Effect of different amino acidic pretreatments that protect the kidney against papillary necrosis induced by bromoethylamine on differential distribution of renal nonprotein sulfhydryls

Content of nonprotein sulfhydryls (NPSH) was found to be higher in rat renal cortex than in external medulla and papilla. Administration of bromoethylamine (BEA), at a dose that produces extensive papillary necrosis and minor effects in the other renal segments, induced a significant reduction in NPSH levels of renal cortex and external medulla, with no changes in the papilla. Treatment with N-acetyl-L-cysteine (NAC) elicited an increase in papillary NPSH and a decrease in the cortex, with opposite changes being observed with an amino acid mixture of glutamine, glycine, and cystine (AM). Similar results were found in animals pretreated with NAC or AM prior to BEA intoxication. These pretreatments protect the cortex, external medulla, and papilla from the necrosis induced by BEA. It is suggested that protection of BEA-induced renal necrosis by NAC or AM pretreatments might be due to different mechanisms, with NPSH playing direct or indirect roles, respectively.     

Early changes in renal function following chemically induced nephropathy

The functional changes in the rat kidney 24 h after administration of 2-bromoethanamine hydrobromide (BEA) have been extensively described. There is, however, little information regarding earlier alterations. The present study was designed to measure early changes in renal function in order to clarify further pathomechanisms of the BEA-induced lesion. Experiments were performed in two groups of Wistar rats with different infusion rates during the first 3 h following injection of 100 mg/kg BW BEA compared to sham-injected rats. Analysis included measuring urine flow, osmolality, urea, sodium and potassium as well as inulin and para-aminohippuric acid clearance. Our studies show a tubular as well as a glomerular involvement in BEA-induced nephropathy. A significantly higher urine flow occurred already in the first 30 min following injection of BEA. Urine osmolality began to decrease after 90 min, Na excretion was elevated at 3 h, K excretion was not significantly different from the control group, urea excretion was increased after 30 min. Contrary to other studies we found a continuously decreasing glomerular filtration rate and PAH clearance during the first 3 h. Our results suggest an early effect of BEA on tubular function (increasing sodium excretion), papillary concentration capacity (increasing urine flow combined with decreasing osmolality) and glomerular function (decreasing glomerular filtration rate).     

Renal function studies in an experimental model of papillary necrosis in the rat

Twenty four hours after i.v. injection of bromoethylamine-hydrobromide (BEA) in rats, a uniform papillary necrosis is observed. The present study investigates the renal functional and the papillary haemodynamics in response to acute volume expansion (12% of body weight) in this model. Renal function studies were performed in hydropenic and volume expanded sham- or BEA-injected rats. In hydropenic normal animals a GFR of 1.97 +/- 0.14 ml/min, an urinary osmolarity (UOsm) of 1 011 +/- 94.5 mOsm/kg and a fractional sodium excretion (FENa) of 0.18 +/- 0.026% were obtained. In contrast, BEA-treated hydropenic animals showed a lower GFR (1.16 +/- 0.14 ml/min), UOsm (469 +/- 30.31 mOsm/kg) and a higher FENa (0.37 +/- 0.06%). In volume expansion a similar UOsm and FENa were obtained in both groups. The papillary plasma flow (PPF) was measured in each of the experimental groups by the albumin accumulation technique. The mean value in hydropenic normal animals was 50.65 +/- 2.12 m 100 g-1 min-1 and increased to 66.02 +/- 2.00 ml 100 g-1 min-1 after volume expansion (P less than 0.001). In BEA rats the PPF was 58.86 +/- 2.33 ml 100 g-1 min-1 in hydropenia (P less than 0.01 vs. control animals) and remained unchanged after volume expansion. Thus, during hydropenia, BEA-induced papillary necrosis results with a salt wasting state and an urinary concentration defect. After volume expansion no disturbance in sodium excretion capacity was observed. These results are compatible with the nephron-heterogeneity concept in the regulation of sodium excretion. The histological lesions cannot be explained by a decreased renal papillary plasma flow.     

