What Your Gut Microbes Actually Need
The carbohydrate piece on this site answered a viewer who said carbs are not essential. He was right about glucose. He was wrong about the organisms that live past the small intestine and eat what our enzymes leave behind. This is the next chapter. Not another argument about bread. A closer look at the community those leftover plant parts actually feed.
I am not selling a gut reset. I am writing down what the better studies show, what we do in the kitchen here in Oswego County, and where the research still stops.
Where the skeptics are right
A lot of microbiome talk is ahead of the evidence.
There is no species list you are supposed to match. Two people can eat well and carry different communities that still do the same jobs: ferment plant carbohydrates, make short-chain fatty acids, and hold a mucus barrier. Stool tests sold online are snapshots. They move with the last few meals. They are not a diagnosis.
The old line that bacteria outnumber our cells ten to one is outdated. A careful recount for a 70-kilogram reference man put the two populations in the same ballpark: about 38 trillion bacteria and 30 trillion human cells, a ratio near 1.3 to 1, and a bacterial mass of about 200 grams. Most of those bacteria are in the colon, not spread through the whole body. The ratio moves with body size, and the authors note that one bowel movement can swing it. Treat the figures as an order of magnitude for that reference man, not a lab value with your name on it. [(1)](https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1002533)
Probiotic capsules are not a substitute for the plants you keep eating. Most strains pass through. They have a place in a few studied situations. They do not rebuild a thinned-out community, and a shelf of unstudied blends will not put back strains that are already gone.
Fecal microbiota transplant has a real, narrow medical use: repeat *Clostridioides difficile* infection, in clinical care, after the usual antibiotics have failed to hold. CDC describes transplant as something tried for reinfection, not as the first treatment and not as a wellness reset. It is not a home remedy. [(2)](https://www.cdc.gov/c-diff/about/index.html)
If someone promised you a 21-day protocol that would give you a Hadza microbiome, they oversold it.
What is actually living there
The gut microbiome is the bacteria, archaea, viruses, and fungi in the digestive tract, densest in the large intestine. Diet is the lever we actually pull every day. A short feeding study showed the community can shift within days when the whole menu changes. [(3)](https://www.nature.com/articles/nature12820) Antibiotics are a stronger lever. They are not an everyday one.
Our own genome encodes only a small number of enzymes for food carbohydrates, mostly starch, sucrose, and lactose. The microbes encode thousands. [(4)](https://www.nature.com/articles/nrmicro3050) That is why fiber is not a side nutrient. It is carbon for the organisms that live past where our enzymes quit.
Researchers call the fermentable fraction microbiota-accessible carbohydrates: fiber, resistant starch, and some oligosaccharides that reach the colon intact. [(5)](https://pubmed.ncbi.nlm.nih.gov/25156449/) Bacteria ferment them into acetate, propionate, and butyrate. Butyrate is the preferred fuel of the cells lining the colon. Most of it is used in the gut wall. Do not expect it to circulate as a miracle molecule, and do not treat it as the only short-chain fatty acid that matters. Acetate is more abundant. Human studies of what butyrate does beyond the colon are still thin. [(6)](https://pubmed.ncbi.nlm.nih.gov/17973645/)
When those carbohydrates disappear, the cleanest mechanistic work we have is still in mice carrying human-derived microbes:
- Given little or no fiber, those bacteria turn to the mucus that protects the colon wall, and a gut pathogen gets through more easily. [(7)](https://pmc.ncbi.nlm.nih.gov/articles/PMC5131798/)
- Across generations on a low-fiber diet, diversity falls. In a single generation, much of that loss comes back when the carbohydrates return. Over several generations it does not, unless the missing microbes are restored as well. [(8)](https://www.nature.com/articles/nature16504)
People are not those mice. A human feeding study still shows how fast the census can move, and how far not to push the lesson. Ten adults ate an all-animal menu for five days and an all-plant menu for five days. The animal menu lowered the microbes that break down plant polysaccharides and raised bile-tolerant organisms. The plant menu moved the community less. About two days after the animal menu stopped, the community headed back toward where it started. That is five days in ten people, not proof that one meal rewrites you. [(3)](https://www.nature.com/articles/nature12820)
Fiber, plants, and fermented foods are not the same lever
Three human findings belong on one page. They do not say the same thing, and the carbohydrate article already used two of them for a different question. Here the point is what each one can and cannot claim.
| Study | Design | What it actually showed |
|---|---|---|
| McDonald et al., mSystems, 2018 — American Gut | Citizen-science stool samples plus diet questionnaires | People who reported more than 30 different plants a week had more diverse gut communities than people who reported 10 or fewer. A vegan or omnivore label explained less than the plant count. Association only, in a self-selected group. Not a disease trial. (9) |
| Wastyk, Sonnenburg, Gardner et al., Cell, 2021 | Stanford feeding study, 18 healthy adults in each arm, 10 weeks on the assigned diet | Fermented foods raised diversity and lowered a set of inflammatory proteins in blood. High fiber raised carbohydrate-degrading enzymes and did not raise diversity. The pre-specified cytokine response score did not change in either arm. In the fiber arm, as intake rose, carbohydrate measured in stool rose too. The authors read that as a community not fully fermenting the extra load. (10) |
| Reynolds et al., The Lancet, 2019 | 185 prospective studies and 58 trials on carbohydrate quality | Higher fiber intake was linked to lower all-cause and cardiovascular mortality and to less coronary heart disease, stroke, type 2 diabetes, and colorectal cancer. The observational contrasts were on the order of 15–30% lower risk, clearest around 25–29 g/day, with the dose-response suggesting more benefit above 30 g. Trials in the same paper are smaller and mostly about risk factors, not deaths. (11) |
Most U.S. adults are not near that range. National survey means have sat around 16 grams a day, and only about 5% of people meet the recommendation. Official targets, the Institute of Medicine figures still behind U.S. guidance, run from 19 to 38 grams a day by age and sex. [(12)](https://pmc.ncbi.nlm.nih.gov/articles/PMC6124841/)
The Stanford trial is the one I come back to, because it separates two levers people keep mashing into one. Fiber is still the fuel. If the microbes that burn that fuel have thinned out, ten weeks of more oats may show up as extra enzymes and extra carbohydrate in the stool, not as a diversity win. Fermented foods — yogurt with live cultures, kefir, sauerkraut, kimchi, other fermented vegetables — were the arm that moved diversity in that window. The practical read is both, started gradually. It is not a contest, and it is not six servings a day as a rule for the grocery run. That was the research dose.
What we do here
We eat a whole-food plant-based pattern at the Sanctuary. That is our choice, not a rule I would write onto someone else's plate. What I would put on almost any plate is plainer: more intact plants, more kinds of plants, and some fermented food if it sits well.
In this kitchen that looks ordinary. Beans and lentils. Onions, garlic, and leeks from the garden. Oats. Apples and berries with the skins. Potatoes and rice cooked, cooled, and eaten later so more of the starch gets past the small intestine. Sauerkraut when we have a crock going. Those are the plants we already grow and cook. They are not a protocol.
The same idea sits in the compost and the Johnson-Su pile. A mix of residues feeds soil life. A mix of plants feeds gut life. I do not claim the same species live in the pile and in the colon. I do claim we should stop treating biology as something that only happens in a capsule.
Antibiotics save lives. They also thin the gut community, and that thinning can last months. CDC notes that the weeks after a course are exactly when *C. diff* is more likely if you meet the spores. Use an antibiotic when a clinician has a reason. Do not use one as a default for a viral cold, and do not expect a drugstore probiotic to put back what a broad-spectrum course took out. Recovery is the food you go back to, plus time. Sometimes a long time. [(2)](https://www.cdc.gov/c-diff/about/index.html)
Honest caveats
- The cell-count paper is a reference man, 20 to 30 years old, 70 kilograms. Women, children, and heavier adults land in a similar range, not on the same number. Uncertainty on the bacterial count is wide.
- Mouse mucus and multi-generation studies are mechanisms. A low-fiber week in a person is not a permanent extinction. The mouse work says the loss compounds when the deprivation runs across generations, and that fiber alone may not restore what was lost.
- American Gut volunteers selected themselves, paid to participate, and mailed swabs. "Thirty plants" is a useful target from an association. It is not a magic number from a randomized trial, and it is not a promise about disease.
- The Stanford study was small and short. Two people in the assignment were not randomized. The headline immune score did not move. Six servings of fermented food a day is a research dose, not a supermarket obligation. Less, if you tolerate it, is still the direction the trial pointed.
- Diversity is a proxy. Jobs — making butyrate, holding the mucus barrier, not letting every new microbe through — matter more than a pretty species chart.
- People with active inflammatory bowel disease, short bowel, or a history of small-intestinal bacterial overgrowth need a clinician, not a fiber dare. Increase slowly. Cooked plants first if raw salads punish you.
- Low-carbohydrate patterns can improve short-term metabolic markers for some people. If that is your tool, the colon still has to be fed with something your own enzymes will not take first.
- Butyrate reviews are stronger on cells and animal models than on long human trials. Keep the claim at the colon lining.
Where we land
You do not need a perfect microbiome. You need a fed one.
1. Eat plants you can name. Beans, intact grains, vegetables, fruit, nuts, seeds. Count kinds across the week. A scoop of one fiber powder is not thirty plants.
2. Move fiber up from where you are, not from a slogan. If you are near the national average, around 16 grams, add one serving of legumes or one piece of fruit with the skin for a week, then another. Gas that fades is adaptation. Pain that does not is a reason to back up and talk to a doctor.
3. Put a fermented food on the table most days if you tolerate it. A few forkfuls of sauerkraut or kimchi, or a cup of unsweetened live-culture yogurt or kefir. You do not need the six-serving research dose to take the hint.
4. Cook and cool some starch. Leftover potatoes, rice, and beans are one of the cheaper resistant-starch sources we use. They are not a consolation prize.
