UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Oxygen free radical scavenging systems in clinical and experimental (chemical and spontaneous) diabetes… Wohaieb, Saleh A. 1987

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1987_A1 W62.pdf [ 9.48MB ]
Metadata
JSON: 831-1.0097524.json
JSON-LD: 831-1.0097524-ld.json
RDF/XML (Pretty): 831-1.0097524-rdf.xml
RDF/JSON: 831-1.0097524-rdf.json
Turtle: 831-1.0097524-turtle.txt
N-Triples: 831-1.0097524-rdf-ntriples.txt
Original Record: 831-1.0097524-source.json
Full Text
831-1.0097524-fulltext.txt
Citation
831-1.0097524.ris

Full Text

OXYGEN FREE RADICAL SCAVENGING SYSTEMS IN CLINICAL AND EXPERIMENTAL (CHEMICAL AND SPONTANEOUS) DIABETES MELLITUS By SALEH A. WOHAIEB B.V.M., The U n i v e r s i t y o f Baghdad, 1975 M.Sc., The U n i v e r s i t y o f Baghdad, 1977 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE-OF DOCTOR OF PHILOSOPHY i n THE FACULTY OF GRADUATE STUDIES (Department o f Pharmacology & T h e r a p e u t i c s , F a c u l t y o f M e d i c i n e ) We a c c e p t t h i s t h e s i s as c o n f o r m i n g t o t he r e q u i r e d s t a n d a r d THE UNIVERSITY OF BRITISH COLUMBIA June 1987 ©-•Saleh A. Wohaieb, 1987 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of Brit ish Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the Head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Sal eh A. Wohaieb Faculty of Medicine Department of Pharmacology & Therapeutics The University of Brit ish Columbia Vancouver, Brit ish Columbia, Canada V6T 1W5 i ABSTRACT The extent to which endogenous free radical-scavenging defense mechan-isms are involved in experimental and human diabetes was investigated in various tissues of animals with chemically-induced or spontaneous diabetes (BB Wistar rats) and in erythrocytes of patients with either Type I or Type II diabetes. Diabetes was induced in female Wistar rats using alloxan (ALX) or streptozotocin (STZ), each administered in a dose of 50 mg/kg body wt., intravenously. The present study also included a group -of animals in which body wt. loss was induced by food-deprivation for 72 h. The effects of pharmacological interventions (insulin or allopurinol (ALP)), on these processes were also investigated in chemically-induced diabetes., The act iv-i t ies of catalase (CAT), CuZn-superoxide dismutase (CuZn-SOD), glutathione peroxidase (GSH-PX) and glutathione reductase .(GSSG-RD) as well as levels of reduced glutathione (GSH) were examined in heart, pancreas, l iver and kidney as well as in erythrocytes. Erythrocytes were also examined for their sus-cept ib i l i ty to in vitro oxidative stress induced by hydrogen peroxide (H2O2). Cr iter ia studied in this regard were GSH-depletion and malondi-aldehyde (MDA) production (an index of l ip id peroxidation). The results obtained showed that tissue antioxidant systems are altered in experimental diabetes and that the magnitude of the alterations increased with the degree of body weight loss. Furthermore, the duration of hypoinsulinemia might contribute to the nature of alterations in antioxidant mechanisms. The complex patterns of the alterations observed varied from one tissue to another and may be the result of compensatory increases, usually involving enzymes whose act iv ity in the particular tissue may be l imiting, and direct inhibitory effects of endogenous oxidants on the enzy-matic components of tissue antioxidant systems. The ab i l i ty of insulin (9-12 U/kg body wt., subcutaneously) to reverse i i the many similar alterations of tissue antioxidant enzymes in diabetes induced by either STZ or ALX suggests that these changes are more l ikely attributable to hypoinsulinemia rather than to direct effects of either diabetogenic drug. The above-mentioned effects indicate that insulin can markedly influence tissue antioxidant status. However, the reason for the persistence of decreased CuZn-SOD act iv ity in both l iver and kidney of ALX-diabetic rats after 12 wk of treatment with insulin is not clear at present, and requires further investigation to determine whether this reflects the presence of a residual def ic i t in tissue antioxidant processes in l iver and kidney despite insulin treatment, or whether i t is the result of a direct effect exerted by ALX. Acute ALP administration (50 mg/kg body wt., intraperitoneally) was associated with reductions in ketonuria and early mortality among ALX-diabetic rats, and long-term ALP treatment (1.9 mg/day in drinking water) resulted in a normalization of renal CuZn-SOD act iv ity in these animals. Comparable (although not identical) changes in tissue antioxidant status are present in insulin-dependent spontaneously diabetic BB (ISDBB) rats and in animals made diabetic by STZ or ALX administration. Our data also demonstate that the alterations in tissue GSH levels characterizing ALX-diabetes more closely paralleled changes seen in the ISDBB rat than did those in the diabetic state induced by STZ. If the alterations in antioxi-dant status in uncontrolled chemically-induced diabetes are attributable to a lack of insul in, the observed changes in ISDBB rats are suggestive of sub-optimal insulin therapy in these animals. The results obtained from BB rats demonstrate two types of alterations in antioxidant status: strain-related differences (increased CAT activity in pancreas and decreased GSH levels in pancreas and l iver of both ISDBB and their non-diabetic littermates (NDLM)) and diabetes-related changes (mani-i i i fested by an increase in cardiac GSH content and increases in act iv i t ies of cardiac CAT and GSSG-RD, pancreatic CuZn-SOD and GSSG-RD, and renal GSH-PX). Whether or not these "strain-related" alterations in antioxidant status increase the susceptibi l ity of these animals to developing diabetes remains unknown. Certain alterations were observed in red cel l s from diabetic patients and from animals with experimental diabetes suggesting that these a ltera-tions are more l ike ly to be diabetes-related than species-dependent. Red cel ls in chemically-induced and c l in ica l diabetes showed an increased res is-tance to peroxide-induced depletion of GSH, an effect attributed to hyper-glycemia, which results in an increased supply of NADPH through the hexose monophosphate shunt for regeneration of GSH from GSSG via the GSSG-RD system. However, the susceptibi l ity of red cel l s from diabetic patients and animals to l i p id peroxidative damage was increased as reflected in augmented MDA production. In addition, insulin treatment did not normalize MDA production in red ce l l s subjected to oxidative challenge and vigorous insulin treatment in both ALX- and STZ-diabetic rats resulted in a markedly decreased MDA production in response to H2O2. Moreover, GSSG-RD act iv-i ty of red cel ls was increased in both uncontrolled and insulin-treated diabetic animals as well as in diabetic patients. However, some differences in erythrocyte antioxidant enzymes were also observed in erythrocytes from diabetic subjects and animals. For example, diabetic patients showed an increased act iv ity of CuZn-SOD, while erythro-cytes from diabetic animals showed no alterations in the activity of this enzyme. Erythrocyte membrane NADH-dehydrogenase act iv i ty was increased only in diabetic patients with Type I diabetes, but not in Type II diabetes or in diabetic animals. Erythrocytes from ALX- and STZ-diabetic animals showed an increase in the activity of GSH-PX and those from NDLM BB rats showed a decrease in CAT act iv i ty, alterations that were not observed in human diabetes. Final ly, as far as antioxidant defense mechanisms are concerned, our results suggest that diabetes is associated with some common alterations in these mechanisms regardless of the model (chemically-induced versus the spontaneous type of diabetes) or the species used (animal versus human diabetes). Some of these alterations seem to be influenced by the degree of diabetic control, while others are apparently independent of i t . Future studies wil l focus on the extent to which alterations in red cel ls of human diabetics can be used to predict the development of long-term sequelae of the disease. V TABLE OF CONTENTS Chapter Page ABSTRACT i TABLE OF CONTENTS v LIST OF TABLES v i i i LIST OF FIGURES x i i LIST OF APPENDICES x i i i LIST OF ABBREVIATIONS xiv ACKNOWLEDGMENTS xv DEDICATION xvi 1. INTRODUCTION 1 1.1 Diabetes Mellitus 1 1.2 Diabetic Complications 1 1.3 Reactive Oxygen Radicals and Diabetic Complications 6 1.4 Free Radical Metabolism and Pancreatic p-Cell Function 10 1.5 Alterations in Free Radical Scavenging Enzymes in Cl inical Diabetes 14 1.6 The Role of Free Radicals in Chemically-Induced Diabetes 16 1.7 Alterations in Oxygen Radical Scavenging Systems in Chemically-Induced Diabetes 21 1.8 Spontaneously Diabetic BB Rat (SDBB) 23 1.9 Antioxidant Status in SDBB 25 1.10 Lipid Peroxidation in Diabetes 25 1.11 Non-Enzymatic Antioxidants in Diabetes 26 1.12 Rationale and Objectives of the Study 28 2. MATERIALS AND METHODS 33 2.1 Animal Studies 33 2.1.1 Induction of diabetes 33 2.1.2 Insulin treatment 33 2.1.3 Allopurinol treatment 34 vi 2.1.4 Food-deprivation 34 2.1.5 Spontaneously diabetic male Wistar BB rats 35 2.2 Human Studies 35 2.2.1 Studies of erythrocytes from diabetic patients 35 2.3 Preparation of Cytosolic Fractions 36 2.4 Preparation of Erythrocyte Hemolysates 37 2.4.1 Hemoglobin assay 37 2.5 Tissue Free Radical Scavenging Enzymes 37 2.5.1 Catalase 37 2.5.2 CuZn-superoxide dismutase 38 2.5.3 Glutathione peroxidase 39 2.5.4 Glutathione reductase 39 2.5.5 Tissue sulfhydryl content 40 2.5.6 Correction for contamination caused by blood 40 2.6 Erythrocyte Membrane Preparation 41 2.6.1 NADH-dehydrogenase 41 2.6.2 Protein assay 42 2.6.3 Cholesterol content 42 2.6.4 Phospholipid phosphorus content 42 2.7 Erythrocyte Membrane and Plasma Phospholipid Profi les 43 2.8 Measurement of Erythrocyte Susceptibil ity to Hydrogen Peroxide-Induced Oxidative Stress 44 2.8.1 Reduced glutathione (GSH) assay 44 2.8.2 Malondialdehyde (MDA) assay 44 2.9 Chemical Assays 45 2.9.1 Glucose, insulin and l ip id measurements 45 2.9.2 Glycosylated hemoglobin assay 46 2.9.3 Water content 47 3. STATISTICAL ANALYSIS 47 4. RESULTS 47 4.1 Distribution of Tissue Antioxidants 47 4.2 Chemically-Induced Diabetes in Female Rats 53 4.2.1 General features 53 4.2.2 Tissue antioxidant status 56 4.2.3 Erythrocyte enzymatic changes 66 4.2.4 Erythrocyte susceptibi l ity to peroxidative damage 66 4.2.5 Erythrocyte membrane and plasma l ip id analyses 73 4.3 Effects of Allopurinol Treatment in ALX-Diabetic Animals 80 4.3.1 General features 80 4.3.2 Tissue antioxidant status 80 4.3.3 Erythrocyte enzymatic changes 80 4.3.4 Erythrocyte susceptibi l ity to peroxidative damage 86 vi i 4.3.5 Erythrocyte membrane and plasma l ip id analyses 85 4.4 Food-Deprivation 86 4.4.1 General features 86 4.4.2 Tissue antioxidant status 86 4.4.3 Erythrocyte enzymatic changes 96 4.4.4 Erythrocyte susceptibi l ity to peroxidative damage 96 4.4.5 Erythrocyte membrane and plasma l ip id analyses 96 4.5 Spontaneously Diabetic BB Wistar Rats 105 4.5.1 General features 105 4.5.2 Tissue antioxidant status 105 4.5.3 Erythrocyte enzymatic changes 110 4.5.4 Erythrocyte susceptibi l ity to peroxidative damage 110 4.5.5 Erythrocyte membrane and plasma l ip id analyses 110 4.6 Diabetic Patients 118 4.6.1 General features 118 4.6.2 Erythrocyte enzymatic changes 118 4.6.3 Erythrocyte susceptibi l ity to peroxidative damage 121 4.6.4 Erythrocyte membrane and plasma l ip id analyses 121 6. DISCUSSION 128 6.1 Evaluation of the Experimental Data: Coefficient of Variation 128 6.2 Differentiation Between Diabetes- and Chemical-Induced Changes in Antioxidant Status 129 6.3 Effect of Body Weight Loss on Tissue Antioxidant Status . 130 6.4 Tissue Antioxidant Status in Chemically-Induced Diabetes 131 6.5 Tissue Antioxidant Status in Spontaneously Diabetic BB Rats 139 6.6 Allopurinol and Experimental Diabetes 142 6.7 Erythrocyte Antioxidant Status 146 6.8 Erythrocyte Susceptibil ity to Oxidative Stress 149 6.9 Erythrocyte and Plasma Lipid Components 151 6.10 . S imilarit ies Between Erythrocyte Alterations in Animal and Human Diabetes 154 6.11 Summary and Conclusions 155 7. REFERENCES 166 VI 11 LIST OF TABLES Table Page 1. General features and plasma levels of glucose, l ipids and insulin in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 54 2. General features and plasma levels of glucose, l ip ids and insulin in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 55 3. Antioxidant status of heart in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 58 4. Antioxidant status of heart in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 59 5. Antioxidant status of pancreas in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 60 6. Antioxidant status of pancreas in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 61 7. Antioxidant status of l iver in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 62 8. Antioxidant status of l iver in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 63 9. Antioxidant status of kidney in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 64 10. Antioxidant status of kidney in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 65 11. Antioxidant enzyme act iv i t ies of erythrocytes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 67 12. Antioxidant enzyme act iv i t ies of erythrocytes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 68 13. H202-induced GSH depletion of erythrocytes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 69 14. H202~induced GSH depletion of erythrocytes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 70 15. H202~induced MDA production of erythrocytes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats 71 16. H 20 2-induced MDA production of erythrocytes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats 72 Cholesterol and phospholipid contents of erythrocyte membranes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats Cholesterol and phospholipid contents of erythrocyte membranes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats Phospholipid composition of erythrocyte membranes • in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats Phospholipid composition of erythrocyte membranes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats Phospholipid composition of plasma in control, untreated STZ-diabetic, and insulin-treated STZ-diabetic rats Phospholipid composition of plasma in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats General features and plasma levels of glucose, l ipids and insulin in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats Antioxidant status of heart in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats Antioxidant status of pancreas in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats Antioxidant status of l iver in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats Antioxidant status of kidney in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats Antioxidant enzyme act iv i t ies of erythrocytes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats H202~induced GSH depletion of erythrocytes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats H202~induced MDA production of erythrocytes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats Cholesterol and phospholipid contents of erythrocyte membranes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats 49. Antioxidant status of kidney in control, spontaneously diabetic male BB rats and their non-diabetic littermates 111 50. Antioxidant enzyme act iv i t ies of erythrocytes in control, spontaneously diabetic male BB rats and their non-diabetic littermates 112 51. H202~induced GSH depletion of erythrocytes in control, spontaneously diabetic male BB rats and their non-diabetic littermates 113 52. H202~induced MDA production of erythrocytes in control, spontaneously diabetic male BB rats and their non-diabetic littermates 114 53. Cholesterol and phospholipid contents of erythrocyte membranes in control, spontaneously diabetic male BB rats and their non-diabetic littermates 115 54. Phospholipid composition of erythrocyte membranes in control, spontaneously diabetic male BB rats and their non-diabetic littermates 116 55. Phospholipid composition of plasma in control, spontaneously diabetic male BB rats and their non-diabetic littermates 117 56. General features and plasma levels of glucose and l ip ids in control and diabetic subjects 119 57. Antioxidant enzyme act iv i t ies of erythrocytes in control and diabetic subjects 120 58. H 2 0 2 -and d 59a. H 2 0 2 -and d 59b. H 2 0 2 -and di nduced GSH depletion of erythrocytes in control abetic subjects 122 nduced MDA production of erythrocytes in control abetic subjects 123 nduced MDA production of erythrocytes in control abetic subjects 124 60. Cholesterol and phospholipid contents of erythrocyte membranes in control and diabetic subjects 125 61. Phospholipid composition of erythrocyte membranes in control and diabetic subjects 126 62. Phospholipid composition of plasma in control and diabetic subjects • 127 xi i LIST OF FIGURES Figure Page 1. Structural formulae of alloxan and streptozotocin 17 2. Structural formula of allopurinol 31 3. Tissue distribution of CAT in female rats 48 4. Tissue distribution of GSH-PX in female rats 49 5. Tissue distribution of CuZn-SOD in female rats 50 6. Tissue distribution of GSSG-RD in female rats 51 7. Tissue distribution of GSH in female rats 52 x n i LIST OF APPENDICES Appendix Page I. Alterations in antioxidant systems of heart in various models of diabetes 161 II. Alterations in antioxidant systems of pancreas in various models of diabetes 162 III. Alterations in antioxidant systems of l iver in various models of diabetes , 163 IV. Alterations in antioxidant systems of kidney in various models of diabetes 164 V. Alterations in antioxidant enzymes of erythrocytes in various models of diabetes 165 xiv LIST OF ABBREVIATIONS ALP Allopurinol ALX Alloxan BB rat Bio-Breeding Wistar rat C/P Cholesterol/Phospholipid ratio CAT Catalase GSH-PX Glutathione peroxidase GSSG-RD Glutathione reductase GSH Reduced glutathione H 20 2 Hydrogen peroxide HbA l c Glycosylated hemoglobin HDL High density lipoprotein HMPS Hexose monophosphate shunt IDDM Insulin-dependent diabetes mellitus (Type I) ISDBB Insulin-treated spontaneously diabetic BB rat LDL Low density lipoprotein LPC Lysophosphatidylchol ine MDA Maiondialdehyde NADPH Reduced nicotinamide adenine dinucleotide phosphate NDLM Non-diabetic littermates of BB rats NIDDM Non-insulin-dependent diabetes mellitus (Type II) PC Phosphatidylcholine PE Phosphatidylethanolamine PI Phosphatidyl inositol PS Phosphatidylserine SM Sphingomyelin SOD Superoxide dismutase STZ Streptozotocin TBA Thiobarbituric acid ACKNOWLEDGMENTS xv I am grateful to Dr. D.V. Godin for giving me the opportunity to work with him, and also to M. Garnett, Mrs. J . Garnett and R. Ko who provided me with their assistance during my work. I am also grateful to Dr. M. C. Sutter, Head of the Department of Pharmacology Therapeutics, and to Ms. C. Bruce and to the faculty and staff members of this department for allowing me to use their f a c i l i t i e s and for their continuous help. I express my appreciation to Dr. J . H. McNeill, Dean of the Faculty of Pharmaceutical Sciences, for kindly supplying the BB rats used in our study. I am also grateful to Dr. J . Frohlich, Director of the Cl inical Chemistry Laboratory at Shaughnessy Hospital for providing us with the diabetic patient blood samples. Final ly, without the continous advice and guidance of Dr. D. V. Godin and the much appreciated financial support of the Iraqi Government this study would never have been completed. Therefore, i t is a great honor and privilege for me to be indebted to them for al l they have done. In thankfulness and gratitude dedicated my parents for their patience, understanding and unlimited support 1 1. INTRODUCTION 1.1 Diabetes Mellitus Diabetes represents a spectrum of metabolic disorders, most of which have an underlying genetic component, with a common f inal effect namely a lack of insulin action on target tissues. Two general types exist, namely, the insulin-dependent juvenile-onset type (IDDM or Type I), comprising something of the order of 20 percent of c l in i ca l diabetes, and the non-insulin-dependent  maturity-onset type (NIDDM or Type II) which constitutes the majority of c l i n -ical cases (1). In the Type I form, s-cells of the is lets of Langerhans are total ly or part ia l ly destroyed, possibly as the result of autoimmune processes or cumulative insults by B-cytotoxic viruses or chemicals (1,2). There is a complex interplay between genetic and environmental factors. In Type II d ia -betes, where a genetic component is present, a considerable degree of hetero-geneity exists, with two major groups of abnormalities being inadequate insu-l in release in response to glucose and insensit iv ity of target tissues (such, as l i ver , adipose tissue and muscle) to insulin (1). 