Restoration of the prolyl-hydroxylase domain protein-3 oxygen-sensing mechanism is responsible for regulation of HIF2α expression and induction of sensitivity of myeloma cells to hypoxia-mediated apoptosis

Abstract

Multiple myeloma (MM) is an incurable disease of malignant plasma B-cells that infiltrate the bone marrow (BM), resulting in bone destruction, anemia, renal impairment and infections. Physiologically, the BM microenvironment is hypoxic and this promotes MM progression and contributes to resistance to chemotherapy. Since aberrant hypoxic responses may result in the selection of more aggressive tumor phenotypes, we hypothesized that targeting the hypoxia-inducible factor (HIF) pathways will be an effective anti-MM therapeutic strategy. We demonstrated that MM cells are resistant to hypoxia-mediated apoptosis in vivo and in vitro, and that constitutive expression of HIF2α contributed to this resistance. Since epigenetic silencing of the prolyl-hydroxylase-domain-3 (PHD3) enzyme responsible for the O2-dependent regulation of HIF2α is frequently observed in MM tumors, we asked if PHD3 plays a role in regulating sensitivity to hypoxia. We found that restoring PHD3 expression using a lentivirus vector or overcoming PHD3 epigenetic silencing using a demethyltransferase inhibitor, 5-Aza-2'-deoxycytidine (5-Aza-dC), rescued O2-dependent regulation of HIF2α and restored sensitivity of MM cells to hypoxia-mediated apoptosis. This provides a rationale for targeting the PHD3-mediated regulation of the adaptive cellular hypoxic response in MM and suggests that targeting the O2-sensing pathway, alone or in combination with other anti-myeloma chemotherapeutics, may have clinical efficacy.

Fig 1. Characterization of O₂-sensing pathway and HIFα-subunit expression in MM cell lines.

(A) Immunoblots of HIF1α, HIF2α, PHD2 and PHD3 expression in OPM2, 8226 and MM1.S cell lines cultured under normoxic (22% O₂) or hypoxic (0.1% 0₂) conditions for 48 hrs. (B) Immunoblots of cleaved PARP and cleaved caspase 9 in 8226 and OPM2 cells. cultured under normoxic (22% O₂) or hypoxic (0.1% 0₂) conditions for 48 hrs. (C) Knockdown of HIFα-subunits sensitizes 8226 cells to hypoxia-mediated apoptosis. 8226 cells were transfected with HIF1α siRNA, HIF2α siRNA or non-specific scrambled control siRNA and then cultured under normoxic (22% O₂ white bars) or hypoxic (0.1% 0₂ black bars) conditions for 48 hrs and apoptosis was measured by flow cytometry for cleaved caspase-3. The data shows mean ± SEM of 3 independent experiments * P<0.05.

Fig 2. Exogenous expression of PHD3 expression inhibits HIF2α and restores sensitivity to hypoxia.

(A) Immunoblots of HIF1α, HIF2α, and PHD3 expression in 8226 cells stably transduced with either empty vector (8226EV) or PHD3 (8226PHD3) lentivirus. Cells were cultured under normoxic (N) or hypoxic (H) conditions for 48hrs, and the cell lysate was collected and immunoblotted for indicated proteins. (B) PHD3 expression inhibits VEGF expression. Isogenic 8226 cells were cultured under normoxic or hypoxic conditions as described above and the supernatant was collected and VEGF levels measured by ELISA. Data is presented as % change in mean VEGF (pg/mg protein) ± STDEV of 3 independent experiments (bracket* = P<0.05). (C) Exogenous expression of PHD3 increases sensitivity of 8226 cells to hypoxia-mediated apoptosis. Representative histogram of isogenic 8226 cells cultured under normoxic conditions for 48 hrs. Apoptosis was measured by flow cytometry for cleaved caspase-3 kit. (D) Isogenic 8226 cells were cultured under normoxic or hypoxic conditions for 48 hrs and apoptosis was measured by flow cytometry for cleaved caspase-3. Values represent mean ± STDEV of % apoptosis of 3 independent experiments.

Fig 3. 5-Aza-dC treatment reverses hypoxia resistance of MM cells.

(A) 5-Aza-dC treatment restored expression of PHD3 mRNA transcripts in various MM cell lines. (B) Treatment with 5-Aza-dC results in upregulation of PHD3 expression and concomitant downregulation of HIF2α protein. Immunoblot of isogenic 8226 cells treated with 5-Aza-dC under normoxic or hypoxic conditions for 48 hours. (C) 5-Aza-dC treatment sensitizes 8226EV cells to hypoxia-mediated apoptosis. Cells were cultured under normoxic (22% O₂) or hypoxic (0.1% O₂) conditions for 48 hrs in the presence or absence of indicated concentrations of 5-Aza-dC and apoptosis was measured by flow cytometry for cleaved caspase-3. The data shows mean ± SEM of 3 independent experiments. Brackets* = P<0.05.

Fig 4. Effects of bortezomib or rapamycin treatment on the induction of apoptosis in myeloma cells cultured under hypoxic conditions.

(A) Stably transduced 8226PHD3 or 8226EV control cells were cultured under normoxic (22% O₂) or hypoxic (0.1% O₂) conditions for 48 hrs in the presence or absence of indicated concentrations of bortezomib. Apoptosis was measured by flow cytometry for cleaved caspase-3. The mean ± SEM of 3 independent experiments is shown. Brackets* = P<0.05. (B) 8226PHD3 or 8226EV control cells were cultured under normoxic (22% O₂) or hypoxic (0.1% O₂) conditions for 48 hrs in the presence or absence of indicated concentrations of rapamycin as described above. Induction of apoptosis was measured by flow cytometry for cleaved caspase-3. The mean ± SEM of 3 independent experiments is shown. Brackets* = P<0.05. (C) Effects of hypoxia on expression of mTOR/AKT signaling pathway. 8226PHD3 or 8226EV control cells were cultured as described above and lysate collected and immunoblotted for indicated proteins. (D) Effects of hypoxia on VEGF expression. 8226PHD3 or 8226EV control cells were cultured under normoxic or hypoxic conditions as described above. VEGF levels were then measured in the supernatant by ELISA. Data is presented as % change in mean VEGF (pg/mg protein) ± STDEV of 3 independent experiments (bracket* = P<0.05).