Aspirin and Gut Biology: Mucosal Proteomics and Metabolomics in a Randomized Trial

The majority of colorectal cancers (CRC) are sporadic, arising from adenomatous polyps in otherwise healthy tissue. Regular aspirin use has been shown to reduce polyp formation and CRC incidence in observational studies and randomized trials. It has been hypothesized that aspirin exerts its effects through inhibition of cyclooxygenases (COX), thereby blocking prostaglandin E2 (PGE2) synthesis. However, other associations in human observational studies and experimental in vitro models suggest that aspirin may also have effects on non-COX-mediated pathways, such as NF-κB and PI3K signaling, Wnt/β-catenin activity, DNA mismatch repair, proliferation and apoptosis, and modulation of antioxidant enzymes. Thus, the understanding of the chemopreventive mechanisms of aspirin is still evolving. Given that hypothesis-driven insights into the chemopreventive effects of aspirin have largely been exhausted, discovery-based approaches are warranted. This proposal, supported by a team of established investigators in proteomics, metabolomics, biochemistry, and biostatistics, aims to evaluate effects of aspirin on biomarkers of cancer-risk pathways, directly in colorectal mucosa, in a controlled trial using state-of-the-art high-dimensional ‘omics’ technologies. We will use samples and data collected as part of a controlled, randomized crossover trial to address the following aims: 1) Measure, in rectal mucosal biopsies, the effects of aspirin supplementation compared to placebo on proteomic biomarkers of regulatory pathways important in cancer susceptibility; 2) Measure the effects of aspirin supplementation compared to placebo on the metabolome and lipidome in rectal mucosal biopsies; and 3) Integrate the various omics data, and compare proteomics and metabolomics results from mucosal biopsies to measures previously assayed in plasma using the same platform. The completed trial, Aspirin and Biology of the Colon (ABC) Study, included 42 participants (18 men and 24 women) who were given aspirin (325 mg/d) and an identical-appearing placebo for 60 days each, with a 3-month washout between intervention periods. Colorectal biopsies were collected at the end of each intervention period. We will use a high-density antibody array to measure the rectal mucosal proteome, and will characterize the rectal mucosal metabolome and lipidome using global approaches via an ultrahigh-performance liquid chromatography-tandem mass spectroscopy platform. We will use well-characterized statistical methods for integrating the omics data generated, and test for associations using single-marker and pathway-enrichment statistical models, which adjust for demographic variation. Further, we will compare our results to the plasma proteomics and metabolomics data we generated previously in the same study participants. In summary, our project aims to elucidate the mechanisms of chemopreventive actions and targets of aspirin, directly in the tissue of interest, which may also identify new pathways for development of other chemotherapeutic agents.