DDRGK1 Enhances Osteosarcoma Chemoresistance via Inhibiting KEAP1-Mediated NRF2 Ubiquitination

Abstract

Chemoresistance is the main obstacle in osteosarcoma (OS) treatment; however, the underlying mechanism remains unclear. In this study, it is discovered that DDRGK domain-containing protein 1 (DDRGK1) plays a fundamental role in chemoresistance induced in OS. Bioinformatic and tissue analyses indicate that higher expression of DDRGK1 correlates with advanced tumor stage and poor clinical prognosis of OS. Quantitative proteomic analyses suggest that DDRGK1 plays a critical role in mitochondrial oxidative phosphorylation. DDRGK1 knockout trigger the accumulation of reactive oxygen species (ROS) and attenuate the stability of nuclear factor erythroid-2-related factor 2 (NRF2), a major antioxidant response element. Furthermore, DDRGK1 inhibits ubiquitin-proteasome-mediated degradation of NRF2 via competitive binding to the Kelch-like ECH-associated protein 1 (KEAP1) protein, which recruits NRF2 to CULLIN(CUL3). DDRGK1 knockout attenuates NRF2 stability, contributing to ROS accumulation, which promotes apoptosis and enhanced chemosensitivity to doxorubicin (DOX) and etoposide in cancer cells. Indeed, DDRGK1 knockout significantly enhances osteosarcoma chemosensitivity to DOX in vivo. The combination of DDRGK1 knockdown and DOX treatment provides a promising new avenue for the effective treatment of OS.

Keywords: DDRGK domain-containing protein 1; chemoresistance; doxorubicin; osteosarcoma; redox homeostasis.

Figure 1

DDRGK1 acts as a cancer‐promoting gene. A) Pan‐cancer analysis of expression of DDRGK1 in different tumors compared with normal tissues using TCGA database. B) Gene dependency analysis of DDRGK1 using Depmap database. Negative values represent gene knockdown would inhibit cell growth. C) mRNA levels of DDRGK1 in muscle, mesenchymal stem cells, and osteosarcoma tissues, data were extracted from GSE18043, GSE38718, and GSE14827 respectively. D) Levels of DDRGK1 in bone and osteosarcoma with different clinical stages by immunohistochemistry staining. The OS tissue array was purchased from Biomax (CAT:OS804d) including 40 cases. E) Kaplan–Meier survival analysis of DDRGK1 for osteosarcoma according to the GSE21257 dataset. (***p < 0.001, ****p < 0.0001).

Figure 2

DDRGK1 promotes cell proliferation and inhibits apoptosis. A) Expression of DDRGK1 in various cell lines including osteosarcoma cell lines 143B, Saos‐2, U2OS and MG63, osteoblast cell line hFOB1.19, and muscle cell line L6 were detected by western blot. B) Effects of DDRGK1 on cell proliferation in osteosarcoma cells. Wild‐type 143B and DDRGK1‐knockout 143B cells were cultured at different times and cell viability was detected by CCK‐8 assay. C) Effect of DDRGK1 on cell clone formation. Wild‐type 143B and DDRGK1‐knockout 143B cells with 800 cells per well were seeded in a plate and cultured for 7 days, followed by crystal violet staining. D) Effects of DDRGK1 on cell cycle. Wild‐type 143B and DDRGK1‐knockout 143B cells were seeded in plate and hungered without serum for 24 h, then cultured for another 24 h with serum and collected for Flow cemetery analysis. E) Statistical analysis of cell cycle. F) Cell toxicity induced by thapsigargin (Tg) with or without DDRGK1. Wild‐type 143B and DDRGK1‐knockout 143B cells were subjected to Tg (10 nM) for the indicated time, and cell viability was detected by CCK‐8 assay. G) Influence of apoptosis after DDRGK1 knockout. Cleaved caspase 3 represents cell apoptosis was detected by western blot in cell lysates from Wild‐type 143B cells and DDRGK1‐knockout 143B with or without subjected to Tg (10 nM, 24 h). H) Flow cemetery to detect apoptosis rate by Annexin V/DAPI staining after being treated with Tg (10 nM) for 24 h. I) Statistical analysis of cell apoptosis. (three samples for each statistical analysis, *p < 0.05; **p < 0.01; ***p < 0.001).

Figure 3

Quantitative Proteomics analyses reveal the function of DDRGK1. A) Volcano plot for differentially expressed genes. Whole‐cell protein extracts of DDRGK1 knockout 143B cells and control cells were harvested, labeled with TMT then followed by mass spectrometry. A total of 2154 differentially expressed proteins were identified with 1053 proteins upregulated and 1101 downregulated. B) Gene Ontology enrichment analysis for differentially expressed genes. C) KEGG analysis for differentially expressed genes. D) Heatmap for expression of oxidative phosphorylation‐related proteins in DDRGK1 knockout cells and the control cells. Expression values were normalized by the “pheatmap” R package. E) Gene Set Enrichment Analysis (GSEA) in DDRGK1 knockout cells and the control cells. F) Mitochondrial metabolism analysis. Cells were seeded in an XFmicroplate at an optimal density and cultured for 24 h. The cellular oxygen consumption rate (OCR) was measured using a Seahorse XF Analyzer. G) Mitochondrial respiratory capacity analysis according to the OCR. (three samples for each statistical analysis, *p < 0.05; ****p < 0.0001).

