Society: AGA

The gastrointestinal tract is a diverse and complex ecosystem shaped by continual interactions between host cells, nutrients, and the gut microbiota. Under homeostatic conditions, the human gut microbiota limits the growth of pathogenic bacteria and restricts the influence of resident pathobionts. However, when the gut microbiota network is disrupted, pathobionts and pathogens can colonize and cause disease. This session focuses on factors which influence the colonization of Clostridioides difficile and other gut pathobionts.

Background. Colonic biofilms are mucus-invasive bacterial aggregates that are associated with colorectal cancer (CRC) in humans and initiate carcinogenesis in mice. Biofilm-positive tissues are also associated with altered metabolism in U.S. and Swedish cohorts. Furthermore, three major biofilm subtypes (polymicrobial, Fusobacteria blooms, and Proteobacteria dominant) have been identified using fluorescent in situ hybridization (FISH). The extent to which prior metabolomic findings extend across diverse global populations and biofilm subtypes is unknown.

Methods. Samples included biopsies from individuals undergoing screening colonoscopy (n = 10, all biofilm-negative) and tumor resections from CRC patients (n = 35) at the University of Malaya in Malaysia. The CRC tissues represented diverse biofilm compositions as determined by FISH: biofilm negative (n = 10), polymicrobial (n = 10), Fusobacteria blooms (n = 10), and Proteobacteria dominant (n = 5). Untargeted metabolomics was performed by HILIC and RPLC-mass spectrometry. Data were processed using ProteoWizard MSConvert, XCMS, CAMERA, MetCleaning, and Metaboanalyst. Enriched pathways were identified using the Mummichog pipeline. Metabolite identification and validation are ongoing and being performed by matching accurate mass and experimental MS/MS data against an in-house library.

Results. Distinct metabolic profiles were detected between biopsy and CRC tissues, biofilm-positive and negative tissues, and biofilm subtypes, p < 0.01, fold-change > 2. Three hundred and fifty-two metabolic features were higher in biofilm-positive tissues, whereas 37 features were higher in biofilm-negative tissues. Twenty-three metabolic pathways were predicted to be differentially enriched across biofilm subtypes, p < 0.05. The ‘C21-steroid hormone biosynthesis’ pathway was most differentially enriched between biofilm-positive and negative CRC tissues, p = 2.0E-14. The ‘prostaglandin formation from arachidonate’ pathway was the only pathway that was differentially enriched between biofilm-positive and negative CRC tissues (p = 2.5E-4), as well as between biofilm-positive subtypes: Fusobacteria blooms vs Proteobacteria dominant (p = 3.5E-3), Fusobacteria blooms vs polymicrobial (p = 1.8E-6), and polymicrobial vs Proteobacteria dominant (p = 0.01). By clustering analysis, the Fusobacteria blooms subtype displayed a distinct pattern versus other subtypes.

Conclusions. Biofilm subtypes have distinct metabolic profiles in a Malaysian cohort. Subtypes may exist upon a continuum or be functionally redundant because certain biofilm subtypes are more metabolically similar than others. Findings align with prior work in international cohorts that demonstrates metabolic differences between biopsy and CRC tissues as well as biofilm-positive and negative samples.

Jamuna Vadivelu

Appearances