Application of high-throughput sequencing for microbial diversity detection in feces of specific-pathogen-free ducks

Application of high-throughput sequencing for microbial diversity detection in feces of... ABSTRACT High-throughput sequencing technologies play important roles in the study of animal enteric microorganisms. Fecal samples collected from 3 strains (9 samples) of specific-pathogen-free (SPF) ducks raised in an isolator. Duck intestinal bacterial flora were analyzed by sequencing distinct regions of the 16S rRNA genes using an Illumina HiSeq2500 platform. The result showed the most abundant primary enteric microbial phyla of 3 different strains of 70-week-old SPF ducks were Firmicutes, Proteobacteria, and Bacteroidetes, the most abundant primary enteric microbial order were Clostridiales, Lactobacillales, and Aeromonadales, and the most abundant primary enteric microbial genera were Bacteroides, Comamonas, and Enterococcus. In addition, the 3 duck strains harbored different compositions of the microorganisms, but these differences were not significant. Duck intestinal bacterial flora were analyzed using a high-throughput sequencing approach to further understand the distribution of intestinal flora and the biology of SPF ducks to ultimately benefit the purification of SPF ducks. INTRODUCTION Experimental poultry for epidemiological research are bred by artificial cultivation and carry specific microorganisms to establish control birds with clear genetic backgrounds. At present, specific-pathogen-free (SPF) poultry are used in poultry research and for production of biological products and experimental raw materials. SPF chickens have been widely used for poultry disease research and development of poultry vaccines, while little research is conducted using SPF ducks (Wu, 2013). Long periods of natural selection under specific environmental conditions in combination with feeding and rearing patterns are needed to produce various types of animals with characteristic intestinal ecosystems that contain diverse and complex microbial populations (Rastall, 2004; Ley et al., 2008; Costello et al., 2009). The mutual influence and interactions of the microbial community with the host constitute the complex and diverse gut ecosystem (gut microbiota) of the host (Savage, 1977). The composition of intestinal flora is closely related to host nutrition, metabolism, and immunity (Ley et al., 2006). The host provides the required conditions and nutrients for growth and reproduction of gut microbes (i.e., oxygen, temperature, and pH), while the intestinal bacteria degrade and ferment some carbohydrates that the host does not produce for production of energy to improve the energy utilization ratio (Nicholson et al., 2005). Gut microbes also play important roles in host metabolism, aid against pathogen attack, promote the development of immune organs, and activate the immune system. In recent years, with the development of the second generation sequencing technologies, the technology has been applied to the study of the structure and function of the gastrointestinal microbial community in humans (Turnbaugh et al., 2009; Qin et al., 2010), chickens (Lu and Domingo, 2008), dogs (Swanson et al., 2011), pigs (Lamendella et al., 2011), cats (Tun et al., 2012), and camels (Bhatt et al., 2013), among others. Soo-Je et al. (Park et al., 2014) employed a high-throughput sequencing technique to sequence the genomes of intestinal flora to identify differences in meat quality and weight of 2 groups of pigs, and found differences in the populations of Oscillibacter, Lactobacillus, Rothia, and Clostridium species between the 2 groups, which indicated that intestinal flora may be associated with fat accumulation in pork. Rothia and Clostridium species produce linoleic acid and short-chain fatty acids, which contribute to the health and growth of pigs. The predominate phylotypes in chicken fecal samples were Firmicutes, Bacteroidetes, and Proteobacteria by 16S rRNA sequencing technique (Lu and Domingo, 2008). A 454 pyrosequencing study of 6 dog fecal samples that employed random sequencing of microbial genomes found that Bacteroidetes/Chlorobi and Firmicutes accounted for about 35% of the analyzed sequences, followed by Proteobacteria (13%–15%) and Fusobacteria (7%–8%) (Swanson et al., 2011). 16S rRNA sequencing technique studies of pig feces indicated that the dominant microorganisms were Firmicutes and Bacteroidetes (Lamendella et al., 2011). A study of cat gut microbes conducted by Tun (1996) using the 454 pyrophosphate sequencing method determined the proportions of intestinal bacteria and identified a potential pathogen associated with antibiotic resistance genes (Tun et al., 2012). The Jin Ding duck, also known as the green head duck and South China duck, is a type of hemp duck that is a traditional variety of poultry in Fu Jian Province. In a previous study, Jin Ding ducks were hatched from progenitor Jin Ding duck eggs obtained from the JiangSu FengDa waterfowl breeding field of the National Waterfowl Germplasm Resource Gene Pool (Taizhou, China) and raised in an isolator from the age of 1 day to 70 weeks to produce ducks free from specific pathogens (i.e., highly pathogenic avian influenza virus, duck hepatitis virus, duck enteritis virus, avian lymphoid leukemia virus, avian reovirus, duck/goose parvovirus, duck circovirus, duck tembusu virus, and group III poultry adenoviruses). However, the bacterial distribution of SPF ducks remains unclear. In the present study, high-throughput sequencing technology was first employed to sequence SPF duck intestinal flora to further elucidate the distribution of intestinal flora and the biology of different strains SPF ducks to provide a reference for the purification of SPF ducks. MATERIALS AND METHODS Ethical Statement This study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and approved by the Harbin Veterinary Research Institute. Extraction of Microbial Genomic DNA Fecal samples were collected from 3 strains of SPF ducks (samples A1, A2, A3, B1, B2, B3, and C1, C2, C3) and stored at −80 °C for subsequent sequencing until assayed. Seventy-week-old SPF ducks were provided by the Center of Laboratory Animal of Harbin Veterinary Research Institute. The bacteria were pelleted by centrifugation and resuspended in phosphate-buffered saline solution. Total microbial genomic DNA from the samples was extracted using the cetyltrimethyl ammonium bromide/sodium dodecyl sulfate (CTAB/SDS) method. DNA concentration and purity was monitored on 1% agarose gels. For the following experiments, DNA was diluted to a concentration of 10 ng/μL in sterile water. Then quantification, qualification, mixing, and purification of the polymerase chain reaction (PCR) products. Amplicon Generation The V4 region of 16S rRNA genes was amplified using an Illumina HiSeq2500 platform (Novogene, Beijing, China) in accordance with the standard protocol of the manufacturer. All PCR reactions were carried out with Phusion® High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA). Library Preparation and Sequencing Sequencing libraries were generated using the TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina, San Diego, CA), following manufacturer's recommendations, and index codes were added. The library quality was assessed using a Qubit® 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA) and an Agilent Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA). The library was sequenced on an Illumina HiSeq2500 platform and 250-bp paired-end reads were generated. Data Analysis Raw sequencing data contains joint information, low-mass bases, and unmeasured bases (N), which can cause significant interference to subsequent information analysis. Therefore, measures were implemented to rule out invalid data to ensure valid analysis of biological information. Through careful filtering methods to get rid of the interference information, the resulting data were validated as clean data or clean reads. Flash software (Magoc and Salzberg, 2011) was used to filter out reads of joint and barcode sequences, and to splice overlapping reads. QIIME software (Caporaso et al., 2011) was used to filter splicing data, low-quality bases, and chimeric sequences in spliced sequences. Clustering and species classification analysis was conducted using the operational taxonomic units (OTUs) of the clean data. According to the clustering result, each representative OTU sequence