Combining metabolomics with genetic analysis is, Choo and Kanno et al argue, a richer way to predict a phenotypic outcome when it comes to the complex universe of inputs and outputs in our densely populated gut.

By mass, 99% of the DNA in and on a person is bacterial – only 1 % is human.

The gut microbiome, a word to describe the bacteria living in your intestines, is defined by its symbiosis, interconnectivity and redundancy. If you take away one species, another might fill the gap in function.  Wipe out a different species and this might have devastating effects down the chain.

Researchers from King’s College London set out to decode this necessary complexity by combining genetics with metabolomics: looking at not just which bacteria are present, but what they consume and produce, too.

Recently, we’ve started to appreciate the far reaching power of the microbiome. Health issues we thought couldn’t be connected to the gut, we now see are linked to the status of a pulsing, flourishing ecosystem of bacterial cells. By looking at the inputs and outputs of such a system, you can get a much more predictive picture of how your gut bacteria work to cause a particular state of health or un-health.

To prove the concept, the researchers treated identical lab mice with Vancomycin-imipenem or Ciprofloxacin, common antibiotics that are proven to cut swathes through different parts of the microbiome. And that are also known to have off target effects on our physiology.

Simple correlation to taxons can be done, and the researchers find that 14 metabolites could be correlated with decreased of increased populations of given bacteria when patients were treated with either of the two antibiotics.

An example of that is arabinitol, an undigestible. This substance is broken down and used as fuel by certain bacteria. Vancomycin-Imipenem antibiotics wiped out these populations, resulting in an abundance of the arabinitol, that the researchers quantified by NMR. This example is important, as arabinitol prevalence provides a fertile niche for harmful bacterial populations like C. Dificile. Here the experimenters recorded an increase in Enterobacteriaceae, which are linked with gut inflammation.

The second case observed was where the microbial population recovers after initial obliteration with ciprofloxacin, but the metabolites produced don’t recover. This just hints at the complexity of the system, and points to a longer term effect on gut health after the patient has stopped taking antibiotics. Cipro also lent itself to increased levels of valine, leucine, and phenylalanine, a phenotype linked to the development of type 2 diabetes.

Research has been hampered by an inability to pick apart the complexity of the seething, living cornucopia of interdependency we have inside us. Often functions of bacteria overlap, and a healthy microbiome differs wildly – even between genetically identical lab mice. So looking at the taxa – the identities of the bacteria present, should be coupled with looking at the outcome: the metabolites themselves.

Research article here.

Image from Elizabeth Pennisi in Science magazine

Written by Michael Shannon