ELSEVIER

ORIGINAL ARTICLES

Ashok Pate1 Martin L. Hibberd Beverley A. Millward Andrew G. Demaine

ABSTRACT

Aldose reductase (ALR2), the first enzyme of the polyol pathway, may plan an important role in the pathogenesis of diabetic microvascular complications. The gene coding for ALR2 has been localized to chromosome 7q35. Using an ALR2 probe in conjunction with the restriction endonuclease Barn-HI, we have investigated the ALR2 locus of 128 patients with type I diabetes. A significant decrease in the frequency of the 8.2 kilobase (kb) Barn-HI ALR2 genotype and 8.2 kb allele was found in patients with nephropathy (nephropaths) compared to those with retinopathy alone (retinopaths) @ < 0.05 and 0.25, respectively). We have previously shown that an RPLP of the T-cell antigen receptor constant B-chain (TCRBC) locus, which is also localized to chromosome 7q35, is strongly associated with

INTRODUCTION

T he reason why only some patients with type I diabetes develop microvascular complica- tions, such as retinopathy and nephropathy, is unclear. Although prolonged exposure to

hypergylcemia is the primary factor associated with the development of most of these complications,1-3 additional risk factors are also required. Good meta- bolic control can help to slow their relentless progres- sion in some patients but does not prevent them en-

Department of Medicine, Faculty of Postgraduate Medicine, Davy Building, University of Plymouth, Plymouth, United Kingdom.

Reprint requests to be sent to: Dr. A.G. Demaine, Department of Medicine, Faculty of Postgraduate Medicine, Davy Building, Uni- versity of Plymouth, Plymouth, PL4 SAA, United Kingdom.

]ournal of Diabetes and Ifs Complications 1996; 10;62-67 0 Elsevier Science Inc., 1996 655 Avenue of the Americas, New York, NY 10010

Chromosome 7q35 and Susce Diabetic Microvascular P

tibility to Comp ications

susceptibility to microvascular complications. The 128 patients were genotyped using the restriction endonuclease Bgl-II and a TCRBC probe. The 101 9.2-8.2 kb TCRBC-ALR2 genotype was significantly decreased in the nephropaths compared to the retinopaths (13.7% versus 43.6%, x2 = 10.1, p < 0.0025). The 10/9.2 and 9.219.2 kb TCRBC-ALR2 haplotypes were increased in the nephropaths compared to the retinopaths (32.5% versus 8.9% x2 = 10.9, p < 0.001). These results suggest that chromosome 7q35 harbors a gene(s) that is involved in the pathogenesis of microvascular complications. Interestingly, the gene coding for endothelial nitric oxide synthase has recently been localized to the same chromosomal region as ALR2. (Journal of Diabetes and Its Complications 10:2:62-67, 1996.)

tirely, suggesting that genetic determinants of tissue susceptibility as well as environmental factors are im- portant in the predisposition to microvascular compli- cations.

Aldose reductase (ALR2; EC 1.1.1.21) is the first en- zyme in the polyol pathway and it catalyzes the NADPH-dependent reduction of hexose sugars to their corresponding alcohols.4 It is part of a family of monomeric aldo-keto reductases, all between 35 and 40 kilodaltons, that show broad substrate specificities for aldehydes and ketones, including xenobiotics as well as endogenous compounds. It is expressed in the cell cytoplasm of a number of human tissues including lens, kidney, liver, placenta, and brain. While under normal physiological conditions, this pathway is not thought to play a major role in glucose metabolism; when cells are placed under hyperglycemic condi-

1056.8727/96/$15.00 SSDI 1056-8727(95)00004-L

] Diab Camp 1996; 10:62-67 CHROMOSOME 7q35 AND DIABETIC COMPLICATIONS 63

tions, significant quantities of sorbitol can be pro- duced. As sorbitol does not readily diffuse out of the cell and its conversion to fructose is slow, the resulting intracellular accumulation of sorbitol may cause hy- perosmotic stress leading to loss of cellular integrity and tissue damage. 4,5 Alternatively, the increased flux through the polyol pathway may lead to decreased levels of NADPH, which will reduce the concentration of glutathione and result in increased oxygen free radi- cal activity and reduced nitric oxide activity.6-8

A number of aldose reductase inhibitors have been shown to reverse decreased axonal transport and im- prove nerve conduction velocity,9 to delay or prevent the onset of sugar cataracts,*o and to ameliorate glo- merular hyperfiltration” in animals with either chemi- cally induced or genetic diabetes. However, the effi- cacy of these drugs on the treatment of diabetic complications in patients has still to be fully demon- strated.12

Human ALR2 gene has been cloned from a number of tissues,13-i7 and strong homology has been found at the nucleotide and protein level to aldehyde reduc- tase, 2.5-diketogluconic acid reductase, and P-crystal- lin. This suggests that these proteins are evolutionarily related and are collectively known as the aldo-keto reductase supergene family.

The gene for the functional human ALR2 has been localized to chromosome 7q35 and consists of ten ex- ons extending over 18 kilobases (kB) of DNA.i6

In a previous study, we have shown that a polymor- phism of the T-cell antigen receptor P-chain (TCRBC) locus, which also maps to chromosome 7q35, is strongly associated with susceptibility to diabetic mi- crovascular complications. “Additionally, in a prelimi- nary study we have shown that a polymorphism of the ALR2 is also associated with microvascular compli- cations.” In this current study we have used probes to both TCRBC and the ALR2 loci in a population of patients with well-defined microvascular complica- tions.

METHODS

Subjects. One hundred and twenty-eight British caucasoid patients with type I diabetes mellitus as de-

fined by the National Diabetes Data Group20 were ob- tained from the Diabetic Clinic (Dr. B.A. Millward). Local ethical committee approval had been obtained. The patients were classified according to their micro- vascular complications as follows:

1. Uncomplicated patients (n = 38): patients who have had type I diabetes for at least 20 years but are free of retinopathy (fewer than 5 dots or blots per fun- dus) and nephropathy (urine Alblustix negative on three consecutive occasions over 12 months)

2. Nephropaths (n = 51): patients with type I diabetes for at least 10 years with persistent proteinuria (urine Albustix positive on at least three consecu- tive occasions over 12 months or three successive total urinary protein excretion rates greater than 0.5g/24h), in the absence of hematuria or infection on midstream urine samples; diabetic nephropathy was always associated with diabetic retinopathy

3. Retinopuths (n = 39): patients with type I diabetes and retinopathy and no proteinuria (urine Albustlx negative on three consecutive occasions over 12 months); diabetic retinopathy defined as more than five dots or blots per eye, hard or soft exudates, new vessels or fluorescein angiographic evidence of maculopathy or previous laser treatment or vi- trectomy for new vessels, maculopathy or vitreous hemorrhage.

Fundoscopy was performed by both a diabetologist and opthalmologist.

The clinical features of these groups are shown in Table 1.

Fifty-one British caucasoid blood donors with no ev- idence of diabetes or renal disease were used to obtain normal control frequencies of ALR2 alleles and geno- types. Normal control frequencies of the TCRBC al- leles and genotypes were obtained from studies pre- viously published by ourselves.18,21

Experimental Methods. High-molecular-weight DNA was prepared from peripheral blood and digested with the appropriate restriction endonuclease following the manufacturer’s protocol (Bethesda Research Labora- tories, Paisley, Scotland). The DNA fragments were

TABLE 1. CLINICAL FEATURES OF THE PATIENT SUBGROUPS

Uncomplicated Nephropaths Retinopaths

Male:Female Age at onset of disease (years) Duration of disease (years)

20:18 28:23 19:20 20.1 (151) 17.3 (3-40) 21.1 (440) 34.8 (20-65) 29.1 (13-49) 34.4 (17-60)

Mean figures and range (in brackets).

Uncomplicated, patients who have had type I diabetes for more than 20 years yet remain free of retinopathy and proteinuria; Nephropaths, patients

with diabetic nephmpathy (and coexistent retinopathy); and Retinopaths, patients with diabetic retinopathy without proteinuti.

64 PATELETAL. ] Diab Comp 1996; lo:6247

TABLE 2. FREQUENCY OF ALR2 GENOTYPES AND ALLELES IN PATIENTS WITH OR WITHOUT. MICROVASCULAR COMPLICATIONS

Uncomplicated n = 30

Patient Subgroup

Nephropaths Retinopaths Controls n = 51 n = 39 n = 51

Genotype (kb) 9.218.2 47.4 9.2 10.5 8.2 42.1

Allele (kb) 9.2 0.342 8.2 0.658

y Nephropaths vs. retinopaths, x2 = 5.29, p < 0.05, PC = NS.

b Nephropaths vs. retinopaths x2 = 6.2, p < 0.025, PC = NS.

c Retinopaths vs. controls x2 = 6.1, p < 0.025, PC = NS.

d Retinopaths vs. controls x2 = 6.2, p < 0.025, PC = NS.

separated by size in 0.6% agarose gels transferred and immobilized in nylon membranes (Hybond-N + ) .

The membranes were hybridized with a denatured random-primer 32P-dCTP labeled probe for 16-20 h in a solution containing 6X SSC, 5X Denhardt’s solution, 5% dextran sulphate, 0.2 mg/mL denatured salmon sperm DNA at 65°C. After hybridization, the filters were washed in 0.2 x SSC, 0.5% SDS at 65°C for 3-45 min and subsequently placed between Cronex lightening plus intensifying screens with Kodak XAR5 film at -70°C. Films were developed after l-4 days.

DNA Probes. The ALR2 probe has been described previously,lg was synthesized using the polymerase chain reaction with oligonucleotide primers specific for genomic ALR2 gene. The TCRBC cDNA probe is derived from the Jurkat T-lymphoblastoid cell line.

The ALR2 probe in conjunction with Pst-I detects

52.9 38.5 56.9 9.8 0.p 7.8

37.3" 61.5 35.3'

0.363 0.193 0.363 0.637b 0.807 0.637d

polymorphic fragments of 9.2 and 8.2 kb.lg The TCRBC probe in conjunction with Bgl-II detects polymor- phisms of 10 and 9.2 kb.“,‘r

Statistical Analysis. The x2 test was used to compare the data obtained for the various groups. p values were corrected for the number of comparisons (PC). The TCRBC-ALR2 haplotype frequencies were obtained using subjects who were homozygous at either the TCRBC or the ALR2 locus or both.

RESULTS

The ALR2 probe in conjunction with Barn HI detects allelic fragments of 9.2 or 8.2 kb and genotypes of ei- ther 9.2; 8.2,9.2, or 8.2 kb. The frequency of the ALR2 alleles and genotypes in the patient and control popu- lations are shown in Table 2. The frequency of the 8.2 kb ALR2 genotype was increased in the retinopaths

TABLE 3. FREQUENCY (%) OF TCRBC GENOTYPES AND ALLELES IN PATIENTS WITH OR WITHOUT MICROVASCULAR COMPLICATIONS

Uncomplicated n = 38

Patient Subgroups

Nephropaths Retinopaths Controls n = 51 n = 39 n = 204

Genotype (kb) 10;9.2 63.1 45.P 69.2 42.2 10 5.3 21.6 7.7 34.3 9.2 31.6 33.3 23.1 24.5

Allele (kb) 10 0.368 0.441 0.423 0.461 9.2 0.632 0.559 0.577 0.539

’ Nephropaths vs. retinopaths x2 = 5.1, p < 0.025, PC = NS.

b Nephropaths vs. nonnephropaths, ,$ = 5.65, p < 0.025, PC = 0.05.

1 Dub Gnnp 1996; 10:62-67 CHROMOSOME 7q35 AND DIABETIC COMPLICATIONS 65

n

Chromosome

l-l lq35

TCRBC Allele

ALR2 Shorthand Allele Nomenclature

io 9.2 A

9.2 9.2 B

10 8.2 C

9.2 8.2 D

FIGURE 1. Schematic representation ofthe T-cell antigen recep- torf%chain (TCRBC) and aldose reductase (ALR2) huplotypes. The TCRBCallelesaredetected using the resfricfion endonuclease B&II and a TCRBC DNA probe. The ALR2 alleles are detected using the restriction endonuclease Barn-HI and an ALR2 probe.

compared to the nephropaths (61.5% versus 37.3%, respectively, p < 0.05); however, this was not signifi- cant after correcting for the number of parameters. The frequency of the 8.2 kb allele was increased in the retinopaths compared to the nephropaths (0.807 versus 0.637 respectively, p < 0.025, PC = NS). The frequency of the 8.2 kb ALR2 allele and genotype was also increased in the retinopaths compared to the nor- mal controls (x2 =6.2 and 6.1, respectively, p < 0.025, PC = NS). There were no other significant differences between any of the patient or control groups.

The TCRBC probe in conjunction with the restriction endonuclease Bgl-II detects allelic fragments of 10.0 and 9.2 kb.‘8,21 The frequencies of the TCRBC alleles and genotypes in the patients and controls are shown in Table 3. As expected from our previous study, there

was an increased frequency of the 10;9.2 kb TCRBC genotype in the retinopaths compared to the nephro- paths (69.2% versus 45.18%, respectively. p < 0.05). This difference was not significant after correction of the p value. There was also a significant difference in the frequency of the 10;9.2 kb TCRBC genotypes between nonnephropaths and nephropaths (x2 = 5.65, p < 0.025, PC = 0.05).

The TCRBC and ALR2 loci are both located on chro- mosome 7q35, and it is possible to identify a minimum of 9 TCRBC-ALR2 genotypes. These are derived from4 TCRBC-ALR2 haplotypes: 10/9.2, 9.2/9.2, 10/8.2, and 9.2/8.2, which are designated ‘A,’ ‘B,’ ‘C,’ and ‘D,’ re- spectively (Figure 1). Of the nine possible genotypes, eight were detected in the patient population (Table 4). There were four commonTCRBC-ALR2 genotypes, which accounted for up to 90% of the genotypes. There was a significant decrease in the frequency of the ‘CID’TCRBC-ALR2 genotype in the nephropaths com- pared to the retinopaths (13.7% versus 43.6%, respec- tively, x2 = 10.1, p < 0.0025, PC = 0.03). This was accompanied by an increase of the ‘A/C’ TCRBC-ALR2 genotype in the nephropaths compared to the retino- paths (15.7% versus 2.6%, respectively). There were no other significant differences between the patient groups.

The frequency of the TCRBC-ALR2 haplotypes in the patient groups was determined in the subjects who were homozygous at either the TCRBC or ALR2 locus. It was possible to assign a total of 192 TCRBC-ALR2 haplotypes. The results are shown in Table 5. There was a significant increase in the frequency of both the ‘A’ and ‘B’ TCRBC-ALR2 haplotypes in the nephro- paths compared to the retinopaths (15.0% and 17.5%

TABLE 4. FREQUENCY OF TCRBC-ALR2 GENOTYPES IN PATIENTS WITH OR WITHOUT MICROVASCULAR COMPLICATIONS

Patient Subgroup

Uncomplicated n = 38

Nephropaths n = 51

Retinopaths n = 39

TCRBC-ALR2 genotypes (Shorthand) 10;9.2/9.2;8.2 A/D 10;9.2/9.2 AIB 10;9.2/8.2 C/D

10/9.2;8.2 AIC 1019.2 AIA lOi8.2 c/c

9.2/9.2;8.2 B/D 9.219.2 BIB 9.218.2 D/D

28.9 21.6 25.6 5.3 7.8 0.0

28.9 13.7” 43.6 5.3 15.7 2.6 0.0 0.0 0.0 0.0 7.8 5.1

13.1 15.7 10.2 5.3 2.0 0.0

13.1 15.7 12.8

The letters “A,B,C,D” refer to the shorthand nomenclature for describing the TCRBC-ALR2 hnplotypes shown in Figure 1.

a Frequency of C/D TCRBC-ALR2 genotype in nephropuths vs. retinopaths x2 = 70.1, p < 0.0025, PC = 0.03. There were no other significant

differences between any of the groups.

66 PATELETAL. J Diub Comp 1996; 10:62-67

TABLE 5. FREQUENCY OF TCRBC-ALRZ HAPLOTYPES IN PATIENTS WITH OR WITHOUT MICROVASCULAR COMPLICATIONS

Patient Subgroup

Uncomplicated Nephropaths tl=54 n = 80

TCRBC-ALR2 haplotypes 10/9.2 (A) 7.4 15.0"

9.2/9.2 (B) 20.4 17.5" 1018.2 (C) 24.1 28.7

9.2/8.2 (D) 48.1 38.8

The letters in brackets refer to the shorthand nomenclature for describing the haplotypes (Figure I).

n, number of chromosomes.

*frequency of Wand ‘B’ TCRBC-ALR2 huplotypes in nephropuths vs. retinopuths x2 = 10.93, p < 0.001, PC = 0.007.

Retinopaths n = 58

1.7 7.2

37.9 53.4

versus 1.7% and 7.2%, respectively, x2 = 10.93, p <

0.001, PC = 0.007). The ‘D’ TCRBC-ALR2 haplotype was the most common haplotype in the retinopaths as well as in the uncomplicated groups (53.4% and 4&l%, respectively).

DISCUSSION

It has previously been suggested that ALR2 may con- tribute toward the pathogenesis of diabetic microvas- cular complications. 4,5,9-11~22-24 Many of these data, however, have been derived from animal models of diabetes and from long-term cell cultures. Now that the gene for ALR2 has been cloned and sequenced, its importance in diabetic microvascular complications in humans can be addressed. We have attempted to answer the question of whether the ALR2 locus con- fers susceptibility, by identifying polymorphisms of the gene and analyzing their frequency in a clinically well-defined patient population. These results suggest that a gene or genes in the region of ALR2 contributes to the susceptibility to diabetic complications and ex- tend our previous results in using probes to this re- gion.18,19 Indeed, a combination of markers of the ALR2 and TCRBC loci was extremely informative in dis- secting ‘susceptibility’ genotype and haplotypes. The results suggest that the genetic susceptibility to dia- betic nephropathy may not be the same as that to dia- betic retinopathy, while the uncomplicated group ap- peared to be different from either the nephropaths or retinopaths. These patients are relatively infrequent and may have a genetic background that ‘protects’ from complications.

Patients with type I diabetes (and type II diabetes) and microvascular complications have a higher eryth- rocyte and neutrophil activity of ALR2 than those without complications. 25-28 The data presented here suggest that either there are different isoforms of ALR2 that have different activities or there is a polymor- phism in the regulatory region of the gene that may result in increased or decreased expression of the

gene. Alternatively, the association may be due to link- age with another gene. For instance, it is possible that the immune system may be involved, and this re- flected in the association with the TCRBC locus. The TCRBC locus has previously associated with suscepti- bility to a number of autoimmune diseases,29,30 and there have been reports showing that patients with microvascular complications have increased levels of activated T-lymphocytes and immune complexes.31,32 Finally, it is extremely interesting that the gene coding for the constitutive endothelial nitric oxide synthase (NOS) has also been localized to chromosome 7q35.33 In experimental models of microvascular disease, it has been suggested that a decreased availability of ni- tric oxide may contribute toward the ischemia present in the damaged tissue. 34,35 ALR2 and NOS are both NADPH-requiring enzymes, therefore, if ALR2 binds NADPH at a higher affinity than NOS, it would prefer- entially convert glucose to sorbitol. This would result in reduced amounts of nitric oxide. The localization of the genes for both these enzymes on chromosome 7q35 may be chance and, in evolutionary terms, a ‘re- cent event.’ Alternatively, there may be good physio- logical reasons for having the two enzymes in rela- tively close proximity.

In conclusion, these results suggest that chromo- some 7q35 contains a gene that appears to be involved in the susceptibility to diabetic microvascular compli- cations. Further work is now required to determine the exact relationship of the TCRBC, ALR2, and NOS genes to each other and whether the enzymatic activity of ALR2 and NOS are related to polymorphisms in the DNA.

ACKNOWLEDGMENT

This work was funded by the British Council for the Preven- tion of Blindness and the British Diabetic Association.

REFERENCES

1. Brownlee M: Gycosylation products as toxic mediators of diabeticcomplications. Annu. RevMed 42:159-166,199O.

1 Diab Comp 2996; 20:62-67 CHROMOSOME 7q35 AND DIABETIC COMPLICATIONS 67

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

Larkins RG, Dunlop ME: The link between hyperglycae- mia and diabetic nephropathy. Diabetologia 35:499-504, 1992.

The DCCT Research Group: The effect of intensive treat- ment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329:977, 1993.

Dvornik D: Hyperglycemia in the pathogenesis of dia- betic complications, in: Porte D (ed), Aldose Reductase Inhibition: An Approach to the Prevention ofDiabetic Compli- cations. New York, McGraw-Hill, pp. 7-151, 1987.

Raskin P: Risk factors for the development of diabetic complications. J Diabet Complications 8:195-200, 1994.

Baynes JW: Role of oxidative stress in the development of complications of diabetes. Diabetes 40:405-412, 1991.

Corbett JA, Tilton RG, Chang K, Hassan KS, Ido Y, Wang JI, et al.: Aminoguanidine, a novel inhibitor of nitric oxide formation, prevents diabetic vascular dys- function. Diabetes 41:552-556, 1992.

McVeigh GE, Brennan GM, Johnston GD, McDermott BJ, McGrath LT, Henry WR, et al.: Impaired endothe- lium-dependent and independent vasodilation in pa- tients with type 2 (non-insulin-dependent) diabetes mellitus. Diabetologia 35:771-776, 1992.

Tomlinson DR, Moriarty RJ, Mayer JH: Prevention and reversal of defective axonal transport and motor nerve conduction velocity in rats with experimental diabetes by treatment with the aldose reductase inhibitor sor- binil. Diabetes 33:470-476, 1984.

Kinoshita JH, Kador I’, Datiles M: Aldose reductase in diabetic cataracts. JAMA 246:257-261, 1981.

Goldfarb S, Simmons DA, Kern E: Amelioration of glo- merular hyperfiltration in acute experimental diabetes by dietary myo-inositol and aldose reductase inhibition. Trans Assoc Am Physicians 99:67-72, 1986.

Masson EA, Boulton AJM: Aldose reductase inhibitors in the treatment of diabetic neuropathy: A review of the rationale and clinical evidence. Drugs 39:190-202, 1990.

Bohren KM, Bullock B, Wermuth B, Gabbay KH: The aldo-keto reductase superfamily. cDNAs and deduced amino acid sequences of human aldehyde and aldose reductases. J Biol Chem 264:9547-9551, 1989.

Chung S, LaMendola J: Cloning and sequence determi- nation of human placental aldose reductase gene. I Biol Chem 264:14775-14777, 1989.

Nishimura C, Matsuura Y, Kokai Y, Akera T, Carper D, Morjana N, et al.: Cloning and expression of human aldose reductase. J Biol Chem 265:9788-9792, 1990.

Graham A, Heath I’, Morten JEN, Markham AF: The human aldose reductase gene maps to chromosome 7q35. Hum Genet 86:509-514, 1991.

Graham A, Brown L, Hedge PJ, Gammack AJ, Markham AF: Structure of the human aldose reductase gene. J Biol Chem 266:6872-6877, 1991.

Hibberd ML, Millward BA, Wong FS, Demaine AG: T-cell receptor constant B-chain polymorphisms and susceptibility to type I diabetes. Diabetic Med 9:929-933, 1992.

Pate1 A, Ratanachaiyavong S, Millward BA, Demaine AG: Polymorphisms of the aldose reductase locus

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

(ALR2) and susceptibility to diabetic microvascular com- plications. Adv Exp Mel Biol 328:325-332, 1993.

National Diabetes Data Group: Classification and diag- nosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 28: 1039-1044, 1979.

Millward BA, Welsh KI, Leslie RDG, l’yke DA, Demaine AG: T-cell receptor beta chain polymorphisms are asso- ciated with insulin dependent diabetes mellitus. Clin Exp lmmunol 70:152-157, 1987.

Gabbay KH: The sorbitol pathway and the complica- tions of diabetes. N Engl J Med 288:831-836, 1973.

Winegrad AI: Does a common mechanism induce the diverse complications of diabetes? Diabetes 36:396-406, 1987.

Kinoshita JH, Nishimura C: The involvement of aldose reductase in diabetic complications. Diabetes Metab Rev 4:323-354, 1987.

Dent MT, Tebbs SE, Gonzalez AM, Ward JD, Wilson RM: Neutrophil aldose reductase activity and its associa- tion with established microvascular complications. Diu- betic Med 8:4349-442, 1991.

Hamada Y, Kitoh R, Raskin I’: Crucial role of aldose reductase activity and plasma glucose level in sorbitol accumulation in erythrocytes from diabetic patients. Di- abetes 40:1233-1240, 1991.

Dent MT, Veves A, Tebbs SE, Gonzalez AM, Malik RA, Boulton AJM, et al. : Neutrophil aldosereductase activity as a potential marker for neuropathy and cataract in diabetes. Diabetic Med 8:911-916, 1991.

Hamada Y, Kitoh R, Raskin I? Association of erythrocyte aldose reductase activity with diabetic complications in type I diabetes mellitus. Diabetic Med 10:33-38, 1993.

Kronenberg M, Siu G, Hood LE, Shastri N: The molecu- lar genetics of the T-cell antigen receptor. Annu Rev lm- munol5:529-591, 1986.

Moss PAH, Rosenberg WMC, Bell JI: The human T cell receptor in health and disease. Annu Rev lmmunol 10: 71-96, 1992.

Bending JJ, Lobo-Yeo A, Vergani D, Viberti GC: Protein- uria and activated T-lymphocytes in diabetic nephropa- thy. Diabetes 37:507-511, 1988.

Di Mario U, Borsey DQ, Contreas G, Prowse CV, Clarke BF, Andreani D: The relationship of soluble immune complexes, insulin antibodies and insulin anti-insulin complexes to platelet and coagulation factors in type I diabetic patients with or without proliferative retinopa- thy. Clin Exp Immunol 65:57-65, 1986.

Marsden PA, Heng HHQ, Scherer SW, Stewart RJ, Hall AV, Shi X-M, Tsui L-C, Schappert KT: Structure and chromosomal localisation of the human constitutive en- dothelial nitric oxide synthase gene. J Biol Chem 268: 17478-17488, 1993.

Oyama Y, Kawasaki H, Hattori Y, Kanno M: Attenua- tion of endothelium-dependent relaxation in aorta from diabetic rats. Eur J Pharmacol 131:75-78.

Kamata K, Miyata N, Kasuya Y: Impairment of endothe- hum-dependent relaxation and changes in levels of cy- clic GMP in aorta from streptozotocin-induced diabetic rats. Br ] Pharmacol 97:614-618, 1989.