What the microbiome does

• The microbiome is a community of microbes inhabiting body surfaces, including the gut lumen, skin, mouth, nose, lungs, and reproductive tract. The gut lumen is continuous with the external environment, and its contents enter the body after absorption.
• A functioning microbiome produces useful metabolites, excludes pathogenic organisms, and contributes to immune-system development. Bacteria predominate, alongside fungi and viruses.
• Microbial fermentation of indigestible fiber produces butyrate and propionate, which can fuel colon cells and support the gut barrier; microbes also synthesize vitamin K and B vitamins. Some microbes generate toxins or methane associated with reduced intestinal motility and constipation; harmful imbalance is dysbiosis.

Fiber and dietary change

• Dietary fiber is not required for a microbial community to persist. Its removal favors organisms capable of using other substrates, and microbiome composition alone does not determine its functional adequacy.
• In a 2014 Nature experiment, ten people ate meat, eggs, and cheese for five days. Their microbiomes shifted within 24 hours, fiber fermenters declined, diversity remained intact, and the community returned toward baseline within 48 hours of resuming an omnivorous diet. [1]
• A 2026 comparison of ten long-term carnivores with matched omnivores found differences in microbial composition without a reduction in overall diversity. Carnivore, vegan, and omnivore diets can support distinct functional microbial communities. [2]
• Dietary transitions can cause temporary gastrointestinal symptoms during microbial adaptation, often over about a week. In a 2024 study, ten athletes consumed 250 g of sauerkraut daily for ten days; bloating, diarrhea, and pain peaked around days five to six before settling. [3]
• An initial large serving of white kimchi caused digestive discomfort, while gradual introduction was followed by tolerance. Lactobacillus sakei, found in some ferments, has demonstrated protection against sinus pathogens in experimental research. [4]

Protection, diversity, and immune training

• Beneficial microbes occupy physical niches, consume resources needed by pathogens, and secrete inhibitory compounds. Dysbiosis involves unwanted organisms or impaired microbial function, not merely lower diversity.
• Gut microbes influence innate and adaptive immune responses. Multiple microbial compositions can supply metabolites, resist pathogens, and support immune function without sharing an identical species profile.
• Estimates are approximately 38 trillion bacterial cells and 30 trillion human cells; microbial biomass is given as 1–3% of body weight. [5]

Reproductive microbiomes and microbial transmission

• Lactobacilli contribute to an acidic reproductive-tract environment that inhibits certain organisms. In a 2016 IVF study of 35 women, a uterine community above 90% Lactobacillus was associated with a 59% live-birth rate (10/17), versus 7% (1/15) without that dominance; uterine microbial composition is a potential reproductive-health factor. [6]
• Fusobacterium was detected in 27/42 women with endometriosis versus 3/42 controls. In mice, its inoculation increased endometrial lesions, whereas antibiotics prevented or reduced lesions; translation of these findings to human disease remains an open clinical possibility. [7]
• Bacterial vaginosis involves diminished vaginal Lactobacillus dominance and frequent recurrence after antibiotics. In a 2025 trial, antimicrobial therapy for both the woman and her male partner reduced 12-week recurrence from 63% with female-only therapy to 35%, consistent with sexual transmission of associated organisms. [8]
• Cohabiting people share microbes without intimate contact; mother-infant pairs share about half of strains among microbial species common to both. Household microbial exchange is one possible factor in the persistence or recurrence of dysbiosis and declining probiotic effects. [9]
• Human studies of antibiotics targeting Fusobacterium in endometriosis yielded smaller effects than mouse experiments. Reinoculation by untreated partners is a hypothesis requiring direct testing.

