Background
Salmonella enterica, serovar Enteritidis (
S. Enteritidis) is a common zoonotic pathogen that causes huge economic losses in the poultry industry. Humans can be infected with
S. Enteritidis by consuming undercooked chicken products [
1].
S. Enteritidis mainly colonizes the chicken cecum [
2]. The cecal microbiome is primarily composed of
Firmicutes,
Bacteroidetes and
Proteobacteria [
3‐
5],
Bifidobacterium provides endogenous sources of vitamins to enhance the chicken’s immune function [
6]. Short-chain fatty acids produced by
Streptococcus faecalis can reduce intestinal pH value and inhibit the growth of pathogens.
The intestinal microbiome matures as the chicken grows, developing rapidly from days [
1‐
3,
7], and then tending to be stable. It has been reported that the early stages of hatching are the critical period for the establishment of chickens’ intestinal microbiota [
8‐
10]. The complex intestinal microbial ecology, especially the microbiota of the gut developed in infancy, is closely intertwined with immune development [
11]. Pathogen infection can affect host intestinal microbial composition.
S. Enteritidis infection in young layer chicks significantly reduces the overall diversity of the microbiota population, promoting expansion of the Enterobacteriaceae family [
12]. The gut microbiome in the ceca of pigs changed with
S. enterica, serovar Typhimurium challenge [
13].
Modern high-throughput deoxyribonucleic acid (DNA) sequencing approaches based on the 16S ribosomal RNA (rRNA) sequence—such as pyrosequencing, gene chip and single-strand conformation polymorphism—have been widely used to characterize the chicken gut microbiome [
12,
14‐
16]. This has sped up understanding of the structural composition of intestinal microbiota as well as the interaction between these microorganisms and their host [
17]. We conducted the current study to assess the diversity of the chicken cecal microbiome induced by
S. Enteritidis inoculation and to provide a scientifically theoretical basis of interaction between pathogens and gut microbiota.
Methods
Animal inoculation
We used Jining Bairi chicken, a regional Chinese breed, in the current study. All chickens were provided by Shandong Bairi Chicken Breeding Co., Ltd. (Shandong, China). We purchased the S. Enteritidis strain (CVCC3377) used for the inoculation from the China Veterinary Culture Collection Center, Beijing.
We collected meconium from each individual chicken and checked it for S. Enteritidis negativity using the plating method. In total, we randomly assigned 168 two-day-old S. Enteritidis–negative chickens into 2 groups of 84 chickens each treated (trt) and control (con) groups and raised them in 2 separate incubators with the same environmental conditions and with access to food and water ad libitum. Each chicken in the treated group was orally inoculated with 0.3 ml 109 colony-forming units (cfu)/ml S. Enteritidis inoculant, while chickens in the control group were mock-inoculated with the same amount of sterile phosphate buffer saline (PBS). Twelve chickens from each of the treated and control groups were euthanized by cervical dislocation for sample collection at 1, 3, 7, 14, 21, 28 and 35 days post-inoculation (dpi). All animal procedures were approved by Shandong Agricultural University Animal Care and Use Committee.
Enumeration of S. Enteritidis in cecal content
We collected fresh cecal content from 1 cecal pouch in each chicken, weighed it, put it on ice and sent it to laboratory for S. Enteritidis enumeration. We then collected the cecal content from another cecal pouch in the same chicken and froze it at − 20 °C for DNA extraction. To assess the amount of S. Enteritidis in the cecal content from each individual chicken, we diluted the samples, plated them on Salmonella–Shigella agar and incubated them for 24 h at 37 °C. Each sample was processed in triplicate.
At 1 and 3 dpi, cecal content from 3 randomly selected chickens were mixed with equal amount to get enough sample for DNA extraction. In total, 3 mixed cecal content samples were obtained from treated group and 3 from control group at 1 and 3 dpi, respectively. At each time point from 7 to 35 dpi, individual cecal content was randomly selected and used for DNA extraction. Genomic DNA was extracted from 500 mg cecal content using a fecal genomic DNA extraction kit (CWBio, Beijing, China). We examined DNA integrity by agarose gel electrophoresis and measured DNA concentration and purity using a DS-11 spectrophotometer (DeNovix, Wilmington, Delaware, US). We stored the qualified DNA samples at − 20 °C for further analysis.
