The effects of human genetic and environmental factors on the gut and airway microbiome

Abstract

The human microbiome consists of an enormous variety of microorganisms including bacteria, archaea, fungi, and viruses that reside in all body surfaces that are exposed to the external environment such as the skin, oral cavity, airway, vagina, and gut. Although the human microbiome has long been known to play important roles in human health and disease, only the recent advances in high-throughput sequencing methods and analytic techniques have enable detailed investigation of the human microbiome composition, function, and host-microbe interactions. Through these studies, it has become apparent that in the healthy state, the human microbiota is involved in a broad range of activities for the maintenance of human health, including nutrient metabolism, epithelial cell proliferation, development and maintenance of the immune system, and protection against pathogens. Meanwhile, disruption of the normal microbiota (dysbiosis) has been linked to the various diseases such as obesity, cancer, a variety of inflammatory diseases of the airways, skin, mouth, and intestinal tract. To direct future therapy for microbiome-related diseases, it is necessary to first characterize the normal microbiota in the healthy population and identify factors that are associated with compositional changes in human microbiome. For these purposes, in this thesis, a large-scale investigation of the gut microbiota and the airway microbiota was conducted in Korean twins and their families. First, we identified that the gut microbiota of healthy Koreans were clustered into two enterotypes using metagenomic sequencing data which was generated from 36 fecal samples of healthy Korean monozygotic (MZ) twins. The two enterotypes were enriched by either Bacteroides or Prevotella. We observed that the enterotype of an individual was stable over time, and that most co-twins shared their enterotypes. We also found that the subjects enterotypes were significantly associated with the serum uric acid level and the long-term intakes of nutrients such as dietary fiber, vitamins, and minerals. These results suggest that both host genetics and dietary habits could be important contributors to the enterotype of an individual. Second, we characterized the gut microbiota of 655 MZ (N=306) and dizygotic (DZ) (N=74) twins, and family members of twin pairs (N=275), of which approximately 18% (121 individuals) had metabolic syndrome (MetS). We found that Methanobrevibacter, Lactobacillus, and Sutterella were significantly enriched in MetS individuals, while Akkermansia, Odoribacter, and Bifidobacterium were significantly enriched in healthy individuals. Among the gut microbes associated with MetS status, Actinobacteria, to which the Bifidobacterium belong, had the highest heritability (45.7%). Even after adjustment for MetS status, reduced abundances of Actinobacteria and Bifidobacterium were significantly linked to the minor allele at the apolipoprotein A-V gene (APOA5) SNP rs651821, which is associated with triglyceride level and MetS. Therefore, our results demonstrate that altered gut microbiota mediated by a specific host genotype can contribute to the development of MetS. Third, we assessed the influences of host genetics and lifestyles such as smoking, alcohol consumption, and physical activity on the airway microbiota composition. A total of 257 sputum samples from 74 MZ twin pairs (n=148), 14 DZ twin pairs (n=28), and their parents and siblings (n=81) were analyzed for airway microbiota composition using next-generation sequencing of partial 16S rRNA gene sequences. We found that several taxa, including Providencia and Bacteroides, were significantly influenced by host genetic factors. Smoking had the strongest effect on the overall microbial community structure among the various host lifestyle factors. The pack-year value was positively associated with abundance of the Veillonella genus, which is known to be related to airway inflammation, but negatively associated with that of Haemophilus. Co-occurrence network analysis showed that the taxa were clustered according to the direction of correlations or trends for smoking, and that the taxa influenced by host genetics were closely correlated with each other. These results demonstrate that the composition of the airway microbiota is shaped by complex interactions among host genetics and lifestyle factors, such as smoking. In conclusion, this study shows that the composition of the healthy human microbiome is influenced not only by extrinsic environmental factors including diet and lifestyle but also by host genetics, and that the changes in the abundances of the specific microbial taxa caused by these factors are associated with metabolic syndrome and possibly airway inflammation. These data suggest that personalized approaches will be required in the prevention and treatment of microbiome-related diseases.