Lenticular cytoprotection. Part 1: the role of hypoxia inducible factors-1α and -2α and vascular endothelial growth factor in lens epithelial cell survival in hypoxia

Abstract

Purpose: The prosurvival signaling cascades that mediate the unique ability of human lens epithelial cells to survive in their naturally hypoxic environment are not well defined. Hypoxia induces the synthesis of the hypoxia inducible factor HIF-1α that in turn, plays a crucial role in modulating a downstream survival scheme, where vascular endothelial growth factor (VEGF) also plays a major role. To date, no published reports in the lens literature attest to the expression and functionality of HIF-2α and the role it might play in regulating VEGF expression. The aim of this study was to identify the functional expression of the hypoxia inducible factors HIF-1α and HIF-2α and establish their role in regulating VEGF expression. Furthermore, we demonstrate a link between sustained VEGF expression and the ability of the hypoxic human lens epithelial cell to thrive in low oxygen conditions and resist mitochondrial membrane permeability transition (also referred to as lenticular cytoprotection).

Methods: Hypoxia inducible factor translation inhibitors were used to demonstrate the role of HIF-1α and HIF-2α and the simultaneous expression of both hypoxic inducible factors to determine their role in regulating VEGF expression. Axitinib, which inhibits lenticular cell autophosphorylation of its VEGF receptor, was employed to demonstrate a role for the VEGF-VEGFR2 receptor complex in regulating Bcl-2 expression. Specific antisera and western blot analysis were used to detect the protein levels of HIF-1α and HIF-2α, as well as the proapoptotic protein, BAX and the prosurvival protein, Bcl-2. VEGF levels were analyzed with enzyme-linked immunosorbent assay (ELISA). The potentiometric dye, 5,5',6,6'-tetrachloro1,1',3,3'-tetraethyl-benzimidazolylcarbocyanine iodide, was used to determine the effect of the inhibitors on mitochondrial membrane permeability transition.

Results: Cultured human lens epithelial cells (HLE-B3) maintained under hypoxic condition (1% oxygen) displayed consistent accumulation of VEGF throughout the 72 h incubation period. Using hypoxia inducible factor translation inhibitors targeting HIF-1α or HIF-2α, the specific inhibition of each protein did not diminish VEGF synthesis. The combined inhibition of HIF-1α and HIF-2α expression, using a double hypoxia inducible factor translation inhibitor, markedly decreased the level of VEGF. The inhibition of VEGF synthesis was associated with a profound deficiency in the level of the prosurvival protein, Bcl-2. Axitinib also prevented the VEGF-mediated expression of Bcl-2. The loss of VEGF coupled with the decrease in intracellular Bcl-2 correlated with marked mitochondrial depolarization, an early predictor of cellular apoptosis.

Conclusions: Our data support a model in which the sustained synthesis of VEGF in human lens epithelial cells, maintained under hypoxic condition, is regulated by a compensatory inter-relationship between HIF-1α and HIF-2α. VEGF acts as a prosurvival factor in hypoxic lens epithelial cells by maintaining consistent expression of the prosurvival protein Bcl-2, which likely prevents the translocation of cytosolic BAX to the outer mitochondrial membrane, thus preventing the initiation of mitochondrial depolarization.

Figure 1

Sustained and cumulative expression of VEGF in hypoxia and atmospheric oxygen. Detection of VEGF levels in hypoxia with ELISA. HLE-B3 cells were cultured in 25 cm² flasks with 20% FBS and switched to serum-free media 24 h before the experiment. The cells were incubated with 3 ml of serum-free media in hypoxia (1% oxygen) or remained in atmospheric oxygen (about 21% oxygen) for up to 72 h. Cell-free supernatants were collected in triplicate at 8, 24, 48, and 72 h and analyzed with ELISA to detect the VEGF levels. VEGF consistently accumulated throughout the 72 h incubation period regardless of whether the cells were maintained in hypoxia or atmospheric oxygen (p<0.05). A Student t test was performed to compare the VEGF levels between hypoxia and normoxia. Significantly higher levels of VEGF were detected at all-time points beyond the initial 8h point in hypoxia compared with atmospheric oxygen. Error bars are not shown at time points because the symbol is larger than the error bar.

