Regulation of-Glutamylcysteine Synthetase Subunit Gene Expression ...

Regulation of g-Glutamylcysteine Synthetase Subunit Gene Expression in Retinal Mu ¨ ller Cells by Oxidative Stress Shelly C. Lu,1 Yuzhou Bao,1 Zong–Zhi Huang,1 Vijay P. Sarthy,2 and Ram Kannan1 PURPOSE. To study regulation of g-glutamylcysteine synthetase (GCS) heavy and light subunit gene expression in Mu ¨ ller cells under conditions of oxidative stress. METHODS. Experiments were carried out with an SV40 transformed cell line (rMC-1) that exhibits the phenotype of rat retinal Mu ¨ ller cells. Endogenous glutathione levels were modified by treating cells with diethyl maleate (DEM), D,L-buthionine sulfoximine (BSO), or tert-butylhydroquinone (TBH). In other experiments, cells were grown in either high (28 mM) or normal (5.5 mM) glucose medium for 1 week to examine the effects of hyperglycemia. Cells were processed for reduced glutathione (GSH) measurement, RNA extraction, cell count, and, in some cases, lactate dehydrogenase activity. The steady state mRNA levels of GCS heavy and light subunits were measured by northern blot analysis using specific cDNA probes. Changes in mRNA levels were normalized to b-actin or 18S rRNA. RESULTS. Treatment with DEM for 30 minutes depleted cell GSH to 20% to 30% of the normal value. GSH content recovered completely 6 hours after returning to normal medium. BSO treatment for 12 hours followed by a medium change for 6 hours resulted in a cell GSH level that was 26% that of untreated cells. If cells were left in BSO for 18 hours, however, GSH levels were reduced to ,1%. Treatment with TBH for 12 hours led to a 77% increase in cellular GSH level. Treatment with DEM, TBH, or BSO for 18 hours led to a significant induction of the mRNA level of the GCS subunits, regardless of glucose concentration in the medium. Shorter BSO treatment exerted no effect. Prolonged hyperglycemia resulted in 30% lower GSH level, 55% lower GCS heavy subunit, and 30% lower GCS light subunit mRNA levels. CONCLUSIONS. Oxidative stress induced the gene expression of GCS heavy and light subunits in Mu ¨ ller cells. The effect of BSO on mRNA levels correlated with the degree of GSH depletion. Prolonged hyperglycemia lowered GCS subunit mRNA and GSH levels. (Invest Ophthalmol Vis Sci. 1999;40:1776 –1782)

T

he mammalian retina is particularly vulnerable to oxidative stress.1 A high polyunsaturated fatty acid level, high oxygen utilization, and continuous exposure to light and environmental toxins render the retina highly susceptible to oxidative damage. It has been recognized for some time that lipid peroxidation, production of free radicals, and associated events in retinal ischemia can be prevented by antioxidant administration.2– 6 There is good experimental evidence that glutathione, an important endogenous antioxidant, plays a major role in regulating oxidative damage in the retina. Reduced glutathione (GSH) is present in high concentrations in the

From the 1Division of Gastrointestinal and Liver Diseases, Department of Medicine, USC Liver Disease Research Center, University of Southern California School of Medicine, Los Angeles; and the 2Department of Ophthalmology, Northwestern University Medical School, Chicago, Illinois. Supported by NIH Grants EY11135, EY03523, and DK-45334 and by unrestricted funds from Research to Prevent Blindness, Inc. Submitted for publication November 9, 1998; revised March 8, 1999; accepted April 8, 1999. Proprietary interest category: N. Reprint requests: Ram Kannan, Division of Gastrointestinal and Liver Diseases, HMR 803A, Department of Medicine, USC School of Medicine, 2011 Zonal Avenue, Los Angeles, CA 90033.

1776

retina, and it has been shown that the retina has the capability to synthesize GSH from its amino acid precursors.7,8 There is good evidence that Mu ¨ ller cells, the major support cells in the retina, play an important role in regulating GSH levels in the retina.9 These cells are present in the retina of all vertebrate species and perform many of the functions carried out by astrocytes, oligodendrocytes, and ependymal cells in the central nervous system.9 Recent studies indicate that retinal Mu ¨ ller cells may actively participate in neuronal protection by providing GSH to neurons, although the mechanism underlying this process is poorly understood. Specifically, GSH has been shown to be predominantly confined to Mu ¨ ller cells, and a recent study suggests that GSH is transferred from Mu ¨ ller cells to neurons under ischemic conditions.10 Despite this important role, very little is known about the regulation of GSH biosynthesis in Mu ¨ ller cells. A major determinant of GSH biosynthesis is the activity of the rate-limiting enzyme g-glutamylcysteine synthetase (GCS).11,12 GCS is composed of heavy (GCS-HS; Mr ;73,000) and light (GCS-LS; Mr ;30,000) subunits that are encoded by different genes in rat and human genomes.13–16 The enzyme may be dissociated under nondenaturing conditions by treatment with dithiothreitol.17 The heavy subunit obtained after this treatment exhibits all the catalytic activity of the isolated Investigative Ophthalmology & Visual Science, July 1999, Vol. 40, No. 8 Copyright © Association for Research in Vision and Ophthalmology

