prebiotics in combination with probiotics also induces differential metabolic responses (Martin et al., 2008b).
Another body of evidence indicating that the microbiome plays a key role in human metabolism comes from research on bariatric surgery in both animal models and humans. Roux-en-Y gastric bypass (RYGB), the gold standard for bariatric surgery, has been associated with an 80 percent reduction in diabetes within 24 hours of surgery. The procedure has also been associated with reduced risks of colonic and other cancers. Nicholson explained that because the diabetes is cured immediately (i.e., not after subsequent weight loss), there must be a biochemical explanation. Part of that explanation likely lies in the microbiome. Zhang et al. (2009) reported a massive increase in Gammaproteobacteria in RYGB patients, compared to normal and obese individuals. Using a rat model, Li et al. (2011b) also reported an increase in Gammaproteobacteria as well as a massive change in bile acid metabolism following RYGB. Other microbiome changes have been detected in RYGB rats as well. Nicholson observed that while the microbiome changes may not be “the key” to understanding the connection between bariatric surgery and changes in diabetes, cancer, or other disease risks, “they are certainly part of the gear box” (Holmes et al., 2011). One possible mechanism is the cytotoxic environment created in the gut following bariatric surgery, as evidenced by fecal extract toxicity (Li et al., 2011a).
Gut Microbial Activities Affect Drug Processing in the Host
One of the goals of a systems-level understanding of human bio-complexity is to realize the vision of personalized or, as Nicholson called it, “precision” health care (Mirnezami et al., 2012). Pharmacometabonomics is one component of that care, in Nicholson’s opinion. He defined pharmacometabonomics as “the prediction of the quantitative outcome or effect of a biomedical intervention based on a pretreatment metabolic model.” The approach is predicated on the concept of “metabolic hyperspace,” where the position of an individual is dependent on a multitude of factors (genes, diet, microbiome) (Nicholson and Wilson, 2003). The closer two individuals are in metabolic hyperspace, the more physiologically similar they are and the more likely they are to behave in the same way when presented with a challenge (e.g., a drug or other therapeutic intervention). As an example of a potential pharmacometabonomic application, Clayton et al. (2006) demonstrated that drug toxicity could be predicted based on pre-intervention metabolic profiles of urine. Other research groups have reported similar findings (Phapale et al., 2010).
In addition to drug toxicity, pre-intervention metabolic profiling has also been used to predict drug metabolism. For example, Clayton et al. (2009) demonstrated an association between gut microbial metabolites and