Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Comparative assessment of faecal microbial composition and metabonome of swine, farmers and human control

Abstract

The gastrointestinal tract of humans and swine consist of a wide range of bacteria which interact with hosts metabolism. Due to the differences in co-evolution and co-adaptation, a large fraction of the gut microbiome is host-specific. In this study, we evaluated the effect of close human-animal interaction to the faecal metagenome and metabonome of swine, farmer and human control. Three distinct clusters were observed based on T-RFLP-derived faecal microbial composition. However, 16S-inferred faecal microbiota and metabolic profiles showed that only human control was significantly different from the swine (P < 0.05). The metabonome of farmers and human controls were highly similar. Notably, higher trimethylamine N-oxide (TMAO) and butyrate were detected in human control and swine, respectively. The relative abundance of TMAO was positively correlated with Prevotella copri. Overall, we compared and established the relationship between the metabolites and microbiota composition of swine, farmers and human control. Based on the data obtained, we deduced that long term occupational exposure to swine and farm environment had affected the gut bacterial composition of farmers. Nonetheless, the effect was less prominent in the metabolite profiles, suggesting the gut bacteria expressed high functional plasticity and are therefore resilience to the level of community shift detected.

Introduction

Gastrointestinal tract of humans and animals harbours a vast community of microorganisms which holds enormous physiochemical and metabolic potentials. The bacterial communities are able to interact with the diet, immune responses, genetic and epigenetic composition of the hosts by compensating numerous biological activities lacking in the host’s biological systems1,2,3. For instance, independent studies carried out using germ-free mice and human volunteers showed that through bacterial fermentation, gut microbiota are able to assist energy harvesting from diet and poorly digestible polysaccharides4,5. In addition, gut microbial community can also influence hosts’ neural development, cognition and behaviour6. It is therefore important to prevent the disruption of gut microbiome to maintain the stability of its functions.

Modulation of gut microbiome can also occur in response to external factors such as environmental stress, antibiotic treatments, diets and exposure to different groups of environmental bacteria3,7. In recent years, it is increasingly recognised that the interaction between humans, animals and their shared environments is an important determinant for public health. Such “One Health” concept has become more important amid the rise of industrial animal production which increased the proximity of the living space between humans and farm animals. For example, long term occupational interactions between humans and swine in swine farms may facilitate the transmission of anthropozoonoses and zooanthroponoses between humans and swine, especially diseases that can be found in both humans and animals such as rabies, brucellosis, salmonellosis and H1N1 virus. For instance, a surge in the prevalence of phylogenetic closely related strains of hepatitis E virus of swine and humans has been reported in animal reservoirs from Uruguay8.

Other than pathogenic microorganisms, an exposure to same microbial source may also results in the reciprocal exchange of non-pathogenic microbial community9,10. Studies had shown that young children who live or being raised in farm environment harbour a wide spectrum of microbes that confers certain degree of protection against the development of asthma and allergies11,12,13. Separately, the usage of antibiotics in the farm may also impact the commensals in humans and animals while increasing the establishment of antimicrobial resistant bacteria in gastrointestinal tract14. To date, many studies on swine-related metagenomics and metabolomics have been carried out mainly to improve the breeding strategies such as animal health assessment, bioproduct characterization, feed efficiency and livestock growth potential15,16,17. Comparatively, few had applied the One Health concept into the microbiome and metabonome to understand the interaction/transmission of gut microbiomes across hosts18.

In this study, we investigated the faecal metagenome and metabonome of swines and swine farmers. To elucidate the influence of farm environment to the gut microbial composition, human subjects who have no direct contact/access to the swine farm was selected as control. By comparing the metagenomics and metabolites profiles of these three groups, our study aimed to understand the interaction between human and animal microbiomes, and its impact to the host metabolisms.

Results

Comparison of faecal bacterial composition of swine, farmer and human control group based on T-RFLP analyses

A clear separation between the bacterial composition of swine and human control was observed in CAP1-axis of the CAP plot (Fig. 1). In comparison, swine and farmers were separated in CAP2-axis. The significance of the separation was statistically tested using PERMANOVA. Based on PERMANOVA, all three pairing including swine vs control, swine vs farmers and control vs farmers were statistically significant with P(MC) <0.05. While T-RFLP was useful in assessing the overall structure of the bacterial community, the method did not provide taxonomic information of the taxa present in the faecal samples collected. Thus, 40 samples (nfarmer = 16; nswine = 16; nhuman control = 8) were randomly selected for 16S pyrosequencing to elucidate the taxonomic composition of the three sampling groups (i.e. swine, human control and farmers).

Figure 1
figure1

Faecal bacterial composition of swine, farmers and human control. The canonical analysis of principal coordinates (CAP) plot shows three distinct groups of bacterial composition were detected in different host (swine (n = 91), farmers (n = 17) and human control (n = 16)) based on T-RFLP.

Metagenome analyses based on 16S pyrosequencing

A total of 304,658 raw reads were obtained from 16S pyrosequencing. The final dataset after trimming, quality filter and chimera removal consisted of 145,752 sequences. The coverage of the sequences ranged from 89–99% and the sequences were clustered into 3268 operational taxonomic units (OTUs). Venn diagram was constructed based on sequence abundance. Swine has the highest level of host specific taxa (n = 1555), followed by farmer (n = 771) and human control (n = 461) (Supplementary Fig. S1). A three times higher overlap in OTUs was observed between swine and farmer (n = 91) as compared to swine and human control (n = 30).

The faecal bacterial diversity, richness and evenness were determined by the Shannon-Weiner diversity index (H’), Simpson diversity index (1−λ) and Pielou’s evenness index (J’). Based on these alpha diversity indices, swine faecal sample had the highest richness and evenness (Hswine = 3.52 ± 0.80; 1−λswine = 0.90 ± 0.06; J’swine = 0.67 ± 0.10), followed by farmers (H’farmer = 3.20 ± 0.51; 1−λfarmer = 0.90 ± 0.06; J’farmer = 0.63 ± 0.08) and human controls (H’human control = 3.09 ± 0.44; 1−λfarmer = 0.90 ± 0.04; J’human control = 0.58 ± 0.07) (Supplementary Fig. S2). A significant higher evenness was found in swine when compared to human control (F = 6.432, P = 0.019).

Taxonomic composition of faecal samples obtained from swine, farmers and human controls

Bacteroidetes, Firmicutes and Proteobacteria made up >93% of the phyla detected in the faecal samples from all three groups. The relative abundance of Bacteroidetes was higher than Firmicutes in all three groups of samples (Figs. 2a,b), with the Firmicutes/Bacteroidetes ratios of 0.53, 0.46 and 0.19 for swine, farmer and human control, respectively.

Figure 2
figure2

Relative abundance of faecal bacterial community. (a) Relative abundance of bacterial phyla in the faecal sample of swine, farmer and human control (b) Relative abundance of core bacterial genera in the faecal sample of swine, farmer and human control.

At phylum level, farmer and swine harboured higher relative abundance of Firmicutes (29% in swine, 27% in farmers and 15% in human control) and Proteobacteria (12% in swine, 8% in farmers and 4% in human control) than human control (Fig. 2a). When the taxonomic composition was examined at genus level, Prevotella was the most dominant genus in all three groups (Fig. 2b). Both human samples (i.e. farmer and human control) showed higher level of Lactobacillus than swine. Overall, the top 20 bacterial genera were more dominant in both farmer and human control than in swine.

Under PLS-DA, clear separation between humans (i.e. human control and farmer) and swine were observed in the axis X-variate 1 while human control clustered separately from swine in axis X-variate 2 (Fig. 3a). Permutation distance analysis of variance (PERMANOVA) with Euclidean distance showed that the metagenomics profiles of human control and swine were significantly different (Pseudo-T = 2.1386, P(MC) = 0.001). However, no significant difference was found between the faecal microbiota of farmers and swine (Pseudo T = 1.2942, P(MC) = 0.112) and between farmers and human controls (Pseudo-T = 1.2999, P(MC) = 0.110).

Figure 3
figure3

Faecal bacterial composition of swine, farmers and human control. (a) PLS-DA score plot showing the distribution of OTUs based on swine, farmer and human control. (b) PLS-DA score plot showing the metabonomes distribution of swine, farmer and human control.

Differentially expressed OTUs

Negative log binomial model was used to identify OTUs that differed significantly across host (Supplementary Fig. S3, Table 1). OTU0007 (P. copri), OTU0018 (P. copri), OTU0034 (Dialister spp.) and OTU0036 (Faecalibacterium prausnitzii) were highly expressed in farmers and human controls when compared to swine. OTU0002 (Enterobacteriaceae), OTU0011 (Escherichia coli), OTU0031 (unclassified bacteria under the class Bacilli) and OTU0044 (unclassified bacteria under Bacteroidales S47 family) were elevated in swine in comparison human control. Lastly, in comparison to farmers, OTU0031 and OTU0044, OTU0055 (Streptococcus alactolyticus) and OTU0062 (Prevotella spp.) were more prevalent in swine.

Table 1 Taxonomy of OTUs with significant expression in different groups of samples.

Faecal metabolic profiles of humans and swine

All metabolites reported in this study are listed in Table 2. Metabolites detected in the two groups of human samples (farmers and human controls) were identical, which included acetate, butyrate, lactate, alanine, lipids in VLDL, lipids in LDL, ornithine, ethanol, propionic acids, taurine, Scyllo-Inositol and β-glucose (Fig. 4a). The identity of the metabolites was validated by 2D-NMR spectroscopy. Except for ethanol, all other detected metabolites were also present in the swine faecal samples (Fig. 4b). PLS-DA and PERMANOVA were used to evaluate the differences in metabolic profiles between groups (Fig. 3b, Supplementary Table S1). Significant difference between metabonome was detected between human control and swine (Pseudo-T = 2.0793, P(MC) = 0.010), but not swine with farmers (Pseudo-T = 1.5397, P(MC) = 0.078), as well as farmers and human control (Pseudo-T = 1.2849, P(MC) = 0.175).

Table 2 1H NMR peak assignments for identified metabolitesa.
Figure 4
figure4

1H NMR spectroscopy showing the metabolites profiles of (a) human faecal sample, (b) swine faecal sample and the (c) covariance plot between the metabolites profiles of human control (upwards) and swine (downwards). Key as indicated in Table 2.

Significantly expressed metabolites were identified using permutation test and presented in the covariance plot (Fig. 4c). Among the detected metabolites, butyrate was found to be significantly elevated in swine as compared to human control (Fig. 4c). On the other hand, trimethylamine-N-oxide (TMAO) was over-expressed in human control in comparison to swine. No distinct metabolite was found to be differentially expressed between human control and farmers, as well as farmers and swine.

Integration of gut microbial composition and metabolomics profiles

The gut microbiota and metabolomics profiles were merged and projected using sPLs plot (Fig. 5). When both 16S gut microbial composition and faecal metabonome was considered together, a stronger clustering based on host species (human vs swine) was observed. A network analysis was further conducted to elucidate the association between the selected OTUs and metabolites (Fig. 6). Positive correlation was found between TMAO with OTU0007 (P. copri). However, butyrate which was significantly elevated in swine in comparison to human control was not correlated to any of the OTUs.

Figure 5
figure5

sPLS score plot showing the distribution of OTUs of all samples in relation to their metabolomics profiles.

Figure 6
figure6

Correlation analysis of all significant OTUs with the two significant metabolites. The network analysis revealed the co-occurrence patterns of bacteria and the metabolites. The blue colour lines represent positive correlation and the thickness of the lines represents the distance.

Discussion

Direct contact is one of the major factors contributing to the transmission of pathogens between animals and humans. Close interaction between animals and humans can also increase the risk for horizontal transfer of antibiotic resistance genes in human microbiome19. Among the different types of “contact”, human-livestock contacts were the most common cause of zoonotic pathogens transmission20. Despite the importance, there is a lack of knowledge on the impact of close contact to the transfer of non-pathogenic commensals. Such notable lack of reports is striking, given the increasing recognition of the importance of both pathogenic and non-pathogenic members of microbiome in health18. In our previous study, we detected the presence of porcine-related Enterococcus faecalis (E. faecalis) in the gut of the humans and human-related strain in the gut of swine21. E. faecalis is a normal microbiota commonly found in the gut of humans and mammals. Consistent with this, porcine-related gentamicin-resistant E. faecalis were also reported in humans in Denmark in year 201022. Such transmission not only present a health burden to the livestock and cause potential economic loss, but also poses a risk of subsequent reinfection in humans23,24.

In this study, we evaluated the impact of close human-swine interaction by integrating the results of 16S metagenomics and 1H-NMR-based metabolomics of faeces collected from swine, farmers and human control. Our result indicated the presence of host-specific gut microbiome between humans and swine (Fig. 1). The latter also showed higher alpha-diversity as compared to the former (Supplementary Fig. S2). In a parallel study by Sun et al.25, the faecal samples of swine farm workers were found to contain lower species diversity, while a clear division in faecal microbiota was observed between swine, farmers and the local villagers. Regardless, farmers harboured relatively more similar gut microbial community to swine in comparison to the human control, who has no direct contact with the livestock (Fig. 3a, Supplementary Fig. 1).

Overall, Firmicutes and Bacteroidetes were the predominant phyla found in all three groups of samples (swine, farmer and human control). The high prevalence of the two phyla (e.g. together attributed 85–90% of the total sequences) observed was consistent with previous reported microbiome assessments on humans and swine2,15,26,27,28. The Firmicutes/Bacteroidetes ratio is commonly related to the health status and diet of humans and swine29,30,31. Interestingly, we found a higher relative abundance of Bacteroidetes than Firmicutes in all three groups (Fig. 2a,b). Bacteroidetes involved in host’s metabolism possibly harvest energy from indigestible polysaccharides and produce short chain fatty acids (SCFAs). The microbiome in the guts of humans undergo a change in the relative abundance of the two major phyla at different stage of life32. An increase in Firmicutes/Bacteroidetes ratio was also reported in the gut microbiota of obese individuals33,34,35.

At genus level, a member of Bacteroidetes, Prevotella spp. dominated the faecal metagenome of all three groups of samples. One of the major species of Prevotella spp. is Prevotella copri (P. copri), which was previously reported to be positively associated with rheumatoid arthritis by favouring Th17 lymphocytes development and induced tissue damage in rheumatoid arthritis patients36,37. Apart from P. copri, a higher level of Faecalibacterium prausnitzii (F. prausnitzii) was detected in farmers. This bacterium was reported as one of the most abundant bacterial species in gut of healthy humans and animals, including swine38. F. prausnitzii is able to control gut epithelial cells metabolism, host immune response and produce important SCFA such as butyrate39. Butyrate is one of the major anti-inflammatory metabolites found in the gut. Previous studies had reported a decrease of F. prausnitzi in patient associated with psoriasis and inflammatory bowel disease such as Crohn’s disease, Coeliac disease and ulcerative colitis40,41,42. In the faeces of swine, we detected high prevalence of Streptococcus alactolyticus, which is a common commensal in animals such as dogs and swine but rarely detected in humans43,44. Nonetheless, zoonotic infection of S. alactolyticus infections in humans was previously reported45,46. Although the role of S. alactolyticus in gut was not clear, the bacterium is known to secrete functional metabolites such as amylase, galactosidase, β-glycoside hydrolase, acidic galactose, αgalactosidase, and urease47.