Autoimmunity, molecular mimicry, and fungi

• Molecular mimicry involves immune recognition of a pathogen component that resembles a human protein. Examples include streptococcal infection associated with rheumatic heart disease and gastrointestinal infections involving Salmonella, Shigella, or Campylobacter followed by reactive arthritis.
• Immune-suppressing medications can limit autoimmune damage without eliminating potential microbial triggers and can impair pathogen clearance. They can also be lifesaving, and changes require physician supervision alongside investigation of potential triggers.
• In a 2021 survey of more than 2,000 carnivore eaters, 36% of respondents with autoimmune disease indicated complete resolution and another 53% indicated substantial improvement. These were self-assessed outcomes. [10]
• Fungi share substantial biological similarity with humans. In a 2015 experiment, human genes functionally rescued 47% of 414 essential yeast-gene substitutions. Shared features can enable molecular mimicry and complicate selective antifungal drug action. [11]
• Lower carbohydrate availability on carnivore diets may limit the growth of sugar-utilizing fungi such as Candida without eliminating them, providing a potential microbiome-mediated mechanism for changes in autoimmune symptoms.

Sweeteners and microbial selection

• Artificial sweeteners affect microbial communities: a 2022 trial of 120 adults found that two weeks of each of four sweeteners changed gut microbial composition. [12]
• Trehalose, an ice-cream ingredient that reduces ice-crystal formation, is usable by some epidemic Clostridioides difficile strains. A genetic mutation increased sensitivity to low concentrations more than 500-fold, and wider food use coincided temporally with outbreaks. [13]
• Reducing a nutrient source can also affect microbes that depend on it. Animal evidence linking carbohydrate availability to microbial growth is part of the proposed low-carbohydrate mechanism involving Candida.

Celiac disease, Candida, and dairy

• Celiac disease often receives a diagnosis in middle age despite a genetic predisposition, leaving room for an environmental trigger. Candida can shift from yeast cells to invasive hyphae, with glucose availability contributing to that change.
• Invasive Candida produces hyphal wall protein 1 (Hwp1), with amino acid sequences resembling gluten epitopes. Immune cross-reactivity with gluten and nearby tissue damage are proposed mechanisms; Hwp1-deficient Candida has reduced adhesion and virulence in cell and mouse experiments. [14][15]
• A 2017 Australian study detected Candida DNA in stool from 33% of 45 people with symptomatic celiac disease versus none of the non-celiac controls. The observation is compatible with a Candida association. [16]
• Dairy can induce a separate inflammatory response in some patients with celiac disease. In a 2007 rectal-challenge study of 20 patients in remission, gluten provoked inflammation in 18 and cow’s milk in 10; additional testing implicated casein, a protein with similarities to gluten, not lactose. [17]
• Lactose malabsorption allows undigested sugar to draw water into the bowel and undergo microbial fermentation, causing gas, bloating, diarrhea, or methane-associated constipation. Injury to intestinal villi in celiac disease can reduce lactase activity, which can recover over a year on a gluten-free diet.
• Dairy products vary in lactose and casein content: hard cheese has little lactose, and butter or ghee has much less casein than milk. Fermentation partly breaks down casein, while added milk solids in commercial yogurt add unfermented proteins; ingredient checking and home fermentation can alter intake.

Tonsils, fungi, and autoimmune risk

• Some recurrent tonsillitis involves fungi, so repeated antibiotics may reduce competing bacteria without clearing Candida. A 2017 study of 75 children undergoing tonsil removal found fungal involvement in roughly one-third of affected tonsils, versus none in controls. [18]
• Tonsils contribute to immune surveillance of organisms passing through the throat. Their removal during recurrent infection eliminates that local immune tissue and may affect defenses against underlying fungal colonization.
• Three observational studies associated tonsil or adenoid removal with subsequent autoimmune diagnoses: a Swedish cohort found increases in 16 diseases; a Korean childhood cohort found 32% higher overall risk; and a US cohort found 52% higher Sjögren’s risk. These associations do not establish causality. [19][20][21]
• Lactobacillus reuteri inhibits some Candida strains and can produce reuterin, a substance with antifungal activity. Oral formulations are a possible area for investigation in throat health. [22]

Diet-matched microbes and testing

• Lactobacillus reuteri can use glycerol and amino acids, which occur in fat and protein metabolism, making it a plausible organism on a fatty carnivore diet. The ten-person carnivore study found no depletion of the broader Lactobacillus group but did not measure individual L. reuteri strains. [2]
• Akkermansia muciniphila consumes gut mucus and can contribute to the intestinal barrier; its abundance was not reduced in the carnivore comparison. Bacillus subtilis preferentially uses protein and could also be sustained by that diet, although the latter prediction lacks direct testing. [2][23]
• Bacillus subtilis has measurable activity beyond the gut: a probiotic trial reduced Staphylococcus aureus colonization in both the nose and intestines. [24]
• Large, rapid changes in probiotic intake can cause temporary gastrointestinal discomfort, and strains without suitable nutrients may fail to persist. Probiotic selection depends on the organism’s function and available dietary substrates.
• Stool microbiome testing samples organisms shed into feces and may miss organisms residing within the intestinal mucus layer, limiting correspondence with the entire gut ecosystem.