was annotated to a specific species. At the same time, the abundance of OTUs, alpha diversity calculation, and petal figure and Venn diagram analyses were conducted to identify species richness and evenness of information of specific or shared OTU information between different samples and groups. System trees were constructed to compare multiple sequence alignments of OTUs and to further identify differences in community structures among different samples and groups. The results are presented as dimension reduction images and sample clustering trees derived from principal component analysis, principal coordinates analysis, and nonmetric multidimensional scaling. Statistical Analysis MetaStat software (MetaStat, Inc., Boston, MA) was used to analyze the 10 most abundant taxonomic sequence tags of the 3 sample groups. A probability (P) value of < 0.05 was considered statistically significant. RESULTS Sequencing and Quality Control Sequencing of the DNA sequences extracted from the feces of SPF ducks to obtain raw sequence (PE reads). The length of the effective sequence tags varied between 420 and 427 bp. Joint information, low mass bases, unmeasured bases (N), and chimeras were removed from the raw sequencing data to obtain clean sequences. A total number of 36,513, 32,533, and 37,193 clean tags were obtained for samples A, B, and C, respectively. Sequencing data quality was mainly distributed above Q20, so as to ensure normality of the subsequent advanced analysis. Redundancy of the clean sequence tags was determined using mothur software (http://www.mothur.org/), which identified 31,296, 29,185, and 30,785 unique sequence tags for strains A, B, and C, respectively (Table 1). Each unique sequence tag was compared to the sequences of the 16S rRNA genes of distinct regions using the Basic Local Alignment Search Tool for Nucleotides (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch). Table 1. Data preprocessing statistics and quality control. Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  1Raw PE means the original PE reads off the computer. 2Raw Tags refers to splice sequence Tags. 3Clean Tags is pointing to the tags that have filtered the low-quality and short length sequence. 4After the Effective Tags refers to sequence tags have filtered chimeras, and eventually for subsequent analysis. 5Base is refers to the final Effective Base number of the data. 6The AvgLen refers to the average length of the Effective Tags. 7Q20 and Q30 refers to the Effective quality value is greater than 20 bases in Tags (sequencing error rate is less than 1%), and 30 (sequencing error rate is less than 0.1%) percentage of bases. View Large Table 1. Data preprocessing statistics and quality control. Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  1Raw PE means the original PE reads off the computer. 2Raw Tags refers to splice sequence Tags. 3Clean Tags is pointing to the tags that have filtered the low-quality and short length sequence. 4After the Effective Tags refers to sequence tags have filtered chimeras, and eventually for subsequent analysis. 5Base is refers to the final Effective Base number of the data. 6The AvgLen refers to the average length of the Effective Tags. 7Q20 and Q30 refers to the Effective quality value is greater than 20 bases in Tags (sequencing error rate is less than 1%), and 30 (sequencing error rate is less than 0.1%) percentage of bases. View Large Data Indices In order to determine the diversity of species among the samples, clusters of the effective tags of all samples, with 97% identities, were used for clustering the OTU sequences. All effective tags of the samples shared sequence identities of > 97% to those of known species, which almost accounted for 100% of the OTUs of samples A and C, suggesting that no new genera or species were detected in samples A and C, while 3 sequence tags were unclassified. Therefore, further analysis is needed to identify microorganisms inhabiting the duck gastrointestinal tract (Figure 1). Figure 1. View largeDownload slide Statistics of each sample OTUs clustering and annotation. Total Tags (red): refers to the number of total tags each sample; Unique Tags (orange): refers to the number of tags that can’t be clustering to OTUs (sequence cannot be clustered OTUs will not be used for subsequent analysis); Taxon Tags (blue): refers to the number of Tags for building OTUs and annotation information; Unclassified Tags (tea green): refers to the number of Tags without annotation information; OTUs (purple): refers to the Number of OTUs in each sample. Figure 1. View largeDownload slide Statistics of each sample OTUs clustering and annotation. Total Tags (red): refers to the number of total tags each sample; Unique Tags (orange): refers to the number of tags that can’t be clustering to OTUs (sequence cannot be clustered OTUs will not be used for subsequent analysis); Taxon Tags (blue): refers to the number of Tags for building OTUs and annotation information; Unclassified Tags (tea green): refers to the number of Tags without annotation information; OTUs (purple): refers to the Number of OTUs in each sample. Further analysis of the sequence tags of samples A, B, and C (770, 587, and 769 OTUs, respectively) showed that 499 OTUs were detected in all samples. Samples A and B shared 540 OTUs, samples A and C shared 664 OTUs, and samples B and C shared 529 OTUs. In addition, 65 OTUs were unique to sample A, 17 OTUs were unique to sample B, and 75 OTUs were unique to sample C (Figure 2). Figure 2. View largeDownload slide Venn Graph. Venn Graph was drawn after homogenization processing for all samples. Each circle in the figure represents a sample, the number in circle and circle overlap between representative samples. Number of OTUs were no overlap number represents the number of unique OTUs in the sample. Figure 2. View largeDownload slide Venn Graph. Venn Graph was drawn after homogenization processing for all samples. Each circle in the figure represents a sample, the number in circle and circle overlap between representative samples. Number of OTUs were no overlap number represents the number of unique OTUs in the sample. The majority (>98%) of sequences were classified at below the domain level and more than 60% of the tags were assigned to a genus. However, only select sequence tags could be classified at the species level (Figure 3). Figure 3. View largeDownload slide The number of sequence at the different taxonomy levels obtained from 3 samples. Each bar represents the number of tags that were assigned to the taxonomic level for the identified bacteria. Figure 3. View largeDownload slide The number of sequence at the different taxonomy levels obtained from 3 samples. Each bar represents the number of tags that were assigned to the taxonomic level for the identified bacteria. Species rarefaction curves of the observed species were firstly very steep and then all curves gradually tended to be flat at the end, which illustrated that the existing sequencing data volume was reasonable (Figure 4a).The rank-abundance curve directly showed relative richness and evenness of microbial species in all 9 samples. Due to the complex intestinal flora in feces, the slopes of curves were small, which indicated a high degree of evenness among the microbial (Figure 4b). Figure 4. View largeDownload slide (a) Species rarefaction curves of 3 strains. (b) Rank-abundance curves of 3 strains. Figure 4. View largeDownload slide (a) Species rarefaction curves of 3 strains. (b) Rank-abundance curves of 3 strains. Alpha diversity was applied to clarify the richness and diversity of microbes in the samples. However, according to the alpha diversity metrics, sample B was also characterized by a lower Shannon entropy and lower ACE index, indicating that the microbial community in sample B was less diverse than in samples A and C. Additionally, calculation of the beta diversity indicated comparative diversities of microbial communities in the different samples. Microbial Community Analysis in the Three Strains Taxonomic information for each representative sequence was determined using the GreenGene database (http://greengenes.lbl.gov/cgi-bin/nph-index.cgi) with Ribosomal Database Project classifiers. Based on the average relative abundance analysis, at the phylum level, the majority of sequences in the 3 groups samples were assigned to 10 phyla: Firmicutes, Bacteroidetes, Proteobacteria, Actinobacteria, Fusobacteria, Synergistetes, Verrucomicrobia, Tenericutes, Cyanobacteria, and Elusimicrobia. The top 3 phyla of strain A were Firmicutes (68.23%), Bacteroidetes (18.29%), and Proteobacteria (10.99%). The top 3 phyla of strain B were Firmicutes (56.39%), Proteobacteria (31.72%), and Bacteroidetes (10.17%), while the top 3 phyla of strain C were Firmicutes (74.04%), Proteobacteria (12.79%), and Bacteroidetes (9%). Although there were few Synergistetes and Elusimicrobia species in strains A and C, none of the sequence tags in group B were assigned to these 2 phyla (Figure 5a). No significant differences (P > 0.05) were observed among the 3 groups of samples in any phylum level. Figure 5. View largeDownload slide (a) Species relative abundance of the phylum level. “Others” means the sum of the relative abundance of all the other phylum. (b) Species relative abundance of the order level. “Others” means the sum of the relative abundance of all the other order. (c) Species relative abundance of the genus level. “Others” means the sum of the relative abundance of all the other genus. Figure 5. View largeDownload slide (a) Species relative abundance of the phylum level. “Others” means the sum of the relative abundance of all the other phylum. (b) Species relative abundance of the order level. “Others” means the sum of the relative abundance of all the other order. (c) Species relative abundance of the genus level. “Others” means the sum of the relative abundance of all the other genus. Based on average relative abundance analysis, at the order level, the majority of sequences in the 3 group samples were assigned to 10 orders: Clostridiales, Lactobacillales, Aeromonadales, Bacteroidales, Burkholderiales, Bacillales, Actinomycetales, Erysipelotrichales, Xanthomonadales, Pseudomonadales. All of the top 10 orders existed in each of the 3 groups. The top 3 orders of group A were Clostridiales (37.36%), Lactobacillales (22.57%), and Bacteroidales (18.02%). The top 3 orders of group B were Lactobacillales (29.19%), Clostridiales (20.27%), and Aeromonadales (17.01%), while the top 3 orders of strain C were Clostridiales (34.46%), Lactobacillales (29.62%), and Bacteroidales (8.15%) (Figure 5b). No significant difference (P > 0.05) was observed among the 3 groups of samples in any order level. Based on average relative abundance analysis, at the genus level, the majority of sequences in the 3 groups of samples were assigned to 10 genera: Bacteroides, Comamonas, Enterococcus, Roseburia, Faecalibacterium, Vagococcus, Streptococcus, Aerococcus, Sporosarcina, and Blautia. All the top 10 genera existed in 3 groups. The top 3 genera of group A were Bacteroides (15.08%), Enterococcus (9.31%), and Faecalibacterium (3.54%). The top 3 genera of strain B were Enterococcus (12.48%), Comamonas (8.0%), and Bacteroides (7.38%), while the top 3 genera of strain C were Enterococcus (10.24%), Bacteroides (5.89%), and Faecalibacterium (5.16%) (Figure 5c). No significant difference (P > 0.05) was observed among the 3 groups of samples in any genus level. The species annotation results showed that the greatest abundance of species were of the phylum Firmicutes (70.51%). Approximately 70.58% of the sequence tags of this phylum were assigned to the class Bacilli and the remaining sequence tags were assigned to the class Clostridia. Microorganisms annotated to the class Bacilli were most abundant in sample C and least abundant in strain A. Most (90.4%) of the sequence tags of this class were assigned to the order Lactobacillales, and the remaining were assigned to the order Bacillales. Enterococcaceae was the most dominate family in strains B and C, accounting for approximately 18.6% and 13.79% of the sequence tags, respectively. Meanwhile, Planococcaceae was the most dominate family in strain A, accounting for approximately 31.46% of all sequence tags. The second-highest abundance of sequence tags in strain A were categorized to the phylum Bacteroidetes, which was ranked third in strains B and C. All sequence tags of all samples were assigned to class Bacteroidia, order Bacteroidales, family Bacteroidaceae, genus Bacteroides, and included the species Bamesiae, Coprophilus, Coprosuls, Fragilis, Ovatus, and Plebeius. Proteobacteria was the second most-prevalent phylum in strains B and C, and the third most-prevalent in strain A. All sequence tags were assigned to class Betaproteobacteria, order Burkholderia, family Comamonadaceae, genus Comamonas, It is worth mentioning that the abundance of phylum Proteobacteria (31.72%) in strain B was much higher than in the other 2 strains (Figure 6). Figure 6. View largeDownload slide Species classification tree in the samples. Different colors fan represented different samples in circle. The size of Fan refers to the microorganism classification of relative abundance in the sample. Classification of the Numbers below all samples means the relative abundance of the average percentage on this classification, there are 2 Numbers, the former said the percentage of all species, the latter said the percentage of selected species. Figure 6. View largeDownload slide Species classification tree in the samples. Different colors fan represented different samples in circle. The size of Fan refers to the microorganism classification of relative abundance in the sample. Classification of the Numbers below all samples means the relative abundance of the average percentage on this classification, there are 2 Numbers, the former said the percentage of all species, the latter said the percentage of selected species. Beta Diversity Index Analysis Beta Diversity Index Analysis was measured to estimate the divergence of microbial species among 3 groups, which based on the weighted UniFrac and Unweighted Unifrac distance cluster analysis, a dissimilarity coefficient for samples (Figure 7). The lower dissimilarity coefficients indicate the less divergence of microbial species. This study, the dissimilarity coefficients were measured to be 0.279 (0.346) bacterial diversities between strains A and B .The dissimilarity coefficients were measured to be 0.189 (0.238) bacterial diversities between strains A and C. The dissimilarity coefficients were measured to be 0.194 (0.314) bacterial diversities between group B and C. The result showed more divergence of microbial species between A and B than between A and C, between B and C. Figure 7. View largeDownload slide Beta diversity (weighted UniFrac) among 3 strains. Beta diversity indexes were measured based on weighted UniFrac and unweighted UniFrac distances. The upper numbers in the grid represent the weighted UniFrac; and the lower numbers in the grid represent the unweighted UniFrac distances, respectively. Figure 7. View largeDownload slide Beta diversity (weighted UniFrac) among 3 strains. Beta diversity indexes were measured based on weighted UniFrac and unweighted UniFrac distances. The upper numbers in the grid represent the weighted UniFrac; and the lower numbers in the grid represent the unweighted UniFrac distances, respectively. DISCUSSION Illumina sequencing has provided system development description of gastrointestinal microbial species. Compared with other molecular biology techniques, data generated by high-throughput sequencing technologies is more direct and comprehensive (Qin et al., 2010). Enteric microbes are closely related to animal feeding, energy metabolism, and health (Ley et al., 2008). Elucidation of the composition of enteric microbes will not only contribute to the understanding of gastrointestinal physiology, but also is conducive to the long-term development of animal husbandry. In the present study, the Illumina genome sequencing platform was used for confirmation of distinct regions of the 16S rRNA genes sequence, and determination of the enteric microbial species of different strains of SPF ducks. To the best of our knowledge, this is the first study to use a high-throughput technique to study enteric microbial species of different strains of SPF ducks. Previous high-throughput sequencing studies have shown that Bacteroidetes, Proteobacteria, Actinobacteria, and Firmicutes were the dominant species in the gastrointestinal tract of humans, loris, and other animals, although the relative proportion of species differed. According to the sequence results obtained in this study, the most common phyla were Firmicutes, Bacteroidetes, and Proteobacteria, which is similar to the gut microbes of many other species. Long-term intake of fat