The 16S rRNA gene sequencing of gut microbiota in chickens infected with different virulent Newcastle disease virus

Newcastle disease virus (NDV) is pathogenic to chickens, which is characterized by dyspnea, diarrhea, nervous disorder and hemorrhages. However, the inuence of different virulent NDV infection on the host gut microbiota composition is still poorly understood. In this study, twenty 21-day-old specic pathogen free chickens were inoculated with either the velogenic Herts33 NDV strain, lentogenic La Sota NDV strain or sterile phosphate buffer solution (PBS). Through 16S rRNA sequencing, the collected fecal samples of control and NDV infected chickens were examined.


Results
The results showed that the gut microbiota were mainly dominated by Firmicutes, Bacteroidetes and Proteobacteria in both healthy and NDV infected chickens. NDV infection altered the structure and composition of gut microbiota. As compared to PBS group, phylum Firmicutes were remarkably reduced, whereas Proteobacteria was signi cantly increased in velogenic NDV infected group. While the gut community structure has no signi cant differences between lentogenic NDV infected group and PBS group at phylum level. At genus level, Escherichia-Shigella was signi cantly increased in both velogenic and lentogenic NDV infected groups, but the lactobacillus was only remarkably decreased in velogenic NDV infected group. Collectively, different virulent NDV infection resulted in a different alteration of the gut microbiota in chickens, including a loss of probiotic bacteria and a expansion of pathogenic bacteria.

Conclusion
These results indicated that NDV with different virulence have different impact on chicken gut microbiota and may provide new insights into the intestinal pathogenesis of NDV.

Background
All vertebrate animals are inhabited by an immense population of microorganisms. The intestinal tracts maintain a particular rich and diverse microbial community with the number over trillions and species more than 1000 [1,2]. These amazing amount of gut microbes were previously thought to be mainly bene t for food sources utilization. Recently, with the developing of research, scholars found that these microbes also play an essential role in many aspects of the host's physiology, including nutrients digestion, immune system development, detoxi cation of some compounds, and resistance to pathogens [3][4][5][6]. Although the diversity, the roles and the importance of these microbes in animal's physiology have been illustrated, the biological signi cance of the presence of intestinal microbes in animals remains largely unclear. As the unique life history trait of birds that are different from other vertebrates, such as hatching from eggs, chickens are an interesting study object for intestinal microbes. However, research of the avian intestinal microbiota was thought to be relatively falled behind that of other vertebrates and the recent study about avian intestinal microbiota mainly focused on composition of gut microbiota at different development stages, different segments of gut, and different living condition. Little is known about the interaction between viral infection and avian gut microbiota. Existed reports were limited to avian in uenza virus, infectious bronchitis, marek's disease virus, infectious bursal disease virus and Newcastle disease virus [7][8][9][10][11][12][13][14]. With the ongoing prohibition of using antibiotic as growth promoter and the recognition of bene ts of a healthy gut microbiota played in promotion growth and resistance of viral and bacterial diseases [15,16], extensive study is still required to understand more about the interaction between virus infection and gut microbiota.
As a highly contagious avian disease, Newcastle disease (ND) causes hemorrhages and necrosis of the respiratory tract and the digestive tract, result in high morbidity and mortality in chicken and has caused great economic losses to the poultry industry. Newcastle disease virus (NDV), the causative agent of ND, belongs to the family Paramyxoviridae and has a single-stranded, non-segmented, negative-sense RNA genome. It's genome is approximately 15.2 kb in length and contains six genes in the order of 3'-NP-P-M-F-HN-L-5' [17].
According to the disease severity in chicken post infection, NDV strains are categorised as three pathotypes:  [13]. But does the impact of different virulent NDVs on chicken gut microbiota are same is still unknown and need further investigation. Here, we evaluated the in uence of different virulent NDV on gut microbiota composition in 21-day-old speci c pathogen free chickens by 16S rRNA sequencing technology.
To our knowledge, this is the rst report that illustrate the impact of different virulent NDV on chicken gut microbiota.

Materials And Methods
Viruses NDV strains La Sota and Herts33 were used in the present study. The La Sota strain is a class II genotype II lentogenic strain, and the Herts33 is a class II genotype IV virulent strain. These two strains were propagated in the allantoic cavity of 9-11 day-old embryonated speci c pathogen-free (SPF) chicken eggs. Allantoic uid was harvested from chicken embryos and stored at -70℃. The virus median tissue culture infective dose (TCID 50 ) was tested on DF-1 cell by Reed-Muench method.

Ethical statement
The experiments were performed in strict accordance with Animal Ethics Procedures and Guidelines of the Ministry of Health in China and the ARRIVE guidelines. All experimental procedures were approved and supervised by the Ethics Committee for the Care and Use of Laboratory Animals in Qinghai University, China.
Informed consent was obtained from the JINAN SAIS POULTRY CO., LTD in advance.

Experiment design
Twenty 2-week-old speci c pathogen free white Leghorns chickens were purchased from JINAN SAIS POULTRY CO., LTD. The chickens were maintained in bio-security isolation units with feed and water administered ad libitum. After acclimatizing for 1 week, all chickens were divided into three groups with seven birds in two experiment groups and six birds in control group, namely group Herts33 (n=7), group La Sota (n=7) and group PBS (n=6). Each bird in groups Herts33, and La Sota was challenged with 10 5 TCID 50 /100 μL of the Herts33 strain or La Sota strain via eye drop (50 μL) and intranasal (50 μL) routes (EI/IN), respectively. Birds in the PBS group were challenged with 100 μL of PBS. All birds were monitored daily for clinical signs (depression, respiratory signs, diarrhea, etc), mortality. Cloacal swabs were used to collect about 200 mg fecal sample from each bird at 3 to 5 days post challenge for fecal DNA isolation.

DNA extraction and library construction
Total genomic DNA was extracted from about 200 mg collected feces using QIAamp 96 PowerFecal QIAcube HT kit (QIAGEN) following the manufacturer's instructions. The concentration and purity of extracted DNA was veri ed with NanoDrop and agarose gel. Then the genome DNA was used as template to amplify V3-V4 variable regions of 16S rRNA genes with universal primers 343F 5'-(TACGGRAGGCAGCAG)-3' and 798R 5'-(AGGGTATCTAATCCT)-3' and Tks G ex DNA Polymerase (Takara). PCR were carried out in a 30 μl reaction mixture containing 2× G ex PCR buffer 15 μl, primer 343F (5 pmol/μl) 1 μl, primer 798R (5 pmol/μl) 1μl, Tks G ex DNA Polymerase 0.6 μl, and 50 ng DNA template. The PCR condition were initial denaturation at 94℃ for 5 min, followed by 26 cycles of denaturation at 94℃ for 30 s, annealing at 56℃ for 30 s and extension at 72℃ for 20 s, with a nal extension phase at 72℃ for 5 min. The PCR products were visualized using gel electrophoresis, and puri ed with AMPure XP beads (Agencourt). The puri ed rst round PCR product was used as template to conduct second round PCR with the index primer pairs adapter I5 primer and adapter I7 primer. The PCR reaction system was carried out in a 30 μl reaction mixture as the rst round PCR. The PCR condition were same as the rst round PCR except for the cycles reduced to seven. After puri cation with the AMPure XP beads again, the nal amplicon was quanti ed using Qubit dsDNA assay kit. Equal amounts of puri ed amplicon were pooled for subsequent sequencing using Illumina MiSeq system by oebiotech (Shanghai, China).