We performed PCR amplification with forward (5′-ACTCCTACGGGAGGCAGCA-3′) and reverse (5′-GGACTACHVGGGTWTCTAAT-3′) primers targeting the V3 and V4 segments of the 16S rRNA gene. PCR conditions were set for initial denaturation at 95 °C for 5 min, followed by 25 cycles of 95 °C for 30 s, 50 °C for 30 s and 72 °C for 40 s, with a final extension step at 72 °C for 7 min. We submitted the amplicons to Biomarker Technologies Co., Ltd (Beijing, China) to generate 250 paired-end reads on the MiSeq sequencing platform (Illumina, Inc., San Diego, California, US). The data has been deposited into Sequence Read Archive (National Center for Biotechnology Information, National Institutes of Health, Bethesda, Maryland, US) [
18,
19].
16S rRNA gene sequencing and data analysis
FLASH [
20] was used to merge paired end reads before assembly. Trimmomatic [
21] was used to remove adapters, low-quality sequences and reads shorter than 36 bases. We predicted the chimeric sequences and excluded them from the analysis [
22] to get high-quality tag sequences. Similar sequences were clustered into operational taxonomic units (OTU) at a 97% identity threshold using UCLUST software version 1.2.22 (
https://www.drive5.com/) [
23]. We filtered the OTUs using 0.005% of the number of all sequences as thresholds [
24].
We analyzed the alpha diversity metrics, including Chao1 (richness estimate) and Shannon and Simpson diversity indices, using mothur software version 1.30 (mothur project, Department of Microbiology & Immunology, University of Michigan, Ann Arbor, Michigan, US) [
25]. Beta diversity was analyzed using unweighted UniFrac distances [
26] followed by principal-component analysis (PCA). We generated a cluster of all samples based on unweighted UniFrac distances using the heatmap function in R 3.4 software (
https://www.r-project.org/), constructed a polygenetic tree of all samples using Molecular Evolutionary Genetics Analysis (MEGA) 7 [
27] software and identified cladograms with statistically significant taxonomic differences between the groups. In our linear discriminant analysis with effect size (LEfSe;
http://huttenhower.sph.harvard.edu/galaxy), we used a linear discriminant analysis (LDA) value of 4.0 and effect size threshold of 2. We performed our redundancy analysis (RDA) at the bacterial-group level.
Statistical analysis
We evaluated OTUs and alpha and beta diversity between the 2 groups at each time point using unpaired t-tests. We determined alpha and beta diversity metrics across different time points for both groups using analysis of variance (ANOVA). We used the Bonferroni method to compare multiple means. P < 0.05 was considered significant.
Discussion
In the current study, we analyzed the cecal microbiota profile at different time points post–
S. Enteritidis inoculation using 16S rRNA sequencing to elucidate temporal microbiota composition and the interaction between
S. Enteritidis and cecal microbiota. The composition of the gut microbiome reflects co-evolution across the inhabiting microbes’ genetic, immune and metabolic interactions with the host [
28]. High-throughput sequencing makes a composition-based microbial time series feasible by permitting analysis of temporal variations.
Development, genetics and
S. Enteritidis inoculation contributed to cecal microbiome diversity. Age and developmental stage can have a significant impact on the microbiota richness and diversity [
14]. In the current study, for the control group, the microbiota richness increased from 1 to 7 dpi and became stable after 14 dpi. Moreover, richness (Chao1) differed significantly between chickens at 3 and 7 dpi. The Simpson and Shannon indices were significantly different between 1 and 3 dpi, then become stable. However, it has been reported that Shannon diversity is significantly different between 2 and 7 dpi in the control group [
12]. The different genetic background and development of chickens used could contribute to the different findings across time points in different studies. It has been reported that both richness and diversity are significantly higher in 6-week-old broiler chickens than in 1- or 3-week-old chicks [
14], which is consistent with our results.