Glyphosate and food residues

• Glyphosate blocks the shikimate pathway used by plants and many bacteria for aromatic amino acid synthesis. Humans lack this pathway, while computational sequencing analyses identified potential susceptibility in more than half of core gut microbial species. [25]
• In a study of 65 UK twin pairs, glyphosate was detected in the urine of 53% of individuals, indicating exposure within that sample. [26]
• Some glyphosate-resistant crops carry a resistance gene isolated from organisms found in glyphosate-manufacturing waste, enabling crops including soy and corn to survive spraying.
• Organic-food interventions reduce urinary pesticide residues. Organic sourcing, peeling, and careful cleaning of conventional produce can reduce some exposures; strawberries are particularly difficult to clean thoroughly. [27]

Restoring a dysbiotic microbiome

• The proposed restoration sequence is removal of unwanted organisms, introduction of beneficial organisms, sustained nutrition for those microbes, and sometimes short-term occupation of ecological niches. Broad antimicrobials alone can leave vacant niches, while introduced organisms need compatible dietary substrates.
• On carnivore diets, Lactobacillus reuteri, Akkermansia muciniphila, and potentially Bacillus subtilis are candidates because they have nutrient sources beyond dietary fiber. This is a diet-matching rationale, not evidence of one universal probiotic formulation.
• Bifidobacterium infantis consumes human milk oligosaccharides (HMOs) that babies cannot digest. Colonization was about 92% among Gambian breastfed infants in one comparison and below 10% in several Western cohorts, illustrating the effects of ecological and dietary context. [28]
• In a 2023 adult experiment, B. infantis plus HMOs given after antibiotic-induced dysbiosis achieved substantial engraftment in 76% of participants, with abundance declining after HMO withdrawal. [29]
• A temporary combination of B. infantis and HMOs after antimicrobial use could occupy niches and limit pathogenic recolonization during establishment of diet-supported organisms. The complete sequence has not undergone direct clinical testing, despite its use in individual practice.
• Probiotic effects are strain- and condition-specific. Lactobacillus paracasei improved dandruff measures in a clinical study; in ulcerative colitis, a 12-month double-blind trial found equivalent relapse outcomes with Escherichia coli Nissle 1917 and mesalazine. [30][31]

