A combined proteomics and Mendelian randomization approach to investigate the effects of aspirin-targeted proteins on colorectal cancer

Nounu, Aayah; Greenhough, Alexander; Heesom, Kate J.; Richmond, Rebecca C.; Zheng, Jie; Weinstein, Stephanie J.; Albanes, Demetrius; Baron, John A.; Hopper, John L.; Figueiredo, Jane C.; Newcomb, Polly A.; Lindor, Noralane M.; Casey, Graham; Platz, Elizabeth A.; Le Marchand, Loïc; Ulrich, Cornelia M.; Li, Christopher I.; van Duijnhoven, Fränzel J. B.; Gsur, Andrea; Campbell, Peter T.; Moreno, Víctor; Vodicka, Pavel; Vodickova, Ludmila; Brenner, Hermann; Chang-Claude, Jenny; Hoffmeister, Michael; Sakoda, Lori C.; Slattery, Martha L.; Schoen, Robert E.; Gunter, Marc J.; Castellví-Bel, Sergi; Kim, Hyeong Rok; Kweon, Sun-Seog; Chan, Andrew T.; Li, Li; Zheng, Wei; Bishop, D. Timothy; Buchanan, Daniel D.; Giles, Graham G.; Gruber, Stephen B.; Rennert, Gad; Stadler, Zsofia K.; Harrison, Tabitha A.; Lin, Yi; Keku, Temitope O.; Woods, Michael O.; Schafmayer, Clemens; Van Guelpen, Bethany; Gallinger, Steven; Hampel, Heather; Berndt, Sonja I.; Pharoah, Paul D. P.; Lindblom, Annika; Wolk, Alicja;...

Abstract

Background: Evidence for aspirin’s chemopreventative properties on colorectal cancer (CRC) is substantial, but its mechanism of action is not well-understood. We combined a proteomic approach with Mendelian randomization (MR) to identify possible new aspirin targets that decrease CRC risk. Methods: Human colorectal adenoma cells (RG/C2) were treated with aspirin (24 hours) and a stable isotope labeling with amino acids in cell culture (SILAC) based proteomics approach identified altered protein expression. Protein quantitative trait loci (pQTLs) from INTERVAL (N=3,301) and expression QTLs (eQTLs) from the eQTLGen Consortium (N=31,684) were used as genetic proxies for protein and mRNA expression levels. Two-sample MR of mRNA/protein expression on CRC risk was performed using eQTL/pQTL data combined with CRC genetic summary data from the Colon Cancer Family Registry (CCFR), Colorectal Transdisciplinary (CORECT), Genetics and Epidemiology of Colorectal Cancer (GECCO) consortia and UK Biobank (55,168 cases and 65,160 controls). Results: Altered expression was detected for 125/5886 proteins. Of these, aspirin decreased MCM6, RRM2, and ARFIP2 expression, and MR analysis showed that a standard deviation increase in mRNA/protein expression was associated with increased CRC risk (OR: 1.08, 95% CI, 1.03–1.13; OR: 3.33, 95% CI, 2.46–4.50; and OR: 1.15, 95% CI, 1.02–1.29, respectively). Conclusions: MCM6 and RRM2 are involved in DNA repair whereby reduced expression may lead to increased DNA aberrations and ultimately cancer cell death, whereas ARFIP2 is involved in actin cytoskeletal regulation, indicating a possible role in aspirin’s reduction of metastasis. Impact: Our approach has shown how laboratory experiments and population-based approaches can combine to identify aspirin-targeted proteins possibly affecting CRC risk.

• Journal Article Type: Article
• Acceptance Date: Dec 9, 2020
• Online Publication Date: Dec 14, 2020
• Publication Date: Mar 1, 2021
• Deposit Date: Jan 19, 2021
• Publicly Available Date: Dec 15, 2021
• Journal: Cancer Epidemiology Biomarkers and Prevention
• Print ISSN: 1055-9965
• Electronic ISSN: 1538-7755
• Publisher: American Association for Cancer Research
• Peer Reviewed: Peer Reviewed
• Volume: 30
• Issue: 3
• Pages: 564-575
• DOI: https://doi.org/10.1158/1055-9965.EPI-20-1176
• Keywords: Cancer, Bioinformatics, Epidemiology, Aspirin,
• Public URL: https://uwe-repository.worktribe.com/output/7002197
• Publisher URL: https://cebp.aacrjournals.org/content/early/2020/12/12/1055-9965.EPI-20-1176.long
• Additional Information: Author contributions. A. Greenhough: Conceptualization, data; curation, investigation, methodology, writing-review and editing.