Diabetes-Induced Mitochondrial Dysfunction in the Retina

Renu A. Kowluru and Saiyeda Noor Abbas

4–9 PURPOSE. Oxidative stress is increased in the retina in diabetes, tase are downregulated. We have reported that the long- and antioxidants inhibit activation of caspase-3 and the devel- term administration of antioxidants inhibits the development opment of retinopathy. The purpose of this study was to of retinopathy in diabetic rats and in galactose-fed rats (another investigate the effect of diabetes on the release of cytochrome model of diabetic retinopathy),3 suggesting an important role c from mitochondria and translocation of Bax into mitochon- for oxidative stress in the development of retinopathy in dia- dria in the rat retina and in the isolated retinal capillary cells. betes. Oxidative stress is involved directly in the upregulation ETHODS of vascular endothelial growth factor in the retina during early M . Mitochondria and cytosol fractions were prepared 10 from retina of rats with streptozotocin-induced diabetes and diabetes. Recent studies from our laboratory have shown that from the isolated retinal endothelial cells and pericytes incu- oxidative stress plays an important role, not only in the development of retinopathy in diabetes, but also in the resistance of bated in 5 or 20 mM glucose medium for up to 10 days in the 11 presence of superoxide dismutase (SOD) or a synthetic mi- retinopathy to arrest after good glycemic control is initiated. metic of SOD (MnTBAP). The release of cytochrome c into the Capillary cells and neurons are lost in the retina before other histopathology is detectable, and apoptosis has been cytosol and translocation of the proapoptotic protein Bax into 12–15 the mitochondria were determined by the Western blot tech- implicated as one of the mechanism(s). Apoptosis execu- nique and cell death by caspase-3 activity and ELISA assay. tion enzyme, caspase-3, and nuclear transcriptional factor (NF- ␬B) are activated in the retina when the duration of diabetes in RESULTS. Diabetes of 8 months’ duration in rats increased the rats is such that the capillary cell death and histopathology are release of cytochrome c into the cytosol and Bax into the detectable, and antioxidants inhibit such activations.4,16,17 Ox- mitochondria prepared from the retina, and this phenomenon idative stress is shown to be closely linked to apoptosis in a was not observed at 2 months of diabetes. Incubation of iso- variety of cell types18,19; however, the signaling steps involved lated retinal capillary cells with 20 mM glucose increased in oxidative-stress–induced retinal capillary cell apoptosis are cytochrome c content in the cytosol and Bax in the mitochon- not clear. dria, and these abnormalities were accompanied by increased Mitochondria are the major endogenous source of superox- cell apoptosis. Inclusion of SOD or its mimetic inhibited glu- ides and hydroxyl radicals.20 Reactive oxidant intermediates cose-induced release of cytochrome c, translocation of Bax, can trigger mitochondria to release cytochrome c, resulting in and apoptosis. activation of caspase-3.21–23 Overproduction of superoxides by CONCLUSIONS. Retinal mitochondria become leaky when the mitochondria is considered as a causal link between elevated duration of diabetes is such that capillary cell apoptosis can be glucose and the major biochemical pathways postulated to be observed; cytochrome c starts to accumulate in the cytosol and involved in the development of vascular complications in dia- Bax into the mitochondria. Inhibition of superoxides inhibits betes.24,25 Increasing evidence indicates that mitochondria are glucose-induced release of cytochrome c and Bax and inhibits intimately associated with the initiation of apoptosis. Mito- apoptosis in both endothelial cells and pericytes. Identifying chondrial changes are associated with the activation of apopto- the mechanism by which retinal capillary cells undergo apo- tic pathways resulting in diabetic neuropathy,26,27 impaired ptosis may reveal novel therapies to inhibit the development of kidney function,28 and myocardial abnormalities.29 However, retinopathy in diabetes. (Invest Ophthalmol Vis Sci. 2003;44: the involvement of mitochondria in the development of reti- 5327–5334) DOI:10.1167/iovs.03-0353 nopathy in diabetes is not clear. In the present study the effect of diabetes on mitochondrial iabetes increases oxidative stress, which plays an impor- dysfunction in the retina of rats and in the isolated retinal Dtant role in the development of diabetic complications.1–3 capillary cells was investigated by measuring the release of Oxidative stress is increased in retina in diabetes and in isolated cytochrome c into the cytosol and translocation of Bax into the retinal capillary cells (both endothelial cells and pericytes) mitochondria. The effect of inhibition of mitochondrial oxida- incubated in high-glucose medium.4 The antioxidant defense tive stress on capillary cell death is also determined. system is impaired in the retina in diabetes, GSH levels are decreased, superoxide production is increased, and mRNA levels of superoxide dismutase (SOD) and glutathione reduc- METHODS Rats From the Kresge Eye Institute, Wayne State University, Detroit, Michigan. Wistar rats (male, 200–220 g) were randomly assigned to normal or Supported in part by grants from the Juvenile Diabetes Research diabetic groups. Diabetes was induced with streptozotocin injection Foundation, The Thomas Foundation, and Research to Prevent Blind- (55 mg/kg body weight, intraperitoneal), and insulin was given as ness. needed to allow slow weight gain while maintaining hyperglycemia Submitted for publication April 7, 2003; revised June 9, 2003; (blood glucose levels of 20–25 mM). The rats were weighed two times accepted June 19, 2003. a week, and their food consumption was measured once every week. Disclosure: R.A. Kowluru, None; S.N. Abbas, None Glycated hemoglobin (GHb) was measured at 2 months of diabetes, The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertise- and every 3 months thereafter, using afﬁnity columns (kit 442-B; ment” in accordance with 18 U.S.C. §1734 solely to indicate this fact. Sigma-Aldrich). Diabetic rats and age-matched normal rats were killed Corresponding author: Renu A. Kowluru, Kresge Eye Institute, at 2 and 8 months of diabetes, and the retina was immediately re- Wayne State University, 4717 St. Antoine, Detroit, MI 48201; moved. These experiments conformed to the ARVO Statement for the [email protected]. Use of Animals in Ophthalmic and Vision Research.

