|Molecular Vision 1999;
Received 20 July 1999 | Accepted 21 December 1999 | Published 22 December 1999
Modelling cortical cataractogenesis XXIV: Uptake by the lens of glutathione injected into the rat
P. Jill Stewart-DeHaan, Tomasz Dzialoszynski,
John R. Trevithick
Department of Biochemistry, Faculty of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada, N6A 5C1
Correspondence to: J. R. Trevithick, PhD, Department of Biochemistry, Faculty of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada, N6A 5C1; Phone: (519) 661-3063; FAX: (519) 661-3175; email: firstname.lastname@example.org
Purpose: Work of several groups including ours has shown that injection of glutathione may help to prevent the formation of cataract in the rat lens both in vitro and in vivo. These experiments were initiated to investigate the mechanism by which injected glutathione reaches the lens in vivo. The route is uncertain, but might involve either aqueous or vitreous humors, in contact with the lens anterior and posterior, respectively. Kannan's work has indicated that glutathione can be taken up ex vivo from the aqueous, by perfused isolated lens, but has not investigated; (1) whole animal glutathione injections, (2) the relative proportion of reduced and oxidized glutathione, and (3) the possibility that uptake can occur from the vitreous (in contrast to the aqueous humor) route.
Methods: 3H- or 35S-glutathione was injected into rats intraperitoneally and the radioactivity in serum and lens homogenates followed.
Results: The 3H-radioactivity reached a peak in the serum approximately 20-30 min after injection. Counts were also found in the lens, aqueous and vitreous humors. HPLC using a C18 Bondapak column (37 x 300 mm) indicated that the majority of the 3H-radioactivity in the lens was found in a component of a lower molecular weight than glutathione, but 8.1% of the counts occurred in the peak corresponding to reduced glutathione. Analysis of the unidentified radioactive component revealed a mobility the same as that of a dipeptide. Further analysis suggested this contained the amino acids cysteine and glycine bound in peptide linkage. These results suggest that glutathione may be degraded by the [gamma]-glutamyl cycle, and the action of transpeptidase produced cysteinylglycine. To confirm these results, similar experiments were undertaken using 35S-glutathione injection, to test whether a differently labelled form would be able to enter the lens. Homogenates prepared from the lens 20 min after 35S-glutathione injection were fractionated by HPLC. The glutathione peak contained 4.5% of the radioactivity in the lens extract. This amount was similar in quantity to the value for 3H-glutathione uptake by the lens. The average of the two values indicated that 6.3% of the total lens label was glutathione. The source of the labelled glutathione taken up by the lens was investigated by determining its concentration in the aqueous and vitreous humors and serum. The dipeptide appeared to be the major radiolabelled form occurring in the serum. This may explain its high level in the lens, as a result of uptake from other sources. Analysis using HPLC revealed that reduced glutathione (GSH) was the predominant chemical species of glutathione in the aqueous humor. In the vitreous humor, oxidized glutathione (GSSG) was the major species. The ratio of GSSG:GSH in the vitreous varied between 2:1 and 4:1.
Conclusions: Over a 4 h period the lens could obtain 12.3% of its total GSH from the injected GSH, using the specific activity of the labelled glutathione to calculate the actual uptake of glutathione by the lens, suggesting a half-time of 16.25 h for replenishing GSH from external sources. The probable route of glutathione entry was by blood plasma and aqueous since the specific activity of the vitreous humor was too low for the vitreous to be a possible source of the lens GSH.
