Hypoxia-inducible factor-2alpha regulates the expression of TRAIL receptor DR5 in renal cancer cells

Abstract

To understand the role of hypoxia-inducible factor (HIF)-2alpha in regulating sensitivity of renal cancer cells to tumor necrosis factor-related apoptosis inducing ligand (TRAIL)-induced apoptosis, we transfected wild-type and mutant von Hippel Lindau (VHL) proteins into TRAIL-sensitive, VHL-negative A498 cells. We find that wild-type VHL, but not the VHL mutants S65W and C162F that do not degrade HIF proteins, cause TRAIL resistance. Knock down of the HIF-2alpha protein by RNA interference (short hairpin RNA) blocked TRAIL-induced apoptosis, decreased the level of TRAIL receptor (DR5) protein and inhibited the transcription of DR5 messenger RNA. By using luciferase constructs containing the upstream region of the DR5 promoter, we demonstrate that HIF-2alpha stimulates the transcription of the DR5 gene by activating the upstream region between -448 and -1188. Because HIF-2alpha is thought to exert its effect on gene transcription by interacting with the Max protein partner of Myc in the Myc/Max dimer, small interfering RNAs to Myc were used to lower the levels of this protein. In multiple renal cancer cell lines decreasing the levels of Myc blocked the ability of HIF-2alpha to stimulate DR5 transcription. PS-341 (VELCADE, bortezomib), a proteasome inhibitor used to treat human cancer, increases the levels of both HIF-2alpha and c-Myc and elevates the level of DR5 in renal cancer, sensitizing renal cancer cells to TRAIL therapy. Similarly, increasing HIF-2alpha in prostate and lung cancer cell lines increased the levels of DR5. Thus, in renal cancer cell lines expressing HIF-2alpha, this protein plays a role in regulating the levels of the TRAIL receptor DR5.

Fig. 1.

Induction of apoptosis in A498 cells by TRAIL. (A) A498 cells were treated with TRAIL (1 μg/ml) at 37°C for 4, 8 and 20 h. Cell lysates collected at each time point were subjected to western blotting using the indicated antibodies. (B) A498 cells treated with TRAIL were harvested at the indicated time points, washed three times with phosphate-buffered saline and fixed with 70% ethanol in phosphate-buffered saline for 2 h at 4°C. Following fixation, cells were stained with propidium iodide as described in Materials and Methods to detect the presence of hypodiploid cells (sub-G₁ peak) by flow cytometry.

Fig. 2.

Inhibition of A498 cell death induced by VHL. (A) A498 cells were transfected with pEGFP-C1 (empty vector) or pEGFP-VHL and (wild-type) vectors. Forty-eight hours after transfection, cells were washed with phosphate-buffered saline, FACS sorted and enhanced green fluorescent protein (EGFP)-positive cells retained. These cells were lysed and the extract was western blotted with the indicated antibodies. (B) The enhanced green fluorescent protein-sorted cells were grown for 24 h at 37°C before adding TRAIL (1 μg/ml). At the indicated hours of treatment, cells were harvested, lysed in sodium dodecyl sulfate sample buffer and the cleavage of caspase 8, caspase 3 and PARP determined by western blot. (C) A portion of the cells were harvested, washed with phosphate-buffered saline and cell cycle analysis (see legend Figure 1B) done by flow cytometry to detect the presence of a sub-G₁ peak.

Fig. 3.

HIF-2α shRNA decreases the level of DR5 mRNA and protein. (A) A498 cell lines containing scrambled shRNA (A498 clone C), HIF-2α shRNA I (clones 1–5), pcDNA3.1 vector alone or HIF-2α shRNA I-containing cells transfected with the wild-type HIF-2α cDNA were homogenized and extracts subjected to sodium dodecyl sulfate–polyacrylamide gel electrophoresis and western blotted with antibodies to HIF-2α, DR5 and CXCR4 and GAPDH as a loading control. (B) A498 clones containing scrambled shRNA or HIF-2α shRNA I were treated with TRAIL (1 μg/ml) for varying time periods and percent viability measured by the acid phosphatase assay as described in Materials and Methods. Each assay is performed in triplicate and the standard error of the mean is shown. (C) RNA was extracted from these A498 clones and subjected to quantitative real-time PCR to measure the levels of the DR5 mRNA using GAPDH as an internal control. The experiment was performed in triplicate and the standard error of the mean is shown.

