Targeting mitochondrial dynamics in KRAS-mutant cancer cells through pharmacological modulation with M1 and Mdivi-1

Abstract

Colorectal cancer (CRC) is a leading cause of cancer-related mortality worldwide. Activating mutations in KRAS, present in approximately 30–40% of CRCs, drive tumour progression and therapy resistance through canonical MAPK signalling and extensive metabolic rewiring, including changes in mitochondrial function and dynamics. [1–3] Mitochondrial fusion and fission, orchestrated by MFN1/2, OPA1 and the fission GTPase DRP1, have emerged as context-dependent vulnerabilities in cancer. [4–7] Small-molecule modulators of these processes – such as the fusion-promoting compound M1 and the fission inhibitor Mdivi-1 – provide a way to test whether mitochondrial morphology and quality control can be therapeutically exploited in KRAS-mutant CRC. [8,9] In this project, we investigated how pharmacological promotion of mitochondrial fusion (M1) and inhibition of fission (Mdivi-1) reshape mitochondrial morphology, fusion– fission protein balance, proliferation and drug response in a small panel of colorectal cell lines: normal colon epithelium (CCD841CoN), KRAS-wild-type tumour cells (SW48), and KRAS-mutant CRC cells (HCT116 KRASG13D, SW620 KRASG12V). Mitochondrial network architecture was quantified using high-content confocal imaging and skeleton analysis to derive a morphology-based fusion index (total branch length per mitochondrial volume) and Branch Form Factor (BFF, a measure of per-mitochondrion network complexity). [15,16] In parallel, western blotting for DRP1, phospho-DRP1 (Ser616/Ser637), MFF, MFN1/2 and OPA1 was used to compute a composite protein-based Fusion–Fission Index (FFIprotein). Real-time proliferation assays (xCELLigence) and apoptosis/viability measurements were used to link mitochondrial dynamics to functional outcomes, including the e!ect of repeated M1 “priming” on sensitivity to an oxidative combination of arsenic trioxide (ATO) and D-vitamin C (DVC), a regimen known to generate synergistic ROS-dependent cytotoxicity in CRC cells. [2,17] Baseline analysis revealed marked heterogeneity in mitochondrial morphology and fusion–fission protein expression that did not segregate simply by KRAS status. CCD841CoN cells displayed extended networks and balanced fusion/fission proteins. SW48 and HCT116 occupied an intermediate “fragmented but not fully collapsed” state, whereas SW620 showed a more punctate, fragmented network with elevated DRP1 and pSer616, consistent with strong DRP1-dependent fission. In CCD841CoN, both M1 and Mdivi-1 behaved “by the book,” decreasing DRP1 and increasing MFN1/2 and OPA1, with a doubling of fusion-related morphology metrics and FFIprotein. In contrast, SW48 responded non-canonically: both drugs lowered DRP1 Ser616 phosphorylation, and selectively depleted OPA1 and MFN1, suggesting a stress-like, compensatory fragmentation response. KRAS-mutant HCT116 and SW620 bu!ered gross mitochondrial architecture against large-scale hyperfusion in response to M1 and Mdivi-1, remaining in a fragmented state. Nonetheless, western blots showed clear shifts in DRP1 phosphorylation, especially in SW620, where Mdivi-1 reduced Ser616 and increased Ser637 phosphorylation. Fusion fission protein indices changed only modestly, indicating that KRAS-mutant cells can decouple DRP1 phosphorylation from dramatic morphological fusion while preserving a metabolically flexible, fragmented network. Functionally, M1 tended to support proliferation of KRAS-mutant cells over 48–72 h, whereas Mdivi-1 exerted only a modest antiproliferative e!ect at the doses used. Strikingly, repeated M1 priming cycles (two to four 12 h pulses) sensitised KRAS-mutant SW620 and HCT116 cells to ATO–DVC. In SW620, four priming cycles approximately doubled the loss of viability induced by ATO5/DVC1 compared with non-primed controls. In HCT116, M1 priming enhanced the cytotoxicity of both ATO7.5/DVC1.5 (approximately 2-fold) and ATO5/DVC1 (approximately 3-fold). These data support a model in which enforced fusion and metabolic “tight-coupling” transiently benefit KRAS-mutant cell growth but leave them less able to bu!er ROS and mitochondrial damage when confronted with oxidative therapy. This work demonstrates that mitochondrial dynamics in CRC are highly cell-line- and oncogene-dependent. Normal colon cells respond to M1/Mdivi-1 with classical hyperfusion, KRAS-wild-type SW48 shows a non-canonical stress-fragmentation response, and KRAS- mutant cells are morphologically bu!ered yet remain dependent on DRP1 phosphorylation and mitochondrial plasticity. Importantly, pharmacologically promoting fusion can be used as a “trap”: M1-driven hyperfusion supports KRAS-mutant proliferation at baseline but sensitises the same cells to oxidative drug combinations such as ATO–DVC, revealing an exploitable conditional vulnerability in KRAS-driven CRC.