References

1. [04:54] Diet rapidly and reproducibly alters the human gut microbiome — https://doi.org/10.1038/nature12820
2. [05:33] Long-term adherence to the carnivore diet and its impact on the gut microbiota: a cross-sectional study — https://doi.org/10.1530/MAH-25-0015
3. [07:11] Short-Term Supplementation of Sauerkraut Induces Favorable Changes in the Gut Microbiota of Active Athletes: A Proof-of-Concept Study — https://doi.org/10.3390/nu16244421
4. [08:52] Sinus microbiome diversity depletion and Corynebacterium tuberculostearicum enrichment mediates rhinosinusitis — https://doi.org/10.1126/scitranslmed.3003783
5. [13:07] Revised Estimates for the Number of Human and Bacteria Cells in the Body — https://doi.org/10.1371/journal.pbio.1002533
6. [13:35] Evidence that the endometrial microbiota has an effect on implantation success or failure — https://doi.org/10.1016/j.ajog.2016.09.075
7. [15:43] Fusobacterium infection facilitates the development of endometriosis through the phenotypic transition of endometrial fibroblasts — https://doi.org/10.1126/scitranslmed.add1531
8. [17:54] Male-Partner Treatment to Prevent Recurrence of Bacterial Vaginosis — https://doi.org/10.1056/NEJMoa2405404
9. [19:36] The person-to-person transmission landscape of the gut and oral microbiomes — https://doi.org/10.1038/s41586-022-05620-1
10. [26:56] Behavioral Characteristics and Self-Reported Health Status among 2029 Adults Consuming a “Carnivore Diet” — https://doi.org/10.1093/cdn/nzab133
11. [28:04] Systematic humanization of yeast genes reveals conserved functions and genetic modularity — https://doi.org/10.1126/science.aaa0769
12. [30:45] Personalized microbiome-driven effects of non-nutritive sweeteners on human glucose tolerance — https://doi.org/10.1016/j.cell.2022.07.016
13. [31:39] Dietary trehalose enhances virulence of epidemic Clostridium difficile — https://doi.org/10.1038/nature25178
14. [35:51] Is Candida albicans a trigger in the onset of coeliac disease? — https://doi.org/10.1016/S0140-6736(03)13695-1
15. [36:41] Reduced virulence of HWP1-deficient mutants of Candida albicans and their interactions with host cells — https://doi.org/10.1128/IAI.68.4.1997-2002.2000
16. [37:13] Significantly higher faecal counts of the yeasts Candida and Saccharomyces identified in people with coeliac disease — https://doi.org/10.1186/s13099-017-0173-1
17. [38:42] Mucosal reactivity to cow’s milk protein in coeliac disease — https://doi.org/10.1111/j.1365-2249.2007.03298.x
18. [45:56] Fungal Infection as Cause of Chronic Tonsillitis: A Histopathological-Microbiological Study — https://doi.org/10.12816/0046257
19. [48:01] Tonsillectomy associated with an increased risk of autoimmune diseases: A national cohort study — https://doi.org/10.1016/j.jaut.2016.06.007
20. [48:19] Association between childhood tonsillectomy and increased risk of autoimmune disease: a nationwide population-based cohort study — https://doi.org/10.3389/fmed.2026.1892240
21. [48:35] Surgical removal of Waldeyer’s ring and long-term risk of Sjögren’s syndrome: a population-based cohort study investigating the mucosal immune link — https://doi.org/10.3389/fimmu.2026.1760386
22. [49:31] Probiotic Lactobacillus reuteri has antifungal effects on oral Candida species in vitro — https://doi.org/10.1080/20002297.2016.1274582
23. [51:05] Dietary Protein Sources Differentially Affect the Growth of Akkermansia muciniphila and Maintenance of the Gut Mucus Barrier in Mice — https://doi.org/10.1002/mnfr.201900589
24. [52:09] Probiotic for pathogen-specific Staphylococcus aureus decolonisation in Thailand: a phase 2, double-blind, randomised, placebo-controlled trial — https://doi.org/10.1016/S2666-5247(22)00322-6
25. [55:25] Classification of the glyphosate target enzyme (5-enolpyruvylshikimate-3-phosphate synthase) for assessing sensitivity of organisms to the herbicide — https://doi.org/10.1016/j.jhazmat.2020.124556
26. [55:50] Impacts of dietary exposure to pesticides on faecal microbiome metabolism in adult twins — https://doi.org/10.1186/s12940-022-00860-0
27. [56:59] Organic diet intervention significantly reduces urinary pesticide levels in U.S. children and adults — https://doi.org/10.1016/j.envres.2019.01.024
28. [60:41] Bifidobacterium Species Colonization in Infancy: A Global Cross-Sectional Comparison by Population History of Breastfeeding — https://doi.org/10.3390/nu14071423
29. [61:37] Precision modulation of dysbiotic adult microbiomes with a human-milk-derived synbiotic reshapes gut microbial composition and metabolites — https://doi.org/10.1016/j.chom.2023.08.004
30. [63:13] The positive benefit of Lactobacillus paracasei NCC2461 ST11 in healthy volunteers with moderate to severe dandruff — https://doi.org/10.3920/BM2016.0144
31. [63:24] Maintaining remission of ulcerative colitis with the probiotic Escherichia coli Nissle 1917 is as effective as with standard mesalazine — https://doi.org/10.1136/gut.2003.037747