Downloaded from iovs.arvojournals.org on 10/01/2021 5328 Kowluru and Abbas IOVS, December 2003, Vol. 44, No. 12

Capillary Cells TABLE 1. Severity of Hyperglycemia in Diabetic Rats Endothelial cells and pericytes were prepared from bovine eyes by a Glycated Body method described by Kennedy et al.30 and routinely used by us.4,31,32 Rats Hemoglobin Urine Volume Weight Endothelial cells were grown to 80% confluence in Petri dishes coated (n) (%) (mL/24 Hours) (g) with 0.1% gelatin in Dulbecco’s modified Eagle’s medium (DMEM) containing heparin, 10% fetal calf serum (heat inactivated), 10% serum Two months Normal 5 4.2 Ϯ 0.3 11 Ϯ 2 375 Ϯ 35 replacement (Nu-serum; BD Biosciences, Lincoln Park, NJ) endothelial Ϯ Ϯ Ϯ ␮ Diabetes 5 11.4 0.7 119 27 272 19 growth supplement (25 g/mL), and antibiotic-antimycotic in an en- Eight months vironment of 95% O2 and 5% CO2. Confluent cells from passages 4 Normal 5 4.5 Ϯ 0.5 14 Ϯ 5 492 Ϯ 53 to 8 were split and incubated under normoglycemic (5 mM glucose) Diabetes 6 11.9 Ϯ 1.7 128 Ϯ 20 313 Ϯ 31 or hyperglycemic (20 mM glucose) conditions for 1 to 10 days in the presence or absence of 20 mU/mL SOD,33,34 200 ␮M MnTBAP Glycated hemoglobin and 24 hours urine volume were measured (Mn(III)tetrakis(4-benzoic acid)porphyrin chloride; a cell-permeable at 2 months of diabetes, and every 3 months thereafter. Values are SOD mimetic; Biomol, Plymouth Meeting, PA35,36), 250 ␮M N-acetyl mean Ϯ SD. cysteine, or 250 ␮M lipoic acid.4 Pericytes were grown in DMEM supplemented with 10% fetal calf serum, antibiotics, and antimycotics, as described by us previ- Translocation of Bax into mitochondria was measured by perform- ously.4,31,32 Pericytes (passages 4–6) were incubated in DMEM con- ing Western blots for Bax in both mitochondrial and cytosolic fractions taining 2.5% fetal bovine serum in 5 or 20 mM glucose in the presence using rabbit polyclonal antibodies (Santa Cruz Biotechnology). and absence of antioxidants. Control incubations containing 20 mM mannitol always were run Cell Death simultaneously to rule out the effect of increased osmolarity. Each Cell death was determined by performing ELISA and by apoptotic DNA experiment was repeated with at least three separate cell preparations. laddering with cell death detection kits (Cell Death Detection ELISA and Apoptotic DNA ladder kits; Roche Diagnostics, Indianapolis, IN). Isolation of Mitochondria and Cytosol Cell death was further confirmed by measuring the activity of the apoptosis executor enzyme caspase-3.4 Mitochondria were isolated from the freshly removed retina or from Relative amounts of mono- and oligonucleosomes generated from cells by centrifugation.37,38 The retina was suspended in the mitochon- apoptotic cells were quantitated with the ELISA kit, using monoclonal dria buffer containing 20 mM HEPES-KOH (pH 7.5), 10 mM KCl, 1.5 antibodies directed against DNA and histones, respectively. The cyto-

