Occupational exposure in swine farm defines human skin and nasal microbiota

Abstract

Anthropogenic environments take an active part in shaping the human microbiome. Herein, we studied skin and nasal microbiota dynamics in response to the exposure in confined and controlled swine farms to decipher the impact of occupational exposure on microbiome formation. The microbiota of volunteers was longitudinally profiled in a 9-months survey, in which the volunteers underwent occupational exposure during 3-month internships in swine farms. By high-throughput sequencing, we showed that occupational exposure compositionally and functionally reshaped the volunteers' skin and nasal microbiota. The exposure in farm A reduced the microbial diversity of skin and nasal microbiota, whereas the microbiota of skin and nose increased after exposure in farm B. The exposure in different farms resulted in compositionally different microbial patterns, as the abundance of Actinobacteria sharply increased at expense of Firmicutes after exposure in farm A, yet Proteobacteria became the most predominant in the volunteers in farm B. The remodeled microbiota composition due to exposure in farm A appeared to stall and persist, whereas the microbiota of volunteers in farm B showed better resilience to revert to the pre-exposure state within 9 months after the exposure. Several metabolic pathways, for example, the styrene, aminobenzoate, and N-glycan biosynthesis, were significantly altered through our PICRUSt analysis, and notably, the function of beta-lactam resistance was predicted to enrich after exposure in farm A yet decrease in farm B. We proposed that the differently modified microbiota patterns might be coordinated by microbial and non-microbial factors in different swine farms, which were always environment-specific. This study highlights the active role of occupational exposure in defining the skin and nasal microbiota and sheds light on the dynamics of microbial patterns in response to environmental conversion.

Keywords: human microbiota; longitudinal investigation; microbial diversity; occupational exposure; swine farm.

Figure 1

Study design and quality control of amplicon sequencing data. (A) Schematic overview of the study design. S/N stands for the skin and nasal samples collected from the foreheads and nasal cavities of volunteer students. (B) Alpha rarefaction curves show a relatively higher bacterial community richness in the skin samples compared to the nasal samples. The average number of the observed features for each sequencing depth is presented. (C) Shannon entropy for skin and nasal swab samples. The t-tests were conducted to compare the diversity between skin and nasal swab samples.

Figure 2

Microbial diversity alteration across three stages. Alpha and beta diversity of microbiota from students’ skin and nasal swab samples displayed in (A) group A and (B) group B. The Wilcoxon tests were conducted across different stages to display the significance of alpha diversity. The PCoA analysis was performed based on Bray-Curtis distances.

Figure 3

The variation of microbiota composition across stages. (A) The bacterial composition on the phyla level across stages is presented via pie charts. The top five phyla based on relative abundance from each stage were filtered and merged within two groups. The volcano plots show the alteration of microbiota composition on genus level of skin and nasal samples during three phases in (B) group A and (C) group B. The log2FoldChange was used to illustrate the variation of microbiota composition with occupational exposure in swine farms (T1-3) compared with T0 and T4-9. The red/blue dots represent the significant genera up−/downregulation in T1-3 compared to both T0 and T4-9, and the red−/blue-bordered, gray-filled dots indicate significant up−/downregulation in T1-3, compared to T0 or T4-9.

Figure 4

Predicted microbial function comparison across stages based on KEGG level-3. The skin (A) and nasal (B) swab results from group A and the skin (C) and nasal (D) swab results from group B across stages are presented via extended error bar charts. The two-sided t-test was conducted and the functions that significantly up−/downregulated in T1-3 compared to both T0 and T4-9 are presented.