The gut and autoimmune issues, and what the evidence actually shows
A gut bacterium that escapes to the liver and triggers self attack. A thyroid protein that looks like a bacterial one. Three real mechanisms, one honest verdict, and the line where the research stops and the marketing starts.
Autoimmune conditions are diagnosed and managed by doctors. This article explains published mechanisms and nothing more. It does not describe treating, curing or preventing any disease, and no food in it is offered as a therapy.
The short answer
The gut is where most of your immune system learns what to attack. Three mechanisms link it to autoimmunity in the published research: bacteria physically escaping the gut, bacterial proteins that resemble your own, and the loss of the fermentation products that keep the immune system calm. All three are documented. None of them is a diet you can buy.
In this article
Why the gut is where this argument happens
An autoimmune condition is the immune system attacking the body it belongs to. The obvious question is how a system built on telling self from not self gets it wrong. The gut keeps appearing in the answer for a structural reason: it is the one place where an enormous foreign population lives permanently, one cell layer away from your blood, and is tolerated.
That tolerance is not passive. It is an active, trained, constantly maintained decision. Which means it is a decision that can be pushed. The research below is about three specific ways it gets pushed, and each of them was demonstrated rather than theorised.
Mechanism one: the bacterium that left the gut
In 2018, a team at Yale published a study in Science that is the cleanest single demonstration in this field. They found that translocation of a gut pathobiont, meaning a resident gut bacterium capable of causing harm in the wrong place, to the liver and other systemic tissues triggered autoimmune responses in mice with a genetic background predisposing them to autoimmunity.1
The organism was Enterococcus gallinarum. What makes the study strong is not the association, it is what happened when they intervened. Antibiotic treatment prevented mortality in the model, suppressed growth of the organism in tissues, and eliminated pathogenic autoantibodies and T cells. A vaccine targeting the organism, given intramuscularly, also prevented mortality.1
Figure 01
The translocation experiment, step by step
What was done and what happened. The reversal is what makes this causal rather than correlational.
Bacterial DNA specific to the organism was recovered from liver biopsies of autoimmune patients, and co-cultures with human hepatocytes replicated the murine findings. The authors conclude similar processes apparently occur in susceptible humans.
Source: Manfredo Vieira S, Hiltensperger M, Kumar V, et al. Science, 2018;359(6380):1156 to 1161, retrieved via PubMed.
Hold on to the qualifier the authors themselves used: in genetically predisposed hosts, and in susceptible humans.1 This is not a mechanism that says a leaky gut gives anybody an autoimmune disease. It says that when the genetic ground is already prepared, a gut organism reaching the wrong tissue can be the thing that lights it.
The barrier that is supposed to prevent that escape is covered in its own article. If you want the anatomy of the gate rather than what comes through it, read what "leaky gut" actually means.
Mechanism two: proteins that look like each other
The second mechanism is called molecular mimicry, and the name is exactly what it sounds like. The immune system recognises shapes. If a bacterial protein happens to share enough structure with one of your own, an immune response trained on the bacterium can also fit the tissue.
A 2025 scoping review in Autoimmunity Reviews lays this out for the thyroid. It reports that Lactobacilli and Bifidobacteria have a role in autoimmune thyroid disease through a molecular mimicry mechanism, because their protein sequences share structural similarity with thyroid peroxidase and thyroglobulin.2 Those two proteins are exactly the ones targeted in Hashimoto's thyroiditis.
The immune system does not read names. It reads shapes. Two unrelated proteins with the same shape get the same response.
The same review describes a second lever alongside mimicry. Lactobacilli can increase T helper 17 cells and modify the number of colonic regulatory T cells, which are the cells largely responsible for maintaining immunological tolerance at the gut barrier.2 Mimicry supplies a target. The shift in cell populations supplies the willingness to act on it.
It is worth naming what this does not mean, because the genera involved are the ones sold as probiotics. The review is describing complex, context dependent relationships in people with or developing autoimmune thyroid disease. It is not a reason to fear yoghurt, and it is not a reason to take a supplement either. It is a reason to be suspicious of anyone who tells you a single strain is straightforwardly good for immunity.
Mechanism three: the fuel that keeps the peace
The third link is the one with the clearest connection to what you eat, and it runs through the products of fermentation rather than through food directly.
The thyroid review describes a disrupted intestinal barrier causing dysbiosis and reducing beneficial metabolites such as short chain fatty acids, with proinflammatory effects, as potentially crucial for the induction of autoimmune thyroid disease.2 It goes further and names organisms: Blautia and Anaerostipes working with butyric acid, promoting antimicrobial peptide synthesis from intestinal cells and supporting the innate immune system.2
Independent work supports the barrier half of that. In weaned mice, butyrate enhanced the mucin layer, improved the integrity of the intestinal brush border, accelerated tight junction development, and increased the proportion of regulatory T cells in the small intestine.3 Regulatory T cells are the brake pedal of the immune response. A fermentation product was strengthening both the wall and the brake at once.
Figure 02
Three mechanisms, graded honestly
What was shown, in what species, and what it does not license anyone to claim.
Swipe to see the whole table
| Mechanism | Evidence | Species | Limit |
|---|---|---|---|
| Bacterial translocation | Causal, reversible by antibiotic and vaccine | Mice, plus human biopsy and cell data | Required a predisposing genetic background |
| Molecular mimicry | Structural similarity described in review | Human proteins | Association, not a demonstrated trigger |
| Loss of short chain fatty acids | Barrier and regulatory T cell effects | Mice, at a developmental stage | Not shown to alter disease in adults |
| A diet that prevents autoimmunity | None presented here | Not applicable | Claim is not supported |
Sources: Manfredo Vieira S et al. Science, 2018, Sessa L, Malavolta E, Sodero G, Cipolla C, Rigante D. Autoimmun Rev, 2025, Wang H et al. Nutrients, 2024. Retrieved via PubMed.
