The Role of Prostaglandins and a High-fat Diet in Cancer Progression and Metastasis

Abstract

Almost all colorectal cancer (CRC) deaths are due to metastasis, yet there are currently no FDA-approved drugs to halt or prevent the spread of cancer cells to distant organs. A better understanding of the underlying biology of CRC metastasis could lead to the next generation of therapeutics designed to block steps in the metastatic cascade. Prior research provided a model to understand tumor metastasis through sequential stages of invasion, migration into the vasculature, circulation, extravasation, and colonization of distant organs. Each step heavily depends on diverse cell-cell communications and the microenvironment, which has often not been well recapitulated in previous in vitro and in vivo model systems to study CRC metastasis. However, the recent development of CRISPR-engineered 3D colorectal cancer organoids (CRCOs) and colonoscopy-guided orthotopic CRC transplantation models have provided a more physiological model system to study metastasis. The CRCOs in the Yilmaz Lab are precisely engineered with genomic alterations found in most human CRCs, such as perturbations in the Apc, Kras, and TP53 genes (AKP). These engineered CRCOs can be transplanted into different models of CRC metastasis (e.g., injected orthotopically into the colonic mucosa or into the portal circulation via the spleen), where they recapitulate all or key steps of interest in the metastatic cascade. We have used the orthotopic transplantation model to identify genes potentially important for CRC metastasis by assessing differentially expressed genes in primary tumor- and liver metastasis-derived AKP CRCOs. Profiling these primary tumor and liver metastasis-derived CRCOs by RNA sequencing (RNA-seq) revealed numerous differentially expressed genes that could have an important role in CRC metastasis, including a significant downregulation of Ptgs1 in liver metastases. Ptgs1 encodes for cyclooxygenase 1 (COX-1), the rate-limiting enzyme for prostaglandin synthesis. Prostaglandin E2 (PGE2) is the primary prostaglandin produced by CRC and is thought to promote tumorigenesis and metastasis. Our finding that CRC liver metastases decrease their production of PGE2 was inconsistent with the prevailing view that PGE2 fosters metastasis. Additionally, recent clinical data suggested prostaglandins may inhibit CRC metastasis, motivating us further to explore the potential inhibitory role of prostaglandins in CRC metastasis.

To study the role of PGE2 in CRC metastasis, we expanded our ex vivo tumor colon organoid systems and assessed them in models of CRC metastasis (Chapter 2). We generated AKP CRCOs deficient in Ptgs1 and Ptgs2 (cannot make PGE2), Hpgd (cannot inactivate PGE2), and Ptger1 and Ptger4 (cannot respond to PGE2) using CRISPR. By assessing these engineered CRCOs in our metastasis models, we have found that prostaglandins promote metastatic initiation but inhibit metastatic outgrowth through a tumor-cell extrinsic mechanism. Consistent with our genetic knockout models, we have identified for the first time that aspirin, an inhibitor of COX-1 and cyclooxygenase 2 (COX-2), promotes the outgrowth of liver metastases. We further applied the physiological model systems in the Yilmaz Lab to study the role of a pro-obesity, high-fat diet (HFD) on CRC metastasis (Chapter 3). We found that an HFD promotes metastasis independent of mature lymphocytes or enhanced insulin signaling. Further, we have sought to identify the mechanism by which an HFD promotes metastasis using single-cell RNA sequencing (scRNA-seq) of primary tumors from the control and HFD cohorts. scRNA-seq revealed that an HFD enhances a tumor-specific stemness program and may increase specific mesenchymal and myeloid populations in the primary tumor. Numerous genes were significantly differentially expressed in the CRC cells on the HFD and control diet (CD). We prioritized and selected several differentially expressed gene candidates that may play a causative role in an HFD promoting metastasis and began to generate CRISPR knockouts in the AKP CRCOs. These knockouts are currently being prepared for assessment in our orthotopic mouse model of metastasis.

Our new understanding of how prostaglandins and an HFD influence metastasis has clinical implications. While prostaglandins have previously only been thought to promote CRC progression and metastasis, we found prostaglandins inhibit metastatic colonization and outgrowth, suggesting aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) may have additional risk in CRC patients. Conversely, PGE2 or PGE2 receptor agonists may prove therapeutic to treat CRC metastases. We have also identified and are knocking out several genes that may be mechanistically linked to an HFD-promoting CRC metastasis. These new insights into the role of prostaglandins and an HFD in CRC metastasis have implications for cancer prevention and therapeutics.