Abstract Background Non-steroidal anti-inflammatory drugs (NSAIDs), despite their widespread analgesic and anti-inflammatory use, can cause NSAID-induced gastrointestinal injury and exacerbate disease activity in patients with chronic inflammatory bowel disease. As no highly effective strategy currently exists to prevent or treat NSAID-related bowel injury, this remains an important unmet clinical need. We hypothesize that extracellular vesicles derived from Bifidobacterium longum subsp. infantis (B. infantis–EVs) may mitigate NSAID-induced colonic injury by modulating immune responses and restoring gut microbial homeostasis. Methods B. infantis–EVs were purified by tangential flow filtration and characterised by nanoparticle tracking analysis and electron microscopy. For in vivo experiments, bowel injury was induced in 7-week-old C57BL/6 mice by oral administration of indomethacin (IND, 20 mg/kg) on day 8 (IND-only injury model). A more severe inflammatory injury model was generated by providing 2% dextran sodium sulfate in drinking water for 5 days followed by IND administration on day 8 (DSS + IND model). Mice in both models received oral B. infantis–EVs (1 × 1010 particles) every other day for a total of five doses. Disease severity was evaluated by Disease Activity Index(DAI), colon length, histopathology, and inflammatory cytokine levels. Fecal microbiota composition was assessed by 16S rRNA metagenomic sequencing. Results B. infantis–EVs showed protective effects in both the IND-only and DSS+IND colitis models. In the IND-only model, EVs treatment reduced DAI progression and significantly lowered histopathological scores, indicating attenuation of NSAID-induced mucosal injury. In the DSS+IND model, which exhibited the most severe inflammation, B. infantis–EVs markedly improved outcomes, including reduced DAI (p 0.01), preserved colon length (p 0.001), and substantially improved histological architecture (p 0.0001), with notable reductions in mucosal erosion, epithelial detachment, and focal hemorrhage. They also decreased the expression of pro-inflammatory cytokines (IL-1β, IL-2, TNF-α) and upregulated the anti-inflammatory cytokine TGF-β in both IND-only and DSS+IND models. Furthermore, 16S rRNA profiling showed that B. infantis–EVs restored gut microbial balance by decreasing pro-inflammatory taxa like Bacteroides vulgatus and increasing beneficial microbes such as Akkermansia, Turicibacter, and Lactobacillus spp. Conclusion Taken together, our results demonstrate that B. infantis-EVs protect against NSAID-induced colonic injury by reducing disease activity, preserving mucosal architecture, and restoring gut microbial homeostasis, highlighting their potential as a microbiota-based therapeutic strategy for NSAID-associated colitis. Conflict of interest: Prof. Dr. Kim, Seong-Eun: No conflict of interest Jeon, Hyeon-Jeong: No conflict of interest Yoon, Min Joo: No conflict of interest Yu, Suna: No conflict of interest Bae, Seryun: No conflict of interest Jeong, Eui Sun: No conflict of interest Song, Eun Mi: No conflict of interest Byeon, Ju Ran: No conflict of interest Park, Yehyun: No conflict of interest Choe, A Reum: No conflict of interest Tae, Chung Hyun: No conflict of interest Hye-Kyung, Jung: No conflict of interest Ki-Nam, Shim: No conflict of interest Jung, Sung-Ae: No conflict of interest
Kim et al. (Thu,) studied this question.