Gut microbiota plays an important role in maintaining the homeostasis of the host’s body and majority of the physiological contributions of gut bacteria are involved in fermentation and production of SCFAs such as acetate, propionate and butyrate. For instance, gut microbes can ferment complex carbohydrate in dietary fibre into SCFAs48. Although a shift in the faecal microbiota of farmers was observed, the overall faecal metabolites of the two groups of humans remained comparable. This shows that there is a high level of functional plasticity in the gut microbial community. In our study, two metabolites were found to be upregulated in specific sample group. A higher level of butyrate was found in the swine faeces. Although our integration study showed that the production of butyrate was not linked to any gut bacteria, an association of butyrate production with high abundance of Firmicutes was reported49. We speculate that the butyrate is produced collectively by a group of bacterial taxa and hence a linear relationship between the metabolite and bacterial OTUs is absent. Butyrate has been reported to associate with many health issues ranging from anti-inflammatory properties, host immunity and enhancement of intestinal barrier function50. On the other hand, P. copri was positively correlated with the level of TMAO found in the human control. TMAO is known to be a by-product of dietary choline digestion. Food rich in dietary choline include egg yolks and meats51. TMAO is vital for platelet responsiveness and thus plays a vital role in increasing the incidence of thrombotic events such as heart attack and stroke52. P. copri has been implicated in a number of autoimmune diseases such as colitis, inflammatory bowel disease and correlated with adverse cardiovascular effects due to the increase of TMAO as microbial by-product53,54,55. Our findings were concordant with previous studies by Scher et al.56 and Koeth et al.51 who reported that abundance of P. copri was correlated with the level of TMAO.

In summary, we showed that occupational contact between farmers and livestock may result to a bacterial community shift in human gut microbiome, as evident in the higher similarity in microbiome between farmers and swine than human control. Despite these changes, no substantial difference in the metabonome was detected between farmers and human control. The lack of effect may suggest that the changes are transient and can be compensated with the high functional plasticity of the gut bacteria. It is however possible that the health effect may only manifest under long-term exposure. As such, a long-term monitoring study of microbiome and health outcomes of farmers is warranted.

Methods

Samples collection

Seven swine farms (five farms located in the northern region coded as PF1, PF2, PF3, PF4 and PF5 and two farms located in central region coded as SF1 and SF2) located in the high-density swine farming areas in Peninsular Malaysia were sampled between August 2013 to December 2013. A total of 91 swine faecal samples were collected from the animals. All samples were collected under the supervision of a veterinarian from the Faculty of Veterinary Medicine, Universiti Putra Malaysia (UPM). Separately, 33 faecal samples were collected from swine farmers (n = 17) who worked in the seven participating farms and non-farmer human control group (n = 16). All swine farmers involved in this study have been working in the swine farms for at least two years. The human subjects were advised to defecate directly into the stool collection bottle or onto a clean surface and immediately transfer the faecal sample into the collection tube by using the scoop on the cap. Background information of the samples was inferred based on questionnaire as well as the observations and advices given by the attending veterinarian. The information included farm locations, farm hygiene practice, gender of the swine (male/female), body temperature and health condition of the swine (healthy/unhealthy). Physical examination (clinical signs, behavior and body temperature) of the swine was performed to determine their health status by the field veterinarian. Swine that presented with abnormal clinical signs, behavior and elevated body temperature were categorized as unhealthy. All the human subjects and swine were in healthy or asymptomatic condition during the sampling. All samples were transported on ice to Kuala Lumpur and stored at −80 °C at the earliest opportunity. This study was conducted following the guidelines as stated in the Code of Practice for Care and Use of Animals for Scientific Purposes as stipulated by UPM (UPM/IACUC/FYP- AUP-T006/), complied with the current guidelines for the care and use of animals, and was approved by the Animal Care and Use Committee (ACUC), Faculty of Veterinary Medicine, UPM. The human samples collection was approved by Medical Research Ethics Committee, University Malaya Medical Centre (UMMC-MREC) (Ethic committee/IRB reference number: 1010.41) and performed in accordance with the UMMC-MREC guidelines. Informed consent was obtained from all human subjects.

Terminal-Restriction Fragment Length Polymorphism (T-RFLP)

DNA the faecal samples were extracted by using QIAamp DNA Stool Mini Kit (Qiagen, Hilden, Germany) according to manufacturers’ instruction. 16S rDNA amplifications were performed on the extracted DNA by using Universal primers 27F-FAM and 1492R-HEX as described in Chong et al.57. Briefly, both forward (27 F) and reverse primer (1492 R) were tagged with different fluorophores (i.e. FAM and HEX) via PCR. The PCR products were purified by using Wizard Genomic DNA purification kit (Promega, USA) and the purified DNA was digested with Msp I restriction enzyme (Promega, USA). The digested products were commissioned to a local commercial company for electrophoretic separation of restriction fragments. The resulting electropherograms were first processed with the Peak Scanner Software v1.0 (Life Technologies, USA). Subsequently, noise filtration, alignment and scoring were conducted using web-based T-REX program (http://trex.biohpc.org/). Peak alignment was carried out by binning the signals at a clustering threshold of 0.5 bp, starting from the smallest fragment length among all the T-RFLP profiles. The scoring of the peaks was recorded as peak area, and normalized by dividing the individual peak over the total peak area of each sample. The scoring datasheet was exported into PRIMER 7 & PERMANOVA (PRIMER-E Ltd, UK) for statistical analyses. Briefly, the beta diversity was assessed using Bray-Curtis Distance based canonical analysis of principal coordinates (CAP) and Permutational Multivariate Analysis of Variance (PERMANOVA).