The generally accepted view of the etiology of cataracts is that many factors contribute to the causation of cataract, i.e., that cataract causes are multifactorial. Although many different types of cataract have been described at different locations in the eye lens, the possibility exists that cortical lens cells have a limited set of responses to stress. Evidence has accumulated that, after an initiation of cellular damage leading to a multi-step mechanism of pathogenesis, the process of cortical cataract formation involves several steps leading to globule formation . One such step appears to be oxidative damage. This has been indicated by antioxidant risk reduction for cataract resulting from many agents or apparent causes: diabetes , steroids , antibiotic drugs , age , cytochalasin D , radiation [7,8], and heat stress . Spector et al. have suggested that hydrogen peroxide, which can be synthesized and degraded by aqueous humor , may be the responsible oxidizing species. The observation that supplemental dietary vitamins E and C in humans  reduce the risk or retard the relatively slow development of aging cataract, shows striking parallelism to the reduction of risk in a much higher stress in several animal models by vitamin E [2-4,6-9]. This gives an indication that a similar oxidative step may be involved in the pathogenesis of both human and animal model cataracts. If this step is prevented, by addition of antioxidants, the cortical cataract formation can be prevented or delayed [2-9,11-15]. Several mechanisms suggested to be responsible for the cortical cataract formation involve changes in the state of polymerization of the lens cell's cytoskeleton; (1) dimerization of spectrin occurs by oxidation of sulfhydryls in the diamide-treated frog lens ; (2) oxidative stress results in calcium influx, which activates lens calpain [17,18]; (3) calpain digests cytoskeletal fodrin, resulting in blebbing of the weakened cell membrane into globules which scatter light .
Several clinical groups have suggested that elevation of glutathione (GSH) could have cataract preventive activity. Such preventive activity has been demonstrated by GSH eye drops [20-22] and by dietary regimens designed to elevate antioxidant levels, including vitamin C and GSH .
Previous work reported by us has indicated that externally administered GSH may assist in preventing diabetic cataract in vitro  or in vivo (by injection) . In vitro, GSH (0.1 mM) was effective in preventing cataract induced by 55.6 mM glucose, although the concentration used was only about one-tenth of the concentration of GSH in the lens . Several diabetic rats were found not to form cataracts when injected with GSH, although in vivo, no significant preventive effect was identified.
Previous work summarized by Reddy  had suggested that reduced GSH cannot pass through the cell membrane to enter the cytoplasm, and further suggested that the oxidized form (glutathione disulfide, GSSG) could traverse the cell membrane. He suggested that all GSH in the lens must be synthesized in situ, since he could not detect any in the aqueous humor. More recently, Rathbun's group investigated the synthetic pathway for GSH in the lens in detail , and concluded that the synthetic pathway was functioning in the lens . The rate-limiting enzyme in the synthetic pathway seemed to be the first enzyme, cysteinylglycine synthetase, for which cysteine is one of the substrates. GSH and cysteine concentrations fall rapidly during the first two weeks, in the precataractous diabetic rat lens: decreased cysteine would impair the functioning of the synthetic pathway which could replenish the lost GSH . Rathbun, Schmidt, and Holleschau  showed that glutathione also decreased in one type of human cortical cataract. Activity loss of glutathione synthesis enzymes was associated with human subcapsular cataract. Consistent with this, in several model systems [30,31] they were also able to show that cysteine precursors, which could result in elevation of the glutathione levels in the lens in rat naphthalene cataract and mouse acetaminophen cataract, were able to prevent the development of cataracts [30,31].
In contrast to the prevailing hypothesis that all lens GSH must be synthesized in the lens [25,26,28], the previous observation of protection in vitro by GSH suggested that exogenous GSH could protect the lens. The GSH could protect oxidation-sensitive sites on the exterior of the lens cell membranes, or could enter the lens either in the reduced or oxidized form to elevate the lenticular GSH. If oxidized GSH (glutathione disulfide) could enter the cell, however, we reasoned that reduction by glutathione reductase could result in an increase in the intracellular glutathione concentration. In the experiments described here (published in abstract form: ), we decided to follow the entry of GSH in the lens following an intraperitoneal injection of radioactive GSH, in order to determine whether, in vivo, exogenous GSH represents an important route for GSH to enter the cell. In performing these experiments two different isotopic forms of GSH, labelled with 3H and 35S, were used in order to confirm that both differently isotopically labelled forms of GSH, labelled in both sulfur and tritium of the GSH, behaved similarly in their uptake by the lens.
Since this work was done, Kannan's group has provided independent evidence that transport of GSH into the aqueous humor occurs and that reduced GSH is actively taken up by the eye lens . This work is important in that it complements his investigations and confirms his conclusions by: (1) performing the work in vivo with whole animal glutathione injections, rather than ex vivo, (2) investigating the relative proportion of reduced and oxidized glutathione in the different eye compartments and fluids at various times after injection into the animal, and (3) checking on the possible hypothesis that uptake can occur from the vitreous (in contrast to the aqueous humor) route.
