267 PCSK9 IS A THERAPEUTIC TARGET IN KRAS-MUTANT COLORECTAL CANCER

• Published in: Gastroenterology (New York, N.Y. 1943) Vol. 160; no. 6; p. S-64
• Main Authors: Wong, Chi Chun, Wu, Jianlin, Chan, Lam-Shing, Ji, Fenfen, Kang, Wei, To, Ka Fai, Sung, Joseph J., Yu, Jun
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.05.2021
• ISSN: 0016-5085; 1528-0012
• DOI: 10.1016/S0016-5085(21)00895-7

KRAS-mutant colorectal cancer (CRC) is an unmet need. Analysis of metabolic gene profile in KRAS-mutant CRC patients and cells revealed the enrichment of cholesterol metabolism. Here, we investigated the cholesterol homeostasis in KRAS-mutant CRC cells and identified the proprotein convertase subtilisin kexin 9 (PCSK9) as a potential therapeutic target. Gene Set Enrichment Analysis was utilized to unravel altered pathways in human KRAS-mutant CRC. Isogenic human normal colon cells (1CT) expressing shAPC (1CT-A), KRASG12V(1CT-K), or both (1CT-AK) were generated. Role of PCSK9 in KRAS-mutant CRC was determined by genetic knockout or PCSK9 inhibitors in cell line and mouse models. GSEA of RNA-seq (TCGA) from KRAS-mutant CRC (N=101) and adjacent normal (N=51) revealed Steroid Biosynthesis as the top enriched metabolic pathway in KRAS-mutant CRC. Up-regulation of PCSK9 mRNA and protein was identified in 1CT-AK compared to 1CT, 1CT-K and 1CT-A cells. PCSK9 was also up-regulated (>5-fold) in tumors of ApcMin/+KrasG12Dvillin-Cre mice. PCSK9 knockout selectively repressed malignant properties of 1CT-AK cells, KRAS-mutant CRC cells (LOVO, SW1116) by inducing apoptosis and cell cycle arrest; but had no effect on normal 1CT cells. PCSK9 knockout attenuated the growth of LOVO and SW1116 xenografts in nude mice, whereas ectopic PCSK9 expression in DLD1 cells exerted opposite effects. Mechanistically, we revealed that PCSK9 repressed LDLR-mediated LDL-cholesterol uptake, which in turn induced cholesterol biosynthesis genes via SREBP2. Stable D2Olabeling and LC-MS/MS revealed increased metabolic flux via mevalonate pathway to synthesize cholesterol in 1CT-AK and SW1116 as compared to 1CT cells, implying a shift from uptake to de novo cholesterol synthesis. While total cholesterol content was unchanged, enhanced mevalonate pathway led to geranylgeranyl diphosphate (GGPP) biosynthesis to higher levels in 1CT-AK and ApcMin/+KrasG12Dvillin-Cre tumors than their normal counterparts. Since GGPP is an intermediate required for KRAS activation, we hypothesized that PCSK9 may regulate KRAS-MEK-ERK via GGPP. Indeed, PCSK9 knockout or its inhibitors (R-IMPP, PF-06446846, Evolocumab) suppressed activated KRAS, p-MEK, p-ERK and Cyclin D1, an effect reversed by GGPP. Consistently, PCSK9 inhibitors suppressed the growth of KRAS-mutant CRC cells, organoids and xenografts. In addition, PCSK9 inhibitors in combination with simvastatin (mevalonate pathway inhibitor) synergized to suppress GGPP, KRAS-MEK-ERK, and tumor growth in vivo. Moreover, PCSK9 mRNA predicts poor survival in KRAS-mutant (N=162, P<.05), but not KRAS-wildtype CRC (N=192) from TCGA dataset. PCSK9 mediates KRAS-mutant CRC growth via GGPP-dependent activation of KRAS-MEK-ERK. PCSK9 inhibitors in combination with simvastatin inhibits KRAS-mutant CRC growth.