Skip to main content

The Gut’s Ecology

Research on the gut microbiome offers promises to the future of health care, even while it shatters widespread myths. Nursing Professor Charlene Compher and her Penn colleagues are at the forefront of discovery.

The human gut is a 28-foot-long ecosystem with a balance (or imbalance) of temperature, acidity, mucous membranes, and thousands of species of microbes. These bacteria bestow a vast diversity to each of us, comprising a genome 150 times greater than our own human code. Our microbiome helps digest our food, contributing to the release of vitamins and minerals, restoring our gut linings, crowding out the more harmful microbes and releasing antimicrobial chemicals to kill the most dangerous bacteria. Gut microbes affect the way we smell. They help our immune system detect dangers, affect the efficacy of drugs like acetaminophen, and may even influence our behavior. Changes in the microbiome have been associated with obesity, asthma, colon cancer, diabetes, and even autism.

But microbes don’t work alone. For that matter, the gut microbiome itself hardly works in isolation. The science is still unclear on how much a changing microbiome’s association with illnesses and weight loss are correlation, and how much are causation. Despite many claims on the Internet, probiotics are no magic cure. As with organisms in a garden or forest, microbes depend on what we feed them. While microbes can be introduced to a gut that lacks them, nutrition helps determine which species flourish and which ones diminish or even die out. Some microbes—including those associated with weight loss, prefer dietary fiber. Others, especially those microbes that often populate the guts of obese patients, feed on saturated fats and food additives such as CMC and P80.

These microbes produce the metabolites that help determine our health. And even here, diet plays a role. Professor of Nutrition Sciences Charlene Compher notes that gut microbes can convert common essential nutrients such as dietary choline and carnitine into trimethylamine. This gets oxidized in the liver to produce trimethylamine oxide, a chemical associated with increased risk of coronary vascular disease.

Dr. Compher and her colleagues are helping discover the ways that nutrition plays a role in the microbiome’s chemical production, and how changes in diet (and the care of infants) can affect our long-term health.

The following is the latest on what we know of our diet-dependent internal environment.


Of the thousands of species of microbes in our gut, fewer than 100 cause diseases. Yet taxonomy alone fails to measure the gut microbiome’s influence on a patient. Metabolites, the chemicals produced by microbes and processed by the liver and other organs, are the critical factor. Nutrition not only helps determine the constituents and variety of microbes in the gut; our diet also influences the production of metabolites.

Akkermansia muciniphila, a gut bacterium common in humans, is associated with weight loss and reduction of type 2 diabetes symptoms in animal studies. A. muciniphila also produces short chain fatty acids, which reduce inflammation. Plant carbohydrates such as inulin—found in many foods such as onions, garlic, and bananas—act as a fertilizer for this microbe. Inulin, along with oxalate, Leucaena, and other microbe promoters, can be given to patients in the form of prebiotics.

The Infant Gut

Full-term fetuses have functional immune systems that get suppressed by certain immune cells. This suppression prepares the gut and other organs for microbe colonization. Human milk provides an antibody bridge, helping control the growth of the baby’s microbiome.

Three-quarters of a newly born baby’s microbes come from the mother’s vagina.

Among the most important mother-acquired microbes in a newborn: Bifidobacterium, which helps digest milk.

As baby picks up more microbes from its environment, the gut gets colonized by Bacteroides and other microbes that digest carbohydrates.

Human milk contains many more multiples of milk oligosaccharides than cow’s milk. Human milk oligosaccharides (HMOs) seem to exist to feed the gut microbiome.

The microbe Bifidobacterium infantis releases short-chain fatty acids that feed gut cells. These cells in turn produce helpful metabolites such as adhesive proteins and anti-inflammatories.

The Microbiome in Health Care

Even short-term antibiotics can dramatically reduce gut flora diversity.
Clostridium difficile (C-diff) tends to affect patients in hospitals and other clinical settings where antibiotics are used widely. C-diff populates the gut where other species have been suppressed.

Probiotics may alleviate infectious diarrhea, and may reduce antibiotic-influenced diarrhea risk. They also have shown to reduce mortality rates in NEC (necrotizing enterocolitis). In NEC, however, human milk may produce the same or even greater benefits.

Unproven claims of probiotics: allergies, asthma, autism, diabetes, eczema, IBD, obesity.

FMT (fecal microbiota transplant) is administered many ways, including by colonoscopy, naso-enteric tubes, and capsules. It has been shown to help patients with IBD. Other claims—obesity, autoimmune diseases, irritable bowel syndrome—have not been proven.

Charlene Compher

Professor of Nutrition Sciences Charlene Compher has a senior role on a Penn-wide team  investigating the microbiome. We asked her to explain their work.

On Penn’s microbiome research:

Penn is a highly regarded leader in this science, both globally and nationally. It’s a new type of team science that includes highly trained experts with different skills working together. Our group includes a gastroenterologist and biochemist who specialize in basic science and animal models; adult and pediatric gastroenterologists who specialize in human trials; a microbiologist in whose laboratory microbiome samples are analyzed for the DNA content of bacteria;
and a statistician who specializes in analysis of high intensity data. As the group’s nutrition scientist, I lead the design, collection, and interpretation of dietary intake information in the overall context of gut microbiome data.

There are also doctoral and postdoctoral trainees involved in each of these processes. At the end of each project, we sit around a table to discuss and interpret the findings along with their biological and nutritional implications.

On the influence of diet on the microbiome:

Typical diet has a great influence on the microbiota. From around age 2, the microbiome remains stable throughout adulthood. The microbiota are really communities of bacteria that compete with each other for nutrients from the human host’s diet; thus typical diet has great influence on the composition of the stable adult microbiome. Studies in children from remote African villages show a different microbiome than European children. Yet, in one of our studies comparing vegans and omnivores, the microbiome was not very different, possibly because of other environmental factors.

I like to think of the relationship between humans and their microbiome as a symbiotic one. We provide a nice niche for the microbes to grow and obtain their own nutrient needs. They help us to break down dietary fiber to produce health-protective compounds; they produce some vitamins; and they help us activate our immune system. This latter role may be especially important for preventing allergy, inflammatory bowel disease, and other conditions.

On diet and metabolites:

Metabolites are the small chemicals that result when humans or bacteria use enzymes to break down (metabolize) nutrients. The metabolites are in many ways the end result of what we eat, after we’ve digested it and absorbed it into the blood. Our team compared the microbiomes of vegans and omnivores. The dietary intake of the two groups was radically different; the vegans ate less protein and fat, and much more fiber. Yet we were surprised to find that the gut microbiome in both groups was not totally dissimilar. However, there was a big difference in  the metabolites from these two groups.

A vegan diet contains more foods with micronutrients not contained in animal foods but that can be obtained when either the person or the microbiota break down the source foods using enzymes. These metabolites generally have positive effects on health.

On the complexity of their research:

While it sounds simple to establish linkages between diet and health, it’s actually incredibly complex. When we do these analyses, we use nearly 200 diet variables. We compare these to literally trillions of bacterial genes and compare them to 800 metabolites. There’s a huge data stream that requires complex statistical approaches. We’re in the early phases of using AI to help us understand these massive data streams in real time. You have to group all those data into patterns. It’s very exciting and great fun.