All procedures followed guidelines of the Canadian Council of Animal Care. Male rats were obtained from commercial sources (Charles River, St.-Constant, Quebec, Canada), and injected with radioactive GSH, either glutathione-2-3H(NEN)- or 35S-glutathione(NEN)-, 0.8 µCi 3H- or for 35S-GSH, 1 µCi, in 2 µ moles of cold GSH in 0.2 ml PBS, diluted 1/10 with cold GSH (0.01 M). For HPLC, (35S-GSH 21.2 µCi) was injected in 0.2 ml of 0.01 M GSH. Following the injection the rat was returned to its cage until anesthesia and sacrifice by guillotine. Blood was collected and either allowed to clot or treated with sodium citrate to prevent clotting. Serum was obtained by aspiration with a pasteur pipette after the clot of blood retracted. Whole blood was frozen and treated with concentrated perchloric acid to deproteinize, at a final concentration of 0.5 M. Lenses were homogenized in 0.1 M sodium phosphate buffer pH 7.4 and the homogenate deproteinized by addition of an equal volume of 1 M HClO4 containing 0.002 M Na EDTA (PEDTA). High performance liquid chromatography was performed using a Bondapak C18 column with a mobile phase of 99% 0.1 M monochloroacetate buffer pH 3, and 1% methanol. The flow rate at 25 °C was 1 ml/min. GSH and GSSG were detected by a Bioanalytical Systems (West Lafayette, IN) electrochemical detector with 2LC-4B controllers with dual Hg/Au electrodes, set at the following potential; upstream -1.0 V, downstream +0.15 V and detector sensitivity 50 n A/V. Standard curves for GSH and GSSG (glutathione disulfide) were obtained, since GSH estimation was more sensitive than GSSG on a per mole basis over the 0-800 picomole range.
Fractions containing GSH or GSSG were collected and assayed by scintillation counting; free amino acids were analysed. Following hydrolysis, peptide combined amino acid residues were identified. The N-terminal amino acid of the peptide was identified using dansylation.
Injected GSH (intraperitoneally) enters the eye lens. Following injection of 3H-GSH with carrier cold GSH into the rat, two rats were sacrificed at each time period. The dpm were plotted for each pair as a function of time after injection. There was an initial rapid rise in plasma counts at 10 min after injection, and falling at 20 min, then gradually increasing again towards the initial value (Figure 1) by 1.5 h. A slight decline at 2 h was followed by a rise to a plateau which remained constant until 24 h.
GSH in whole blood showed an increase over the first two h to a plateau (Figure 2). Table 1 shows the radioactivity found in the plasma and eye fluids and parts and in the kidney at 0.5 h and 1 h after injection.
3H-GSH in the lens reached a maximum concentration in dpm after 10-20 min, decreasing by 2 h to initial levels (Figure 3).
The deproteinized homogenate from the maximum peak at 20 min was analysed by HPLC using the C18 column to obtain the fractions containing GSH, and each fraction was analyzed for free amino acids (Table 2) and amino acids in peptide linkage (bound, Figure 4, Table 3). These fractions were also counted to determine the radioactivity in each fraction. Although a majority of counts was obtained in fraction 7, and a lesser number of counts in fraction 9 (8.1% of total), the composition of the fractions indicated that fraction 9 contained glutamate, cysteine and glycine in peptide linkage. This fraction eluted from the C18 column at the position of GSH, consistent with this composition. The majority of the counts found in fraction 7 appeared to be associated with a dipeptide, cysteinylglycine, as indicated by the finding of cysteine and glycine in peptide linkage in this fraction. Fraction 13 was similar in composition to fraction 9, consistent with it containing oxidized glutathione (disulfide form: GSSG).
Because the counts taken up by the lens were too low for HPLC separation, in another experiment 21 µCi were injected i.p. into 2 rats (254 and 271 g) and the rats sacrificed after 20 min and serum and lens prepared. The serum counts were 4.35 x 105 dpm/ml. The lens counts were 1.47 x 104/g ±0.33 x 104 ie. 3.36% of the serum level, or 0.00133% of GSH injected initially.
