Metformin, Berberine and the Intestine: What New Research Actually Shows

A new study is putting the intestine at the center of a long-running question: how does metformin lower blood sugar?

Metformin is commonly associated with reducing glucose production by the liver and improving the body’s response to insulin. Those effects remain part of the drug’s established biology. But researchers now report evidence that metformin also acts on mitochondria inside intestinal epithelial cells—the cells lining the gut—to increase local glucose use.

The same study found that berberine, a plant-derived compound sold as a dietary supplement, produced a similar glucose-lowering effect in mice. The finding is scientifically notable, but it does not show that berberine is equivalent to metformin, or that people should use it as a replacement for prescribed diabetes treatment.

A new intestinal mechanism for metformin

The research, published in Nature Metabolism, focused on mitochondrial complex I, a component of the respiratory chain. Mitochondria normally help cells generate energy through oxidative metabolism. In the study, metformin inhibited complex I in intestinal epithelial cells, shifting those cells toward glycolysis—the process of breaking down glucose for energy.

In genetically modified mice, this shift made the intestinal lining behave like a temporary glucose “sink.” The cells took up more glucose and converted it into lactate and other metabolites. When the researchers restored complex I activity specifically in the intestinal epithelium, metformin’s effects on glucose tolerance and post-meal blood sugar were weakened. That result supports a causal role for intestinal complex I inhibition in the mice studied. (Primary study (https://pmc.ncbi.nlm.nih.gov/articles/PMC13303093/))

The researchers also analyzed human metabolomic data. Metabolomics measures patterns of small molecules in the body and can reveal whether a drug appears to affect particular biological pathways. In people taking metformin, the data were consistent with changes in intestinal mitochondrial metabolism, including lower levels of citrulline, a molecule produced in the small intestine.

But the human data did not directly reproduce the mouse experiments. The genetic tests that established whether intestinal complex I was necessary for metformin’s glucose-lowering effect were performed in mice. That distinction is important: human metabolic signatures can support a mechanism, but they do not by themselves prove that the same molecular sequence explains the drug’s effects in every patient.

Human studies already pointed toward the gut

The idea that metformin affects the intestine is not entirely new. A randomized clinical trial involving people with newly diagnosed type 2 diabetes found that metformin increased glucose uptake in the small intestine and colon. The investigators measured tissue-specific glucose handling with fluorodeoxyglucose positron-emission tomography, or FDG-PET, during controlled insulin exposure. Intestinal uptake increased substantially in the metformin group and was associated with improved glycemic control. (Randomized human trial (https://pubmed.ncbi.nlm.nih.gov/28778047/))

Other human imaging research has found increased accumulation of glucose-like tracers in the bowel during metformin treatment. A 2020 PET-MRI study reported greater tracer accumulation in several intestinal segments among people taking metformin, particularly in the intestinal lumen. The authors said this was consistent with metformin promoting movement of glucose from the circulation into the intestine. (Human PET-MRI study (https://pubmed.ncbi.nlm.nih.gov/32493754/))

A 2025 study further examined metformin-regulated glucose movement from the circulation into the intestinal lumen using PET-MRI-based imaging. Taken together, these studies make increased intestinal glucose handling a reproducible clinical observation, even though the precise cellular pathway is still being clarified. (Human glucose-flux study (https://pubmed.ncbi.nlm.nih.gov/40033038/))

The new complex-I research therefore adds a possible explanation to an existing human finding. It does not erase metformin’s known effects on the liver or insulin sensitivity; instead, it suggests that the gut may be a more important site of action than previously appreciated.

What the study says about berberine

Berberine is an alkaloid found in plants including barberry, goldenseal and Oregon grape. It is widely marketed as a dietary supplement and is sometimes described online as “natural metformin.” That comparison can be misleading.

In the new study, berberine improved glucose tolerance in ordinary mice, but not in mice engineered to maintain mitochondrial complex I activity in intestinal epithelial cells. This result suggests that, in that experimental system, berberine’s acute glucose-lowering effect depended on intestinal complex I inhibition. The authors also used experiments involving the intestinal drug transporter P-glycoprotein, which helps explain why berberine may remain concentrated in the gut despite limited absorption into the bloodstream. (Primary mechanistic study (https://pmc.ncbi.nlm.nih.gov/articles/PMC13303093/))

However, the berberine mechanism has not been demonstrated in a comparable human clinical trial. The key evidence came from mice and laboratory experiments. The doses and exposure patterns used in animal studies also cannot be translated directly into supplement instructions for people.

Berberine may affect several biological processes, including glucose transport, insulin signaling, cellular energy metabolism and the gut microbiome. Earlier mouse research, for example, found that berberine reduced intestinal glucose absorption through effects involving the GLUT2 transporter. Such findings may be compatible with a broader, multi-pathway effect, but they do not establish which mechanism matters most in humans. (Earlier mouse study (https://pmc.ncbi.nlm.nih.gov/articles/PMC8745600/))

Clinical evidence for berberine remains limited

Some clinical studies suggest that berberine may modestly improve blood glucose measures in people with type 2 diabetes. But the evidence is uneven. The National Center for Complementary and Integrative Health says studies have differed in participant characteristics, product formulations, treatment protocols, study quality and results. Many were conducted in Chinese populations, and the available research is not strong enough to support definite conclusions for all patients. (NCCIH evidence summary (https://www.nccih.nih.gov/health/diabetes-and-dietary-supplements-what-you-need-to-know))

That uncertainty matters because dietary supplements are not interchangeable with standardized prescription medicines. Products can differ in purity, concentration and labeling accuracy. Berberine can also cause gastrointestinal symptoms, including nausea, constipation, diarrhea and abdominal discomfort, and it may interact with medications. NCCIH advises people who take medicines to consult a health professional before using berberine. It also warns that berberine should not be used during pregnancy or breastfeeding and should not be given to infants. (NCCIH safety information (https://www.nccih.nih.gov/health/berberine-and-weight-loss-what-you-need-to-know))

What can reasonably be concluded?

The strongest conclusion is that the intestine is an important target of metformin. Human imaging and metabolic studies show increased intestinal glucose handling during treatment. The new mouse and cell research proposes that inhibition of mitochondrial complex I in intestinal epithelial cells helps drive that effect by increasing local glucose uptake and glycolysis.

For berberine, the evidence is more preliminary. The compound appears to share the intestinal complex-I pathway in the study’s mouse models, but researchers have not yet established that the same mechanism lowers blood sugar in people. Nor does the study prove that berberine offers the same benefits, safety profile, reliability or clinical outcomes as metformin.

The findings could eventually guide development of drugs designed to target intestinal metabolism more precisely. They may also help explain why gastrointestinal side effects are common with metformin and why the drug can alter bowel activity and glucose-tracer results during medical imaging.

For now, the research supports a promising biological model—not a new self-treatment strategy. People should not stop metformin, substitute berberine for it or combine the two without advice from a qualified clinician. This article is for general information and is not a diagnosis or treatment recommendation.