Free radical scavenger depletion in post-ischemic reperfusion brain damage

In the present study the influence of pretreatment with various GSH depletors such as buthionine sulfoximine (BSO) and diethylmaleate (DEM) was investigated in rats following cerebral postischemic reperfusion. Moreover, the effect of diethyldithiocarbamic acid (DDC), inhibitor of endogenous Cu,Zn-SOD, was evaluated. A significant depletion (40% of control value) of GSH levels was observed 24 h after DEM administration; after 48 h the value reached control levels. BSO showed maximal GSH depletion (59%) 24 h after administration and it was constant for almost 48 h. DDC administration caused a marked decrease (60%) of Cu,Zn-SOD activity 4 h after the injection and induced a marked decrease in percentage of survival with respect to control (untreated, ischemic) rats, when administered 4 h before ischemia. BSO and DEM prolonged the survival time of animals when administered 24 h before ischemia. This last paradoxical effect is unclear at present, but it might be due to an influence on glutamate cascade.     

Renal and hepatic glutathione pool modifications in response to depletion treatments

In this study we examined the response of the renal and hepatic glutathione (GSH) pool in rats to drastic GSH depletion treatments. For this purpose, we used a protein-free diet, starvation, and the injection of varying doses of diethyl maleate as depleting agents. We analysed GSH levels in both kidney and liver tissue homogenates after rats were fed a protein-free diet for 2 or 7 days or starved for 1, 2, or 3 days, as well as after diethyl maleate administration in a single maximal dose or in varying doses. The results indicated that the liver GSH pool was always more labile than the kidney GSH pool. Moreover, kidney GSH levels were almost unchanged after 7 days on a protein-free diet or after 2 days of starvation, while liver showed significant changes in GSH levels. When we analysed the repletion rate, kidney had higher kinetic parameters (k = 0.148 h-1) than liver (0.097 h-1). We conclude that efficient mechanisms of maintaining GSH levels exist in the kidney and these may serve to avoid GSH diminution and hence preserve renal function during states of GSH depletion.     

N-Acetyl-l-cysteine abolishes the bromoethylamine-induced choline incorporation into renal papillary tissue

The role of regenerative processes in the protective effect of N-acetyl-L-cysteine (NAC) against bromoethylamine-induced renal papillary necrosis was assessed in rats given bromoethylamine (BEA) (1.2 mmol/kg), N-acetylcysteine (6 mmol/kg), or N-acetylcysteine plus BEA. Renal papillary slices were dissected after 15 hours of treatment, and ¹⁴C-choline incorporation into total phospholipid, lysophosphatidylcholine, sphingomyelin, and phosphatidylcholine was measured. Bromoethylamine elicited an increase in the incorporation of ¹⁴C-choline into choline-containing phospholipid, an effect that was abolished when N-acetylcysteine was administered prior to bromoethylamine. These studies indicate that the defensive mechanism of N-acetylcysteine against bromoethylamine-induced renal papillary necrosis is not related to regenerative processes and that N-acetylcysteine abolishes the bromoethylamine-induced choline incorporation into papillary phospholipid. © 1996 John Wiley & Sons, Inc.     

Renal necrosis and the involvement of a single enzyme of the de novo pathway for the biosynthesis of platelet-activating factor in the rat kidney inner medulla

2-Bromoethylamine hydrobromide (BEA), when administered to rats, induces a highly specific papillary necrosis associated with the inner medulla. PAF levels in the blood were lowered by 50% and of the three enzymes that comprise the de novo route for PAF in the cortex/medulla, only the cholinephosphotransferase activity in the inner medulla microsomes was reduced (33%) by the BEA treatment. Moreover, BEA did not affect phosphatidylcholine synthesis in either the cortex or inner medulla. Our studies indicate that the de novo pathway for PAF synthesis in the renal inner medulla is responsible for the secretion of newly formed PAF into the blood stream and that a single enzyme in the de novo route accounts for the decreased rate of PAF synthesis during the development of renal necrosis.     