can increase the proportion of Firmicutes to Bacteroidetes, while the intake of dietary fiber may lead to a decrease in the ratio of Firmicutes to Bacteroidetes species (Peng et al., 2015). The proportion of Firmicutes bacteria was higher than in previous reports and this change can affect the metabolism and function of gut microbes, thereby altering the ability of the host to derive energy from food. In addition, Firmicutes may play important roles in the gastrointestinal health of ducks. The gastrointestinal microbial system of poultry is an organic body made up of various bacteria species that can be divided into harmful microbes and mutualistic microorganisms (Jeurissen et al., 2002). The results of the present showed the absence of other reported pathogenic microbes, such as Ornithobacterium rhinotracheale, Salmonella sp., Riemerella anatipestifer, Pasteurella multocida, and Staphylococcus aureus, while Enterococcus and Streptococcus species were found in all 3 samples. However, further research is needed to determine whether Enterococcus and Streptococcus are pathogenic to ducks. The animal gut is a dynamic ecosystem composed of a diversity of microbes that is influenced by the host itself as well as the external environment. At present, it is generally believed that the diet, age, and genotype of the host are important factors that affect the diversity of the intestinal microflora (Toivanen et al., 2001; Zoetendal et al., 2006). In this study, the main factor affecting the diversity of the intestinal microflora was thought to be the host genotype, although there was no significant difference observed in the proportion of 3 different strains, implying that the duck genotype may not necessarily influence the composition of gut microbes. In conclusion, the microbiome composition and diversity of the 3 strains of SPF ducks were investigated to compare the similarities and differences between duck strains. This study is the first to use high-throughput technology to investigate the enteric microbial diversity in ducks. Sequencing of the duck intestinal flora by high-throughput sequencing should provide useful information to further understand the distribution of intestinal flora and the biology of SPF ducks. The results of this study also help to advance our understanding of host and microbe symbiosis to facilitate improvements in animal health and production. ACKNOWLEDGMENTS This study was supported by the National Key Research and Development Program of China (2016YFD0500800) and the Chinese Academy of Agricultural Sciences Fundamental Scientific Research Funds (Y2016PT41). REFERENCES Bhatt V. D., Dande S. S., Patil N. V., Joshi C. G.. 2013. Molecular analysis of the bacterial microbiome in the fore stomach fluid from the dromedary camel (Camelus dromedarius). Mol. Biol. Rep . 40: 3363– 3371. Google Scholar CrossRef Search ADS PubMed  Caporaso J. G., Lauber C. L., Walters W. A., Berg-Lyons D., Lozupone C. A., Turnbaugh P. J., Fierer N., Knight R.. 2011. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl. Acad. Sci. U. S. A . 108( Suppl 1): 4516– 4522. Google Scholar CrossRef Search ADS PubMed  Costello E. K., Lauber C. L., Hamady M., Fierer N., Gordon J. I., Knight R.. 2009. Bacterial community variation in human body habitats across space and time. Science . 326: 1694– 1697. Google Scholar CrossRef Search ADS PubMed  Jeurissen S. H., Lewis F., van der Klis J. D., Mroz Z., Rebel J. M., ter Huurne A. A.. 2002. Parameters and techniques to determine intestinal health of poultry as constituted by immunity, integrity, and functionality. Curr. Issues Intest. Microbiol . 3: 1– 14. Google Scholar PubMed  Lamendella R., Domingo J. W., Ghosh S., Martinson J., Oerther D. B.. 2011. Comparative fecal metagenomics unveils unique functional capacity of the swine gut. BMC Microbiol . 11: 103. Google Scholar CrossRef Search ADS PubMed  Ley R. E., Hamady M., Lozupone C., Turnbaugh P. J., Ramey R. R., Bircher J. S., Schlegel M. L., Tucker T. A., Schrenzel M. D., Knight R., Gordon J. I.. 2008. Evolution of mammals and their gut microbes. Science . 320: 1647– 1651. Google Scholar CrossRef Search ADS PubMed  Ley R. E., Peterson D. A., Gordon J. I.. 2006. Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell . 124: 837– 848. Google Scholar CrossRef Search ADS PubMed  Lu J., Domingo J. S.. 2008. Turkey fecal microbial community structure and functional gene diversity revealed by 16S rRNA gene and metagenomic sequences. J. Microbiol . 46: 469– 477. Google Scholar CrossRef Search ADS PubMed  Magoc T., Salzberg S. L.. 2011. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics . 27: 2957– 2963. Google Scholar CrossRef Search ADS PubMed  Nicholson J. K., Holmes E., Wilson I. D.. 2005. Gut microorganisms, mammalian metabolism and personalized health care. Nat. Rev. Microbiol . 3: 431– 438. Google Scholar CrossRef Search ADS PubMed  Park S. J., Kim J., Lee J. S., Rhee S. K., Kim H.. 2014. Characterization of the fecal microbiome in different swine groups by high-throughput sequencing. Anaerobe . 28: 157– 162. Google Scholar CrossRef Search ADS PubMed  Peng S., Yin J., Liu X., Jia B., Chang Z., Lu H., Jiang N., Chen Q.. 2015. First insights into the microbial diversity in the omasum and reticulum of bovine using Illumina sequencing. J. Appl. Genet . 56: 393– 401. Google Scholar CrossRef Search ADS PubMed  Qin J., Li R., Raes J., Arumugam M., Burgdorf K. S., Manichanh C., Nielsen T., Pons N., Levenez F., Yamada T., Mende D. R., Li J., Xu J., Li S., Li D., Cao J., Wang B., Liang H., Zheng H., Xie Y., Tap J., Lepage P., Bertalan M., Batto J. M., Hansen T., Le Paslier D., Linneberg A., Nielsen H. B., Pelletier E., Renault P., Sicheritz-Ponten T., Turner K., Zhu H., Yu C., Li S., Jian M., Zhou Y., Li Y., Zhang X., Li S., Qin N., Yang H., Wang J., Brunak S., Dore J., Guarner F., Kristiansen K., Pedersen O., Parkhill J., Weissenbach J., Meta H. I. T. C., Bork P., Ehrlich S. D., Wang J.. 2010. A human gut microbial gene catalogue established by metagenomic sequencing. Nature . 464: 59– 65. Google Scholar CrossRef Search ADS PubMed  Rastall R. A. 2004. Bacteria in the gut: friends and foes and how to alter the balance. J. Nutr . 134: 2022S– 2026S. Google Scholar CrossRef Search ADS PubMed  Savage D. C. 1977. Microbial ecology of the gastrointestinal tract. Annu. Rev. Microbiol . 31: 107– 133. Google Scholar CrossRef Search ADS PubMed  Swanson K. S., Dowd S. E., Suchodolski J. S., Middelbos I. S., Vester B. M., Barry K. A., Nelson K. E., Torralba M., Henrissat B., Coutinho P. M., Cann I. K., White B. A., Fahey G. C., Jr. 2011. Phylogenetic and gene-centric metagenomics of the canine intestinal microbiome reveals similarities with humans and mice. ISME J . 5: 639– 649. Google Scholar CrossRef Search ADS PubMed  Toivanen P., Vaahtovuo J., Eerola E.. 2001. Influence of major histocompatibility complex on bacterial composition of fecal flora. Infect. Immun . 69: 2372– 2377. Google Scholar CrossRef Search ADS PubMed  Tun H. M., Brar M. S., Khin N., Jun L., Hui R. K., Dowd S. E., Leung F. C.. 2012. Gene-centric metagenomics analysis of feline intestinal microbiome using 454 junior pyrosequencing. J. Microbiol. Methods . 88: 369– 376. Google Scholar CrossRef Search ADS PubMed  Turnbaugh P. J., Hamady M., Yatsunenko T., Cantarel B. L., Duncan A., Ley R. E., Sogin M. L., Jones W. J., Roe B. A., Affourtit J. P., Egholm M., Henrissat B., Heath A. C., Knight R., Gordon J. I.. 2009. A core gut microbiome in obese and lean twins. Nature . 457: 480– 484. Google Scholar CrossRef Search ADS PubMed  Wu Y. 2013. MHC haplotype screening of HBK-SPF duck and bioinformatics analysis of TAP gene . M. D. thesis. Chinese Academy of Agriculture Sciences, Beijing, China (in Chinese). Zoetendal E. G., Booijink C. C., Klaassens E. S., Heilig H. G., Kleerebezem M., Smidt H., de Vos W. M.. 2006. Isolation of RNA from bacterial samples of the human gastrointestinal tract. Nat. Protoc . 1: 954– 959. Google Scholar CrossRef Search ADS PubMed  © 2017 Poultry Science Association Inc. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Poultry Science Oxford University Press

Application of high-throughput sequencing for microbial diversity detection in feces of specific-pathogen-free ducks

Loading next page...
 
/lp/ou_press/application-of-high-throughput-sequencing-for-microbial-diversity-0THkH9bbP0
Publisher
Oxford University Press
Copyright
© 2017 Poultry Science Association Inc.
ISSN
0032-5791
eISSN
1525-3171
D.O.I.