Figure 4

DDRGK1 regulates ROS production by regulating NRF2 stability. A) Sensitivity of DDRGK1‐knockout cells to H₂O₂. The controlled and DDRGK1‐knockout cells were subjected to H₂O₂ for 20 min, then loaded with a DHE probe for 30 min and detected by flow cemetery. B) Regulation of NRF2 by DDRGK1. The protein levels of NRF2 in controlled cells and DDRGK1‐knockout 143B cells were detected by western blot. C) Regulation of NRF2 downstream mRNA levels by DDRGK1. The mRNA levels of NRF2 and its downstream genes in controlled cells and DDRGK1‐knockout 143B cells were detected by qPCR. D) Regulation of downstream proteins of NRF2 by DDRGK1. The expression of SOD1, SOD2, GPX1, and GPX4 in controlled cells and DDRGK1‐knockout 143B cells were detected by western blot. E) NRF2 overexpression rescues the ROS accumulation induced by H₂O₂. NRF2‐HA plasmid was transfected into DDRGK1‐knockout 143B cells, cells were subject to H₂O₂ and ROS levels were detected by flow cemetery. F) Kinematic decrease of NRF2 under H₂O₂ stimulation. The cells were treated with H₂O₂ for the indicated times and NRF2 expression was detected by western blot. G) Stability of NRF2 after DDRGK1 knockout. The DDRGK1 knockout and controlled 143B cells were treated with cycloheximide (50 µg mL⁻¹) and MG132 (10 µM) for indicated times. NRF2 levels were detected by western blot. H) Ubiquitination of NRF2. NRF2‐HA, ubiquitin‐Myc plasmids were transfected into HEK293T cells with or without co‐transfected with sh‐DDRGK1 plasmid. Ubiquitination level was detected by immunoprecipitation with HA antibody followed by western blot with anti‐ubiquitin antibody. (MFI, mean fluorescence intensity; Three samples for each statistical analysis, ****p < 0.0001).

Figure 5

DDRGK1 regulates NRF2 in a UFMylation‐independent pathway. A) Interaction between NRF2 and UFM1. The NRF2‐HA and ufm1‐HA plasmids were transfected into HEK293T cells. The interaction between NRF2 and ufm1 was detected by co‐immunoprecipitation with HA antibody followed by western blot with NRF2 and ufm1 antibody. B) Interaction between NRF2 and DDRGK1. The NRF2‐HA and DDRGK1‐Flag plasmids were transfected into HEK293T cells. The interaction between NRF2 and DDRGK1 was detected by co‐immunoprecipitation using anti‐HA immunomagnetic beads and anti‐Flag immunomagnetic beads. C) NRF2 levels of heart, lung, kidney, and cartilage in WT and K268R mouse detected by western blot. D) Re‐express of WT‐DDRGK1 and K267R‐DDRGK1 in DDRGK1 deficient cells rescued level of NRF2. E) Mitochondrial metabolism analysis. Cells were seeded in an XF microplate at an optimal density and cultured for 24 h. The cellular oxygen consumption rate (OCR) was measured using a Seahorse XF Analyzer. F) ROS level measurement. The controlled, DDRGK1‐knockout cells, and WT‐DDRGK1/K267R‐DDRGK1 re‐expressed cells were subjected to H₂O₂, then loaded with a DHE probe for 30 min and detected by flow cemetery. (MFI, mean fluorescence intensity; Three samples for each statistical analysis ****p < 0.0001).

