Implications of aldose reductase in cataracts in human diabetes

Transcription

1 Implications of aldose reductase in cataracts in human diabetes Shambhu D. Varma, Stanley S. Schocket, and R. D. Richards Cataracts removed intracapsularly by cryoprobe technique from human diabetics were analyzed for sugars and polyols by gas liquid chromatography. The contents of sorbitol and fructose of lenses followed blood glucose levels at least up to 250 mg/dl. Studies indicate thai human lens is capable of synthesizing substantial amounts ofpolyol pathway metabolites given exposure to high glucose levels such as are prevalent in diabetes. The synthesis of sorbitol was found to be susceptible to quercitrin, an inhibitor of aldose reductase. The implications of these findings in the formation of cataracts in diabetic individuals have been discussed. Key words: cataract, diabetes, sorbitol, polyol pathway, aldose reductase Lenses of experimentally diabetic animals have been shown to contain excessive amounts of sorbitol and fructose. 1 ~ 3 Sorbitol is synthesized by the action of aldose reductase on glucose, the reducing equivalents being derived from NADPH. Fructose is derived by the action of sorbitol dehydrogenase, NAD being the electron acceptor. The conversion of glucose to sorbitol and the transformation of the latter to fructose through the above-mentioned series of reactions constitute the polyol pathway. Excessive accumulation of sorbitol and fructose in the lens fibers and epithelium is considered osmotically traumatic, leading to tissue swelling and opacity. Since the first reaction of the From the Department of Ophthalmology, University of Maryland Medical School, Baltimore, Md. Supported by National Eye Institute (EY and its CCRG Supplement), Research to Prevent Blindness Inc., Fight for Sight Inc., and the Upjohn Co. These findings were part of the material presented at the ARVO meeting 1978 (INVEST. OPHTHALMOL. VISUAL SCI. 17(ARVO Suppl.):300, 1978). Submitted for publication May 12, Reprint requests: Dr. Shambhu D. Varma, Department of Ophthalmology, School of Medicine, University of Maryland, Baltimore, Md polyol pathway is catalyzed by aldose reductase, Kinoshita 4 considers this enzyme as the key initiator of cataractogenesis in diabetes and galactosemia. The above concept has been amply proven in experimental animals. 5 ' 6 However, so far it has not been possible to assess as completely as in experimental animals the role of aldose reductase in the development of cataracts in human diabetics, 7 wherein the incidence of lens opacity as well as maturation of such opacity into frank clinical cataracts has been suggested to be accelerated. 8 Thus there is an obvious need for further studies toward determining the role of this enzyme in cataractogenesis in human diabetics. This report, along with that of Chylack et al., 9 presents evidence that aldose reductase may play a substantial role in the development of lens opacity in diabetic individuals. Direct analysis of cataracts from diabetic patients and the in vitro incubation experiments indicated that human lens is capable of synthesizing appreciable amounts of sorbitol and fructose in response to high levels of glucose. The synthesis of sorbitol is susceptible to inhibition by quercitrin, an aldose reductase inhibitor /79/ $00.50/ Assoc. for Res. in Vis. and Ophthal., Inc. 237

2 238 Varnia, Schocket, and Richards Invest. Ophthalmol. Visual Sci. March 1979 Table I. Sugars and polyols in cataracts extracted from human diabetics Group A (19) B(12) C(16) Significance of differences (f) Blood sugar (mg/dl) 115 ± ± ± 45 A vs. B B vs. C A vs. C Fructose (F) 0.50 ± ± ± 5.00 Sorbitol (S) 0.25 ± ± ± 2.10 Total F and S Glucose Inositol 0.75 ± ± ± ± ± ± 1.50 <0.30 < ± ± ± <0.27 <0.77 <0.13 Values are in /xm/gm wet weight, expressed as means ± SD. Statistical analysis consisted of a two-tailed t-test; A p value <0.001 was considered significant. Sorbitol (S) and fructose (F) data in the table have been treated together but are significant to the same level when treated individually also. Figures in parenthesis indicate the number of samples. The levels of individual sugars and polyols in normal lens do not exceed 0.5 /u,m/gm. The average age was 62 ± 15 years. Blood sugar was determined by the familiar glucose oxidase peroxidase method; these values thus represent blood glucose levels observed after an overnight fast, on the morning of the day of surgery. Materials and methods Cataracts were extracted surgically by the cryoprobe technique. The extracted cataracts were singly weighed and homogenized in 1 ml of 0.3N zinc sulphate solution as soon as possible (within 10 min). Deproteinization was accomplished by addition of an equivalent amount of 0.3N Ba(OH) 2 solution (titrated to phenolphthalein end point) and further homogenization. The uniform homogenate was centrifuged to obtain a clear supernatant. A measured amount of the supernatant (about 1.5 ml) was lyophilized. Sugars in the lyophilized sample were analyzed by gas chromatography following their derivitization to TMS (trimethylsilyl) ethers. The TMS ethers were prepared by allowing the samples to react with a mixture of hexamethyldisilazane, trimethylchlorosilane, and pyridine (2:1:10 v/v), obtainable as Trisil from Pierce Chemical Co., Rockford, 111. This method has been used by the authors previously and has given satisfactory results. 