HPLC analysis of the samples containing the highest concentration of 35S-GSH revealed that the peak of 35S-GSH (contained in fraction 9) was 4.5 ±1.9% of total counts. This indicates that of the 0.00133% of the injected 35S-label entering the lens, 4.5% was actually still present as GSH. Combining results, the 3H-(8.1%) and 35S-GSH (4.5%) content as percent of total counts in lens gave 6.3 ±1.21% for the average percent GSH in the total isotope taken up by lens.
The source of the GSH was investigated in a preliminary way, by counting the activity in the aqueous and vitreous humors after 3H-GSH injection (Table 4). This data indicates that both aqueous and vitreous may serve as sources for GSH. In this regard, it is interesting to observe the ratio of GSH/GSSG found in aqueous and vitreous; HPLC analysis indicates that no GSSG is detectable in the aqueous, which contains traces (picomolar levels) of GSH but not detectable GSSG. By contrast, the major component in the vitreous is GSSG, in a ratio of GSH/GSSG: 0.246. The average ratio of GSH/GSSG increases from 0.5 h to 4 h. The increases (in plasma 1.84, lens 1.31, and vitreous 1.91), are consistent with GSH entering the vitreous, since the actual amounts of GSSG decreased only slightly (7.08 to 6.45 nmol/ml), but the amount of GSH increased (1.74 to 3.03 nmol/ml).
In the lens, by contrast (Table 4), the GSH/GSSG ratio is very high, due to the higher concentration of GSH in the lens. The levels of GSH increased (111-128 nmol/lens) while the GSSG decreased (39.7-35 nmol/lens). Consistent with this, blood levels changed for GSH from 312 nmol/ml at 30 min to 345 nmol/ml at 4 h. GSSG changed from 256 nmol/ml at 30 min to 163 nmol/ml at 4 h.
To check on the amount of GSH entering the lens, and the route of entry into the lens (by way of the aqueous or vitreous), the amount of labelled GSH at 30 min and 4 h was calculated for the lens, aqueous and vitreous humors (Table 5). The assumption made in this calculation was that the percent of the label occurring as GSH, which remains after the initial injection, is approximately 6.3% of the total label. This average was found (see above) by HPLC fractionation of label in the lens for both labelled forms of GSH. These values were then compared with the actual chemical concentrations in µmol/ml in the various fractions, using the following approximations; plasma, aqueous and vitreous were considered to be essentially aqueous solutions, so concentrations calculated in µmol/ml could be used directly. In the case of the lens, its protein content resulted in a calculated water content of 75% of the lens weight, so the concentrations were corrected to allow for a 75% water content. This resulted in slightly higher concentrations being calculated for the lens than if it was assumed to be 100% water.
Using this data the plasma GSH label which had entered the circulation was approximately 5.4% (30 min) and 4.6% (4 h) of the total reduced GSH. In the lens the labelled GSH which had entered was 7.2% (30 min) and 12.3% (4 h) of total reduced GSH. The possible sources of this lens label were the aqueous and vitreous humors. Only the aqueous humor, with levels of newly entering GSH at 18.6% (30 min) and 26.2% (4 h) was capable of providing precursor for the lens reduced GSH. Vitreous humor GSH with newly entering GSH at 0.39% (30 min) and 0.38% (4 h) of the total GSH would not be capable of serving as a precursor of the lens GSH to give the levels above. Even if oxidized GSSG were serving as a precursor of the lens GSH, the vitreous could not be considered as a precursor because of the low percent of label in the vitreous GSSG. This data is somewhat surprising since it indicates that a significant source of lens GSH may be exogenous GSH not made within the lens. No matter what the route of the GSH to the lens, the lens GSH provided from the exogenous label was significant; if 12.3% of the GSH could be provided to the lens from exogenous sources in 4 hr, then in 24 h, 74% of the total GSH could enter the lens from outside sources. To replace 100% of the lens GSH would require only 32.5 h (1 1/3 days). This is surprising and may provide an additional source of GSH in addition to the synthetic pathway so well elucidated by Rathbun's group. In agreement with their previous estimates that the half life of GSH in the incubated rat lens is 23 h , the half life of replacement in the rat lens in vivo using the data from these experiments would be of similar magnitude (16.25 h).
This could be important since Rathbun and co-workers have shown that the synthesis of GSH is limited by the enzymes of the synthetic pathway .
