Characterization and Regulation of Carrier Proteins of Mitochondrial Glutathione Uptake in Human Retinal Pigment Epithelium Cells

Abstract

Purpose: To characterize two mitochondrial membrane transporters 2-oxoglutarate (OGC) and dicarboxylate (DIC) in human RPE (hRPE) and to elucidate their role in the regulation of mitochondrial glutathione (mGSH) uptake and cell death in oxidative stress.

Methods: The localization of OGC and DIC proteins in confluent hRPE, polarized hRPE monolayers and mouse retina was assessed by immunoblotting and confocal microscopy. Time- and dose-dependent expression of the two carriers were determined after treatment of hRPE with H2O2, phenyl succinate (PS), and butyl malonate (BM), respectively, for 24 hours. The effect of inhibition of OGC and DIC on apoptosis (TUNEL), mGSH, and mtDNA was determined. Silencing of OGC by siRNA knockdown on RPE cell death was studied. Kinetics of caspase 3/7 activation with OGC and DIC inhibitors and effect of cotreatment with glutathione monoethyl ester (GSH-MEE) was determined using the IncuCyte live cell imaging.

Results: OGC and DIC are expressed in hRPE mitochondria and exhibited a time- and dose-dependent decrease with stress. Pharmacologic inhibition caused a decrease in OGC and DIC in mitochondria without changes in mtDNA and resulted in increased apoptosis and mGSH depletion. GSH-MEE prevented apoptosis through restoration of mGSH. OGC siRNA exacerbated apoptotic cell death in stressed RPE which was inhibited by increased mGSH from GSH-MEE cotreatment.

Conclusions: Characterization and mechanism of action of two carrier proteins of mGSH uptake in RPE are reported. Regulation of OGC and DIC will be of value in devising therapeutic strategies for retinal disorders such as AMD.

Figure 1

Gene and protein expression of OGC and DIC in human RPE. (A) mRNA expression of OGC and DIC in human RPE cells. Reverse-transcription PCR of DIC and OGC was performed using two intron-spanning primer pairs, respectively, for two RPE cDNA from two different donors. RPE1 from donor M; RPE2 from donor N; Neg, negative control. (B) Western blots showing both carriers are expressed in mitochondria of human RPE. Western blot analysis was performed along with positive control cell lysates, 721B (OGC) and MCF7 (DIC). Subunit IV of cytochrome c oxidase (COX IV) was used as the mitochondrial marker, and α-tubulin was used as the cytosolic marker.

Figure 2

Confocal images of immunofluorescence staining of OGC and DIC in RPE cells. (A, B) OGC and DIC carrier proteins (green) colocalized with mitochondria (Mitotracker, red). DAPI (blue) was used to counterstain nucleus. No primary antibody negative control is shown in lower panels of (A) and (B). Scale bar: 5 μm.

Figure 3

Effect of polarity on OGC and DIC expression in hRPE cells and confocal microscopy images of ZO-1 staining showing the breaks in tight junction induced by OGC and DIC inhibitors. The restoration of tight junction with GSH-MEE cotreatment was also shown. Western blot analysis were from confluent nonpolarized and polarized RPE cells (A–D), and ZO-1 confocal images were from polarized RPE cells (E, F). The area within the yellow line shows the breaks in tight junction which decreased in the GSH-MEE cotreatment groups as compared to PS/BM treated alone. Data are shown as mean ± SEM (n = 3), the mean TER of polarized cells was 380 ± 60 Ω·cm². ***P < 0.001. Scale bar: 50 μm. GSH-MEE, GSH monoethyl ester.