Amplicons -Next-Generation Sequencing (16S-NGS)

Forty samples including 16 swine, 16 farmers and 8 human control were selected for 16S-NGS. The 16S rRNA genes fragments from variable V3 regions were amplified using primer set 27 F (GAGTTTGATCMTGGCTCAG) and 518 R (WTTACCGCGGCTGCTGG) containing sample specific barcodes. Amplicon pyrosequencing was performed by Macrogen Inc. (Seoul, South Korea) using Roche 454 GS-FLX system (Roche, NJ, USA). The pyrosequencing produced a total of 808,275 sequence reads with an average read length of 374 bp. The sequences obtained were processed using Mothur software (v.1.34.3)58 according to the 454 SOP (http://www.mothur.org/wiki/454_SOP). In brief, the raw sequences were first processed by “sff.multiple” command. The sequences were denoised and filtered by removing sequence shorter than 250 bp and longer than 550 bp. In addition, maximum homopolymer count was set at 6 bp while the maximum allowable differences in primer and barcode sequences were set at 2 bp. The sequences were aligned to SILVA-compatible alignment reference database (Version 132). Sequences which were poorly aligned and overhangs at the both ends were removed so that the sequences overlapped at the same region. Unique sequences were screened and chimeric and ambiguous sequences classified to unrelated taxon were removed by using “chimera.uchime” and “remove.lineage” commands. The dataset was clustered into OTU by using 97% cut-off. The final aligned dataset contained 17,660 unique sequences. Alpha diversity was assessed with Shannon diversity index, Simpson index, Pielou’s evenness. The “DIVERSE” option in the PRIMER 7 data analyses packages (PRIMER-E Ltd, UK) was used to obtain the alpha diversity index. The beta diversity among the samples were elucidated using Partial Least Square - Discriminant Analysis (PLS-DA) and PERMANOVA. Prior to the analysis, the data was ‘regularised log’ transformed. PLS-DA implemented in the mixOmics R package59 was used to visualise the separation between different groups of samples while the compositional differences was compared using PERMANOVA. Separately, differentially expressed OTUs were identified based on negative binomial distribution using DESeq. 2 R package60.

Sample preparation and 1H NMR spectroscopic analysis

Faecal samples were processed by using the NMR buffer [1 mM of 3-(trimethylsilyl) propionate (TSP) and 3 mM sodium azide (D2O: H2O, v/v, 8:2; pH 7.4)]. TSP was used as a reference for chemical shift. For each sample, 0.05 g of faecal matter was homogenized and vortexed in one ml of NMR buffer. The mixture was sonicated for 30 min and centrifuged at 14,000 rpm for 10 min. Six hundred µl of supernatant were transferred to 5 mm-diameter NMR tubes (Norell, USA). The processed samples were stored at −80 °C until analysis.

A standard 1-dimensional (1-D) 1H NMR spectrum was acquired by using Bruker AVIII 600 MHz spectrometer (Bruker Biospin, Fallenden, Switzerland) with a 5 mm PABBO BB probe operating at 600.17 MHz. The field frequency was locked on the D2O solvent and water peak suppression was performed during RD of 2 s and mixing time (tm) of 10 s. In addition, 2-D NMR using 1H-1H correlation spectroscopy (COSY) and 1H-1H J-resolved (JRES) were performed on selected representative samples to assist metabolite identification.

The NMR spectra were manually phase- and baseline-corrected using Bruker TopSpin 4.0.6 and imported into MATLAB (version 2014b). All the spectra were referenced to the TSP resonance at δ 0.00. The spectra were digitized into data point using in-house developed MATLAB script (O. Cloarec, Imperial College London). The region containing noise (δ 0.0–0.5 and δ 9.2–10.0) and water resonance (δ 4.5–6.5) were removed. Spectra normalization was performed and the regions with TSP peaks, water presaturation imperfection and the end regions containing only noise were removed. PLS-DA was used to illustrate the relationship between groups. The significance and validity of statistical differences were calculated using permutation test (number of permutations = 1000). Covariance plots were generated to visualize the significance of each metabolite from the permutation test. The colour scheme projected onto the spectrum indicate the significance of the metabolites. Blue indicating to no significant difference (P > 0.05 confidence level) and red indicating significant difference (P < 0.01 confidence level). The relative concentrations of the significant metabolites were further calculated by using in-house developed MATLAB script (O. Cloarec, Imperial College London).

Linking faecal metabolites with gut microbiota composition

The integration and visualization of OTUs and metabonomes was performed using sparse partial least squares (sPLS) regression method implemented in R mixOmics package. sPLS allows the integration of heterogeneous omics data from the same set of samples, OTUs (matrix X) and metabonomes (matrix Y). The relationship was projected using sPLS plot and network diagram.

Declarations of ethical approval and consent to participate

This study was approved by the Animal Care and Use Committee (ACUC), Faculty of Veterinary Medicine, UPM and conducted according to the guidelines as stated in the Code of Practice for Care and Use of Animals for Scientific Purposes as stipulated by UPM (UPM/IACUC/FYP- AUP-T006/). The human samples collection was approved by Medical Ethics Committee, University Malaya Medical Centre (Ethic committee/IRB reference number: 1010.41) and performed with the informed consent of human subjects.

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

References

  1. 1.

