Selenoprotein O Promotes Melanoma Metastasis and Regulates Mitochondrial Complex II Activity

Abstract

Evolutionarily conserved selenoprotein O (SELENOO) catalyzes a posttranslational protein modification known as AMPylation that is essential for the oxidative stress response in bacteria and yeast. Given that oxidative stress experienced in the blood limits survival of metastasizing melanoma cells, SELENOO might be able to affect metastatic potential. However, further work is needed to elucidate the substrates and functional relevance of the mammalian homolog of SELENOO. In this study, we revealed that SELENOO promotes cancer metastasis and identified substrates of SELENOO in mammalian mitochondria. In patients with melanoma, high SELENOO expression was correlated with metastasis and poor overall survival. In a murine model of spontaneous melanoma metastasis, SELENOO deficiency significantly reduced metastasis to distant visceral organs, which could be rescued by treatment with the antioxidant N-acetylcysteine. Mechanistically, SELENOO AMPylated multiple mitochondrial substrates, including succinate dehydrogenase subunit A, one of the four key subunits of mitochondrial complex II. Consistently, SELENOO-deficient cells featured increased mitochondrial complex II activity. Together, these findings demonstrate that SELENOO deficiency limits melanoma metastasis by modulating mitochondrial function and oxidative stress. Significance: SELENOO alters mitochondrial function and supports metastasis in melanoma, highlighting the impact of SELENOO-mediated posttranslational modification of mitochondrial substrates and selenoproteins in cancer progression.

Graphical abstract

Figure 1.

NAC treatment modulates melanoma metastasis. A–D, Growth of subcutaneous tumors and endpoint tumor diameter in mice transplanted with YUMM3.3 or YUMM5.2 cells and treated with vehicle control or NAC. E–H, Frequency of circulating melanoma cells in the blood as determined by flow cytometry (E and F) and metastatic disease burden based on ex vivo bioluminescence imaging (G and H). Results are shown as mean $\mathit{\pm }$ SD obtained from three independent experiments. *, P < 0.05; **, P < 0.01; unpaired two-tailed Student t test. Only significant results are shown.

Figure 2.

Role of SELENOO in ROS and stress responses. A, Proliferation assay. YUMM3.3 and YUMM5.2 cells were quantified by absolute cell counts after 1 week in culture. B, Migration assay. Cells were monitored for 100 hours using the xCELLigence Real-Time Cell Analyzer instrument. Slope was calculated from 4 to 100 hours. C, Invasion assay. Cells were monitored for 139 hours using the xCELLigence Real-Time Cell Analyzer instrument. Slope was calculated from 4 to 139 hours. Slopes (1/hour) are presented as mean $\mathit{\pm }$ SD of biological triplicates. D and E, Absolute mean fluorescence intensity (MFI) for CellROX Green (D) or CellROX Red (E) of the cell lines YUMM3.3 WT, YUMM3.3 SELENOO-KO clone 1 (SELO^−/−#1), YUMM3.3 SELENOO-KO clone 2 (SELO^−/−#2), YUMM3.3 SELO^−/−#2 re-expressing full-length SELENOO (SELO^−/−#2_FL), YUMM3.3 SELO^−/−#2 with the SELENOO inactive catalytic mutation D338A (SELO^−/−#2_D338A), YUMM5.2 WT, YUMM5.2 SELENOO-KO clone 1 (SELO^−/−#1), and YUMM5.2 SELENOO-KO clone 2 (SELO^−/−#2). Results are shown as mean $\mathit{\pm }$ SD obtained from three independent experiments. *, P < 0.05; unpaired two-tailed Student t test; NS, not significant. Only significant results are shown.

Figure 3.

SELENOO-deficient cells increase GSH and NADPH levels. A–F, Quantitative analysis of NADPH, NADP⁺, and relative NADP⁺/NADPH ratios in the cell lines YUMM3.3 WT, YUMM3.3 SELENOO-KO clone 1 (SELO^−/−#1), YUMM3.3 SELENOO-KO clone 2 (SELO^−/−#2), YUMM3.3 SELO^−/−#2 re-expressing full-length SELENOO (SELO^−/−#2_FL), YUMM3.3 SELO^−/−#2 with the SELENOO inactive mutant D338A (SELO^−/−#2_D338A), YUMM5.2 WT, YUMM5.2 SELENOO-KO clone 1 (SELO^−/−#1), and YUMM5.2 SELENOO-KO clone 2 (SELO^−/−#2); YUMM3.3 and YUMM5.2 WT and shRNA SELENOO knockdown cell lines (shRNA #1 and shRNA #2). G and H, Relative GSH/GSSG ratios of YUMM3.3 and YUMM5.2 cells as described previously. I, Diagram of the Seahorse XF Analyzer displaying the OCR measurements across the same cell lines. Results are shown as mean $\mathit{\pm }$ SD obtained from three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; NS, not significant; unpaired two-tailed Student t test (A–H); only significant results are shown.