Figure 4

Dose- and time-dependent changes in OGC and DIC expression in hRPE cells exposed to H₂O₂-induced oxidative stress. hRPE cells were treated with H₂O₂ doses (50, 100, 200, 300 μM) for 24 hours (A, E), and for varying durations (2, 4, 6, 8, 24 hours) with 200 μM H₂O₂ (C, G). OGC and DIC protein were analyzed by Western blot analysis. β-actin was used as the normalizing protein for quantification shown in bar graphs. Both OGC and DIC expression decreased dose dependently with treatment with H₂O₂ compared with untreated control (B, F). OGC expression decreased significantly with time (P < 0.001 vs. control) with 200 μM H₂O₂ (D), and this trend was also observed in DIC (H). Data are mean ± SEM; n = 3. *P < 0.05. **P < 0.01. ***P < 0.001. NS, not significant.

Figure 5

Dose- and time-dependent inhibition of OGC and DIC expression by chemical inhibitors. hRPE cells were incubated with PS and BM, the chemical inhibitors of OGC and DIC respectively, at varying doses (2, 5, 10 mM) for 24 hours. The right panels show representative gels while the left panels show fold changes calculated by normalization of band density with β-actin from three independent experiments. Western blot analysis shows that the chemical inhibitors caused a depletion of OGC and DIC expression, especially at higher doses. Both OGC and DIC expression decreased significantly at 2 hours after treatment with the chemical inhibitors, respectively (P < 0.001 versus control). Data are mean ± SEM; n = 3. **P < 0.01. ***P < 0.001.

Figure 6

Inhibition of OGC and DIC expression in mitochondrial fraction by pharmacologic inhibitors and H₂O₂ in hRPE cells. (A–D) The mitochondrial fraction protein was extracted from the confluent hRPE cells, after incubation with 5 mM PS, 5 mM BM, and 200 μM H₂O₂ for 24 hours. OGC, DIC protein expression was analyzed using COX IV as a mitochondrial-specific marker. (B, D) OGC and DIC protein expression was significantly decreased in mitochondrial fraction with the chemical inhibitors treatment and H₂O₂, respectively (P < 0.001 versus control). The total amount of mitochondrial DNA (mtDNA) was determined from hRPE cells after treatment with 5 mM PS, 5 mM BM (E, F). (E) Shows the total amount of mitochondrial DNA (μg/mL) while (F) shows fold changes calculated by normalization of control. No significant difference of mtDNA yield with chemical inhibitors as compared to control was observed. Data are mean ± SEM; n = 3. *** P < 0.001.

Figure 7

Induction of apoptosis by OGC and DIC inhibitors and suppression by GSH-MEE. (A, C) hRPE cells were treated with a single 5 mM dose of either PS or BM in the absence/presence of GSH-MEE (2 mM) cotreatment for 24 hours, and cell death was determined by TUNEL staining (red). Nuclei were stained with DAPI (blue). Cell death was significantly higher in PS- and BM-treated cells than in cells cotreated with GSH-MEE. (B, D) Quantification of the TUNEL-positive cells. Cotreatment with GSH-MEE significantly inhibited cell apoptosis induced by OGC and DIC inhibitors. Data are mean ± SEM from three independent experiments performed in duplicate. ***P < 0.001. Scale bar: 50 μm.

Figure 8

Exacerbation of RPE cell death by silencing of OGC and attenuation of cell death by cotreatment with GSH-MEE. (A, B) Expression of OGC mRNA and protein levels in hRPE cells following knockdown of OGC. The mRNA was extracted 24 hours after transfection and the protein in mitochondrial fraction was obtained 48-hour posttransfection. OGC protein was analyzed by Western blotting using COX IV as a mitochondrial biomarker (B). Gene silencing significantly decreased OGC expression, a 75% decrease of OGC mRNA compared to that of siRNA control, P < 0.001 (A) and a 70% decrease of OGC protein compared to siRNA control, P < 0.001 (B), respectively. (C) After 24 hours transfection with either control siRNA or OGC siRNA, cells were treated with or without GSH-MEE (2 mM) for 24 hours. (D) Cells were treated with 300 μM H₂O₂ for 24 hours after transfection with either control siRNA or OGC siRNA. Cell death was determined by TUNEL staining (red). Nuclei were stained with DAPI (blue). Quantification of the TUNEL-positive cells in the experimental groups was shown in right lanes of (C, D). (C) Compared with control siRNA-treated cells, OGC knockdown significantly increased cell death while GSH-MEE cotreatment protected cells from apoptosis. (D) No significant difference of cell death between H₂O₂ treated alone and cotreated with control siRNA, while cotreatment with OGC silencing siRNA significantly increased cell death. Data are mean ± SEM from three independent experiments performed in duplicate. **P < 0.01. ***P < 0.001.