Figure 2

HIF-1α inhibition does not affect VEGF expression. A: western blot analysis of HIF-1α expression in HLE-B3 cells treated with topotecan. Cell lysates were collected from cells treated with 500 nM topotecan in 0.01% DMSO after 8 h of hypoxic incubation. Control cells were mock treated with 0.01% DMSO and maintained in hypoxia as the topotecan-treated cells. Twenty μg protein/lane of cell lysates were analyzed with western blot analysis, and lane loading was normalized using a 1:1,000 dilution of rabbit anti-pan-actin antibody. Topotecan inhibited the expression of HIF-1α (1:1,000 dilution of rabbit anti- HIF-1α antibody) while a compensatory increase in HIF-2α was noted (1:1,000 dilution of rabbit anti- HIF-2α antibody). B, C: Densitometry analysis of HIF-1α and HIF-2α, respectively. D: Effect of HIF-1α inhibition on VEGF synthesis in hypoxia. HLE-B3 cells were cultured in 25 cm² flasks with 20% FBS and switched to serum-free media 24 h before the experiment. The cells were incubated with 3 ml of serum-free media containing 500 nM topotecan or 0.01% DMSO for 8 h of hypoxic exposure. Cell-free supernatants were collected in triplicate at the end of hypoxic incubation and analyzed for VEGF levels with ELISA. There was no significant difference in the VEGF levels between the topotecan-treated cells and control cells. (A Student t test was performed to compare the VEGF levels between control and treated sample, and the p value was greater than 0.05.) E: HIF-2α protein from HLE-B3 control cells compared with a standard lysate prepared from human embryonic kidney cells (HEK) 293 (Novus Biologicals, Litton, CO). The HIF-2α found in the standard lysate migrated at about 97 kDa. The HIF-2 α from HLE-B3 cells was about 80 kDa.

Figure 3

Loss of HIF-1α does not influence mitochondrial membrane potential. JC-1 analysis of HLE-B3 cells treated with topotecan. Cells were incubated in serum-free media containing 500 nM of topotecan in 0.01% DMSO for 3 h in hypoxia. Control cells were treated with 0.01% DMSO in serum-free media and likewise exposed for 3 h in hypoxia. After 3 h of hypoxic exposure, fresh, oxygenated media without the inhibitor but with the addition of 5 µg/ml of JC-1 dye were added and incubated at 37 °C for 30 min. JC-1 is a potentiometric dye that exhibits a membrane potential dependent loss as J-aggregates (polarized mitochondria) when transitioned to JC-1 monomers (depolarized mitochondria), as indicated by a fluorescence emission shift from red to green. Therefore, mitochondrial depolarization can be indicated by an increase in the green/red fluorescence intensity ratio. The media were removed, and fresh serum-free media without inhibitor or potentiometric dye were again added to the cells. A: Confocal imaging of mitochondrial membrane depolarization after inhibition of HIF-1α. Note the proportionally equivalent red and green fluorescence between topotecan-treated and mock-treated cells, indicating that the membrane potential was not altered by inhibiting HIF-1α expression. These images were taken from a randomly chosen field. (The bar represents 20 µm.) B: There was no significant difference in the green/red fluorescence ratio between the control and topotecan-treated cells. Student t test, p>0.05.