IOVS, July 1999, Vol. 40, No. 8

g-Glutamylcysteine Synthetase Expression in Mu ¨ ller Cells

enzyme and feedback inhibition by GSH.17 The light subunit is enzymatically inactive but plays an important regulatory function by lowering the Km of GCS for glutamate and raising the Ki for GSH.14,18 Thus, the holoenzyme is catalytically more efficient and less subject to inhibition by GSH than the heavy subunit. Regulation of GCS has been a subject of intense study in many cell types.19 –32 However, very little is known about GCS gene regulation in neural cells. Transcriptional and posttranscriptional regulation of both subunits have been described. We and others have shown that the gene expression of both GCS subunits are upregulated under oxidative stress and treatment with diethyl maleate (DEM) and tert-butylhydroquinone (TBH).23,25,28,32 We have also shown that in hepatocytes, hormones such as insulin and glucocorticoids upregulated the expression of GCS-HS without affecting GCS-LS.25,29 In addition, during periods of rapid liver growth such as plating hepatocytes under low cell density and liver regeneration after partial hepatectomy, the gene expression of GCS-HS but not GCS-LS was upregulated.25,29,30 Thus, the two subunits of GCS appear to be differentially regulated. Recently, Urata and colleagues31 reported decreased GCS-HS gene expression in mouse endothelial cells after prolonged hyperglycemia. Whether the light subunit is similarly affected was not examined. In the present study, we examined the effect of oxidative stress and prolonged hyperglycemia on the gene expression of the heavy and light subunits of GCS. These conditions were chosen because of the propensity of the Mu ¨ ller cells to be under oxidative stress and the role of hyperglycemia in diabetic retinopathy. These present studies represent the initial characterization of the regulation of the two GCS subunits in this cell type and the first study to examine the effect of prolonged hyperglycemia on the GCS light subunit.

MATERIALS

AND

METHODS

Materials GSH, NADPH, 5,59-dithiobis (2-nitrobenzoic acid), diethyl maleate (DEM), D,L-buthionine sulfoximine (BSO), GSH reductase, fetal bovine serum (FBS), and TBH were purchased from Sigma Chemical (St. Louis, MO). Dulbecco’s minimal essential medium (DMEM) was purchased from Mediatech’s Cellgro (Tustin, CA). [32P]-dCTP (3000 Ci/mM) was purchased from New England Nuclear–Du Pont (Boston, MA). Messenger RNA isolation kit (Fast Track 2.0) was obtained from Invitrogen (Carlsbad, CA). All other reagents were of analytical grade and were obtained from commercial sources.

Culture and Treatment of Mu ¨ ller Cells These studies were carried out using an immortalized Mu ¨ ller cell line that was previously characterized.33 Mu ¨ ller cell cultures were passaged in DMEM (28 mM glucose) supplemented with FBS (10%), penicillin (100 U/ml), and streptomycin (100 mg/ml) in a humidified atmosphere of 5% CO2/95% air. Medium was changed every 2 to 3 days, and cells were grown to confluence using T-75 flasks. Cells at passages 11 to 22 were used in the current studies. To study the effects of DEM, BSO, and TBH, 10 3 106 Mu ¨ ller cells were plated per 100-mm 3 15-mm Falcon Primaria culture dish (Becton Dickinson, Lincoln Park, NJ) using cells

1777

grown in DMEM containing 10% FBS and 28 mM or 5.5 mM glucose for at least 7 days. To examine the effect of DEM, cells were treated with 0.2 mM DEM or vehicle for 20 minutes, followed by a medium change to DMEM for 6 hours. To examine the effect of BSO, cells were treated with 1 mM BSO or vehicle for 12 hours followed by medium change to DMEM for an additional 6 hours or BSO continuously for 18 hours. To examine the effect of TBH, cells were treated with 50 mM TBH or vehicle for 12 hours. At the end of the treatment, cells were processed for GSH measurement by the recycling method of Tietze.34 RNA extraction and lactate dehydrogenase measurement were performed as described elsewhere.35 The dosage and treatment duration of these agents are based on our previous findings using cultured rat hepatocytes.25 To study the effects of prolonged exposure to a high glucose concentration, Mu ¨ ller cells were grown in DMEM containing 10% FBS and 5.5 mM glucose (normal glucose) or 28 mM glucose (high glucose) for 7 days. To determine the effects of glucose on cell growth, 1.8 3 106 cells were plated per 100-mm 3 15-mm dish, and the number of cells and viability were determined daily using a hemocytometer. Viability was determined by 0.2% trypan blue exclusion as described previously.25 At the end of the seventh day, cells were processed for GSH determination, RNA extraction, and lactate dehydrogenase measurement.

Nucleic Acid Extraction Poly(A)1–RNA (mRNA) was isolated from Mu ¨ ller cells according to the method provided by Invitrogen (Carlsbad, CA) along with the messenger RNA isolation kit (Fast Track 2.0). RNA concentration was determined spectrophotometrically before use.

Northern Hybridization Analysis Northern hybridization analysis was performed on poly(A)1– RNA (3 mg) using standard procedures as described previously.25 The GCS-HS cDNA probe is comprised of a 390-bp fragment corresponding to nucleotides 79 to 468 of the published rat kidney GCS-HS sequence,13 and the GCS-LS cDNA probe is comprised of a 1.1-kb fragment corresponding to nucleotides 122 to 1232 of the published rat kidney GCS-LS sequence.14 Both were labeled with [32P]dCTP using a random-primer kit (Primer-It II Kit; Stratagene, La Jolla, CA). To ensure equal loading of RNA samples, the same membrane was rehybridized with 32P-labeled human b-actin or 18S cDNA probes (Clontech, Palo Alto, CA). Autoradiography and densitometry (Gel Documentation System, Scientific Technologies, Carlsbad, CA and NIH Image 1.60 software program) were used to quantitate relative RNA content. In the case of GCS-LS in which multiple mRNA species exist, all the bands were quantitated before being used for comparison. Results of northern blot analysis were normalized to b-actin or 18S RNA levels.

Statistical Analysis Mean values from duplicate plates were used for comparison, and the mean values of multiple experiments were compared by paired two-tailed Student’s t-test (two comparisons) or ANOVA followed by Fisher’s test (multiple comparisons). For cell GSH levels, actual values were compared. For changes in mRNA levels, ratios of GCS-HS or GCS-LS to b-actin or 18S densitometric values were compared. The criterion for signif-

1778

Lu et al.


TABLE 1. Effects of DEM, BSO, TBH, and Prolonged Hyperglycemia on GSH Level of Mu ¨ ller Cells Condition DEM BSO (1 mM for 12 h followed by medium change for 6 h) BSO (1 mM for 18 h) TBH High glucose

% of Control Cell GSH 132 6 17 26 6 7* 0.9 6 0.3* 177 6 11* 70 6 3*

Results are expressed as mean 6 SEM from 3 to 4 experiments. Mu ¨ ller cells were treated with DEM (0.2 mM for 20 minutes followed by medium change for 6 hours), BSO (1 mM for 12 hours, followed by medium change for an additional 6 hours or continuously for 18 hours), TBH (50 mM for 12 hours), or exposed to high glucose (28 mM) concentration for 7 days as described in the Methods section. Control cell GSH value for DEM, BSO, and TBH was 13.5 6 0.8 nmol/106 cells, and control cell GSH value for the high glucose experiments (cells grown on medium with normal glucose level) was 19.4 6 0.3 nmol/ 106 cells. These control cell GSH values were higher, because the glucose concentration in the medium was 5.5 mM, whereas the control for DEM, BSO, and TBH actually had 28 mM glucose in the medium. * P , 0.05 vs. control by ANOVA followed by Fisher’s test.

icance was P , 0.05. Results are shown as mean 6 SEM of n, the number of experiments.