Figure 6

DDRGK1 interacts with KEAP1. A) Interactions between DDRKG1 and NRF2, KEAP1, or CUL3. HEK293T cells were transfected with Myc‐DDRGK1 plasmid followed by co‐immunoprecipitation with Myc antibody. B) Interactions between DDRKG1 and KEAP1. HEK293T cells were transfected with HA‐KEAP1 plasmid then co‐immunoprecipitation with HA antibody followed by western blot with DDRGK1 antibody. C) Interactions between DDRKG1 and KEAP1. HEK293T cells were co‐transfected with HA‐KEAP1 and Myc‐DDRGK1 plasmids followed by co‐immunoprecipitation with Myc antibody and western blot with HA antibody. D) Interactions between DDRKG1 and KEAP1. HEK293T cells were co‐transfected with HA‐KEAP1 and Myc‐DDRGK1 plasmids then co‐immunoprecipitation with HA antibody followed by western blot with Myc antibody. E) Co‐localization of DDRGK1 and KEAP1 by immunofluorescence staining and observed under confocal microscopy. F) Domains and binding sites in Keap1 protein and truncating strategy for constructing different lengths of KEAP1 plasmids and DDRGK1 plasmids. G) DDRGK1 affects affinity between KEAP1 and NRF2, rather than with CUL3. HEK293T cells were transfected with HA‐KEAP1 and with or without co‐transfected with Myc‐DDRGK1 plasmid, then co‐immunoprecipitation with HA antibody. H) KI696 rescued NRF2 levels in DDRGK1 knockout cells. I) Identifying the binding site in Keap1 to interact with DDRGK1. Different lengths of KEAP1 plasmids and flag‐DDRGK1 plasmid were transfected into HEK293T cells respectively, and then co‐immunoprecipitation with HA antibody followed by western blot with flag antibody. J) Identifying the binding site in DDRGK1 to interact with KEAP1. Different lengths of DDRGK1 plasmids were transfected into HEK293T cells respectively, and then co‐immunoprecipitation with Flag antibody followed by western blot with KEAP1 antibody. K) Fc pulldown assay. 10 µg purified Fc, Fc‐DDRGK1 were mixed with KEAP1 and NRF2 protein, incubated with protein A/G beads, then detected by Co‐IP procedures. L) GST pulldown assay. 10 µg GST‐NRF2 protein was mixed with KEAP1 and DDRGK1 protein, incubated with protein GST beads, then detected by Co‐IP procedures. M) His pulldown experiment. 10 µg His‐KEAP1 protein was mixed with NRF2 and 5 or 10 µg DDRGK1 protein, incubated with His beads, then detected by Co‐IP procedures.

Figure 7

DDRGK1 knockout enhances chemosensitivity to doxorubicin. A,B) Sensitivity of DDRGK1‐knockout cells to DOX. The controlled and DDRGK1‐knockout cells were subjected to DOX(0.8 µg mL⁻¹) for 24 h, then loaded with a DHE probe for 30 min and detected ROS level by flow cemetery. C,D) NRF2 overexpression rescues the ROS accumulation induced by DOX. cells were subject to DOX(0.8 µg mL⁻¹) for 24 h and ROS levels were detected by flow cemetery. E) Half‐maximal inhibitory concentration (IC₅₀) of doxorubicin in the controlled and DDRGK1‐knockout 143B cells detected by CCK8 assays. F,G) Flow cemetery to detect cell apoptosis by Annexin V/DAPI staining in the controlled and DDRGK1‐knockout 143B cells in the presence or absence of DOX (0.8 µg mL⁻¹). H) Protein levels of cleaved caspase 3 and cleaved PARP in the controlled and DDRGK1‐knockout 143B cells in the presence or absence of DOX (0.8 µg mL⁻¹) detected by western blot. I,J) Flow cytometry to analyze cell apoptosis by Annexin V/DAPI staining in controlled cells, DDRGK1‐knockout cells, and Nrf2 overexpressed DDRGK1‐knockout cells in the presence of DOX (0.8 µg mL⁻¹). K,L) Flow cemetery to detect cell apoptosis by Annexin V/DAPI staining in the controlled and DDRGK1‐knockout 143B cells in the presence or absence of KI696. (MFI, mean fluorescence intensity; Ten samples for each statistical analysis, *p < 0.05; **p < 0.01; ***p < 0.001,****p < 0.0001).

Figure 8

DDRGK1 knockout inhibits osteosarcoma growth in vivo A) BALB/c nude mice were transplanted with controlled cells at the left back and DDRGK1‐knockout 143B cells at the right back. Two weeks later doxorubicin (15 mg kg⁻¹) was administered via intraperitoneal injection (n = 10). After another 2 weeks, the mice were sacrificed (n = 10). B) Tumor tissues dissected from mice at 4 weeks after transplantation. C) Tumor size measured every week after being treated with doxorubicin. D) Tumor weight measured at 4 weeks after transplanted. E) Relative decrease of tumor weight in NC + DOX, and KO + DOX groups. F) Immunofluorescence staining of Nrf2 in different tumor tissues. G) Immunofluorescence staining of Ki‐67 in different tumor tissues. H) Immunofluorescence staining of TUNEL in different tumor tissues. (*p < 0.05; **p < 0.01; ***p < 0.001,****p < 0.0001).

Figure 9

A schematic diagram of regulation of NRF2 activity by DDRGK1. At basal conditions, DDRGK1 binds competitively to the Kelch domain of KEAP1 and stops NRF2 ubiquitin‐mediated degradation, allowing its nuclear translocation, thus exerting its role as a transcription factor and promoting transcription of ARE genes to eliminate ROS. When DDRGK1 is knocked out, the inhibition of NRF2 binding to KEAP1 is abolished thereby promoting its ubiquitination and impairing the cell antioxidant capacity, leading to cell apoptosis and chemosensitivity induced by ROS‐mediated DNA damage.