Dr. Justin Sonnenburg on Building and Repairing Gut Health: Fiber, Fermented Foods and the Microbiome

High-Level Overview

Andrew Huberman sits down with Dr. Justin Sonnenburg, Professor of Microbiology and Immunology at Stanford, who runs his lab with his wife, Dr. Erica Sonnenburg, and co-wrote the book The Good Gut with her. Huberman calls himself a true novice on the microbiome, and the conversation builds from first principles: what the gut microbiome is, where the microbes live along the digestive tract, how an infant acquires them, and why industrialized populations carry very different communities from traditional ones. From there it moves to practical questions about fasting, cleanses, processed foods, sweeteners, probiotics and prebiotics, and centers on Sonnenburg's study with Christopher Gardner's group comparing a high-fiber diet with a high-fermented-food diet.

The tone is curious, friendly and carefully evidence-based. Sonnenburg repeatedly separates what is known from what has not yet been studied, and both men talk about their own eating habits. The discussion flows from basic biology to experimental findings and then to everyday choices anyone can make on almost any budget.

The Discussion Explores:

• What the microbiome is: a dense ecosystem of trillions of bacteria, archaea, fungi and viruses concentrated in the colon.
• How we acquire microbes: birth mode, breastfeeding, pets, antibiotics, and the lasting effects of early colonization.
• Traditional versus industrialized microbiomes: whether the modern Western microbiome is healthy or "gone off the rails."
• Resilience and extinction: why microbiomes snap back, and a multi-generation mouse study showing permanent species loss.
• Fasting, cleanses and processed food: what the evidence does and does not say.
• The fiber versus fermented food study: why fermented foods increased diversity and lowered inflammation.
• How gut microbes talk to the body: immune sampling, short-chain fatty acids, and molecules that reach the brain.
• Probiotics, prebiotics and practical tools: buyer beware, whole plants over purified fiber, and homemade fermented foods.

Deep Dive into Main Ideas

What the Gut Microbiome Is and Where It Lives

Sonnenburg explains that "microbiome" and "microbiota" are often used interchangeably for our microbial communities, which live anywhere the environment touches the body: nose, mouth, skin, and especially the digestive tract. The vast majority live in the distal gut and colon, and the density is astonishing: roughly 30 to 50% of fecal matter is microbes. There are trillions of cells from hundreds to around a thousand species. Most are bacteria, but there are also archaea, eukaryotes such as fungi, and bacteriophages, viruses that infect bacteria and outnumber them about 10 to 1, creating predator-prey dynamics. He mentions the view some hold that humans are, in effect, an elaborate culturing flask designed by evolution to pass microbes from generation to generation.

The communities differ along the tract. The mouth's microbes handle oxygen and grow in stationary mats on teeth. The stomach's sparse community includes Helicobacter pylori, which can cause ulcers and occasionally gastric cancer. The small intestine is still something of a black box, now being explored with sampling capsules. The colon is the densest and best-studied community, partly because stool is easy to collect and fairly representative of it. Along the way, pH, bile, bicarbonate, nutrient availability and immune activity all change, so different microbes thrive in different places.

Huberman asks why microbes aren't simply flushed out. The answer lies in the gradient from the gut wall inward. The gut is lined with mucus, a carbohydrate mesh that keeps bacteria out while letting nutrients through, and it turns over more slowly than the passing contents. Microbes that can cling to mucus, and some that eat it, such as Akkermansia muciniphila, resist washout. Some reach the crypts, folds where the stem cells that renew the lining live. Sonnenburg's work suggests that if a microbe can localize to a crypt, "you've hit the jackpot" and can exclude similar newcomers.

How We Acquire Our Microbes and Why Early Life Matters

The womb appears to be essentially sterile, Sonnenburg says, so each newborn is like an island rising out of the ocean, with a land rush for open territory. The first year is a very dynamic period of assembly, with further shifts when children move to solid food at two to three years of age. Birth mode has a striking effect: babies born by C-section have gut microbes resembling human skin, while vaginally born babies are colonized by their mother's vaginal and stool microbes. Breastfeeding versus formula, pets in the home and antibiotic exposure all push development down different trajectories, and animal studies show early microbes can steer the immune system and metabolism in different directions.