Coeliac disease, the one where the whole chain is visible
Coeliac disease deserves its own section here, because it is the single condition where every link in the chain has been observed rather than inferred, and it is therefore the template the rest of the field is measured against.
It is a systemic, immune mediated disorder triggered by gluten in genetically susceptible individuals, and it is associated with altered gut permeability.4 Gliadin peptides pass through the epithelial layer to launch an adaptive immune response in the lamina propria, and that response damages the absorptive surface of the small intestine.5 The permeability step itself has a traced signal: gluten peptides bind CXCR3 and induce zonulin release, which disassembles tight junctions.6 The tight junctions themselves do not fully recover even on a gluten free diet, which is why the barrier remains a research target rather than a solved problem.7
Four things had to be true at once. A genetic susceptibility. A specific dietary trigger. A measurable barrier change with a named mediator. An adaptive immune response with identifiable tissue damage. When people say the gut is involved in autoimmunity, this is the standard of evidence that sentence was earned on, and almost nothing else in the field has all four yet.
What this does not mean
Three claims routinely get built on this research that the research does not carry, and it is worth naming them directly.
- It does not mean a leaky gut caused your condition. Every strong result above is conditional on a predisposing genetic background, and the authors say so themselves.1 Predisposition is not something a diet changes.
- It does not mean a supplement can fix it. The interventions that worked in the translocation study were an antibiotic and a targeted vaccine.1 Not a food, and not a probiotic.
- It does not mean probiotics are straightforwardly protective. In the thyroid review the genera implicated in molecular mimicry are Lactobacilli and Bifidobacteria, the two most commonly sold.2 The relationship is context dependent, and a marketing claim of "supports immunity" is not a finding.
Frequently asked questions
Can gut bacteria really cause an autoimmune reaction?
In a 2018 study in Science, translocation of the gut organism Enterococcus gallinarum to the liver and other tissues triggered autoimmune responses in mice with a genetic background predisposing them to autoimmunity. Antibiotic treatment eliminated the pathogenic autoantibodies and T cells, and a targeted vaccine prevented mortality. Bacterial DNA was also recovered from liver biopsies of autoimmune patients.
What is molecular mimicry?
It is when a bacterial protein shares enough structural similarity with one of your own proteins that an immune response raised against the bacterium can also recognise your tissue. A 2025 review reports that Lactobacilli and Bifidobacteria have protein sequences sharing structural similarity with thyroid peroxidase and thyroglobulin, the two proteins targeted in autoimmune thyroid disease.
Should I take probiotics for an autoimmune condition?
That is a question for the doctor managing the condition, not for an article. Worth knowing before you ask: the genera most implicated in the molecular mimicry described for autoimmune thyroid disease are the same two genera most commonly sold as probiotics, which is a reason for care rather than confidence.
Does fibre help the immune system?
Indirectly, through fermentation rather than through the fibre itself. Gut bacteria convert fermentable material into short chain fatty acids, and in animal work butyrate strengthened the mucin layer, accelerated tight junction development and increased regulatory T cells in the small intestine. That is a mechanism, in mice, and it is not a claim that fibre treats any condition.
Why is coeliac disease used as the example?
Because all four links are documented: a genetic susceptibility, a specific dietary trigger in gluten, a measured barrier change with a named mediator in zonulin, and an adaptive immune response causing identifiable damage to the intestinal surface. Very few proposed gut and immune links have all four, which is what makes it the fair standard to judge others against.
Sources
- Manfredo Vieira S, Hiltensperger M, Kumar V, Zegarra-Ruiz D, Dehner C, Khan N, et al. Translocation of a gut pathobiont drives autoimmunity in mice and humans. Science, 2018;359(6380):1156 to 1161. doi.org/10.1126/science.aar7201
- Sessa L, Malavolta E, Sodero G, Cipolla C, Rigante D. The conspiring role of gut microbiota as primer of autoimmune thyroid diseases: A scoping focus. Autoimmunity Reviews, 2025;24(5):103780. doi.org/10.1016/j.autrev.2025.103780
- Wang H, Qiu J, Zhou M, Luo Y, Li X, Wang M. Monobutyrin Can Regulate the Gut Microbiota, Which Is Beneficial for the Development of Intestinal Barrier Function and Intestinal Health in Weaned Mice. Nutrients, 2024;16(13):2052. doi.org/10.3390/nu16132052
- Valitutti F, Fasano A. Breaking Down Barriers: How Understanding Celiac Disease Pathogenesis Informed the Development of Novel Treatments. Digestive Diseases and Sciences, 2019;64(7):1748 to 1758. doi.org/10.1007/s10620-019-05646-y
- Khaleghi S, Ju JM, Lamba A, Murray JA. The potential utility of tight junction regulation in celiac disease: focus on larazotide acetate. Therapeutic Advances in Gastroenterology, 2016;9(1):37 to 49. doi.org/10.1177/1756283X15616576
- Rallabhandi P. Gluten and celiac disease, an immunological perspective. Journal of AOAC International, 2012;95(2):349 to 355. doi.org/10.5740/jaoacint.sge_rallabhandi
- Jauregi-Miguel A. The tight junction and the epithelial barrier in coeliac disease. International Review of Cell and Molecular Biology, 2020;358:105 to 132. doi.org/10.1016/bs.ircmb.2020.09.010
All references retrieved via PubMed.
