ERYTHROCYTE GLYCERALDEHYDE- REDUCTASE LEVELS IN DIABETICS WITH RETINOPATHY AND CATARACT

1268 I thank Prof. F.

Eisenberger, Dr J. Jocham, and Miss Hertha Bayer for their

help. Requests

for

reprints

should be addressed

to

W.

B., Institute for Surgical

Research, Klinikum Grosshadern, Marchioninistr. 15, Munchen 70, West

Germany.

specificity.

present evidence that the mean levels of erythrocyte glyceraldehyde reductase are raised in patients with diabetic retinopathy and cataracts. We also measured levels of polyol dehydrogenase, which has been found in erythrocytes of several mammals,18, 19 but found them to be not significantly altered in patients with diabetes or cataract. We

REFERENCES 1.

previously reported the presence of an enzyme, glyceraldehyde reductase, in human erythrocytes which has kinetic properties similar to those of lens aldose reductase, although 14-17 differing from this enzyme in substrate We have

Wanner K, et al. The use of shock waves for the destruction of renal calculi without direct contact. Urol Res 1976; 4: 175. 2. Chaussy Ch, Schmiedt E, Forssmann B, Brendel W Contact-free renal stone destruction by means of shock waves. Eur Surg Res 1979; 11: 36. 3. Brendel W, Chaussy Ch, Forssmann B, Schmiedt E. A new method of non-invasive destruction of renal calculi by shock waves. Br J Surg 1979, 66: 12. 4. Forssmann B, Hepp W, Chaussy Ch, Eisenberger F. Wanner K. Eine Methode zur berührungsfreien Zertrümmerung von Nierensteinen mit Stosswellen. Biomed Technik 1977; 22: 164.

Chaussy Ch, Eisenberger F,

now

Subjects and Methods

Subjects ERYTHROCYTE GLYCERALDEHYDEREDUCTASE LEVELS IN DIABETICS WITH RETINOPATHY AND CATARACT

Patients attending the Oxford Eye Hospital or the diabetic clinic the John Radcliffe Hospital were examined for the presence of cataract or retinopathy and put into one of five groups (see at

accompanying table). Groups

1 and 3

were

age-matched,

as were

GLYCERALDEHYDE-REDUCTASE AND POLYOL DEHYDROGENASE ACTIVITIES

M.

JAMES C. CRABBE

C. ORDE PECKAR HUNG CHENG

ANJANA BASAK HALDER

Nuffield Laboratory of Ophthalmology, University of Oxford, Oxford Eye Hospital, Oxford

and

levels of glyceraldehyde aldose-reductase-like enzyme present in the erythrocyte, were determined in 104 subjects, who were divided into five groups—diabetics with retinopathy and cataract, diabetics with retinopathy and no cataract, diabetics with no retinopathy and no cataract, nondiabetics with senile cataract, and non-diabetic normal controls. Diabetics with retinopathy and cataract had significantly higher mean enzyme activity levels than normal control subjects (2·5 fold increase, p<0·001); so had diabetics with retinopathy but no cataract (2 fold, p<0·01) and patients with senile cataract (1·5 fold, p<0·05). Juvenile diabetics had a significantly higher enzyme level than maturity onset diabetics. There was no significant difference in glyceraldehyde reductase between normal controls and diabetics without retinopathy or cataract, and no significant difference in polyol-dehydrogenase activity levels between any group studied. Enzyme activity did not correlate with age or glycosylated haemoglobin (HbA1c) levels. The increase in levels of erythrocyte glyceraldehyde reductase were due to increased amounts of active enzyme, rather than to elaboration of new kinetic pathways.

Summary

The

activity

reductase,

an

Introduction THE metabolic conversion of aldoses to ketoses is catalysed in the lens,I,2 liver3, seminal vesicles4, placenta5, brainy sciatic nerve’, pancreas’, kidney8, and aorta9 by the enzymes of the polyol pathway, aldose reductase and polyol dehydrogenase. Sorbitol, the intermediate in the metabolism of glucose to fructose by this pathway, diffuses poorly out of the cell, so it accumulates in certain tissues where there is hyperglycaemia. This, together with the reduced nucleotide and ketose levels, contributes to the pathological changes in diabetes mellitus and galactosaemia," particularly cataract formation, because high sugar alcohol concentration in the lens exerts a strong osmotic effect, causing ingress of water and eventual disruption of lens fibre membranes. 11-13

*Tested in 14 subjects. f Tested in 28 subjects. p =significance of difference from group (1).

groups 2, 4, and 5. Patients attending the John Radcliffe diabetic clinic also had their height and weight recorded. Cataract was defined as lens opacities visible on axial or oblique illumination when viewed with a "slit-lamp" or on direct ophthalmoscopy with a + 10 dioptre lens. Retinopathy was defined as the presence of any of the features of background or proliferative diabetic retinopathy visible on direct or indirect ophthalmoscopy. 104 subjects, aged 15-87 years, entered the study, which was blind.