1.2 Diabetic Complications Although the introduction of insulin therapy some 60 years ago has revol-utionized the management of diabetes, the problems posed by this common and progressively debil itating disease are far from resolved (3). Thus, deaths resulting from ketoacidosis are very much less common and insulin use has resulted in a markedly increased life-span of diabetic patients. However, the University Group Diabetes Program (UGDP), a long-term prospective c l in ica l t r i a l , provided l i t t l e evidence that insulin treatment was any better than diet alone in altering the course of vascular complications in stable adult-onset diabetes (4). This was true whether insulin was given at a fixed-dose based on the patient's height and weight or at doses adjusted 2 to maintain blood glucose within defined levels. Furthermore, evidence is emerging which indicates that insulin may actually be capable of exerting a permissive effect on the development of atherosclerosis (5). As yet, no definit ive treatment has become available for the preven-tion of diabetic complications. Whether or not long-term glycemic control wil l prevent the chronic vascular complications of diabetes mellitus remains unknown (6). At the present time, the therapeutic management of diabetic patients is based on efforts to control blood sugar levels, whether by dietary measures or pharmacological interventions (6). Recently, however, Fr ier and Hilsted (7) indicated that the acute physiological changes induced by hypoglycemia, although short l ived, may have deleterious effects on the blood vessels of patients with long-standing diabetes and may perhaps aggra-vate the complications of diabetes. They also postulated that acute hypo-glycemia provokes sudden hemodynamic changes and induction of hemoconcentra-tion as well as hypercoagulability, factors that are l ikely to exacerbate the pre-existing complications by increasing blood viscosity and decreasing capi l lary blood flow, thereby further compromising an already jeopardized microcirculation which may increase hypoxia and ischemia in tissues such as the retina and the renal glomerulus, both of which are affected by diabetic microangiopathy. Diabetic patients develop many chronic secondary complications of the disease, namely retinopathy, nephropathy, neuropathy and peripheral vascular disease (8). Microangiopathy and accelerated macroangiopathy are prevalent in both Type I and Type II diabetes mellitus. Microangiopathy is predomin-antly responsible for the excessive morbidity and mortality in Type I diabe-tes patients, whereas accelerated macroangiopathy is direct ly relevant to the excessive morbidity and mortality in Type II diabetic patients (6). Thus, microangiopathy leads to the retinopathy, nephropathy and the 3 capi l lary basement membrane thickening associated with diabetes. Macroang-iopathy results in accelerated atherogenesis with its cardiovascular abnor-malities, the most important being coronary heart disease (CHD)*, which is approximately 2.5 times more frequent in diabetics and is responsible for about 50% of a l l deaths in diabetics (9). The prognosis following acute myocardial infarction is also poorer in the diabetic, as compared with the hon-diabetic population (10). Diabetes mellitus is recognized as an independent risk factor for CHD. The mechanisms underlying the association between hyperglycemia and atherosclerotic heart' disease are incompletely understood, and i t is even unclear whether a causal link exists between abnormal blood glucose levels and CHD (11). However, hyperlipidemia constitutes another CHD risk factor often associated with diabetes (11). Despite the increased risk of athero-sclerosis in diabetes, the role of alterations in l ipids and lipoproteins in the pathogenesis of atherosclerosis in diabetes is s t i l l not entirely clear. In particular, much remains to be learned concerning the role of changes in the composition of low density lipoprotein (LDL) cholesterol and high density lipoprotein (HDL) cholesterol, whose levels when increased and decreased, respectively, are believed to be associated with an increased risk of atherosclerosis (12). However, i t has been suggested that qualita-tive changes in plasma lipoproteins in diabetes, e.g., those involving oxidation, may promote arterial ce l l injury which init iates the events leading to atherosclerosis (13,14). In a recent review of the pathogenetic mechanisms of the long-term complications of diabetes, Greene et a l . (15) suggested that these complica-tions presumably result from a complex interplay between multiple direct and indirect metabolic consequences of insulin deficiency, hyperglycemia, or 4 both, and additional poorly defined genetic or environmental factors. Thus, three potentially pathogenic processes resulting from the exposure of tissues to hyperglycemia have been proposed: non-enzymatic glycosylation of proteins, altered microvascular hemodynamics, and abnormal polyol-inositol metabolism (15). The finding that inhibitors of aldose reductase, the f i r s t enzyme in the polyol pathway, are able to both prevent the accumulation of sorbitol (a product of the polyol pathway) and the depletion of myo-inositol (the most abundant stereoisomer of inositol) has resulted in the formulation of the "sorbitol-myo-inositol hypothesis" with possible therapeutic applica-tions in the prevention of pathological changes not only in lenses and peripheral nerves but also in other tissues susceptible to diabetic compli-cations (15). However, another recent art ic le has pointed out the lack of evidence to indicate that therapeutic interventions aimed at the level of sorbitol-myo-inositol metabolism have any influence upon the natural course of large blood vessel disease in patients with diabetes mellitus (16). Macroangiopathy in diabetic patients has been suggested to be age-dependent, rather than related to the duration and severity of diabetes (17). Further-more, in view of the hypothesis that diabetes accelerates the aging process, or results in premature vascular disease through effects on platelets, glycosylation of vascular tissues, or ce l lu lar prol i ferat ion, Minaker (18) suggested that the interactions of aging and diabetes may determine the prevalence and incidence of vascular complications. He presented several examples of the tissue effects of diabetes known to be characteristic of the aging process. For example, the age-related thickening of the basement membrane of tissues may be accelerated by diabetes. Also, aging of collagen is greatly enhanced by the diabetic state (18). Although the exact molecu-lar mechanism(s) behind such a proposed interaction between aging and 5 macrovascular complications of diabetes is not as yet clear, reactive oxygen radical processes, e.g., l ip id peroxidation, may provide one common link (14,19,20). Another potentially pathogenic process implicated in the development of diabetic complications is the non-enzymatic glycosylation of proteins (15). It has been suggested that non-enzymatically glycosylated proteins, such as tissue collagen, undergo a "browning" reaction involving non-enzy-matically glycosylated proteins which gradually give rise to cross-linked protein adducts and the appearance of brown fluorescent pigments. These fluorescent products have been detected in skin-biopsy specimens from type I diabetics (21). This "browning" is believed to be the same process which has been recognized as the discoloration of stored and heated food (the Mai Hard reaction) (22). Monnier et a l . (21) have demonstrated that the levels of fluorescent pigments increased with the severity of diabetic comp-l icat ions, particularly retinopathy, suggesting that an overall correlation may exist between the severity of diabetic complications and cumulative glycemia over many years, based on the assumption that the fluorescence results from a "browning" product of glucose (21). On the other hand, sub-sequent study has suggested that protein browning appears to be a property of non-enzymatically glycosylated proteins irrespective of their ha l f - l i f e and is not directly proportional to their degree of glycosylation, since no significant correlation between protein "browning", giving rise to f luores-cence, and the degree of blood glucose control, assessed by HbA-^ measure-ment, was found (23). Instead, i t has been demonstrated that fluorescence of proteins can be influenced by free radical processes and that non-enzymatic glycosylation may enhance protein oxidation and aggregation by virtue of the ab i l i ty of glucose bound to protein amino groups by ketoamine 6 linkage to participate in free-radical reactions (23). This finding was interpreted as being in support of the suggestion that free-radical proces-ses may be involved in the pathogenesis of diabetic microangiopathy (23). 1.3 Reactive Oxygen Radicals and Diabetic Complications During the course of normal metabolism, approximately 98% of molecular oxygen undergoes complete reduction to water via the "oxidase" pathway (24). The remainder is converted by an "oxygenase" pathway to potentially toxic reactive species (including superoxide and hydroxyl radicals, as well as hydrogen peroxide) capable of causing oxidative damage to membranes and alterations in subcellular organelle structural and functional integrity (25,26). The levels of these intermediate reduction products of oxygen metabolism are controlled by a variety of ce l lu lar defense mechanisms consisting of enzymatic [superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GSH-PX)] and non-enzymatic [glutathione (GSH) and a-tocopherol] scavenger systems (24,27). Generally speaking, the enzyme SOD converts superoxide radicals to H ,^ and 0^. H2O2, in turn, is metabolized by CAT and also by GSH-PX. There is increasing evidence that, in certain pathological states (19,20,27-30), the enhanced production and/or ineffective scavenging of such reactive oxygen species may play an important role in tissue injury. In an extensive review dealing with biochemical aspects of complica-tions of diabetes mellitus, Brownlee and- Cerami suggested that relative tissue hypoxia might play an important role in the development of several diabetic complications (31). Thus, diabetics may exhibit increased erythro-cyte aggregation, with increased microviscosity (32) and decreased deform-abi l i ty (33), alterations in levels of 2,3-diphosphoglyceric acid (34) or glycosylated hemoglobin (35), and disorders of coagulation and platelet 7 function (35) — al l of which, in combination with atherosclerotic vascular changes, could potentially contribute to inadequate tissue oxygenation. Although there have been indications that the severity of some of these abnormalities might correlate with the development of certain diabetic complications, the evidence is incomplete and sometimes contradictory. Thus, while one study showed that diabetics with retinopathy had s i g n i f i -cantly greater mean blood viscosit ies than those without (37), others fai led to demonstrate such a correlation (38). To further complicate matters, recent studies of microvascular disease of -the foot in diabetes suggest that diabetic microangiopathy does not produce microvascular occlusive disease and induce skin hypoxia (39), but that neuropathy and ischemia due to large-vessel occlusive disease are the only causative factors that have been identif ied in this regard (40). Although a l l theories advanced concerning the cause of damage induced by hypoxia are largely inferent ia l , tissue hypoxia could favor the enhanced production of reactive products of the partial reduction of molecular oxygen which might contribute to the development of pathological alterations in c l i n i ca l l y affected tissues. Variations in the act iv i ty of processes involved in the generation or detoxification of these reactive radicals (arising either as the result of inter-individual biochemical differences, alterations in diabetic control or other factors) might, therefore, be important in determining susceptibi l i ty of individual patients to various types of diabetic complications. It has been suggested that reduced oxygen supply (hypoxia) might be one of the factors predisposing to l ip id peroxida-tion under these conditions (41). There is much suggestive evidence indica-ting that oxygen lack may favour atherosclerosis (42,43) and the high prevalence of atherosclerosis in diabetics may in part be a reflection of 8 increased oxygen free radical act iv i ty. Selenium is a cofactor for GSH-PX, an important component of ce l lu lar oxidation defence mechanisms, and a significant inverse correlation between plasma selenium levels and the severity of coronary atherosclerosis has recently been demonstrated (44). The severity of atherosclerosis has also been shown to be related to the degree of elevation in l ip id peroxide levels in human arterial walls (45). Similarly, the oxidation of plasma low density lipoproteins has been shown to increase their atherogenicity (46). Recently, the existence of a posi-tive correlation between l ip id peroxide concentration in plasma and in the arterial wall was found in patients with atherosclerotic lesions of per i -pheral arteries (47). Moreover, plasma levels of malondialdehyde (MDA) (a by-product of l i p id peroxidation) were found to be s ignif icantly higher in hyperlipidemic subjects with arterial lesions than in those without arterial lesions, supporting the hypothesis that l i p id peroxidation may play a role in atherogenesis (48). At present, the detailed mechanism of atherogenesis is not known; however, as indicated above there is evidence to suggest a possible relationship between l ip id peroxides and atherosclerosis. Three possible mechanisms have been suggested to explain the relationship between athero-sclerosis and elevated levels of l ip id peroxide in the plasma (49). F i r s t , l ip id peroxides enhance platelet aggregation, which has been implicated in the process of atherosclerosis. Second, there is the poss ib i l i ty of direct damage to endothelial or intimal cel ls by l ip id peroxides, and the third is the fact that high l ip id peroxide levels suppress biosynthesis within the arterial wall of prostacyclin, which is thought to inhibit the aggregation of platelets and the contraction of blood vessels, thereby protecting against the development of vascular disease. Sato et a l . (50) observed that 9 NIDDM patients with angiopathy have higher plasma l ip id peroxide levels than other diabetic patients, suggesting that a high level of l i p id peroxide in plasma might give rise to an increase in l ip id peroxide levels in the intima of the blood vessel, thereby favoring the process of atherogenesis. Plasma l ip id peroxide levels were also higher in NIDDM patients with retinopathy than in diabetic individuals without retinopathy (51). Although the serum thiobarbituric acid (TBA)-reactive material (an index of l ip id peroxidation) is concentrated mainly in the plasma LDL fraction in both normal and diabe-t i c patients, the increase in TBA-reactive material in diabetes occurs largely in the HDL fraction (52). Triglyceride content was also increased in the HDL serum fraction of both IDDM and NIDDM patients, and a positive correlation existed between TBA-reactive material and HDL fraction t r i g l y -ceride levels in diabetic subjects, but not in normal subjects (52). These and other results (49) suggest that l ip id peroxides may play some role in the in i t iat ion of atherogenesis and that an increase in l ip id peroxide levels in the blood could be an important factor in the early stages of atherosclerotic lesion formation. It is well known that ischemic heart disease occurs ear l ier and is more marked in diabetics as compared with non-diabetics and is a major cause of death in these patients (53,54). The role of reactive oxygen radicals in determining myocardial ischemic/reperfusion injury now seems firmly estab-lished and free radical-induced l ip id peroxidation has been implicated in the process of ischemic injury (30,55,56). Under conditions of relative oxygen lack, electron transport carriers wil l exist preferential ly in the reduced form and the univalent reduction of molecular oxygen to reactive species will be favoured. This could result in the oxidative deterioration of essential macromolecules, especially the unsaturated l ip id structural 10 components of ce l lu lar membranes, with resulting tissue necrosis (56-58). Aznar et a l . have shown that serum levels of MDA increase following myocar-dial infarction and these show a significant correlation with the release of myocardial enzymes, which is used c l i n i ca l l y as an index of the extent of necrosis ( i .e . , infarct size) (59). Besides atherosclerosis and ischemic heart disease, other complica-tions of diabetes may also involve oxygen-radical-related processes. Ross et a l . have demonstrated that injection of antioxidants such as vitamin E (60) or GSH (61) can prevent or diminish the severity of sugar cataractogen-esis in diabetic rats, suggesting that, in addition to osmotic damage, diabetic cataract formation may also involve oxidative damage to cel l membranes, although the former may be the primary cause (61). Superoxide radicals also have been implicated in retinal damage produced by diabeto-genic doses of streptozotocin (STZ) in rats (62). Thus, abnormalities in the generation and/or inactivation of oxygen free radicals may play an important role in determining the development of the adverse changes asso-ciated with the chronic diabetic state. 