16S rRNA gene sequencing reveals an altered composition of the gut microbiota in chickens infected with a nephropathogenic infectious bronchitis virus

Infectious bronchitis virus (IBV), a member of the Coronaviridae family, causes serious losses to the poultry industry. Intestinal microbiota play an important role in chicken health and contribute to the defence against colonization by invading pathogens. The aim of this study was to investigate the link between the intestinal microbiome and nephropathogenic IBV (NIBV) infection. Initially, chickens were randomly distributed into 2 groups: the normal group (INC) and the infected group (IIBV). The ilea were collected for morphological assessment, and the ileal contents were collected for 16S rRNA gene sequencing analysis. The results of the IIBV group analyses showed a significant decrease in the ratio of villus height to crypt depth (P < 0.05), while the goblet cells increased compared to those in the INC group. Furthermore, the microbial diversity in the ilea decreased and overrepresentation of Enterobacteriaceae and underrepresentation of Chloroplast and Clostridia was found in the NIBV-infected chickens. In conclusion, these results showed that the significant separation of the two groups and the characterization of the gut microbiome profiles of the chickens with NIBV infection may provide valuable information and promising biomarkers for the diagnosis of this disease.


Results
Viral load and histopathology. All the NIBV-infected chickens in the IIBV group were listless, huddled together, had watery faeces, and displayed ruffled feathers (score = 2, according to the methods described by Avellaneda et al. 24 ) from 3 to 9 days post-inoculation (dpi), and the morbidity was higher than 90%. These clinical symptoms were not observed in the INC group. The quantification of the viral load by RT-qPCR showed values ranging from 2.96 × 10 3 to 3.84 × 10 8 genomic copies per 0.2 μg of RNA, with the lowest viral load in the jejuna and the highest in the kidneys (Fig. 1B). It was noted that the viral load in the ilea was much higher than that in the jejuna, reaching 1.39 × 10 5 genomic copies per 0.2 μg of RNA. No viral RNA was detected in the INC group. There were more goblet cells, which appeared as vacuoles when stained by H&E, in the IIBV group than in the INC group (Fig. 1C, red arrow). In addition, there were many shed villus epithelial tissue that appeared in the ilea of the IIBV group (Fig. 1C, black arrow). Among the ilea, there was no significant difference in villus height (Student's t test, P = 0.094, Fig. 1D), but crypt depth was significantly increased (Student's t test, P = 0.0003, Fig. 1E), and the V/C ratio was significantly decreased (Student's t test, P = 0.016, Fig. 1F).
Description of the sequencing data. After two separate runs on an Illumina HiSeq. 2500 platform and quality-filtering as described in the methods, 698,855 total sequences were identified for further analysis. The rarefaction curves (Fig. 2a) showed that the sequencing depth was near saturation and that the sequencing data included most of the 16S rRNA gene information in the samples.
A decrease in the microbial diversity in the ilea of chickens with NIBV infection. The sample community richness was evaluated based on the operational taxonomic units (OTU) 25 counts in each single sample as shown in Table 1. As shown by the OTU counts, the number of OTUs in the ilea of the IIBV group chickens was less than that of the INC group chickens (Fig. 2b). According to the lower ACE and Chao1 indices, NIBV infection reduced the community richness compared to that in the INC group (Table 1). In addition, neither the Shannon nor the Simpson indices were significantly different between the IIBV group and the INC group within the experiment (Wilcox, P = 0.5887 and P = 0.9372, respectively). The higher Shannon and Simpson index indicated a higher bacterial diversity, which meant that the NIBV infection decreased the microbial diversity compared to that in the INC group (Fig. 2c,d). In conclusion, these data indicated a smaller bacterial diversity in the ileal microbiota of NIBV infected chickens.

NIBV infection altered the ileal bacterial microbiome composition in chickens.