Surprisingly, only part of an infant's microbiome comes from the mother. Much comes from surfaces, caregivers and even strangers over the first year. Pets may contribute directly (Sonnenburg notes his dog, Louis Pasteur, occasionally licks his mouth) and indirectly by bringing in environmental microbes from outside.

What Is a Healthy Microbiome? Traditional Versus Industrialized

Sonnenburg calls this the "million dollar question" in the field. Context matters, the microbiome is plastic, and many configurations are compatible with health. The NIH-funded Human Microbiome Project, which began around 2008 and 2009, set out to define healthy versus diseased microbiomes. The collective genome of our gut microbes has 100 to 500 times more genes than the human genome. The project revealed enormous individual variation. At the same time, studies of hunter-gatherers and rural agricultural populations showed healthy people whose microbiomes looked nothing like a healthy American's.

Quoting Dobzhansky, "nothing in biology makes sense except in the light of evolution", Sonnenburg notes that these traditional microbiomes, confirmed by sequencing ancient feces, better represent the microbes humans evolved with. One possibility is that the industrialized microbiome is adapted to modern life. Another is that it has deteriorated under antibiotics and a Western diet, and that what the Human Microbiome Project documented as "healthy" is actually a perturbed state predisposing people to inflammatory and metabolic disease. Huberman compares this to neuroscience's discovery that standard lab cages were deprived environments.

On whether a poor microbiome can be fixed, Sonnenburg describes "reprogramming" as more feasible than editing the human genome, but notes that microbiomes tend to sit in stable states, like a well with gravity. After antibiotics or a dietary change, communities often snap back close to where they were. In a key mouse experiment, switching mice to a low-fiber diet reduced diversity, which returned when fiber came back. But keeping mice on a low-fiber diet for four generations left only about 30% of species, and restoring fiber no longer helped, because the microbes were gone. A fecal transplant from mice kept on high fiber for four generations restored the lost diversity. Lasting change, he concludes, requires both access to the right microbes and the right diet to feed them.

Fasting, Cleanses and the Simplest Healthy Diet

Sonnenburg says there is little high-quality science on fasting or cleanses. Fasting and diets like ketogenic eating can help people with metabolic syndrome, but in hibernating animals and long fasts, mucus-eating bacteria bloom, and in mice deprived of fiber those mucus eaters get too close to host tissue and inflammatory markers rise. Short-term fasting may be fine, but the long-term effects on the microbiome are unknown. Cleanses and flushes he personally considers a terrible idea: wiping out the resident community without a plan for what replaces it is "a little bit like playing Russian roulette." He can imagine a future clinic that deliberately flushes and repopulates the gut with a known healthy consortium, but today it seems half-baked.

He cuts through the complexity with advice he heard Christopher Gardner give a dietitian. When she listed plant-based fiber as rule number one and moved on, Gardner told her: "if people do number one well, you don't need to know any other rules." If you fill up on whole grains, legumes, vegetables and high-fiber fruit, there is little room left for too much meat or sweets. It echoes Michael Pollan's line, "eat food not too much mostly plants." Huberman adds that all-or-none rules are often easier for people to follow than nuanced ones, which may explain the popularity of fasting and elimination diets.

On low-carb diets, Sonnenburg's hypothesis is that people feel better cutting carbohydrates because the carbs in industrialized diets are mostly simple starches and sugars that spike blood glucose. He prefers the term "microbiota accessible carbohydrates" for complex fibers we cannot digest but gut microbes can ferment, to stress that not all carbs are bad. Genetics and culture matter too: lactase persistence evolved in some populations within about 10,000 years, and many people in Southeast Asia carry gut microbes with genes for digesting porphyran in seaweed. Still, he notes the Hadza hunter-gatherers he studies eat an estimated 100 to 150 grams of fiber a day, versus about 15 grams for a typical American. They actually prefer meat and honey, but eat plants because that is what is reliably available; only about one in 20 to 30 hunts lands big game.