Renal ischemia induces an increase in nitric oxide levels from tissue stores

Tissue nitric oxide (NO) levels increase dramatically during ischemia, an effect that has been shown to be partially independent from NO synthases. Because NO is stored in tissues as S-nitrosothiols and because these compounds could release NO during ischemia, we evaluated the effects of buthionine sulfoximine (BSO; an intracellular glutathione depletor), light stimulation (which releases NO, decomposing S-nitrosothiols), and N-acetyl-L-cysteine (a sulfhydryl group donor that repletes S-nitrosothiols stores) on the changes in outer medullary NO concentration produced during 45 min of renal artery occlusion in anesthetized rats. Renal ischemia increased renal tissue NO concentration (+223%), and this effect was maintained along 45 min of renal arterial blockade. After reperfusion, NO concentration fell below preischemic values and remained stable for the remainder of the experiment. Pretreatment with 10 mg/kg nitro-L-arginine methyl ester (L-NAME) decreased significantly basal NO concentration before ischemia, but it did not modify the rise in NO levels observed during ischemia. In rats pretreated with 4 mmol/kg BSO and L-NAME, ischemia was followed by a transient increase in renal NO concentration that fell to preischemic values 20 min before reperfusion. A similar response was observed when the kidney was illuminated 40 min before the ischemia. The coadministration of 10 mg/kg iv N-acetyl-L-cysteine with BSO + L-NAME restored the increase in NO levels observed during renal ischemia and prevented the depletion of renal thiol groups. These results demonstrate that the increase in renal NO concentration observed during ischemia originates from thiol-dependent tissue stores.     

Conjugated linoleic acid isomers have differential effects on triglyceride secretion in Hep G2 cells

Treatment for 2 h with 200 microM cadmium chloride, followed by recovery, caused apoptosis and induced heat-shock protein 70 (HSP70) expression in U-937 promonocytic cells. However, pre-incubation with the GSH depleting agent L-buthionine-[S,R]-sulfoximine (BSO, 1 mM for 24 h) caused necrosis instead of apoptosis and failed to induce HSP70 expression. This failure was a consequence of necrosis instead of GSH depletion, since BSO allowed or even potentiated HSP70 induction when used in combination with heat shock (2 h at 42.5 degrees C) or with 50 microM cadmium, which caused apoptosis. The administration of N-acetyl-L-cysteine (NAC) at the beginning of recovery after BSO/200 microM cadmium treatment prevented the execution of necrosis and restored the execution of apoptosis, but did not restore HSP70 induction, indicating that the inhibition by BSO of HSP70 expression is an early regulated event. This contrasted with the capacity of NAC to prevent the alterations caused by BSO/200 microM cadmium in other proteins, namely the suppression of Bax expression and the increase in Bcl-2 and HSP-60 expression. Finally, it was observed that treatment with 200 microM cadmium rapidly increased the HSP70 mRNA level and stimulated heat-shock factor 1 (HSF1) trimerization and binding, and that these effects were prevented by pre-incubation with BSO. Taken together, these results indicate that the stress response is compatible with apoptosis but not with necrosis in cadmium-treated promonocytic cells. The suppression of the stress response is specifically due to the early inhibition of HSF1 activation.     

Protection by ebselen against cisplatin-induced nephrotoxicity: Antioxidant system

This study was designed to investigate the cisplatin-induced alteration in renal antioxidant system and the nephroprotection with ebselen. Male Wistar rats were injected with (1) vehicle control; (2) cisplatin; (3) ebselen; and (4) cisplatin plus ebselen. Rats were sacrificed three days post-treatment and plasma as well as kidney were isolated and analyzed. Plasma creatinine increased 598% following cisplatin administration alone which decreased by 158% with ebselen pretreatment. Cisplatin-treated rats showed a depletion of renal glutathione (GSH) levels (52% of control), while cisplatin plus ebselen injected rats had GSH values close to the controls. Antioxidant enzymes superoxide dismutase (SOD), catalase (CAT) and glutathione peroxidase (GSH-Px) activities decreased 38, 75 and 62% of control, respectively, and malondialdehyde (MDA) levels increased 174% of control following cisplatin administration, which were restored to control levels after ebselen treatment. The renal platinum level did not significantly change with ebselen pretreatment. This study suggests that the protection offered by ebselen against cisplatin-induced nephrotoxicity is partly related to the sparing of antioxidant system.     

