Avian leukosis virus subgroup J infection inuences the gut microbiota composition in chicken

Background: Avian leukosis virus (ALV) is a major cause of disease in poultry. Probiotics play a critical role in maintaining animal health. Studies have indicated that viral infection can alter the composition of the chicken gut microbiota. We hypothesized that ALV-J infection alters the probiotics composition in the chicken fecal bacterial microbiome. We performed high-throughput 16S rRNA gene sequencing and evaluated the gut microbiota pro ﬁ les using feces from ALV-J-infected and healthy chickens. Results: The relative abundance at the phylum and species levels was calculated. The phylum Proteobacteria was more abundant in ALV-J-infected chickens than in healthy chickens. Additionally, the abundance of the opportunistic pathogen Propionibacterium acnes was signicantly increased in ALV-J-infected chickens. Interestingly, ALV-J infection tended to be signicantly decreased by the probiotics Lactobacillus helveticus and Lactobacillus reuteri Conclusions: The study indicates that ALV-J infection signicantly altered the gut microbiota distribution in chickens. Additionally, ALV-J infection signicantly inuenced the abundance of L. helveticus and L. reuteri in the chicken gut.


Results
Phyla composition exhibited signi cant microbial differences between ALV-J-infected chickens and controls Each sample was rarefied to 17,595 sequences; using a threshold of 97% sequence identity, 16,740 unique operational taxonomic units (OTUs) were identified in the samples. Total sequences were assigned to 38 phyla (3 archaeal phyla and 35 bacterial phyla). Bacterial phyla isolated from the samples included Firmicutes, Proteobacteria, Bacteroidetes, and Actinobacteria. The distribution of the four phyla in group A (viral control) showed that the gut microbiota was dominated by Firmicutes (average relative abundance: 98.03%), whereas other phyla were present in smaller quantities including Proteobacteria (average relative abundance: 1.03%), Bacteroidetes (average relative abundance: 0.27%), and Actinobacteria (average relative abundance: 0.16%). In group B (ALV-J-infected chickens), the gut microbiota was dominated by both Firmicutes (average relative abundance: 52.51%) and Proteobacteria (average relative abundance: 38.69%), with other phyla found in smaller quantities including Bacteroidetes (average relative abundance: 1.46%), Actinobacteria (average relative abundance: 0.68%) and a few other unknown phyla (Fig 1, S1).
Interestingly, two phyla, Firmicutes and Proteobacteria, were proportionally significantly different between groups A and B (P < 0.05). Among the two phyla, the Proteobacteria concentration was much higher in group B than in group A (Fig 1). These results indicate that ALV-J infection significantly affected the proportion of Firmicutes and Proteobacteria at the phylum level.
Bacterial taxonomic clades showed significant differences between ALV-J-infected chickens and controls Principal coordinate analysis was conducted based on weighted UniFrac distances to assess the microbial distribution between the two groups. The results of weighted UniFrac analysis revealed a notable distribution difference for PC2; However, no difference in the distribution of PC1 was observed. The gut microbial community of group A was substantially separated from that of group B ( Fig. 2A). In group A (control), all samples clustered together, whereas in group B (ALV-J-infected chickens), all samples except for B2 clustered together. This indicates that ALV-J infection signi cantly altered the gut microbiota distribution in chickens. Although the microbiome of B2 was rather different from that of the other samples in group B, a higher abundance of Proteobacteria was observed in all samples in group B than in group A. Notably, B2 was characterized by a higher abundance of the phylum Proteobacteria (60.89% of relative abundance) and lower abundance of the phylum Firmicutes (13.48% of relative abundance) than those in the other chickens in group B, resulting in an obvious separation of B2 from the other chickens in group B ( Fig. 2A).
To investigate which OTUs can serve as biomarkers in an unbiased manner, we used the linear discriminant analysis effect size (LEfSe) classi cation tool. The analysis revealed that 15 bacterial taxonomic clades significantly differed among the two groups (P < 0.05). In group B, the key phylotypes were Proteobacteria, Helicobacter, Helicobacteraceae, Comamonas, Betaproteobacteria, Burkholderiales, Gammaproteobacteria, Comamonadaceae, Actinomyc and Lactobacillaceae were present in group A (Fig. 2B). The heat map displayed a similar pattern, as shown in Fig. 2C. These results suggest that the composition of the chicken gut microbiota was signi cantly altered by ALV-J infection.

