Blockage of EGFR/AKT and mevalonate pathways synergize the antitumor effect of temozolomide by reprogramming energy metabolism in glioblastoma

Abstract

Background: Metabolism reprogramming plays a vital role in glioblastoma (GBM) progression and recurrence by producing enough energy for highly proliferating tumor cells. In addition, metabolic reprogramming is crucial for tumor growth and immune-escape mechanisms. Epidermal growth factor receptor (EGFR) amplification and EGFR-vIII mutation are often detected in GBM cells, contributing to the malignant behavior. This study aimed to investigate the functional role of the EGFR pathway on fatty acid metabolism remodeling and energy generation.

Methods: Clinical GBM specimens were selected for single-cell RNA sequencing and untargeted metabolomics analysis. A metabolism-associated RTK-fatty acid-gene signature was constructed and verified. MK-2206 and MK-803 were utilized to block the RTK pathway and mevalonate pathway induced abnormal metabolism. Energy metabolism in GBM with activated EGFR pathway was monitored. The antitumor effect of Osimertinib and Atorvastatin assisted by temozolomide (TMZ) was analyzed by an intracranial tumor model in vivo.

Results: GBM with high EGFR expression had characteristics of lipid remodeling and maintaining high cholesterol levels, supported by the single-cell RNA sequencing and metabolomics of clinical GBM samples. Inhibition of the EGFR/AKT and mevalonate pathways could remodel energy metabolism by repressing the tricarboxylic acid cycle and modulating ATP production. Mechanistically, the EGFR/AKT pathway upregulated the expressions of acyl-CoA synthetase short-chain family member 3 (ACSS3), acyl-CoA synthetase long-chain family member 3 (ACSL3), and long-chain fatty acid elongation-related gene ELOVL fatty acid elongase 2 (ELOVL2) in an NF-κB-dependent manner. Moreover, inhibition of the mevalonate pathway reduced the EGFR level on the cell membranes, thereby affecting the signal transduction of the EGFR/AKT pathway. Therefore, targeting the EGFR/AKT and mevalonate pathways enhanced the antitumor effect of TMZ in GBM cells and animal models.

Conclusions: Our findings not only uncovered the mechanism of metabolic reprogramming in EGFR-activated GBM but also provided a combinatorial therapeutic strategy for clinical GBM management.

Keywords: EGFR; combinatorial therapeutic strategy; energy metabolism; glioblastoma.

FIGURE 1

GBM patients with high levels of EGFR accompanied by ACSS3, ACSL3, and ELOVL2 expressions show worse prognoses. (A) Cell types were annotated and visualized as a UMAP plot in GBM single‐cell RNA sequencing data. (B) Expressions of GFAP, CHI3L1, OLIG2, SOX6, EGFR, and PDGFA in subcellular populations of tumor cells from GBM single‐cell data were stained. (C) A total of 43,215 tumor cells were grouped into 4 cell populations according to the expression features of key genes. (D) Kaplan‐Meier curve was generated to evaluate the overall survival time of GBM patients in distinct groups from the CGGA cohort, based on the expressions of GFAP, CHI3L1, OLIG2, SOX6, EGFR, and PDGFA. G1 (red line) represented patients with GFAP^high, CHI3L1^high, OLIG2^low, SOX6^low, EGFR^high, and PDGFA^high; G2 (green line) represented patients with GFAP^high, CHI3L1^high, OLIG2^low, SOX6^low, EGFR^low, and PDGFA^low; and G3 (cyan line) represented patients with GFAP^low, CHI3L1^low, OLIG2^high, SOX6^high. ^∗∗∗ P < 0.001 (Log‐rank test). (E) Differentially expressed genes in CP1 were identified and visualized as a volcano plot. The dotted lines represent 0.25 and ‐0.25 of log2‐fold change. (F) The correlation analysis among the expressions of EGFR, PDGFA, ACSS3, ACSL3, and ELOVL2 in the CGGA cohort. ^∗∗∗ P < 0.001 (Pearson). (G) Kaplan‐Meier curve analysis showed that patients with high RFA scores in the CGGA cohort had a poor prognosis. ^∗∗∗ P < 0.0001 (Log‐rank test). Abbreviations: GBM, glioblastoma; UMAP, uniform manifold approximation and projection; GFAP, glial fibrillary acidic protein; CHI3L1, chitinase 3 like 1; OLIG2, oligodendrocyte transcription factor 2; SOX6, SRY‐box transcription factor 6; EGFR, epidermal growth factor receptor; PDGFA, platelet‐derived growth factor subunit A; ACSS3, acyl‐CoA synthetase short‐chain family member 3; ACSL3, acyl‐CoA synthetase long‐chain family member 3; ELOVL2, long‐chain fatty acid elongation‐related gene ELOVL fatty acid elongase 2; CGGA, Chinese glioma genome atlas.