5. Skip the weekly cleanse. The community shifts in days and settles on the pattern you keep. A juice week teaches it the wrong lesson.
6. Treat antibiotics as a tool, not a habit. When you need them, take them. Then go back to food, not to a cart of blends nobody has tested for the job.
If you read the carbohydrate article, this is the same farm from the other side of the plate. We grow the plants. The microbes finish the job our enzymes do not.
Further Reading
If you want the papers instead of a protocol, start with three: the American Gut plant-count paper, the Stanford fiber-versus-fermented trial, and the Lancet carbohydrate-quality review. Read the cell-count paper if you are still carrying the ten-to-one line, and the 2016 Nature extinction paper if you want to see why "just add fiber later" can fail after the loss has compounded. Those will tell you more than a year of supplement ads.
On Our Website
Start with the carbohydrate piece if you have not. It is the argument this chapter assumes, and it already covers the survival-versus-quality distinction. Then walk the soil articles if you want the same idea outdoors. Different organisms. Same refusal to feed a living system with one input and a sales pitch.
- [Are Carbohydrates Essential?](https://www.mindfullivingsanctuary.com/articles/are-carbohydrates-essential) — the chapter this one continues. Not a section of it.
- [Johnson-Su Bioreactor: Fungal Compost Without the Turning](https://www.mindfullivingsanctuary.com/articles/the-johnson-su-bioreactor-compost-system-a-modern-solution-for-aerobic-composting) — how we feed biology in the pile, not in a jar.
- [Transforming Degraded Properties in Cold Temperate Climates](https://www.mindfullivingsanctuary.com/articles/transforming-degraded-properties-in-cold-temperate-climates-a-guide-to-soil-restoration) — what "feed the community" looks like on worn Central New York ground.
- [Soil Succession](https://www.mindfullivingsanctuary.com/articles/soil-succession-from-barren-soil-to-a-thriving-forest-ecosystem) — why a mix of residues is a long game above and below the surface.
References
1. Sender R, Fuchs S, Milo R. Revised estimates for the number of human and bacteria cells in the body. *PLOS Biology*. 2016;14(8):e1002533. https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1002533
2. Centers for Disease Control and Prevention. About *C. diff*. Updated May 13, 2026. Treatment and reinfection, including fecal microbiota transplant. https://www.cdc.gov/c-diff/about/index.html
3. David LA, Maurice CF, Carmody RN, et al. Diet rapidly and reproducibly alters the human gut microbiome. *Nature*. 2014;505:559–563. https://www.nature.com/articles/nature12820
4. El Kaoutari A, Armougom F, Gordon JI, Raoult D, Henrissat B. The abundance and variety of carbohydrate-active enzymes in the human gut microbiota. *Nature Reviews Microbiology*. 2013;11:497–504. https://www.nature.com/articles/nrmicro3050
5. Sonnenburg ED, Sonnenburg JL. Starving our microbial self: the deleterious consequences of a diet deficient in microbiota-accessible carbohydrates. *Cell Metabolism*. 2014;20(5):779–786. https://pubmed.ncbi.nlm.nih.gov/25156449/
6. Hamer HM, Jonkers D, Venema K, Vanhoutvin S, Troost FJ, Brummer RJ. Review article: the role of butyrate on colonic function. *Alimentary Pharmacology & Therapeutics*. 2008;27(2):104–119. https://pubmed.ncbi.nlm.nih.gov/17973645/
7. Desai MS, Seekatz AM, Koropatkin NM, et al. A dietary fiber-deprived gut microbiota degrades the colonic mucus barrier and enhances pathogen susceptibility. *Cell*. 2016;167(5):1339–1353. https://pmc.ncbi.nlm.nih.gov/articles/PMC5131798/
8. Sonnenburg ED, Smits SA, Tikhonov M, Higginbottom SK, Wingreen NS, Sonnenburg JL. Diet-induced extinctions in the gut microbiota compound over generations. *Nature*. 2016;529:212–215. https://www.nature.com/articles/nature16504
9. McDonald D, Hyde E, Debelius JW, et al. American Gut: an open platform for citizen science microbiome research. *mSystems*. 2018;3(3):e00031-18. https://pmc.ncbi.nlm.nih.gov/articles/PMC5954204/
10. Wastyk HC, Fragiadakis GK, Perelman D, et al. Gut-microbiota-targeted diets modulate human immune status. *Cell*. 2021;184(16):4137–4153. https://pmc.ncbi.nlm.nih.gov/articles/PMC9020749/
11. Reynolds A, Mann J, Cummings J, Winter N, Mete E, Te Morenga L. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. *The Lancet*. 2019;393(10170):434–445. https://pubmed.ncbi.nlm.nih.gov/30638909/
12. Quagliani D, Felt-Gunderson P. Closing America's fiber intake gap: communication strategies from a Food and Fiber Summit. *American Journal of Lifestyle Medicine*. 2016;11(1):80–85. https://pmc.ncbi.nlm.nih.gov/articles/PMC6124841/
