Liraglutide temporarily reshapes gut bacteria in mice, with changes largely reversing after treatment stops

A team of researchers gave obese mice liraglutide—a drug now widely prescribed for weight loss and diabetes—for two weeks, then watched what happened to their gut bacteria when the treatment stopped. The answer was striking: the microbiome shifted dramatically while the drug was active, but largely bounced back to normal within a week of stopping. The finding raises a fundamental question about how these medications work and whether their effects on the bacterial ecosystem inside us are meant to be temporary or lasting.

The study, published in Microbiology Spectrum, tracked 24 male mice across three groups over 21 days. One group received daily liraglutide injections while eating a high-fat diet; another high-fat group got saline injections as a control; a third ate a low-fat diet with saline. Researchers collected fecal samples at the start, after 14 days of treatment, and again seven days after stopping the drug. They sequenced the bacterial DNA using a method that targets a specific region of the 16S ribosomal gene, generating 7.1 million high-quality genetic reads to map the microbial landscape.

The drug's effect on the mice themselves was clear: liraglutide-treated animals weighed significantly less than their high-fat-fed peers by day four and stayed lighter throughout. But when the injections stopped, their weight began creeping back up, though not all the way to where it started. The microbiome told a similar story of temporary change. During active treatment, the bacterial community shifted noticeably—liraglutide favored bacteria related to Lactobacillus species, which produce lactic acid, and a genus called Leptogranulimonas. At the same time, it suppressed bacteria that ferment proteins and carbohydrates, like Romboutsia, Faecalicatena, and Oscillibacter. These suppressed bacteria dropped from 11.49 percent of the community to just 2.61 percent by day 14.

But the reversal was swift. By day 21, after a week without the drug, the gut bacterial community had largely returned to its baseline state. The suppressed fermentative bacteria bounced back to 8.54 percent. The bacterial diversity measures—Shannon diversity and Chao1 richness—also recovered, though not completely. One exception: bacteria related to Bacteroidales remained reduced even after washout, suggesting not every change was fully reversible. The researchers identified 21 bacterial variants that accounted for most of the variation in the liraglutide-treated group, with nine increasing during treatment and 12 declining.

The findings matter because liraglutide and other GLP-1 receptor agonists are now among the most prescribed drugs for weight loss and metabolic disease, yet scientists still don't fully understand how they work. Previous studies had shown these drugs alter the microbiome, but it was unclear whether those changes were fleeting side effects or fundamental rewiring that persisted after treatment ended. This mouse study suggests the former—at least in the short term. The drug appears to temporarily favor bacteria associated with mucin and bile acid metabolism, which could influence how the gut functions, but the effect seems to wear off once dosing stops.

The researchers were careful to note the limits of their work. The study used only male mice on a high-fat diet, so the findings may not apply to females or to people eating different diets. They didn't measure whether the bacterial shifts actually benefited or harmed the animals' health, nor did they identify the mechanisms driving the changes. They also couldn't determine whether similar effects occur in humans, where the microbiome is far more complex and the timeline of change might be different. The study lasted only three weeks, so it says nothing about what happens if someone takes liraglutide for months or years.

The authors called for longer studies that include female animals, measure actual food intake, perform functional analyses of what the bacteria are doing metabolically, and most importantly, test the same questions in human volunteers. They want to know whether microbiome changes persist with extended treatment, whether they differ based on diet or sex, and whether they actually contribute to the drug's weight-loss and metabolic benefits or are merely incidental. Until those studies are done, the microbiome remains one of the drug's least understood territories—a temporary landscape that shifts and then settles back, its true significance still unmapped.