Bioinformatic analysis
Raw sequencing data were in FASTQ format. Paired-end reads were then preprocessed using Trimmomatic software to detect and cut off ambiguous bases (N). We also cut off low quality sequences with average quality score below 20 using sliding window trimming approach. After trimming, paired-end reads were assembled using FLASH software. Assemble parameters were: 10 bp of minimal overlapping, 200 bp of maximum overlapping and 20% of maximum mismatch rate. To obtain high quality clean tags, quality ltering of the raw tags was performed using QIIME software (version 1.8.0) quality-controlled process. The chimeric sequences were removed by using the UCHIME Algorithm. Then the clean reads were clustered to generate operational taxonomic units (OTUs) using VSEARCH software with 97% similarity. The representative read of each OTU was selected using QIIME package. All representative reads were annotated and blasted against Silva database Version 123 using RDP classi er (con dence threshold was 70%). In the present study, all sequences have been deposited to the National Center for Biotechnology Information (NCBI) database under accession number PRJNA700718.

Statistical Analyses
Differences between populations had been analyzed using parametric (ANOVA) and non-parametric statistical methods. All results were presented as the mean value (± SE). Differences between groups were declared signi cant at P< 0.05.

Sequencing results overview
In the present study, twenty fecal samples (seven Herts33 challenged, seven La Sota challenged, six PBS negative control) were collected and processed for 16S rRNA gene sequencing and analysis. After quality control, about 60617 to 72639 clean tags were obtained. And the valid tags were distributed between 46932 and 69273 post removing chimera. The average length of valid tags is 406.47 to 425.62 bp and the OUT number of each sample was distributed between 471 and 1477. The Good's coverage ranged from 99.32% to 99.59%, indicating a good sequencing depth enough to cover the majority of the gut microbiota in each sample.
A decrease in the microbial diversity in gut microbiota with NDV infection.
The fecal sample richness was evaluated by operational taxonomic units (OUT) counts in each sample. As shown in gure 1, the average number of observed OUT in the NDV infected chicken samples was more than that of the PBS control group (Fig.1 Fig.2b and Fig.2c).

Gut bacterial beta-diversity analysis
To analysis the similarities and differences of bacterial communities among these three groups chickens, the Bray-Curtis similarity were calculated. And the Bray-Curtis based analysis of similarities indicated that the microbiota among three groups were signi cantly different from each other (R= 0.3908, P= 0.001).
Furthermore, principal coordinate analysis (PCoA) were performed based on Bray-Curtis distances to visualize the similarity of the microbial community structure in different groups. As shown in gure 3, PC1 and PC2 account for 33.23 and 18.89% of the total variation. And there was distinguishing clustering of the samples from each group. However, partial samples from La Sota and PBS were close to each other. The PCoA result suggested distinct differences in the bacterial composition among the three groups.

NDV infection alter the gut microbiome composition in chickens
To elucidate the effect of NDV infection on gut bacterial composition, we evaluated the gut microbiota at different taxonomical levels. The overall bacterial composition of different groups at the phylum level was showed in gure 4A and 4a, sequences that accounted for very small proportions were combined as others.
From gure 4A and 4a, we found that Firmicutes, Proteobacteria and Bacteroidetes were the three most abundant phyla in all groups. The average relative abundance of phylum Firmicutes in Hetts33 group was signi cantly lower than that in other two groups (Fig.5a, ANOVA P 0.01), while the relative abundance of phylum Proteobacteria was signi cantly higher than that in La Sota and PBS groups (Fig.5a, ANOVA P 0.01).
When analyzed at the genus level, as shown in gure 4B and 4b, the main genera in these three group included lactobacillus, Escherichia-Shigella, Enterococcus and Bacteroides. The top 10 signi cantly different genus were lactobacillus, Escherichia-Shigella, enterococcus, GCA-900066575, Clostridium, Pseudomonas, Azospirillum, Pseudogracilibacillus, Weissella and Brachybacterium (Fig.5b, ANOVA, P 0.05). The relative abundance of genus lactobacillus in Hetts33 group was signi cantly lower than other two groups. While the relative abundance of genus Escherichia-Shigella in Herts33 group and La Sota group was signi cantly higher that of PBS group ( Fig.6a and 6b, T test P<0.05). However, the abundance of genus lactobacillus had no signi cant difference between La Sota group and PBS group (Fig.6b, T test P>0.05). And the relative abundance of genus enterococcus in hetts33 group and La Sota group was signi cantly lower that of PBS group.