FIGURE 2

EGFR/AKT pathway activation enhances the expression of ACSS3, ACSL3, and ELOVL2 via NF‐κB activation. (A) TBD0220 cells were treated with 0, 0.1, 0.5, 1, 5, and 10 μmol/L MK‐2206 for 24 h. U‐87 MG cells were stimulated with 0, 0.5, 5, and 50 ng/mL of EGF for 24 h, or overexpressed with EGFR‐vIII mutant. The expression levels of EGFR, p‐EGFR, AKT, p‐AKT, ACSS3, ACSL3, ELOVL2, and β‐actin were tested by western blotting. Protein was normalized to their respective β‐actin loading control and expressions were quantified by ImageJ software. (B) Western blotting analysis of EGFR, p‐EGFR, AKT, p‐AKT, NF‐κB, p‐NF‐κB, ACSS3, ACSL3, ELOVL2, and GAPDH expressions in TBD0220, U‐87 MG and U‐87 MG‐EGFR‐vIII cells treated with DMSO or 5 μmol/L of MK‐2206 for 24 h. (C) Western blotting to check the expression levels of ACSS3, ACSL3, ELOVL2, NF‐κB, p‐NF‐Κb, and β‐actin in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells treated with DMSO or 100 μmol/L of JSH‐23 treatment. Protein was normalized to their respective β‐actin loading control and expressions were quantified by ImageJ software. (D‐F) TBD0220, and U‐87 MG‐EGFR‐vIII cells were treated with DMSO or 5 μmol/L MK‐2206 for 24 h. ChIP analysis of the regulatory regions of p‐NF‐κB binding to the promoters of ACSS3 (D), ACSL3 (E), and ELOVL2 (F). ^∗ P < 0.05, ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (independent‐sample Student's t‐test for TBD0220; one‐way ANOVA for U‐87 MG). Abbreviations: EGFR, epidermal growth factor receptor; p‐EGFR, phosphorylation of epidermal growth factor receptor; EGF, epidermal growth factor; AKT, AKT serine/threonine kinase 1; GBM, glioblastoma; ANOVA, analysis of variance; ACSS3, acyl‐CoA synthetase short‐chain family member 3; ACSL3, acyl‐CoA synthetase long‐chain family member 3; ELOVL2, long‐chain fatty acid elongation‐related gene ELOVL fatty acid elongase 2.