The microbial composition varies with development.
Firmicutes,
Bacteroidetes and
Proteobacteria are the 3 most abundant phyla in ceca, respectively, representing 44–56, 23–46 and 1–16% of all taxa in the cecum [
29], dominating microbiota composition in 1-week-old control chickens [
12]. In the current study, in the control group,
Firmicutes dominated microbiota composition from 1 to 35 dpi, followed by
Proteobacteria (1 dpi) and
Bacteroidetes (at 14–35 dpi); whereas in chickens inoculated with
S. Enteritidis at 7 dpi, more-abundant
Firmicutes was observed.
Firmicutes followed by
Bacteroidetes were the 2 most common phyla found in pigs after
S. typhimurium infection [
13]. Similar results have been reported previously [
15]. The results of our PCA analysis indicated that
S. Enteritidis inoculation moderately affected microbial community structure and composition in cecal content. Microbiome diversity was more affected by age than by treatment, which is consistent with previous results [
7].
The chicken gut has two tasks that often interfere with one another: nutrient absorption and defense against pathogens. The microbial community plays an important role in maintaining normal physiological homeostasis, modulating the host immune system and influencing organ development and host metabolism [
30]. Competitive exclusion (physical occupation, resource competition and direct physical or chemical insult to the potential colonist) is the main strategy by which gut microbiota exclude pathogens [
31]. Normal microbiota contribute to the susceptibility of chicks to bacterial infection [
32].
S. Enteritidis inoculation can affect the composition of the microbiome by changing the relative abundance of certain microbes. But the changes in cecal microbiota after
S. Enteritidis inoculation were quite weak, which was similar to previous reports [
15,
33]. Such inoculation significantly affected the abundance of microbiota at the genus level at each time point except for 1 and 21 dpi. It could take some time for
S. Enteritidis to alter the abundance of cecal microbiota. Significantly different abundance in the microbiome at the genus level could be seen between the treated and control groups. The cecal microbiome community changes over time to protect the gut from
S. Enteritidis inoculation.
Bifidobacterium,
Rikenella,
Coprococcus and Lachnospiraceae
incertae sedis played major roles in protecting against
S. Enteritidis inoculation at an early stage (before 7 dpi), while
Bacillus,
Blautia,
Shuttleworthia and
Flavonifractor did so at a later stage (after 7 dpi; Additional file
4). Moreover,
Bacillus positively correlated with
Blautia and
Flavonifractor (Fig.
12). It has been reported that older chickens are more resistant to
Salmonella infection than are younger ones [
34,
35], suggesting that gut microbiota play an important role in host resistance and the mature host immune system. Some studies also support this idea that early colonizers influence the relative abundance of the microbiome but the effect weakens over the long term [
36].
We observed a greatly significant change in
Bifidobacterium after
S. Enteritidis inoculation. We assume that
S. Enteritidis inoculation stimulates the immune system and
Bifidobacterium proliferates as a biofilm to defend against pathogen infection. Increased
Bacillus was also found in the current study (Additional file
4). It has been reported that
S. Enteritidis can use organic acid produced by
Bacillus as an energy source [
37].
Bacillus appears only in the ceca of old chickens [
10], which is identical to our finding that
Bacillus significantly increased at 28 dpi.
Blautia, as a functional core group of intestinal flora, produces short-chain fatty acids by fermentation in the intestine; this benefits the host by lowering cecal pH value [
38]. This can explain why
Blautia content in the control group was higher than in the treated group only at late ages (28 and 35 dpi). The dramatic increase may benefit the host via resistance to pathogens. The volatile fatty acids produced by fermentation of the beneficial bacteria help control the amount of
Salmonella in poultry [
39].
Authors’ contributions
XL designed the experiment and revised the manuscript, LLiu did animal trial and drafted the manuscript. LLin performed the experiment, analyzed the data, and drafted the manuscript. HT and XF analyzed the data and revised the manuscript. LZ and NX helped performed the experiment, collected the samples. MLi and MLiu extracted DNA and analyzed the sequencing data. All authors read and approved the final manuscript.