Figure 9

Time-dependent increase in caspase-3/7 generation with OGC and DIC inhibitors and suppression by GSH-MEE. (A) Kinetic measure of the number of caspase 3/7 positive cells shown over time in response to specified treatments. hRPE cells were plated and either left unstimulated or stimulated with 5 mM PS (green), 5 mM BM (yellow), or cotreated with 2 mM GSH-MEE (PS + GSH-MEE: blue and BM + GSH-MEE: gray). Images were acquired every 15 minutes. Automated real-time assessment by IncuCyte ZOOM, measured as green object count for all cells stained green with SYTOX Green, which was allowed to generate graphics of the data as soon as image analysis was complete. PS and BM induced significant cell apoptosis compared with control, which was markedly suppressed by GSH-MEE. (A) Caspase-3/7 activation increased for the first 2 hours with PS and BM which was inhibited by cotreatment with GSH-MEE. (B) IncuCyte ZOOM image (phase contrast and green fluorescence overlaid) of hRPE cells with stimulus at 6 hours. Cells undergoing apoptosis have membrane compromise, and their DNA are stained with SYTOX Green dye that is already present in the media along with the stimuli (original magnification ×20). (C) Quantification of the activation of caspase-3/7 (green fluorescent staining of nuclear DNA) at 6 hours in the treatment groups showing the suppression of inhibitor-induced caspase-3/7 activation by GSH-MEE. H₂O₂ treatment of RPE cells was used as a positive control. Data are shown as mean ± SEM (n = 4). ****P < 0.0001.

Figure 10

Selective depletion of mitochondrial GSH with OGC and DIC inhibitors and repletion with GSH-MEE cotreatment. GSH levels were measured in whole cell lysates (A, B), cytosolic fraction (C, D), and mitochondrial fraction (E, F). GSH levels were significantly decreased with inhibitors but is restored to above control levels with GSH-MEE (E, F). Whole cell and cytosolic fractions do not show such a change (A–D). The GSH levels were normalized to control level (=1). All experiments were carried out in triplicate. Data are shown as mean ± SEM of four independent experiments, *P < 0.05. **P < 0.01.

Figure 11

Competitive inhibition of GSH uptake shown by selective depletion of mitochondrial GSH with dimethyl 2-oxoglutarate and diethyl malate. GSH levels were measured in mitochondrial fraction. GSH levels were significantly decreased with competitive inhibitors, Dimethyl 2-oxoglutarate, P < 0.01 (A), and Diethyl malate, P < 0.05 (B), respectively. GSH levels were normalized to that of control (= 1). All experiments were carried out in triplicate. Data are shown as mean ± SEM of four independent experiments. *P < 0.05. **P < 0.01.

Figure 12

Expression of OGC and DIC (green) in mouse retinal layers in mouse RPE/choroid. (A) Immunofluorescence staining of OGC and DIC (green) of the posterior eye cup of mice. DAPI (blue) was used to counterstain nuclei. In upper panels diffuse green fluorescence can be seen in RPE. Strong green fluorescence was noted at the photoreceptor inner segment which harbor abundant mitochondria. Bottom panels show negative control devoid of primary antibody for OGC and DIC. Scale bar: 20 μm. (B) Protein expression of OGC and DIC in RPE/choroid of mice. Mitochondrial and cytosolic fractions of RPE/choroid from mice were isolated and immunoblotted for DIC and OGC. Both carriers are expressed in mitochondrial fraction of RPE/choroid and are absent in the cytosol. IS: photoreceptor inner segments; ONL, outer nuclear layer; OS, photoreceptor outer segments.