    Kau, A. L., Ahern, P. P., Griffin, N. W., Goodman, A. L. & Jeffrey, I. Human nutrition, the gut microbiome, and immune system: envisioning the future. Nature 474, 327–336 (2012).

    Article  CAS  Google Scholar 

  2. 2.

    Isaacson, R. & Kim, H. The intestinal microbiome of the pig. Anim. Heal. Res. Rev. 13, 100–109 (2012).

    Article  Google Scholar 

  3. 3.

    Marchesi, J. et al. The gut microbiota and host health: A new clinical frontier. Gut 65, 330–339 (2016).

    PubMed  Article  Google Scholar 

  4. 4.

    Kaoutari, A. E., Armougom, F., Gordon, J. I., Raoult, D. & Henrissat, B. The abundance and variety of carbohydrate-active enzymes in the human gut microbiota. Nat. Rev. Microbiol. 11, 497–504 (2013).

    PubMed  Article  CAS  Google Scholar 

  5. 5.

    Valdes, A. M., Walter, J., Segal, E. & Spector, T. D. Role of the gut microbiota in nutrition and health. BMJ 361, 36–44 (2018).

    Google Scholar 

  6. 6.

    Rogers, G. B. et al. From gut dysbiosis to altered brain function and mental illness: Mechanisms and pathways. Mol. Psychiatry 21, 738–748 (2016).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  7. 7.

    Candela, M., Biagi, E., Maccaferri, S., Turroni, S. & Brigidi, P. Intestinal microbiota is a plastic factor responding to environmental changes. Trends in Microbiol 20, 385–391 (2012).

    CAS  Article  Google Scholar 

  8. 8.

    Mirazo, S. et al. Serological and virological survey of hepatitis E virus (HEV) in animal reservoirs from Uruguay reveals elevated prevalences and a very close phylogenetic relationship between swine and human strains. Vet Microbiol 213, 21–27 (2018).

    CAS  PubMed  Article  Google Scholar 

  9. 9.

    Song, S. et al. Cohabiting family members share microbiota with one another and with their dogs. Elife 2, e00458 (2013).

    PubMed  PubMed Central  Article  Google Scholar 

  10. 10.

    Pehrsson, E. et al. Interconnected microbiomes and resistomes in low-income human habitats. Nature 533, 212–216 (2016).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  11. 11.

    Vestergaard, D. et al. Pig farmers’ homes harbor more diverse airborne bacterial communities than pig stables or suburban homes. Front. Microbiol. 9, 1–14 (2018).

    Article  Google Scholar 

  12. 12.

    Roduit, C., Frei, R., von Mutius, E. & R, L. The hygiene hypothesis. In Environmental Influences on the Immune System (ed. Esser, C.) 77–96, https://doi.org/10.1007/978-3-7091-1890-0 (Springer, 2016).

    Google Scholar 

  13. 13.

    Tasnim, N., Abulizi, N., Pither, J., Hart, M. & Gibson, D. Linking the gut microbial ecosystem with the environment: Does gut health depend on where we live? Front. Microbiol. 8, 1–8 (2017).

    Article  Google Scholar 

  14. 14.

    Jakobsson, H. et al. Short-term antibiotic treatment has differing long-term impacts on the human throat and gut microbiome. PLoS One 5, e9836 (2010).

    ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

  15. 15.

    Lamendella, R., Domingo, J., Ghosh, S., Martinson, J. & Oerther, D. Comparative fecal metagenomics unveils unique functional capacity of the swine gut. BMC Microbiol. 11, 1–17 (2011).

    Article  CAS  Google Scholar 

  16. 16.

    Deusch, S., Tilocca, B., Camarinha-Silva, A. & Seifert, J. News in livestock research - Use of Omics-technologies to study the microbiota in the gastrointestinal tract of farm animals. Comput. Struct. Biotechnol. J. 13, 55–63 (2015).

    CAS  PubMed  Article  Google Scholar 

  17. 17.

    Wang, W., Hu, H., Zijlstra, R. T., Zheng, J. & Gänzle, M. G. Metagenomic reconstructions of gut microbial metabolism in weanling pigs. Microbiome 7, 1–11 (2019).

    CAS  Article  Google Scholar 

  18. 18.

    Trinh, P., Zaneveld, J. R., Safranek, S. & Rabinowitz, P. M. One health relationships between human. animal, and environmental microbiomes: A mini-review. Front. Public Heal. 6, 1–9 (2018).

    Google Scholar 

  19. 19.

    Smillie, C. S. et al. Ecology drives a global network of gene exchange connecting the human microbiome. Nature 480, 241–244 (2011).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  20. 20.

    Klous, G., Huss, A., Heederik, D. J. J. & Coutinho, R. A. Human-livestock contacts and their relationship to transmission of zoonotic pathogens, a systematic review of literature. One Heal. 2, 65–76 (2016).

    Article  Google Scholar 

  21. 21.

    Tan, S., Chong, C., Teh, C., Ooi, P. & Thong, K. Occurrence of virulent multidrug - resistant Enterococcus faecalis and Enterococcus faecium in the pigs, farmers and farm environments in Malaysia. PeerJ 6, e5353 (2018).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

  22. 22.

    Larsen, J. et al. Porcine-origin gentamicin-resistant Enterococcus faecalis in humans, Denmark. Emerg. Infect. Dis. 16, 682–684 (2010).

    PubMed  PubMed Central  Article  Google Scholar 

  23. 23.

