used in the 16th century to diagnose and treat disease (based on the color, smell, and taste of urine). Today, scientists use advanced metabolic profiling tools, namely nuclear magnetic resonance (NMR) spectroscopy and mass spectrometry, that yield a tremendous amount of complex data—a single run generates data on hundreds or thousands of molecules. Researchers use various mathematical modeling tools to extract and convert relevant data into biologically useful information (e.g., information that can be used to identify “normal” versus “abnormal” metabolism). The complexity of the information generated by advanced metabolic profiles is due to the fact that not only are all human cells producing metabolites (with more than 500 functionally distinct cell types), but so too are all microbial cells (Nicholson et al., 2005). Microbes produce short-chain fatty acids, bile acids and related oxysterols, vasoactive (aromatic) amines, cresols and aromatic acids, endocannabinoids, and other molecules.

Many microbial metabolites participate in human metabolism in what Nicholson referred to as “combinatorial metabolism.” For example, bile acids, which are critically important host signaling molecules, are co-metabolized by microbes, with significant implications for liver and colonic disease risk (Nicholson and Wilson, 2003). Bile acids are synthesized in the liver on a daily basis and then secreted into the mammalian gut, where they are deconjugated into cholic acid by Lactobacillus and other gut microbiota. The cholic acid, in turn, can be dehydroxylated by yet other microbes into deoxycholic acid. Deoxycholic acid is both hepatoxic and carcinogenic. Microbial co-metabolism of bile acids also impacts lipid bioavailability.

Modifying Host-Microbiome Metabolic Interactions: Mouse and Rat Models

When thinking about how the gut microbiome impacts human metabolism, the tendency is to think about distal colon processing and the production of short-chain fatty acids, but the microbiome plays an important role in the upper gut as well—for example, with lipid bioavailability. Investigators have used both mouse and rat models to study bile acid and other host-microbiome metabolic interactions related to lipid absorption. For example, Martin and colleagues (2007) measured bile acid signaling after transferring human baby microbiomes into gnotobiotic (germ-free) mice and reported an increased emulsification potential and greater lipid bioavailability in the humanized mice compared to the normal mice. Studies with conventional versus germ-free rats have yielded similar findings (Swann et al., 2011).

Also using the mouse model, investigators have demonstrated that introducing probiotics, such as Lactobacillus paracasei or L. rhamnosus, can induce differential metabolic responses (Martin et al., 2008a). Introducing



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement