C. butyricum have different influences on the expression of intestinal barrier function and inflammatory genes. Highdose C. butyricum can promote the expression of Claudin and Occludin, whereas low-dose C. butyricum decreases the mRNA levels of these genes . Symbiotic products and the time-point at which administration takes place also affect its efficacy against NE in chicken. In general, mixed products, such as probiotics and prebiotics, have a greater influence than individual components alone. To the best of our knowledge, probiotics, such as Bifidobacterium, Bacillus subtilis, Bacillus licheniformis, Lactobacillus salivarius, and Lactobacillus plantarum (Murry et al., 2004;Teo and Tan, 2005;Schoster et al., 2013;Lin et al., 2017), and prebiotics, such as chicory fructans, can inhibit C. perfringens growth (Kleessen et al., 2003). In addition, the time-point at which it is administered is also important for probiotic effectiveness. Nakphaichit et al. (2011) reported that dietary Lactobacillus reuteri supplementation during the first week of life has a beneficial effect on the composition of intestinal bacteria and that the effects last for up to 6 weeks. Therefore, it is difficult to predict the specific reason for the failure of the CB treatment based only on this single experimental trial. Therefore, further studies will be conducted to find a more reasonable scheme of administration of probiotics to prevent and treat NE in chickens.
In conclusion, C. butyricum can inhibit the increase of IL-17A gene expression and the decrease of Claudin-1 gene expression induced by C. perfringens challenge. Moreover, dietary C. butyricum promotes the expression of IL-10 in infected chickens. This finding suggests that C. butyricum has a beneficial effect on the immune response and intestinal barrier function of NE chickens. C. butyricum can significantly shift the structure and composition of intestinal bacteria during the stages of basal diet and high-fishmeal-supplementation diet. C. butyricum has a particularly large beneficial effect on the structure of intestinal bacteria during the basal diet stage. Meanwhile, C. butyricum decreases the abundance of C. perfringens in the gut. However, C. butyricum cannot significantly alter the structure of intestinal microbiota in C. perfringens-challenged chickens and does not have a significant effect on the occurrence of intestinal lesions induced by NE. Therefore, schemes that are more effective than C. butyricum supplementation need to be identified to prevent and treat NE in chickens.

DATA AVAILABILITY STATEMENT
The datasets generated for this study can be found in Sequence Read Archive database, SRP159781.

SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fmicb. 2019.02309/full#supplementary-material FIGURE S1 | Maximum-likelihood tree based on 16S rRNA gene sequences showing the phylogenetic relationships between strain C. butyricum YH 018 and closely related species of the genus Clostridium. The tree was constructed using the program MEGA 6 with bootstrap values based on 1000 replications. Filled circles mark nodes that were also present in respective maximum-parsimony and neighbor-joining trees. Scale bar, 0.01 nucleotide substitutions per position.