Clinical, molecular, metabolic, and immune features associated with oxidative phosphorylation in melanoma brain metastases

Abstract

Background: Recently, we showed that melanoma brain metastases (MBMs) are characterized by increased utilization of the oxidative phosphorylation (OXPHOS) metabolic pathway compared to melanoma extracranial metastases (ECMs). MBM growth was inhibited by a potent direct OXPHOS inhibitor, but observed toxicities support the need to identify alternative therapeutic strategies. Thus, we explored the features associated with OXPHOS to improve our understanding of the pathogenesis and potential therapeutic vulnerabilities of MBMs.

Methods: We applied an OXPHOS gene signature to our cohort of surgically resected MBMs that had undergone RNA-sequencing (RNA-seq) (n = 88). Clustering by curated gene sets identified MBMs with significant enrichment (High-OXPHOS; n = 21) and depletion (Low-OXPHOS; n = 25) of OXPHOS genes. Clinical data, RNA-seq analysis, and immunohistochemistry were utilized to identify significant clinical, molecular, metabolic, and immune associations with OXPHOS in MBMs. Preclinical models were used to further compare melanomas with High- and Low-OXPHOS and for functional validation.

Results: High-OXPHOS MBMs were associated with shorter survival from craniotomy compared to Low-OXPHOS MBMs. High-OXPHOS MBMs exhibited an increase in glutamine metabolism, and treatment with the glutaminase inhibitor CB839 improved survival in mice with MAPKi-resistant, High-OXPHOS intracranial xenografts. High-OXPHOS MBMs also exhibited a transcriptional signature of deficient immune activation, which was reversed in B16-F10 intracranial tumors with metformin treatment, an OXPHOS inhibitor.

Conclusions: OXPHOS is associated with distinct clinical, molecular, metabolic, and immune phenotypes in MBMs. These associations suggest rational therapeutic strategies for further testing to improve outcomes in MBM patients.

Keywords: brain metastases; immune therapy; melanoma; oxidative phosphorylation; targeted therapy.

Figure 1.

Identification of High-OXPHOS and Low-OXPHOS melanoma brain metastases. (A) Hierarchical clustering of the OXPHOS-Indices from 88 melanoma brain metastases (MBMs) with available RNA-seq data resulted in the formation of 3 clusters: MBMs with significant enrichment (High-OXPHOS; n = 21); depletion (Low-OXPHOS; n = 25); or intermediate (Intermediate-OXPHOS; n = 42) OXPHOS gene set enrichment. Results are represented as a heatmap of median-centered values from each of the 8 components of the OXPHOS-Index. (B) GSEA-P confirming enrichment of OXPHOS in MBMs identified as High-OXPHOS versus Low-OXPHOS. (C) Kaplan–Meier analysis of overall survival from craniotomy for patients with High-OXPHOS versus patients with Low-OXPHOS MBMs. Hazard ratio determined via Mantel-Haenszel test and significance by log-rank test. (D–G) Comparison of clinical variables between patients with High- and Low-OXPHOS MBMs, including mean age, gender, mean body mass index (BMI), and frequency of previous radiation XRT). Significance was determined for age and BMI via 2-sided Student’s t-test; gender and frequency of prior XRT were compared via 2-sided Fisher’s exact test.

Figure 2.

Molecular associations of OXPHOS in melanoma brain metastases. (A) Gene expression analysis of PGC1α and MITF (by RNAseq) in High-(Red) versus Low-OXPHOS (Blue) melanoma brain metastases (MBMs). Each plot is a simple box and whisker plot. Median values (lines) and interquartile range (whiskers) are indicated. Adjusted P values calculated via generalized linear model analysis are listed. (B) Cumulative GSEA-P enrichment plot demonstrating significant enrichment or depletion (false discovery rate [FDR] q-val < 0.0001) of MSigDB Hallmarks gene sets in High-OXPHOS versus Low-OXPHOS MBMS identified via clustering methods. The 10 most up-regulated gene sets are shown in red. All significantly down-regulated gene sets are shown in blue. The normalized enrichment score forms the x-axis. (C) Comparison of the prevalence of complete PTEN loss by IHC for High-OXPHOS and Low-OXPHOS MBMs. Y-axis represents the frequency (%) of MBMs with complete absence of PTEN expression. Significance determined via 2-sided Fisher’s exact test. (D) Comparison of P-PRAS40 IHC expression by H-scores for High-OXPHOS versus Low-OXPHOS MBMs. Lines represent mean ± SD; each dot represents a single tumor. Significance determined via 2-sided Student’s t-test. (E) Comparison of P-S6 IHC expression by H-scores for High-OXPHOS versus Low-OXPHOS MBMs. (F) Representative P-S6 staining in High- and Low-OXPHOS MBMs. Samples selected reflect the median H-scores in the High- (median = 35) and Low- (median = 10) OXPHOS MBMs. Tumor cells are present throughout the entirety of both samples.