Enzyme Assays Glyceraldehyde reductase erythrocytes and bovine

polyol dehydrogenase in human were assayed as described previously’ 14,20 by the use of DL-glyceraldehyde and NADPH in pH 6-2 potassium phosphate buffer and sorbitol and NAD+ in pH 8-9 ’Tris’-HC1 buffer, respectively, at constant ionic strength (0 - 2 mol/1). Each assay was run with a blank containing buffer, enzyme, and NADPH or NAD+, but without substrate, at the appropriate concentration. All progress curves were linear for at least 10 min. A plot of initial rate versus enzyme concentration was linear for glyceraldehyde reductase from human erythrocytes and for aldose reductase from bovine lens. ’Cary 210’ and ’Cary 119’ and

lens

1269

(Varian Instruments, London, U.K.) and Pye Unicam SP8000 (Pye Instruments, Cambridge, U.K.) double-beam spectrophotometers used in this study. The coefficient of variation of the procedure better than ± 5%.

were was

Non-fasting blood samples (10 ml) were withdrawn with minimum veno-occlusion and placed in heparinised tubes. Fresh blood was centrifuged at 15 000 rpm for 10 min, the plasma decanted, and erythrocytes washed twice with 0.9% saline. Erythrocytes were purified by column chromatography in alphacellulose fibre and micro-crystalline cellulose (Sigma Chemical Co., London, U.K.)15Packed cells were lysed by freezing and thawing. An equal, volume of 0 -1mmol/1 potassium phosphate buffer pH 6’ 2 was then added, and the suspension was diluted twenty-fold with distilled water. pH was monitored throughout. Enzyme activities are expressed as U ml-1 of packed red cells. Estimations were done at least twice in most cases. One U of activity is defined as that amount of enzyme catalysing the oxidation/reduction of 111mol of substrate min-Iat 30°C±0- 1°C. Statistical and Kinetic Methods Both parametric (Student’s t test) and non-parametric (Wilcoxon’s rank sum test) methods were used in the statistical analysis of the data, together with linear regression analysis for age, HbA, and height-weight product correlations. All analyses were done on a ’PET 16N’ microcomputer (Commodore Electronics). Kinetic methods were used as described previously.14-17 Initial rate data at each substrate concentration were fitted to alternative rate equations by the use of a Gauss-Newton non-linear regression programme on an ’ICL 2980’ computer.

Individual levels of erythrocyte groups 1-5.

Horizontal bars indicate the

x x

Results reductase activity was found in the of the 104 subjects examined. There was no correlation between glyceraldehyde reductase activity and age (r<0’4). The table shows the mean ± SD levels of glyceraldehyde reductase and polyol dehydrogenase in human erythrocytes in the five groups, and the p values for each group were compared with those for group 1, the normal control group; significance was calculated according to the Student’s t test. There was little difference between the p values calculated by this method and the distribution-free rank sum test. Compared with levels in normal controls (group 1) and diabetic controls (group 2), the mean erythrocyte glyceraldehyde reductase activity was significantly higher in diabetic subjects with retinopathy and cataract (2’5 fold increase, p<0-001), diabetics with retinopathy but no cataract (2 fold increase, p<0-01), and non-diabetic subjects with There was no "senile" cataract (1-5 fold increase, p<0’05). between difference (p>0-9) significant glyceraldehyde reductase levels in the normal control (group 1) and the diabetic controls (group 2). There was little difference in the mean values between groups 3 and 4 (p<0’2) and 4 and 5 (p<0’2), but groups 2 and 4 were significantly different (p<

Glyceraldehyde erythrocytes of 95

out

0.01). The

accompanying figure shows the individual levels of glyceraldehyde reductase activity in the five groups. Although there was some overlap in activity levels between each group, all normal and diabetic control subjects had

glyceraldehyde reductase levels lower than 1-8 x 10-1 U ml-1, whereas 55% of the diabetics with retinopathy and cataract had glyceraldehyde reductase levels above this value. The group with senile cataract all had glyceraldehyde reductase levels above 0-45 x 10-1U ml-l, whereas 44% of normal controls had activity levels below this value. Juvenile diabetics had a significantly higher mean enzyme activity (1-55

glyceraldehyde

mean

reductase

activity in

levels of enzyme activity for each group.