RESULTS Effect of DEM, BSO, TBH, and Glucose Concentration on Cell GSH Level The glucose concentration in the normal medium used for culturing Mu ¨ ller cells is 28 mM glucose.33 In initial experiments, we examined the effects of DEM, BSO, and TBH on GSH levels in Mu ¨ ller cells using normal medium. Table 1 summarizes the results obtained. Treatment with DEM for 30 minutes depleted cell GSH to approximately 20% to 30% of normal (not shown), which recovered completely 6 hours after a medium change. BSO treatment for 12 hours followed by a medium change for 6 hours resulted in a cell GSH level that was 26% of baseline, whereas cell GSH was ,1% of baseline if BSO was left in for 18 hours. Despite this dramatic reduction in cell GSH, no significant cell loss was observed (data not shown). Treatment with TBH for 12 hours led to a 77% increase in cell GSH level. Because prolonged exposure to high glucose concentration (28 mM) is known to affect GCS-HS expression in mouse endothelial cells,31 we examined the effect of the same treatment in Mu ¨ ller cells. Prolonged exposure to a high glucose concentration led to a 30% decrease in cell GSH level. This did not result in any cell death (data not shown).

Effects of DEM, BSO, TBH, and Glucose Concentration on GCS Subunit Gene Expression Next, we examined mRNA levels in Mu ¨ ller cells treated with agents that altered GSH levels. Figure 1 shows that treatment of Mu ¨ ller cells with TBH and DEM led to a significant increase in the steady state GCS-HS mRNA level. However, with respect to BSO, the steady state mRNA level of GCS-HS increased only after 18 hours of continuous treatment. The difference between the two BSO treatments was in the magnitude of GSH depletion. Thus, increased GCS-HS expression occurred only

FIGURE 1. Effects of TBH, DEM, and BSO on the steady state mRNA level of GCS-HS. Poly(A)1–RNA (3 mg each lane) samples obtained from Mu ¨ ller cells treated with TBH (50 mM for 12 hours), DEM (0.2 mM for 20 minutes followed by medium change for 6 hours), BSO (BSO1 5 1 mM for 18 hours; BSO2 5 1 mM for 12 hours followed by medium change for an additional 6 hours), or vehicle control (CON) were analyzed by northern blot hybridization with GCS-HS cDNA probe as described in the Materials and Methods section. The same membranes were subsequently hybridized with b-actin or 18S cDNA probes. A representative northern blot from 3 to 4 separate experiments is shown.

when the cell GSH level was reduced to a very low level (,1%). However, no effect was seen when cell GSH level was reduced to 26% of control. Moreover, these treatments also induced comparable changes in the mRNA level of GCS-LS (Fig. 2). Prolonged exposure to high glucose was previously shown to reduce the GCS-HS gene transcriptional rate in mouse endothelial cells.31 Figure 3 shows that in Mu ¨ ller cells, prolonged exposure to high glucose also exhibited lower GCS-HS mRNA level (55% lower) compared with cells cultured in medium containing normal glucose concentration. GCS-LS mRNA level was also lower (30% lower), although to a lesser extent than in the case of the heavy subunit. When exposure to high glucose was limited to only 24 hours, no change was seen in GCS subunit expression (data not shown). Figure 4 shows the effect of glucose on cell growth over a 7-day period. There was no difference in cell growth up to

FIGURE 2. Effects of TBH, DEM, and BSO on the steady state mRNA level of GCS-LS. Poly(A)1-RNA (3 mg each lane) samples obtained from Mu ¨ ller cells treated with TBH (50 mM for 12 hours), DEM (0.2 mM for 20 minutes followed by medium change for 6 hours), BSO (BSO1 5 1 mM for 18 hours; BSO2 5 1 mM for 12 hours followed by medium change for an additional 6 hours), or vehicle control (CON) were analyzed by northern blot hybridization with GCS-LS cDNA probe as described in the Materials and Methods section. The same membranes were subsequently hybridized with b-actin or 18S cDNA probes. A representative northern blot from 3 to 4 separate experiments is shown.



FIGURE 3. Effect of glucose concentration on the steady state mRNA level of GCS subunits. Poly(A)1-RNA (3 mg each lane) samples obtained from Mu ¨ ller cells exposed to high (28 mM) or normal (NL, 5.5 mM) glucose for 7 days were analyzed by northern blot hybridization with GCS-HS, GCS-LS, and b-actin cDNA probes as described in the Materials and Methods section. The same membrane was sequentially hybridized with all three probes. A representative northern blot from 3 to 4 separate experiments is shown.

the fourth day after plating. However, cells grown in 28 mM glucose exhibited much higher rates of growth after the fourth day. To verify that effects of BSO (1 mM for 18 hours), DEM, and TBH on GCS subunit gene expression are independent of the glucose concentration, cells were treated with the same agents after being grown in medium containing 5.5 mM glucose for 7 days. Figure 5 shows that the same agents induced the expression of both GCS subunits to comparable levels.

FIGURE 4. Effect of glucose concentration on cell growth. 1.8 3 106 Mu ¨ ller cells were plated on 100-mm 3 15-mm dishes in medium containing either 28 mM (closed circle) or 5.5 mM (open square) glucose, and the number of cells was determined daily as described in the Materials and Methods section. Viability of the cells remained greater than 95% in both conditions throughout the 7-day period. Results represent mean 6 SD from n 5 3.

1779

FIGURE 5. Effects of TBH, DEM, and BSO on the steady state mRNA level of GCS subunits in cells grown on normal glucose concentration (5.5 mM). Poly(A)1-RNA (3 mg each lane) samples obtained from Mu ¨ ller cells treated with TBH (50 mM for 12 hours), DEM (0.2 mM for 20 minutes followed by medium change for 6 hours), BSO (BSO 5 1 mM for 18 hours), or vehicle control (CON) were analyzed by northern blot hybridization with GCS-HS cDNA probe as described in the Materials and Methods section. The same membrane was sequentially hybridized with GCS-LS and b-actin cDNA probes. A representative northern blot is shown.

Table 2 summarizes the effect of these various treatments on the mRNA levels of GCS-HS and GCS-LS.

DISCUSSION GSH is synthesized by all mammalian cells and plays a vital defensive role against oxidative stress.11,12 Major determinants of the rate of biosynthesis of GSH are the availability of cysteine, its sulfur amino acid precursor, and the activity of the rate-limiting enzyme GCS.11,12 Several conditions or treatments are known to alter GCS activity in mammalian tissues. A partial list consists of drug-resistant tumor cell lines,19,20 oxidative stress,21–25 antioxidants,26,27 formation of Michael reaction acceptors (containing an electrophilic electron-deficient center that is susceptible to nucleophilic attack) that yield GSH conjugates25,28 that can also induce oxidative stress, hormones,25,29 and rapid liver cell growth.25,29,30 In addition, cell GSH itself can alter GCS activity. 25,31 The above treatments caused an upregulation of GCS. On the other hand, prolonged hyperglycemia was reported to cause decreased transcription of GCS-HS and GSH level in mouse endothelial cells.31 Whether the light subunit of GCS is affected by prolonged hyperglycemia was not examined; other than mouse endothelial cells, the effect of glucose concentration on GCS expression has not been reported. The major physiological and pathologic stress that the retina is exposed to is oxidative stress, which is known to influence the expression of several key enzymes involved in GSH metabolism and other detoxifying enzymes. These enzymes include GSH-peroxidase, GSSG-reductase, g-glutamyltranspeptidase, GCS, superoxide dismutase, and catalase.40 Although it is known that the Mu ¨ ller cells are the major support cells in the retina and that they are one of the major sources of retinal GSH,9,10 no information is available on the regulation of GSH biosynthesis in these cells. The present study focused on

1780

Lu et al.


TABLE 2. Effects of DEM, BSO, TBH, and Prolonged Hyperglycemia on GCS Subunit mRNA Levels % of Control mRNA Level GCS-HS Condition DEM BSO (1 mM for 12 hours followed by medium change for 6 hours) BSO (1 mM for 18 hours) TBH High glucose

GCS-LS*

(Control Medium 5 28 mM Glucose) 262 6 20*

182 6 10*

90 6 6 204 6 33* 240 6 23*

113 6 15 199 6 12* 189 6 15*

GCS-HS

GCS-LS

(Control Medium 5 5.5 mM Glucose) 203 6 25*

213 6 22*

200 6 16* 275 6 16* 45 6 4*

229 6 40* 285 6 15* 71 6 2*

Results are expressed as mean 6 SEM from 3 to 4 experiments. Mu ¨ ller cells were grown in medium containing either 28 mM or 5.5 mM glucose for 7 days and treated with DEM (0.2 mM for 20 minutes followed by medium change for 6 hours), BSO (1 mM for 12 hours followed by medium change for an additional 6 hours or continuously for 18 hours), TBH (50 mM for 12 hours), or vehicle control and subsequently processed for northern blot analysis as described in the Materials and Methods section. * P , 0.05 vs its respective control by ANOVA followed by Fisher’s test.