    Messenger, A. M., Barnes, A. N. & Gray, G. C. Reverse zoonotic disease transmission (Zooanthroponosis): A systematic review of seldom-documented human biological threats to animals. PLoS One 9, 1–9 (2014).

    Article  CAS  Google Scholar 

  24. 24.

    Delahoy, M. J. et al. Pathogens transmitted in animal feces in low- and middle-income countries. Int. J. Hyg. Environ. Health 221, 661–676 (2018).

    PubMed  PubMed Central  Article  Google Scholar 

  25. 25.

    Sun, J. et al. Comparison of fecal microbial composition and antibiotic resistance genes from swine, farm workers and the surrounding villagers. Sci. Rep. 7, 1–7 (2017).

    ADS  Article  CAS  Google Scholar 

  26. 26.

    Chong, C. et al. Effect of ethnicity and socioeconomic variation to the gut microbiota composition among pre-adolescent in Malaysia. Sci. Rep. 5, 1–12 (2015).

    Google Scholar 

  27. 27.

    Chae, J. P., Pajarillo, E. A. B., Oh, J. K., Kim, H. & Kang, D.-K. Revealing the combined effects of lactulose and probiotic enterococci on the swine faecal microbiota using 454 pyrosequencing. Microb. Biotechnol. 9, 486–495 (2016).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  28. 28.

    Holman, D., Brunelle, B., Trachsel, J. & Allen, H. Meta-analysis to define a core microbiota in the swine gut. mSystems 2, e00004–17 (2017).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  29. 29.

    Greenhalgh, K., Meyer, K. M., Aagaard, K. M. & Wilmes, P. The human gut microbiome in health: establishment and resilience of microbiota over a lifetime. Environ. Microbiol. 18, 2103–2116 (2016).

    PubMed  Article  Google Scholar 

  30. 30.

    Walsh, H., Haq, H., Cersosimo, L., Kien, C. L. & Kraft, J. Decreased abundance of Firmicutes in the gut microbiota after consumption of a diet containing milk fats. FASEB J. 30, 683.11 (2016).

    Google Scholar 

  31. 31.

    Liu, B. et al. Response of gut microbiota to dietary fiber and metabolic interaction with SCFAs in piglets. Front. Microbiol. 9, 1–12 (2018).

    Article  Google Scholar 

  32. 32.

    Mariat, D. et al. The firmicutes/bacteroidetes ratio of the human microbiota changes with age. BMC Microbiol. 9, 1–6 (2009).

    Article  CAS  Google Scholar 

  33. 33.

    Ley, R., Turnbaugh, P., Klein, S. & Gordon, J. Human gut microbes associated with obesity. Nature 444, 1022–1023 (2006).

    ADS  CAS  PubMed  Article  Google Scholar 

  34. 34.

    Koliada, A. et al. Association between body mass index and Firmicutes / Bacteroidetes ratio in an adult Ukrainian population. BMC Microbiol. 17, 1–6 (2017).

    Article  CAS  Google Scholar 

  35. 35.

    Guo, X. et al. Development of a real-time PCR method for Firmicutes and Bacteroidetes in faeces and its application to quantify intestinal population of obese and lean pigs. Lett. Appl. Microbiol. 47, 367–373 (2008).

    CAS  PubMed  Article  Google Scholar 

  36. 36.

    Moreno, J. Prevotella copri and the microbial pathogenesis of rheumatoid arthritis. Reumatol. Clin. 11, 61–63 (2015).

    PubMed  Article  Google Scholar 

  37. 37.

    Maeda, Y. & Takeda, K. Role of gut microbiota in rheumatoid arthritis. J. Clin. Med. 6, 60 (2017).

    PubMed Central  Article  CAS  Google Scholar 

  38. 38.

    Miquel, S. et al. Faecalibacterium prausnitzii and human intestinal health. Curr. Opin. Microbiol. 16, 255–261 (2013).

    CAS  PubMed  Article  Google Scholar 

  39. 39.

    Ferreira-Halder, C. V., Faria, A. V., de, S. & Andrade, S. S. Action and function of Faecalibacterium prausnitzii in health and disease. Best Pract. Res. Clin. Gastroenterol. 31, 643–648 (2017).

    CAS  PubMed  Article  Google Scholar 

  40. 40.

    Eppinga, H. et al. Similar depletion of protective Faecalibacterium prausnitzii in psoriasis and inflammatory bowel disease, but not in hidradenitis suppurativa. J. Crohn’s Colitis 10, 1067–1075 (2016).

    Article  Google Scholar 

  41. 41.

    Lopez-Siles, M., Duncan, S. H., Garcia-Gil, L. J. & Martinez-Medina, M. Faecalibacterium prausnitzii: From microbiology to diagnostics and prognostics. ISME J. 11, 841–852 (2017).

    PubMed  PubMed Central  Article  Google Scholar 

  42. 42.

    Quévrain, E. et al. Identification of an anti-inflammatory protein from Faecalibacterium prausnitzii, a commensal bacterium deficient in Crohn’s disease. Gut 65, 415–425 (2016).

    PubMed  Article  CAS  Google Scholar 

  43. 43.

    Rinkinen, M. L., Koort, J. M. K., Ouwehand, A. C., Westermarck, E. & Björkroth, K. J. Streptococcus alactolyticus is the dominating culturable lactic acid bacterium species in canine jejunum and feces of four fistulated dogs. FEMS Microbiol. Lett. 230, 35–39 (2004).