FIGURE 3

EGFR/AKT pathway regulates mitochondrial respiration and proliferation in GBM cells. (A‐B) TBD0220 (A), and U‐87 MG‐EGFR‐vIII cells (B) were treated with DMSO or 5 μmol/L MK‐2206 for 24 h. The mitochondrial functions were monitored by Seahorse XF Cell Mito Stress test. The OCR, basal respiration, proton leak, and ATP production rates were measured as illustrated. ^∗ P < 0.05, ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (independent‐sample Student's t‐test for TBD0220; one‐way ANOVA for U‐87 MG). (C) ATP levels in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells were analyzed after 24 h of treatments with DMSO or 5 μmol/L of MK‐2206. ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (independent‐sample Student's t‐test for TBD0220; one‐way ANOVA for U‐87 MG). (D) The cell growth assay for TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII lines treated with DMSO, 5 μmol/L MK‐2206, or 5 μmol/L MK‐2206 plus 50 μmol/L ATP were performed. ^∗∗∗ P < 0.001 (two‐way ANOVA). (E) The colony formation assay of TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII lines treated with DMSO or 1 μmol/L MK‐2206. (F) Cell cycle distributions were analyzed by flow cytometry in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells treated with DMSO or 5 μmol/L MK‐2206. (G) Western blotting to show changes in expressions of CDK2, CDK4, CDK6, Cyclin D, RB, p‐RB, and GAPDH in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells treated with DMSO or MK‐2206. Abbreviations: EGFR, epidermal growth factor receptor; AKT, AKT serine/threonine kinase 1; GBM, glioblastoma; ANOVA, analysis of variance; DMSO, dimethyl sulfoxide; CDK2, cyclin‐dependent kinase 2; CDK4, cyclin‐dependent kinase 4; CDK6, cyclin‐dependent kinase 6; OCR, oxygen consumption rate.

FIGURE 4

The hyperactivated EGFR/AKT pathway correlates with phospholipid metabolism and accelerated cholesterol biosynthesis in GBM cells. (A) A total of 66 GBM samples were analyzed using RNA sequencing and untargeted metabolomics. The expressions of EGFR, ACSS3, ACSL3, ELOVL2 and corresponding metabolites such as LysoPC, PC, LysoPE, PE, PG, PI, and PS were analyzed. (B) Untargeted metabolomic analysis of PC and LysoPC correlated with the expressions of EGFR, ACSS3, ACSL3, and ELOVL2 in GBM samples. (C) The long‐chain fatty acids and cholesterol levels were positively correlated with EGFR, ACSS3, ACSL3, and ELOVL2 expressions in GBM samples. (D) The schematic diagram of fatty acid beta‐oxidation and TCA cycle. (E‐J) The levels of crucial intermediates in the TCA cycle, such as citrate (E), cis‐aconitate (F), isocitrate (G), α‐ketoglutarate (H), ATP (I), and VLCFA behenic acid (J) were downregulated in TBD0220 cells by 24 h of treatment with DMSO, 5 μmol/L MK‐2206 or 5 μmol/L MK‐803. ^∗ P < 0.05, ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (One‐way ANOVA). Abbreviations: EGFR, epidermal growth factor receptor; AKT, AKT serine/threonine kinase 1; GBM, glioblastoma; ANOVA, analysis of variance; ACSS3, acyl‐CoA synthetase short‐chain family member 3; ACSL3, acyl‐CoA synthetase long‐chain family member 3; ELOVL2, long‐chain fatty acid elongation‐related gene ELOVL fatty acid elongase 2; TCA, citric acid cycle; VLCFA, very long‐chain fatty acid; LysoPC, lysophosphatidylcholine; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PG, phosphatidyl glycerol; PI, phosphatidylinositol; PS, phosphatidylserine.