mM MgCl2, 0.5 mM EDTA, 0.5 mM EGTA, 1 mM phenylmethylsulfonyl plasmic fraction of the cells was transferred to a streptavidin-coated fluoride, 10 ␮g/mL leupeptin, 10 ␮g /mL aprotinin, and 250 mM microtiter plate and incubated for 2 hours at room temperature with a sucrose, and gently homogenized with a glass homogenizer. The cells mixture of peroxidase-conjugated anti-DNA and biotin-labeled anti- were removed from the incubation Petri dishes by trypsin digestion, histone. The plate was then thoroughly washed, incubated with ABTS washed with ice-cold PBS, and homogenized in the mitochondria (2,2Ј-Azino-di[3-ethylbenzithiazolinesulfonate(6)] disodium salt; Roche buffer. The homogenate was centrifuged at 750g for 10 minutes at 4°C Diagnostics), and absorbance was measured at 405 nm against ABTS to remove nuclei and unbroken cells, and the supernatant was centri- solution as a blank. After separation of the cytoplasmic fraction, the fuged at 10,000g for 15 minutes. The resultant mitochondrial pellet nuclear pellet was suspended in 50 mM sodium phosphate buffer (pH ␮ was lysed in 50 L of 20 mM Tris (pH 7.4), 100 mM NaCl, 1 mM 7.5) containing 2 mM NaCl and 0.05 mM Na2HPO4 (pH 7.5) and phenylmethylsulfonyl fluoride, 10 ␮g/mL leupeptin, and 10 ␮g/mL sonicated. DNA was measured in this fraction, and apoptosis was aprotinin, and the supernatant was centrifuged at 100,000g for 60 normalized to micrograms of DNA. minutes to obtain the cytosolic fraction. Protein was determined in DNA fragmentation was detected with a kit (Apoptotic DNA Ladder both mitochondrial and cytosolic fractions by the bicinchoninic acid Kit; Roche Diagnostics). Briefly, the cells were washed twice with PBS, assay (Sigma-Aldrich). resuspended in PBS, and incubated for 10 minutes with an equal volume of lysis buffer containing 10 mM Tris-HCl (pH 7.4), 6 M guanidine-HCl, 10 mM urea and EDTA, and 0.2% Triton X-100. The Cytochrome c Release and Bax Translocation samples were passed through glass fiber fleece by centrifugation and Release of cytochrome c was quantitated by Western blot techniques the nucleic acid bound to the glass fibers was eluted. The DNA was by measuring the expression of cytochrome c in mitochondrial and applied to a 1.5% agarose gel, and the bands were then visualized by cytosolic fractions. Mitochondrial (20 ␮g) and cytosolic (40 ␮g) pro- ethidium bromide staining and photographed. teins were separated on 15% reducing polyacrylamide gel and then transferred to nitrocellulose membranes. The membranes were Statistical Analysis blocked in 5% milk, followed by incubation with a polyclonal antibody Data are reported as the mean Ϯ SD, and experimental groups were against cytochrome c (Santa Cruz Biotechnology, Santa Cruz, CA). After compared using the nonparametric Kruskal-Wallis test followed by the washing, the membranes were incubated with anti-rabbit IgG horse- Mann-Whitney test for multiple-group comparison. Similar conclusions radish peroxidase–conjugated antibody in blocking buffer for 1 hour, were reached also by using ANOVA with the Fisher or Tukey test. washed, and developed using a Western blot chemiluminescence detection kit (ECL-Plus; Amersham Biosciences, Arlington Heights, IL). To ensure that the same content of mitochondrial or cytosolic RESULTS protein was loaded in each lane, after they were blotted for cyto- Diabetic Rats chrome c, the membrane were incubated in stripping buffer (62.5 mM Tris-HCl [pH 6.8], 100 mM mercaptoethanol, and 2% sodium dodecyl Glycated hemoglobin (an index of severity of hyperglycemia) sulfate) at 50°C for 30 minutes and washed (three times for 10 minutes and urine volumes were elevated by two- to threefold in the each). The membranes were then incubated with anti-cytochrome c diabetic rats compared with their age-matched normal rats. oxidase subunit IV (Cox IV; Molecular Probes, Eugene, OR) or ␤-actin The body weights were significantly lower in diabetic rats at 2 (Santa Cruz Biotechnology) and developed with the Western blot and 8 months of diabetes compared with the age-matched analysis detection kit. normal control (Table 1).