10.3382/ps/pex348
Publisher site
See Article on Publisher Site

Abstract

ABSTRACT High-throughput sequencing technologies play important roles in the study of animal enteric microorganisms. Fecal samples collected from 3 strains (9 samples) of specific-pathogen-free (SPF) ducks raised in an isolator. Duck intestinal bacterial flora were analyzed by sequencing distinct regions of the 16S rRNA genes using an Illumina HiSeq2500 platform. The result showed the most abundant primary enteric microbial phyla of 3 different strains of 70-week-old SPF ducks were Firmicutes, Proteobacteria, and Bacteroidetes, the most abundant primary enteric microbial order were Clostridiales, Lactobacillales, and Aeromonadales, and the most abundant primary enteric microbial genera were Bacteroides, Comamonas, and Enterococcus. In addition, the 3 duck strains harbored different compositions of the microorganisms, but these differences were not significant. Duck intestinal bacterial flora were analyzed using a high-throughput sequencing approach to further understand the distribution of intestinal flora and the biology of SPF ducks to ultimately benefit the purification of SPF ducks. INTRODUCTION Experimental poultry for epidemiological research are bred by artificial cultivation and carry specific microorganisms to establish control birds with clear genetic backgrounds. At present, specific-pathogen-free (SPF) poultry are used in poultry research and for production of biological products and experimental raw materials. SPF chickens have been widely used for poultry disease research and development of poultry vaccines, while little research is conducted using SPF ducks (Wu, 2013). Long periods of natural selection under specific environmental conditions in combination with feeding and rearing patterns are needed to produce various types of animals with characteristic intestinal ecosystems that contain diverse and complex microbial populations (Rastall, 2004; Ley et al., 2008; Costello et al., 2009). The mutual influence and interactions of the microbial community with the host constitute the complex and diverse gut ecosystem (gut microbiota) of the host (Savage, 1977). The composition of intestinal flora is closely related to host nutrition, metabolism, and immunity (Ley et al., 2006). The host provides the required conditions and nutrients for growth and reproduction of gut microbes (i.e., oxygen, temperature, and pH), while the intestinal bacteria degrade and ferment some carbohydrates that the host does not produce for production of energy to improve the energy utilization ratio (Nicholson et al., 2005). Gut microbes also play important roles in host metabolism, aid against pathogen attack, promote the development of immune organs, and activate the immune system. In recent years, with the development of the second generation sequencing technologies, the technology has been applied to the study of the structure and function of the gastrointestinal microbial community in humans (Turnbaugh et al., 2009; Qin et al., 2010), chickens (Lu and Domingo, 2008), dogs (Swanson et al., 2011), pigs (Lamendella et al., 2011), cats (Tun et al., 2012), and camels (Bhatt et al., 2013), among others. Soo-Je et al. (Park et al., 2014) employed a high-throughput sequencing technique to sequence the genomes of intestinal flora to identify differences in meat quality and weight of 2 groups of pigs, and found differences in the populations of Oscillibacter, Lactobacillus, Rothia, and Clostridium species between the 2 groups, which indicated that intestinal flora may be associated with fat accumulation in pork. Rothia and Clostridium species produce linoleic acid and short-chain fatty acids, which contribute to the health and growth of pigs. The predominate phylotypes in chicken fecal samples were Firmicutes, Bacteroidetes, and Proteobacteria by 16S rRNA sequencing technique (Lu and Domingo, 2008). A 454 pyrosequencing study of 6 dog fecal samples that employed random sequencing of microbial genomes found that Bacteroidetes/Chlorobi and Firmicutes accounted for about 35% of the analyzed sequences, followed by Proteobacteria (13%–15%) and Fusobacteria (7%–8%) (Swanson et al., 2011). 16S rRNA sequencing technique studies of pig feces indicated that the dominant microorganisms were Firmicutes and Bacteroidetes (Lamendella et al., 2011). A study of cat gut microbes conducted by Tun (1996) using the 454 pyrophosphate sequencing method determined the proportions of intestinal bacteria and identified a potential pathogen associated with antibiotic resistance genes (Tun et al., 2012). The Jin Ding duck, also known as the green head duck and South China duck, is a type of hemp duck that is a traditional variety of poultry in Fu Jian Province. In a previous study, Jin Ding ducks were hatched from progenitor Jin Ding duck eggs obtained from the JiangSu FengDa waterfowl breeding field of the National Waterfowl Germplasm Resource Gene Pool (Taizhou, China) and raised in an isolator from the age of 1 day to 70 weeks to produce ducks free from specific pathogens (i.e., highly pathogenic avian influenza virus, duck hepatitis virus, duck enteritis virus, avian lymphoid leukemia virus, avian reovirus, duck/goose parvovirus, duck circovirus, duck tembusu virus, and group III poultry adenoviruses). However, the bacterial distribution of SPF ducks remains unclear. In the present study, high-throughput sequencing technology was first employed to sequence SPF duck intestinal flora to further elucidate the distribution of intestinal flora and the biology of different strains SPF ducks to provide a reference for the purification of SPF ducks. MATERIALS AND METHODS Ethical Statement This study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and approved by the Harbin Veterinary Research Institute. Extraction of Microbial Genomic DNA Fecal samples were collected from 3 strains of SPF ducks (samples A1, A2, A3, B1, B2, B3, and C1, C2, C3) and stored at −80 °C for subsequent sequencing until assayed. Seventy-week-old SPF ducks were provided by the Center of Laboratory Animal of Harbin Veterinary Research Institute. The bacteria were pelleted by centrifugation and resuspended in phosphate-buffered saline solution. Total microbial genomic DNA from the samples was extracted using the cetyltrimethyl ammonium bromide/sodium dodecyl sulfate (CTAB/SDS) method. DNA concentration and purity was monitored on 1% agarose gels. For the following experiments, DNA was diluted to a concentration of 10 ng/μL in sterile water. Then quantification, qualification, mixing, and purification of the polymerase chain reaction (PCR) products. Amplicon Generation The V4 region of 16S rRNA genes was amplified using an Illumina HiSeq2500 platform (Novogene, Beijing, China) in accordance with the standard protocol of the manufacturer. All PCR reactions were carried out with Phusion® High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA). Library Preparation and Sequencing Sequencing libraries were generated using the TruSeq® DNA PCR-Free Sample Preparation Kit (Illumina, San Diego, CA), following manufacturer's recommendations, and index codes were added. The library quality was assessed using a Qubit® 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA) and an Agilent Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA). The library was sequenced on an Illumina HiSeq2500 platform and 250-bp paired-end reads were generated. Data Analysis Raw sequencing data contains joint information, low-mass bases, and unmeasured bases (N), which can cause significant interference to subsequent information analysis. Therefore, measures were implemented to rule out invalid data to ensure valid analysis of biological information. Through careful filtering methods to get rid of the interference information, the resulting data were validated as clean data or clean reads. Flash software (Magoc and Salzberg, 2011) was used to filter out reads of joint and barcode sequences, and to splice overlapping reads. QIIME software (Caporaso et al., 2011) was used to filter splicing data, low-quality bases, and chimeric sequences in spliced sequences. Clustering and species classification analysis was conducted using the operational taxonomic units (OTUs) of the clean data. According to the clustering result, each representative OTU sequence was annotated to a specific species. At the same time, the abundance of OTUs, alpha diversity calculation, and petal figure and Venn