Discussion
The intestine tract of chickens is populated with a relatively rich and diverse microbial community, including bacteria, viruses, fungi and protozoa, as other animals. These incredibly complex microbial community possess important functions for host on many aspects. At the same time, the intestinal microbiota is dynamic and in uenced by environment, diet, age, antibiotics, pathogen infection and other factors [19]. So the maintenance of a health gut microbiota is very important and contributes signi cantly to the overall health and performance of a ock [20]. If the structure and composition of gut microbiota is disturbed, this may have a severe impact on the chickens' grow performance and may enhance the risk for systemic diseases including infectious diseases [21]. Viruses and bacteria could interact with each other in the gut, and thus affect the virus replication and transmission [22,23]. Therefore, this study was designed to evaluate whether NDV infection could cause the alteration of chicken gut microbiota.
Previous study reported that the infection of NDV resulted in the disproportion of intestinal microbiota [13]. In the present study, we compared the gut microbiota between different virulent NDV infected chickens and noninfected chickens by 16S rRNA gene sequencing and found that NDV could alter the gut microbiota composition at different levels, which is in line with previous observations [13]. To examine whether vertical infection of NDV in uence the formation of intestinal community, Cui evaluated the effect of NDV infection on chick embryos at hatch. Their result showed that NDV infection decreased the richness and overall diversity of duodenal ora, but the richness and diversity of cecal micro ora was not affected. Our result is in accordance with Cui's result on cecal as the alpha diversity indexes were not signi cantly different between NDV infection groups and control group in the present work. The results of PCoA indicated that the NDV infection altered the structure of gut microbiota, which is consistent with the results of previous study [13].
And from the PCoA results we concluded that different virulent NDV infection have a vary in uence on chicken microbiota.
Many studies have demonstrated that chicken gut microbiota consisting of three major bacterial phyla, namely the Firmicutes, the Proteobacteria, and the Bacteroidetes. Our present study also found that the above three bacterial phyla were the predominant observed bacterial taxa, which con rmed previous observations [24]. However, the relative abundances of these three phyla was quantitatively different among our three groups. In PBS control group, the Firmicutes, the Bacteroidetes, and the Proteobacteria accounted for 80.98%, 16.07% and 2.31% of the total bacterial, respectively. While the proportion of Firmicutes, Bacteroidetes, and Proteobacteria in the velogenic Herts33 NDV infected group was 13.41%, 10.73% and 68.35%, and in the lentogenic La Sota NDV infected group was 74.12%, 3.7% and 21.37%. The functions of Firmicutes and Bacteroidetes are closely related with carbohydrate and protein metabolism and paly a role in energy production [25,26]. At the same time, some members in phyla Firmicutes could regulate the in ammation by induction of anti-in ammatory cytokines [27]. As a minor constituent in the fecal microbial community, the Proteobacteria accounted for only 2.31% in PBS group (Fig. 5) and this group included many pathogenic bacteria, such as Escherichia, Shigella, Salmonella, Clostridium Cluster XI, Vampirovibrio and so on [28]. As compared to PBS group, the increase of the Proteobacteria and decrease of Firmicutes in two NDV infected group (velogenic Herts33 VS PBS, P<0.01; lentogenic La Sota VS PBS, P>0.05 ) may be a sign of disease in chickens.
Lactobacillus are one of the predominant bacterial genera in the gastrointestinal tract of chicken [24], and great bene ts for chickens, such as help in carbohydrate fermentation, restriction of the replicate of other bacteria species by production of lactate, bacteriostatic and bactericidal substances [29][30][31][32]. In addition, lactobacillus could modulate the immune system and signi cant enhancement of the immune response was also observed in chicken [33]. Now, lactobacillus strains are actually considered as safe organisms and has been widely used as a probiotics to improve grow performance and inhibit the potential pathogenic microorganisms such as Salmonella and Escherichia-coli [34,35]. In this study, as the most abundant genus and top one different genus, the relative abundance of Lactobacillus in velogenic NDV infection group was signi cantly lower than that of PBS group (T test, P<0.01), but that has no signi cant differences between the lentogenic NDV infection group and PBS group (T test, P>0.05). The decline of Lactobacillus were also observed in chickens post Eimeria tenella or H9N2 avian in uenza virus infection [8,36]. It has been shown that some Lactobacillus can enhance the IFN and IL-22 production and response [37,38], and higher abundance of the Lactobacillus was associated with restoration of the epithelial barrier integrity [39,40]. Furthermore, Oral administration of Lactobacillus can effectively relieve diarrhea by regulating intestinal micro ora and improving immune system function [41]. So we speculated that the differences in the In contrast, opportunistic pathogen Escherichia-Shigella, which belongs to family Enterobacteriaceae, was signi cantly increased both in two NDV infected groups. In velogenic NDV infected group, the average relative abundance of Escherichia-Shigella increase from 1.4% to 53.3%, while that increased from 1.4% to 19.8% in lentogenic NDV infected group. The increasement of Escherichia-Shigella was also observed in the infection of H9N2 avian in uenza virus, ALV-J, duck reovirus and Eimeria tenella [42,43]. And some reports suggested a positive correlation between the abundance of Escherichia-Shigella and the development of necrotic enteritis in chickens [44]. Moreover, previous studies have shown found that IFN-α, IFN-β, IFN-γ, and IL-22 expression were negatively correlated with Clostridium cluster-XI, Escherichia, and Shigella species post AIV infection [9,37]. In H9N2 AIV infected chickens, elevated level of IFNs caused the dysbiosis of commensal gut microbiota and decreased the number of lactic acid producing bacteria due to an increased relative abundance of pathogenic Proteobacteria, including Shigella, which produce in ammation in GIT [45]. These date indicate that NDV infection might increase the possibility of subsequent infection by other pathogens.
Does the different expression level of cytokines, such as IFNs, IL22, IL17, which were induced by NDV infection, account for differences in gut microbiota alteration and clinical symptoms post different virulent NDV infection need further study.