Figure 3.

Metabolic profiling of High-OXPHOS melanoma brain metastases (MBMs) identifies glutamine metabolism as a therapeutic target. (A) Cumulative GSEA-P enrichment plot demonstrating significant enrichment or depletion (false discovery rate [FDR] q-val < 0.0001) of KEGG metabolism gene sets in High-OXPHOS versus Low-OXPHOS MBMS. Enriched gene sets are shown in red. No depleted gene sets met the criteria for significance. X-axis shows the normalized enrichment score for each pathway. (B) Differentially expressed metabolites (FDR q-val < 0.25) between A375 (Low-OXPHOS; n = 5) and A375-R1 (High-OXPHOS; n = 4) human melanoma intracranial xenografts, as determined by liquid chromatography mass spectrometry (LC-MS). Heatmap shows median-centered log2-tranformed concentrations of these metabolites. (C) Pathway analysis of metabolites significantly upregulated (log2FC>0 and FDR q-val < 0.25) in A375-R1 versus A375 intracranial xenografts. All pathways listed are significantly enriched in A375-R1 versus A375 (FDR q-val < 0.05). (D,E) Seahorse mitochondrial stress test results for A375 and A375-R1 cells treated for 12 h with vehicle or 100 nM of CB839 in vitro. The figures show basal, oligomycin-inhibited (“O”), FCCP-activated (“F”), and Antimycin/Rotenone-inhibited (“A&R”) oxygen consumption rate levels. Data are representative of quadruplicates and SD. (F) Cell proliferation inhibition of A375 and A375-R1 cell lines treated with CB839 for 72 h in vitro. Data are representative of triplicates and SD. (G) Kaplan–Meier analysis of overall survival (OS) for mice bearing intracranial A375-R1 xenografts treated with vehicle or CB839 (200 mg/kg p.o. twice daily). Significance was determined by log-rank testing. (H) Kaplan–Meier analysis of OS for mice bearing intracranial MEL624 xenografts treated with vehicle or CB839 (200 mg/kg p.o. twice daily). Significance determined by log-rank testing.

Figure 4.

Oxidative phosphorylation associates with immunosuppression in melanoma brain metastases. (A) CIBERSORT analysis of High-OXPHOS (n = 21) and Low-OXPHOS (n = 25) MBMs. Data are presented as a heatmap of median-centered log2-tranformed estimates of the 22 immune cell populations listed on the right side of the heatmap. FDR q-values are listed to the left of the graph along with determination of significance (FDR q-val < 0.05). No immune cell population significantly differed between the groups. (B–D) IHC analysis for CD3-, CD8-, and PAX5-positive cells in High- and Low-OXPHOS MBMs. H-scores shown, as described in Figure 1. (E,F) Comparison of a 6-gene IFN-γ mRNA signature and 18-gene T-cell inflamed gene expression profile (GEP) for High- versus Low-OXPHOS MBMs. (B–F) Lines represent mean ± SD, and each dot represents a single tumor. Significance determined by 2-sided Student’s t-test.

Figure 5.

Metformin improves the response of High-OXPHOS, syngeneic intracranial melanoma xenografts to anti-PD1 immunotherapy. B16-F10 cells were implanted intracranially in C57BL/6 mice and treated for 96 h with low-dose metformin (50 mg/kg i.p. every other day; n = 3) or PBS (n = 4). qRT-PCR analysis was used to assess mRNA levels of (A) panel of genes previously shown to correlate with the presence of activated CD8⁺ T cells in melanomas, and (B) panel of IFNγ-related genes previously shown to predict response to anti-PD1 in metastatic melanoma patients. Values represent mean ± SD. ****P < .0001; ***P < .001; **P < .01, *P < .05 by 2-sided Student’s t-test. (C) Kaplan–Meier analysis of overall survival for C57BL/6 mice bearing intracranial B16-F10 xenografts treated with isotype antibody control (200 µg i.p. 3×/week) + PBS (10 µL/g body weight i.p. every other day) (Group 1; n = 5); anti-PD1 (200 µg i.p. 3×/week) + PBS (10 µL/g body weight i.p. every other day) (Group 2; n = 7); isotype control (200 µg i.p. 3×/week) + metformin (50 mg/kg i.p. every other day) (Group 3; n = 6); and anti-PD1 (200 µg i.p. 3×/week) + metformin (50 mg/kg i.p. every other day) (Group 4; n = 7). Significance of differences determined by log-rank testing.