Exploring the microbial community (microflora) associated with ovine Haemonchus contortus (macroflora) field strains

High-throughput sequencing technology has shown tremendous promise for microbial community composition and diversity. Illumina MiSeq platform was exploited to study the microbial community associated with the different stages of the life-cycle of ovine Haemonchus contortus field strains using two distinct amplification primer sets (targeting V3–V4, and V5–V7). Scanning electron microscope and polymerase chain reaction coupled with Illumina MiSeq platform were employed to confirm the absence of any parasite surface contamination by intact bacteria or their DNA products. Results showed 48 (V3–V4 tags) and 28 (V5–V7 tags) bacterial genera comprised the microbial flora of H. contortus microbiome. The dominant bacterial genera belonged to Escherichia-Shigella, Pseudomonas and Ochrobactrum, which were shared in all the stages of the parasite life-cycle using V3–V4 and V5–V7 amplicons. Moreover, the parasite microbiome could reflect the external micro-organisms (i.e. micro- and macro-habitats). There is abundant room for further progress in comparing microbiome of different helminths, which has, and will continue to offer considerable potential for better understanding a wide-variety of devastating animal and human diseases.

the bacterial communities in a sample without the bias or cultivation effort. Currently, this is the most reliable approach to explore the diversity of microbial community and to determine its composition. We employed 16S rRNA gene amplification and Illumina MiSeq platform to portray, for the first time to our knowledge, the differences in composition and relative abundance of diverse H. contortous microbiome. This study provides new insights into the microbial community naturally associated with field strains of H. contortus life-cycle stages, with the long-term novel approach of manipulating them to control ruminant gastrointestinal parasites.

Materials and Methods
Ethics approval and consent to participate. Animal experiments were conducted in accordance with the guidelines of Beijing the Municipality on the Review of Welfare and Ethics of Laboratory Animals approved by the Beijing Municipality Administration Office of Laboratory Animals (BAOLA), and under the protocol (CAU-AEC-2010-0603) approved by the China Agricultural University Animal Ethics Committee. All experimental procedures were also approved by the Institutional Animal Care and Committee of China Agricultural University (The certificate of Beijing Laboratory Animal employee, ID: 15883).