To compare the global differences in the bacterial community composition between the NIBV infected chickens and the INC group chickens, the Bray-Curtis similarity and unweighted UniFrac 26 were calculated (Fig. 3a). We also performed nonmetric multidimensional scaling (NMDS) 27 of all the samples using count-based Bray-Curtis similarity at the OTU level to explore the differences in the bacterial community composition among the two main groups of chickens 28 . As shown in Fig. 3b, the results provided good discrimination between the groups, suggesting that the bacterial community composition of the IIBV group was different from that of the INC group. Simper (similarity percentage) is a decomposition of the Bray-Curtis difference index, which quantifies the contribution of each species to the difference between the two groups. The results showed the top 10 species in the genus level and their contributions to the differences between the two groups (Fig. 3c).
We utilized principal coordinate analysis (PCoA) to visualize visualizing the similarity of the microbial community structures and the phylogenetic distances between the samples by using the phylogenetic-tree-based weighted UniFrac metric. Figure 3d is a UniFrac PCoA-based comparison of the microbial community from the ilea, showing that the NIBV-infected chickens and the normal chickens had a clear separation. In addition, analysis of similarity (ANOSIM) based on the Bray-Curtis distance (R-value = 0.413, P = 0.002, Fig. 3e) and analysis of molecular variance 29 based on the weighted UniFrac distance (P-value = 0.027, AMOVA) showed a significant separation of the two groups. These results suggested distinct differences in the bacterial composition between the INC group and IIBV group.
An inspection of the predicted taxonomic profiles at the phylum level for all the samples indicated that Firmicutes (82.3%) was the major phylum of the ileal community, exceeding Proteobacteria (9.6%) and Bacteroidetes (4.3%). Notably, at the phylum level, the abundance of Firmicutes decreased (from 89.21% to 75.44%) in the IIBV group compared to that in the INC group, whereas there was an increase in the abundances of Proteobacteria (from 2.36% www.nature.com/scientificreports www.nature.com/scientificreports/ to 16.89%) and Bacteroidetes (from 1.91% to 6.69%) (Fig. 4a). Evidently, Lactobacillaceae, belonging to the phylum Firmicutes, was the most abundant family in the ileum. The family Clostridiaceae 1 had significantly lower abundance in the IIBV group than in the INC group (P < 0.05) (Fig. 4b). Lactobacillus, belonging to the family Lactobacillaceae, was the most abundant genus in the ilea. The genus Candidatus Arthromitus, belonging to the phylum Firmicutes, and the genus unidentified Chloroplast, belonging to the phylum Cyanobacteria, were significantly less abundant in the IIBV group than in the INC group (P < 0.05) (Fig. 4c).
The linear discriminant analysis effect size (LEfSe) analysis was performed in this study to identify distinctive features between the two groups 30 . The differentially abundant phyla detected showed that the Cyanobacteria phylum was predominant in the INC group, while the most abundant phylum in the IIBV group was Proteobacteria (Fig. 5a). For the IBV-infected chickens, there was an overrepresentation of Enterobacteriaceae and underrepresentation of Chloroplast and Clostridia (Fig. 5b).   www.nature.com/scientificreports www.nature.com/scientificreports/ metagenomes from the community profiles of the normalized 16S rRNA genes (Fig. 6). The results showed that there were sixteen pathways from level 3 of the Kyoto Encyclopaedia of Genes and Genomes (KEGG, http://www. genome.jp/kegg/) enriched in the INC group (P < 0.01, White's non-parametric t-test): atrazine degradation; www.nature.com/scientificreports www.nature.com/scientificreports/ stilbenoid, diarylheptanoid and gingerol biosynthesis; meiosis-yeast; flavonoid biosynthesis; amoebiasis; amino acid metabolism; ether lipid metabolism; plant-pathogen interaction; lipid biosynthesis proteins; tuberculosis; steroid biosynthesis; sporulation; fatty acid biosynthesis; ABC transporters; chlorocyclohexane and chlorobenzene degradation; and N-glycan biosynthesis. Six pathways were enriched in the IIBV group: other ion-coupled transporters; secondary bile acid biosynthesis; primary bile acid biosynthesis; nucleotide metabolism; nicotinate and nicotinamide metabolism; and toluene degradation (Fig. 5). Altogether, these data suggest that NIBV infection alters the metabolic functions of ileal microbiota and therefore deserves further investigation.

