VIII. 3. Metformin

Metformin is one of the rare drugs that works partly through your gut bacteria — it feeds Akkermansia, and fiber and exercise amplify its benefit.

Metformin and the Gut – A Drug That Speaks to Bacteria

Metformin is one of the few medications where the gut microbiota may be central to its therapeutic mechanism [24].

The story of metformin begins not in a pharmaceutical laboratory but in a medieval herb garden. For centuries, the plant Galega officinalis – known in English as French lilac or goat’s rue – had been used across Europe as a folk remedy for the symptoms we now recognise as diabetes: excessive thirst, frequent urination, and the wasting of the body. Physicians in the sixteenth century recorded its use for ‘the pestilence of sweet urine.’ In the 1920s, chemists identified that the plant’s active compounds were guanidine derivatives, and biguanides – a more stable chemical class – were synthesised soon after. Most were too toxic for clinical use. In 1957, a French physician named Jean Sterne published the results of the first clinical trials of a biguanide he called ‘Glucophage’ – literally, ‘glucose eater’. Sterne’s drug was metformin. It went on to become the most prescribed oral diabetes medication in the world. What Sterne could not have anticipated was that six decades later, researchers would discover that metformin’s primary site of action is not the liver, as long assumed, but the gut itself – and that it works, at least in part, by remodelling the microbial communities living there.

Metformin's relationship with the gut microbiota is among the most scientifically productive medication–microbiome^[G] stories in modern medicine, and it overturned a long-standing assumption about how the drug works. For decades, metformin was understood primarily as a liver-acting drug: it inhibits hepatic gluconeogenesis through activation of AMPK, reducing fasting blood glucose. Gut effects were noted – the drug causes gastrointestinal side effects in many patients – but were considered secondary. [221] A 2019 study by Wu and colleagues published in Nature Medicine challenged this view directly. In a randomized controlled trial of type 2 diabetes patients, metformin's blood glucose effects were substantially mediated by gut microbiota changes. When germ-free^[G] mice were transplanted with microbiota from metformin-treated patients, the recipients showed improved glucose tolerance compared to mice receiving microbiota from untreated controls – demonstrating that the microbiota changes caused by metformin were themselves therapeutically active, not merely coincidental. [222] The specific microbial changes metformin induces are now well characterized. The most consistent finding is an increase in Akkermansia muciniphila – the mucus-associated species linked to improved barrier integrity and reduced metabolic endotoxemia. Metformin also consistently increases several short-chain fatty acid–producing taxa and reduces bile acid–metabolizing species in ways that increase the proportion of secondary bile acids with favorable metabolic effects. [144] The clinical significance is bidirectional. First, understanding that metformin works partly through the microbiota helps explain its variable response across patients: individuals with depleted Akkermansia or other key responders may show weaker glycemic benefit. Second, interventions that support Akkermansia – a high-fiber diet, polyphenol-rich foods – may enhance metformin's efficacy. The drug and the diet are not competing interventions for type 2 diabetes microbiome management; they are synergistic ones.

Metformin is the most widely prescribed oral medication for type 2 diabetes and insulin resistance. It belongs to the biguanide class and works primarily by reducing hepatic glucose production and improving peripheral insulin sensitivity. What makes metformin uniquely interesting from a microbiota perspective is accumulating evidence that a significant portion of its glycaemic effect is mediated not through classical pharmacological pathways but through the gut microbiota itself [221].

Metformin achieves very high concentrations in the intestinal lumen relative to plasma levels. Unlike most systemic drugs that affect the gut as a secondary consequence, metformin appears to directly interact with the gut microbial community at pharmacologically relevant concentrations. This luminal presence is now considered central to understanding both its therapeutic effects and its gastrointestinal side effect profile.

The primary mechanism by which metformin influences the microbiota involves inhibition of complex I of the mitochondrial electron transport chain in intestinal cells, altering local energy metabolism and bile acid secretion. These changes create a modified luminal environment that selectively favours certain bacterial groups over others, shifting community composition in ways that may contribute to metabolic improvement.

