Today was a busy day but, at the end of the day, the one talk that stands out in my mind is Ruth Ley’s talk on the human microbiome. Personally, I find this topic to be extremely interesting and Ruth’s presentation did not disappoint. The talk began with an introduction to a phenomenon in human history: the introduction of agriculture. Now, we can all imagine that during the shift from hunting and gathering to farming the human microbiome changed to adjust to the new diet. Perhaps one of the most important changes was the introduction of starch heavy foods that were not the basis of hunter-gather diets. One gene in particular, AMY1, codes for amylase, an enzyme responsible for starch digestion in humans. So, from a microbiome standpoint where one is continually trying to find correlations between gut microbiota and genotype/phenotype/diet etc, the natural question becomes whether or not the microbiomes of hunter-gathers differ from the microbiomes of agriculturalists.
Previous work by Perry et al. (2007) reported a significant correlation between the number of AMY1 copies a person has and the amylase concentration in their saliva. Ruth wanted to follow-up this question with a study on whether or not the microbiomes of people with low AMY1 copy numbers differed from those with high AMY1 copy numbers (CN).
In order to undertake such a project, Ruth’s group recruited over 100 Cornell undergraduates for AMY1 screening. From this batch, they selected 12 students with low AMY1 copy numbers and 13 students with high AMY1 copy numbers that were given the same diet and sampled over a number of weeks. From this time series dataset, Ruth’s group screened the samples for 16S and used linear discriminant analysis in order to determine what organisms separated those with low AMY1 CN to those with high AMY1 CN. Perhaps as expected, Ruth’s group found that there were more organisms separating low and high AMY1 CN microbiomes at time zero (n=49) than at the end of the study (n=16) with several of the distinguishing species capable of amylase activity. However, an even more intriguing experiment was performed next…
As exhibited by the amylase story, an over-arching theme of Ruth’s talk was whether or not the a person’s genotype is correlated to the microbiome. Following this line, they next wanted to ask whether or not the microbiome can influence phenotype. To test this hypothesis, Ruth’s group performed a study on ~500 twins to identify what are the most inheritable taxa using the ACE model of heritability. The Firmicutes, Christensenella, was identified to be the most highly heritable taxon. Christensenella was also found to be depleted in obese subjects. To test whether or not this microorganism influences phenotype, they inoculated germ free mice with microbiomes containing Christensenella and found that the mice enriched with Christensenella did not experience weight gain like the mice lacking Christensenella. Looks like you may really be what you eat!