10 One tenth of a milliliter of the mixture was found to be more than adequate to silylate the sugars present in the lyophilized sample. The separation and quantitation of the TMS ethers was achieved on a glass column, 8 feet x 0.4 mm, packed with 3% SE-30 on mesh chromosorb W and mounted in Perkin Elmer Gas Chromatograph 3920 B equipped with a flame ionization detector and peak integrator. The carrier gas was N 2 (40 ml/ min). The column temperature was maintained at 170 C. The temperatures of the injection port and the detector were 190 and 255 C, respectively. The effect of quercitrin on the synthesis of sorbitol was determined by incubating intact and visibly clear human lenses in medium with and without quercitrin and determining the postincubation levels of sorbitol, fructose, glucose, and inositol. The medium used for incubation was supplemented with glucose so that the final concentration of this sugar was 30 mm.' The eyes used in these experiments were from the local eye bank. Such eyes, after donation, were kept on ice and sent to the laboratory as soon as obtained by the bank. The time lapse between the death of the person and enucleation of the eye was between 2 and 6 hr. Travel between the eye bank and the laboratory took approximately 1 hr. Lenses from the contralateral eyes were used as controls in the present experiments. The effect of the time interval, if any, applies equally to each member of the lens pair. Following incubation, the lenses were briefly rinsed with 0.5 ml of 0.9% NaCl solution and transferred to homogenizing tubes containing 1 ml of 0.3N ZnSO 4. Further processing of the lenses for sugar analyses was identical to that described for direct human lens analyses in the preceeding paragraph. Results and discussion The results of analyses of cataracts for fructose, sorbitol, glucose, and inositol are presented in Table I. The data have been divided into three groups on the basis of the patients' fasting blood sugar levels tabulated against the levels of the carbohydrates in their cataracts so as to obtain the best fit. The blood sugar level was found to overlap by 10 to 15 mg/dl between the three groups thus formed. The total number of cases was 47. Group A was designated for patients with mild hyperglycemia (mean group blood sugar, 115 ± 23 mg/dl), group B was designated for patients having intermediate level hyperglycemia (mean blood sugar, 175 ± 50

3 Volume 18 Number 3 Aldose reductase, human diabetes cataracts BLOOD SUGAR LEVEL(mg%) BLOOD SUGAR MGA* SD 19 Fig. 1. Sorbitol + fructose contents in cataracts extracted from diabetic patients. The extracted cataracts were processed within 10 min of the extraction for analysis. Analysis was done on Nelson-Semyogi (ZnSO 4, Ba(OH) 2 ) protein-free filtrate following lyophilization. The Ts on the bars represent the standard error. Figures on the bars represent number of samples. mg/dl), and group C was assigned to patients having the highest degree of hyperglycemia (mean blood sugar, 280 ± 45 mg/dl). As is evident from the data presented in Table I, the most marked change associated with hyperglycemia was an elevation in lenticular sorbitol and fructose, the two metabolites of the polyol pathway. Another significant change was an elevation of lens glucose (compare groups A and C). Lens inositol was minimally different among the three groups. The level of sorbitol ranged from 0.25 /LtM/gm in group A, with the lowest blood sugar level, to about 6.0 //.M/gm in group C, with the highest blood sugar level. The level of lens sorbitol in group B, with intermediate-level hyperglycemia, was higher than in the group with mild hyperglycemia (A) and approached that in group C, with the highest Fig. 2. Relationship between blood sugar level and the content of sorbitol + fructose in cataracts extracted from diabetic patients. The number at the top indicates the number of samples at the respective point. The blood sugar is the mean of corresponding samples with standard deviations lying between 1 and 10 mg/dl except at the point of 280 ing, where the deviation was 