Processed Foods, Sweeteners and Emulsifiers

Both men agree that avoiding heavily processed food is good for the microbiome across every dietary philosophy, and Sonnenburg says this is compatible with Gardner's rule. Fiber matters because gut microbes ferment it into short-chain fatty acids such as butyrate, which fuel colon cells, strengthen the gut barrier, keep inflammation low, and help regulate the immune system and metabolism. Simple carbs spike blood sugar, while complex fibers keep it low and generate these beneficial compounds. He notes ketogenic and high-fat diets may bring short-term metabolic benefits but that there is also literature linking them to heart disease.

Processed foods add other problems. Sonnenburg cites work from the Weizmann Institute showing artificial sweeteners can harm the microbiome and push toward metabolic syndrome, and says many emulsifiers used for shelf stability disrupt the mucus layer, which can lead to inflammation. Huberman mentions research from Diego Bohórquez's lab at Duke showing gut neuropod cells distinguish real sugar from artificial sweeteners and signal the brain differently. On plant-based sweeteners like stevia and monk fruit, Sonnenburg says very little is known; they may be less harmful because so little is needed, but the studies have not been done.

Personally, Sonnenburg avoids artificial sweeteners but is a flexible eater and does not think an occasional diet soda will cascade into disease. His bigger lesson is to change slowly: over five or more years he weaned himself off an afternoon muffin or cookie habit until sweet foods became unpalatable. Huberman describes losing his own taste for sugar: "I like sweet people. I don't like sweet food anymore."

The High-Fiber Versus High-Fermented-Food Study

Sonnenburg explains the motivation. Western medicine was built for infectious disease, waiting for illness and then treating it, which works poorly for chronic inflammatory diseases like metabolic syndrome, autoimmune disease and heart disease. A body of research showed the microbiome sets the immune system's tone, raising the possibility that depleted modern microbiomes keep people in a state of simmering inflammation. Diet studies could start in humans right away without drug approval. The team partnered with Christopher Gardner's group, experts in human diet studies, and Stanford's Human Immune Monitoring Center, run by Mark Davis and Holden Maecker, which can track hundreds to thousands of immune parameters from a blood draw. The study was funded by philanthropy.

One group raised plant fiber from roughly 15 to 20 grams a day to over 40 grams through whole grains, legumes, vegetables and nuts, which also led them to eat less animal protein and saturated fat. The other ate naturally fermented foods with live microbes: unsweetened yogurt, kefir, sauerkraut, kimchi, brine-fermented vegetables and kombucha. Sonnenburg stresses that shelf-stable pickles and canned sauerkraut are usually made with vinegar or have had their microbes killed; you need refrigerated products, and you should avoid sugar-loaded versions. Store-bought beer, being filtered and canned, did not count. A dietitian, Dahlia Perelman, guided participants, who ramped up over four weeks and then held for six weeks, reaching over six servings of fermented food a day on average.

The results surprised the researchers. They expected fiber to produce the big signal, and it was Gardner who had to twist their arms to include the fermented food arm. Yet the fermented food group showed increased microbiome diversity over the six weeks and a stepwise drop in a couple dozen inflammatory markers, including interleukin-6 and interleukin-12, with less activated signaling inside immune cells. In the washout phase, as people ate less of these foods, diversity began to plateau and reverse, suggesting maintenance is needed. The fiber group responded individually: participants who started with the most diverse microbiomes were most likely to see inflammation fall, while others had mixed responses. Sonnenburg links this to the mouse extinction experiment and to a University of Minnesota study showing immigrants to the US lose microbiome diversity and fiber-degrading capacity within months to years. Many industrialized people may simply lack the microbes to use the fiber. A new study will test fiber and fermented foods together.

Both groups reported better bowel habits, but formal questionnaires and a cognitive test did not show significant changes in mood or cognition, although Sonnenburg hears many anecdotes of more energy, clearer thinking, better sleep and better skin and thinks lower inflammation could plausibly explain them.