Difference in composition of probiotics in chicken gut microbiota in ALV-J-infected chickens and controls
We further identi ed the dominant species in the gut microbiota between the two groups. The results revealed signi cant differences (P < 0.05) between eight species including Propionibacterium acnes, Lactobacillus coleohominis, Lactobacillus helveticus, Lactobacillus reuteri, and rarely identi ed species such as Mycoplana spp., Comamonas spp., Delftia spp., and Helicobacter spp. (Figure 3). In group B (ALV-J-infected chickens), three species exhibited a signi cant reduction in abundance including L. coleohominis, L. helveticus, and L. reuteri compared to in group A (control). Two of these species, L. helveticus and L. reuteri, are probiotics. These results suggest that at the species level, ALV-J infection signi cantly altered the composition of probiotics among the chicken gut microbes.

Discussion
Effect of ALV-J infection on composition of chicken gut microbiota An increasing number of studies has indicated that viral infection can alter the composition of the chicken gut microbiota [10][11][12][13]; these results are consistent with our research. Our results also agreed with a recent study of microbial diversity in chickens showing that Firmicutes, Actinobacteria, Proteobacteria, and Bacteroidetes were the top four phyla in the intestinal tract of chickens [18].
The de ned taxa are potential biomarkers for ALV-J-infected and healthy chickens. For example, Proteobacteria can serve as biomarkers for ALV-J infection in chickens at the phylum, order, class, family and genus levels (S1,S2, S3, S4 and S5), whereas a few taxa were markers for healthy chickens, most prominently members of the family Lactobacillaceae. Further, at the species level, 8 species exhibited significant differences between ALV-J-infected and healthy control chickens. The abundance of L. coleohominis, L. helveticus, and L. reuteri was signi cantly decreased in ALV-J-infected chickens, whereas signi cant increases were observed in the abundance of P. acnes and four unidenti ed species, Mycoplana spp., Comamonas spp., Delftia spp., and Helicobacter spp. Our results clearly illustrate that this viral infection can signi cantly alter the composition of the host gut microbiota, which is consistent with previous ndings [19][20][21]. However, further studies are required to evaluate whether changes in the microbiome play a role in disease complications.
The abundance of P. acnes was signi cantly increased in the chicken gut microbiota after ALV-J infection. Propionibacterium acnes is an opportunistic pathogen that may play a role in other conditions, including in ammation of the prostate leading to cancer [22,23]. These results suggest that ALV-J infection can result in increased expression of opportunistic pathogenic bacteria, which is consistent with a previous study [13].

Effect of ALV-J infection on composition of probiotics in the chicken gut microbiota
Our results indicate that ALV-J infection signi cantly in uenced the composition of probiotics in the chicken gut microbiota. This also is the rst study to investigate the potential effects of ALV-J infection on the composition of probiotics in the fecal bacterial microbiome of chickens. The mechanisms by which probiotics affect infection, disease, and immunity are an active area of study. Different strains of lactobacilli can decrease in ammation in the gastrointestinal tract. For example, L. acidophilus interacts with dendritic cells to induce the production of interleukin-10 [24]. Further, L. paracasei can degrade highly in ammatory interferon γ-induced protein 10 [25]. Probiotics are being developed as a nonpharmacological means for preventing or ameliorating gastroenteritis caused by enteropathogens. Lactobacillus rhamnosus GG-supplemented pigs showed a signi cant reduction in diarrhea following rotavirus challenge [26,27]. Moreover, L. reuteri and L. acidophilus with human rotavirus infection produced an additive effect on TLR2-and TLR9-expressing antigens in Gn pigs [28]. Lactobacillus reuteri strains produced an array of antimicrobial compounds that inhibited pathogens in vitro [29]. Increasing evidence has shown that strains belonging to L. helveticus species have health-promoting properties [30]. Interestingly, our study indicated that ALV-J infection inhibits the growth of bene cial bacteria such as Lactobacillus.
The abundance of the probiotics L. helveticus and L. reuteri was signi cantly decreased in the chicken gut microbiota following ALV-J infection; this may be useful for assisting with the diagnosis of this illness post-mortem. Moreover, further studies are required to understand the mechanism by which ALV-J infection signi cantly decreased the abundance of these two probiotics. Our study indicates that to relieve avian leucosis, ALV-J multiplication must be prevented and microbiota-targeted therapies such as probiotic supplements are required.