diagram analyses were conducted to identify species richness and evenness of information of specific or shared OTU information between different samples and groups. System trees were constructed to compare multiple sequence alignments of OTUs and to further identify differences in community structures among different samples and groups. The results are presented as dimension reduction images and sample clustering trees derived from principal component analysis, principal coordinates analysis, and nonmetric multidimensional scaling. Statistical Analysis MetaStat software (MetaStat, Inc., Boston, MA) was used to analyze the 10 most abundant taxonomic sequence tags of the 3 sample groups. A probability (P) value of < 0.05 was considered statistically significant. RESULTS Sequencing and Quality Control Sequencing of the DNA sequences extracted from the feces of SPF ducks to obtain raw sequence (PE reads). The length of the effective sequence tags varied between 420 and 427 bp. Joint information, low mass bases, unmeasured bases (N), and chimeras were removed from the raw sequencing data to obtain clean sequences. A total number of 36,513, 32,533, and 37,193 clean tags were obtained for samples A, B, and C, respectively. Sequencing data quality was mainly distributed above Q20, so as to ensure normality of the subsequent advanced analysis. Redundancy of the clean sequence tags was determined using mothur software (http://www.mothur.org/), which identified 31,296, 29,185, and 30,785 unique sequence tags for strains A, B, and C, respectively (Table 1). Each unique sequence tag was compared to the sequences of the 16S rRNA genes of distinct regions using the Basic Local Alignment Search Tool for Nucleotides (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch). Table 1. Data preprocessing statistics and quality control. Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  1Raw PE means the original PE reads off the computer. 2Raw Tags refers to splice sequence Tags. 3Clean Tags is pointing to the tags that have filtered the low-quality and short length sequence. 4After the Effective Tags refers to sequence tags have filtered chimeras, and eventually for subsequent analysis. 5Base is refers to the final Effective Base number of the data. 6The AvgLen refers to the average length of the Effective Tags. 7Q20 and Q30 refers to the Effective quality value is greater than 20 bases in Tags (sequencing error rate is less than 1%), and 30 (sequencing error rate is less than 0.1%) percentage of bases. View Large Table 1. Data preprocessing statistics and quality control. Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  Sample Name  Raw PE1  Raw Tags2  Clean Tags3  Effective Tags4  Base5  AvgLen6  Q207  Q307  GC  Effective    (#)  (#)  (#)  w(#)  (nt)  (nt)      (%)  (%)  A  61,260  48,798  36,513  31,296  13,065,674  418  97.11  94.23  51.90  51.00  B  57,221  45,590  32,533  29,185  12,324,631  423  96.94  93.89  52.12  50.79  C  62,146  49,888  37,193  30,785  12,865,454  419  97.07  94.17  52.19  49.16  1Raw PE means the original PE reads off the computer. 2Raw Tags refers to splice sequence Tags. 3Clean Tags is pointing to the tags that have filtered the low-quality and short length sequence. 4After the Effective Tags refers to sequence tags have filtered chimeras, and eventually for subsequent analysis. 5Base is refers to the final Effective Base number of the data. 6The AvgLen refers to the average length of the Effective Tags. 7Q20 and Q30 refers to the Effective quality value is greater than 20 bases in Tags (sequencing error rate is less than 1%), and 30 (sequencing error rate is less than 0.1%) percentage of bases. View Large Data Indices In order to determine the diversity of species among the samples, clusters of the effective tags of all samples, with 97% identities, were used for clustering the OTU sequences. All effective tags of the samples shared sequence identities of > 97% to those of known species, which almost accounted for 100% of the OTUs of samples A and C, suggesting that no new genera or species were detected in samples A and C, while 3 sequence tags were unclassified. Therefore, further analysis is needed to identify microorganisms inhabiting the duck gastrointestinal tract (Figure 1). Figure 1. View largeDownload slide Statistics of each sample OTUs clustering and annotation. Total Tags (red): refers to the number of total tags each sample; Unique Tags (orange): refers to the number of tags that can’t be clustering to OTUs (sequence cannot be clustered OTUs will not be used for subsequent analysis); Taxon Tags (blue): refers to the number of Tags for building OTUs and annotation information; Unclassified Tags (tea green): refers to the number of Tags without annotation information; OTUs (purple): refers to the Number of OTUs in each sample. Figure 1. View largeDownload slide Statistics of each sample OTUs clustering and annotation. Total Tags (red): refers to the number of total tags each sample; Unique Tags (orange): refers to the number of tags that can’t be clustering to OTUs (sequence cannot be clustered OTUs will not be used for subsequent analysis); Taxon Tags (blue): refers to the number of Tags for building OTUs and annotation information; Unclassified Tags (tea green): refers to the number of Tags without annotation information; OTUs (purple): refers to the Number of OTUs in each sample. Further analysis of the sequence tags of samples A, B, and C (770, 587, and 769 OTUs, respectively) showed that 499 OTUs were detected in all samples. Samples A and B shared 540 OTUs, samples A and C shared 664 OTUs, and samples B and C shared 529 OTUs. In addition, 65 OTUs were unique to sample A, 17 OTUs were unique to sample B, and 75 OTUs were unique to sample C (Figure 2). Figure 2. View largeDownload slide Venn Graph. Venn Graph was drawn after homogenization processing for all samples. Each circle in the figure represents a sample, the number in circle and circle overlap between representative samples. Number of OTUs were no overlap number represents the number of unique OTUs in the sample. Figure 2. View largeDownload slide Venn Graph. Venn Graph was drawn after homogenization processing for all samples. Each circle in the figure represents a sample, the number in circle and circle overlap between representative samples. Number of OTUs were no overlap number represents the number of unique OTUs in the sample. The majority (>98%) of sequences were classified at below the domain level and more than 60% of the tags were assigned to a genus. However, only select sequence tags could be classified at the species level (Figure 3). Figure 3. View largeDownload slide The number of sequence at the different taxonomy levels obtained from 3 samples. Each bar represents the number of tags that were assigned to the taxonomic level for the identified bacteria. Figure 3. View largeDownload slide The number of sequence at the different taxonomy levels obtained from 3 samples. Each bar represents the number of tags that were assigned to the taxonomic level for the identified bacteria. Species rarefaction curves of the observed species were firstly very steep and then all curves gradually tended to be flat at the end, which illustrated that the existing sequencing data volume was reasonable (Figure 4a).The rank-abundance curve directly showed relative richness and evenness of microbial species in all 9 samples. Due to the complex intestinal flora in feces, the slopes of curves were small, which indicated a high degree of evenness among the microbial (Figure 4b). Figure 4. View largeDownload slide (a) Species rarefaction curves of 3 strains. (b) Rank-abundance curves of 3 strains. Figure 4. View largeDownload slide (a) Species rarefaction curves of 3 strains. (b) Rank-abundance curves of 3 strains. Alpha diversity was applied to clarify the richness and diversity of microbes in the samples. However, according to the alpha diversity metrics, sample B was also characterized by a lower Shannon entropy and lower ACE index, indicating that the microbial community in sample B was less diverse than in samples A and C. Additionally, calculation of the beta diversity indicated comparative diversities of microbial communities in the different samples. Microbial Community Analysis in the Three Strains Taxonomic information for each representative sequence was determined using the GreenGene database (http://greengenes.lbl.gov/cgi-bin/nph-index.cgi) with Ribosomal