Relationships between alterations in glutathione metabolism and the disposition of inorganic mercury in rats: effects of biliary ligation and chemically induced modulation of glutathione status

Influences of biliary ligation and systemic depletion of glutathione (GSH) or modulation of GSH status on the disposition of a low, non-nephrotoxic i.v. dose of inorganic mercury were evaluated in rats in the present study. Renal and hepatic disposition, and the urinary and fecal excretion, of inorganic mercury were assessed 24 h after the injection of a 0.5-micromol/kg dose of mercuric chloride in control rats and rats pretreated with acivicin (two 10-mg/kg i.p. doses in 2 ml/kg normal saline, 90 min apart, 60 min before mercuric chloride), buthionine sulfoximine (BSO; 2 mmol/kg i.v. in 4 ml/kg normal saline, 2 h before mercuric chloride) or diethylmaleate (DEM; 3.37 mmol/kg i.p. in 2 ml/kg corn oil, 2 h before mercuric chloride) that either underwent or did not undergo acute biliary ligation prior to the injection of mercury. Among the groups that did not undergo biliary ligation, the pretreatments used to alter GSH status systemically had varying effects on the disposition of inorganic mercury in the kidneys, liver, and blood. Biliary ligation caused the net renal accumulation of mercury to decrease under all pretreatment conditions. By contrast, biliary ligation caused significant increases in the hepatic burden of mercury in all pretreatment groups except in theacivicin-pretreated group. Blood levels of mercury also increased as a result of biliary ligation, regardless of the type of pretreatment used. The present findings indicate that biliary ligation combined with methods used to modulate GSH status systemically have additive effects with respect to causing reductions in the net renal accumulation of mercury. Additionally, the findings indicate that at least some fraction of the renal accumulation of inorganic mercury is linked mechanistically to the hepato-biliary system.     

Induction of hepatic and renal ornithine decarboxylase by cobalt and other metal ions in rats.

We previously showed that Cd2+ is able to induce hepatic and renal ornithine decarboxylase (ODC). In addition to Cd2+, the administration of Co2+ and other metal ions such as Se2+, Zn2+ and Cr2+ produced a significant increase of hepatic and/or renal ODC activity. Of the metal ions used in this study, Co2+ produced the greatest increase of ODC activity. The maximum increases in hepatic and renal ODC activity, to respectively 70 and 14 times the control values in male rats, were observed 6 h after the administration of Co2+. A similar response was seen in the liver, but not in the kidney, of female rats. Thereafter, ODC activity gradually returned to control values in the liver, but it was profoundly decreased to 7% of the control value at 24 h in the kidney. The pretreatment of animals with either actinomycin D or cycloheximide almost completely blocked the Co2+-mediated increase of ODC activity. Co2+ complexed with either cysteine or glutathione (GSH) failed to induce ODC. Depletion of hepatic GSH content by treatment of rats with diethyl maleate greatly enhanced the inducing effect of Co2+ on ODC. The inhibitors of ODC, 1,3-diaminopropane and alpha-difluoromethylornithine, were able to inhibit the induction of the enzyme, without affecting the induction of haem oxygenase by Co2+. Methylglyoxal bis(guanylhydrazone), an inhibitor of S-adenosylmethionine decarboxylase, significantly inhibited the Co2+-mediated induction of both ODC and haem oxygenase. It is suggested that the inducing effects of Co2+ on ODC and haem oxygenase are brought about in a similar manner.     

Urinary markers of nephrotoxicity following administration of 2 bromoethanamine hydrobromide a comparison with hexachlorobutadiene

2 bromoethanamine hydrobromide (BEA) has been widely considered to be a target selective nephrotoxin that causes necrosis of the medulla in 24-48 h, but recent reports suggest that early cortical injury is also associated with this lesion. In order to assess the cortical effects of BEA (100 mg kg^-1 bw single ip injection), several urinary markers of renal injury were evaluated over a 7 day period in male Wistar Albino rats. Hexachlorobutadiene (HCBD 150 mg kg^-1 bw in peanut oil ip), a renal toxin which targets selectively for the proximal tubule, was used as a comparison. After BEA treatment, urinary levels of alanine aminopeptidase, gamma-glutamyl-transpeptidase, alkaline phosphatase and glucose increased transiently. Each of the proximal tubule marker enzymes peaked earlier following HCBD treatment and elevation of alanine aminopeptidase and gamma glutamyl transpeptidase was sustained for longer periods than for BEA. Following BEA treatment, lactate dehydrogenase rose prominently on day 1 followed by a return to control values on day 2 and a further rise on day 3 and remained high until the end of the study. BEA also increased the urinary excretion of total protein and albumin. After HCBD treatment, lactate dehydrogenase showed a transient elevation and glucose levels were slightly increased. Based on the present observations the changes induced by BEA administration on urinary markers of renal injury are different from those observed following HCBD treatment. These findings suggest that BEA toxicity also involves other parts of the kidney besides the papilla.     