Animals and parasites. Field microscopic investigation on H. contortus in 107 sheep in an animal farm in
Jin Zhan, Cun, Chaoyang, Beijing, China to determine whether any general characteristics of the sheep were associated with infections were performed to select H. contortus-infected animals for the current study. Parasitological examination of rectal fecal samples from the sheep was analyzed with the modified McMaster technique 13 . In addition, FAMACHA © scores were performed for detecting anemic sheep on the farm by matching the color of the ocular conjunctival mucous membranes of each animal to the FAMACHA © eye color chart, with the scale ranging from 1 to 5, where a score of 1 is valued not anemic and 5 is valued highly anemic 14 . Body condition scoring (BCS) was assessed by observing and palpating the loin area (behind the ribs and in front of the pelvis), and assigning a score of 1-10 according to the procedure used by 15 with 1 being emaciated and 10 being extremely fat. Blood was collected for haemato-biochemical to reconfirm the presence of H. contortous infection. The aforementioned parasitological tests, FAMACHA © scores and body condition scoring and haemato-biochemical analyses were reassessed after two weeks to ensure the consistency of the results.
A total of three out of fourteen H. contortus-infected sheep; two-month-old sheep (Ovis aries) that were randomly selected and brought from the local farmer in Jin Zhan village, Chaoyang, Beijing, China. Upon arrival in China Agricultural University facilities, all sheep were tagged (each sheep was assigned an arbitrary number for sample identification purposes). Feces from the infected sheep were collected and cultured at room temperature for 7 days to allow the third stage larvae (the infectious stage) to develop from the eggs, and larvae were harvested using the Baermann technique 16 . The remaining feces together with the abomasal contents were collected from H. contortus-infected sheep immediately after death, and eggs were extracted employing sucrose step-density gradient centrifugation (unpublished data). The collected eggs were washed on a 35 µm sieve with distilled water to remove the sugar under sterile conditions. The mesh sieve is present at the top of ceramic funnel and the flask with one side-arm attached to rubber tube connected with another double-side arm flask. The latter is connected with vacuum pump, which has an advantage over the ordinary gravitational force. The adult worms were recovered from the abomasal mucosa using forceps and dissecting microscope. Before the DNA was extracted H. contortus adult worm, exsheathed L 3 and eggs were washed with phosphate-buffered saline (PBS, pH 7.4) and incubated in an antibiotic solution (1 mg/ml ampicillin/1 mg/ml gentamicin) for 2 h to kill external bacteria. Then, the parasites were washed twice with 2% sodium hypochlorite for 20 s each followed by five times with sterile PBS.
Ruling out bacterial contamination associated with the parasite surfaces. Several tests, such as scanning electron microscope (SEM) and polymerase chain reaction (PCR) were conducted to rule out any abomasal microbial contamination associated with the surface of the distinct life-cycle stages of H. contortus.
Parasites were prepared for SEM as previously reported by 17 and Hitachi S-3400N SEM was employed to assure the absence of any surface bacteria on the selected life-cycle stages of H. contortus (Fig. 1). So in addition to SEM, PCR further confirmed the absence of bacterial contamination from the abomasal microbial community. The approach to distinguishing between unwashed and washed individual female worms, surface washes, which were washed following the procedure above, involved PCR and next-generation sequencing. Parasites were stored at −20 °C for downstream applications. The procedures launched are described in detail in the Materials and Methods section below. DNA extraction, amplification and processing of samples for high-throughput sequencing. Microbial DNA extraction. H. contortus genomic DNA together with microbial genomic DNA from samples was extracted using the E.Z.N.A. ® Bacterial DNA Kit (Omega Bio-tek, Norcross, GA, US) according to manufacturer's instruction. DNA products were assessed by electrophoresis in 1% agarose gel run at 100 V for 30 min, and the sizes of the products (23 kb) were validated by comparison with a molecular size standard. The quantity and quality of the DNA were evaluated by measuring the absorbance employing an ultraviolet spectrophotometer (DU1 640, Beckman Instruments, Inc., CA, USA) at 260 and 280 nm. All microbial genomic DNA was stored at −20 °C before further analysis.
Amplicon generation and high-throughput sequencing. The 16S ribosomal RNA (rRNA) gene contains of nine hypervariable regions enclosed by regions of more conserved sequence. As shown in Fig. 2, regions of approximately 468 bp and 394 bp encircling the V3-V4 and V5-V7 hypervariable regions respectively within the 16S rRNA gene was subjected to high-throughput sequencing to maximize the effective length of sequencing reads 18 .
The V3-V4 and V5-V7 regions of the bacterial 16S rRNA gene sequences were amplified employing a set of commonly used primers for the analysis of H. contortus-associated bacterial communities using Illumina MiSeq platform (Table 1)   Statistical and bioinformatics analysis. After trimming the adaptor and primer sequences from Illumina reads, the raw sequences were assembled for each sample according to the unique barcode using Qiime (V1.7.0, http:// qiime.org/scripts/split_libraries_fastq.html). Paired-end reads from the original DNA fragments were merged using FLASH (V1.2.7, http://ccb.jhu.edu/ software/FLASH/).
Quality filtering was performed under specific filtering conditions to obtain the high-quality clean tags according to the pipeline tool quantitative insights into microbial ecology Qiime (V1.7.0, http://qiime.org/scripts/ split_libraries_fastq.html).
The clean tags were compared against the reference database (Gold database, http://drive5.com/uchime/ uchime_download.html) employing UCHIME algorithm (UCHIME Algorithm, http://www.drive5.com/ usearch/manual/uchime_algo.html) to detect and remove chimeric sequences. Paired-end reads were assigned to each sample according to the unique barcodes. Sequence analysis was performed by UPARSE software package (Uparse v7.0.1001, http://drive5.com/uparse/) using the UPARSE-OTU and UPARSE-OTU ref algorithms. Alpha (within samples) and beta (among samples) diversities were analyzed employing in-house Perl scripts. Sequences with ≥97% identities were assigned to the same OTUs. Representative sequences for each OTU were picked up to be used in the ribosomal database project (RDP) classifier to annotate the taxonomic information for each representative sequence. In order to estimate alpha diversity, we rarified the operational taxonomic units (OTUs) and calculated five metrics: Phylogenetic diversity (PD) Whole tree, Good's coverage, Chao1, Shannon and Observed species. Rarefaction curves were created based on this five metrics.
The Kruskal-Wallis test 19 was used to compare diversity metrics, such as genus evenness, Simpson and Shannon-Wiener species diversity indices. The level of significance was determined at P < 0.05.
The Principal Coordinate Analysis (PCoA) was conducted to explore the differences in the bacterial community structures and was displayed with the WGCNA, stat and ggplot2 packages in the R software (Version 2.15.3). Weighted and un-weighted unifrac distances were calculated employing QIIME 20 . Graphical representation of the relative abundance of bacterial diversity from phylum to genus was visualized using a bar plot with horizontal bars.