These experiments show that two different isotopically labelled forms of GSH can enter the lens after intraperitoneal injection. Although the percent of GSH counts in the lens is only 6.3% when both forms are averaged, the majority of the loss of GSH appears to be due to metabolism of GSH after its absorption, probably in the liver. Metabolism by the gamma-glutamyl cycle converts it to cysteinylglycine.
The dipeptide level as a proportion of total label is greater than 90% in both serum and lens. Both dipeptide and GSH are present in serum and lens in similar proportions. The similarity of these proportions is consistent with the hypothesis that the serum is the original source of both GSH and dipeptide, and that both enter the lens via the aqueous humor, which is formed by ultrafiltration from the plasma by the ciliary body, supporting in vivo the ex vivo results of Kannan's group in which the perfused eye model was used .
The conversion of GSH to cysteinylglycine after injection is very rapid, taking place in only a few minutes. Presumably a portion of this dipeptide may be metabolized to reform GSH in the lens, or if not metabolized, would serve as a source of reduced sulfhydryl groups intracellularly. A similar situation may occur in other organs, since similar uptake of 3H-GSH was shown by the kidneys of these rats.
While previous research results pointed to a role for exogenous injected GSH in the prevention or amelioration of cataract [1,20-24], the work of Reddy  was frequently cited to indicate that GSH per se could not enter the lens unless present as the oxidized form, glutathione disulfide. In fact, Reddy's paper  does not produce any evidence to prove his statement "No GSH appears to be present or has been detected in the aqueous humor, and it was natural to expect that the tripeptide may be synthesized in the lens." It appears, in fact, that the insensitivity of the techniques in use at the time this statement was written may have resulted in failure to detect GSH levels in the aqueous which we have measured by modern HPLC techniques.
Taken with Kannan's independent investigations of GSH uptake by the lens, the data presented here support Kannan's results, indicating that reduced GSH can be taken up by the lens from the aqueous. This finding may permit a rationalization of the efficacy of GSH in preventing cataracts previously demonstrated by our papers [1,24] and by Japanese workers who used it in eye drops to arrest and reverse cataractogenesis [20-22]. GSH in the aqueous could protect the exterior sites in the lens cell membrane as in our previous in vitro experiment , or enter the lens to maintain intracellular GSH concentrations.
GSH is known to perform an important role in maintaining lens integrity, since its abrupt loss precedes formation of the opacity in a well-described model of cataracts, radiation cataract  and in several other animal models [29-31]. Its early loss in precataractous diabetic rat lenses precedes opacification and may lead to oxidative stress as a result of antioxidant deficit . Consistent with this observation, in model diabetic lenses in which R-[alpha]-lipoic acid reduced opacity and protein leakage , the biologically active R-[alpha]-lipoic acid maintained the GSH concentration , but the biologically inactive isomer S-[alpha]-lipoic acid and racemic lipoic acid which were not protective, did not maintain lens GSH.
These data suggest that it may be possible to manipulate experimentally levels of GSH in order to maintain them. Although a caveat must be observed since it is not always possible to extrapolate the results obtained in animal models to the human, this may permit prevention of some of the oxidative stress leading to the pathological changes occurring in the final stages of cataract formation.
Investigations of GSH administration in the diet have also indicated that the tripeptide can be successfully administered in the diet, and increase the concentration of GSH in the plasma . Although it is surprising that it will survive the proteolytic attack of the digestive enzymes, significant absorption was shown to occur in the small intestine. This suggests that use of GSH as a dietary supplement may be a useful alternative to its administration in eye drops. In this mode it may be a useful free radical scavenger of the glycation-related free radicals produced by the Maillard reaction.
This paper is dedicated to the memory of Walter Chung, long serving technical officer in the University of Western Ontario Department of Biochemistry, who cheerfully contributed ideas and effort to the work described in this paper.
This research was funded by the Medical Research Council of Canada and the U.S. Army Medical Research and Development Command.
A portion of this work was presented as part of a symposium at the International Congress of Eye Research in 1990.
1. Bhatnagar A, Ansari NH, Wang L, Khanna P, Wang C, Srivastava SK. Calcium-mediated disintegrative globulization of isolated ocular lens fibers mimics cataractogenesis. Exp Eye Res 1995; 61:303-10.