Comparative Nephrotoxicity of Aspirin and Phenacetin Derivatives

Both aspirin and phenacetin derivatives were shown to be nephrotoxic when administered to rats as a single intravenous injection. Phenacetin derivatives tended to produce more severe renal damage and to be nephrotoxic in smaller doses than aspirin derivatives. With the exception of a single derivative, the renal lesions were confined to the proximal convoluted tubule, even after administration of compounds which under other conditions have induced renal papillary necrosis.     

Urinary markers of nephrotoxicity following administration of 2 bromoethanamine hydrobromide a comparison with hexachlorobutadiene

2 bromoethanamine hydrobromide (BEA) has been widely considered to be a target selective nephrotoxin that causes necrosis of the medulla in 24-48 h, but recent reports suggest that early cortical injury is also associated with this lesion. In order to assess the cortical effects of BEA (100 mg kg^-1 bw single ip injection), several urinary markers of renal injury were evaluated over a 7 day period in male Wistar Albino rats. Hexachlorobutadiene (HCBD 150 mg kg^-1 bw in peanut oil ip), a renal toxin which targets selectively for the proximal tubule, was used as a comparison. After BEA treatment, urinary levels of alanine aminopeptidase, gamma-glutamyl-transpeptidase, alkaline phosphatase and glucose increased transiently. Each of the proximal tubule marker enzymes peaked earlier following HCBD treatment and elevation of alanine aminopeptidase and gamma glutamyl transpeptidase was sustained for longer periods than for BEA. Following BEA treatment, lactate dehydrogenase rose prominently on day 1 followed by a return to control values on day 2 and a further rise on day 3 and remained high until the end of the study. BEA also increased the urinary excretion of total protein and albumin. After HCBD treatment, lactate dehydrogenase showed a transient elevation and glucose levels were slightly increased. Based on the present observations the changes induced by BEA administration on urinary markers of renal injury are different from those observed following HCBD treatment. These findings suggest that BEA toxicity also involves other parts of the kidney besides the papilla.     

Effect of acetaminophen on hepatic content and biliary efflux of glutathione disulfide in mice

The increased expiration of ethane and pentane by mice treated with hepatotoxic doses of acetaminophen suggests the possibility of oxidant mechanisms associated with the necrosis. However, studies in rats are not consistent with oxidant stress mechanisms causing the damage, because acetaminophen given to rats does not increase GSSG efflux, a sensitive index of intrahepatic oxidant stress. To compare the extent of oxidant stress generated by acetaminophen in mice versus rats, hepatic content and biliary efflux of GSSG and GSH in mice have been examined. Bile was collected from anesthetized male ICR mice before and after intraperitoneal administration of acetaminophen (325 mg/kg, 2.15 mmol/kg), t-butyl hydroperoxide (TBHP) (1.5 mmol/kg), diethyl maleate (400 mg/kg, 2.33 mmol/kg, in corn oil) or saline (control) and GSH and GSSG were measured by the enzymatic recycling method of Tietze. An increase in biliary GSSG efflux was produced by t-butyl hydroperoxide, but not by the other agents. Biliary GSH/GSSG ratios decreased in acetaminophen-treated animals, presumably reflecting the marked depletion of hepatic GSH, since a similar decrease was observed with non-hepatotoxic doses of diethyl maleate. The failure of acetaminophen to increase the hepatic content or biliary efflux of GSSG in ICR mice is not consistent with the view that oxidant stress mechanisms cause the damage, despite the increases in alkanes expired after acetaminophen administration in this specific animal model.     

Depletion of renal cortical glutathione and nephrotoxicity by cephaloridine,cephalothin and gentamicin in male sprague-dawley rats

Cephaloridine and gentamicin are selectively accumulated in renal cortex and produce necrosis of proximal tubular cells. However, the mechanisms responsible for renal cortical accumulation of these two antibiotics are quite different; therefore the early pathogenetic processes may not be the same. In the present study, effects of two cephalosporins (cephaloridine and cephalothin) and an aminoglycoside (gentamicin) on rat renal cortical glutathione were determined. Cephaloridine produced a dose-related depletion of renal cortical glutathione one hour following a single administration of the drug. In contrast, cephalothin in equivalent doses did not reduce renal cortical glutathione. Gentamicin had no effect on renal cortical glutathione, even when an acutely lethal dose (1000 mg/kg) was used. Pretreatment of rats with diethyl maleate (0.4 ml/kg) markedly depleted renal cortical glutathione and this pretreatment also potentiated cephaloridine nephrotoxicity. These results suggest that glutathione may play a protective role against cephaloridine but not gentamicin nephrotoxicity.     