Results
Parasitological tests, FAMACHA © scores and body condition scoring. For coprological examinations, the rectal fecal samples were collected for fecal egg count examination employing the modified McMaster technique. H. contortus were found in 14 (13.08%) of the 107 sheep studied. Additionally, FAMACHA © scores and body condition scoring for all 14 infected animals were assessed. Our results showed that FAMACHA © scores of all 14 infected animals fell between 2 and 3 out of 5 and no sheep decreased below 6 or increased above 8 of all 14 infected sheep for body condition scores. Blood samples were collected for haemato-biochemical analyses to confirm the presence of H. contortus infection. Interestingly, these results further support the coprologic examinations (unpublished data). The aforementioned parasitological tests, FAMACHA © scores, body condition scoring and haemato-biochemical analyses were reassessed after two weeks; the results are consistent with the previously obtained data.

Morphological forms of the ovine H. contortus life-cycle stages. The life-cycle of H. contortus
consists of six stages: the egg, four larval stages (L 1 , L 2 , L 3 , L 4 ) and the adult. Each stage in the life cycle is separated by a molt except for the egg -L 1 . The embryo develops to a 1 st -stage larva within the egg, hatches and undergoes four molts before developing into the sexually mature adult. H. contortus passes through diverse environments during completion of its life-cycle. The adults and 4 th -stage larvae are parasitic and adapted to the environment within the host, while the other stages are free-living (Fig. 3). Three distinct stages of the life-cycle of H. contortus are selected for the characterization of their microbiota, which are: eggs, L 3 larval stage and adults (Fig. 4). Investigation of parasite surfaces for abomasal microbial contamination. Combining the evidence from the two distinct approaches, including SEM and next-generation sequencing of 16S rRNA gene (V3-V4 region), the obvious result is there is no contamination (i.e. intact bacteria and their DNA products) originating from the abomasal microbial community and consequently; the results of the Haemonchus microbiome are sufficient to remove an important obstacle about the abomasal bacterial contamination ( Figure S1 and Table S1).

Larval-stage of H. contortus displaying internalized transgenic Escherichia coli.
The expression cassette plasmid, H4-DHFR-EYFP-ACT (Fig. 5), was constructed from the recombinant plasmid pd2EYFP-N1+ formerly described by 21 . The mic1-EYFP and RFP were replaced with H4-DHFR. Therefore, the tandem dihydrofolate reductase thymidylate synthase-enhanced yellow fluorescent protein (DHFR-EYFP) coding region is flanked by the Histone 4 (H4) promoter from E. tenella and 3′region of actin. With respect  to plasmid transformation, the procedure was conducted according to the manufacturer's recommendations using BL21 (DE3) chemically competent cell (Beijing TransGen Biotech Co., Ltd.). A fluorescent microscopy image (Olympus, Tokyo, Japan) using the filter set at 488-nm excitation, 508-nm emission 22 of the free-living larval-stage that incubated for 72 hours with yellow fluorescent protein (YFP)-expressing E. coli is presented in Fig. 6.
Diversity index analysis of microbial community. Samples were evaluated for alpha diversity (microbial diversity within samples) and beta diversity (community diversity between samples) analysis.
Alpha rarefaction curves. It can be seen from the data in Table 2 Beta diversity. Computing differences between microbial communities were calculated exploiting the default beta diversity metrics of weighted and unweighted UniFrac 23 (beta_diversity_through_plots.py). Principal Coordinate Analysis (PCoA) was plotted using the result of UniFrac distance matrices to determine the similarity between groups of samples/time-points. The three-dimensional PCoA plots were visualized employing the Emperor tool 24 . The beta diversity analysis showed three very distinct clusters separating the different life-cycle stages of H. Contortus (Fig. 9).
The microbial community composition of the ovine abomasum and rumen in the context of H. contortus infection at the genus level. Two-month-old sheep (Ovis aries) that were raised under confined living conditions to avoid worm exposure were moved from a local farmer in Jin Zhan village, Chaoyang, Beijing, China with an average liveweight of 25.02 ± 2.2. Twelve parasite-naive sheep were divided into four treatments, uninfected-control groups and 7-and 50-day post-infection (dpi) Haemonchus-infected groups, in triplicate. 6 sheep were inoculated with 5,000 of H. contortus infective larvae laboratory strain and followed for 7 and 50 days. The animals were fed a fixed-formula diet containing forage and concentrate in a ratio of 1 to 1. The abomasal and ruminal microbial communities (V3-V4 amplicon) of the early (7-day post-infection) and late (50-day post-infection) H. contortus -infected sheep at the genus level are presented in Tables S4 and S5.