20 mg/dl. This was done to obtain significant numbers at that point. A 0.5 to 1 mm difference in glucose concentration will affect the trend of lines minimally. The vertical line at each point represents the standard deviation with the middle point as mean. The vertical line on the last point corresponding to ±355 mg of glucose represents the range, since the sample size was small for calculating the deviation. degree of hyperglycemia. Similar correlation was observed for fructose levels. The elevation in the levels of lens sorbitol and fructose with increase in hyperglycemia is more strikingly apparent from the histogram in Fig. 1. In group B (intermediate hyperglycemia) the level was at least four times that in the group with mild hyperglycemia (A). In the group with advanced hyperglycemia, the level was likewise at least twelvefold higher. However, in this group of cataracts (C), the concentration of the polyol pathway metabolites was more variable in comparison to that encountered in the cataracts extracted from patients with a lower degree of hyperglycemia. This variation is more apparent in Fig. 2, which depicts graphically the relationship between the two variables at smaller intervals. The blood sugar values in this graph (abscissa) have been grouped to obtain the most meaningful representation. As is again evident, the

4 240 Varma, Schocket, and Richards Invest. Ophthalmol. Visual Set. March 1979 Table II. Effect of quercitrin on postincubation levels of sorbitol and fructose in human lens Glucose Fructose Sorbitol eriment No. i l o A Q O Mean ± SD o (two-tailed) (fim) q U g 10= ± 9.28 NS (til \1) ± 8.88 <0.001 (HM) X AQ Jl AR 4D 1UU 41.5 ± 7.59 <0.001 (fim) Inositol QO Ov) 1 OR 1ZD RR yu ± 20.0 NS Age of donors between 70 and 89 years. On dissection only those lenses were selected which were clear to naked eye. Majority of available eyes had lenses not entirely clear and were thus discarded. The controls consisted of contralateral lenses incubated in medium containing 30 mm glucose. In the experimental group the medium contained in addition 10~ 4 M quercitrin. Period of incubation was overnight. Other incubation conditions were similar to those used previously. 10 Analytical procedures have been described in the text. NS = not significant, p is significant. It refers to the comparison of / x 100, with the contralateral control taken as 100. = rol; = erimental. sorbitol and fructose contents in the lens rose predictably up to 240 ± 5 mg/dl blood glucose concentration. In samples obtained from patients with higher blood sugar there was a leveling tendency. Whether this plateau was a result of an increased leakiness in lens due to swelling induced by excessive sorbitol and fructose levels could not be ascertained. However, this possibility is indicated by an earlier report on incubation of lens in hyperglycemic medium. 11 Despite this uncertainty, however, the figures clearly demonstrate an overall rise in lens polyols (sorbitol and fructose) commensurate with a rise in blood glucose in the range of 100 to 250 mg/dl. In an earlier report the highest concentration of polyols reported in diabetic cataracts was 1.7 /xm/gm, and therefore the authors could not arrive at any conclusion about the significance of their data. 7 In only one noncataractous postmortem young (35 years) diabetic lens did they observe a concentration of sorbitol plus fructose similar to that found in the present cataracts. In that case, the net intracellular sorbitol and fructose rendered the intracellular milieu 21 mosm/ Liter more hypertonic, and this might have been a significant cataractogenic stress. However, in another case, where blood sugar was very high (190 mg/dl), no sorbitol could be detected. 7 Further studies on the status of the polyol pathway in human lens are thus desirable in order that the role of this pathway in the development of cataracts may be more completely understood. Kinoshita 4 suggested that aldose reductase, the first enzyme of the pathway, initiates the cataractous process by catalyzing the synthesis of polyols, the excessive polyols in the lens cells causing osmotic hydration of the tissue and its final opacification. This hypothesis implicating aldose reductase has been verified in galactosemic and diabetic animals through the use of aldose reductase inhibitors. 5 ' 6 The polyol levels (sorbitol and fructose) in the human cataracts observed in the present series of investigations appeared high enough to induce significant osmotic hydration and cataractous changes analogous to those observed in animal models. The levels of sorbitol and fructose occurring in actual situations may even be higher than reported here, since with maturation of diabetic cataracts the lens is known to become leaky, 12 and the levels picked up in our analysis may in fact represent residual amounts. The great difference between the lenses of the animal models and the human lens, however, is that the latter has a very low activity of aldose reductase (unpublished observa-