1.4 Free Radical Metabolism and Pancreatic e-Cell Function The act iv i t ies of free radical metabolizing enzymes in the e-cells of the pancreas affect the ab i l i ty of these cel ls to withstand oxidative stress, whether induced by endogenous free radicals or by free rad ica l -producing drugs. The act iv i t ies of these various enzyme systems have been measured in e-cel ls, and compared to those in other organs. The suscept-i b i l i t y of pancreatic i s let cel ls (63,64) or intact pancreas (65,66) to oxidant injury is greater than that of other tissues. This may predispose the pancreas to free radical injury, especially i f free radical-producing substances accumulate in the 6 -cel l . The free radical-generating properties 11 of alloxan (ALX), and to a lesser extent of STZ, are well characterized and have been reviewed recently (67). Malaisse et a l . (64) observed a good correlation between the act iv ity of GSH-PX and the resistance to oxidative damage induced in vitro by exogenous tert-butyl hydroperoxide, the lowest values for both variables being found in pancreatic is lets and among other tissues studied the highest values were seen in l iver . A comparable distribution in the activity of GSH-PX was reported by Grankvist et a l . (63) and Asayama et a l . (68). Malaisse (69) has concluded that both the capacity of cel ls to take up ALX and their sensit iv ity to active oxygen species are important in determining their susceptibi l i ty to ALX. Islet B-cells might be highly vulnerable to ALX because of the capacity of these cel ls for rapid uptake of ALX and their sensit iv ity to active oxygen species. On the other hand, muscle could be relat ively resistant to ALX, despite its high sensit iv ity to peroxides, because of poor penetration of ALX into muscle ce l l s , while l iver which readily accumulates ALX could d i f fer from is let cel l s by virtue of its resistance to peroxides (69). Another possible reason for the sensit iv ity of the B-cell to free radical injury is the presence of high levels of a thioredoxin-thioredoxin reductase system. This system, which has been implicated in glucose-induced insulin release (70), may favor the reduction of ALX to dialur ic acid, from which reactive oxygen radicals can readily arise (71). There is increasing evidence that the susceptibi l ity of i s let B - ce l l s to redox insult may contribute to the pathogenesis of diabetes mellitus in certain experimental models. Asayama et a l . (68) investigated the effect of vitamin E deficiency, selenium deficiency, and their combined deficiency on i s let ce l l function, and free radical scavenging systems. They demonstrated 12 that Mn-SOD concentrations in is lets were s ignif icantly lower and CuZn-SOD concentrations in is lets s ignif icantly higher than controls in response to deficiency of vitamin E or selenium, alone or in combination. GSH-PX act iv i ty was markedly decreased in is lets of selenium-deficient animals, while no difference in i s let cel l CAT act iv ity was observed between groups. Also, insulin secretory reserve decreased in each of these three deficient groups (68). This decrease was associated with glucose intolerance only in the group with combined deficiency. These authors suggested that the effect of vitamin E and selenium deficiency on"insulin secretion could be related to changes in the redox state of the B -cel ls (68). Indeed, the insu l in -releasing capacity of glucose is influenced by the redox state of i s let thiols (72), and the inhibition of glucose-stimulated insulin release in vitro by t-butyl hydroperoxide is associated with a reduction in the intracel lu lar GSH level in i s let cel l s (64). The decreases in pancreatic i s let Mn-SOD and GSH-PX in vitamin E deficiency, selenium deficiency, or both (68) could explain the protective effects of vitamin E on the diabeto-genicity of ALX (73) and STZ (74). Several antioxidant pathways (notably the GSH-PX-GSSG-RD systems) require reduced nicotinamide adenine dinucleotide phosphate (NADPH), which is derived from the hexose monophosphate shunt (HMPS). It appears that the i s let redox state, as reflected by NADPH/NADP and GSH/GSSG ratios, affects glucose-induced insulin release (72), suggesting that, in addition to protecting the B-cells from free radical injury, the antioxidant enzyme systems may also play a role in insulin release. On the other hand, i t has also been proposed that insulin i t se l f is able to stimulate an NADPH oxidase enzyme in rat adipocyte plasma membranes (75), with the generation of H202« Autoimmune mechanisms have been implicated in the pathogenesis of 13 diabetes on the basis of studies in non-obese diabetic mice (76) and in BB Wistar rats (77), as well as in human IDDM (78). Sub-diabetogenic doses of STZ also gave rise to a diabetic state via a T-cell-mediated immune response (79). The possible involvement of immunological (possibly cell-mediated) processes in STZ-diabetes was suggested by the observation of Like and Rossini (80) that the repeated administration of sub-diabetogenic doses of STZ to mice produced a syndrome which, unlike that resulting from a single injection of STZ, was characterized by pancreatic insu l i t i s (mononuclear cel l in f i l t rat ion) and the appearance of large numbers of Type C virus particles in the B - ce l l s . On the other hand, ALX, administered similarly, produced none of these effects (81). The involvement of free radical production in cell-mediated cytotoxicity suggests this as one possible mechanism of B - ce l l destruction (82). With regard to the possible involve-ment of free radical damage in diabetic pancreas, indirect evidence does exist in support of such a poss ib i l i ty. There is often inflammation of the pancreatic is lets at the onset of insulin-dependent diabetes mellitus (IDDM) (83) and there are indications that diabetes may occur as the result of a viral infection (84). In this context, i t is interesting to note that most cel ls involved in the process of inflammation (polymorphonuclear leukocytes, monocytes, and macrophages) are capable of producing toxic oxygen reduction products when activated (85). Their ab i l i ty to produce superoxide radicals and H2O2 may also be of importance in the tissue damage caused by the inflammatory process (86). Based on such findings, one point which has emerged clearly is the vulnerabil ity of the pancreas to oxidative injury, i s let B -ce l l s being relat ively deficient in enzymes acting to scavenge reactive oxygen radicals. Indeed, i t has been suggested that radical-mediated B - ce l l damage induced by a number of apparently unrelated noxious agents may be important 14 in the etiology of insulin-dependent diabetes (2). Also, Metz (87) has proposed that the release of h^C^ and l ip id peroxides by in f i l t ra t ing leukocytes involved in autoimmune " in su l i t i s " could be important in the destruction of e-cells in Type I diabetes. 1.5 Alterations in Free Radical Scavenging Enzymes in Cl inical Diabetes Although direct evidence that free radical-related processes are altered in c l in ica l diabetes is rather sparse, some data in support of this hypothesis are available in the l i terature. Although there was no s i g n i f i -cant difference between plasma CuZn-SOD act iv i t ies in IDDM children and controls (88), small but significant decreases in red ce l l CuZn-SOD and GSH-PX were reported in one study involving nine IDDM children (89). In NIDDM patients, erythrocyte GSH-PX act iv i t ies were also decreased (51), especially in patients with retinopathy. In contrast, Matkovics et a l . found that hemolysates from human diabetics showed a decrease in CuZn-SOD but an increase in GSH-PX (66). No alteration in CAT act iv i ty has been reported in erythrocytes of patients with either Type I (89) or Type II diabetes (66) but erythrocyte GSSG-RD act iv i ty was reported to be increased in patients with either type of diabetes (90). Previous reports concerning the concentration of GSH in blood of diabetic patients have given confl ict ing results. Thus, Stock and Currence (91) found no significant difference in GSH concentrations in blood from fasting diabetic patients as compared with normal controls. I l l ing et a l . (92) reported similar results in stable diabetic patients although a significant reduction in GSH was found in individuals with severe ketosis. The levels of superoxide anion were found to be s ignif icantly elevated in polymorphonuclear leukocytes from male diabetic subjects (aged 40-55 yr) as compared with those from normal subjects (93) with the greatest changes being observed in NIDDM. This elevation was attributed to the significant 15 reduction in the act iv i t ies of both cytosolic and mitochondrial SOD. Poly-morphonuclear leukocytes obtained from insulin-treated diabetic patients showed less of a reduction in SOD act iv i t ies . A detailed investigation by Shah et a l . (94) has shown that while the resting level of superoxide radical production in diabetic leukocytes is increased, the chemilumines-cence (a direct measure of oxygen free radical production) response and superoxide radical production in the presence of soluble or particulate stimuli is s ignif icantly decreased in leukocytes from diabetic patients. The latter alteration may give rise to impaired ab i l i ty to resist infection, while the increased resting superoxide production may possibly have some bearing on pathological changes associated with diabetes. Indeed, reactive oxygen species produced by leukocytes could give rise to a number of patho-logical changes associated with the diabetic state, including alterations in platelet function (95), vascular endothelial damage (96) and myocardial ischemic injury (97-99). The enhanced tendency to aggregation of platelets in patients with diabetes mellitus is considered to be one factor contributing to . the development of vascular complications (100). Increased synthesis by plate-lets of thromboxane A£ (TXA^), a potent vasoconstrictor and plate let-aggregating substance, is probably important in this regard (100,101). GSH in the presence of GSH-PX can exert an inhibitory effect on TXA^ synthesis at the cyclooxygenase step by reacting with the hydroperoxide intermediate (102). Thomas et a l . (103) reported that platelets from patients with either Type I or Type II diabetes showed increased TXA2 synthesis which has been proposed to be the result of low intracel lu lar GSH levels. It has also been reported that platelets from IDDM (100) and NIDDM (104) patients synthesize more TXA2 than age-matched controls. Platelets from subjects with coronary heart disease were also found to synthesize more TXA? than 16 normal patients, and patients with diabetes and coronary heart disease synthesized more TXA£ than did patients with coronary heart disease alone (104). However, in another study (105) higher plasma levels of the antiaggregatory prostacyclin metabolite 6-keto-PGF-^ and TXB2 (the stable metabolite of the aggregatory TXA^) were found in both IDDM and NIDDM as compared with non-diabetic controls, but the extent of micro-angiopathy correlated only with levels of the 6-keto-PGF^ (proposed to mediate a vascular "self-protecting" response against platelet adhesion and aggregation) and not TXB2 levels. Cl inical studies designed to test the role of platelet aggregation in the pathogenesis of atherosclerosis in diabetes concluded that antiplatelet agents (aspirin and dipyridamole) have no effect on the course of atherosclerotic vascular disease in NIDDM patients except for a marginal protective effect (found only after multiple analyses of the data) against strokes and transient ischemic attacks in a diabetic group with a s ignif icantly greater risk for cerebrovascular disease than other diabetics (106). 1.6 The Role of Free Radicals in Chemically-Induced Diabetes Chemical agents able to exert an immediate, toxic effect on B -cel ls of the is lets of Langerhans and cause a chronic primary insulin deficiency state (107) have been used experimentally to produce diabetes, thereby permitting detailed studies of the biochemical, hormonal and pathological events in the diabetic state. The two agents which have been most exten-sively studied are ALX and STZ. Both are quite selective e-cytotoxins which, in diabetogenic doses, are relat ively free of non-specific toxic effects (81). ALX (Fig. 1) is a derivative of uric acid. The capability of ALX to selectively destroy pancreatic B-cells was f i r s t described by Dunn et a l . in 1943 (108). The elucidation of the mechanism(s) involved in the induction 17 HN—C=0 t>=C C=0 2 ,4,5,6-tetraoxohexa-hydropyrimidine (Alloxan) CrfcOH C=0 N-N=0 CH3 2-deoxy-2-(methyl-3-n i t rosourea) ) l -d -gluco-pyranose (Streptozotocin) Figure 1. St ructura l formulae of al loxan and s t reptozotoc in 18 of diabetes by ALX has been the subject of much investigation. However, in spite of recent advances in our knowledge of i ts direct cytotoxicity in insulin-producing ce l l s , the exact mechanism of its toxicity is not completely understood. STZ is a metabolite of the soil organism Streptomyces achromogenes. It consists of a methyl nitrosourea side chain-linked to the position of D-glucose (Fig. 1). This substance was f i r s t reported to be diabetogenic in studies in dogs and rats by Rakieten et a l . in 196 3 (109). This agent, l ike ALX, selectively destroys insulin-secreting B-cells of the pancreas. Also, as is the case with ALX, its mode of action is not f u l l y understood, although alkylation of c r i t i c a l cel l components has been implicated (67). The fundamental chemical differences between these two agents (Fig. 1) make i t reasonable to speculate that the mechanism of the i n i t i a l attack on the B-cel l is probably different (67). STZ contains a N-methylnitrosourea moiety and is an alkylating agent, a property that is l ike ly responsible for its cytotoxic action. ALX is rapidly reduced in the body to form dialuric acid which undergoes autoxidation to yield the extremely reactive hydroxyl free radical, the species believed to be direct ly responsible for the toxicity to B-cells (110). The evidence implicating oxygen free radicals in the action of ALX primarily derives from experiments showing that, a variety of scavenging enzymes (e.g., SOD), radical scavengers (e.g., dimethylurea and butylated hydroxyanisole) and chelators of metal catalysts involved in the production of hydroxyl radicals are protective in vitro and in vivo against ALX-induced damage to pancreatic B-cells (111-115). Some evidence of a similar nature indicates that oxygen-free radicals may also be involved in STZ-induced diabetes (116,117). The basis for the speci f ic i ty of the attack on B-cells by these agents is not entirely clear. Evidence from a number of studies suggesting that 19 ALX and STZ may act at the site for sugar transport into the c e l l , or at a glucoreceptor site separate from the transport s ite and responsible for insulin release, or at a location separate from either the transport or glucoreceptor s i te, has recently been reviewed (118). For STZ, i t has been suggested that the glucose moiety present in the molecule directs the alkylating agent to the pancreatic B-cell which may have a special a f f in i ty for glucose (67). It has been suggested that ALX and STZ may react with sulfhydryl-containing ce l lu lar components. There is abundant evidence in vivo that modification of SH groups is an important aspect of ALX toxic ity (119) and in vitro studies have demonstrated that a number of SH-binding reagents can mimic the effects of ALX on B-cell permeability and glucose-induced insulin release (120). STZ administration at diabetogenic doses has been shown to decrease GSH levels in both red ce l l s (121) and pancreatic B-cells (116). The subcellular site for the i n i t i a l attack by reactive products derived from STZ (methylating species) or ALX (oxygen free radicals) is not clear; however, Uchigata et a l . (122) have shown that both ALX and STZ cause DNA damage in isolated pancreatic is lets and that this in it iates a chain of events leading to B-cell dysfunction and inhibition of proinsulin synthesis. A key event in both cases is the depletion of cel lu lar pyridine nucleotides, particularly NAD , caused by the activation of the NAD+-dependent nuclear enzyme poly-(ADP-ribose) synthetase which is activated by double-stranded breaks in DNA. In the same study of Uchigata et a l . (132), a combination of SOD and CAT was found to protect against i s let DNA strand breaks and inhibition of proinsulin synthesis induced by ALX but not STZ. This suggests the involvement of both superoxide anion and hydrogen peroxide (and therefore hydroxyl radicals) in the action of ALX on i s let DNA, and that, in contrast to ALX, i t is unlikely that STZ acts on 20 i s let DNA through these reactive oxygen moieties. However, a recent study by Schraufstatter et a l . (123) suggests that the mechanism of DNA-strand break does involve the presence of an oxidant. They observed that exposure of cel l s to H ^ induces rapid DNA-strand breaks, increases poly-(ADP-ribose) synthetase act iv i ty, and that inhibition of this enzyme considerably reduced H202~induced depletion of NAD+. Nicotinamide, an inhibitor of poly-(ADP-ribose) synthetase, has been shown to protect against the diabetogenic effects of ALX (124) and STZ (125) by preventing the . fa l l in NAD+ associated with ALX- and STZ-diabetes (122). C l in ica l ly , n icot in-amide has been administered to newly diagnosed Type I diabetic patients, and has been shown to extend remission time (defined as the period of time during which fasting plasma glucose remains below 7.8 mM, postprandial glucose below 9 mM, and HbA^ below 7.5% in the absence of insulin or sulphonyl urea treatment) in these patients, suggesting that i t has the abi l i ty to decrease the rate of destruction of B-cells and enhances their regeneration (126). Grankvist et a l . (63) have proposed that the high susceptibi l ity of B-cells to ALX may be due to the presence in B-cells of binding sites or ef f ic ient reductive mechanisms for ALX, or to a greater sensit ivity of B-cells than other cel ls to superoxide radicals and H2O2. Asayama et a l . (127) have demonstrated that i s let cel ls in the presence of ALX show a greater degree of early chemiluminescence than either red cel l s or hepato-cytes. These authors (128) have also demonstrated an enhanced production of chemi luminescence in isolated is lets from rat pancreas by ALX, but not STZ, suggesting that ALX acts as an exogenous generator of free radicals, possibly hydroxyl radicals, in pancreatic i s le ts , but that STZ does not. The poss ibi l i ty that free radicals are responsible for STZ-induced B-cell toxicity has also been considered, although the case is not as well 21 made as for ALX. STZ does inhibit CuZn-SOD in red cel ls in vivo (62) and, l ike ALX, exerts an inhibitory effect on i s let CuZn-SOD act iv ity in vitro, suggesting that such an effect might contribute to its cytotoxicity (129). However, experiments to determine whether or not SOD administration can prevent STZ-diabetes have yielded confl ict ing results (116,130,131). The fact that SOD can provide some amelioration of the diabetogenic effects of STZ (116) raises the poss ibi l i ty that the diabetogenic effect of STZ may be mediated, in part, by STZ-induced impairment in the ab i l i ty of the B-cells to withstand oxidant stress. Therefore, i t has been hypothesized that STZ somehow interferes with ce l lu lar protective mechanisms that normally scavenge endogenously produced free radicals (128). Further support for the hypothesis that the cytotoxic effects of STZ are related to free rad ica l -induced peroxidation is provided by the ab i l i ty of acetyl-homocysteine-thiol actone (CYT), an organic thio compound that exerts free radical scavenging act iv i ty, to increase pancreatic CuZn-SOD act iv i ty and attenuate the damage in pancreatic B-cells induced by single diabetogenic or multiple subdiabetogenic doses of STZ (132). Administration of the antioxidant vitamin E to rats prior to the administration of either STZ or ALX provided protection against the diabetogenic effects of both these agents (74). In addition, rats whose antioxidant status was compromised by means of a vitamin E- and selenium-deficient diet, demonstrated increased diabetogenic susceptibi l i ty to normally non-diabetogenic doses of STZ (74), providing indirect support for the suggestion that STZ and ALX may exert their diabetogenic effect by acting as oxidants or free radical producers. 1.7 Alterations in Oxygen Radical Scavenging Systems in Chemically-Induced  Diabetes Evidence for the involvement of alterations in reactive oxygen radical scavenging mechanisms in experimental diabetes has emerged from studies of 22 animals treated with ALX or STZ. Matkovics (65) reported a decreased CuZn-SOD act iv i ty in homogenates of l iver, kidney, pancreas and heart of rats made diabetic for 2 months with STZ. In another study, Matkovics et a l . (66) also demonstrated comparable decreases in CuZn-SOD act iv i t ies in several organ homogenates studied in chemically-induced diabetes in rats and s ignif icantly higher CAT act iv i t ies in the l iver and kidney of diabetic rats. Further studies have shown that CuZn-SOD act iv i ty in red cel ls and retina from rats with STZ-diabetes of 5 day duration is decreased, although no change in CuZn-SOD act iv ity was noted in other organs, including l iver and kidney (62). The fa i lure of insulin treatment (3 days) to restore CuZn-SOD act iv i ty in retina and red cel ls has been suggested to indicate a direct toxic effect of STZ on these tissues (62). Decreases in the act iv i ty of CuZn-SOD have been reported in renal cortex and l i ver of rats with STZ-diabetes of 9-10 days duration (133). Insulin or oral glutathione treatments restored act iv ity of this enzyme toward levels found in normal tissues (133). These results have been interpreted to indicate that insulin deficiency may result in inhibition of CuZn-SOD act iv i ty of some, but not a l l , organs and that the similar effect of glutathione to insulin may relate to the role of glutathione in removing hydrogen peroxide (133). This finding, which would tend to suggest that H^O^ is an inhibitor of SOD, contradicts that of Matkovics et a l . (65) who suggested that acts as an inducer of SOD, based on the finding that rats provided with drinking water containing 0.5% ^ 02 showed increased CuZn-SOD act iv ity in various tissues. However, i t was later suggested (82) that this oral H2O2 might not enter the blood directly but rather lead to increased oxygen production in the stomach. This oxygen would then enter the blood stream and increase the oxygen tension in tissues which, in turn, could lead to increased CuZn-SOD act iv i t ies due to increased oxidative metabolism. 23 CAT act iv i ty in rats of both sexes with STZ-induced diabetes was increased in some tissues (namely l iver, kidney and erythrocytes) but decreased in spleen (66). On the other hand, another study has reported that hepatic CAT act iv ity of female rats with STZ-diabetes was not s i g n i f i -cantly depressed at 30 or 60 days after the induction of diabetes, but was decreased at 90 days (134). Hepatic GSH levels were decreased in male (133) and female (135) rats with STZ-diabetes. Oral glutathione administration had no effect on these levels, but insulin treatment restored GSH levels to normal (133). In contrast to these findings and as an indication of the fact that hepatic GSH levels have shown a variable response toward STZ diabetes, an increase in hepatic GSH content of male STZ-diabetic rats that was reversed by insulin treatment was reported (136). In view of the foregoing information, i t is clear that alterations in free radical defense mechanisms in experimental diabetes are complex in nature and vary from tissue to tissue. Further, i t seems that the duration of diabetes as well as the type of diabetogenic agent used are also involved in these complex patterns of alterations. 1.8 Spontaneously Diabetic BB Rat (SDBB) The spontaneously diabetic BB (Bio-Breeding) Wistar rat was discovered at the Bio-Breeding Laboratories Limited in Ottawa, Ontario, Canada in 1974. These animals provide a model of Type I diabetes (137). The incidence of diabetes in established colonies is approximately 40% (77). Onset of diabetes is abrupt, early (48-120 days), and progressive: in older or younger animals the syndrome develops only rarely. Diabetes occurs with equal frequency in both sexes. This model of insulin-deficient diabetes is characterized by the development of hyperglycemia, hypoinsulinemia, and ketosis at a relatively 24 early age, and death wil l usually occur at an early age unless insulin therapy is introduced (137). Hypertriglyceridemia and plasma lipoprotein changes were observed not only in BB rats with overt diabetes (SDBB) but also in their non-diabetic littermates (NDLM) (138), suggesting the presence of metabolic abnormalities in the latter group of rats usually referred to as "littermate controls". The poss ib i l i ty that changes in plasma l ip id concentrations and apolipoprotein distribution are strain-related rather than disease-related (138) was supported by the finding of common patholog-ical changes in the immune system in various organs (139) of both SDBB and NDLM. The spontaneously diabetic BB Wistar rat has provided an extremely useful model of insulin-dependent diabetes. The diabetic state of these animals is believed to be autoimmune in origin and is considered a useful model of juvenile onset insulin-dependent diabetes, although metabolic abnormalities and several sequelae of the human disorder (e.g., diabetic nephropathy and large vessel abnormalities) are absent (139,140). For our purpose, this model offers an appreciable advantage over those involving ALX or STZ administration in that these diabetogenic agents may exert effects which could interfere with efforts to assess the role of oxygen rad ica l -related processes in the pathogenesis of diabetes as i t relates to the human disorder. A long-term study by Nakhooda et a l . (141) explored the pathophysio-logic alterations prior to and during the development of diabetes. In this report, the evolution of overt diabetes from a normoglycemic state was found to be condensed into several days. Mild insu l i t i s and abnormal glucose tolerance were demonstrated in normoglycemic littermates. Such a finding has been suggested to represent a variant of the syndrome with insufficient B -ce l l destruction to manifest frank diabetes, or a stage in the evolution 25 to the overt syndrome, indicating the existence of a "chemical form" of diabetes, a term used instead of "impaired glucose tolerance". Histologic examination of the pancreas of diabetic animals did show pronounced insu l i t i s (a manifestation of the onset of juvenile-onset diabetes mellitus in man) (139) making this model an intriguing one in its parallelism to the Type 1 form of human diabetes. 1.9 Antioxidant Status in SDBB The currently available l iterature contains minimal information concerning tissue antioxidant systems in this important model of diabetes. One of the only publications in this regard is that of Behrens et a l . (142) which showed that elevated levels of the antioxidant vitamin E are present in the plasma and various tissues (e.g., RBC, pancreas, heart and l iver) of both NDLM and SDBB rats. Insulin treatment was shown to restore normal tissue levels of this antioxidant. Among the tissues investigated, pancreas had the lowest vitamin E concentration in control rats, and pancreata of NDLM and SDBB rats accumulated six times more vitamin E than pancreata of control rats (142). 1.10 Lipid Peroxidation in Diabetes As mentioned ear l ier, elevated levels of l i p id peroxides have been reported in various blood components of diabetic patients (50-52,66). It is a well-known fact that l ip id peroxides result from oxidative attack on l ipids of biomembranes. Although the source of l ip id peroxides in diabetes is not known, possible sources could be lipoxygenase- or cyclooxygenase-dependent pathways of arachidonic acid (AA) metabolism. Lipoxygenase act iv i ty, which produces hydroperoxy-fatty acids from AA, is increased in platelets from diabetic subjects (143). Cyclooxygenase act iv i ty, which produces endoperoxides of AA, is also increased in platelets from diabetic humans (144). There are several possible effects of excessive generation of 26 these highly reactive compounds. These include B-cell destruction (64), peroxide-induced endothelial damage (145) and inhibition of prostacyclin generation (146), which together could contribute to early lesions of diabetic vascular disease and increased platelet adhesion as well as decreased polymorphonuclear leukocyte function, due to oxidative inactiva-tion of white ce l l s (147). Peroxide levels have also been reported to be increased in various tissues in diabetic rats (66,148-151) as well as obese hyperinsulinemic mice (152). Augmented levels of tissue l ip id peroxides were found in kidneys and retina of STZ-diabetic rats and these increases were completely abolished by insulin treatment (150). Rats with STZ-diabetes of 1 wk duration, but not of 20 wk duration, had elevated levels of serum TBA-reactive material, while hepatic TBA-reactive material was increased at both periods of diabetes (148). ALX-diabetic animals also show a marked increase of l ip id peroxide concentrations in l i ver and kidney, but not other organs, suggesting that the elevation of plasma l ip id peroxide levels may reflect peroxidation in these organs (149). 1.11 Non-Enzymatic Antioxidants in Diabetes GSH has been shown to be of importance during oxidative stress because i t supplies GSH-PX (which ut i l i zes H2O2 and fatty acid hydroperoxides as substrates) with reducing equivalents, maintains protein sulfhydryl groups in reduced form and may non-enzymically detoxify free radicals (153). Alterations in GSH levels in diabetes are discussed elsewhere. Vitamin E is considered to provide protection against free radical toxicity by stabi l iz ing membranes through formation of complexes with membrane fatty acids (154). Vitamin E is also capable of reducing super-oxide and hydroxyl radicals, singlet oxygen, l ip id peroxy radicals and other radical species (155). There have been suggestions of an association 27 between vitamin E and diabetes in humans. For example, there are reports that human diabetics have higher concentrations of plasma tocopherol compared to normal subjects (155) although there was one study in which no difference was observed (157). Animals deficient in vitamin E are part icu-lar ly susceptible to tissue damage by peroxidation (73,158). Vitamin E or C were reported to prevent erythrocyte membrane l i p id peroxidation (detected by MDA measurements) in rats treated with the oxidant drug phenyl hydrazine (159). Recently, Pritchard et a l . (151) reported that supplementation of the diets of STZ-induced diabetic rats with high doses of vitamin E e l im i -nates the accumulation of l ip id peroxides in the plasma and the l iver, returns plasma triglycerides toward normal levels, and increases the act iv ity of lipoprotein lipase, which was not affected by the diabetic state. Vitamin E exerted these effects without affecting insulin or glucose levels. It has been postulated that ascorbate serves as an important antioxi-dant because i t has been shown to scavenge superoxide radicals (160) and is a free-radical quencher (161). In addition, i t is postulated that ascorbate reduces tocopheryl radical to tocopherol and augments the role of the latter as an inhibitor of l i p id peroxidation (162). Previous studies have established that plasma ascorbic acid levels are low in STZ-diabetic rats (163), BB diabetic rats (164) and in human diabetic subjects (165). This latter finding correlates with the concern that some Type I diabetics might be in a state of chronic vitamin C deficiency (166). Som et a l . (165) reported that dehydroascorbic acid levels in diabetic subjects are markedly elevated. Dehydroascorbic acid results from oxidation of ascorbic acid during its metabolism and the former has been proposed to result in damage to the pancreatic i s let cel l membrane (164). It has been suggested that since ascorbic acid is dependent on insulin for active transport, hypo-28 insulinemia or hyperglycemia (which could cause competitive inhibition at the level of the transport complex) associated with diabetes should result in an ascorbate deficiency in the cel l (167). This deficiency has been proposed to lead to intimal matrix breakdown, vascular wall f r ag i l i t y , and subsequent increase in atherogenesis and the microangiopathies associated with diabetes may be related to this mechanism (167). Uric acid is another endogenous antioxidant which protects red cel l s from l ip id peroxidative damage. Because of its ab i l i ty to react with singlet oxygen and hydroxyl radicals, urate can act as an effective free radical scavenger (168). Despite this protective effect of uric acid, alterations in its level in diabetes do not seem consistent. Although one study observed a high incidence of diabetes in patients with gout (169), another study (170) has shown that serum uric acid levels are higher in prediabetic individuals but are lower in diabetics than in non-diabetics and continue to f a l l with increasing duration of the diabetes. The mechanism responsible for these alterations in uric acid levels in diabetics is unknown (170). 1.12 Rationale and Objectives of the Study The general aim of this study was to determine the extent to which reactive oxygen-radical-related processes are involved in experimental and human diabetes and the effects of pharmacological interventions on these processes in the hope of deriving information concerning molecular factors which might be involved in diabetes as well as i ts associated complica-tions. A major hypothesis to be tested is that biochemical processes involved in the generation and/or inactivation of reactive oxygen radical species may play an important role in this regard. Although direct evidence that free radical-related processes are altered in diabetes is rather sparse, some data in support of this hypothe-29 sis are available in the l i terature. One point which has emerged clearly is the vulnerability of the pancreas to oxidative injury, i s let beta cel l s being relatively deficient in enzymes acting to scavenge reactive oxygen radicals (63,127). Isolated pancreatic cel l s exposed to the diabetogenic agent ALX in vitro sustain damage which is preventable by SOD, CAT, and hydroxyl radical scavengers (113). There is evidence to suggest that the ab i l i ty of ALX to generate reactive oxygen radicals (including hydroxyl radicals) is a primary determinant of its diabetogenic properties (127,128). STZ exerts diabetogenic actions which, l ike ALX, may be prevented by prior administration of SOD (131), but its actions may not involve the direct formation of reactive oxygen radicals (122,128). Clearly, much remains to be learned about the mechanisms by which these two commonly used diabetogenic agents exert their effects. Based on the foregoing considerations, we have measured the act iv i ty of oxygen-radical scavenging systems in various tissues (heart, pancreas, and kidney) known to be involved either in the development of diabetes or in diabetic complications as well as in l i ver , a major site of detoxification of toxic substances, in rats with STZ- or ALX-induced diabetes. A major question that has been pursued in this study is the extent to which analysis of red cel l biochemical characteristics in poorly controlled diabetes can provide a useful model system to investigate the role of oxygen-derived free radicals in diabetes. The information so derived might not only provide further insights into the mechanisms of diabetic complica-tions, but also indicate the possible predictive value of red cel l measure-ments in assessing the risk of their development in uncontrolled c l in ica l diabetes. Toxic oxygen reduction products are known to in i t ia te l ip id peroxidation and disrupt cel lu lar membrane structures. If experimental diabetes is associated with a generalized alteration in tissue oxidative 30 defense mechanisms, this might be reflected in changes in the susceptibi l ity of erythrocytes to peroxidative damage in vitro. In the present study, therefore, we have measured the act iv ity of various oxygen radical scaveng-ing enzymes in red cel ls of STZ- and ALX-treated animals. We have also examined the susceptibi l ity of their red cel ls to H202~induced oxidative damage. We also planned two general types of experiments to determine the effects of two pharmacological interventions (insulin treatment and allopur-inol administration) on diabetic rats. Allopurinol (Fig. 2) is an inhibitor of xanthine oxidase (one source of superoxide radicals in the body) and has shown beneficial effects in a variety of pathological states where reactive oxygen-derived radicals have been implicated (171). For comparison and in an effort to distinguish alterations attr ibut-able to the diabetic state from incidental actions of the diabetogenic chemicals, we have performed the same analyses on tissues and red blood cel ls from spontaneously diabetic male Wistar BB rats. Investigations into the molecular mechanisms determining the develop-ment of diabetes and its complications have necessitated the use of animal models of the disease. Untreated diabetic animals usually show a marked decrease in body weight that seems to correlate with the severity of diabetes. Experiments in rats have shown that weight loss induced by food-deprivation is associated with alterations in cardiac CAT and CuZn-SOD (172) and* hepatic GSH (173). Interestingly, genetically obese mice have been found to exhibit increased act iv ity of GSH-PX in blood and brain but decreased act iv i ty in l iver (152). It would seem, then, that the act iv i t ies of enzymatic components of antioxidant systems in various tissues are influenced in a complex manner by metabolic status. Therefore, in the present study, we have investigated in detail the effect of a loss in body OH 4-Hydroxypyrazolo-(3,4-d)-pyrimidine gure 2. Structural formula of a l lopurinol. 32 weight (induced by a 72 h period of food-deprivation) on free radical tissue defense mechanisms in rats. In our attempt to investigate the extent to which the chemical and biochemical properties of red cel ls from diabetic animals are representative of the situation in human diabetes, we have also examined red cel ls of patients with Type I and Type II diabetes mellitus. The following were the major questions to be pursued in the study: 1. To what extent are the biochemical processes involved in the genera-tion and/or inactivation of reactive oxygen radical species altered in experimental diabetes mellitus? 2. What is the extent to which alterations in these processes in red cel l s ref lect those observed in tissues? 3. What is the extent to which the chemical and biochemical properties of red cel l s from diabetic animals are representative of the situation in human diabetes? 4. Can pharmacological interventions (insulin and allopurinol) exert an effect on the reactive oxygen scavenging systems in experimental diabetes? 33 2. MATERIALS AND METHODS 2.1 Animal Studies 2.1.1 Induction of diabetes Female Wistar rats (body wt 230-250 g) were fasted for 48 h before inducing diabetes with streptozotocin (STZ) or alloxan (ALX) (Sigma Chemical Co., St. Louis, Missouri). Animals were anesthetized with ether and injec-ted via the ta i l vein with 0.23-0.25 ml of a freshly prepared solution of STZ (50 mg/ml in 0.1 M citrate buffer, pH 4.5) or ALX (50 mg/ml in saline) to give a f inal dose of 50 mg/kg body wt. Control animals received 0.23-0.25 ml of c itrate buffer or saline. Animals treated with either STZ and ALX were allowed to drink 5% glucose solution overnight to prevent drug-induced hypoglycemia. Both treated and control animals were allowed access to food and water ad libitum. During the 12 wk experimental period, animals treated with either diabetogenic agent were monitored by periodic testing for glucosuria using L i l l y Tes-Tape and ketonuria (only in ALX-injected animals) using Chemstrip uG 5000/k (Boehringer Mannheim, West Germany) for the f i r s t week or for glucosuria only for the remainder of the 12 wk period. At the end of this period, seven animals from each group were randomly selected to serve as the STZ- and ALX-diabetic group. 2.1.2 Insulin treatment After an 8 wk period of overt diabetes, a group of STZ-diabetic animals (n = 7) were randomly selected to receive insulin treatment. Prot-amine-zinc insulin (L i l l y , Indianapolis, IN.) was injected subcutaneously at a daily dose of 9-12 U/kg body wt for 4 wk. ALX-diabetic rats received insulin either immediately after the development of glucosuria (24 h after ALX injection), i .e . , the prevention study (n = 7), or after 6 wk of diabetes, i . e . , the reversal study (n = 7). Insulin treatment was continued in both groups to the end of the 12 wk diabetic period. The dosage of 34 insulin was adjusted by monitoring urinary glucose daily and blood glucose weekly to prevent glucosuria and hyperglycemia. 2.1.3 Allopurinol treatment In an experiment in which two groups of animals (8 each) were made diabetic with ALX, one group of rats was injected with allopurinol (ALP) (50 mg/kg body wt. in 0.2-0.23 ml saline solution pH 12.0, intraperitoneally) 30 min before the injection of ALX. During the f i r s t week of injections, periodic testing for glucosuria and for ketones was performed. Then the ALP-treated diabetic rats received ALP in drinking water (25 mg/L) for the 12 wk period of diabetes during which periodic testing for glucosuria was performed. Another group of rats (n = 7) received the vehicle intraperi -tonea l^ , and ALP in drinking water at a concentration of 50 mg/L instead of 25 mg/L because of the increased water intake in diabetic rats. These animals were used as ALP-treated control rats. Because diabetic animals exhibited a wide range of severity of the disease, a fact that may result in varying degrees of polydipsia, and were housed in groups of 3-4 animals per cage, i t was impossible to perform measurements of individual water intake/24 h. However, based on available information in l i terature, the daily water intake of control rats is approximately 30 ml, while that of diabetic rats is approximately 75 ml (174). Accordingly, control rats are estimated to have received a daily dose of 1.5 mg and.diabetic rats a daily dose of 1.9 mg ALP in drinking water. 2.1.4 Food-deprivation Female Wistar rats (body weight 200 - 230 g) were obtained from Charles River Canada Inc., Quebec, randomly divided into two groups of six each and housed in hanging wire mesh cages. Animals received Purina Chow #5012 (Ralston Purina Co., St. Louis) and water ad libitum. One group of rats was food-deprived for 72 h and their body weights recorded before 35 and after food-deprivation. 2.1.5 Spontaneously diabetic male Wistar BB rats The BB Wistar rats used in this study were kindly supplied by Dr. J . H. McNeill, Dean, Faculty of Pharmaceutical Sciences, UBC, from a breeding stock obtained from the Animal Resources Division, Health Protection Branch, Health & Welfare Canada, Ottawa, Ontario, Canada. Our experiments were res-tr icted to male rats as these were the only animals we were able to obtain. Rats were tested for glucosuria daily using Testape (Eli L i l y & Co. Ltd., Toronto, Ontario). Diabetes was confirmed on the basis of +2 glucosuria and plasma glucose of 12.0 mM or more. Diabetic animals (ISDBB) (mean age of onset: 110 ± 1 1 days) were treated with protamine-zinc insulin (Connaught Laboratories Ltd., Ontario) subcutaneously at a dose of 3-9 U/day to main-tain body weight, control hyperglycemia and prevent glucosuria. The dura-tion of treatment was 7-12 wk, at the end of which the animals (mean age 6.19 ^ o.75 months) were k i l led (n = 8). Equivalent numbers (n = 8) of nor-moglycemic diabetes-prone BB rats (non-diabetic littermates) (NDLM), whose ages (mean 7.62 ± 0.44 m) and weights were comparable to those of the treated diabetic BB rats, were obtained from the same stock. For comparison and in an effort to identify alterations attributable to the BB-strain, eight weight-matched control male Wistar rats (obtained from Charles River, Quebec) were also used in the study. Al l rats were maintained in a constant-temperature (22°C) environment with a constant 12 h light schedule (light on at 0700 h, off at 1900 h). 2.2 Human Studies 2.2.1 Studies of erythrocytes from diabetic patients Blood samples from 39 male and female diabetic patients (with either Type I or Type II diabetes) referred to the Diabetic Speciality Center, Shaughnessy Hospital, for evaluation and treatment were used in this study. 36 The diabetic subjects included 24 men and 15 women with a mean age of 53 yr and an age range of 21-83 yr. The mean duration of diabetes was 7.3 yr with a range of 1 month to 31 yr. Seventeen patients were diagnosed as Type I (IDDM) and 22 patients as Type II (NIDDM). Fourteen patients were control-led with insul in, nine with diet alone and sixteen with oral hypoglycemic drugs. The incidence and nature of diabetic complications (such as retino-pathy, peripheral vascular disease (e.g., atherosclerosis), neuropathy and renal functional impairment) were evaluated by examination of the patients' detailed medical histories in a blind manner ( i .e . , in the absence of infor-mation concerning the results of chemical or biochemical assays). These evaluations were carried out by a fourth-year medical student, Mr. A. D. Goumeniouk, in a directed studies project carried out in our laboratory. Accordingly, complications were c lass i f ied on the basis of c l in ica l judge-ment as Hard (H) or Soft (S). Patients with 3H, 2H, or 1H and 2S complications were considered to have marked complications, while those with 1H and/or IS, or several S complications were regarded as patients with moderate complications. Twenty-seven patients had complications, f ive only had hypertension, and seven had no c l in ica l evidence of complications. Control blood samples were obtained from 19 healthy volunteers of both sexes and were matched, as closely as possible, in age with the diabetic pat i -ents. The control subjects included 12 men and 7 women with a mean age of 48 yr and an age range of 26-70 yr. As with the patients, a l l control sub-jects were fasting overnight when blood samples were drawn. Blood samples were drawn into vacutainer tubes (Becton Dickinson Canada Inc.) containing ethyl-enediaminetetraacetic acid (EDTA) (0.07 ml of 15% EDTA). 2.3 Preparation of Cytosolic Fractions At the time of sacr i f ice, rats were anesthetized with ether, blood (8-10 ml) was obtained by cardiac puncture and collected into two heparin-37 i z e d t u b e s , one o f which was used f o r e r y t h r o c y t e membrane p r e p a r a t i o n and enzyme a n a l y s e s , w h i l e t h e o t h e r was u t i l i z e d t o a s s e s s s u s c e p t i b i l i t y t o ^ C ^ - i n d u c e d o x i d a t i v e s t r e s s . The a n i m a l s were k i l l e d by c a r d i a c e x c i -s i o n . T i s s u e s ( h e a r t , p a n c r e a s , l i v e r and k i d n e y ) were removed, b l o t t e d d r y and homogenized i n 10 v o l o f 50 mM t r i s (hydroxymethy1) aminomethane ( T r i s ) -0.1 mM EDTA, pH 7.6 (175), a t 4°C, f o r 30 sec (2 x 15 sec w i t h 15 sec c o o l -i n g i n t e r v a l s ) u s i n g a P o l y t r o n homogenizer at 25% maximum speed (Brinkmann I n s t r u m e n t s , I n c . , Westbury, N.Y.). C y t o s o l i c f r a c t i o n s were p r e p a r e d by c e n t r i f u g i n g homogenates f o r 15 min a t 105,000 x g i n a Beckman L2-65 u l t r a -c e n t r i f u g e . 2.4 P r e p a r a t i o n o f E r y t h r o c y t e Hemolysates Hemolysates were p r e p a r e d i n a 1:10 (v/v) d i l u t i o n i n d o u b l e - d i s t i l l e d w ater, w i t h t h r e e t i m e s f r e e z i n g and thawing i n a d r y - i c e / a c e t o n e m i x t u r e . The hemoglobin c o n t e n t o f hemolysates was d e t e r m i n e d by t h e cyanomethemoglo-b i n method o f D r a b k i n and A u s t i n (176) as o u t l i n e d below. 2.4.1 Hemoglobin a s s a y One ml o f hemolysate s o l u t i o n was mixed w i t h 1.0 ml 1.8 mM KgFe(CN)g and 1.0 ml 2.4 mM KCN. A f t e r 30 min, absorbance was r e c o r d e d a t 540 nm. Hemo-g l o b i n (Hb) c o n c e n t r a t i o n was d e t e r m i n e d u s i n g known amounts o f Hb s t a n d a r d assayed s i m i l a r l y . 2.5 T i s s u e F r e e R a d i c a l S c a v e n g i n g Enzymes T i s s u e c y t o s o l i c f r a c t i o n s and e r y t h r o c y t e hemolysates were used f o r t h e measurement o f f r e e r a d i c a l s c a v e n g i n g enzymes u s i n g a Beckman ACTA C2 s p e c t r o p h o t o m e t e r . A l l assays were c a r r i e d out a t room t e m p e r a t u r e (22°C-25°C) u n l e s s o t h e r w i s e i n d i c a t e d . 2.5.1 C a t a l a s e CAT a c t i v i t y was assayed by t h e method o f Aebi (177) a t 240 nm and e x p r e s s e d as e i t h e r K/g wet t i s s u e o r K/g Hb, where K i s t h e f i r s t o r d e r 38 rate constant. Cytosolic CAT was assayed by incubating 2.7 ml of cytosolic fraction with 27 pi of 95% ethanol for 30 min at 4°C. Then 0.3 ml 10% Triton (in 50 mM Tris-0.1 mM EDTA pH 7.6) was added and the solution was thoroughly mixed. From this solution 100 yl were diluted to 10 ml with 50 mM phosphate buffer pH 7.0 immediately prior to assaying. In a 3 ml cuvette, 2.0 ml of this diluted solution were added and the reaction was started by adding 1.0 ml of freshly prepared 30 mM After rapid mixing, the reaction rate (decomposition of H2O2) was determined from the absorbance changes at 240 nm after 15 and 30 sec relative to a control cuvette containing 1.0 ml of phosphate buffer instead of H2O2. Erythrocyte hemolysates were assayed by di luting 20 ul to 10 ml with 50 mM phosphate buffer pH 7.0 immediately prior to assaying. This diluted solution was then assayed as above. 2.5.2 CuZn-superoxide dismutase CuZn-SOD act iv i ty was measured by the method of Winterbourn et a l . (178) and expressed as units of SOD/g wet tissue or units/g Hb, 1.0 unit being defined as that amount of enzyme causing half-maximal inhibition of nitro blue tetrazolium (NBT) reduction. Cytosolic extracts were prepared by adding to 1.0 ml of cytosolic fraction 3.0 ml of H20 and 1.6 ml of chloroform:ethanol mixture (3:5, v/v). The mixture was shaken vigorously for 5 min and centrifuged at 3,000 g for 10 min. The resulting supernatant was then centrifuged at 12,000 g for 10 min to remove any contaminants. The assay mixture contained 1.0 ml of 75 mM phosphate buffer pH 7.8, 0.2 ml of 0.1 M Na2EDTA-1.5 mg%NaCN, 0.1 ml of 1.5 mM NBT, 50 ul of 0.12 mM r ibof lav in, 25-500 yl of the clear supernatant, and H2O in a f inal volume of 3.0 ml. Riboflavin was added immediately before illuminating the tubes to start the reaction. The tubes were placed in a box illuminated for 2 X 39 2.5 min (with vortexing at beginning, between and at end of the illumina-tion) with fluorescent light at a constant distance, and protected from any other source of l ight. The rate of inhibition of NBT reduction by super-oxide generated via photoreduction of riboflavin was determined by measuring the absorbance at 560 nm. For red cel l analysis, 0.5 ml of erythrocyte hemolysate was combined with 3.5 ml h^ O and 1.6 ml chloroformrethanol mixture (3:5, v/v) and assayed as described above. 2.5.3 Glutathione peroxidase GSH-PX activity was measured by the method of Paglia and Valentine (179) and the act iv i ty was expressed as umol of NADPH oxidized to NADP per min per g wet tissue or per g Hb using a molar extinction coefficient for NADPH at 340 nm of 6.22 x 10 6 . Cytosolic GSH-PX was assayed in a 3 ml cuvette containing 2.0 ml of 75 mM phosphate buffer pH 7.0. The following solutions were then added: 50 ul of 60 mM glutathione, 0.10 ml of glutathione reductase solution (30 units/ml), 50 ul of 0.12 M NaN3, 0.1 ml of 15 mM Na2 EDTA, 0.1 ml of 3.0 mM NADPH, various aliquots of cytosolic fractions (50-150 ul) and H2O (0.45-0.35 ml) in a total volume of 2.9 ml. The reaction was started by the addition of 0.1 ml of 7.5 mM H 20 2 and the conversion of NADPH to NADP was monitored by continuous recording of the change in absorbance. of the system at 340 nm every min for 5 min. Erythrocyte GSH-PX was assayed as described above except that the hemolysate was mixed with an equal volume of double-strength Drabkin's reagent to convert a l l Hb to the stable cyanomethemoglobin form. 2.5.4 Glutathione reductase GSSG-RD act iv i ty was assayed by the method of Long and Carson (90) and expressed as umoles of NADPH oxidized to NADP per min per g wet tissue or per g Hb using a molar extinction coefficient for NADPH at 340 nm of 40 6.22 X 10 . The assay system for measuring the enzyme act iv ity in cytoso-l i c fractions contained 2.2 ml of 0.177 M Tris-43.5 mM EDTA buffer (pH 7.6), 0.5 ml of 11.0 mM oxidized glutathione, and 0.2 ml of cytosolic fraction. The reaction was started by the addition of 0.1 ml of 3.0 mM NADPH and the absorbance at 340 nm recorded every minute for 5 min. Erythrocyte GSSG-RD act iv i ty was measured as described above with l i t t l e modification. Stroma were removed by centrifuging hemolysates at 30,000 g for 10 min. After equilibration of the assay system at 37°C for 5 min, 3.0 mM NADPH was added to in i t iate the reaction. The decrease in absorbance at 340 nm was determined every 5 min for up to 20 min of incuba-tion at 37°C. 2.5.5 Tissue sulfhydryl content Tissue sulfhydryl group content (as an indirect measure of tissue GSH) was measured by a modification of the method of Moron et a l . (180). Aliquots of tissue homogenate were treated with 25% trichloroacetic acid (TCA) to a f inal concentration of 5%. The supernatant obtained by centr i -fuging the mixture at 12,000 xg for 10 min was used. The assay mixture contained 1.0 ml of 0.15 M imidazole pH 7.4, 1.7 ml double d i s t i l l ed water, 0.2 ml supernatant and 0.1 ml of 3 mM 5,5'-dithio-bis(2-nitrobenzoic acid) (DTNB) in 0.15 M imidazole. The reaction was started by the addition of DTNB and readings at 412 nm were obtained at 3 min. Known amounts of GSH were assayed by the same method and used for calculation of GSH quantities in tissues. 2.5.6 Correction for contamination caused by blood In al l cases, the Hb content of tissue cytosolic fractions was deter-mined in order to allow correction of measured enzyme act iv i t ies for the contribution from contaminating blood (which ranged from 1.0 to 16.0 mg Hb/g wet tissue). For this purpose, cytosolic fractions were c la r i f ied by adding 41 1% Triton X-100-50 mM Tris pH 7.6 and assayed for Hb (176). These data were used to subtract the contribution that contaminating blood would make to the measured values of tissue antioxidant enzymes. Depending on the specific parameter evaluated, blood contamination contributed a .major (CAT), an intermediate (GSH-PX), or a minor (GSSG-RD, CuZn-SOD and GSH) fraction of the total amount of a given cytosolic antioxidant. The addition of Triton to cytosolic fractions was found not to interfere with the Hb assay. 2.6 Erythrocyte Membrane Preparation Red cel l s were separated from plasma by centrifugation at 3,000 rpm for 10 min at 4°C. Red cel ls were washed twice with isotonic saline and the buffy coat carefully removed. An aliquot of packed erythrocytes was used for the preparation of hemoglobin-free membranes by a step-wise hemolytic procedure employing progressively more hypotonic sodium chloride solutions at 4°C as described by Godin and Schrier (181). The ratio of packed cel ls to hemolyzing solution was 1:10 for human red cel ls and 1:50 for rat red cel l s at each stage of the step-wise hypotonic lysis procedure. A f inal wash in Tris buffer (10 mM, pH 7.4) was used in order to remove residual Hb. Membrane suspensions or plasma samples were quick-frozen in small aliquots using a mixture of dry-ice and acetone and were stored at -70°C until chemical analyses could be performed. 2.6.1 NADH-dehydrogenase The NADH-dehydrogenase act iv i ty of red ce l l membranes was measured by the method of Howland et a l . (182). To a 3 ml cuvette the following solu-tions were added: 1.0 ml of 48 mM Tris-HCl (pH 7.4), 0.1 ml of 15 mM EDTA, 0.1 ml of 6 mM K^Fe^N)^, 5 0 ul red cel l membrane suspension and H£0 to make a f inal volume of 3.0 ml. The reaction was started by the addition of the membranes. Act iv i t ies were determined by measuring the decrease in absorbance at 340 nm each 30 sec for 3 min. Act iv i t ies were expressed as 42 nmol NADH oxidized to NAD per min per mg protein using a molar extinction coefficient of NADH at 340 nm of 6.22 x 10 6 . 2.6.2 Protein assay Protein concentration was measured by the method of Lowry et a l . (183) using bovine serum albumin (Armour Pharmaceutical Company, I l l inois ) at a concentration of 1.0 mg/ml, as standard. 2.6.3 Cholesterol content The cholesterol content of red cel l membranes was determined using a colorimetric assay kit obtained from Boehringer Mannheim (West Germany). This . assay uses oxidation products generated by cholesterol oxidase following cholesterol ester hydrolysis by cholesterol esterases. The hydro-gen peroxide generated reacts with 4-aminoantipyrine and p-hydroxybenzene-sulfonate in the presence of peroxidase to yield a quinoneimine dye, which has a maximal absorbance at 500 nm. The amount of color produced is directly proportional to the concentration of cholesterol in the sample. 2.6.4 Phospholipid phosphorus content Phospholipid content was determined by assaying- for phospholipid phosphorus as described by Higgins and Dawson (184). Stock solution A (Fiske-Subbarow reagent) was prepared by adding 0.25 g l-amino-2-napthol-4-sulfonic acid (ANS) to 100 ml freshly prepared 15% (w/v) sodium b isu l f i te and then adding 0.5 g anhydrous sodium su l f i te . The solution was f i l tered and then stored in a dark bottle for no more than a week. Stock solution B consisted of 5% (w/v) ammonium molybdate in water. To a 50 yl membrane sample, 1.5 ml of 70% perchloric acid were added and the mixture digested at 230°C in a sand bath for 30 min. After the samples had cooled, 7.6 ml d i s t i l l ed water were added and the solution mixed. A 4.5 ml. aliquot was combined with 0.5 ml d i s t i l l ed water, 0.2 ml of stock solution B, 0.2 ml of stock solution A, and the solutions were mixed and placed in a boiling water 43 bath for 7 min for color development. After cooling, the absorbance of the sample at 830 nm was determined. Sample absorbance was compared to the absorbance obtained from a standard solution containing a known quantity of inorganic phosphate (20 ug/ml). 2.7 Erythrocyte Membrane and Plasma Phospholipid Profi les The phospholipid profiles of plasma and membrane samples (stored no longer than 24 h at -70°C) were determined as described by Godin and Garnett (185) using chloroformrmethanol (2:1, v/v) as the extraction solvent. To a 0.5 ml plasma or membrane suspension was added 4.0 ml of chloroformrmethanol and the mixture was incubated at 4°C for 1 hr with frequent s t i r r ing . Extracts were washed twice with 0.5 ml of 0.2 N KC1 to remove inorganic phosphates and the chloroform layer was separated, its volume measured and known aliquots were used for the analysis of phospholipid as described above. Another aliquot of each sample was dried under a stream of nitrogen, redissolved in 200 ul of the above solvent and an aliquot (40-60 ul) was spotted on precoated s i l i c a gel F254 plates (0.2 mm thickness, Merck). Phospholipids from animal samples were separated using a two-dimensional system: chloroform:methanol:ammonia (14:6:1, v/v) in the f i r s t dimension and chloroform:methanol:acetone:acetic acid:water (10:2:4:3:1, v/v) in the second dimension. Phospholipids from human plasma and membrane samples were separated using a one-dimensional system: chloroform:methanol:ammonia (14:6:1, v/v). Individual phospholipids were detected by exposure to iodine vapor, identif ied by comparison with standards, and scraped from the plates and their phosphorus content determined by the method of Higgins and Dawson (184). Known quantities of standard phospholipids [phosphatidylcholine (PC), lysophosphatidylcholine (LPC), phosphatidyl inositol (PI), phospha-tidylserine (PS), phosphatidylethanolamine (PE), sphingomyelin (SM), and phosphatidic acid] were processed and chromatographed in the same manner as 44 the membrane and plasma samples. The standard phospholipids were obtained from Sigma (St. Louis, Missouri). The amount of phosphorus in each phospho-l ip id fraction was calculated from the amount of phospholipid phosphorus per mg protein of red cel l membrane or per l i ter plasma and the corresponding percent phosphorus content (186). 2.8 Measurement of Erythrocyte Susceptibil ity to Hydrogen Peroxide-Induced  Oxidative Stress Blood was centrifuged at 3,000 rpm for 10 min at 4°C, plasma and buffy coat were discarded, and erythrocytes were washed twice with isotonic saline containing 2.0 mM sodium azide to inhibit CAT act iv i ty (187). 2.8.1 Reduced glutathione (GSH) assay The loss of intracel lular sulfhydryl group content (as an indirect measure of GSH) in response to increasing concentrations of H2O2 was measured as follows. Two ml aliquots of a ten percent suspension of red cel ls in saline-azide were preincubated for 5 min at 37°C. H2O2 in saline-azide solution was added (in a volume of 2.0 ml) to obtain f inal concentrations of H 20 2 between 0.025 and 1.0 mM. Following a 30 min incubation at 37°C, the mixture was centrifuged and erythrocytes were washed with saline-azide solution (5.0 ml) and recentrifuged. The packed erythrocytes were combined with 0.2 ml H20 and 1.3 ml of 5% TCA-lmM EDTA, and the supernatant analysed for sulfhydryl group content at 412 nm using 3 mM DTNB in phosphate buffer. The assay mixture contained 2.6 ml of 0.1 M phosphate buffer (pH 8.0), 0.3 ml supernatant and 0.1 ml DTNB. Readings at 412 nm were obtained at 2 min. 2.8.2 Malondialdehyde (MDA) assay A sl ight modification of the method of Stocks and Dormandy (187) as modified by Gilbert et a l . (188) was used to measure malondialdehyde (MDA) production. A one ml aliquot of a ten percent suspension of red cel l s in 45 saline-azide was pre-incubated for 5 min at 37 C. Peroxidative challenge was induced by the addition of an equal volume (1.0 ml) of various concen-trations of H2O2 in isotonic saline-azide solution (final ^ 2 concentration 0.5-3.0 mM for rat erythrocytes and 2.5-15 mM for human erythrocytes). Following a 30 min incubation at 37°C, the reaction was terminated by addition of 1.0 ml of 28% (TCA)-O.l M sodium arsenite (w/v). The mixture was centrifuged, and a 2.0 ml aliquot of supernatant was combined with 1.0 ml of 0.5% TBA in 0.05 M sodium hydroxide. Color develop-ment was achieved by boiling for 15 min. The tubes were cooled under tap water and the extent of MDA production was estimated from the absorbances at 532 and 453 nm as described previously (187,188). 2.9 Chemical Assays 2.9.1 Glucose, insulin and l ip id measurements Plasma glucose was measured by a quantitative enzymatic (glucose oxidase) determination at 450 nm using a Sigma k i t . The test is based on the simultaneous use of glucose oxidase and peroxidase coupled with a chromogenic oxygen acceptor (e.g., o-dianisidine) for the colorimetric determination of glucose. Plasma insulin levels were measured by an enzyme-linked immunosorbent assay obtained from Boehringer Mannheim (West Germany). This enzyme-immunological test for the quantitative determination of plasma insulin involves measuring the amount of antibody-insulin-POD complex formation by the addition of substrate and a chromogen which results in formation of a dye, the absorbance of which is proportional to the enzyme act iv ity of the antibody-insulin-POD complex. It follows from the competi-tion principle that increasing serum insulin concentrations in the sample wil l result in less insulin-POD conjugate-binding by the antibodies and hence in lower measured enzyme act iv i t ies . Plasma cholesterol and t r i g l y -cerides were measured by quantitative enzymatic determinations using assay 46 kits obtained from Sigma. The principle of the cholesterol assay has been described previously (2.6.3). The enzymatic reactions involved in the triglyceride assay are tr iglyceride hydrolysis by lipase and measurment of the glycerol produced by coupled enzyme reactions catalyzed by glycerol kinase, glycerol-l-phosphate dehydrogenase and diaphorase. The intensity of the color of the resulting formazan produced is direct ly proportional to the triglyceride concentration of the sample. Total plasma l ipids were measured by a colorimetric assay kit obtained from Boehringer Mannheim. The assay is based on the ab i l i ty of l ipids to react with sulfuric acid and phosphoric acid and vani l l in to form a pink-colored complex. The intensity of the pink color (measured at 530 nm) is proportional to the concentration of the total l ipids present in the sample. 2.9.2 Glycosylated hemoglobin assay Hemoglobin glycosylation of rat red cel ls was measured colorimetri-cal ly according to the method of Winterhalter (189). Approximately 0.5 ml of saline-washed red blood cel ls were lysed with 1.8 ml of water and 0.4 ml of carbon tetrachloride with vigorous shaking at room temperature. The hemolysate was freed of cel lular debris by centrifugation, and the Hb concentration was subsequently adjusted to 50 mg/ml with H2O. To 2.0 ml of this hemolysate was added 1.0 ml of 1.0 N oxalic acid. After mixing, the solution was incubated at 100°C for 4.5 h and subsequently cooled to room temperature. One ml of 40% TCA was added, and, after mixing, the prec ip i -tate was removed by centrifugation. From the resulting supernatant, 2.0 ml are mixed with 0.5 ml of saturated TBA solution. After mixing, the solution was kept at 40°C for 30 min, then cooled to room temperature. The reaction of liberated 5-hydroxymethyl-furfural with TBA was measured at 443 nm, and the mean glycosylation of a l l hemoglobins was determined using an extinction coeff icient of 6 x 10 at 443 nm for HbA, (quantitatively is the most 47 prominent glycosylated form of Hb). HbA^c measurements for diabetic patients were performed at Shaughnessy Hospital (Diabetic Specialty Centre) by a more specif ic method, ion-exchange column chromatography. Because this method was not available in our laboratory, we were unable to measure HbA l c values in the red cel ls of healthy control subjects. 2.9.3 Water content Tissue samples were blotted l ight ly with f i l t e r paper and weighed in porcelain crucibles. The tissues were dried at 110°C overnight to constant weight and the dry weight was determined by subtraction. Water content is expressed as a percent of tissue wet weight. 3. STATISTICAL ANALYSIS Each value represents mean ± standard deviation (SD). Stat ist ical analyses were performed using either Student's t-test (for two group experi-ments) or one-way analysis of variance (ANOVA) using a significance level of P < 0.05. Specific group differences were determined using Tukey's test. In order to establish the relat ive variation in the values of each parameter studied, the coefficient of variation was employed (190). 4. RESULTS 4.1 Distribution of Tissue Antioxidants Experiments in control male rats indicated that, of the various tissues studied, the heart and pancreas possess the lowest act iv i t ies of CAT, CuZn-SOD, GSH-PX, and GSSG-RD as well as the lowest contents of the non-enzymatic antioxidant GSH. In female rats, the same pattern was observed except for GSH-PX in the kidney which showed re lat ive ly low act iv ity (Figs. 3-7). The l i ver , on the other hand, contained the highest antioxidant act iv ity regardless of animal gender. The kidney showed the 48 5 0 -1 4 0 -ffl X tr 3 0 -LtJ D t n t n • - I I - SO-L I \ T ID-HEART PANCREAS LIVER KIDNEY RBC Figure 3. Tissue d i s t r i bu t i on of CAT in female ra t s . Results shown are mean of 7 animals with the error bars ind icat ing SD. Values for heart and pancreas, which were too low to depict on the present sca le , are given in Tables 3 and 5. 49 60-i CD I a 48-bJ D in in - 36-I-U N Z x N I a. • < z _r a 24-12-0-HEART PANCREAS LIVER KIDNEY RBC Figure 4. Tissue distr ibution of GSH-PX in female rats. Values shown are mean of 7 animals with the error bars indicating SD. 50 3 5 0 0 - 1 2 8 0 0 -2 1 0 0 -o I tr • hi D Ul Ul I- 1 4 0 0 -O 23 7 0 0 -HEART PANCREAS LIVER KIDNEY RBC Figure 5. Tissue distr ibution of CuZn-SOD in female rats. Values shown are mean of 7 animals with the error bars indicating SD. 5 1 I CD I • UJ D w CJ N Z I 0. • < z X • i -HEART PANCREAS LIVER KIDNEY RBC Figure 6. Tissue distr ibut ion of GSSG-RD in female rats. Values shown are mean of 7 animals with the error bars indicating SD. 52 l O - i 8 -Ul D Ul U) 6 -13 • a. 4 -2 -HEART PANCREAS LIVER KIDNEY Figure 7. Tissue d i s t r i b u t i o n of GSH in female ra t s . Values shown are mean of 7 animals with the error bars ind icat ing SD. 53 highest act iv i ty of GSSG-RD in both male and female rats, other act iv i t ies (with the exception of low GSH-PX act iv ity in female rats) being higher than those of the heart and pancreas, but lower than those in the l iver (Figs. 3-7). 4.2 Chemically-Induced Diabetes in Female Rats 4.2.1 General features Induction of diabetes by STZ or ALX was confirmed by the presence of glucosuria within 24 h. In the f i r s t week of diabetes, the mortality rate was approximately 25%in the STZ-injected animals while i t was 75% in ALX-injected animals. Animals which survived this period (25/34 for STZ-treated and 20/85 for ALX-injected animals (not including 7 ALX-diabetic rats injected with insulin immediately after detection of diabetes) survived the remaining 12 wk period of diabetes. Three animals from the STZ-treated group and one from the ALX-treated group showed neither glucosuria nor hyperglycemia on subsequent periodic testings. In order to assess the possible direct toxic effects exerted by STZ and ALX, these animals were housed separately and after a 12 wk period al l the parameters reported in this study were measured in these animals. No abnormal values were found (data not shown) and these non-diabetic STZ- or ALX- treated animals were, therefore, excluded from our analysis. o Tables 1 and 2 show the various parameters measured to assess the diabetic state at the time of sacr i f ice. While control animals gradually gained weight during the 12 wk experimental period, the diabetic animals showed a variable response. Approximately 40% of the diabetic animals showed considerable emaciation and weight loss; others either did not show increases in body wt (40%) or had gained some weight after 12 wk of diabetes (20%). In general, diabetic animals had a lower mean body wt compared to either their own pretreatment value or to the body weights of control 54 Table 1. General features and plasma levels of glucose, l ipids and insulin in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. Control Diabetic untreated insulin-treated Body wt. (g) 290 ± 16 207 ± 49* 306 ± 8.50 + Heart wt. (g) 0.94 ± 0. 03 0.89 ± 0.07 1.03 ± 0.07* + Heart wt./body wt. (g/g x IO"2) 0.33 ± 0. 01 0.38 ± 0.04* 0.33 ± 0.02 + Left kidney wt. (g) 0.99 ± 0. 04 1.48 ± 0.14* 1.32 ± 0.08* f L. kidney wt./body wt. (g/g x IO - 2 ) 0,35 ± 0. 01 0.63 ± 0.09* 0.43 ± 0.02* + Glucose (mM) 7.99 ± 0. 64 32.38 8.72* 7.38 ± 3.59 + Cholesterol (mM) 1.97 ± 0. 38 4.19 ± 0.97* 1.92 0.46 + Triglycerides (mM) 1.15 ± 0. 29 10.48 6.41* 1.65 ± 0.76 + Phospholipid P (mM) 1.31 ± 0. 14 2.02 ± 0.68* 1.28 ± 0.23 + Insulin (uU/ml) 25.29 ± 4. 51 6.64 3.24* 115 ± 43* + HbA l c {%) 3.47 ± 0. 23 8.78 ± 1.20* 4.96 dz 0.37* + Values are expressed as mean ± SD, n = 7 in each group. *p < 0.5, s ignif icantly different from control, p < 0.5, s igni f icant ly different from diabetic. 55 Table 2. General features and plasma levels of glucose, l ipids and insulin in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk Body wt. (g) 291 ± 26 218 ± 29* 312 dz 19 + 306 dz 13 + Heart wt. (g) 0.92 ± 0.04 0.86 dz 0.08 0.98 dz 0.05 + 0.97 ± 0.08 + Heart wt./body wt. (g/g x 10" 2) 0.33 ± 0.02 0.40 zt 0.04* 0.31 dz 0.02+ 0.32 ± 0.03 + Left kidney wt. (g) 0.98 ± 0.08 1.40 dz 0.21* 1.15 dz 0.09+ 1.16 dz 0.09+ L. kidney wt./body wt. (g/g x IO"2) 0.35 ± 0.02 0.65 dz 0.10* 0.37 dz 0.03 + 0.38 dz 0.03 + Glucose (mM) 8.27 1.08 31.54 dz 4.56* 5.81 dz 1.79+ 8.81 dz 7.29+ Cholesterol (mM) 1.87 ± 0.32 2.72 dz 1.31 2.26 dz 0.37 2.06 dz 0.56 Triglycerides (mM) 1.19 ± 0.34 5.00 dz 5.08 1.28 dz 0.24 1.94 dz 1.77 Phospholipid P (mM) 1.31 ± 0.19 1.73 ± 0.29* 1.43 dz 0.27 1.46 dz 0.28 Insulin (iiU/ml) 25.93 dz 4.35 7.50 dz 2.53* 136 dz 44* + 131 dz 58* + HbA l c ( % ) 3.87 dz 0.18 8.18 dz 1.16* 4.27 dz 0.70 f 4.14 dz 0.49 + Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. 4-p < 0.05, s ignif icant ly different from diabetic. 56 animals. Diabetic animals showed higher kidney weights expressed either as absolute wet wt or wet wt/100 g body wt. Heart wet weights in diabetic animals were not s ignif icantly different from control, but, when expressed per 100 g body wt, s ignif icantly higher values were observed in the diabetic animals. Insulin treatment in STZ-diabetic animals (4 wk) normalized body weight and reduced, but did not abolish, the increased kidney weights. Insulin treatment in ALX-diabetic animals (6 and 12 wk) also normalized kidney weights. Total water contents of tissues examined were measured and normal values (presented as mean ± SD and expressed as%of tissue wet weight) are as follows: heart, 79.74 ± 1.32; pancreas, 77.34 ± 2.62; l i ver , 74.66 ± 2.95 and kidney, 82.06 ± 1.25). No changes in tissue water content were observed in either ALX- or STZ-diabetic animals (data not shown). In addition to hypoinsulinemia and hyperglycemia, the diabetic animals showed elevated Hb glycosylation (HbA-^) and a number of plasma l ip id abnormalities (Tables 1 and 2). The magnitude of these changes was variable, but tended to correlate with the degree of emaciation and, there-fore, was l ike ly an indication of the severity of the diabetic state. This was particularly evident in the plasma triglyceride levels in diabetic animals; emaciated diabetic animals were markedly hypertriglyceridemic while those diabetic animals which gained some weight showed relat ively normal plasma tr iglyceride values. Insulin treatment completely prevented plasma l ip id alterations, but these animals were found to be hyperinsulinemic and s l ight ly hypoglycemic. However, the levels of glycosylated Hb remained increased relat ive to controls in STZ-diabetic animals (4 wk treatment) but were normalized after 6 and 12 wk insulin treatment in ALX-diabetic animals. 4.2.2 Tissue antioxidant status The general pattern of changes seen in diabetic animals was a tendency I 57 of enzyme act iv i t ies which were low in control tissues to be increased and some of those which were high in control tissues to show some degree of reduction in diabetic animals. Thus, the hearts of both groups of diabetic animals showed an increased act iv ity of CAT and GSSG-RD (Table 3 and 4), while the pancreas showed increased act iv i t ies of CAT and GSSG-RD compared to controls (Table 5 and 6). No significant differences were observed in either cardiac GSH-PX, CuZn-SOD or pancreatic GSH-PX act iv i t ies in diabetic tissues as compared with corresponding control tissues. However, cardiac GSH levels were s ignif icantly increased and pancreatic levels s l ight ly decreased in ALX-diabetic rats, while pancreatic act iv i t ies of CuZn-SOD were s ignif icantly increased in STZ-diabetic rats. A similar pattern of changes was seen in pancreatic CuZn-SOD act iv i ty in ALX-diabetic rats, although these did not quite attain s tat i s t ica l significance. Liver showed the most marked decrease in antioxidant capacity of a l l the tissues studied. With the exception of GSSG-RD, the act iv i t ies of a l l the enzymes as well as GSH levels were s ignif icantly lower in diabetic than in control l ivers (Table 7 and 8). Less severe impairment was observed in the kidneys of diabetic rats, which exhibited reductions in CAT and CuZn-SOD act iv i t ies but increases in GSH-PX act iv ity (Table 9 and 10). Renal GSSG-RD act iv i t ies were not s ignif icant ly different from those of control rats. Insulin treatment was found to normalize hepatic GSH levels and a l ter -ations in the act iv i t ies of tissue scavenging enzymes (Tables 3-10). However, the CuZn-SOD act iv i ty in both l iver and kidney of insulin-treated diabetic rats, although normalized following 4 wk of treatment, was not s ignif icantly different from that in the corresponding non-treated diabetic tissues (Tables 7 and 9), and after 12 wk of insulin treatment was even s ignif icantly lower than the control values (Tables 8 and 10). 58 Table 3. Antioxidant status of heart in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. Control Diabetic untreated insulin-treated CAT (K/g wet wt.) 0.07 ± 0.04 0.94 ± 0.50* 0.05 ± 0.03 + CuZn-SOD (U/g wet wt.) 744 ± 73 781 ± 126 722 ± 55 GSH-PX (pmol NADPH/min /g wet wt.) 3.84 ± 0.90 4.11 ± 0.96 4.44 ± 0.60 GSSG-RD (pmol NADPH/min " /g wet wt.) 0.48 ± 0.07 0.67 0.11* 0.52 ± 0.07 + GSH (pmol/g wet wt.) 1.52 ± 0.11 1.57 ± 0.16 1.39 ± 0.15 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. 4-p < 0.05, s ignif icantly different from untreated diabetic. CAT, catalase; K, f irst-order rate constant; CuZn-SOD, CuZn-superoxide dismutase; GSH-PX, glutathione peroxidase; GSSG-RD, glutathione reductase; GSH, reduced glutathione. 59 Table 4. Antioxidant status of heart in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk CAT (K/g wet wt.) 0.09 ± 0. 03 0.87 ± 0.61* 0.06 dz 0.02 f 0.06 dz 0.05+ CuZn-SOD (U/g wet wt.) 745 ± 62 698 ± 113 768 dz 105 772 dz 84 GSH-PX (ymol NADPH /min/g wet wt.) 3.72 ± 0. 99 3.69 ± 0.36 4.23 dz 1.02 4.05 dz 0.54+ GSSG-RD (ymol NADPH /min/g wet wt.) 0.50 ± 0. 08 0.67 dz 0.04* 0.47 dz 0.04+ 0.49 dz 0.06+ GSH (ymol/g wet wt.) 1.46 ± 0. 09 1.84 ± 0.21* 1.53 ± 0.16+ 1.62 dz 0.16'' Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 60 Table 5. Antioxidant status of pancreas in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. Control Diabetic untreated insulin-treated CAT (K/g wet wt.) 0.15 ± 0.05 1.02 ± 0.38* 0.06 ± 0.04+ CuZn-SOD (U/g wet wt.) 675 ± 122 919 ± 137* 691 ± 162+ GSH-PX (umol NADPH/min /g wet wt.) 5.22 ± 1.20 5.94 ± 1.29 5.19 ± 0.78 GSSG-RD (umol NADPH/min /g wet wt.) 1.64 ± 0.19 2.09 ± 0.22* 1.67 ± 0.17+ . GSH (umol/g wet wt.) 1.55 ± 0.16 1.49 ± 0.22 1.59 ± 0.18 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 61 Table 6. Antioxidant status of pancreas in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk CAT (K/g wet wt.) 0.12 ± 0.05 0.65 ± 0.28* 0.08 ± 0.05+ 0.14 ± 0.08+ CuZn-SOD (U/g wet wt.) 678 ± 115 884 188 675 ± 213 608 ± 145+ GSH-PX (pmol NADPH/min /g wet wt.) 5.61 ± 1.32 6.63 ± 1.83 5.67 ± 1.71 5.61 ± 1.89 GSSG-RD (pmol NADPH/min /g wet wt.) 1.67 ± 0.17 2.01 ± 0.25* 1.58 ± 0.16+ 1.64 ± 0.13+ GSH (pmol/g wet wt.) 1.69 ± 0.16 1.37 ± 0.14 1.79 ± 0.42 1.83 ± 0.35+ Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control, ^p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 62 Table 7. Antioxidant status of l iver in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. . Control Diabetic untreated insulin-treated CAT (K/g wet wt.) 22.42 2.59 13.88 ± 3.91* 20.82 ± 4.43+ CuZn-SOD (U/g wet wt.) 3341 ± 490 2405 ± 545* 2934 ± 202 GSH-PX (pmol NADPH/min /g wet wt.) 31.23 ± 2.88 19.86 ± 8.64* 29.34 ± 4.05 + GSSG-RD (pmol NADPH/min /g wet wt.) 2.92 ± 0.36 3.21 ± 0.26 3.03 ± 0.18 GSH (pmol/g wet wt.) 6.64 ± 0.18 4.17 ± 1.04* 5.96 ± 0.82+ Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. +p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 63 Table 8. Antioxidant status of l iver in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk CAT (K/g wet wt.) 22.93 ± 2.87 14.63 ± 6.96* 19.30 ± 3.58 17.15 ± 6.65 CuZn-SOD (U/g wet wt.) 3392 ± 680 2585 ± 306* 3055 ± 342 2560 ± 470* GSH-PX (pmol NADPH/min /g wet wt.) 29.88 ± 3.69 19.80 ± 8.67* 29.58 ± 3.48+ 33.18 ± 7.53' GSSG-RD (ymol NADPH/min /g wet wt.) 3.03 ± 0.26 2.97 ± 0.52 2.82 ± 0.25 2.99 ± 0.24 GSH (pmol/g wet wt.) 6.88 ± 0.54 4.50 ± 1.41* 5.84 ± 1.06 5.68 ± 1.69 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. 4-p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 64 Table 9. Antioxidant status of kidney in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. Control Diabetic untreated insulin-treated CAT (K/g wet wt.) 2.77 ± 0.54 1.54 ± 0.46* 2.82 ± 0.43 + CuZn-SOD (U/g wet wt.) 1657 ± 371 1151 ± 207* 1472 ± 214 GSH-PX (pmol NADPH/min /g wet wt.) 4.02 ± 1.11 6.99 ± 0.99* 4.62 ± 0.57 + GSSG-RD (ymol NADPH/min /g wet wt.) 4.83 ± 0.35 5.20 ± 0.39 4.84 ± 0.38 GSH (pmol/g wet wt.) 2.86 ± 0.23 2.90 ± 0.43 3.26 ± 0.59 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 65 Table 10. Antioxidant status of kidney in. control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk CAT (K/g wet wt.) 2.64 ± 0.47 1.70 ± 0.36* 2.37 ± 0.76 2.28 ± 0.63 CuZn-SOD (U/g wet wt.) 1668 ± 344 1301 ± 113* 1381 ± 182 1334 ± 149* GSH-PX (ymol NADPH/min lq wet wt.) 4.41 ± 0.78 6.15. ± 0.78* 4.32 ± 1.35+ 4.26 ± 1.201 GSSG-RD (ymol NADPH/min /g wet wt.) 5.04 ± 0.33 5.12 ± 0.43 4.93 ± 0.36 4.98 ± 0.25 GSH (ymol/g wet wt.) 2.90 ± 0.33 2.88 ± 0.44 3.37 ± 0.48 3.28 ± 0.57 Values are expressed as mean * SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. f p < 0.05, s ignif icant ly different from untreated diabetic. The abbreviations used are the same as those in Table 3. 66 0 4.2.3 Erythrocyte enzymatic changes The act iv i t ies of CAT, CuZn-SOD, and GSH-PX as well as GSSG-RD were measured in the erythrocytes of control and diabetic rats. Significantly elevated act iv i t ies of GSSG-RD and GSH-PX were observed in erythrocytes of both groups of diabetic rats when compared to controls (Tables 11 and 12). Act iv i t ies of red cell CAT, CuZn-SOD and the membrane-associated oxidoreductase NADH-dehydrogenase in diabetic rats were not different from controls (Tables 11 and 12). Although insulin treatment was able to normalize red cel l GSH-PX act iv i t ies in diabetic rats, i t fa i led to prevent the diabetes-related increase in red cel l GSSG-RD act iv i t ies . This was evident in both STZ- and ALX-diabetes regardless of the duration of insulin treatment (Tables 11 and 12). 4.2.4 Erythrocyte susceptibi l ity to peroxidative damage In another series of experiments, red cel l s from control and diabetic rats were subjected to conditions of oxidative stress in v i tro. In the absence of H2O2, there were no differences in GSH or MDA levels between control and diabetic red cel ls (Tables 13,15 and 14,16). In control red ce l l s , H2O2 caused a concentration-dependent depletion of GSH followed, at somewhat higher concentrations, by a progressive increase in the production of MDA. In red cel ls from STZ-diabetic rats, MDA levels at the lowest concentrations of H2O2 were s ignif icant ly increased relat ive to controls, while GSH depletion was markedly reduced. Red cel ls from ALX-diabetic rats behaved similarly to those of STZ-treated rats and the data are summarized in Tables 15 and 16. GSH depletion was normalized in insulin-treated rats. Insulin treatment, regardless of duration, s i gn i f i -cantly lowered MDA production in red cel ls exposed to H2O2 (Tables 15,16). However, one of the 12 wk insulin-treated diabetic rats which showed hyperglycemia and relat ive hypoinsulinemia showed MDA levels elevated 67 Table 11. Antioxidant enzyme act iv i t ies of erythrocytes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. Control Diabetic untreated insulin-treated CAT (K/g Hb) 40.97 ± 8.15 38.47 ± 6.10 38.87 ± 2.81 CuZn-SOD (U/g Hb) 2979 ± 335 2996 ± 717 2995 ± 362 GSH-PX (ymol NADPH/min /g Hb) 52.74 ± 7.92 70.41 ± 6.39* 57.18 ± 4.38 + GSSG-RD (ymol NADPH/min /g Hb) 0.88 ± 0.10 1.43 ± 0.30* 1.36 ± 0.08* NADH-dehydrogen ase (nmol/min/mg protein) 26.04 ± 3.98 29.20 ± 2.40 28.90 ± 4.17 Values are expressed as mean ± SD, n 7 in each group. *p < 0.05, s ignif icantly different from control. +p < 0.05, s ignif icantly different from untreated diabetic. CAT, catalase; K, f irst-order rate constant; CuZn-SOD, CuZn-superoxide dismutase; GSH-PX, glutathione peroxidase; GSSG-RD, glutathione reductase. 68 Table 12. Antioxidant enzyme act iv i t ies of erythrocytes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk CAT (K/g Hb) 39.85 ± 6.74 40.38 ± 6.50 38.93 ± 2.03 39.02 3.95 CuZn-SOD (U/g Hb) 2989 ± 551 3181 ± 481 3025 ± 259 3240 ± 197 GSH-PX (umol NADPH /min/g Hb) 52.44 ± 10.1 69.06 ± 11.9* 55.05 ± 14.0 54.42 ± 5.01 GSSG-RD (pmol NADPH/min/g Hb) 0.90 ± 0.13 1.50 ± 0.20* 1.18 J t 0.16* + 1.21 ± 0.141 NADH-dehydrogenase (nmol/min/mg protein) 28.24 ± 3.23 27.73 ± 4.83 27.04 ± 4.51 26.90 ± 3.35 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icant ly different from control. +p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 11. 69 Table 13. H202-induced GSH depletion of erythrocytes in control, untreated STZ--diabetic and insulin-treated STZ-diabetic rats. GSH Content (%) Control Diabetic H 20 2 (mM) untreated insulin-treated 0.000 - 100 100 100 0.025 81.8 ± 6.4 96.1 ± 4.8* 81.3 ± 8.1 + 0.050 46.6 ± 9.5 91.3 ± 7.8* 51.7 ± 11.4+ 0.250 7.4 ± 4.2 64.7 ± 7.4* 6.0 ± 4.4+ 0.500 5.3 ± 1.7 48.0 ± 11.9* 4.0 ± 2.0+ 1.000 4.1 ± 2.3 27.9 ± 13.6* 4.3.± 1.7+ Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icant ly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. GSH levels in red cells incubated in the absence of added H 2 0 2 (see Methods, 2.8.1) were 3.19 t °- 5 7» 3.17 + 0.44, and 3.38 + 0.36 ymol/g Hb for control, untreated diabetic and insulin-treated diabetic animals, respectively. 70 Table 14. ^2^2-^ n c ' u c e c ' S^H depletion of erythrocytes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. GSH Content (%) Control Di abetic H 20 2 (mM) untreated insulin--treated 6 wk 12 wk 0.000 100 100 100 100 0.025 76.1 ± 9.4 99.1 ± 5.0* 74.3 ± 8.1 + 79.7 ± 18.3 + 0.050 45.8 ± 12.3 86.4 ± 9.9* 43.3 ± 6.3 + 61.1 ± 30.4 + 0.250 5.6 ± 2.2 57.0 ± 12.3* 4.3 ± 2.0 + 23.7 ± 34.6+ 0.500 5.0 ± 1.5 40.0 ± 7.3* 2.8 ± 1.5 + 12.8 ± 16.5+ 1.000 3.8 ± 2.5 23.1 ± 8.9* 3.0 ± 1.9 + 5.9 ± 3.2 + Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. t p < 0.05, s ignif icantly different from untreated diabetic. GSH levels in red cells incubated in the absence of added H 2 0 2 (see Methods, 2.8.1) were 3.39 + 0.51, 3.64 + 0.76, 3.59 + 0.48 and 3.53 + 0.54 ymol/g Hb for control, untreated diabetic and insulin-treated (6 wk and 12 wk) diabetic animals, respectively. 71 Table 15. H 20 2-induced MDA production of erythrocytes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. MDA Content7 Control Diabetic 2U2 v " " , ; untreated insulin-treated 0.00 4.7 ± 4.4 7.7 7.3 4.8 ± 3.0 0.50 19.6 ± 4.5 56.9 ± 11.6* 19.2 ± 4.7 + 1.00 106 ± 73 482 ± 165* '• 27.7 ± 7.8+ 1.50 354 ± 83 622 ± 90* 140 ± 63 * + 2.00 563 ± 76 740 ± 75* 305 ± 77*+ 3.00 761 ± 102 868 ± 115 540 ± 79*+ Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. +p < 0.05, s ignif icantly different from untreated diabetic. VMDA production is expressed as nmol/g Hb. 72 Table 16. h^C^-induced MDA production of erythrocytes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. MDA Content7 Control Diabetic untreated insulin-treated H2°2 ^ • 6^k" i7~ 0.00 3.6 ± 4.4 3.0 ± 2.8 8.3 ± 5.9 5.5 ± 4.8 (10.5) 0.50 22.1 ± 5.5 34.9 ± 8.1* 22.5 ± 8.8 + 23.2 5.9 + (39.8) 1.00 138 ± 98 252 ± 69* 54.2 44.7+ 28.1 ± 15.5 (292) 1.50 380 134 594 ± 82* 168 ± 91*+ 124 ± 84*+ (608) 2.00 554 ± 90 734 ± 69* 350 ± 9 5 * t 274 ± 130*+ (642) 3.00 729 ± 108 929 89* 604 ± 78+ 551 ± 184+ (952) Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. VMDA production is expressed as nmol/g Hb. Values shown in parentheses are results obtained with one aberrant insul in-treated hyperglycemic rat - these data were not included in our analysis. 73 to a similar extent as those observed in red cel ls from non-treated diabetic rats (Table 16). 4.2.5 Erythrocyte membrane and plasma l i p id analyses The previously noted abnormalities in plasma l ip id components and the increased susceptibi l i ty of red cel l membrane l ipids to I^G^-induced peroxidation in STZ- or ALX-diabetic rats prompted us to undertake a detailed study of plasma and red cel l membrane l i p id characteristics. Despite the increases in plasma cholesterol and phospholipid concentrations in diabetic rats, the levels of these l ipids in red cel l s were not s i g n i f i -cantly altered, although a small (non-significant) decrease in the molar cholesterol (C)/phospholipid (P) rat io was found in both groups of diabetic rats (Tables 17 and 18). In red ce l l s , SM content was s ignif icantly reduced in both groups of diabetic rats, leading to an increased PC/SM rat io, and PE was increased in erythrocytes of ALX-diabetic rats (Tables 19 and 20). Analyses of plasma phospholipid prof i les, on the other hand, revealed significant alterations in both ALX- and STZ-treated rats (Tables 21 and 22). These included significant elevations in PC (resulting in an increase in PC/SM rat io) , PE and PI fractions in both groups of diab'etic rats and a decrease in LPC that attained stat i s t ica l significance only in STZ-diabetes. Insulin treatment for 4 wk fai led to normalize red cel l membrane SM levels or PC/SM ratios in STZ-diabetic rats, while longer treatment (6-12 wk) of ALX-diabetic rats did prevent the decrease in SM levels and the increase in PC/SM ratios (Tables 19 and 20). Alterations in plasma phospho-l ip id profi les were normalized by insulin treatment except for a significant increase in PC/SM ratio after 12 wk of insulin treatment in ALX-diabetic rats (Tables 21 and 22). 74 Table 17. Cholesterol and phospholipid contents of erythrocyte membranes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. Control Diabetes untreated insulin-treated Cholesterol (C) (umol/mg protein) 0.54 ± 0.06 0.49 ± 0.04 0.52 ± 0.03 Phospholipid P (P) (umol/mg protein) 0.53 ± 0.05 0.52 ± 0.06 0.53 ± 0.02 C/P ratio 1.03 ± 0.07 0.95 ± 0.05 0.98 ± 0.04 Values are. expressed as mean ± SD, n = 7 in each group. 75 Table 18. Cholesterol and phospholipid contents of erythrocyte membranes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. Control Diabetic untreated insulin-treated 6 wk 12 wk Cholesterol (C) 0.51 ± 0.06 0.50 ± 0.04 0.51 ± 0.02 0.53 ± 0.04 (umol/mg protein) Phospholipid P (P) 0.52 ± 0.07 0.55 ± 0.05 0.54 ± 0.05 0.54 ± 0.04 (umol/mg protein) C/P ratio 0.98 ± 0.05 0.91 ± 0 . 0 6 0.94 ± 0.05 0.98 ± 0.05 Values are expressed as mean ± SD, n = 7 in each group. 76 Table 19. Phospholipid composition of erythrocyte membranes in control, untreated STZ-diabetic and insulin-treated STZ-diabetic rats. nmol Phospholipid Phosphorus/mg protein Control Diabetes untreated insulin-treated SM 52.6 ± 12.3 34.8 ± 8.4* 39.7 ± 5.0* PI 36.2 ± 7.5 36.1 ± 10.2 29.3 ± 5.7 PS 70.5 ±. 8.7 59.3 ± 11.9 70.5 ± 4.1 PC 236.0 ± 18.0 248.0 ± 35.5 250.0 ± 20.0 PE 136.0 ± 18.0 146.0 ± 12.4 142.0 ± 5.0 PC/SM 4.5 ± 1.1 7.3 ± 1.2* 6.4 ± 0.9* Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icant ly different from control. SM, sphingomyelin; PI, phosphatidyl inos ito l ; PS, phosphatidylserine; PC, phosphatidylcholine; PE, phosphatidylethanolamine. 