Elevation of renal glutathione enhances ischemic injury.

In a previous study, we tested the hypothesis that an elevated level of renal glutathione (GSH) would protect the kidney from ischemic injury. However, prior elevation of GSH with GSH monoethylester enhanced then injury induced by 35 min of ischemia and blood reflow [Scaduto RC Jr, Gattone VH, Grotyohann LW, et al; Effect of an altered glutathione content on renal ischemic injury. Am J Physiol 1988;255:F911-F921]. Additionally, GSH monoethylester produced morphologic alterations in the absence of ischemia. Thus the greater ischemic injury observed after GSH ester pretreatment could have been due to a synergistic effect between the events caused by ischemia and the pretreatment. The present study was conducted to evaluate the utility of elevating renal GSH levels by administration of GSH. Administration of GSH (1 mmol/kg body weight) caused a 3-fold elevation of renal GSH levels and a 6-fold elevation of renal cysteine levels after 60 min without causing changes in renal morphology or GFR. After 35 min of renal artery occlusion and 90 min of blood reflow, animals pretreated with GSH had a much greater decline in GFR than untreated control animals. This enhancement of renal ischemic injury in GSH-treated animals was similar to that observed following administration of GSH monoethylester. We conclude that administration of GSH is the method of choice for elevation of renal GSH and that elevation of renal GSH leads to an enhanced ischemia-induced injury which is independent of the method employed to elevate renal GSH.     

Effect of glutathione depletion on urinary acidification in the rat

Glutathione (GSH) depletion by diethyl maleate (DEM) administration and its rapid repletion were associated with the development of a moderate acidosis in the rat. The acidosis observed after DEM treatment could be a consequence of an impairment of lactate metabolism. GSH-depleted rats also showed an increased urine pH and a higher bicarbonate fractional excretion compared with control rats. Renal bicarbonate excretion was magnified when blood bicarbonate levels were normalized by means of a bicarbonate infusion in GSH-depleted rats; however, the amount of bicarbonate excreted in the urine was a very small fraction (less than 5%) of the calculated filtered load. GSH-depleted rats failed to elevate the relation urine minus blood (U-B) pCO2 as compared with control rats when they were subjected to a high bicarbonate load to the distal portions of the nephron. All these data were consistent with a distal renal tubular acidosis due to GSH depletion which could participate in the maintenance of the systemic acidosis, although it is unlikely that it is the primary cause of the acidosis.     

Intracerebroventricular Administration of l-Buthionine Sulfoximine: A Method for Depleting Brain Glutathione

Sprague-Dawley rats (200-260 g) were anesthetized with chloral hydrate (400 mg/kg) and polyethylene cannulae were permanently implanted into the lateral ventricles. One or two days later, L-buthionine-[S,R]-sulfoximine (L-BSO), an apparently selective inhibitor of gamma-glutamylcysteine synthetase, was administered intracerebroventricularly through the cannulae. The brain content of glutathione (GSH) was determined by HPLC with electrochemical detection (gold/mercury electrode) using N-acetylcysteine as internal standard. A time-course study of the changes in the striatum following a single dose of L-BSO (3.2 mg) revealed a maximal depletion of GSH (-60%) approximately 48 h after the administration. The effects of various doses of L-BSO on GSH in the striatum, in the limbic region, and in the cortex were assessed at 24 h and 48 h after the administration. L-BSO (0.02-3.2 mg) produced dose-dependent reductions of GSH in all brain regions studied at both time intervals. In a long-term experiment L-BSO (3.2 mg) was administered every second day. After 4 days, i.e., after two injections, striatal GSH was reduced by approximately 70%. No further depletion of GSH was obtained by additional injections of L-BSO, but GSH was maintained at this low level for the 12 days studied. These results suggest that L-BSO, administered intracerebroventricularly, would serve as a useful tool for evaluation of the biological role of GSH in the CNS.