Discussion
The bacterial profiles of complex microbial communities associated with the different life-cycle stages of H. contortus field strains were explored with the aid of Illumina MiSeq sequencing technology. Furthermore, in this study the authors have advocated a set of measures that resulted in the absence of the cross-contamination by unwanted abomasal microbial community. Bacteria are presumably transmitted vertically (from parents), horizontally (from the environment or unrelated animals) or in some cases a mixture of horizontally and vertically. As the barber's pole worm, H. contortus is exposed to the microbiota of the rumen and abomasum; the nematode gut microbiota may reflect the external micro-organisms (i.e. from microhabitat [abomasum and rumen] and macrohabitat [pasture and manure]), which continually pass through its alimentary canal as evidenced by haemonchine larvae with internalized transgenic Escherichia coli. These results match those observed in earlier studies. Relatedly, significant changes in the abundance and composition of abomasal and ruminal microbiota were observed throughout H. contortus infections under the same diet regimen (forage (grass silage): concentrate (barley, maize, wheat, wheat bran and bruised soya Similarly, for all stages, the parasite microbiome is highly dominated by microbes of the following bacterial phyla, Proteobacteria and Firmicutes for egg-stage microbiome, Proteobacteria for adult-stage microbiome and Proteobacteria, Firmicutes and Bacteroidetes for larval-stage microbiome using V3-V4 tags. However, the predominant phyla observed in the three stages of the H. contortus life-cycle were Proteobacteria for egg-and adultstage microbiome and Proteobacteria 96.29% and Bacteroidetes 1.53% for larval stage microbiome employing V5-V7 tags. Furthermore, the free-living stages of H. contortus, such as L 1 (rhabditiform), and L 2 (rhabditiform) and L 3 filiariform infective larvae on pasture may be a vector or reservoir, of environmental microbial community 26 . These results corroborate the ideas of 27 , who suggested that the representative members from these phyla are universal and are not limited to the peculiar environment. Among the aforementioned genera (see result section), Pseudomonas, Ochrobactrum and Escherichia-Shigella were predominantly shared by all the life-cycle stages of H. contortus (eggs, larvae and adults) using V3-V4 and V5-V7 amplicons. Ochrobactrum spp., which are observed as emerging opportunistic pathogens, have been isolated from soil, hospital water sources, the rhizosphere of potato (Solanum tuberosum), rice (Oryza sativa), and wheat (Triticum sativum). The expression of AmpC beta-lactamases immensely complicates the therapy of infections caused by Ochrobactrum spp. 28 . Similarly, in 2009, Lacharme-Lora, et al. 26 reported the potential role of helminths as pathogen vectors. Intriguingly, α-Proteobacteria group has the ability to avoid the host innate recognition 29,30 . Additionally, it has been demonstrated that O. anthropi infection provokes a mix of Th1 and Th2 type responses in the mice 30 . It can therefore be assumed that the H. contortus microbiome play a role in shaping the ovine immune response. Of special concern is the supporting evidence that the repression of Beta There is a fierce competition for obtaining iron among microorganisms inside the ovine body to keep bacterial invaders under control. Iron is an indispensable element for many reduction-oxidation (redox) reactions and the virulence of pathogenic bacteria 31 . The predominant microbial genera, especially, Pseudomonas and Escherichia have developed an innovative way to confront with a variety of iron limitations by colonization of all the life-cycle stages of the blood-sucking parasite, H. contortus (eggs, larvae and adults). The probable association between the predominant microbial genera, such as Pseudomonas and Escherichia, and ovine hosts were further observed by data from H. contortus in naturally infected caprine hosts (unpublished data). The next few years will likely witness an immense number of studies uncovering the exact nature of this microbiome-H. contortus interaction, and finally reveal the specific molecular mechanisms underlying the nematode microbiome as an achilles' heel of the gastrointestinal parasites. There is abundant room for further progress in comparing microbiome of different helminths, which has, and will continue to offer an important parallel goal for the removal of a wide-variety of devastating animal and human diseases.