Fig. 4.

DR5 promoter activity detected by luciferase assay. A498 clones were transfected with the pGVB2-based reporter plasmids (5 μg DNA per 60 mm dish) containing various lengths of 5′-human DR5 promoter gene (see reporter plasmid structure). Ten microliters of each cell lysate (protein concentration: 5 μg/μl) was used to determine luciferase activity that the values obtained were normalized by assaying the activity of a cotransfected pSV-β-galactosidase plasmid. The data shown are the mean with the standard error from the mean of three independent transfections.

Fig. 5.

Decreasing Myc protein levels regulates HIF-2α control of DR5 transcription. (A) Western blot of extracts from A498 cells studied in (B) were probed with antibodies to Myc, HIF-2α, p21, cyclin D2 or GAPDH as a loading control. (B) Regulation of DR5 promoter activity in A498 HIF-2α-knockdown cells. A498 cells containing HIF-2α shRNA were transfected with DR5 promoter construct (−1188) with or without the cDNA expressing HIF-2α Mut, MYC siRNA or MadMyc protein were assayed after 48 h. HIF-2α Mut contains a mutation (see Materials and Methods) in the coding sequence so it cannot be recognized by the shRNA. Luciferase activity was obtained using 5 μg of cell protein lysate that was normalized to cotransfected pSV-β-galactosidase plasmid. The luciferase values shown are the mean ± SD from three independent observations. (C) Schematic representation of the human DR5 promoter with seven non-canonical E-box elements is shown. The specific primers for each element used in the PCR in chromatin immunoprecipitation assay are described in the Materials and Methods. The sizes (bp) of the amplified regions for BS1–BS7 are 182, 130, 143, 145, 141, 154 and 112, respectively. A control reaction shown demonstrates c-Myc binding to a known site in GAPDH. The specificity of binding of c-Myc to each site is shown as the fold difference between c-Myc immunoprecipitation and rabbit IgG control (background) as demonstrated by densitometry. (D) The effect of mutation in the BS5 region on DR5 baseline promoter activity. A498 cells were transfected for 48 h with wild-type or E-box mutant DR5/−1188 promoter constructs. The luciferase activity was measured by using 5 μg of cell protein lysate and values normalized to cotransfected pSV-β-galactosidase plasmid. The luciferase values shown are the mean ± SD from three independent observations.

Fig. 6.

Regulation of DR5 transcription in PV-10 and RCC4 cells. (A) Levels of HIF-2α and Myc, MadMyc after siRNA and pCMVMadMyc transfections into RCC4 cells. RCC4 cells were transfected with siRNA or plasmids. Forty-eight hours later, the cells were homogenized and western blots were carried out. (B) The effect of lowering HIF-2α and Myc levels on DR5 transcription. RCC4 and PV-10 cells were transfected with DR5 (−1188)-Luc and HIF-2α siRNA, Myc siRNA or the Mad/Myc plasmid DNA. Forty-eight hours after transfection, luciferase assays were carried out. The values shown are the mean ± SD from three independent observations. (C) Induction of HIF-2α regulates DR5 levels. Following transfection with the indicated siRNAs for 48 h, A549 lung and PC3 prostate cancer cells were grown at 20 or 0.8% O₂. Cells were lysed and subjected to western blot analysis with HIF-2α, DR5 and GAPDH antibodies. (D) siRNA to HIF-1α does not decrease DR5 protein. RCC4 cells were transfected with siRNA to HIF-1α and 48 h later, homogenates were run on sodium dodecyl sulfate–polyacrylamide gel electrophoresis and western blotted for DR5, HIF-2α and HIF-1α. (E) The effect of knock down of HIF-1α on DR5 promoter transcription. RCC4 cells were transfected with DR5 (−1188)-Luc and HIF-1α or HIF-2α siRNA and 48 h after transfection, luciferase assays were carried out. The values shown are the mean ± SD from three independent observations.