Abstract: Objective To investigate the alterations in lipid metabolism characteristics of temozolomide⁃resistant glioblastoma cells. Methods Human glioblastoma U87 cells were utilized to develop a temozolomide⁃resistant cell model (U87TR) through a concentration gradient induction method, with resistance confirmed by CCK⁃8 assay. The clonogenic, invasive, and migration abilities of the cells were assessed using the plate colony formation assay, Transwell invasion assay, and wound healing assay, respectively. Lipids were extracted from both U87 and U87TR cells employing a modified Bligh⁃Dyer method, and lipid metabolites were subsequently identified, quantified, clustered, and analyzed for differences using ultra high performance liquid chromatography⁃tandem mass spectrometry. Principal component analysis and orthogonal partial least squares⁃discriminant analysis (OPLS⁃DA) were utilized to identify lipid metabolites associated with temozolomide⁃acquired resistance. Results The temozolomide⁃acquired resistant U87TR model was successfully established. Compared to the sensitive parental cells, U87TR cells demonstrated significantly enhanced clonogenic formation, invasion, and migration capabilities (all P<0.05). Lipidomic profiling identified a total of 417 lipid metabolites in 32 categories (mainly including phosphatidylcholines, triglycerides, and sphingomyelin), among which 260 lipid metabolites showed statistically significant differences in content. The expression levels of PG34∶2(18∶1_16∶1), PG38∶5 (18∶1_20∶4), and BMP34∶3 (18∶2_16∶1) were significantly decreased in U87TR cells compared to U87 cells (all P<0.05). Conversely, the expression levels of PG34∶2 (18∶2_16∶0), PG36∶4 (18∶2_18∶2), PG36∶3 (18∶2_18∶1)， etc. were significantly increased (all P<0.05). OPLS⁃DA confirmed that lipid metabolites such as PI36∶3 (18∶2_18∶1), PA32∶0, and LacCer d18∶1/14∶0 effectively differentiate resistant cells from sensitive ones. Conclusions Temozolomide⁃resistant glioblastoma cells exhibit distinct lipid metabolite profiles compared to sensitive cells，which may be associated with the drug resistance observed in  glioblastoma.

Key words: Glioblastoma, Lipidomics, Temozolomide, Drug resistance； Ultra high performance liquid chromatography?tandem mass spectrometry