2. Creighton MO, Trevithick JR. Cortical cataract formation prevented by vitamin E and glutathione. Exp Eye Res 1979; 29:689-93.
3. Creighton MO, Sanwal M, Stewart-DeHaan PJ, Trevithick JR. Modeling cortical cataractogenesis. V. Steroid cataracts induced by solumedrol partially prevented by Vitamin E in vitro. Exp Eye Res 1983; 37:65-75.
4. Creighton MO, Trevithick JR, Sanford SE, Dukes TW. Modelling cortical cataractogenesis. IV. Induction by hygromycin B in vivo (swine) and in vitro (rat lens). Exp Eye Res 1982; 34:467-76.
5. Robertson JM, Donner AP, Trevithick JR. Vitamin E intake and risk of cataracts in humans. Ann N Y Acad Sci 1989; 570:372-82.
6. Kilic F, Trevithick JR. Vitamin C reduces cytochalasin D cataractogenesis. Curr Eye Res 1995; 14:943-9.
7. Ross WM, Creighton MO, Inch WR, Trevithick JR. Radiation cataract formation diminished by vitamin E in rat lenses in vitro. Exp Eye Res 1983; 36:645-53.
8. Ross WM, Creighton MO, Trevithick JR. Radiation cataractogenesis induced by neutron or gamma irradiation in the rat lens is reduced by vitamin E. Scanning Microsc 1990; 4:641-50.
9. Stewart-DeHaan PJ, Creighton MO, Sanwal M, Ross WM, Trevithick JR. Effects of vitamin E on cortical cataractogenesis induced by elevated temperature in intact rat lenses in medium 199. Exp Eye Res 1981; 32:51-60.
10. Spector A, Ma W, Wang RR. The aqueous humor is capable of generating and degrading H2O2. Invest Ophthalmol Vis Sci 1998; 39:1188-97.
11. Ross WM, Creighton MO, Stewart-DeHaan PJ, Sanwal M, Hirst M, Trevithick JR. Modelling cortical cataractogenesis: 3. In vivo effects of vitamin E on cataractogenesis in diabetic rats. Can J Ophthalmol 1982; 17:61-6.
12. Linklater HA, Dzialoszynski T, McLeod HL, Sanford SE, Trevithick JR. Modelling cortical cataractogenesis VIII: effects of butylated hydroxytoluene (BHT) in reducing protein leakage from lenses in diabetic rats. Exp Eye Res 1986; 43:305-14.
13. Trevithick JR, Linklater HA, Mitton KP, Dzialoszynski T, Sanford SE. Modeling cortical cataractogenesis: IX. Activity of vitamin E and esters in preventing cataracts and gamma-crystallin leakage from lenses in diabetic rats. Ann N Y Acad Sci 1989; 570:358-71.
14. Linklater HA, Dzialoszynski T, McLeod HL, Sanford SE, Trevithick JR. Modelling cortical cataractogenesis. XI. Vitamin C reduces gamma-crystallin leakage from lenses in diabetic rats. Exp Eye Res 1990; 51:241-7.
15. Linklater HA, Dzialoszynski T, McLeod HL, Sanford SE, Trevithick JR. Modelling cortical cataractogenesis. XII: Supplemental vitamin A treatment reduces gamma-crystallin leakage from lenses in diabetic rats. Lens Eye Toxic Res 1992; 9:115-26.
16. Gandolfi SA, Tagliavini J, Belpoliti M, Duncan G, Maraini G. Oxidative cross-linking of fodrin parallels a membrane conductance increase in the mammalian lens. Curr Eye Res 1988; 7:747-54.
17. David LL, Varnum MD, Lampi KJ, Shearer TR. Calpain II in human lens. Invest Ophthalmol Vis Sci 1989; 30:269-75.
18. David LL, Shearer TR. Calcium-activated proteolysis in the lens during selenite cataractogenesis. Invest Ophthalmol Vis Sci 1984; 25:1275-83.
19. Kilic F, Trevithick JR. Modelling cortical cataractogenesis. XXIX. Calpain proteolysis of lens fodrin in cataract. Biochem Mol Biol Int 1998; 45:963-78.