Akkermansia muciniphila, a mucus-residing bacterium associated with gut barrier integrity and metabolic health, is consistently increased by metformin in multiple human and animal studies. This species produces propionate and acetate, supports mucus layer maintenance, and is associated with improved glucose metabolism and reduced intestinal permeability – effects that align with metformin's clinical benefits.

Metformin also influences bile acid metabolism in the gut. It reduces intestinal bile acid reabsorption and alters the composition of the bile acid pool reaching the colon. Since bile acids are potent modulators of microbial community composition, these changes contribute to metformin's downstream microbiota effects.

Gastrointestinal side effects – nausea, diarrhoea, abdominal discomfort – affect 20 to 30 percent of patients initiating metformin and are the primary reason for discontinuation. These side effects are largely attributed to metformin's luminal effects on gut motility, bile acid dynamics, and microbial fermentation patterns. Extended-release formulations reduce side effect frequency by slowing luminal drug release and distribution.

Emerging research suggests that the microbiota composition at the time of metformin initiation predicts treatment response. Patients with higher baseline Akkermansia muciniphila abundance tend to show better glycaemic responses to metformin, positioning the microbiota as both a mediator and a predictor of metformin efficacy.

Managing Gut Health During Metformin Therapy

Gastrointestinal side effect management is the primary clinical challenge during metformin initiation. Starting at a low dose (typically 500 mg once or twice daily) and titrating slowly over four to eight weeks substantially reduces side effect burden while allowing gut adaptation.

Extended-release (ER) metformin formulations are recommended over immediate-release for patients with significant gastrointestinal sensitivity. ER formulations achieve equivalent glycaemic control with significantly lower rates of nausea and diarrhoea, and their slower luminal release profile may also produce a more gradual microbiota transition.

Taking metformin with or immediately after meals reduces peak luminal concentrations and minimises gastric irritation. This simple timing adjustment significantly reduces nausea in most patients.

Dietary fiber intake supports the microbiota effects of metformin. Prebiotic fibers that feed Akkermansia muciniphila and other metformin-favoured species amplify the beneficial microbiota shifts associated with the drug. Inulin, pectin, and resistant starch are particularly relevant substrates.

Fermented foods and probiotic supplementation can complement metformin's microbiota effects. Some clinical trials have combined metformin with probiotic supplementation and observed superior glycaemic outcomes compared to metformin alone, suggesting additive or synergistic microbiota modulation.

Vitamin B12 monitoring is essential during long-term metformin therapy. Metformin impairs vitamin B12 absorption in the terminal ileum through mechanisms involving calcium-dependent membrane transporters, and B12 deficiency affects approximately 10 to 30 percent of long-term users. B12 deficiency has its own effects on gut health, neurological function, and energy metabolism.

Physical activity amplifies metformin's microbiota effects. Exercise independently increases Akkermansia muciniphila abundance and supports the butyrate-producing species that metformin also favours. The combination of metformin and regular aerobic exercise produces greater microbiota and metabolic benefits than either alone.

In patients undergoing FMT for conditions alongside insulin resistance or diabetes, metformin continuation during FMT is a clinical decision requiring individual assessment. Metformin's own microbiota-modifying effects interact with FMT engraftment^[G] dynamics in ways not yet fully characterised. Clinical teams manage this on a case-by-case basis.

Microbiota Effects

• Metformin consistently increases Akkermansia muciniphila abundance across multiple human and animal studies, with this effect considered one of the primary microbiota mechanisms underlying its metabolic benefits [221].
• Metformin enriches short-chain fatty acid-producing bacteria including Bifidobacterium, Lactobacillus, and butyrate producers in the Lachnospiraceae and Ruminococcaceae families, contributing to improved SCFA output and gut barrier support [24].
• Metformin alters bile acid metabolism by reducing intestinal reabsorption and changing the composition of the secondary bile acid (bile acids chemically modified by gut bacteria, important for colonisation resistance) pool, which selectively reshapes microbial community composition in the colon.
• Metformin reduces the abundance of pro-inflammatory Pseudomonadota (formerly Proteobacteria) and Enterobacteriaceae members, contributing to a less endotoxaemic microbial profile and lower systemic LPS exposure.
• The gastrointestinal side effects of metformin (diarrhoea, altered motility) are partly driven by its luminal effects on fermentation dynamics and bile acid-driven gut secretion, mechanisms that overlap with its microbiota-modifying actions.
• Metformin's microbiota effects partially explain its glycaemic efficacy: some germ-free animal studies suggest metformin may lose part of its glucose-lowering efficacy in the absence of gut microbiota, positioning the microbiota as a functional mediator; the magnitude of this effect and its human clinical relevance remain under investigation.
• Long-term metformin use produces durable microbiota shifts that persist beyond the acute pharmacological effect, suggesting that microbiota restructuring contributes independently to its sustained metabolic benefits.