FIGURE 5

MK‐803 inhibits EGFR/AKT pathway transduction via decreasing EGFR levels on the cell membrane. (A) The components of cytosol and membrane in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells treated with DMSO, MK‐803, or MK‐803 plus 50 μmol/L water‐soluble cholesterol were fractionated. The distribution levels of EGFR, Na‐K‐ATPase, and β‐Tubulin were measured by western blotting. (B) TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells were stimulated by 50 ng/mL EGF at different time points after 24 h pre‐treatment with DMSO or MK‐803. Western blotting analysis was performed to detect the expression and phosphorylation levels of EGFR, and AKT. GAPDH was used as the loading control. (C) Western blotting analysis of ACSS3, ACSL3, ELOVL2, and β‐Tubulin expressions in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells treated with DMSO or 5 μmol/L MK‐803 for 24h. Proteins were normalized to their respective β‐Tubulin loading control and expressions were quantified by ImageJ software. (D‐E) TBD0220 (D), and U‐87 MG‐EGFR‐vIII cells (E) were treated with 5 μmol/L of MK‐803 for 24 h. The mitochondrial functions in these cells were monitored by Seahorse XF Cell Mito Stress assay. The OCR, basal respiration, proton leak, and ATP production rates were measured as illustrated. ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (independent‐sample Student's t‐test for TBD0220; one‐way ANOVA for U‐87 MG). (F) ATP levels in TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells were analyzed after 24 h of treatment with DMSO or 5 μmol/L MK‐803. ^∗∗∗ P < 0.001 (independent‐sample Student's t‐test for TBD0220; one‐way ANOVA for U‐87 MG). (G) The proliferation rates of TBD0220, U‐87 MG, and U‐87 MG‐EGFR‐vIII cells were determined after treatment with DMSO or 5 μmol/L MK‐803. ^∗∗∗ P < 0.001 (Two‐way ANOVA). Abbreviations: EGFR, epidermal growth factor receptor; AKT, AKT serine/threonine kinase 1; ANOVA, analysis of variance; ACSS3, acyl‐CoA synthetase short‐chain family member 3; ACSL3, acyl‐CoA synthetase long‐chain family member 3; ELOVL2, long‐chain fatty acid elongation‐related gene ELOVL fatty acid elongase 2; DMSO, dimethyl sulfoxide; OCR, oxygen consumption rate.

FIGURE 6

Targeting EGFR/AKT and mevalonate pathways suppresses GBM proliferation and prolongs the survival time of tumor‐bearing mice. (A) The flow diagram of nude mice xenograft model. (B) Representative brain bioluminescence images of nude mice on Day 7, Day 14, and Day 21 after implantation. (C) Tumor growth curves were quantitated and illustrated. ^∗ P < 0.05, ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (two‐way ANOVA). (D) Kaplan‐Meier curves for different experimental and control groups. ^∗ P < 0.05, ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (log‐rank test). (E) Representative images of FFPE brain tissues for H&E and IHC staining of Ki67, p‐AKT, ACSS3, ACSL3, and ELOVL2. Scale bar = 1 mm for H&E, 50 μm for IHC. Abbreviations: EGFR, epidermal growth factor receptor; AKT, AKT serine/threonine kinase 1; GBM, glioblastoma; ANOVA, analysis of variance; FFPE, formalin‐fixed and paraffin‐embedded; ACSS3, acyl‐CoA synthetase short‐chain family member 3; ACSL3, acyl‐CoA synthetase long‐chain family member 3; ELOVL2, long‐chain fatty acid elongation‐related gene ELOVL fatty acid elongase 2.

FIGURE 7

RFA score positively correlates with the immunosuppressive GBM microenvironment. (A) Expressions of immunosuppressors were positively correlated with respective RFA scores in glioma samples of the CGGA cohort. The clinical information of gender, tumor grade, IDH mutation status, 1p/19q co‐deletion status, and MGMT promoter methylation status in GBM samples are displayed. (B) Correlations between RFA scores and the expressions of CASP4, ELF4, LAIR1, LYN, MR1, MSR1, NFKB1, RAB27A, SWAP70, and TGFB1 were analyzed in the CGGA cohort. (C) The ssGSEA for correlation of RFA scores with corresponding immune cell lineages in the CGGA cohort. (D) Tumor purity was negatively correlated with RFA scores in GBM samples of the CGGA cohort. ^∗∗∗ P < 0.001 (Pearson) (E) CIBERSORT analysis of immune cell compositions in GBM samples of the CGGA cohort. The M2 macrophage levels were directly correlated with RFA scores. ^∗∗∗ P < 0.001 (independent‐sample Student's t‐test). Abbreviations: GBM, glioblastoma; RFA, RTK‐fatty acid‐gene signature; CGGA, Chinese glioma genome atlas; MGMT, O6‐methylguanine DNA methyltransferase; CASP4, caspase 4; ELF4, E74‐like factor 4, LAIR1, leukocyte‐associated Ig‐like receptor 1; LYN, LYN Proto‐Oncogene, Src family tyrosine kinase; MR1, major histocompatibility complex, class I‐related; MSR1, macrophage scavenger receptor 1; NFKB1, nuclear factor kappa B subunit 1; RAB27A, RAB27A, member RAS oncogene family; SWAP70, switching B cell complex subunit; TGFB1, transforming growth factor beta 1; ssGSEA, single‐sample gene set enrichment analysis.