Downloaded from iovs.arvojournals.org on 10/01/2021 IOVS, December 2003, Vol. 44, No. 12 Retinal Mitochondrial Dysfunction in Diabetes 5329

Diabetes of 2 months’ duration in rats had no effect on the release of cytochrome into the cytosol. The cytochrome c content of the cytosolic fraction of the retina was similar in diabetic rats and age-matched normal rats, but 8 months of diabetes resulted in an approximate twofold increase in the cytosolic content of cytochrome c, compared with the age- matched normal rats (Fig. 1A). However, the expression of ␤-actin among the lanes did not vary. Similarly, the expression of Bax was significantly increased in the mitochondria obtained from the retina of rats diabetic for 8 months compared with the age-matched normal control rats (P Ͻ 0.02). However, there was no significant difference in the Bax expression in the mitochondria obtained from the retina of rats diabetic for 2 months compared with that from age-matched normal rats (Fig. 1B). In the same retina, as reported previously,4 caspase-3 activity was increased by 40% at 8 months of diabetes compared with age-matched normal control rats, but 2 months of diabetes had no effect on retinal caspase-3 activity.4 Isolated Capillary Cells: Studies with Endothelial Cells The cytosolic content of cytochrome c was similar in the cells incubated in 5 or 20 mM glucose for 3 days, but was increased by more than fourfold when the incubation in 20 mM glucose medium was extended to 5 days. The release of cytochrome c was not further increased when the duration of incubation with 20 mM glucose was further extended to 10 days (Fig. 2A). Despite the significant differences in the content of cytochrome c among various cytosol preparations, the content of ␤-actin did not vary. As shown in Figure 2B, the expression of Bax in the mitochondria obtained from the same preparations was increased by 40% in the cells incubated with 20 mM glucose for 3 days compared with the cells incubated in 5 mM glucose, but was increased by fourfold when the incubation was extended to 5 days. There was no additional increase in Bax content in the mitochondria when the incubation with glucose was allowed to continue for 10 days. The expression of Cox IV in various lanes was identical. Cell Death Apoptosis levels, as determined by measuring the cytoplasmic nucleosomal DNA, were not changed in the endothelial cells incubated in 20 mM glucose for 3 days compared with the cells incubated in 5 mM, but were increased by 50% when the incubation with 20 mM glucose was increased to 5 days (Fig. 3). Similarly, increased DNA laddering was observed in these cells at 5 days of incubation with 20 mM glucose (data not shown). Inclusion of SOD or MnTBAP in the medium for 5 days inhibited glucose-induced release of cytochrome c from the mitochondria into the cytosol, and in the same mitochondrial

‹

FIGURE 1. Effect of diabetes on the (A) release of cytochrome c in the cytosol and (B) translocation of Bax into mitochondria. The mitochondrial and cytosolic fractions were prepared from retina freshly har- vested from rats diabetic for 2 or 8 months and age-matched normal control rats. Cytochrome c and Bax contents were determined by Western blot analyses, and the band intensities were adjusted to the expression of the ␤-actin or Cox IV in cytosolic and mitochondrial fractions, respectively. The Western blots are representative of ﬁve rats in each group, and the retina from each rat was analyzed in duplicate in two separate experiments. *P Ͻ 0.05 and #P Ͼ 0.05 compared with the age-matched normal control.

Downloaded from iovs.arvojournals.org on 10/01/2021 5330 Kowluru and Abbas IOVS, December 2003, Vol. 44, No. 12

FIGURE 3. Glucose-induced apoptosis of retinal endothelial cells. Ap- optosis was measured by performing ELISA for cytoplasmic histone- associated DNA fragments. Data were obtained from cells incubated with glucose for 5 days, and the values were adjusted to the total DNA. No signiﬁcant apoptosis was observed in these cells when the glucose exposure was 3 days or less. *P Ͻ 0.05 compared with 5 mM glucose; P Ͻ 0.05 compared with 20 mM glucose.

fractions, inhibited increased Bax expression. As stated earlier, despite the significant differences in the content of cytochrome c and Bax among various preparations, the content of ␤-actin and Cox IV did not vary in their respective fractions (Fig. 3). Other antioxidants, including N-acetyl cysteine and lipoic acid had beneficial effects similar to those of SOD or its mimetic, MnTBAP (data not shown). Similarly, the addition of SOD or MnTBAP inhibited glucose- induced increased apoptosis of endothelial cells, and this was confirmed by ELISA assays (Fig. 4). Both SOD and MnTBAP also inhibited activation of caspase-3 induced by glucose.

