MFN2 loss is associated with EMT-like invasive phenotypes and Ca2+/Mitochondrial stress–PERK/eIF2α/ATF4 signaling in lung adenocarcinoma

Objective This study investigated whether mitofusin 2(MFN2) loss is associated with invasive phenotypes, Ca 2+ dysregulation, mitochondrial dysfunction, and PERK/eIF2α/ATF4-mediated stress signaling in lung adenocarcinoma. Methods Stable MFN2 knockdown, overexpression, and rescue models were established in A549 and H1975 cells. MFN2-silenced cells were treated with GSK2606414 to pharmacologically attenuate PERK-related signaling. Cell viability, EMT markers, migration, invasion, pathway activation, Ca 2+ signals, ROS, and mitochondrial membrane potential were assessed using molecular, imaging, and flow-cytometric assays. A tail-vein lung colonization model was used to evaluate tissue-level changes following MFN2 knockdown and GSK2606414 treatment; this design bypassed primary-tumor formation and the earlier steps of the metastatic cascade. Results MFN2 knockdown increased the CCK-8 viable-cell signal, migration, and invasion in both cell lines and shifted EMT markers toward reduced E-cadherin and increased N-cadherin, Vimentin, and Snail, whereas MFN2 overexpression produced the opposite effects. MFN2 loss also increased GRP78 and ATF4 expression and elevated PERK and eIF2α phosphorylation; genetic restoration of MFN2 attenuated these responses. In A549 cells, MFN2 silencing increased cytosolic Ca 2+ fluorescence and reduced mitochondria-associated Ca 2+ fluorescence in mitochondria-enriched fractions, while both cell lines showed increased ROS accumulation and mitochondrial depolarization. GSK2606414 attenuated the Ca 2+ -associated fluorescence abnormalities in A549 cells and reduced ROS accumulation and mitochondrial depolarization in both A549 and H1975 cells, together with attenuation of PERK/eIF2α/ATF4 signaling and migration/invasion-associated readouts. In mouse lungs, MFN2 knockdown was associated with greater histopathological alterations, including increased cellularity and tissue disruption, weaker MFN2 staining, and increased p-PERK/PERK, p-eIF2α/eIF2α, and Vimentin; these alterations were attenuated by GSK2606414. Conclusion MFN2 loss was associated with EMT-like molecular remodeling, enhanced migratory and invasive behavior, Ca 2+ redistribution, mitochondrial dysfunction, and PERK/eIF2α/ATF4 pathway activation in lung adenocarcinoma models. Pharmacological attenuation by GSK2606414 supports the involvement of PERK-related stress signaling but does not establish a fully linear or PERK-specific causal axis.