Gradual Ramps, Hygiene and Environmental Microbes

Both men stress ramping up slowly to avoid bloating and discomfort as the microbiome reconfigures. Huberman describes drinking sauerkraut brine after exercise, which he built up to gradually, and warns of the time he drank a 12-ounce bottle of a similar product that was meant to be 24 servings. Sonnenburg notes that a brine product called Gut Shots was among the foods used in the study and that his lab is now studying brine in mice because it is rich in lactate and other metabolites. For those on a budget, Huberman points to the sauerkraut recipe in Tim Ferriss's The 4-Hour Chef (cabbage, water and salt, following the protocol carefully), and Sonnenburg explains how he keeps homemade kombucha going with a SCOBY, a symbiotic culture of bacteria and yeast.

On hygiene, Sonnenburg says handwashing matters, especially during a pandemic, but that sanitization has gone overboard, with antimicrobials built into shopping carts and toothbrushes. When raising his daughters, he weighed context: a bit of garden dirt before lunch was fine, while a public playground, grocery store or subway called for washing. Exposure to environmental microbes likely helps educate the immune system, and fermented foods may tap into the same pathways safely.

How Gut Microbes Communicate With the Body and Brain

Most of the body's immune cells are in the gut, because microbes that are beneficial in the right place can become opportunistic pathogens if they get out of place. Specialized structures called Peyer's patches deliberately take up microbes so immune cells can learn what is on the other side of the barrier, which Huberman likens to "border patrol", and dendritic cells reach into the gut to sample microbes. Gut-lining cells carry receptors for microbial patterns like lipopolysaccharide, and cells similar to taste cells sense microbial metabolites. The enteric nervous system signals the brain and controls motility, and educated immune cells can travel to other mucosal surfaces.

Microbial molecules also enter the bloodstream directly: short-chain fatty acids, plus compounds like indole and phenol derivatives from amino acid metabolism, which the body further modifies. Some of these metabolites are found in cerebrospinal fluid, so they do reach the brain. Many are excreted in urine, and people with kidney disease accumulate them in the blood, which may contribute to the brain fog common in kidney disease.

Probiotics and Prebiotics

Sonnenburg's first advice on probiotics is buyer beware. The supplement market is largely unregulated, studies of off-the-shelf products have found contents that do not match their labels, and some companies prey on people's fears. Look for independent validation or well-established brands. Evidence is mixed: some studies suggest probiotics slow recovery of the gut's microbiota after antibiotics, though meta-analyses suggest they may help prevent diarrheal disease. His recommended approach is to find a well-designed study showing a specific probiotic achieves what you want, ideally with a doctor's help, and stick with that product, or experiment carefully to see what works for you.

Prebiotics, meaning purified fibers, are also a mixed bag. Some studies show diversity falls as a few bacteria that specialize in one fiber bloom. A mouse study found that highly fermentable purified fiber on top of a Western diet led some mice to develop liver cancer, though whether this applies to humans is unknown. Whole plants ferment slowly along the colon, while purified soluble fiber can produce a rapid burst. The broad view in the field, he says, is that a diverse variety of plants beats purified fibers for fostering microbial diversity, though some people do benefit from specific fiber supplements.

Resources and the Good Gut

Sonnenburg and his wife wrote The Good Gut after noticing that even well-informed scientist friends were not making the changes they were. He tells a story from a conference center that hosts a different scientific meeting every week: during the microbiome conference, the dining staff asked what was going on because "we can't keep the salad bar stocked." The Stanford Center for Human Microbiome Studies and the Sonnenburg lab website list ongoing studies, and they are always looking for participants.

Conclusion

Sonnenburg's message is that the gut microbiome is a dense, malleable ecosystem shaped from birth by how we are born, what we are exposed to and above all what we eat, and that industrialized life may have depleted it in ways that keep inflammation simmering. The most practical takeaways are simple and cheap: eat mostly whole plants with plenty of fiber, avoid heavily processed foods with added sugars, artificial sweeteners and emulsifiers, add live fermented foods such as plain yogurt, kefir, sauerkraut, kimchi and kombucha, ramp up gradually, and be skeptical of cleanses and unverified probiotics. His own study found that fermented foods increased microbiome diversity and lowered inflammatory markers within weeks, while fiber helped most those who already had diverse microbes, suggesting the two together may be the best approach.

This is a summary of "How to Build, Maintain & Repair Gut Health | Dr. Justin Sonnenburg" from Huberman Lab, giving you the GIST so you can decide if you want the REST. Listen to the full episode.