DOP023 Bifidobacterium-candida warfare in the gut of patients with ulcerative colitis

Abstract Background C. albicans is identified more frequently in the gut of patients with active ulcerative colitis (UC), and may drive inflammation through activation of specific fungal receptors in the intestine. Pathogenic expansion of C. albicans in UC may occur due to loss of specific protective bacterial strains. Prior work has suggested that Bifidobacterium species may antagonize Candida species. However, relationships of Bifidobacterium species with Candida species in patients with UC are not known, and whether Bifidobacterium may mechanistically suppress C. albicans growth is unclear. Methods We utilized data from the Study of a Prospective Adult Research Cohort with IBD to interrogate relationships between fungal and bacterial species during varying inflammatory phases of UC. Relative abundances of fungal and bacterial taxa was obtained from 18 internal transcribed spacer sequences and shotgun sequences. Spearman’s correlations between fungal and bacterial relative abundance were performed. To further assess observed bacterial-fungal correlations in vitro, cell-free supernatants from lab-derived and patient-derived strains were utilized to assess their suppressive ability in competitive assays. Results We found that C. albicans is increased in active UC vs during remission (q-val <0.05). Across all patients, 15 unique Bifidobacterium species were identified, including B. adolescentis, B. longum, and B. bifidum [Fig 1A]. The relative abundance of fecal B. adolescentis was negatively associated with C. albicans abundance in UC (p<0.05) [1B]. Multiple strains of B. adolescentis were detected [1C], although these strains did not demonstrate significant genetic variability when compared to a reference marker [1D]. We further observed that cell-free supernatants from lab-derived strains of both Bifidobacterium adolescentis (ATCC 15703) and Bifidobacterium bifidum (ATCC 29521) were able to inhibit C. albicans (ATCC 18804) growth [2CD], while cell-free supernatants from control species Blautia producta (ATCC 27340) only showed partial inhibition of C. albicans [2B]. Using donated fecal samples from 7 patients with active UC, we were able to isolate in culture 11 separate isolates of Bifidobacterium strains (confirmed by MALDI-TOF), and 1 isolate of Candida species, with inhibitory assays between patient derived strains ongoing. Conclusion Bifidobacterium adolescentis is negatively associated with C. albicans in fecal samples from UC patients. Both B. adolescentis and B. bifidum strains completely suppress C. albicans growth in vitro. Bifidobacterium species may provide colonization resistance against C. albicans in the UC gut. Loss of Bifidobacterium strains may allow for C. albicans expansion which may contribute to inflammation and relapse.