Abstract
Colorectal cancer (CRC) is the second leading cause of cancer death in New Zealand. Patients with inflammatory bowel disease (IBD) are at an elevated risk of developing CRC. Chronic inflammation in IBD is a critical predisposing factor in CRC progression. Aberrant immune response and chronic inflammation can emerge from changes in gut microbiota composition in genetically susceptible hosts. Microbial dysbiosis subverts intestinal homeostasis and impairs epithelial barrier function, thereby perpetuating inflammation and promoting carcinogenesis. High-throughput multi-omics analyses have discovered significant distinctions in microbial diversity and taxonomic and functional composition between healthy subjects and patients with IBD or CRC. However, it remains unknown if the microbial abundance and distribution differ between disease statuses and how that may influence the host responses. In addition, spatial information regarding bacterial distribution and abundance within the colonic tissue of IBD and IBD-CRC patients is currently unavailable.
The present study used RNAscope, a novel in situ RNA hybridisation assay, to characterise bacterial distribution in normal, inflamed and progressed (dysplastic/tumour) tissues in IBD and IBD-CRC patients. In silico quantification of 16S rRNA signal was performed using QuPath bioimage analysis platform to determine relative bacterial abundance. RNAscope results demonstrated extensive intra-tissue heterogeneity in bacterial distribution and interindividual variation in relative bacterial abundance. Bacterial 16S rRNA signals were primarily present in the mucus layer and colonic crypts. In addition, differences in mucus thickness and crypt morphology were observed across normal, inflamed and progressed tissue.
Contrary to our hypothesis, there was no evidence suggesting that the relative bacterial abundance drastically differed across normal, inflamed and progressed tissues in progressors and IBD controls. However, the presence of immune cells was significantly associated with local bacterial abundance. An attempt to correlate bacterial abundance and host expression of the lysozyme gene (LYZ) found no significant relationship between relative bacterial abundance and host expression of LYZ across all tissue types. RT-qPCR detected a low level of pan-bacterial 16S rRNA and Fusobacterium nucleatum RNA (Fn RNA) in CRC and IBD- CRC tumours. Relative 16S rRNA and Fn RNA level did not vary significantly between CRC and IBD-CRC tumours.
This preliminary study presents a unique picture of bacterial distribution and abundance in the colonic tissues of IBD and IBD-CRC patients at single-cell resolution. However, the small sample size (n =14) and high variability in the dataset may affect the reliability of statistical analyses. Future work should include larger cohorts to ascertain the findings from this study and to measure the impact of IBD subtypes, disease duration, extent of inflammation, and treatment history on bacterial abundance and host immune response, within the relevant spatial context.