The next therapeutic frontier may be in your gut

Scientists are trying to harness the microbiome for healing

Kevin Pels

Chemical Biology

Dana-Farber Cancer Institute

Our microbiome helps us break down toxic compounds, produces vitamins that our cells can’t, and break down fiber from our diet. Symbiotic strains like B. thetaiotaomicron even induce our cells to produce antibiotic molecules that weed out disease-causing bacteria like Listeria. Imbalances in the community can damage intestinal tissue and lead to chronic inflammatory diseases such as ulcerative colitis and Crohn’s disease.

With such strong connections to overall health, it is no surprise that people want to gain more control over their gut ecology through probiotic supplements. Their labels tout claims of better health, though both the FTC and the European Commission have made efforts to limit this advertising. Suppressing the marketing of healthy microbes might seem nefarious, but it’s currently impossible to identify the exact benefits of any particular ensemble of bacteria. We don’t even know if these strains persist in our bodies or simply wash out. But lagging ever so slightly behind the probiotic craze, the time of prebiotics is ascendant.

The idea is to supplement our diet not with bacteria themselves, but with larger doses of the nutrients they love. Our gut bacteria eat after us, feasting on the remaining carbs after a meal exits our stomach. Starch already broke down in our mouth and stomach, and simple sugars are long gone, having been swiftly ushered into our bloodstream. But certain carbohydrates will make the digestive journey untouched – dietary fiber, the perfect fuel for our resident anaerobic microbes. In fact, bacterial genes for fiber metabolism are critical for colonization of the intestinal tract.

So prebiotics are just dietary fiber supplements. One example is inulin, a type of dietary fiber found in many common vegetables. Eating inulin can boost levels of gut symbionts Bifidobacteria and Lactobacilli; conversely, low-fiber diets are associated with a disrupted microbial ecology incorporating greater abundance of pathogenic species.

As natural laxatives, prebiotics have some benign but unseemly side effects that have perhaps limited their popularity as nutritional supplements. But they appear to be a key determinant of the composition and health of the human microbiome. This begs the question: can prebiotics (dietary fiber) be used to manipulate probiotics (bacteria) to favor the viability of non-pathogenic members of the gut microbiome? And more specifically, can bacteria be selected metabolically by ingesting a specific type of dietary fiber?

The Sonnenburg research group at Stanford has been pursuing questions like these for over a decade. They previously showed that adding inulin to the diet of germ-free mice enabled them to sustain Bacteroides colonies. A crucial first step, but humans are usually not germ-free, and any newly introduced bacterial species has to confront the current residents on arrival. Therefore they sought to do the same thing mice with three different pre-established microbiomes, two of which came from human donors. To create these humanized mice that serve as models of our own microbiomes, germ-free mice are inoculated with human fecal samples.

The introduced species is a genetically engineered green fluorescent strain of Bacteroides (NB001) that can metabolize an exotic type of dietary fiber called porphyran, which is uniquely abundant in seaweed. Roughly 10 mice for each of the three microbiota were given NB001; then, these mice were fed a low-fiber diet for seven days, followed by seven days of a diet enriched with either inulin or porphyran. Each day, the researchers dutifully collected the mice's fecal samples and cultured the bacteria within to observe how much NB001 was hanging around.

During the low-fiber period, NB001 was observed in two of the three microbiota tested, while it dropped below the limit of detection in the other. Once the mice began eating porphyran, their NB001 levels spiked by at least 10,000-fold across all three microbiota. Inulin could not produce the same effect. In two of the three microbiota, NB001 levels spiked and crashed by day four back to their previous low level, and NB001 were undetectable in the other. The impact of porphyran on NB001 growth was not blunted in the context of a fiber-rich diet, indicating that no other gut bacteria could metabolize the porphyran.

Further results showed that the genes responsible for porphyran metabolism were vital for NB001 to overcome the priority effects exerted by other entrenched Bacteroides strains in the gut. Mutating these genes abolished NB001's competency to colonize animals fed porphyran-supplemented diets. Likewise, transforming these genes into two other strains of Bacteroides allowed them to replicate the results observed with NB001. Finally, the researchers showed that the introduced bacteria's rate of growth was directly related to the amount of porphyran in the diet: thus, the prebiotic can be dosed like a drug.

Taken together, these data suggest that one day we could introduce genetically engineered bacterial species with desirable features into our microbiome and control their fate via a specific prebiotic. Just what those features are is limited only by scientists' imaginations and the speed at which they can manipulate those traits. This goes beyond probiotics and into the realm of bacteria as drugs to treat disease. For example, patients with inflammatory bowel diseases could colonize themselves with bacteria that produce potent anti-inflammatory compounds.

Even if therapeutic bacteria can be engineered to treat diseases, porphyran might not be the best prebiotic to pair with a probiotic drug. The few human gut microbes that can metabolize porphyran are believed to have acquired the trait via horizontal gene transfer from the marine bacteria present on the dried seaweed, which forms an important component of the Japanese diet. So while the rarity of porphyran metabolism in most populations may provide a solid niche for engraftment, it may be weaker in people that frequently consume seaweed.

The same could limits could apply to any dietary fiber. How then to make a strain of bacteria that could rapidly colonize any person regardless of their diet? The answer may not be to search for a more unique fiber than porphyran, but to invent one. Perhaps in the not-so-distant future, researchers will create abiotic carbohydrates and evolve enzymes that degrade them. Synthetic fiber-bacteria pairs will allow us to rapidly engraft ourselves with therapeutic microbes free from metabolic competition with other gut species.

This idea isn’t without risk. Just as the genes for porphyran metabolism migrated into the gut microbiota, genes from engineered bacteria could escape into other species in the microbiome and lead to unknown consequences. Most companies developing therapeutic bacteria are deliberately designing them to not persist in the microbiome to avoid this dilemma. But as the mechanisms that control our bacterial ecology become more evident, those companies may shift their focus to long-term colonization to treat chronic disease.

Finally, the methods from this research could also empower us to determine which probiotic supplements actually provide beneficial health effects. Supplement companies have long capitalized on the hype surrounding the healthful applications of gut microbiome research, but now cherished strains can be empirically tested in humanized mice. We can ensure that the bacteria from probiotic sources stay in our guts after we consume them: the key is to feed them their favorite carbs. After all, you are what they eat.