FIGURE 8

The combination therapeutic strategy of TMZ + OSI + ATO reduces tumor proliferation and improves GBM microenvironment. (A) The flow diagram of the C57BL/6J mice xenograft model is shown. (B) Representative brain bioluminescence images of C57BL/6J mice on Day 3, Day 7, and Day 14 post‐implantation. (C) Quantitation of tumor growth curves. ^#not significant, ^∗ P < 0.05, ^∗∗ P < 0.01, ^∗∗∗ P < 0.001 (two‐way ANOVA). (D) IHC staining of MHC‐II, CD206, and CD8a in FFPE tumor tissue sections. Scale bar = 100 μm. (E) Flow cytometry analysis to evaluate the infiltration rate of cytotoxic T‐lymphocytes under different combination treatments of TMZ, OSI, and ATO. ^∗ P < 0.05, ^∗∗ P < 0.01 (one‐way ANOVA). (F) Flow cytometry analysis to evaluate M1/M2 ratios of macrophages under different combination treatments of TMZ, OSI, and ATO. ^∗ P < 0.05 (one‐way ANOVA). Abbreviations: GBM, glioblastoma; TMZ, temozolomide; OSI, Osimertinib; ATO, Atorvastatin; MHC‐II, major histocompatibility complex, class II; FFPE, formalin‐fixed paraffin‐embedded; ANOVA, analysis of variance.

FIGURE 9

TMZ synergizes ATO's antitumor potency in clinical GBM treatment. (A‐B) Kaplan‐Meier curve of OS (A) and PFS (B) in GBM patients with statins (n = 9) or without statins (n = 51). ^∗ P < 0.05, ^∗∗ P < 0.01 (log‐rank test). (C) Intracranial images of the typical GBM case from diagnosis to postoperative follow‐ups by MRI and clinical examinations. (D) Representative images of FFPE sections of primary tumor tissues from patient #1 for H&E and IHC staining of Ki67, EGFR, p‐EGFR, p‐AKT, CD34, ACSS3, ACSL3, ELOVL2, CD8, MHC‐II, and CD163. Scale bar = 20 μm. Abbreviations: TMZ, temozolomide; ATO, Atorvastatin; GBM, glioblastoma; MRI, magnetic resonance imaging; FFPE, formalin‐fixed paraffin‐embedded; OS, overall survival; PFS, progression‐free survival; EGFR, epidermal growth factor receptor; p‐EGFR, phosphorylation of epidermal growth factor receptor; p‐AKT, phosphorylation of AKT serine/threonine kinase 1; ACSS3, acyl‐CoA synthetase short‐chain family member 3; ACSL3, acyl‐CoA synthetase long‐chain family member 3; ELOVL2, long‐chain fatty acid elongation‐related gene ELOVL fatty acid elongase 2; MHC‐II, major histocompatibility complex, class II.

FIGURE 10

Schematic illustration depicting the mechanism of hyperactivation of EGFR/AKT and mevalonate pathway promoting energy metabolism, leading to malignant progression of GBM. Combinational therapeutic strategy of TMZ, TKI, and statins benefited GBM patients. Abbreviations: EGFR, epidermal growth factor receptor; AKT, AKT serine/threonine kinase 1; GBM, glioblastoma; TMZ, temozolomide; TKI, tyrosine kinase inhibitor.