Nanoclustering and signaling of KRAS G12C and KRAS G12D respond to lipid acyl chain remodeling in an allele-specific manner

Abstract

Small GTPase KRAS mutated at hotspots, such as G12, G13 and Q61, are major drivers of cancer and display allele-specific oncogenic properties, which are not well understood. KRAS mutants require precise spatiotemporal distribution to the proteolipid nanoclusters on the plasma membrane (PM) for efficient signaling. We recently reported allele-specific lipid sensing of KRAS mutants. KRASG12D, KRASG12V and KRASQ61H favor the unsaturated phosphatidylserine (PS), while KRASG12C and KRASG13D gain enrichment of the saturated PS, cholesterol and/or phosphoinositol 4,5-bisphosphate (PIP2). We, here, examined how the allele-specific lipid sensing of KRAS mutants contributes to their allele-specific signaling and activities. We now show that the stable expression of lysophosphatidylcholine acyltransferase 1 (LPCAT1) elevates the saturated phospholipids and reduces the mixed-chain lipids, especially PS species. Our super-resolution electron microscopy (EM)-spatial analysis revealed that the LPCAT1 expression perturbs the PM nanoclustering of KRASG12D, without affecting that of KRASG12C. LPCAT1 suppresses the KRAS-dependent mitogen-activated protein kinases (MAPKs) signaling and the MAPK-regulated proliferation and colony formation of the KRASG12D-expressing human pancreatic PANC1 cells, while promoting those of the KRASG12C-expressing MiaPaCa-2 cells. Mouse embryonic fibroblasts (MEF) transformed with KRASG12C contain more saturated lipids than those expressing KRASG12D. Concordantly, patient tumor genomics analysis illustrated that expression of LPCAT1 and KRAS mutants negatively correlate in pancreatic adenocarcinoma with KRASG12D as a dominant driver, but loses correlation in lung adenocarcinoma with KRASG12C as a major driver. Thus, the allele-specific lipid sensing of KRAS mutants contributes to their pathological activities.