Mechanisms of colonic tumor initiation suggest that cancer can develop through multiple genetic and epigenetic pathways. Human epidemiologic studies have shown the presence of cancer is associated with changes in the complex gut microbiota (GM). To determine if differences in the complex GM directly cause differences in cancer susceptibility, we used complex microbiota targeted rederivation (CMTR) to rederive isogenic embryos of the F344/NTac-ApcPirc/+ (Pirc) rat model of familial adenomatous polyposis, using three different strains of surrogate dams (F344/NHsd (F344), LEW/SsNHsd (LEW), and Crl:SD (SD)) each harboring distinct gut microbiota. Fecal samples from the rederived pups were collected from weaning up to 180 days. The GM composition was characterized to the operational taxonomic unit (OTU) level by extracting bacterial DNA from fecal material and sequencing the 16S rRNA gene V4 hypervariable region on the Illumina MiSeq platform. Adenoma development was also monitored longitudinally via colonoscopy and tumor multiplicity was determined at 6 months.
We found that the GM varied between dams, and that the pups’ microbiota resembled their surrogates by 1.5 months. Two rats with the LEW GM did not develop any colonic tumors, decreasing the Pirc phenotype penetrance from 100% to 87%. All rats with this GM also had a significantly reduced colonic tumor burden compared to both F344 GM (p<0.001) and SD GM (p<0.05). The relative abundance of certain OTUs including family Peptococcaceae and Akkermansia muciniphila were positively correlated with tumor multiplicity in the F344 GM (p<0.05). A previously reported sex bias in colonic tumor burden was also confirmed with males showing more adenomas than females (p<0.05). To assess functional changes associated with variation in GM, PICRUSt (Phylogenetic Investigation of Communities by Reconstruction of Unobserved States) was used and showed elevated expression of H2S-related bacterial genes in the LEW GM.
To assess whether tumors from rats with diverse GM differed in loss of heterozygosity (LOH), we used a quantitative allele-specific pyrosequencing assay to determine the allelic ratio of the genomic ApcPirc loci in colonic tumors. Approximately 70% of tumors from rats with F344 GM underwent LOH with 30% tumors maintaining heterozygosity of the Apc allele, consistent with previous reports. Interestingly, rats with SD and LEW GM showed 100% and 96% LOH respectively, suggesting a possible link between differing complex microbiota and loss of the wildtype Apc allele. Our study provides new insight into the role of gut microbiota as a modulator and predictor of disease phenotype in this rat model.