Bariatric Surgery Rewires the Gut Microbiome at the Protein Level, Study Finds

Two years after bariatric surgery, the gut microbiome looks profoundly different, not just in which microbes are present, but in what those microbes are actually doing. A new study published June 2026 in mSystems tracked 45 patients with severe obesity through two types of gastric bypass and found sweeping changes in both microbial and human proteins in fecal samples, offering one of the most detailed pictures yet of how this surgery reshapes gut biology at a functional level.

Most earlier work on bariatric surgery and the gut has focused on which bacterial species increase or decrease. This team, led by Antonio Tanca and colleagues, asked a harder question: what are those bacteria actually producing, and how do the host's own secreted proteins change in response?

Microbial protein shifts after Roux-en-Y and OAGB bypass

The researchers analyzed fecal samples from 45 patients with severe obesity (BMI above 35 kg/m2) at a preoperative baseline and again two years after surgery. About half underwent Roux-en-Y gastric bypass (RYGB) and half underwent one anastomosis gastric bypass (OAGB), two procedures that both reroute the small intestine to reduce nutrient absorption. Using shotgun metaproteomics, a technique that identifies proteins directly from complex biological mixtures rather than inferring function from DNA alone, the team catalogued more than 101,600 non-redundant peptide sequences, the overwhelming majority of them microbial.

The approach, described as metaproteomics for gut microbiome function, captures active biology rather than genetic potential. A microbe can carry a gene without expressing the protein; metaproteomics catches only what the gut is actually doing at the moment of sampling.

Akkermansia and Streptococcus rise as Faecalibacterium retreats

Five genera increased significantly two years after surgery: Akkermansia, Anaerotignum, Desulfovibrio, Streptococcus, and Veillonella. Two genera dropped: Faecalibacterium and Romboutsia. These are not arbitrary names on a list; each shift carries metabolic consequences.

The rise in Streptococcus is coupled with elevated glycolytic activity, and its increase aligns logically with Veillonella, a genus that consumes the lactate that Streptococcus produces. Anaerotignum, now more abundant, generates acetate and propionate, two short-chain fatty acids with roles in appetite regulation and intestinal barrier function. At the same time, Faecalibacterium, widely recognized as a keystone producer of butyrate, a short-chain fatty acid associated with gut anti-inflammatory activity, declined substantially. The authors suggest that elevated bile acid concentrations following surgical rerouting, particularly chenodeoxycholic acid and deoxycholic acid, may inhibit Faecalibacterium growth. This mirrors patterns documented in long-term bariatric microbiota research, where metabolic improvement and microbial reorganization do not always move in the same direction.

Increased Desulfovibrio is particularly notable given its association with altered bile acid metabolism, which the authors flag as a potential contributor to gallstone formation, one of the more common complications following bariatric procedures.

What the host's own proteins reveal about surgical rerouting

The human side of the fecal proteome told an equally detailed story. Seventy-four human proteins increased significantly two years post-surgery, the majority immune-related. Lactotransferrin, an antimicrobial protein with immunomodulatory properties, rose substantially. So did meprin A, involved in mucosal homeostasis, along with neutrophil elastase, myeloperoxidase, and S100-A8, a cluster that points toward heightened innate immune activity in the gut lining.

REG3A, an antimicrobial peptide secreted by intestinal epithelial cells, also increased. Beyond its defensive role, REG3A has reported functions in glucose homeostasis and insulin resistance, suggesting a potential link between microbiome remodeling and the metabolic improvements that bariatric surgery reliably produces. Earlier mouse model work showed that gut microbiota after bariatric surgery can directly lower intestinal glucose absorption independently of weight loss, and the REG3A finding adds a plausible host-side mechanism to that picture.

On the other side of the ledger, fourteen human proteins decreased, primarily digestive enzymes originating from pancreatic and gastric secretions. This makes anatomical sense: surgical rerouting physically excludes the stomach and proximal duodenum from the food stream, reducing the stimulus for those secretions.

Limitations and the gap from a healthy gut

The study's 45-patient sample gave adequate statistical power for within-group comparisons, but the authors pooled RYGB and OAGB because analyzing them separately would leave too few participants per arm. Whether the two procedures drive different protein-level outcomes remains an open question.

More importantly, the post-surgical microbiome did not resemble that of a healthy, normal-weight population. The post-operative state represented a distinct functional equilibrium, diverging further from healthy controls than the pre-surgical baseline did in some respects. This fits a broader pattern: as research has noted on incomplete microbial richness restoration after bariatric surgery, metabolic improvement and microbiome normalization are related but separable outcomes. The authors also compared their cohort to a healthy reference drawn from heterogeneous external datasets, a necessary but imperfect approach that can introduce batch effects between studies.

Correlations between specific microbial proteins and clinical blood markers add an intriguing layer: a transport protein from Dorea correlated with cholesterol levels, and an enzyme from Prevotella correlated with HDL. The authors are careful to note that current data cannot establish a direct mechanistic link between those microbial pathways and host cholesterol metabolism.

The path toward functional biomarkers and mechanistic studies

The clearest near-term implication of this work is methodological. Taxonomy alone, knowing which species are present, misses the protein-level reorganization documented here. As future directions for gut metaproteomics move toward standardization and multi-omics integration, studies like this one provide the functional baseline that larger mechanistic trials will need.

For the roughly 45 patients tracked here, two years of follow-up captured a biological state that is metabolically improved but microbiologically distinct from health. Whether longer follow-up brings the gut proteome closer to a normal-weight reference, or whether the surgically rerouted gut stabilizes in its own distinct state, is the question that next-generation studies will need to answer. Patients who undergo bariatric surgery are already committed to lifelong nutritional supplementation; understanding the microbial protein environment they are managing could eventually inform more targeted dietary and probiotic interventions.

Evidence rating

EMERGING — 80/100

Caveats:

• Prospective cohort design with no non-surgical control group; changes cannot be attributed to surgery alone versus concurrent dietary and lifestyle changes.
• Sample size of 45 patients limits statistical power, and the two surgical procedures (RYGB and OAGB) were pooled rather than analyzed separately.
• No funding source was declared by the authors.
• Comparison with healthy-weight populations relied on external datasets with heterogeneous methods, which may introduce batch effects.
• Mechanistic links between observed microbial and host protein changes and long-term clinical outcomes (e.g., cholesterol, insulin resistance) remain correlational — not yet established as causal.

Prospective multicenter cohort (n=45) applying shotgun metaproteomics to fecal samples before and two years after bariatric bypass surgery. The study identifies significant changes in microbial and host protein abundance but lacks a non-surgical control group, so observed shifts cannot be attributed exclusively to the surgical procedure versus accompanying dietary and behavioral changes. Findings are consistent with prior microbiome compositional studies but have not yet been independently replicated using comparable metaproteomic methods.