OBJECTIVE: Dietary therapies for irritable bowel syndrome (IBS) have received increasing interest but predicting which patients will benefit remains a challenge due to a lack of mechanistic insight. We recently found evidence of a role for the microbiota in dietary modulation of pain signalling in a humanised mouse model of IBS. This randomised cross-over study aimed to test the hypothesis that pain relief following reduced consumption of fermentable carbohydrates is the result of changes in luminal neuroactive metabolites. DESIGN: IBS (Rome IV) participants underwent four trial periods: two non-intervention periods, followed by a diet low (LFD) and high in fermentable carbohydrates for 3 weeks each. At the end of each period, participants completed questionnaires and provided stool. The effects of faecal supernatants (FS) collected before (IBS FS) and after a LFD (LFD FS) on nociceptive afferent neurons were assessed in mice using patch-clamp and ex vivo colonic afferent nerve recording techniques. RESULTS: Total IBS symptom severity score and abdominal pain were reduced by the LFD (N=25; p<0.01). Excitability of neurons was increased in response to IBS FS, but this effect was reduced (p<0.01) with LFD FS from pain-responders. IBS FS from pain-responders increased mechanosensitivity of nociceptive afferent nerve axons (p<0.001), an effect lost following LFD FS administration (p=NS) or when IBS FS was administered in the presence of antagonists of histamine receptors or protease inhibitors. CONCLUSIONS: In a subset of IBS patients with improvement in abdominal pain following a LFD, there is a decrease in pronociceptive signalling from FS, suggesting that changes in luminal mediators may contribute to symptom response.