Isolated Retinal Pericytes Incubation of retinal pericytes with 20 mM glucose showed mitochondrial dysfunction similar to that observed with endothelial cells (Fig. 5). Cytochrome c release into the cytosol and Bax translocation into the mitochondria were increased by three to fourfold in the cells incubated in 20 mM glucose for 5 days compared with that in 5 mM glucose. This was accompanied by a twofold increase in apoptosis. No signiﬁcant translocation of cytochrome c or Bax and apoptosis was observed in these cells when the glucose exposure was of 3 days’ or less duration. Addition of MnTBAP inhibited glucose-induced mitochondrial dysfunction and apoptosis in these retinal pericytes (Fig. 5).

days. At the end of the desired time of incubation, the cells were trypsinized and mitochondrial and cytosolic fractions were isolated by differential centrifugation. The purity of mitochondrial and cytosolic fractions was determined by measuring Cox IV expression. Each measurement was performed in duplicate with three to four separate cell FIGURE 2. Time course of (A) glucose-induced release of cytochrome preparations. The results obtained in 5 mM glucose medium did not c and (B) translocation of Bax in retinal endothelial cells. Bovine retinal change with the duration of incubation and are considered as 100%. endothelial cells were incubated in 5 or 20 mM glucose for up to 10 *P Ͻ 0.05 compared with 5 mM glucose.

Downloaded from iovs.arvojournals.org on 10/01/2021 IOVS, December 2003, Vol. 44, No. 12 Retinal Mitochondrial Dysfunction in Diabetes 5331

FIGURE 4. Effect of SOD on glucose- induced mitochondrial dysfunction. Retinal endothelial cells were incubated with 5 or 20 mM glucose for 5 days in the presence or absence of 20 mU/mL SOD or 200 ␮M MnTBAP. The cytochrome c and Bax contents were determined in the mitochondrial and cytosolic fractions by West- ern blot analysis. The experiments were repeated with at least three separate cell preparations, and each measurement was made in duplicate. Similar beneﬁcial effects were observed when the cells were incubated in the presence of 250 ␮M N- acetyl cysteine or 250 ␮M ␣-lipoic acid.

18,19 DISCUSSION processes. The mechanism by which oxidative stress can increase apoptosis may involve increased membrane lipid per- The results presented herein show for the ﬁrst time that the oxidation, increased oxidative injury to other macromolecules, retinal mitochondria become leaky and the cytochrome c alterations in signal transduction, change in cellular redox starts to accumulate in the cytosolic fraction when the potentials or depletion of glutathione (GSH).40,41 An altered duration of diabetes in rats is such that capillary cell apo- gene proﬁle of scavenging enzymes is reported in the retinal ptosis can be detected in the retina. Further, both in retinal pericytes obtained from patients with diabetes, and this corre- endothelial cells and pericytes, high glucose increases the lates with the overexpression of the cell death protease gene, release of cytochrome c into the cytosol and Bax into the suggesting an important role for oxidative stress in the pericyte mitochondria, which can be prevented by reducing super- dropout that occurs in diabetic retinopathy.10 Increased oxi- oxide levels, thus suggesting that retinal mitochondria ex- periences dysfunction in diabetes. Retina and its isolated dative stress in diabetes is shown to play a critical role in capillary cells experience increased oxidative stress in high- advanced glycation end-product (AGE)–induced and palmitate- induced apoptosis of retinal capillary cells that can be inhibited glucose conditions, and increased oxidative stress is postu- 42 lated to play an important role in the development of dia- by antioxidants. Our previous studies have shown that in- betic complications.1–3 High glucose can induce apoptosis creased oxidative stress plays an important role in the activa- and activate caspases, including caspase-3, in retinal capil- tion of retinal caspase-3 and in the development of retinopathy 3,4 lary and Mu¨ller cells,4,15–17,39 and we provide data showing in diabetes. The results of the present study show that the that glucose-induced apoptosis can be inhibited by lowering inhibition of glucose-induced release of cytochrome c into the superoxides levels. cytosol and translocation of Bax into the mitochondria in Oxidative stress is closely linked to apoptosis in a variety of retinal endothelial cells and pericytes by SOD is accompanied cell types. It can alter both signal transduction and genomic by the inhibition of their apoptosis.

Downloaded from iovs.arvojournals.org on 10/01/2021 5332 Kowluru and Abbas IOVS, December 2003, Vol. 44, No. 12

FIGURE 5. Glucose-induced mitochondrial dysfunction and apoptosis in retinal pericytes. Pericytes isolated from bovine retina were incubated in 5 or 20 mM glucose for up to 10 days in the presence or absence of SOD or MnTBAP. Top: release of cytochrome c into the cytosol and Bax into mitochondria after a 5-day incubation with 20 mM glucose. The results are representative of ﬁve separate experiments using three different preparations. Bottom: apoptosis measured by ELISA. Results are the mean Ϯ SD obtained from three separate preparations, and each measurement was made in duplicate. *P Ͻ 0.05 compared with 5 mM glucose and #P Ͻ 0.05 compared with 20 mM glucose.