one of the major determinants of GSH biosynthesis, namely the expression of GCS, in Mu ¨ ller cells. Treatment of Mu ¨ ller cells with agents known to induce oxidative stress such as TBH, DEM, and BSO also induced the expression of both subunits of GCS to a comparable extent. This is similar to the response of hepatocytes and occurred independent of glucose concentration in the medium. The effect of BSO is particularly noteworthy. No effect was observed until the cellular GSH level was profoundly depressed (,1%). Moreover, the effect of BSO was dependent on the magnitude of GSH depletion and was similar to that observed with hepatocytes, in which depletion to less than 10% was required to detect significant changes in GCS subunit expression.25 Whether this reflects the level of oxidative stress or the GSH level itself remains unclear and will require further study. GCS subunits appear also to be under differential regulation. Gipp et al.16 found no correlation between the steady state mRNA levels of the two GCS subunits. We and others have shown that oxidative stress and DEM upregulated both GCS subunits in a variety of cell types.23,25,32 On the other hand, only the heavy subunit was upregulated by hormones such as insulin and glucocorticoids and during periods of rapid liver growth.25,29,30 Our results suggested that in liver cells, there may be more GCS light subunit than heavy subunit because modulation of the heavy subunit alone resulted in changes in cell GCS activity and GSH level.29,30 We have not examined the effect of insulin and glucocorticoid on GCS regulation in Mu ¨ ller cells. Our data show that GCS subunit expression in Mu ¨ ller cells is significantly affected by the exogenous glucose concentration. After prolonged exposure to a high glucose concentration, the mRNA level of both the heavy and light subunits of GCS fell. Interestingly, the heavy subunit exhibited a more significant decrease than the light subunit (55% versus 30%). Under these conditions, the cell GSH only fell by 30%. This would suggest that in contrast to liver cells in which there may be more GCS light subunit than heavy subunit, these two subunits may be present in an equal amount in the Mu ¨ ller cell. Unfortunately, antibodies to the GCS-LS are not currently avail-

able; hence, the actual light chain levels could not be determined. The effect of prolonged hyperglycemia on cell growth deserves brief mention. Under this condition, Mu ¨ ller cells exhibited faster rates of cell growth after the fourth day of plating, and the expression of GCS-HS and GCS-LS were lower. Mouse endothelial cells also exhibited a similar magnitude of fall in the expression of GCS-HS, but whether glucose concentration affected the growth was not mentioned.31 In contrast, we found that when hepatocytes are under periods of rapid growth, the expression of GCS-HS is induced while that of the GCS-LS is unchanged.29,30 Whether GCS expression is also induced during periods of rapid growth in other types of cells is unclear. Our speculation is that in the case of prolonged hyperglycemia, it is the glucose effect rather than a difference in cell growth that predominated in the overall GCS expression. Although our findings on GCS expression in prolonged hyperglycemia may be of general interest with respect to diabetes, one must exercise caution in extrapolating these findings to diabetic retinopathy. The molecular mechanisms by which hyperglycemia induces retinopathy are still poorly understood. Diabetes is associated with increased production of free radicals and lipid peroxides in the retina.36,37 Increased oxidative stress and decreased antioxidant defense systems have been implicated.38 – 41 Some of the complications of diabetic retinopathy were prevented by treatment with antioxidants.39 One study reported no change in the retinal GCS activity.40 A plausible explanation for this observation is that the stimulatory effect of oxidative stress, which induces GCS, may be nullified by hyperglycemia, which has the opposite effect on GCS level. Thus, because oxidative stress is known to induce GCS, the lack of any change in GCS activity in the diabetic retina is probably due to hyperglycemia. Furthermore, our study suggests that this impaired response of GCS may also contribute to the decrease in retinal GSH observed in the diabetic retina. In regard to the molecular mechanism(s) of oxidative stress–induced changes in GCS subunit expression, several studies have implicated AP-1 and ARE elements present in the



59-flanking regions of the human GCS-HS and GCS-LS.24,26,32,42 The mechanism of glucose modulation of GCS remains unclear. Although prolonged hyperglycemia can induce oxidative stress via the formation of glycated products,41 oxidative stress is clearly not the mechanism. Additional work will be required to elucidate the molecular mechanism of glucose effect on GCS gene expression. In summary, we have characterized, for the first time, the regulation of GCS subunit gene expression in Mu ¨ ller cells. Oxidative stress was found to induce the expression of both GCS subunits, a response similar to that of other cell types. Prolonged hyperglycemia, however, resulted in the decreased expression of both subunits and also decreased cell GSH levels. These studies constitute an important first step in our understanding of the role of Mu ¨ ller cells in GSH homeostasis in the retina.

Acknowledgments

17.

18.

19.

20.

21.

22.

Mu ¨ ller cells were cultured and cared for by the Cell Culture Core of the USC Liver Disease Research Center (DK48522).

References 1. Handelman GJ, Dratz DJ. The role of antioxidants in the retina and retinal pigment epithelium and the nature of prooxidant induced damage. Adv Free Radic Biol Med. 1986;2:1– 89. 2. Winkler BS, Giblin FJ. Glutathione oxidation in retina: effects on biochemical and electrical activities. Exp Eye Res. 1983;36:287– 297. 3. Sternberg P, Davidson PC, Jones DP, Hagen TM, Reed RL, Drews– Botsch C. Protection of retinal pigment epithelium from oxidative injury by glutathione and precursors. Invest Ophthalmol Vis Sci. 1993;3 4:3661–3668. 4. Organisciak DT, Wang HM, Li ZY, Tso MOM. The protective effect of ascorbate in retinal light damage of rats. Invest Ophthalmol Vis Sci. 1985;26:1580 –1588. 5. Hayes KC. Retinal degeneration in monkeys induced by deficiencies of vitamin E. Invest Ophthalmol Vis Sci. 1974;13:499 –510. 6. Stone WL, Henderson RA, Howard GH Jr, Hollis AL, Payne PH, Scott RL. The role of antioxidant nutrients in preventing hyperbaric oxygen damage to the retina. Free Radic Biol Med. 1989;6: 505–512. 7. Kannan R, Tang D, Mackic JB, Zlokovic BV, Fernandez–Checa JC. A simple technique to determine glutathione (GSH) levels in ocular tissues as GSH bimane adduct: application to normal and galactosemic guinea-pigs. Exp Eye Res. 1992;56:45–50. 8. Davidson C, Sternberg P, Jones DP, Reed RL. Synthesis and transport of glutathione by cultured human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci. 1994;35:2843–2849. 9. Newman E, Reichenbach A. The Mu ¨ ller cell: a functional element of the retina. Trends Neurosci. 1996;19:307–312. 10. Schuette M, Werner P. Redistribution of glutathione in the ischemic rat retina. Neurosci Lett. 1998;246:53–56. 11. Meister A, Anderson ME. Glutathione. Annu Rev Biochem. 1983; 52:711–760. 12. Ferna´ndez–Checa J, Lu SC, Ookhtens M, et al. The regulation of hepatic glutathione. In: Tavoloni N, Berk PD, eds. Hepatic Anion Transport and Bile Secretion: Physiology and Pathophysiology. New York, NY: Marcel Dekker; 1992:363–395. 13. Yan N, Meister A. Amino acid sequence of rat kidney g-glutamylcysteine synthetase. J Biol Chem. 1990;265:1588 –1593. 14. Huang C, Anderson ME, Meister A. Amino acid sequence and function of the light subunit of rat kidney g-glutamylcysteine synthetase. J Biol Chem. 1993;268:20578 –20583. 15. Gipp JJ, Chang C, Mulcahy RT. Cloning and nucleotide sequence of a full-length cDNA for human liver g-glutamylcysteine synthetase. Biochem Biophys Res Commun. 1992;185:29 –35. 16. Gipp JJ, Bailey HH, Mulcahy RT. Cloning and sequence of the cDNA for the light subunit of human liver g-glutamylcysteine

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

1781

synthetase and relative mRNA levels for heavy and light subunits in human normal tissues. Biochem Biophys Res Commun. 1995;206: 584 –589. Seelig GF, Simondsen RP, Meister A. Reversible dissociation of g-glutamylcysteine synthetase into two subunits. J Biol Chem. 1984;259:9345–9347. Huang C, Chang L, Anderson ME, Meister A. Catalytic and regulatory properties of the heavy subunit of rat kidney g-glutamylcysteine synthetase. J Biol Chem. 1993;268:19675–19680. Mulcahy RT, Untawale S, Gipp JJ. Transcriptional up-regulation of g-glutamylcysteine synthetase gene expression in melphalan-resistant human prostate carcinoma cells. Mol Pharmacol. 1994;46: 909 –914. Ishikawa T, Bao JJ, Yamane Y, et al. Coordinated induction of MRP/GS-X pump and g-glutamylcysteine synthetase by heavy metals in human leukemia cells. J Biol Chem. 1996;271:14981– 14988. Woods JS, Davis HA, Baer RP. Enhancement of g-glutamylcysteine synthetase mRNA in rat kidney by methyl mercury. Arch Biochem Biophys. 1992;296:350 –353. Rahman I, Bel A, Mulier B, Lawson MF, Harrison DJ, Macnee W, Smith CA. Transcriptional regulation of g-glutamylcysteine synthetase heavy subunit by oxidants in human alveolar epithelial cells. Biochem Biophys Res Commun. 1996;229:832– 837. Tian L, Shi MM, Forman HJ. Increased transcription of the regulatory subunit of g-glutamylcysteine synthetase in rat lung epithelial L2 cells exposed to oxidative stress or glutathione depletion. Arch Biochem Biophys. 1997;342:126 –133. Morales A, Garcı´a–Ruiz C, Miranda M, et al. Tumor necrosis factor increases hepatocellular glutathione by transcriptional regulation of the heavy subunit chain of g-glutamylcysteine synthetase. J Biol Chem. 1997;272:30371–30379. Cai J, Huang Z, Lu SC. Differential regulation of g-glutamylcysteine synthetase heavy and light subunit gene expression. Biochem J. 1997;326:167–172. Mulcahy RT, Wartman MA, Bailey HH, Gipp JJ. Constitutive and b-naphthoflavone-induced expression of the human g-glutamylcysteine synthetase heavy subunit gene is regulated by a distal antioxidant response element/TRE sequence. J Biol Chem. 1997;272: 7445–7454. Borroz KI, Buetler TM, Eaton DL. Modulation of g-glutamylcysteine synthetase large subunit mRNA expression by butylated hydroxyanisole. Toxicol Appl Pharmacol. 1994;126:150 –155. Sekhar KR, Long M, Long J, Xu ZQ, Summar ML, Freeman ML. Alteration of transcriptional and post-transcriptional expression of g-glutamylcysteine synthetase by diethyl maleate. Radiat Res. 1997;147:592–297. Cai J, Sun W, Lu SC. Hormonal and cell density regulation of hepatic g-glutamylcysteine synthetase gene expression. Mol Pharmacol. 1995;48:212–218. Huang Z, Li H, Cai J, Kuhlenkamp J, Kaplowitz N, Lu SC. Changes in glutathione homeostasis during liver regeneration in the rat. Hepatology. 1998;27:147–153. Urata Y, Yamamoto H, Goto S, et al. Long exposure to high glucose concentration impairs the responsive expression of g-glutamylcysteine synthetase by interleukin-1a and tumor necrosis factor-b in mouse endothelial cells. J Biol Chem. 1996; 271:15146 –15152. Galloway DC, Blake DG, Shepherd AG, McLellan LI. Regulation of human g-glutamylcysteine synthetase: co-ordinate induction of the catalytic and regulatory subunits in HepG2 cells. Biochem J. 1997; 328:99 –104. Sarthy VP, Brodjian SJ, Dutt K, Kennedy B, French RP, Crabb JW. Establishment and characterization of a retinal Mu ¨ ller cell line. Invest Ophthalmol Vis Sci. 1998;39:212–216. Tietze F. Enzymic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues. Anal Biochem. 1969;27:502– 522. Runnegar M, Kong S, Zhong Y, Lu SC. Inhibition of GSH synthesis by cyanobacterial alkaloid cylindrospermopsin in cultured rat hepatocytes. Biochem Pharmacol. 1995;49:219 –225.

1782

Lu et al.

36. Murata T, Nishida T, Eto S, Mukai N. Lipid peroxidation in diabetic rat retina. Metab Pediatr Ophthalmol. 1981;5:83– 87. 37. Armstrong D, Al–Awadi F. Lipid peroxidation and retinopathy in streptozotocin-induced diabetes. Free Radic Biol Med. 1991;11: 433– 436. 38. Kern, TS, Kowluru RA, Engerman RL. Abnormalities of retinal metabolism in diabetes or galactosemia: ATPase and glutathione. Invest Ophthalmol Vis Sci. 1994;35:2962–2967. 39. Kowluru RA, Kern TS, Engerman RL, Armstrong D. Abnormalities of retinal metabolism in diabetes or experimental galactosemia, III: effects of antioxidants. Diabetes. 1996;45:1233–1237.

IOVS, July 1999, Vol. 40, No. 8 40. Kowluru RA, Kern TS, Engerman RL. Abnormalities of retinal metabolism in diabetes or experimental galactosemia, IV: antioxidant defense system. Free Radic Biol Med. 1997;22:587–592. 41. Paget C, Lecomte M, Ruggiero D, Wiernsperger N, Lagarde M. Modification of enzymatic antioxidants in retinal microvascular cells by glucose or advanced glycation end products. Free Radic Biol Med. 1998;25:121–129. 42. Moinova H, Mulcahy RT. An electrophile responsive element (EpRE) regulates b-naphthoflavone induction of the human g-glutamylcysteine synthetase regulatory subunit gene. J Biol Chem. 1998;273:14683–14689.