5 Volume 18 Number 3 Alclose reductase, human diabetes cataracts 241 tions). The results presented in this investigation, however, indicate that despite low levels of aldose reductase, the metabolic makeup of the human lens allows significant synthesis of sorbitol and its transformation to fructose, given exposure to high concentrations of glucose as in diabetes. Previous studies have shown that in addition to aldose reductase, the various other factors affecting the polyol synthesis are the activities of hexokinase (the initiator of glycolysis) and glucoses-phosphate dehydrogenase (required for NADPH supply), as well as the absolute concentrations and the ratios of the oxidized to reducted forms of the pyridine nucleotides. 10 ' l3 Proper synchronization of these factors can lead to substantial polyol synthesis even if aldose reductase is low. Investigations along these lines in human lens are in progress. Nevertheless, the response of the lens to high glucose of synthesizing substantial sorbitol and fructose is apparent. Further, it was observed that the human lens sorbitol synthesis is susceptible to aldose reductase inhibitors, since the level of this polyol attained in lenses incubated in high-glucose medium containing 10~ 4 M quercitrin, an aldose reductase inhibitor, 14 was substantially lower than in contralateral controls incubated in high-glucose medium not containing quercitrin. The same was true for fructose levels. The results of these experiments are shown in Table II. The postincubation level of sorbitol in lenses incubated in presence of quercitrin was about 40% of that attained in contralateral lenses incubated without quercitrin. These observations on the susceptibility of human lens sorbitol synthesis to aldose reductase inhibitors are similar to those of Chylack et al., 9 who used 1,3-dioxo-lHbenz (de)-isoquinoline-2 (3H) acetic acid, an inhibitor of aldose reductase 5 structurally quite different from the one used in these experiments. 14 In summary, these studies suggest a clear correlation between the level of blood sugar and the contents of polyol pathway metabolites attainable in the lenses of diabetic individuals. The levels observed in cataracts extracted from patients with intermediate and advanced levels of hyperglycemia suggest that the precataractous levels of polyols may be sufficient to initiate osmotic hydration of the lens and the concomitant cataractous changes. Thus, the osmotic theory of cataractogenesis found true in animal studies may be applicable to human diabetics also. Such cataractous changes can obviously be attenuated by aldose reductase inhibitors. REFERENCES 1. Van Heyningen, R.: Formation of polyols by the lens of the rat with sugar cataracts, Nature 164:194, Kinsey, V. E., VVatchl, C, Kuck, J. R. Jr., and Reddy, D. V. N.: Current research on culture in lenses. In Acta XVIII Concilium Ophthalmologicum, p. 865, Kuck, J. F. R. Jr.: The formation of fructose in ocular lens, Arch. Ophthalmol. 65:840, Kinoshita, J. H.: Mechanisms initiating cataract formation, Proctor Lecture, INVEST. OPHTHAL- MOL. 13:713, Dvomik, D., Simard-Duquesne, N., Krami, M., Sestanj, K., Gabbay, K. H., Kinoshita, J. H., Varma, S. D., and Merola, L. O.: Polyol accumulation in galactosemic and diabetic rats: rol by an aldose reductase inhibitor, Science 182:1146, Varma, S. D., Mizuno, A., and Kinoshita, J. H.: Diabetic cataracts and flavonoids, Science 195:205, Pirie, A., and Van Heyningen, R.: The effect of diabetes on the content of sorbitol, glucose, fructose, and inositol in the human lens,. Eye Res. 3:124, Caird, P., Hutchinson, M., and Pirie A.: Cataract and diabetes, Br. Med. J. 2:665, Chylack, L. T. Jr., Henriques, H., and Tung, W.: Inhibition of sorbitol production in human lenses by an aldose reductase inhibitor, INVEST. OPHTHAL- MOL. VISUAL SCI. 17(ARVO Suppl):300, Vanna, S. D., and Kinoshita, J. H.: Sorbitol pathway in diabetic and galactosemic lenses, Biochem. Biophys. Acta 338:632, Chylack, L. T., and Kinoshita, J. H.: A biochemical evaluation of cataracts induced in high glucose medium, INVEST. OPHTHALMOL. 8:401, Patterson, J. W., and Bunting, K. W.: Changes associated with the appearance of mature sugar cataracts, Invest. Ophthalmol. 4:167, Varma, S. D., and Kinoshita, J. H.: Absence of cataracts in mice with congenital hyperglycemia,. Eye Res. 19:577, Vanna, S. D., Mikuni, L., and Kinoshita, J. H.: Flavonoids as inhibitors of lens aldose reductase, Science 188:1215, 1975.