Mitochondria play a key role in regulating apoptosis, reac- thelial cells and pericytes are accompanied by inhibition of tive oxidant intermediates can trigger mitochondria to release apoptosis in these cells. cytochrome c and apoptosis-inducing factor, and increased Bax, a proapoptotic protein, enhances the release of cyto- lipid peroxidation itself can damage mitochondrial membrane chrome c by translocating to the mitochondria and by inducing potential and provoke apoptosis.21,22 Once cytochrome c is a mitochondrial permeability transition.43,44 Others have re- released from mitochondria, it activates caspase-9, which ini- ported that the expression of Bax is increased in the retina in tiates a cascade of events that activates caspase-3 and results in diabetes and in retinal pericytes incubated in high-glucose DNA fragmentation. Cytochrome c can induce apoptosis if it is medium, and this overexpression in retinal pericytes is associ- present in the cytoplasm in the oxidized state, and under ated with their apoptosis.16 In retinal sections, Bax immuno- normal conditions cytoplasmic GSH maintains cytochrome c in staining is shown to be present in ganglion and vascular cells, the reduced state.41 In diabetes, retinal GSH levels are de- the cell types known to undergo accelerated cell death in creased and glutathione redox cycle enzymes are impaired,5–7 diabetes.13,15,16 The results of the present study demonstrate which raises the possibility that the reduction of cytochrome c that it is the mitochondrial fraction of the retina where Bax is also impaired, and here we provide data that clearly show expression is increased in diabetes, and we have conﬁrmed our that the release of cytochrome c into the cytosol is increased by in vivo results using isolated retinal endothelial cells and peri- approximately twofold. The increased release of cytochrome c cytes incubated in high-glucose medium for 5 days. This is seen also when the endothelial cells and pericytes from the strengthens the possible involvement of mitochondria in the retina are incubated in high-glucose medium for 5 days, but not apoptosis of both pericytes and endothelial cells that occurs in for 3 days. This time course of mitochondrial dysfunction is diabetes.12,13 Romeo et al.39 have suggested a possible involve- similar to the activation of caspase-3 that we have reported ment of Bax in retinal capillary cell apoptosis in diabetes, but previously.6 Thus, the present study provided data to show their study did not identify the effect of diabetes on subcellular that the inhibition of mitochondrial changes in retinal endo- distribution of Bax in the retina or its capillary cells. Our study

Downloaded from iovs.arvojournals.org on 10/01/2021 IOVS, December 2003, Vol. 44, No. 12 Retinal Mitochondrial Dysfunction in Diabetes 5333