Figure 4

Inhibition of both HIF-1α and HIF-2α elicits the loss of VEGF expression. HLE-B3 cells were cultured in 25 cm² flasks with 20% FBS and switched to serum-free media 24 h before the experiment. The cells were incubated with 3 ml of serum-free media containing 0.5 µm, 5 µm, and 50 µm HIF-1α inhibitor, HIF-2α inhibitor, and HIF-1α/HIF-2α double translation inhibitor for 3 or 8 h in hypoxia. The effect of the inhibitors on HIF-1α and HIF-2α protein expression was analyzed using western blot analysis. Cell lysates (20 ug protein/lane) were identified using either anti-rabbit HIF-1α or HIF-2α at 1:1000 dilutions and lane loading was normalized using a 1:1000 dilution of rabbit anti- pan-actin antibody. Effect of HIF-1α and/or HIF-2α inhibition on VEGF levels in hypoxia. HLE-B3 cells were cultured in 25 cm² flasks with 20% FBS and switched to serum-free media 24 h before the experiment. The cells were incubated with 3 ml of serum-free media containing 0.5 µm, 5 µm and 50 µm HIF-1α inhibitor, HIF-2α inhibitor and HIF-1α/HIF-2α double translation inhibitor for 3 or 8 h in hypoxia. Cell free supernatants collected in triplicate were analyzed for VEGF levels by ELISA. The HIF-1α translation inhibitor at all concentrations inhibited HIF-1α without affecting the HIF-2α protein synthesis (A). Figure (B) and (C) represent the densitometry analysis for HIF-1α and HIF-2α protein expression. There was no significant difference in the VEGF levels between the control cells and cells treated with HIF-1α inhibitor (D). One-way ANOVA analysis was performed to compare the VEGF levels between the control and the three concentrations of HIF-1α inhibitor and the p value was >0.05. The HIF-2α translation inhibitor at all concentrations inhibited HIF-2α without affecting the HIF-1α protein synthesis (E). Figure (F) and (G) represent the densitometry analysis for HIF-1α and HIF-2α protein expression. There was no significant difference in the VEGF levels between the control cells and cells treated with HIF-2α inhibitor (H). One -way ANOVA analysis was performed to compare the VEGF levels between the control and the three concentrations of HIF-2α inhibitor and the p value was >0.05.The HIF-1α/HIF-2α double translation inhibitor at all concentrations inhibited HIF-2α and HIF-1α protein synthesis (I). Figure (J) and (K) represent the densitometry analysis for HIF-1α and HIF-2α protein expression. There was significant difference in the VEGF levels between the control cells and cells treated with HIF-1α/ HIF-2α double translation inhibitor at 3 h and 8 h of hypoxia. L: One-way ANOVA analysis was performed to compare the VEGF levels between the control and the three concentrations of HIF-1α/ HIF-2α double translation inhibitor and the p value was <0.05.

Figure 5

Inhibition of both HIF-1α and HIF-2α elicits mitochondrial membrane depolarization. Cells were treated with 50 μM of HIF-1α/HIF-2α translation inhibitor in 0.05% DMSO for 3 h in hypoxia. Control cells were treated with 0.05% DMSO. After the hypoxic exposure, the media was replaced with fresh oxygenated serum-free media containing 5 µg/ml JC-1 for 30 min in atmospheric oxygen. The media were removed, and fresh serum-free media were added to the cells. Control cells incubated with DMSO only were treated in a similar manner. We used serial confocal imaging to monitor mitochondrial membrane depolarization in HLE-B3 cells after treatment with the HIF-1α/HIF-2α double translation inhibitor. Sequential images of a random field of cells were taken every 150 s throughout the 60 min duration (Bar=20 µm). Confocal images of the HIF-1α/HIF-2α double translation inhibitor-treated cells indicated that there was a marked increase in green fluorescence intensity (indicative of depolarization) at both 30 and 60 min (B) compared with the control cells (A). C: HLE-B3 cells treated with the HIF-1α/HIF-2α double translation inhibitor exhibited a significantly increased green/red ratio compared with control, untreated cells.

