Zierer et al. 2018 The fecal metabolome as a functional readout of the gut microbiome. Nature genetics.

This is a summary of our DalMUG journal club discussion of the above paper written by Akhilesh Dhanani.

Summary

The gut microbiome is a complex community of microbial entities and is known to maintain gut homeostasis and host metabolism. Also, secreted molecules of microbes contain several host-beneficial compounds such as hormones and vitamins. Disruption of microbiome interaction with the host plays a significant role in metabolic disorders such as obesity and insulin resistance. Use of most next-generation sequencing platforms has been limited to profiling microbial communities and not to decipher their functional and transcriptional capacities. In this study, Zierer et. al, used the fecal metabolome as a functional readout of the microbiome to study metabolic interplay among the host, diet and the gut microbiota. Untargeted fecal metabolomics profiling of 786 predominantly female twins of TwinsUK cohort identified 1,116 metabolites, of which 866 had known chemical identity. Of the 1,116 metabolites, 570 were common and detected in at least 80% of the samples and 345 were detected in more than 20% but less than 80% of the samples. Although there was no significant association between 915 fecal metabolites and age, a multivariate partial-least-squares discriminant analysis incorporating just common 570 metabolites separated the oldest decile (>75 years) from the youngest decile (<56 years) and metabolite, phytanate was significantly different between those two deciles. An imprecise measure of adiposity - BMI - was associated with eight metabolites, whereas visceral-fat mass (corrected for BMI), a measure of abdominal obesity found a total of 102 statistically significant associations and explained 28.4% of the observed total variance in visceral fat. The presence of significant amino acids enriched visceral-fat-associated metabolites supports the previous observations of the involvement of microbial amino acid metabolism in obesity. Additionally, the association of 43 fecal metabolites and 32 OTUs with visceral-fat mass was believed to be mediated by amino acid availability. A genome-wide association study of metabolites with heritable component identified three metabolites significantly associated with genetic loci. Replication of study with other cohort of 230 individuals, metabolic ratio of 5-acetylamino-6-amino-3-methyluracil to 1,3-dimethylurate was significant and associated to the NAT2 gene, which encodes an N-acetyltransferase. Furthermore, to study the extent to which the fecal metabolome reflects the functional capacity of gut microbiome, they estimated the proportion of variance in each metabolite explained by microbiome composition by using the unweighted UniFrac beta-diversity metric. Gut microbiome was found to explained on average 67.7% of observed variance of 710 metabolites.

Points of Interest

• Fecal metabolome provides more direct evidence of microbial activities in host metabolism than composition of gut microbiome.
• Large number of metabolites were identified which were unique to feces when compared to blood profiling of same individual.
• Visceral-fat mass had more statistically significant association (102 metabolites) compared to BMI (8 metabolites).
• Large number of fecal metabolites association with visceral fat supports the previous observation of influence of gut metabolic processes on abdominal adiposity.
• For 428 metabolites, heritable variance component explained modest average 17.9% of metabolite variation and long-chain fatty acid-containing metabolites were among the most heritable metabolites.
• Genome-wide association study found NAT2 gene associated with fecal metabolic trait.
• Gut microbial composition explained notable observed variance of 710 metabolites ranging from 22.1% -100%.

Points of Confusion

• Functional role of most significant metabolites is not discussed in detail with context to host metabolism. For example, fourteen significant fatty acid metabolites associated to visceral-fat.
• Significant 32 OTUs had poor taxonomy identification and most of them classified up to family level. Use of amplification fragment covering more variable regions would have improved identification.
• Amino acids availability could be associated to visceral-fat but limited taxonomy identification of OTUs may not support that observation when two OTUs belong to same family group has opposite association.