is the first to provide data that demonstrate the possible in- 11. Kowluru RA. Effect of re-institution of good glycemic control on volvement of mitochondrial dysfunction in the apoptosis of retinal oxidative stress and nitrative stress in diabetic rats. Diabe- both retinal endothelial cells and pericytes in diabetes. tes. 2003;52:818–823. Release of cytochrome c is considered a key event in the 12. Mizutani M, Kern TS, Lorenzi M. Accelerated death of retinal activation of caspase-3, a downstream pivotal step in the initi- microvascular cells in human and experimental diabetic retinopa- ation of apoptosis.29 Cells deficient in caspase-3 are resistant to thy. J Clin Invest. 1996;97:2883–2890. apoptosis,45 and activation of caspase-3 alone is sufficient to 13. Kern TS, Tang J, Mizutani M, et al. Response of capillary cell death 46 to aminoguanidine predicts the development of retinopathy: com- cause cell death in cardiac muscle. The results presented parison of diabetes and galactosemia. Invest Ophthalmol Vis Sci. herein demonstrate that caspase-3 activation in diabetes is 2000;41:3972–3978. associated with mitochondrial dysfunction. 14. Barber AJ, Lieth E, Khin SA, et al. Neural apoptosis in the retina Hyperglycemia-induced overproduction of superoxides by during experimental and human diabetes: early onset and effect of mitochondria is considered as a causal link between elevated insulin. J Clin Invest. 1998;102:783–791. glucose and the major biochemical pathways postulated to be 15. Podesta F, Romeo G, Liu WH, et al. Bax is increased in the retina involved in the development of vascular complications in dia- of diabetic subjects and is associated with pericyte apoptosis in betes,25,47 and overexpression of Mn-SOD is reported to sup- vivo and in vitro. Am J Pathol. 2000;156:1025–1032. press glucose-induced collagen accumulation in cultured mes- 16. Kowluru RA, Koppolu P, Chakrabarti S, Chen S. Diabetes-induced angial cells.48 We have provided evidence that mitochondrial activation of nuclear transcriptional factor in the retina, and its dysfunction, apoptosis, and caspase-3 activation induced by inhibition by antioxidants. Free Radical Research. In press. high glucose in both retinal endothelial cells and pericytes are 17. Mohr S, Xi X, Tang J, Kern TS. Caspase activation in retinas of inhibited by SOD and its mimetic, MnTBAP suggesting that a diabetic and galactosemic mice and diabetic patients. Diabetes. mitochondria-dependent pathway is operating in both the di- 2002;51:1172–1179. abetic retina and its isolated capillary cells, and SOD produc- 18. Baumgartner-Parzer SM, Wagner L, Pettermann M, et al. High- tion is causally involved in the hyperglycemia-induced apopto- glucose-triggered apoptosis in cultured endothelial cells. Diabetes. 1995;44:1323–1327. sis of retinal capillary cells. 19. Du X, Stocklauser-Farber K, Rosen P. Generation of reactive oxy- Thus, our data strongly suggest that hyperglycemia-induced gen intermediates, activation of NF-kappaB, and induction of apo- retinal capillary cell death most likely is initiated by the mito- ptosis in human endothelial cells by glucose: role of nitric oxide chondrial cytochrome c–mediated caspase-3 activation path- synthase? Free Radic Biol Med. 1999;27:752–763. way. Understanding the signaling pathway(s) involved in the 20. Sandbach JM, Coscun PE, Grossniklaus HE, et al. Ocular pathology retinal capillary cell death will elucidate important molecular in mitochondrial superoxide dismutase (Sod2)-deficient mice. In- targets for future pharmacological interventions. vest Ophthalmol Vis Sci. 2001;42:2173–2178. 21. Anuradha CD, Kanno S, Hirano S. Oxidative damage to mitochondria is a preliminary step to caspase-3 activation in fluoride-in- Acknowledgments duced apoptosis in HL-60 cells. Free Radic Biol Med. 2001;31:367– 373. The authors thank Prashant Koppolu for technical assistance. 22. Phaneuf S, Leeuwenburgh C. Cytochrome c release from mitochondria in the aging heart: a possible mechanism for apoptosis with age. Am J Physiol. 2002;282:R423–R430. References 23. Kristal BS, Koopmans SJ, Jackson CT, et al. Oxidant-mediated 1. Baynes JW. Role of oxidative stress in development of complica- repression of mitochondrial transcription in diabetic rats. Free tions in diabetes. Diabetes. 1991;40:405–412. Radic Biol Med. 1997;22:813–822. 2. Armstrong D, Abdella N, Salman A, et al. Relationship of lipid 24. Nishikawa T, Edelstein D, Brownlee M. The missing link: a single peroxides to diabetic complications. J Diab Comp. 1992;6:116– unifying mechanism for diabetic complications. Kidney Int (sup- 122. plement). 2000;77:S26–S30. 3. Kowluru RA, Tang J, Kern TS. Abnormalities of retinal metabolism 25. Brownlee M. Biochemistry and molecular cell biology of diabetic in diabetes and experimental galactosemia. VII. Effect of long-term complications. Nature. 2001;414:813–820. administration of antioxidants on the development of retinopathy. 26. Russell JW, Golovoy D, Vincet A, et al. High glucose-induced Diabetes. 2001;50:1938–1942. oxidative stress and mitochondrial dysfunction in neurons. FASEB 4. Kowluru RA, Koppolu P. Diabetes-induced activation of caspase-3 J. 2002;16:1738–1748. in retina: effect of antioxidant therapy. Free Rad Res. 2002;36: 27. Schmeichel AM, Schmelzer JD, Low PA. Oxidative injury and 993–999. apoptosis of dorsal root ganglion neurons in chronic experimental 5. Kowluru RA, Kern TS, Engerman RL, Armstrong D. Abnormalities diabetic neuropathy. Diabetes. 2003;52:165–171. of retinal metabolism in diabetes or experimental galactosemia. III. 28. Verzola D, Bertolotto MB, Villaggio B, et al. Taurine prevents Effects of antioxidants. Diabetes. 1996;45:1233–1237. apoptosis induced by high ambient glucose in human tubule renal 6. Kowluru RA, Kern TS, Engerman RL. Abnormalities of retinal cells. J Invest Med. 2002;50:443–451. metabolism in diabetes or experimental galactosemia. IV. Antiox- 29. Cai L, Li W, Wang G, et al. Hyperglycemia-induced apoptosis in idant defense system. Free Radic Biol Med. 1996;22:587–592. mouse myocardium: mitochondrial cytochrome C-mediated 7. Szabo ME, Haines D, Garay E, et al. Antioxidant properties of caspase-3 activation pathway. Diabetes. 2002;51:1938–1948. calcium dobesilate in ischemic/reperfused diabetic rat retina. Eur 30. Kennedy A, Frank RN, Sotolongo LB, et al. Proliferative response J Pharmacol. 2001;428:277–286. and macromolecular synthesis by ocular cells cultured on extra- 8. Grant MB, Ellis EA, Wachowski MB, Murray FT. Free radical de- cellular matrix materials. Curr Eye Res. 1990;9:307–322. rived oxidant localization in retinas of BBZ/WOR diabetic rats 31. Kowluru RA, Engerman RL, Case GL, Kern TS. Retinal glutamate in (Abstract). Diabetes. 1996;45(suppl 2):192A. diabetes and effect of antioxidants. Neurochem Int. 2001;38:385– 9. Li W, Yanoff M, Jian B, He Z. Altered mRNA levels of antioxidant 390. enzymes in pre-apoptotic pericytes from human diabetic retinas. 32. Kowluru A. Diabetes-induced elevations in retinal oxidative stress, Cell Mol Biol. 1999;45:59–66. protein kinase C and nitric oxide are inter-related. Acta Diabetol. 10. Obrosova IG, Minchenko AG, Marinescu V, et al. Antioxidants 2001;38:179–185. attenuate early up regulation of retinal vascular endothelial growth 33. Valencia A, Moran J. Role of oxidative stress in the apoptotic cell factor in streptozotocin-diabetic rats. Diabetologia. 2001;44: death of cultured cerebellar granule neurons. J Neurosci Res. 1102–1110. 2001;64:284–287.