Database Project classifiers. Based on the average relative abundance analysis, at the phylum level, the majority of sequences in the 3 groups samples were assigned to 10 phyla: Firmicutes, Bacteroidetes, Proteobacteria, Actinobacteria, Fusobacteria, Synergistetes, Verrucomicrobia, Tenericutes, Cyanobacteria, and Elusimicrobia. The top 3 phyla of strain A were Firmicutes (68.23%), Bacteroidetes (18.29%), and Proteobacteria (10.99%). The top 3 phyla of strain B were Firmicutes (56.39%), Proteobacteria (31.72%), and Bacteroidetes (10.17%), while the top 3 phyla of strain C were Firmicutes (74.04%), Proteobacteria (12.79%), and Bacteroidetes (9%). Although there were few Synergistetes and Elusimicrobia species in strains A and C, none of the sequence tags in group B were assigned to these 2 phyla (Figure 5a). No significant differences (P > 0.05) were observed among the 3 groups of samples in any phylum level. Figure 5. View largeDownload slide (a) Species relative abundance of the phylum level. “Others” means the sum of the relative abundance of all the other phylum. (b) Species relative abundance of the order level. “Others” means the sum of the relative abundance of all the other order. (c) Species relative abundance of the genus level. “Others” means the sum of the relative abundance of all the other genus. Figure 5. View largeDownload slide (a) Species relative abundance of the phylum level. “Others” means the sum of the relative abundance of all the other phylum. (b) Species relative abundance of the order level. “Others” means the sum of the relative abundance of all the other order. (c) Species relative abundance of the genus level. “Others” means the sum of the relative abundance of all the other genus. Based on average relative abundance analysis, at the order level, the majority of sequences in the 3 group samples were assigned to 10 orders: Clostridiales, Lactobacillales, Aeromonadales, Bacteroidales, Burkholderiales, Bacillales, Actinomycetales, Erysipelotrichales, Xanthomonadales, Pseudomonadales. All of the top 10 orders existed in each of the 3 groups. The top 3 orders of group A were Clostridiales (37.36%), Lactobacillales (22.57%), and Bacteroidales (18.02%). The top 3 orders of group B were Lactobacillales (29.19%), Clostridiales (20.27%), and Aeromonadales (17.01%), while the top 3 orders of strain C were Clostridiales (34.46%), Lactobacillales (29.62%), and Bacteroidales (8.15%) (Figure 5b). No significant difference (P > 0.05) was observed among the 3 groups of samples in any order level. Based on average relative abundance analysis, at the genus level, the majority of sequences in the 3 groups of samples were assigned to 10 genera: Bacteroides, Comamonas, Enterococcus, Roseburia, Faecalibacterium, Vagococcus, Streptococcus, Aerococcus, Sporosarcina, and Blautia. All the top 10 genera existed in 3 groups. The top 3 genera of group A were Bacteroides (15.08%), Enterococcus (9.31%), and Faecalibacterium (3.54%). The top 3 genera of strain B were Enterococcus (12.48%), Comamonas (8.0%), and Bacteroides (7.38%), while the top 3 genera of strain C were Enterococcus (10.24%), Bacteroides (5.89%), and Faecalibacterium (5.16%) (Figure 5c). No significant difference (P > 0.05) was observed among the 3 groups of samples in any genus level. The species annotation results showed that the greatest abundance of species were of the phylum Firmicutes (70.51%). Approximately 70.58% of the sequence tags of this phylum were assigned to the class Bacilli and the remaining sequence tags were assigned to the class Clostridia. Microorganisms annotated to the class Bacilli were most abundant in sample C and least abundant in strain A. Most (90.4%) of the sequence tags of this class were assigned to the order Lactobacillales, and the remaining were assigned to the order Bacillales. Enterococcaceae was the most dominate family in strains B and C, accounting for approximately 18.6% and 13.79% of the sequence tags, respectively. Meanwhile, Planococcaceae was the most dominate family in strain A, accounting for approximately 31.46% of all sequence tags. The second-highest abundance of sequence tags in strain A were categorized to the phylum Bacteroidetes, which was ranked third in strains B and C. All sequence tags of all samples were assigned to class Bacteroidia, order Bacteroidales, family Bacteroidaceae, genus Bacteroides, and included the species Bamesiae, Coprophilus, Coprosuls, Fragilis, Ovatus, and Plebeius. Proteobacteria was the second most-prevalent phylum in strains B and C, and the third most-prevalent in strain A. All sequence tags were assigned to class Betaproteobacteria, order Burkholderia, family Comamonadaceae, genus Comamonas, It is worth mentioning that the abundance of phylum Proteobacteria (31.72%) in strain B was much higher than in the other 2 strains (Figure 6). Figure 6. View largeDownload slide Species classification tree in the samples. Different colors fan represented different samples in circle. The size of Fan refers to the microorganism classification of relative abundance in the sample. Classification of the Numbers below all samples means the relative abundance of the average percentage on this classification, there are 2 Numbers, the former said the percentage of all species, the latter said the percentage of selected species. Figure 6. View largeDownload slide Species classification tree in the samples. Different colors fan represented different samples in circle. The size of Fan refers to the microorganism classification of relative abundance in the sample. Classification of the Numbers below all samples means the relative abundance of the average percentage on this classification, there are 2 Numbers, the former said the percentage of all species, the latter said the percentage of selected species. Beta Diversity Index Analysis Beta Diversity Index Analysis was measured to estimate the divergence of microbial species among 3 groups, which based on the weighted UniFrac and Unweighted Unifrac distance cluster analysis, a dissimilarity coefficient for samples (Figure 7). The lower dissimilarity coefficients indicate the less divergence of microbial species. This study, the dissimilarity coefficients were measured to be 0.279 (0.346) bacterial diversities between strains A and B .The dissimilarity coefficients were measured to be 0.189 (0.238) bacterial diversities between strains A and C. The dissimilarity coefficients were measured to be 0.194 (0.314) bacterial diversities between group B and C. The result showed more divergence of microbial species between A and B than between A and C, between B and C. Figure 7. View largeDownload slide Beta diversity (weighted UniFrac) among 3 strains. Beta diversity indexes were measured based on weighted UniFrac and unweighted UniFrac distances. The upper numbers in the grid represent the weighted UniFrac; and the lower numbers in the grid represent the unweighted UniFrac distances, respectively. Figure 7. View largeDownload slide Beta diversity (weighted UniFrac) among 3 strains. Beta diversity indexes were measured based on weighted UniFrac and unweighted UniFrac distances. The upper numbers in the grid represent the weighted UniFrac; and the lower numbers in the grid represent the unweighted UniFrac distances, respectively. DISCUSSION Illumina sequencing has provided system development description of gastrointestinal microbial species. Compared with other molecular biology techniques, data generated by high-throughput sequencing technologies is more direct and comprehensive (Qin et al., 2010). Enteric microbes are closely related to animal feeding, energy metabolism, and health (Ley et al., 2008). Elucidation of the composition of enteric microbes will not only contribute to the understanding of gastrointestinal physiology, but also is conducive to the long-term development of animal husbandry. In the present study, the Illumina genome sequencing platform was used for confirmation of distinct regions of the 16S rRNA genes sequence, and determination of the enteric microbial species of different strains of SPF ducks. To the best of our knowledge, this is the first study to use a high-throughput technique to study enteric microbial species of different strains of SPF ducks. Previous high-throughput sequencing studies have shown that Bacteroidetes, Proteobacteria, Actinobacteria, and Firmicutes were the dominant species in the gastrointestinal tract of humans, loris, and other animals, although