Discussion
The gut microbiome contains very diverse bacteria and influences myriad host functions 28 . Under pathological conditions, the microbial community composition changes, including an increase and decrease in bacterial populations, causing changes in a wide rage of metabolic functions 32,33 . Accordingly, it is certainly necessary to understand the bacterial community composition and changes in gut microbiota with IBV infection. However, the composition of the gut microbiota and how the coronavirus can affect the gut microbiota are still unclear. Therefore, the effects of IBV infection on chicken ileum damage and the ileum microbiome were investigated in our study by using 16S rRNA gene sequencing. 16S rRNA gene sequencing has been applied extensively for assessing the phylogenetic distribution of metagenomes. It is known that the bacterial 16S rRNA gene consists of nine hypervariable regions, and sequences generated by applying different combinations of these regions generally present differing profiles of microbial diversity. Therefore, the optimal choice of the hypervariable region(s) and primer combination vary between various ecological communities. Previous research proved that the highest bacterial diversity and species richness in chicken gut microbiomes were obtained using the primer set corresponding to the V3-V4 region 34,35 . Based on these investigations, in our study, we selected primers targeting the V3-V4 regions of the 16S rRNA gene. www.nature.com/scientificreports www.nature.com/scientificreports/ The complete intestinal structure is essential for digestive and absorptive functions which are intimately associated with the morphological transformations in villus length and crypt depth 36,37 . In the current study, NIBV infection increased the crypt depth and lowered the ratio of villus height to crypt depth in the ileum compared to those in the INC group. These findings suggest that NIBV infection can destroy intestinal mucosal integrity and may slow the growth of intestinal villus epithelial cells. Furthermore, it was reported that intestinal goblet cells play a crucial defensive role in the intestine by synthesizing and secreting several mediators, including mucins and trefoil factor family peptides, which serve defensive and healing functions in the gut [38][39][40] . In this study, NIBV infection obviously increased the number of goblet cells compared to that in the normal group, which indicated hyperplasia of the goblet cells in the ileum during NIBV infection and may be associated with microbial challenges.
Higher diversity and integrity of gut microbiota is favourable to the intestinal ecosystem 41,42 . According to recent research, viral infection can lower microbial diversity in the gut microbiota according to recent research [43][44][45] . Consistent with previous findings, in this study, our results demonstrated that NIBV infection decreases the diversity and richness of gut microbiota (Fig. 2c and Table 1). The analysis of the taxonomic composition displayed a reduction in the Firmicutes and Cyanobacteria phyla and an enrichment of the phylum Proteobacteria and Bacteroidetes in the ilea. Importantly, the relative abundance of Enterobacteriaceae, particularly E. coli, was increased in the ilea www.nature.com/scientificreports www.nature.com/scientificreports/ of NIBV-infected chickens. Moreover, the LEfSe results showed that Enterobacteriaceae belonging to the phylum Proteobacteria were over-represented in the IIBV group. Previous research shows that in the case of intestinal inflammation, epithelial cells reduce their capacity to undergo beta-oxidation, with the consequence of an increased availability of oxygen, which is thought to promote the bloom of Proteobacteria and the dysbiosis of the gut microbiota 46,47 . Further investigations acknowledged that nitrate produced by the host during inflammatory conditions can be exploited by commensal Enterobacteriaceae, which thus become predominant 48 . Therefore, the increased enrichment of Proteobacteria revealed that the intestinal inflammation in the ilea of NIBV-infected chickens represent a "microbial signature" of NIBV infection. The results also showed that NIBV infection decreased the abundance of the members of the Cyanobacteria in the ilea compared to that in the INC group chickens. The Cyanobacteria group has great biodiversity due to its secondary metabolites, which show a broad variety of biological activities, including antibacterial, antiviral, anticancer and immunomodulator or protease inhibitor activities 28,49,50 . Our results showed a reduction in Cyanobacteria, which indicated that the ability of the chicken ileum to clear viruses was weakened by NIBV infection. Clearly, in the case of viral infection, changes in general metabolic function are closely related to the gain and loss of microbial populations.
Multiple studies suggest that gut microbiota influences the host metabolism 51-53 . Cox et al. showed that disruption of microbiota (reduction in Allobaculum, Lactobacillus, Rikenellaceae and Candidatus Arthromitus (segmented filamentous bacteria, SFB 54 )) can induce obesity 55 . In the present study, we found that Candidatus Arthromitus was dramatically reduced in the IIBV group compared to that in the INC group. However, due to the lack of records of body weight and body fat percentage (BFP) data, it was difficult to verify the direct effect of the reduction in the abundance of Firmicutes, such as Candidatus Arthromitus, on the body in this experiment. Fortunately, in the present study, we predicted the unobserved character states in the bacterial community by using PICRUSt, which is generally applied to investigate animal intestinal function 56 . With respect to the PICRUSt results, the "primary bile acid biosynthesis" and "secondary bile acid biosynthesis" pathways were significantly enriched in the ilea of the IIBV group chickens. Bile acids are cholesterol-derived compounds synthesized in the liver, and gut microbiota can transfer primary bile acids in the small intestine via deconjugation, dehydroxylation, or dehydrogenation to form so-called secondary bile acids, which promote the intestinal absorption of lipids and affect energy metabolism mainly via the farnesoid X receptor (FXR) and G protein-coupled receptor (TGR5) [57][58][59] . A great number of observational studies have indicated that meta-inflammatory disorders, such as obesity and type 2 diabetes (T2D), are always associated with an increase in total bile acid (BA) concentrations 60,61 . In addition, Annika et al. showed that the levels of iso-DCA, which is a common BA produced by bacteria, were significantly increased in humanized mice with low Clostridia abundance. As highlighted above, we know that the enrichment of bile acid biosynthesis is closely related to the loss of Clostridia, which may act as an important part of host metabolism in response to NIBV infections.
In conclusion, existing literature highlights the vital role of gut microbiota in the defence against viral infection. Our results suggested that relative abundance data from the ileal microbiota may differentiate the INC group chickens from the IIBV group chickens infected with NIBV. Microbiome profiles could be a biological indicator Figure 6. Predicted differential KEGG pathways in the INC and IIBV groups. The PICRUSt-predicted relative abundance of the KEGG pathways (KEGG level 3) was compared between the INC and IIBV groups. The statistical analysis was conducted using White's nonparametric t-test between the two groups, and only the significant differences in the KEGG pathways (with a P-value < 0.01) are shown. www.nature.com/scientificreports www.nature.com/scientificreports/ of IBV infection; therefore, further investigation into the mechanism of this shift will help us understand IBV infection and provide a new approach to diagnose, prevent and treat infections.

Viral strains. The IBV strain was isolated in Nanchang, China in 2011 and stored by the College of Animal
Science and Technology, Jiangxi Agricultural University and is designated SX9 throughout this study. At the affected farm, the diseased chickens had the same clinical symptoms: the kidney parenchyma of the dead birds was pale, swollen and mottled. These results show that the IBV strain SX9 has strong nephropathogenic characteristics. The IBV strain SX9 used in this study was propagated in 10-day-old specific-pathogen-free (SPF) embryonated eggs (Jinan sais poultry Co., Ltd., Shandong, China) by the allantoic route. The IBV viral titer (median embryo lethal dose, ELD 50 ) was evaluated in the 10-day-old SPF eggs and calculated by the Reed-Muench method.
Experimental design. A total of 240 healthy Hy-Line variety brown chickens were randomly divided into two experimental groups: a normal group (INC) (30 chickens per subgroup) and a diseased group (IIBV) (30 chickens per subgroup). As shown in Fig. 1A, 0.2 ml of 10 5 ELD 50 of SX9 was intranasally inoculated into each chicken in the IIBV group at 28 days of age, while 0.2 ml of sterile physiological saline was administered to each chicken in the INC group. Two chickens were randomly chosen per subgroup at 10 dpi and were euthanized by carbon inhalation. The livers, spleens, kidneys, ilea and jejuna were quickly separated from the bodies in a sterile environment. The ilea were isolated for morphological assessment, and ileal content samples were gathered for bacterial V3-V4 16S rRNA gene sequencing. The rest of the animals were euthanized according to the animal care guidelines of Jiangxi Agricultural University.

Viral load quantification by RT-qPCR.
Total RNA was extracted from the livers, spleens, kidneys, ilea and jejuna with RNAiso Plus (Takara, Japan) following the manufacturer's protocol. The RNA was quantified using a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific, USA). The cDNA was prepared from 2 μg of RNA. The QPCR was performed with primers (forward: 5′-CCATGGCAAGCGGTAAAGCAR-3′; reverse: 5′-CCACTCAAAGTTCATTCTCTCC-3′) and a QuantStudio 7 Flex Real-Time PCR system (ABI Thermo Fisher Scientific, USA). In brief, the amplification was performed using 10 μl volume reactions in a 96-well plate format with the following conditions: 94 °C for 30 s, then 36 cycles of 94 °C for 5 s, 60 °C for 30 s and 72 °C for 30 s. The IBV RNA copy levels were quantified by comparison with a standard curve generated using dilutions (6.67 × 10 3 -6.67 × 10 10 copies) of the IBV N gene containing plasmid. The IBV RNA levels are expressed as IBV genome copies per 0.2 μg of RNA using the standard curve.
Histopathology. Sections of the ileal tissues were stained with haematoxylin and eosin (H&E) for assessment of the general ileal morphometry. For each slice, fields were randomly selected, from which all villi were quantified by Image-Pro Plus 6.0.
16S rRNA gene sequencing. 16S rRNA gene sequencing was performed as described previously 62 .
In short, the total genomic DNA in the ileal contents was first extracted, and then, using the genomic DNA as a template, the 16S rRNA gene of the V3-V4 region was amplified using specific primers (forward 341 F: CCTAYGGGRBGCASCAG and reverse 806 R: GGACTACNNGGGTATCTAAT) tagged with the unique barcode. Then the desired size (approximately 400-450 bp) of the PCR products was selected to prepare the sequencing libraries, and the index codes were added. Finally, the samples were sequenced on an Illumina HiSeq. 2500 platform.
Statistical analysis of the data. Samples were created from the paired-end reads by cutting off their unique barcode and primer sequence. The paired-end reads were merged using Fast Length Adjustment of SHort reads (FLASH) software (version 1.2.7, http://ccb.jhu.edu/software/FLASH/) 63 , and raw FASTQ data has been submitted in the SRA database of the NCBI with accession number PRJNA 533742. To obtain high-quality clean tags, quality filtering of the raw tags was performed using the Quantitative Insights Into Microbial Ecology software (QIIME, version 1.9.1, http://qiime.org/scripts/split_libraries_fastq.html) 64 quality-controlled process. The chimaeric sequences were removed by using the UCHIME Algorithm (http://www.drive5.com/usearch/manual/uchime_algo.html), and then the effective tags were obtained 65 . The sequences were clustered into OTUs at 97% similarity by UPARSE software (version 7.0.1001, http://drive5.com/uparse/) 25 . In addition, the GreenGene database and MUSCLE software (version 3.8.31, http://www.drive5.com/muscle/) were used to annotate the taxonomic information and conduct the multiple sequence alignment, respectively 66 .
The alpha diversity index was calculated: Chao1 and ACE were used to calculate the bacterial community richness and the Shannon and Simpson indices were calculated to reflect the community diversity 67 . The beta diversity index was calculated using UniFrac and Bray-Curtis distances in QIIME to compare and analyse the composition of the bacterial communities in different samples. PCoA was analysed by the "WGCNA", "stats" and "ggplot2" packages in R. NMDS and ANOSIM were conducted using the "vegan" package, and a simper analysis was performed by the "simper" package. To find the differences in the biomarkers between the uninfected chickens and chickens infected with the NIBV, we performed a LEfSe analysis (http://huttenhower.sph.harvard.edu/lefse/) 30 with the following prerequisites: the p value was 0.05, and the threshold on the logarithmic linear discriminant analysis (LDA) score was 4.0 30,68 . In addition, a Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt, version 1.0.0, http://picrust.github.com/) 31 was performed to predict the ileal microbial community function of the two groups of chickens. Finally, a statistical analysis of the taxonomic and functional profiles (STAMP, version 2.1.3, http://kiwi.cs.dal.ca/Software/STAMP) 69 was used for further exploration of the results that were clustered in level 3 of the KEGG analysis module in the PICRUSt analysis.