Downloaded from iovs.arvojournals.org on 10/01/2021 5334 Kowluru and Abbas IOVS, December 2003, Vol. 44, No. 12

34. Anane R, Creppy EE. Lipid peroxidation as pathway of aluminium programmed cell death. Free Radic Biol Med. 2001;31:697– cytotoxicity in human skin ﬁbroblast cultures: prevention by su- 703. ϩ peroxide dismutase catalase and vitamins E and C. Hum Exp 42. Denis U, Lecomte M, Paget C, et al. Advanced glycation end- Toxicol. 2001;20:477–481. products induce apoptosis of bovine retinal pericytes in culture: 35. Szabo C, Day BJ, Salzman AL. Evaluation of the relative contribu- involvement of diacylglycerol/ceramide production and oxidative tion of nitric oxide and peroxynitrite to the suppression of mito- stress induction. Free Radic Biol Med. 2002;33:236–247. chondrial respiration in immunostimulated macrophages using a manganese mesoporphyrin superoxide dismutase mimetic and 43. Adachi M, Ishii H. Role of mitochondria in alcoholic liver injury. peroxynitrite scavenger. FEBS Lett. 1996;381:82–86. Free Radic Biol Med. 2002;32:487–491. 36. Patel M. Inhibition of neuronal apoptosis by a metalloporphyrin 44. Karpinich NO, Tafani M, Rothman RJ, et al. The course of etopo- superoxide dismutase mimic. J Neurochem. 1998;71:1068–1074. side-induced apoptosis from damage to DNA and p53 activation to 37. Tafani M, Cohn JA, Karpinich NO, et al. Regulation of intracellular mitochondrial release of cytochrome c. J Biol Chem. 2002;277: pH mediates Bax activation in HeLa cells treated with staurospor- 16547–16562. ine or tumor necrosis factor-alpha. J Biol Chem. 2002;277:49569– 45. Yang XH, Sladek TL, Liu X, et al. Reconstitution of caspase-3 49576. sensitizes MVF-7 breast cancer cells to doxorubicin- and etoposide- 38. Kowluru A, Tannous M, Chen HQ. Localization and characteriza- induced apoptosis. Cancer Res. 2001;60:348–354. tion of the mitochondrial isoform of the nucleoside diphosphate 46. Wu W, Lee WL, Wu YY, et al. Expression of constitutively active kinase in the pancreatic beta cell: evidence for its complexation phosphatidylinositol 3-kinase inhibits activation of caspase-3 and with mitochondrial succinyl-CoA synthetase. Arch Biochem Bio- apoptosis of cardiac muscle cells. J Biol Chem. 2000;275:40113– phys. 2002;398:160–169. 40119. 39. Romeo G, Liu WH, Asnaghi V, et al. Activation of nuclear factor- kappaB induced by diabetes and high glucose regulates a proapop- 47. Nishikawa T, Edelstein D, Du XL, et al. Normalizing mitochondrial totic program in retinal pericytes. Diabetes. 2002;51:2241–2248. superoxide production blocks three pathways of hyperglycaemic 40. Matsura T, Kai M, Fujii Y, et al. Hydrogen peroxide-induced apo- damage. Nature. 2000;404:787–790. ptosis in HL-60 cells requires caspase-3 activation. Free Radic Res. 48. Craven PA, Phillips SL, Melhem MF, et al. Overexpression of 1999;30:73–83. manganese superoxide dismutase suppresses increases in collagen 41. Hancock JT, Desikan R, Neill SJ. Does the redox status of accumulation induced by culture of mesangial cells in high-media cytochrome C act as a fail-safe mechanism in the regulation of glucose. Metabolism. 2001;50:1043–1048.

Downloaded from iovs.arvojournals.org on 10/01/2021