the relative proportion of species differed. According to the sequence results obtained in this study, the most common phyla were Firmicutes, Bacteroidetes, and Proteobacteria, which is similar to the gut microbes of many other species. Long-term intake of fat can increase the proportion of Firmicutes to Bacteroidetes, while the intake of dietary fiber may lead to a decrease in the ratio of Firmicutes to Bacteroidetes species (Peng et al., 2015). The proportion of Firmicutes bacteria was higher than in previous reports and this change can affect the metabolism and function of gut microbes, thereby altering the ability of the host to derive energy from food. In addition, Firmicutes may play important roles in the gastrointestinal health of ducks. The gastrointestinal microbial system of poultry is an organic body made up of various bacteria species that can be divided into harmful microbes and mutualistic microorganisms (Jeurissen et al., 2002). The results of the present showed the absence of other reported pathogenic microbes, such as Ornithobacterium rhinotracheale, Salmonella sp., Riemerella anatipestifer, Pasteurella multocida, and Staphylococcus aureus, while Enterococcus and Streptococcus species were found in all 3 samples. However, further research is needed to determine whether Enterococcus and Streptococcus are pathogenic to ducks. The animal gut is a dynamic ecosystem composed of a diversity of microbes that is influenced by the host itself as well as the external environment. At present, it is generally believed that the diet, age, and genotype of the host are important factors that affect the diversity of the intestinal microflora (Toivanen et al., 2001; Zoetendal et al., 2006). In this study, the main factor affecting the diversity of the intestinal microflora was thought to be the host genotype, although there was no significant difference observed in the proportion of 3 different strains, implying that the duck genotype may not necessarily influence the composition of gut microbes. In conclusion, the microbiome composition and diversity of the 3 strains of SPF ducks were investigated to compare the similarities and differences between duck strains. This study is the first to use high-throughput technology to investigate the enteric microbial diversity in ducks. Sequencing of the duck intestinal flora by high-throughput sequencing should provide useful information to further understand the distribution of intestinal flora and the biology of SPF ducks. The results of this study also help to advance our understanding of host and microbe symbiosis to facilitate improvements in animal health and production. ACKNOWLEDGMENTS This study was supported by the National Key Research and Development Program of China (2016YFD0500800) and the Chinese Academy of Agricultural Sciences Fundamental Scientific Research Funds (Y2016PT41). REFERENCES Bhatt V. D., Dande S. S., Patil N. V., Joshi C. G.. 2013. Molecular analysis of the bacterial microbiome in the fore stomach fluid from the dromedary camel (Camelus dromedarius). Mol. Biol. Rep . 40: 3363– 3371. Google Scholar CrossRef Search ADS PubMed  Caporaso J. G., Lauber C. L., Walters W. A., Berg-Lyons D., Lozupone C. A., Turnbaugh P. J., Fierer N., Knight R.. 2011. Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample. Proc. Natl. Acad. Sci. U. S. A . 108( Suppl 1): 4516– 4522. Google Scholar CrossRef Search ADS PubMed  Costello E. K., Lauber C. L., Hamady M., Fierer N., Gordon J. I., Knight R.. 2009. Bacterial community variation in human body habitats across space and time. Science . 326: 1694– 1697. Google Scholar CrossRef Search ADS PubMed  Jeurissen S. H., Lewis F., van der Klis J. D., Mroz Z., Rebel J. M., ter Huurne A. A.. 2002. Parameters and techniques to determine intestinal health of poultry as constituted by immunity, integrity, and functionality. Curr. Issues Intest. Microbiol . 3: 1– 14. Google Scholar PubMed  Lamendella R., Domingo J. W., Ghosh S., Martinson J., Oerther D. B.. 2011. Comparative fecal metagenomics unveils unique functional capacity of the swine gut. BMC Microbiol . 11: 103. Google Scholar CrossRef Search ADS PubMed  Ley R. E., Hamady M., Lozupone C., Turnbaugh P. J., Ramey R. R., Bircher J. S., Schlegel M. L., Tucker T. A., Schrenzel M. D., Knight R., Gordon J. I.. 2008. Evolution of mammals and their gut microbes. Science . 320: 1647– 1651. Google Scholar CrossRef Search ADS PubMed  Ley R. E., Peterson D. A., Gordon J. I.. 2006. Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell . 124: 837– 848. Google Scholar CrossRef Search ADS PubMed  Lu J., Domingo J. S.. 2008. Turkey fecal microbial community structure and functional gene diversity revealed by 16S rRNA gene and metagenomic sequences. J. Microbiol . 46: 469– 477. Google Scholar CrossRef Search ADS PubMed  Magoc T., Salzberg S. L.. 2011. FLASH: fast length adjustment of short reads to improve genome assemblies. Bioinformatics . 27: 2957– 2963. Google Scholar CrossRef Search ADS PubMed  Nicholson J. K., Holmes E., Wilson I. D.. 2005. Gut microorganisms, mammalian metabolism and personalized health care. Nat. Rev. Microbiol . 3: 431– 438. Google Scholar CrossRef Search ADS PubMed  Park S. J., Kim J., Lee J. S., Rhee S. K., Kim H.. 2014. Characterization of the fecal microbiome in different swine groups by high-throughput sequencing. Anaerobe . 28: 157– 162. Google Scholar CrossRef Search ADS PubMed  Peng S., Yin J., Liu X., Jia B., Chang Z., Lu H., Jiang N., Chen Q.. 2015. First insights into the microbial diversity in the omasum and reticulum of bovine using Illumina sequencing. J. Appl. Genet . 56: 393– 401. Google Scholar CrossRef Search ADS PubMed  Qin J., Li R., Raes J., Arumugam M., Burgdorf K. S., Manichanh C., Nielsen T., Pons N., Levenez F., Yamada T., Mende D. R., Li J., Xu J., Li S., Li D., Cao J., Wang B., Liang H., Zheng H., Xie Y., Tap J., Lepage P., Bertalan M., Batto J. M., Hansen T., Le Paslier D., Linneberg A., Nielsen H. B., Pelletier E., Renault P., Sicheritz-Ponten T., Turner K., Zhu H., Yu C., Li S., Jian M., Zhou Y., Li Y., Zhang X., Li S., Qin N., Yang H., Wang J., Brunak S., Dore J., Guarner F., Kristiansen K., Pedersen O., Parkhill J., Weissenbach J., Meta H. I. T. C., Bork P., Ehrlich S. D., Wang J.. 2010. A human gut microbial gene catalogue established by metagenomic sequencing. Nature . 464: 59– 65. Google Scholar CrossRef Search ADS PubMed  Rastall R. A. 2004. Bacteria in the gut: friends and foes and how to alter the balance. J. Nutr . 134: 2022S– 2026S. Google Scholar CrossRef Search ADS PubMed  Savage D. C. 1977. Microbial ecology of the gastrointestinal tract. Annu. Rev. Microbiol . 31: 107– 133. Google Scholar CrossRef Search ADS PubMed  Swanson K. S., Dowd S. E., Suchodolski J. S., Middelbos I. S., Vester B. M., Barry K. A., Nelson K. E., Torralba M., Henrissat B., Coutinho P. M., Cann I. K., White B. A., Fahey G. C., Jr. 2011. Phylogenetic and gene-centric metagenomics of the canine intestinal microbiome reveals similarities with humans and mice. ISME J . 5: 639– 649. Google Scholar CrossRef Search ADS PubMed  Toivanen P., Vaahtovuo J., Eerola E.. 2001. Influence of major histocompatibility complex on bacterial composition of fecal flora. Infect. Immun . 69: 2372– 2377. Google Scholar CrossRef Search ADS PubMed  Tun H. M., Brar M. S., Khin N., Jun L., Hui R. K., Dowd S. E., Leung F. C.. 2012. Gene-centric metagenomics analysis of feline intestinal microbiome using 454 junior pyrosequencing. J. Microbiol. Methods . 88: 369– 376. Google Scholar CrossRef Search ADS PubMed  Turnbaugh P. J., Hamady M., Yatsunenko T., Cantarel B. L., Duncan A., Ley R. E., Sogin M. L., Jones W. J., Roe B. A., Affourtit J. P., Egholm M., Henrissat B., Heath A. C., Knight R., Gordon J. I.. 2009. A core gut microbiome in obese and lean twins. Nature . 457: 480– 484. Google Scholar CrossRef Search ADS PubMed  Wu Y. 2013. MHC haplotype screening of HBK-SPF duck and bioinformatics analysis of TAP gene . M. D. thesis. Chinese Academy of Agriculture Sciences, Beijing, China (in Chinese). Zoetendal E. G., Booijink C. C., Klaassens E. S., Heilig H. G., Kleerebezem M., Smidt H., de Vos W. M.. 2006. Isolation of RNA from bacterial samples of the human gastrointestinal tract. Nat. Protoc . 1: 954– 959. Google Scholar CrossRef Search ADS PubMed  © 2017 Poultry Science Association Inc.

Journal

Poultry ScienceOxford University Press

Published: Apr 20, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off