Conclusion
ALV-J infection signi cantly altered the gut microbiota distribution in chickens. The abundance of two probiotics, L. helveticus and L. reuteri, was signi cantly decreased in the chicken gut microbiota following ALV-J infection.

Methods
Animal and fecal sample collection Female Huiyang bearded chickens at approximately 25 weeks of age were used in this study. The chickens were obtained from a local broiler in Huizhou City, China. They were randomly collected from the national Huiyang bearded chicken breeding ground at Guangdong Jinzhong Agriculture and Animal Husbandry Technology Co., Ltd. (Huizhou, China). The birds were housed in a commercial caging system (each cage was 40 × 40 × 30 cm in height, width, and depth, respectively). Chickens were randomly assigned to the cages, with three chickens in each unit. Water was supplied via two "on-demand" nipples per cage. The company used eradication programs to minimize the transmission of ALV-J from hens to their progenies. To evaluate the eradication programs, based on a previously described method [31], a Taq Man-based real-time PCR method was performed to detect and quantify ALV-J with proviral DNA from 400 swab samples. Using a real-time PCR method to detect ALV-J, only 12 chickens were found to be ALV-J-infected (B group), and 12 uninfected chickens were randomly selected as controls (A group). All 24 experimental chickens were sacri ced by cervical dislocation. Their gut contents were instantly collected from the ceca within 5 min of sacri ce, immediately placed in cryogenic vials, frozen in liquid nitrogen, and transported to the laboratory, where they were stored at −80°C until DNA extraction.
DNA extraction, PCR, and 16S rRNA gene sequencing A genomic DNA extraction kit, the TIANamp Stool DNA Kit, was utilized to extract DNA from the gut contents (TIANGEN, Beijing, China). Twelve DNA samples from each group were randomly divided into four pools to produce three DNA samples per pool. The DNA concentration and purity were determined using a Nanodrop 2000 Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). Extracted DNA was diluted to 10 ng/µL for PCR amplification. The universal primers 515F and 909R, which have been described previously, were used to amplify the V4-V5 hypervariable region of the microbial 16S rRNA gene [32]. The PCR amplifications and puri cation procedures were performed as described in our previous study [33]. All amplicons were sequenced on an Illumina MiSeq system (San Diego, CA, USA) at Guangdong Meilikang Bio-Science, Ltd. (Shaoqing, China).

Bioinformatics analysis
The raw reads were merged using FLASH-software [34]. QIIME Pipeline-Version 1.9.0 was used to process the merged sequence data. The UCHIME algorithm was used to lter the clean data [35]. Effective sequences were grouped into OTUs at a user-de ned level of sequence similarity (such as e.g., 97% to approximate species-level phylotypes). The alpha diversity indices and weighted UniFrac distance metrics were determined using the QIIME pipeline.
Taxonomy was assigned by the Ribosomal Database Project classifier using Greengenes 13_8 [36] (http://qiime.org/home_static/dataFiles.html) as a reference database. Statistical comparisons of microbial communities between treatments were determined using the LEfSe.

Data analysis
Principal coordinate analysis was conducted based on weighted UniFrac distances using QIIME Pipeline-Version 1.9.0. LEfSe analysis was conducted using the Galaxy platform [37]. Wilcoxon rank-sum test was used to test the signi cance of differences between groups using R 3.5.1 software. P < 0.05 was considered as statistically signi cant.

Consent for publication
Not applicable.

Availability of data and materials
The 16S rRNA gene sequencing raw data of the dataset were deposited at the BIG Sequence Read Archive with BioSample accessions CRA002114.

Competing interest
The authors declare that they have no competing interests.

Funding
This study was supported by the Educational Commission of Guangdong Province, China (grant 2018KTSCX217), Science and Technology Planning Project of Huizhou City, China (2019X0701012), and International training project for outstanding young scienti c research talents in Guangdong Universities. All funders supported the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
Authors' contributions HL contributed to the study design and prepared the manuscript. HL and YC participated in all experiments and contributed to data interpretation. HL and JN contributed to the bioinformatics analysis. All authors have read and approved the manuscript   Dominant species of gut microbial microbiota found in groups A and B Dominant species found in the gut microbiota between the two groups. Eight species showed signi cant differences (P < 0.05) between the two groups. Colors represent different groups; group A is blue and group B is red. *indicates signi cant differences between the two indicated groups (P < 0.05).

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.