10- U ml-l, p<0’05) than maturity onset diabetics (0-96 10-lU ml- 1). There was no significant relation between

enzyme activity and age or sex either overall or in any one group, with values for r, the correlation coefficient, being in all cases. There was no significant change in polyol <0.37

in diabetics with retinopathy or cataract, or non-diabetics with senile cataract. There was no correlation between glyceraldehyde reductase levels and glycosylated haemoglobin (HbAIc) levels either in diabetic patients with retinopathy and cataract (r=0’53) or with retinopathy only (r=0’41). Within each group there was no significant correlation between enzyme activity and type or duration of diabetes. The kinetics of glyceraldehyde reductase, with DLglyceraldehyde as substrate, were similar whether the enzyme was from normal erythrocytes, erythrocytes from diabetics with retinopathy and cataract, or erythrocytes from senile cataract patients. In all cases, the steady-state rate equation was 2:2 in degree, and curve-shapes were similar to those obtained with the bovine lens enzyme when DLglyceraldehyde was used as substrate. 2,16,17 Similar kinetic parameters were obtained throughout. Quercitrin, a flavonoid used to delay cataract formation in rodents,2i totally inhibited enzyme activity in all cases at a concentration

dehydrogenase activity

of 10-’ mol/l. above ideal body Diabetic subjects who were >20% had weight22 significantly higher glyceraldehyde reductase levels (1-5 fold, p< 0’01) than diabetics with ideal body weight who had had diabetes for similar periods. However, there was no correlation between enzyme activity and raised body-weight in any of the diabetic groups or the normal control group (r< 0-25 in all cases). There was also no significant difference in the mean body-weight between any of the groups.

Discussion This report shows levels of

erythrocyte

an

association between increased reductase and the

glyceraldehyde

1270 presence of both diabetic retinopathy and cataract. Reports of abnormal constituents of urine and blood that are peculiar to cataract patients have never been substantiated,23,24 although recent studies suggest raised levels of plasma urea and

glucose, and lowered levels of albumin, total calcium, and cholesterol in cataract patients.25 Our kinetic studies show that the increase in mean level of enzyme activity observed in cataract and retinopathy patients is due to increased amounts of active enzyme rather than the elaboration of new kinetic pathways in disease. Thus, these results suggest that, although erythrocyte glyceraldehyde reductase need not necessarily be raised in ocular diabetic diseases or senile cataract, subjects with high levels (> 1-7 IU ml- 1) of glyceraldehyde reductase may be predisposed to the development of cataract and/or retinopathy, or they may get high enzyme levels as a result of the disease. Haemoglobin Ale blood levels, which may be used as a measure of diabetic control,26 do not correlate with glyceraldehyde reductase levels, and were not significantly different between diabetics with retinopathy only and diabetics with retinopathy and cataract. The presence of high levels of sugars does not affect the kinetics of aldose reductase in vitro (M.J.C. Crabbe and A. Basak Halder, unpublished). Human erythrocyte glyceraldehyde reductase resembles the bovine lens aldose reductase.14-17, 27 rather than the red cell NADP-1-gulonate dehydrogenase (E.C.1.1.1.19) in its chromatographic and kinetic properties with DLglyceraldehyde as substrate.28 Its inhibition by quercitrin is also similar to that of the bovine lens enzyme. Aldose reductase has been implicated in the aetiology of diabetic retinopathy, 29,30 and the evidence for its involvement in diabetic cataract is very strong. High aldose reductase levels allied to low polyol dehydrogenase levels have been found in human cataract lenses3l and diabetic rat lenses.3z Diabetic cataracts cannot be induced in the mouse, an animal which does not have measurable levels of aldose reductase in the lens33 or the blood (M.J.C. Crabbe and A. Basak Halder,

unpublished). Erythrocyte glyceraldehyde reductase, in common with lens aldose reductase, catalyses the reduction of glyceraldehyde-3-phosphate to glycerol-3-phosphate, which can then be used in phospholipid synthesis (M. J. C. Crabbe and H. H. Ting, unpublished). The increased levels of glyceraldehyde reductase activity seen in older

erythrocytesl5," would account for the increase in amount of membrane and changes in membrane conformation found in ageing erythrocyte membranes.34,35 Similar changes may occur in diabetics with retinopathy where high enzyme levels may contribute to changes in red-cell elasticity and high blood viscosity. Since glyceraldehyde reductase is probably one of a family of similar aldose-reductase enzymes which may be involved in lipid synthesis, activity levels of this enzyme may reflect those of similar enzymes in other tissues, and this is currently under investigation in our laboratory. Increases in levels of such enzymes may also contribute to hypertriglyceridaemia in diabetics,36,37 and it is interesting that streptozotocin-diabetic rats given a high-fat diet do not get cataracts even though sorbitol levels remain high. 38-40. Monitoring of the levels of glyceraldehyde reductase may be very useful in the clinical assessment of aldose reductase inhibitors. We thank Wolfson College, Oxford, the Wellcome Trust, the Oxford Area Health Authority, and Fight for Sight, Inc., U.S.A. for financial support; Dr Tim Dornan for clinical assistance and advice; Mr Philip Poon and Ms Cathy Whittle for the haemoglobin assays; Dr Robert Turner and Dr John Harding

for helpful conversations; and Mr Anthony Bron for support, encouragement, provision of laboratory facilities.

and

Requests for reprints should be addressed to M. J. C. C., University of Oxford, Nuffield Laboratory of Ophthalmology, Walton Street, Oxford, OX2 6AW. REFERENCES

polyols by the lens of the rat with sugar cataract Heyningen Nature 1959; 184: 194-95. Crabbe MJC, Basak Halder A. The kinetics of bovine lens aldose reductase under defined assay conditions. Clin Biochem 1979; 12: 281-83. Attwood MA, Doughty CC. Purification and properties of calf liver aldose reductase. Biochim Biophys Acta 1974; 370: 358-68. Hers HG. Le metabolism de fructose. Brussells. Editions Arcia, 1957. Clements RS, Winegrad AI. Purification of alditol: NADP oxidoreductase from human placenta. Biophys Res Comm 1972; 47: 1473-79. Moonsammy GI, Stewart MA. Purification and properties of brain aldose reductase and L-hexonate dehydrogenase. J Neurochem 1967; 14: 1187-93. Clements RS, Winegrad AI. The polyol pathway in sciatic nerve. Biochem Biophys Res Commun 1969; 36: 1006-109 Corder CN, Collins JG, TS, Sharma J. Aldose reductase and sorbitol dehydrogenase in rat kidney. J Histochem Cytochem 1977; 25: 1-8. Winegrad AI, Morrison AD, Clements RS. Alterations in the composition and metabolism of the inner aortic wall associated with increased polyol pathway activity. Horm Metals Res 1972; 4 (suppl) 169-71. van Heyningen, R. Galactose cataract: a review. Exptl Eye Res 1971; 11: 415-28 Kinoshita JH, Merola LO, Dikmak E. Osmotic changes in experimental galactose

1. Van

2. 3. 4. 5. 6. 7.

8. 9.

10. 11.

R. Formation of

Brannan

cataract.

Exptl Eye Res 1962; 1: 405-10.