    CAS  PubMed  Article  Google Scholar 

  44. 44.

    Devriese, L. A., Hommez, J., Pot, B. & Haesebrouck, F. Identification and composition of the streptococcal and enterococcal flora of tonsils, intestines and faeces of pigs. J. Appl. Bacteriol. 77, 31–36 (1994).

    CAS  PubMed  Article  Google Scholar 

  45. 45.

    Almeida, P., Railsback, J. & Gleason, J. B. A rare case of Streptococcus alactolyticus infective endocarditis complicated by septic emboli and mycotic left middle cerebral artery aneurysm. Case Rep. Infect. Dis. 2016, 1–3 (2016).

    Google Scholar 

  46. 46.

    Toepfner, N. et al. Fulminant neonatal sepsis due to Streptococcus alactolyticus -A case report and review. Apmis 122, 654–656 (2014).

    PubMed  Article  Google Scholar 

  47. 47.

    Sheng, Q. K., Yang, Z. J., Zhao, H. B., Wang, X. L. & Guo, J. F. Effects of L-tryptophan, fructan, and casein on reducing ammonia, hydrogen sulfide, and skatole in fermented swine manure. Asian-Australasian. J. Anim. Sci. 28, 1202–1208 (2015).

    CAS  Google Scholar 

  48. 48.

    Nicholson, J. K. et al. Host-Gut Microbiota Metabolic Interactions. Science (80-.). 108, 1262–1268 (2012).

    ADS  Article  CAS  Google Scholar 

  49. 49.

    Den Abbeele, P. V. et al. Butyrate-producing Clostridium cluster XIVa species specifically colonize mucins in an in vitro gut model. ISME J. 7, 949–961 (2013).

    PubMed  Article  CAS  Google Scholar 

  50. 50.

    Meijer, K., De Vos, P. & Priebe, M. G. Butyrate and other short-chain fatty acids as modulators of immunity: What relevance for health? Curr. Opin. Clin. Nutr. Metab. Care 13, 715–721 (2010).

    CAS  PubMed  Article  Google Scholar 

  51. 51.

    Koeth, R. A. et al. Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat. Med. 19, 576–585 (2013).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  52. 52.

    Zhu, W., Zeneng, W., Tang, W. H. W. & Hazen, S. L. Gut microbe-generated TMAO from dietary choline is prothrombotic in subjects. Circulation 135, 1671–1673 (2017).

    PubMed  PubMed Central  Article  Google Scholar 

  53. 53.

    Pianta, A. et al. Evidence of the immune relevance of Prevotella copri, a gut microbe, in patients with rheumatoid arthritis. Arthritis Rheumatol. 69, 964–975 (2017).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  54. 54.

    Bajer, L. et al. Distinct gut microbiota profiles in patients with primary sclerosing cholangitis and ulcerative colitis. World J. Gastroenterol. 23, 4548–4558 (2017).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  55. 55.

    Brusca, S. B., Abramson, S. B. & Scher, J. U. Microbiome and mucosal inflammation as extra-articular triggers for rheumatoid arthritis and autoimmunity. Curr. Opin. Rheumatol. 26, 101–107 (2014).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  56. 56.

    Scher, J. U. et al. Expansion of intestinal Prevotella copri correlates with enhanced susceptibility to arthritis. Elife 2, 1–20 (2013).

    Article  CAS  Google Scholar 

  57. 57.

    Chong, C. W. et al. Environmental influences on bacterial diversity of soils on Signy Island, maritime Antarctic. Polar Biol. 32, 1571–1582 (2009).

    Article  Google Scholar 

  58. 58.

    Schloss, P. D. et al. Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75, 7537–7541 (2009).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  59. 59.

    Rohart, F., Gautier, B., Singh, A. & Lê Cao, K. A. mixOmics: An R package for ‘omics feature selection and multiple data integration. PLoS Comput. Biol. 13, e1005752 (2017).

    ADS  PubMed  PubMed Central  Article  CAS  Google Scholar 

  60. 60.

    Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq. 2. Genome Biol. 15, 550 (2014).

    PubMed  PubMed Central  Article  CAS  Google Scholar 

Download references

Acknowledgements

We would like to thank Dr. Ooi Peck Toung from University Putra Malaysia for helping in sample collection. This research was funded by International Antimicrobial Resistance Research Program IF0042020 and University Malaya High Impact Research (HIR) grant [UM.C/625/1/HIR-MOHE/CHAN/11/02]. The funders did not play any role in study design, data collection and interpretation, or the decision to submit the work for publication.

Author information

Affiliations

Authors

Contributions

C.S.J.T., C.W.C., I.K.S.Y. and K.L.T. supervised the project. S.C.T. designed the study, performed the experiment and wrote the manuscript. S.C.T. and C.W.C. analysed and interpreted the data. All authors contributed to drafting, editing and critically reviewed the manuscript and contributed important intellectual input. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Cindy Shuan Ju Teh.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Tan, S.C., Chong, C.W., Yap, I.K.S. et al. Comparative assessment of faecal microbial composition and metabonome of swine, farmers and human control. Sci Rep 10, 8997 (2020). https://doi.org/10.1038/s41598-020-65891-4

Download citation

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing