Common Gut Bacteria Can Break Down Collagen, Study Finds

If you have read that your gut lining depends on a scaffold of collagen and other structural proteins to stay intact, a new laboratory study adds a wrinkle: some of the bacteria that live in your intestines carry the enzymatic tools to break that scaffold down. Researchers systematically tested 11 species of Bacteroides, one of the most common bacterial genera in the human gut, and found that all of them carry genes for enzymes capable of degrading extracellular matrix (ECM, the network of structural proteins and sugars that holds tissue together) components including collagen, elastin, and fibronectin. The work was done entirely in the lab, not in a living gut, but it reframes a question that matters for anyone managing inflammatory bowel disease (IBD) or curious about how gut bacteria interact with the tissue around them.

Key takeaways:

• All 11 tested Bacteroides species carry genes for enzymes that target extracellular matrix (ECM) proteins, and every species degraded at least one ECM component in lab dish tests.
• Metalloprotease and serine protease enzymes were identified as the main drivers of collagen and elastin breakdown; hyaluronic acid was the single most-degraded component, with some species breaking down more than 90% of it.
• This is a lab-based genomic and biochemical study only. It was not tested in animals or people, so it does not show this happens inside a living gut.
• A species genomic potential did not reliably predict how much ECM it degraded in the dish, meaning gene counts alone cannot be used to infer real-world activity.

Bacteroides Gut Bacteria Encode Enzymes That Digest Collagen and Elastin

Every one of the 11 Bacteroides species tested carries genes for enzymes that target the extracellular matrix (ECM), the meshwork of collagen, elastin, fibronectin, and sugar-based molecules called proteoglycans and glycosaminoglycans that gives intestinal tissue its structure, according to a study published in Cellular and Molecular Bioengineering. Comparative genomics found each genome encoded between 33 and 77 ECM-associated carbohydrate-active enzymes (CAZymes) and 27 to 45 ECM-associated proteases, genes that in some species made up close to a fifth of their entire protease repertoire.

Genomic potential is one thing; actual activity is another, so the researchers also incubated culture fluid from each species with purified ECM proteins in the lab. Every species degraded multiple ECM components, though which ones and by how much varied widely. Hyaluronic acid (a lubricating, water-binding molecule found throughout connective tissue) was the most heavily degraded substrate on average, with an average breakdown of 60.7% and one species, Bacteroides caccae, degrading 95.3% of the initial amount. Chondroitin sulfate and heparan sulfate, two other sugar-based ECM components, were also broken down, and several species were able to use chondroitin sulfate as their sole food source for growth. Collagen I, collagen IV, elastin, and fibronectin were degraded as well, with Bacteroides fragilis showing the strongest collagen- and gelatin-degrading activity of the group.

Why This Could Matter for IBD and Gut Barrier Health

The intestinal ECM is not just scaffolding. It shapes how the gut lining heals after injury, how immune cells patrol the tissue, and how the epithelial barrier that separates gut contents from the rest of the body stays sealed. Disturbed ECM remodeling is already a recognized feature of inflammatory bowel disease, and the study's authors note that prior research has linked proteolytic (protein-degrading) activity from Bacteroidales bacteria to the onset and severity of ulcerative colitis, to IBD progression, and even to leaks at surgical bowel reconnection sites. That backdrop is why a systematic look at which gut bacteria can chew through ECM proteins, and how, is relevant to anyone who researches IBD, ulcerative colitis, Crohn's disease, or gut barrier integrity for themselves or a family member, even though this particular study did not test any of those conditions directly.

How Researchers Tested 11 Gut Bacteria Species in the Lab

The team combined two approaches. First, they scanned the genomes of 11 Bacteroides species, including well known gut residents such as B. fragilis, B. thetaiotaomicron, B. ovatus, and B. uniformis, for genes matching a curated list of 67 known ECM-associated enzymes. Second, they grew each species, collected the fluid the bacteria had secreted into (their culture supernatant), and mixed it with purified ECM proteins, including collagen I, collagen IV, elastin, gelatin, fibronectin, and three glycosaminoglycans, to directly measure how much of each substrate was broken down. Finally, they used chemical inhibitors that selectively block different classes of protein-cutting enzymes (metallo-, serine-, and cysteine-proteases) to identify which enzyme types were doing the work. Metalloprotease inhibition cut gelatin degradation to as little as 31.4% to 50.6% of normal activity across four tested species, and serine protease inhibition had a similarly large effect, pinning down metalloproteases and serine proteases as the primary collagen-degrading enzymes, and metalloproteases as the primary elastin-degrading enzymes.

What This Could Mean for Someone Managing Gut Health

Nothing here translates into a specific action today. The study did not test whether this ECM-degrading activity happens inside a living intestine, whether it contributes to disease, or whether any current diet, supplement, or probiotic changes it. What it does establish is a mechanism worth watching: a widespread, previously underappreciated capability of common gut bacteria that overlaps with tissue processes already implicated in IBD and gut barrier disruption. If you manage a condition involving gut barrier integrity or inflammation, this is the kind of research that may eventually inform questions worth raising with a gastroenterologist, such as how tissue remodeling and the microbiome interact, rather than a reason to change anything about your care now.

Caveats: This Was Lab Dish Work, Not a Living Gut

Three limits matter here. First, and most important, the entire study was conducted with purified proteins, bacterial cultures, and genomic data. No animals or people were involved, so nothing here shows that this degradation happens inside an actual intestine, where physical structure, other microbes, and host cells all change the picture. Second, the correlation between a species genomic potential and its measured degrading activity was weak (a statistical test called a Mantel test returned r = 0.047, p = 0.36, essentially no meaningful relationship), meaning you cannot predict which bacteria will be most active just by counting their genes. Third, the functional experiments relied on small replicate counts, three to five biological replicates per species, across all 11 species, which limits how confidently the exact degradation percentages can be generalized.

What Comes Next

The study's authors say the logical next step is testing ECM degradation using more realistic conditions: intact tissue rather than purified proteins, mixed microbial communities rather than single species, and bacterial strains isolated from people with IBD or other ECM-related diseases rather than reference lab strains. No timeline for that follow-up work was given.

Common Questions

Does this mean gut bacteria are damaging my gut lining right now? No. This study tested purified proteins and bacterial samples in lab dishes, not a living gut, so it cannot say whether or how much this occurs inside an actual intestine.

Which gut bacteria were studied? Researchers tested 11 species in the genus Bacteroides, one of the most abundant bacterial groups in the human gut, including Bacteroides fragilis, Bacteroides thetaiotaomicron, and Bacteroides ovatus.

Which enzymes break down collagen? Protease inhibitor experiments showed metalloproteases and serine proteases were the main enzymes degrading collagen, while metalloproteases were the biggest contributor to elastin breakdown.

Is this connected to inflammatory bowel disease? The study did not test IBD patients directly, but its authors point to earlier research linking Bacteroidales proteolytic activity to ulcerative colitis severity and IBD progression, part of why they argue this capability deserves further study.

References

Karen M Mancera Azamar, Krishna Rajesh, James Garner, et al. (2026). Comparative Genomic and Functional Profiling of ECM-Associated Enzymes in Bacteroides, a Key Genus of the Human Gut Microbiome. Cellular and Molecular Bioengineering. https://doi.org/10.1007/s12195-026-00938-6

Inflammatory Bowel Disease-Associated Gut Commensals Degrade Components of the Extracellular Matrix. mBio. https://doi.org/10.1128/mbio.02201-22

Current understanding of the interplay between extracellular matrix remodelling and gut permeability in health and disease. Cell Death Discovery. https://www.nature.com/articles/s41420-024-02015-1