One more study has shown that metformin that is commonly prescribed to people with Type-2 diabetes alters the gut bacteria. While earlier studies have shown a difference in the gut bacteria population with metformin intake, the latest study published in Nature Medicine finds the altered gut bacteria mediates some of therapeutic benefits of the drug.

To test the therapeutic benefits arising from altered gut bacteria, a team led by Fredrik Backhed from the University of Gothenburg, Sweden and José Manuel Fernàndez-Real at the University of Girona, Spain carried out a four-month-long trial involving 40 volunteers who were recently diagnosed with Type-2 diabetes and had not started treatment. The volunteers where randomly assigned to get either metformin (22 volunteers) or a placebo (18 people). A subset of the control group (13 people) was given metformin six months after the start of the study.

Both the groups consumed a calorie-restricted diet for the entire duration of the study. Neither the volunteers nor the researchers knew who received the medication and who got a placebo.

Owing to calorie-restricted diet, the body-mass index of subjects in both the groups reduced. As expected, the groups that received metformin witnessed a significant decrease in HbA1c and fasting blood glucose levels.

But what came as a surprise was that at the end of four months, metformin was found to have “strong effects” on the gut microbiome. Whole-genome sequencing of faecal samples revealed that only one bacterial strain was altered in subjects receiving placebo while significant alterations in relative abundance of over 80 bacterial strains were seen in the group receiving the medicine.

The drug appeared to promote the growth of Akkermansia muciniphila and Bifidobacterium adolescentis strains of bacteria. Even in in vitro studies, metformin promoted the growth of B. adolescentis. The team found no direct correlation between HbA1c level and A. muciniphila.

To further test the effect of changed gut bacteria in promoting therapeutic effects, the researchers transferred faecal samples from treated participants to germ-free mice that were fed high-fat diet for a week before stool transfer and during 18 days of colonisation. There was no reduction in body weight in mice but improvements in glucose tolerance were seen in animals that received the transplant. But improved glucose metabolism was not seen in all mice that got the stool transplant — only two of the three animals showed improvement. The large differences in the gut bacteria in humans might have resulted in the key bacterial species not being transplanted into mice, the researchers write.

Metformin-induced alterations in microbially-regulated metabolites were seen and this suggests that the metabolites might be “partly responsible for the stronger glucose-lowering effect”. But it “cannot be concluded that the major mechanism of action of metformin is through the microbiota”, they write.