12. Kinoshita JH. Cataracts in galactosaemia. Invest Ophthalmol 1965; 4: 786-99. 13. Varma SD, Shocket SS, Richards RD. Implications of aldose reductase in cataractsin human diabetes. Invest Ophthalmol Vis Sci 1979; 18: 237-41. 14. Crabbe MJC, Basak Halder A. Affinity chromatography of bovine lens aldose reductase and a comparison of some kinetic properties of the enzyme from lens and human erythrocyte. Biochem Soc Trans 1980; 8: 194-95 15. Basak Halder A, Wolff SP, Ting HH, Crabbe MJC. An aldose reductase from the human erythrocyte. Biochem Soc Trans 1980; 8: 652-53. 16. Crabbe MJC, Wolff SP, Basak Halder A, Ting HH. Diabetic cataracts-Is aldose reductase important? Metab Pediatr Ophthalmol (in press). 17. Crabbe MJC, Basak Halder A, Wolff SP, Ting HH. The role of aldose reductase in diabetic complications. In: Burlina A, Galzigna L, eds Clinical Enzymology Symposia: 3. Milan Piccin Books, (in press). 18. Agar NS. The activity of sorbitol dehydrogenase in some mammalian erythrocytes. Expertentia 1979; 35: 790-91. 19. Barretto OCO, Beutler E. The sorbitol oxidising enzyme of red blood cells J Lab Clin Med 1975; 85: 645-49. 20. Collins JG, Corder CN. Aldose reductase and sorbitol dehydrogenase in normal and diabetic rat lens. Invest Ophthalmol Vis Sci 1977; 16: 242-43 21. Varma SD, Mizuno A, Kinoshita JH. Diabetic cataracts and flavonoids. Science 1977; 195: 205-06. 22. Metropolitan Life Assurance Co. Ideal body weight tables. Statis Bull 1959; 40: 1. 23. Testa M, Fiore C, Bocci N, Calabro S. Senile cataract: serum factors producing irreversible precipitation ofa lens protein. Exptl Eye Res 1969; 8: 461-66 24. Harding JJ, Dilley KJ. Structural proteins of the mammalian lens: a review with emphasis on changes in development, aging and cataract. Exptl Eye Res 1976, 22: 1-74. 25. Bartholomew RS, Clayton RM, Cuthbert J, et al. Analysis of individual cataract patients and their lenses: preliminary observations on a population basis In Regnault F, Hockwin O, Courtois Y, eds. Ageing of the lens. Elsevier/North Holland, Amsterdam, 1980 241-61. 26. Koenig RJ, Peterson CN, Jones RL, Saudek C, Lehrman M, Cerami A. Correlation of in diabetes mellitis. N Engl J Med 1976, glucose regulation and haemoglobin 295: 417-20. 27. Sheaff CM, Doughty CC. Physical and kinetic properties of homogenous bovine lens aldose reductase. J Biol Chem 1976; 251: 2696-702. 28. Beutler E, Guinto E. The reduction of glyceraldehyde by human erythrocytes L-hexonate dehydrogenase activity. J Clin Invest 1974; 53: 1258-64. 29. Buzney SM, Frank RN, Varma SD, Tanishima T, Gabbay KH. Aldose reductase in retinal mural cells. Invest Ophthalmol Vis Sci 1977; 16: 392-96. 30. Boot-Handfond RP, Heath H. Fructose as a retinopathic agent IRCS Med Sci 1977;5: 583. 31. Pine A, van Heyningen R. The effect of diabetes on the content of sorbitol, glucose, fructose and inositol in the human lens Exptl Eye Res 1964; 3: 124-31. 32. Kinonhita JH, Fukushi S, Kador P, Merola LO. Aldose reductase in diabetic complications of eye. Metabolism 1979; 28: 462-69. 33. Kuck JFR. Response ofthe mouse lens to high concentrations of glucose and galactose Ophthalmol Res 1970; 1: 166-74. 34. Allan D, Michell RH. Accumulation of 1,2 diacylglycerol in the plasma membrane may lead to echinocyte transformation of erythrocytes Nature 1975, 253: 348-49. 35. Allan D, Thomas P, Michell RH. Rapid transbilayer diffusion of 1,2 diacylglycerol and its relevance to control of membrane curvature. Nature 1978; 276: 298-90 36. Albrink MJ. Dietary and drug treatment of hyperlipidaemia in diabetes. Diabetes 1974, 23: 913-18. 37. Weiland D, Mondon CE, Reaven GM. Evidence for multiple causality in the development of diabetic hypertriglyceridaemia. Diabetologia 1980; 18: 335-40. 38. Patterson JW, Patterson ME, Everett-Kinsey V, Reddy, DVN Lens assays on diabetic and galactosaemic rats receiving diets that modify cataract development Invest Ophthalmol 1965; 4: 98-103. 39. Hutton JC, Schofield PH, Williams JF, Regtop HL, Hollows FC. The effect of an unsaturated-fat diet on cataract formation in diabetic rats. Br J Nutr 1976, 36: 161-77 40. Baxter LCA, Schofield PJ. The effects of a high fat diet on chronic streptozotocindiabetic rats. Diabetologia 1980; 18: 239-45.

A1c