Conclusions
In conclusion, our study demonstrated that signi cant dysbiosis occurs in the gut microbiota of chickens post NDV infection. The alteration of gut microbiota was dominant by an increased relative abundance of the pathogen Escherichia-Shigella, and apparent decrease in the level of the non-pathogenic bacteria, for example lactobacillus. These observation indicate that a fundamental alteration in the chicken gut microbiota post NDV infection. Further investigation of the mechanisms underlying these interactions could help reveal useful targets ant treatment approaches for restoring the gut microbiota to help combat NDV.

Declarations
This work was supported by the Natural Science Foundation of Qinghai Province of China (Grant No. 2018-ZJ-952Q).

Availability of data and materials
All data generated or analyzed during this study are included in this published article, and also available from the corresponding author on reasonable request.

Ethics approval and consent to participate
The experiments were performed in strict accordance with Animal Ethics Procedures and Guidelines of the Ministry of Health in China and the ARRIVE guidelines. All experimental procedures were approved and supervised by the Ethics Committee for the Care and Use of Laboratory Animals in Qinghai University, China.
Informed consent was obtained from the animal owners in advance.

Consent for publication
Not applicable.   The microbial diversity index analysis. a Chao1 index. b Shannon index. c Simpson index.   The relative abundance of top 10 different bacteria in three groups (ANOVA), expressed as an average percentage of the total. a At phylum level. b At genus level.