Figure 4.

SELENOO catalyzes AMPylation of mitochondrial protein substrates and reduces mitochondrial complex II activity. A, Representative immunoblots of SELENOO cell lysates from YUMM3.3 SELENOO-deficient cells (SELO^−/−#2) and YUMM3.3 SELENOO-deficient cells (SELO^−/−#2) expressing SELENOO (SELO^−/−#2_FL) or SELENOO inactive mutant D338A (SELO^−/−#2-D338A). Cell extracts were analyzed for AMPylation, SELENOO, and GAPDH. Bottom, total protein levels in the Ponceau-stained membrane. B, Representative blots using avidin-HRP to detect biotinylated proteins following incubation of enriched mitochondria isolated from subcutaneous tumors with bio-17-ATP and SELENOO or SELENOO inactive mutant D348A. Bottom, Ponceau-stained membrane. C, Smear plot depicting sum PEP score (protein ion score) vs. log₂ (abundance ratio (SELO/SELO D348A) of all mitochondrion-associated proteins identified by LC/MS-MS of in vitro AMPylation assay depicted in B using recombinant SELENOO or SELENOO D348A with enriched mitochondrial extracts from subcutaneous tumors. Proteins displayed in red are enriched in SELENOO relative to SELENOO D348A (fold change >1.5). D, GO biological process terms for mitochondrial proteins enriched in SELENOO (SELO) vs. SELENOO D348A (SELO D348A; enriched protein set from C). E, Representative immunoblots of flag immunoprecipitates from YUMM3.3 SELENOO-deficient cells expressing SdhA-flag or Aco2-flag and SELENOO or SELENOO inactive mutant D338A. Cell extracts were analyzed for AMPylation, Flag, SELENOO, and GAPDH. F, Ratios of mitochondrial complex II activity in the cell lines YUMM3.3 WT, YUMM3.3 SELENOO-KO clone 1 (SELO^−/−#1), YUMM3.3 SELENOO-KO clone 2 (SELO^−/−#2), YUMM3.3 SELO^−/−#2 re-expressing full-length SELENOO (SELO^−/−#2_FL), YUMM3.3 SELO^−/−#2 with the SELENOO inactive mutant D338A (SELO^−/−#2_D338A), YUMM5.2 WT, YUMM5.2 SELENOO-KO clone 1 (SELO^−/−#1), and YUMM5.2 SELENOO-KO clone 2 (SELO^−/−#2). G, YUMM3.3 and YUMM5.2 WT and shRNA SELENOO-KO cell lines (shRNA #1 and shRNA #2). Results are shown as mean $\mathit{\pm }$ SD obtained from three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001; NS, not significant; unpaired two-tailed Student t test (F and G).

Figure 5.

SELENOO deficiency and catalytic activity modulate metastasis of melanoma cells, and antioxidant treatment rescues SELENOO deficiency in the YUMM3.3 cell line. A and B, Subcutaneous tumor growth in mice transplanted with WT or SELENOO-deficient YUMM3.3 and YUMM5.2 cells and incidence of subcutaneous tumor formation (percentages). C and D, Metastatic burden based on ex vivo bioluminescence imaging at the experimental endpoint. E and F, Frequency of circulating melanoma cells in the blood at the experimental endpoint. G–J, CellROX Red (G and I) or CellROX Green (H and J) staining of primary tumors at the experimental endpoint. K, Representative images of lung sections stained with hematoxylin and eosin (H&E) from mice transplanted intravenously with YUMM3.3 WT, YUMM3.3 SELENOO^−/−#2 (SELO^−/−#2), YUMM3.3 SELO^−/−#2 re-expressing full-length SELENOO (SELO^−/−#2_FL), and YUMM3.3 SELO^−/−#2 with the SELENOO inactive mutant D338A (SELO^−/−#2_D338A). Metastatic areas are highlighted with a black line. L, Higher magnification images of metastatic areas from the same groups. Metastases were stained with antiluciferase antibody (red) and cell nuclei with DAPI (blue). Scale bars, 50 μm. M and N, Quantification of the percentage of the tumor area covering the lung in hematoxylin and eosin sections obtained from the lungs of mice from the same groups mentioned above. O and P, Metastatic burden quantification (O) based on ex vivo bioluminescence imaging (P) in C57BL/6 mice intravenously injected with YUMM3.3 WT or SELENOO-deficient cells untreated (control) or pretreated with NAC. Results are shown as mean $\mathit{\pm }$ SD obtained from three independent experiments. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; NS, not significant; unpaired two-tailed Student t test. LN, lymph node; MFI, mean fluorescence intensity.