Patient Guidance

• Start metformin at a low dose and titrate slowly to minimise gastrointestinal side effects.
• Take metformin with or immediately after meals to reduce nausea and gastric irritation.
• Consider extended-release formulations if immediate-release causes persistent gastrointestinal discomfort.
• Increase dietary fiber intake – especially inulin, resistant starch, and pectin – to support metformin's microbiota effects.
• Include fermented foods or discuss probiotic supplementation with your clinician for additive metabolic benefit.
• Monitor vitamin B12 levels annually during long-term metformin use; supplement if levels decline.
• Combine metformin with regular aerobic exercise for synergistic microbiota and glycaemic benefits.
• Report persistent diarrhoea, nausea, or abdominal pain to your clinician – dose or formulation adjustment may help.
• If undergoing FMT, discuss metformin continuation with your clinical team before the procedure.

References

[24] Sonnenburg JL, Bäckhed F. Diet–microbiota interactions as moderators of human metabolism. Nature. 2016. Link

Review of mechanisms linking the gut microbiota to obesity and type 2 diabetes drawing on translational animal models and human studies. The microbiota emerges as a mediator of dietary impact on host metabolic status, with growing efforts to establish causal relationships in people and develop therapeutic interventions including personalised nutrition.

[144] Cani PD, Amar J, Iglesias MA et al. Metabolic endotoxemia initiates obesity and insulin resistance. Diabetes. 2007. Link

Cani and colleagues' 2007 Diabetes paper introduced the concept of 'metabolic endotoxemia' as a microbiota-driven trigger of obesity and insulin resistance. In mice, they show that a high-fat diet increases intestinal permeability and circulating lipopolysaccharide (LPS) levels, which activate TLR4-CD14 signalling and induce low-grade inflammation in adipose tissue, liver and muscle. Chronic subcutaneous LPS infusion in mice was sufficient to reproduce diet-induced obesity, insulin resistance and hepatic steatosis. CD14-knockout mice were protected. The paper established a mechanistic axis linking gut microbiota, barrier function and metabolic disease that has shaped subsequent obesity-microbiome research.

[221] Wu H, Esteve E, Tremaroli V et al. Metformin alters the gut microbiome of individuals with treatment-naive type 2 diabetes, contributing to the therapeutic effects of the drug. Nat Med. 2017. Link

This double-blind randomized trial assigned treatment-naive type 2 diabetes patients to placebo or metformin for 4 months and showed metformin had strong effects on the gut microbiome, replicated in a placebo crossover subgroup. FMT from metformin-treated donors into germ-free mice improved glucose tolerance. In vitro gut-simulator experiments showed metformin affected pathways encoding metalloproteins and metal transporters in species across two phyla. The findings provide causal evidence that gut microbiota mediate part of metformin's antidiabetic effect.

[222] Forslund K, Hildebrand F, Nielsen T et al. Disentangling type 2 diabetes and metformin treatment signatures in the human gut microbiota. Nature. 2015. Link

Using 784 human gut metagenomes, this study disentangled type 2 diabetes (T2D) microbiome signatures from antidiabetic drug effects and showed antidiabetic medication, particularly metformin, confounds prior T2D dysbiosis associations. The authors provide evidence for microbial mediation of metformin's therapeutic effects through short-chain fatty acid production, and for microbiota-mediated mechanisms behind known GI side effects, including a relative increase in Escherichia species. The findings highlight that treatment status must be controlled when characterizing disease-associated microbiomes.