20. Oguchi M, Seki H. [Glutathione and eye diseases]. Ganka 1970; 12:125-32.
21. Fujii S, Takahashi K, Namba K. [Effects of an eyedrop solution, Tathion, on cataract]. Nippon Ganka Kiyo 1968; 19:136-42.
22. Kandori F, Kurimoto S, Fukunaga K, Suyama T, Honda I. [Clinical use of topical glutathione for cataract]. Nippon Ganka Gakkai Zasshi 1967; 71:689-97.
23. Weigelin E, Hockwin O. Rapport sur l'etude clinique controlee randomisee de Phakan/Phakolen. In: Vieillissement du cristallin: pathogenie des opacites du cristallin au cours du vieillissement et possibilite d'un traitement medicamenteux. Hockwin O, editor. Montpellier, France: Chauvin-Blache; 1982. p. 213-32.
24. Ross WM, Creighton MO, Trevithick JR, Stewart-DeHaan PJ, Sanwal M. Modelling cortical cataractogensis: VI. Induction by glucose in vitro or in diabetic rats: prevention and reversal by glutathione. Exp Eye Res 1983; 37:559-73.
25. Reddy VN. Metabolism of glutathione in the lens. Exp Eye Res 1971; 11:310-28.
26. Rathbun WB, Holleschau AM. The effects of age on glutathione synthesis enzymes in lenses of Old World simians and prosimians. Curr Eye Res 1992; 11:601-7.
27. Murray DL, Rathbun WB. Conditions for maximizing and inhibiting synthesis of glutathione in cultured rat lenses: an application of HPLC with radioisotope detection. Curr Eye Res 1990; 9:55-63.
28. Mitton KP, Dean PA, Dzialoszynski T, Xiong H, Sanford SE, Trevithick JR. Modelling cortical cataractogenesis. 13. Early effects of lens ATP/ADP and glutathione in the streptozotocin rat model of the diabetic cataract. Exp Eye Res 1993; 56:187-98.
29. Rathbun WB, Schmidt AJ, Holleschau AM. Activity loss of glutathione synthesis enzymes associated with human subcapsular cataract. Invest Ophthalmol Vis Sci 1993; 34:2049-54.
30. Nagasawa HT, Shoeman DW, Cohen JF, Rathbun WB. Protection against acetaminophen-induced hepatotoxicity by L-CySSME and its N-acetyl and ethyl ester derivatives. J Biochem Toxicol 1996; 11:289-95.
31. Rathbun WB, Nagasawa HT, Killen CE. Prevention of naphthalene-induced cataract and hepatic glutathione loss by the L-cysteine prodrugs, MTCA and PTCA. Exp Eye Res 1996; 62:433-41.
32. Stewart-DeHaan PJ, Dzialoszynski T, Chung W, Trevithick JR. Uptake by the lens of glutathione injected into the rat. Proceedings of the International Society for Eye Research; 1990 July 24-August 4; Helsinki, Finland. New York: International Society for Eye Research; 1990. p. 248.
33. Zlokovic BV, Mackic JB, McComb JG, Kaplowitz N, Weiss MH, Kannan R. Blood-to-lens transport of reduced glutathione in an in situ perfused guinea-pig eye. Exp Eye Res 1994; 59:487-96.
34. Matsuda H, Giblin FJ, Reddy VN. The effect of x-irradiation on cation transport in rabbit lens. Exp Eye Res 1981; 33:253-65.
35. Kilic F, Handelman GJ, Serbinova E, Packer L, Trevithick JR. Modelling cortical cataractogenesis 17: in vitro effect of a-lipoic acid on glucose-induced lens membrane damage, a model of diabetic cataractogenesis. Biochem Mol Biol Int 1995; 37:361-70.
36. Kilic F, Handelman GJ, Traber K, Tsang K, Packer L, Trevithick JR. Modelling cortical cataractogenesis XX. In vitro effect of alpha-lipoic acid on glutathione concentrations in lens in model diabetic cataractogenesis. Biochem Mol Biol Int 1998; 46:585-95.
37. Hagen TM, Wierzbicka GT, Bowman BB, Aw TY, Jones DP. Fate of dietary glutathione: disposition in the gastrointestinal tract. Am J Physiol 1990; 259: G530-5.