Figure 6

Inhibition of pERK or pAkt does not affect VEGF expression. Western blot analysis of ERK1/2 and Akt phosphorylation inhibition on HIF-1α, HIF-2α, or downstream VEGF expression in hypoxia. HLE-B3 cells were incubated with 25 µM of LY294002 or 10 µM UO126 in 0.05% DMSO for 8 h under hypoxia. Control cells were incubated in serum-free media with 0.05% DMSO. Cell lysates were collected for western blot analysis, and cell-free supernatants were collected for ELISA. LY294002 blocked Akt phosphorylation without interrupting pERK, while UO126 prevented ERK phosphorylation without interfering with pAkt (A). Inhibition of Akt phosphorylation suppressed HIF-1α expression, but HIF-2α levels remained unchanged. Inhibition of ERK1/2 phosphorylation did not repress HIF-1α and HIF-2α expression (B). C and D represent the corresponding densitometry analysis. Neither phosphorylation inhibitor diminished VEGF synthesis relative to control cells (E). One-way ANOVA analysis was performed to compare the VEGF levels between the control and the LY294002- or UO126-treated samples, and the p value was >0.05.

Figure 7

Inhibition of pERK or pAkt does not affect mitochondrial membrane potential. JC-1 analysis of HLE-B3 cells treated with LY294002 or UO126. Cells were initially seeded onto 35 mm dishes and allowed to grow to semiconfluence. The cells were then pretreated with 25 µM LY294002 or 10 µM UO126 in 0.05% DMSO for 3 h in hypoxia to determine the state of mitochondrial membrane potential. Control cells were treated with 0.05% DMSO in serum-free media. After hypoxic exposure, the media were replaced with fresh, oxygenated serum-free media containing 5 µg/ml JC-1 for 30 min in atmospheric oxygen. The media were removed, and fresh serum-free media without inhibitor were added to the cells. Serial confocal imaging indicated no significant difference in the green/red ratio between the cells treated with LY294002 (A) or UO126 (B) and control cells (Student t test, p>0.05).

Figure 8

Inhibition of HIF-1α does not influence BAX or Bcl-2 levels. Cell lysates collected from cells treated with 500 nM topotecan were analyzed with western blot analysis for BAX and Bcl-2 levels. There was no change in the protein levels of BAX and Bcl-2 (A). (B) represents the corresponding densitometry analysis.

Figure 9

Inhibition of HIF-2α does not influence BAX or Bcl-2 levels. Cell lysates collected from cells treated with 0.5 µM, 5 µM, and 50 µM of HIF-2α translation inhibitor were analyzed with western blot analysis for BAX and Bcl-2 levels. There was no change in the protein levels of BAX and Bcl-2 (A). B and C represent the corresponding densitometry analysis.

Figure 10

Inhibition of pERK or pAkt does not influence BAX or Bcl-2 levels. Cell lysates collected from cells treated with 25 µM of LY294002 or 10 µM of UO126 were analyzed with western blot analysis for BAX and Bcl-2 levels. There was no change in the protein levels of BAX and Bcl-2 (A). B and C represent the corresponding densitometry analysis.

Figure 11

Inhibition of both HIF1α and HIF-2α diminishes Bcl-2 levels without affecting BAX levels. Cell lysates collected from cells treated with 0.5 µM, 5 µM, and 50 µM of HIF-1α/HIF-2α translation inhibitors were analyzed with western blot analysis for BAX and Bcl-2 levels. There was no change in the protein levels of BAX, and a significant decrease in the levels of Bcl-2 coupled with the loss of VEGF (Figure 4L) was observed with the double translation inhibitor (A). B and C represent the corresponding densitometry analysis.

Figure 12

Axitinib blocks binding of VEGF to its receptor, VEGFR2 and prompts a loss of Bcl-2 levels. Cell lysates collected from cells treated with 0.05 µM, 0.5 µM, and 5 µM of axitinib (AG013736) were analyzed with western blot analysis for the BAX and Bcl-2 levels. There was no change in the protein levels of BAX, and a significant decrease in the levels of Bcl-2 was observed with the VEGF receptor inhibitor (A). B and C represent the corresponding densitometry analysis.