77 Table 20. Phospholipid composition of erythrocyte membranes in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. nmol Phospholipid Phosphorus/mg protein Control Diabetic untreated insulin-treated 6 wk 12 wk SM 50.9 ± 7.2 32.7 ± 8.4* 41.0 ± 5.6 42.5 dz 6.5 PI 32.1 ± 7.3 37.4 ± 5.6 34.7 ± 9.5 32.8 dz 7.6 PS 70.4 ± 15.6 69.3 ± 8.9 72.9 ± 7.4 71.7 dz 5.6 PC 236 ± 36 253 ± 22 253 ± 18 248 dz 20 PE 132 ± 15 164 ± 21* 144 ± 15 141 dz 17 PC/SM 4.7 ± 1.2 8.9 ± 3.0* 6.3 ± 1.0 + 5.9 dz 0.7' Values are expressed as mean ± SD, n 7 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 19. 78 Table 21. Phospholipid composition of plasma in control, untreated STZ-diabetic, and insulin-treated STZ-diabetic rats. umol Phospholipid Phosphorus/L Control Diabetic untreated insulin-treated LPC 260 ± 36.0 164 ± 65.9* 281 46.0+ SM 97.9 ± 31.8 89.8 ± 20.7 67.7 ± 22.7 PI 72.5 ± 19.9 113 ± 26.3* 67.3 ± 27.9"+ PC 784 ± 80.0 1512 ± 644* 844 ± 160+ PE 33:5 22.0 106 67.0* 22.1 ± 6.9+ PC/SM 8.99 ± 3.73 16.7 ± 4.4* . 13.0 ± 2.8 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icant ly different from control. + p < 0.05, s ignif icantly different from untreated diabetic. LPC, lysophosphatidylcholine; SM, sphingomyelin; PI, phosphatidyl inos ito l ; PC, phosphatidylcholine; PE, phosphatidylethanolamine. 79 Table 22. Phospholipid composition of plasma in control, untreated ALX-diabetic and insulin-treated ALX-diabetic rats. pmol Phospholipid Phosphorus/L Control Diabetic untreated insulin-treated 6 wk 12 wk LPC 271 ± 34.0 182 ± 76 282 ± 63 + 278 ± 74+ SM 97.4 ± 22.7 73.9 ± 22.8 80.8 ± 14.7 80.4 ± 15.3 PI 73.0 ± 11.9 116 ± 47.0* 75.6 ± 25.2 76.6 26.3 PC 824 ± 155 1261 ± 231* 982 ± 174 998 ± 184 PE 32.9 ± 14.6 82.4 ± 34.2* 21.9 ± 6.7+ 28.1 ± 15.4+ PC/SM 8.7 ± 1.9 17.7 ± 3.1* 12.2 ± 2.6+ 13.0 2.9*+ Values are expressed as mean ± SD, n 7 in each group. *p < 0.05, s ignif icantly different from control, p < 0.05, s ignif icantly different from untreated diabetic. The abbreviations used are the same as those in Table 21. 80 4.3 Effects of Allopurinol Treatment in ALX-Diabetic Animals 4.3.1 General features Although the intraperitoneal injection of ALP (50 mg/kg body wt.) prior to the administration of ALX did not prevent diabetes, i t did lower the level of associated ketonuria. While a l l of the 8 non-ALP-treated rats died in ketosis during the f i r s t week, only one of the ALP-treated rats died in ketosis and the remaining 7 animals showed glucosuria and slight ketonuria. The continuous administration of ALP in drinking water (approximately 1.9 mg/day) for the remaining 12 wk period of diabetes had, generally, no effect on the various parameters observed in ALX-diabetic rats, except for significant increases in plasma levels of triglycerides and phospholipid in ALP-treated ALX-diabetic rats (Table 23). It is important to mention that in order to conveniently i l lus t rate the effects of ALP-treatment in both control and ALX-diabetic animals, Tables 23-33 contain the values for non-ALP-treated control and ALX-diabetic animals which were previously presented in ear l ier tables. 4.3.2 Tissue antioxidant status ALP treatment did not affect the alterations in antioxidant status of most tissues in ALX-diabetic rats (Tables 24-27). ALP treatment did, however, prevent the observed decrease in renal CuZn-SOD act iv i t ies in ALX-diabetic rats (Table 27). Surprisingly, pancreatic GSH levels in ALP-treated control rats were s ignif icantly lower than those of non-ALP-treated controls, while the expected decrease in pancreatic GSH levels in ALX-diabetic rats was not seen in ALP-treated ALX-diabetic rats. Instead, a sl ight increase in pancreatic GSH levels was observed in ALP-treated diabetic rats (Table 25). 4.3.3 Erythrocyte enzymatic changes The administration of ALP did not alter the act iv i t ies of antioxidants 81 Table 23. General features and plasma levels of glucose, l ipids and insulin in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control di abetic Body wt. (g) 291 26 218 29* 290 ± 15 234 ± 39* Heart wt. (g) 0.92 ± 0.04 0.86 ± 0.08 0.90 ± 0.07 0.96± • 0.04 t + Heart wt./body wt. (g/g x IO - 2 ) 0.33 0.02 0.40 ± 0.04* 0.31 0.02 0.38± 0.05* Left kidney wt. (g) 0.98 ± 0.08 1.40 ± 0.21* 1.08 ± 0.08 1.61 ± 0.26* L. kidney wt./body wt. (g/g x IO - 2 ) 0.35 ± 0.02 0.65 ± 0.10* 0.37 ± 0.02 0.64 ± 0.16* Glucose (mM) 8.27 ± 1.08 31.54 ± 4.56* 8.22 ± 0.57 31.7 ± 9.34* Cholesterol (mM) 1.87 ± 0.32 2.72 ± 1.31 1.72 ± 0.45 2.97 ± 1.56 Triglycerides (mM) 1.19 ± 0.34 5.00 ± 5.08 1.43 ± 0.35 7.72 ± 4.80* Phospholipid P (mM) 1.31 ± 0.19 1.73 ± 0.29 1.47 0.27 2.71 ± 1.14* Insulin (viU/ml) 25.93 ± 4.35 7.56 ± 2.53* 30.9 ± 10.54 7.78 ± 4.18* HbA l c {%) 3.87 ± 0.18 8.18 ± 1.16* 3.82 ± 0.16 8.04 ± 1.81* tt Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. + + p < 0.05, s ignif icantly different from untreated ALX-diabetic. Values for control and untreated ALX-diabetes are taken from those in Table 2 for comparison with ALP-treated groups. 82 Table 24. Antioxidant status of heart in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control diabetic CAT (K/g wet wt.) 0.09 ± 0.03 0.87 ± 0.61* 0.08 ± 0.04 0.80 ± 0.49' CuZn-SOD (U/g wet wt.) 745 ± 62 698 ± 113 855 ± 56 760 ± 67 GSH-PX (ymol NADPH /min/g wet wt.) 3.72 0.99 3.69 ± 0.36 3.99 ± 0.45 3.87 ± 0.45 \ GSSG-RD (pmol NADPH /min/g wet wt.) 0.50 ± 0.08 0.67 ± 0.04* 0.50 ± 0.06 0.68 ± 0.09' GSH (ymol/g wet wt.) 1.46 ± 0.09 1.84 ± 0.21* 1.46 ±. 0.09 1.74 ± 0.19' Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 3. Values for control and ALX-diabetes are taken from those in Table 4 for comparison with ALP-treated groups. 83 Table 25. Antioxidant status of pancreas in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control diabetic CAT (K/g wet wt.) 0.12 ± 0.05 0.65 ± 0.28* 0.17 ± 0.07 0.80 ± 0.45* CuZn-SOD (U/g wet wt.) 678 ± 115 884 ± 188* 674 ± 126 827 ± 100 GSH-PX (umol NADPH /min/g wet wt.) 5.61 ± 1.32 6.63 ± 1.83 5.37 ± 0.66 6.06 ± 0.78 GSSG-RD (pmol NADPH /min/g wet wt.) 1.67 ± 0.17 2.01 0.25* 1.54 ± 0.12 2.07 ± 0.33* GSH (umol/g wet wt.) 1.69 ± 0.16 1.37 0.14* 1.29 ± 0.21** 1.53 ± 0.21 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. **p < 0.05, s ignif icant ly different from non-ALP-treated control. The abbreviations used are the same as those in Table 3. Values for control and ALX-diabetes are taken from those in Table 6 for comparison with ALP-treated groups. 84 Table 26. Antioxidant status of l iver in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control diabetic CAT (K/g wet wt.) 22.93 ± 2.87 14.63 6.96* 21.49 ± 5.16 16.66 ± 5.39 CuZn-SOD (U/g wet wt.) 3392 ± 680 2585 ± 306* 3400 236 2835 ± 509 GSH-PX (wmol NADPH /min/g wet wt.) 29.88 ± 3.69 19.80 ± 8.67* 31.59 ± 5.94 17.13 ± 5.67' GSSG-RD (umol NADPH /min/g wet wt.) 3.03 ± 0.26 2.97 ± 0.52 3.01 ± 0.27 3.23 dz 0.22 GSH (umol/g wet wt.) 6.88 ± 0.54 4.50 ± 1.41* 5.84 ± 0.54 4.67 dz 1.27 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 3. Values for control and ALX-diabetes are taken from those in Table 8 for comparison with ALP-treated groups. 85 Table 27. Antioxidant status of kidney in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control diabetic CAT (K/g wet wt.) 2.64 ± 0.47 1.70 ± 0.36* 2.96 0.43 1.86 ± 0.62* CuZn-SOD (U/g wet wt.) 1668 ± 344 1301 ± 113* 1607 ± 156 1546 ± 196 GSH-PX (pmol NADPH /min/g wet wt.) 4.41 ± 0.78 6.15 ± 0.78* 4.23 ± 0.51 6.54 ± 1.65* GSSG-RD (pmol NADPH /min/g wet wt.) 5.04 ± 0.33 5.12 ± 0.43 4.94 0.37 5.26 ± 0.29 GSH (pmol/g wet wt.) 2.90 ± 0.33 2.88 ± 0.44 2.87 ± 0.23 3.28 ± 0.35 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 3. Values for control and ALX-diabetes are taken from those in Table 10 for comparison with ALP-treated groups. 86 in erythrocytes of either control or diabetic rats compared to those observed in non-ALP-treated animals (Table 28). 4.3.4 Erythrocyte susceptibi l ity to peroxidative damage Similar patterns of GSH depletion were observed in ALP-treated control and diabetic red cel ls compared with those of non-ALP-treated groups (Table 29). However, ALP-treatment per se lowered MDA production in response to H2O2 in control rats compared with control non-ALP-treated rats although this only attained stat i s t ica l significance at a concentration of 2.0 mM H2O2. On the other hand, the increase in MDA production at a concentra-tion of 1.0 mM H2O2 in ALX-diabetic rat erythrocytes was normalized by ALP-treatment (Table 30). 4.3.5 Erythrocyte membrane and plasma l ip id analyses No changes were observed in red cel l membrane contents of cholesterol, phospholipid or C/P ratio in ALP-treated groups as compared with those of non-ALP-treated animals (Table 31). The decrease in SM content and the increases in PC/SM ratio and PE content of diabetic red cel l membranes were not altered by ALP-treatment (Table 32). Plasma phospholipid components of ALP-treated diabetic rats showed significant increases in LPC, PC and in PC/SM ratio^compared with non-ALP-treated diabetic plasma (Table 33). 4.4 Food-Deprivation 4.4.1 General features Table 34 shows the various parameters measured in both control and food-deprived rats. Food-deprivation for 72 h produced a weight loss of 16%. In addition to hypoglycemia and hypoinsulinemia, food-deprived animals showed a significant decrease in plasma phospholipid and total l i p i d levels. 4.4.2 Tissue antioxidant status In food-deprived animals, CAT act iv ity in heart (Table 35) and pancreas (Table 36) was s ignif icantly increased and that of l iver decreased 87 Table 28. Antioxidant enzyme act iv i t ies of erythrocytes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control diabetic CAT (K/g Hb) 39.85 ± 6.74 40.38 ± 6.50 40.49 ± 4.39 39.16 ± 3.22 CuZn-SOD (U/g Hb) 2989 ± 551 3181 ± 481 3342 ± 352 3200 ± 334 GSH-PX (umol NADPH /min/g Hb) 52.44 ± 10.1 69.06 ± 11.9* 53.49 ± 7.59 65.07 ± 10.3J GSSG-RD (umol NADPH /min/g Hb) 0.90 0.13 1.50 ± 0.20* 0.97 ± 0.06 1.44 ± 0.23' NADH-dehydrogen ase (nmol/min/mg protein) 28.24 ± 3.23 27.73 ± 4.83 32.77 ± 0.89 32.84 ± 5.25 Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icant ly different from control. The abbreviations used are the same as those in Table 11. Values for control and ALX-diabetes are taken from those in Table 12 for comparison with ALP-treated groups. 88 Table 29. H 20 2-induced GSH depletion of erythrocytes in control, untreated'ALX-diabetic and allopurinol-treated control and diabetic rats. GSH Content [%) Control ALX-diabetic Allopurinol -treatment H 20 2 (mM) control diabetic 0.000 100 100 100 100 0.050 45.8 ± 12.3 86.4 ± 9.9* 41.4 ± 4.2 89.7 ± 18.8* 0.250 5.6 ± 2.2 57.0 ± 12.3* 2.7 ± 1.4 44.4 ± 21.2* 0.500 5.0 ± 1.5 40.0 ± 7.3* 1.9 ± 0.7 29.9 ± 15.5* 1.000 3.8 ± 2.5 23.1 ± 8.9* 2.2 ± 0.4 14.6 ± 8.4* Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. Values for control and ALX-diabetes are taken from those in Table 14 for comparison with ALP-treated groups. GSH levels in red cel ls incubated in the absence of added H 2 0 2 (see Methods, 2.8.1) were 3.39 + 0.51,, 3.64 1 0.76, 3.63 1 0.46 and 3.68 + 0.59 ymol/g Hb for control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats, respectively. 89 Table 30. H202-induced MDA production of erythrocytes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. MDA Content7 Control ALX-diabetic Allopurinol-treatment 2°2 { m M ) control diabetic 0.00 3.5 ± 4.4 3.0 ± 2.8 4.3 3.6 6.7 ± 6.5 0.50 22.1 5.5 34.9 ± 8.1* 23.5' ± 5.5 37.8 ± 12.6* 1.00 138 ± 98 252 ± 69* 48.7 ± 2.8 136 ± 5 9 t + 1.50 380 ± 134 594 ± 82* 225 ± 49 659 ± 149* 2.00 554 ± 90 734 ± 69* 396 ± 64** 830 ± 133* 3.00 729 ± 108 929 ± 89* 643 ± 68 920 ± 93* Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. **p < 0.05, s ignif icantly different from non-ALP-treated control. T T p < 0.05, s ignif icant ly different from ALX-diabetic. v MDA production is expressed as nmol/g Hb Values for control and ALX-diabetes are taken from those in Table 16 for comparison with ALP-treated groups. 90 Table 31. Cholesterol and phospholipid contents of erythrocyte membranes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. Control ALX-diabetic Allopurinol-treatment control diabetic Cholesterol (C) (umol/mg protein) 0.51 ± 0.06 0.50 ± 0.04 0.52 ± 0.04 0.50 ± 0.04 Phospholipid P (P) 0.52 ± 0.07 0.55 ± 0.05 0.53 ± 0.03 0.55 ± 0.03 (umol/mg protein) C/P ratio 0.98 ± 0.05 0.91 ± 0.06 0.98 ± 0.05 0.91 ± 0.06 Values are expressed as mean ± SD, n = 7 in each group. Values for control and ALX-diabetes are taken from those in Table 18 for' comparison with ALP-treated groups. 91 Table 32. Phospholipid composition of erythrocyte membranes in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. nmol Phospholipid Phosphorus/mg protein Control ALX-diabetic Allopurinol-treatment control diabetic SM 50.9 ± 7.2 32.7 ± 8.4* 47.6 ± 8.6 26.1 ± 3.4* PI 32.1 ± 7.3 37.4 ± 5.6 31.1 ± 4.8 30.4 ± 4.2 PS 70.4 ± 15.6 69.3 ± 8.9 70.2 ± 10.6 74.4 ± 7.9 PC 236 ± 36 253 ± 22 242 ± 16 252 ± 14 PE 132 ± 15 164 ± 21* 138 ± 9.0 161 ± 17* PC/SM 4.7 ± 1.2 8.9 ± 3.0* 5.2 ± 1.0 9.9 ± 0.9* Values are expressed as mean ± SD, n = 7 in each group, *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 19. Values for control and ALX-diabetes are taken from those in Table 20 for comparison with ALP-treated groups. 92 Table 33. Phospholipid composition of plasma in control, untreated ALX-diabetic and allopurinol-treated control and diabetic rats. ymol Phospholipid Phosphorus/L Control ALX-diabetic Allopurinol-treatment control diabetic LPC 271 ± 34.0 ,182 ± 76* 305 57 298 ± 62++ SM 97.4 ± 22.7 73.9 ± 22.8 110 ± 21.2 85.4 ± 25.2 PI 73.0 ± 11.9 116 ± 47.0 66.6 ± 18.9 147 ± 53.0* PC 824 ± 155 1261 ± 231 962 ± 184 2066 ± 965* + t PE 32.9 ± 14.6 82.4 ± 34.2 30.2 ± 11.9 118 ± 74.0* PC/SM 8.7 ± 1.9 17.7 ± 3.1* 8.8 ± 1.3 23.8 ± 6.4*++ Values are expressed as mean ± SD, n = 7 in each group. *p < 0.05, s ignif icantly different from control. T T p < 0.05, s ignif icantly different from ALX-diabetic. The abbreviations used are the same as those in Table 21. Values for control and ALX-diabetes are taken from those in Table 22 for comparison with ALP-treated groups. 93 Table 34. General features and plasma levels of glucose, l ipids and insulin in control and food-deprived rats. Control Food-deprived P value Body weight (g) 213 ± 14 169 ± 4 < 0.05 Heart wt. (g) 0.74 ± 0. 09 0 .64 ± 0. 03 < 0.05 Heart wt./body wt. (g/g x IO - 2 ) " 0.35 ± 0. 03 0 .38 0. 02 N.S. Left kidney wt. (g) 0.76 ± 0. 09 0 .67 ± 0. 04 < 0.05 L. kidney wt./body wt. (g/g x 1(T 2) 0.36 ± 0. 02 0 .40 ± 0. 03 < 0.05 Glucose (mM) 5.16 ± 0. 13 4 .94 ± 0. 09 < 0.05 Cholesterol (mM) 1.61 ± 0. 29 1 .23 ± 0. .38 N.S. Triglycerides (mM) 0.68 0. 29 0 .58 ± 0. 16 N.S. Phospholipid P (mM) 0.88 ± 0. 10 0 .72 ± 0. 14 < 0.05 Total l ipids (g/L) 4.39 ± 0. 65 3 .21 ± 0. ,70 < 0.05 Insulin (pU/ml) '23.58 ± 7. 73 15 .58 ± 2. ,15 <0.05 Values are expressed as mean ± SD, n N.S. = non-significant. = 6 in each group. 94 Table 35. Antioxidant status of heart in control and food-deprived rats. Control Food-deprived P value CAT (K/g wet wt.) 0.05 ± 0.03 0.35 ± 0.09 < 0.05 CuZn-SOD (U/g wet wt.) 732 ± 30 619 ± 92 < 0.05 GSH-PX (umol NADPH/min/g wet wt.) 4.17 ± 1.53 4.59 ± 1.08 N.S. GSSG-RD (ymol NADPH/min/g wet wt.) 0.44 ± 0.07 0.44 dz 0.03 N.S. GSH (ymol/g wet wt.) 1.60 ± 0.09 1.62 ± 0.16 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations used are the same as those in Table 3. 95 Table 36. Antioxidant status of pancreas in control and food-deprived rats. Control Food-deprived P value CAT (K/g wet wt.) 0.12 ± 0.10 0.25 ± 0.09 < 0.05 CuZn-SOD (U/g wet wt.) 568 ± 125 857 ± 251 < 0.05 GSH-PX (pmol NADPH/min/g wet wt.) 5.31 ± 1.32 4.89 dz 1.20 N.S. GSSG-RD (pmol NADPH/min/g wet wt.) 1.80 ± 0.10 1.86 dz 0.13 N.S. GSH (ymol/g wet wt.) 1.46 ± 0.12 1.64 dz 0.23 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations used are the same as those in Table 3. 96 (Table 37), while the kidney showed no change (Table 38). CuZn-SOD act iv i ty in the heart was decreased (Table 35) and that in the pancreas (Table 36) and kidney was increased (Table 38), while no changes were observed in l iver (Table 37). The act iv ity of GSH-PX was s ignif icantly increased only in the kidneys of food-deprived rats (Table 38), while GSH levels were decreased only in l ivers of food-deprived rats (Table 37). Tissue GSSG-RD act iv it ies were indistinguishable from those of controls (Tables 35-38). 4.4.3 Erythrocyte enzymatic changes The act iv i t ies of CAT, CuZn-SOD and GSH-PX as well as the act iv ity of the membrane-bound oxido-reductase NADH-dehydrogenase were not affected by food-deprivation. However, s ignif icant ly lower act iv i t ies of GSSG-RD were observed in erythrocytes of food-deprived rats (Table 39). 4.4.4 Erythrocyte susceptibi l i ty to peroxidative damage When red cel ls from both groups were subjected to in v itro challenge with H2O2, MDA levels in food-deprived animals were s ignif icantly lower than control values at H 20 2 concentrations as low as 1.0 mM, while GSH depletion was ' markedly enhanced. In the absence of H 2 0 2 , there were no differences between GSH or MDA levels in erythrocytes of control and food-deprived rats (Tables 40 and 41). 4.4.5 Erythrocyte membrane and plasma l i p id analyses No signif icant changes were observed either in erythrocyte membrane contents of cholesterol or phospholipid (Table 42), or in individual phospholipid components in food-deprived animals (Table 43). Despite the decrease in plasma phospholipid concentration in food-deprived animals, the individual phospholipid profiles were indistinguishable from those of control plasma (Table 44). 97 Table 37. Antioxidant status of l iver in control and food-deprived rats. Control Food-deprived P value CAT (K/g wet wt.) 15.38 ± 5.34 6.10 ± 1.98 < 0.05 CuZn-SOD (U/g wet wt.) 2859 ± 489 3197 ± 713 N.S. GSH-PX (umol NADPH/min/g wet wt.) 31.02 ± 7.38 27.42 ± 8.37 N.S. GSSG-RD (umol NADPH/min/g wet wt.) 2.60 ± 0.26 2.53 ± 0.59 N.S. GSH (umol/g wet wt.) 5.80 ± 0.83 4.32 ± 0.88 < 0.05 Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations used are the same as those in Table 3. 98 Table 38. Antioxidant status of kidney in control and food-deprived rats. Control Food-deprived P value CAT (K/g wet wt.) 2,89 ± 0.36 2.95 ± 0.42 N.S. CuZn-SOD (U/g wet wt.) 1368 ± 99 1605 ± 216 N.S. GSH-PX (pmol NADPH/min/g wet wt.) 3.99 ± 0.93 5.46 ± 0.99 < 0.05 GSSG-RD (pmol NADPH/min/g wet wt.) 5.50 ± 0.73 5.77 0.48 N.S. GSH (pmol/g wet wt.) 3.50 ± 0.30 3.33 ± 0.32 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations used are the same as those in Table 3. 99 Table 39. Antioxidant enzyme act iv i t ies of erythrocytes in control and food-deprived rats. Control Food-deprived P value CAT (K/g Hb) 39.45 ± 3.11 37.68 ± 4.37 N.S. CuZn-SOD (U/g Hb) 3346 ± 342 3057 ± 450 N.S. GSH-PX (umol NADPH/min/g Hb) 48.15 ± 13.11 46.68 ± 4.59 N.S. GSSG-RD (umol NADPH/min/g Hb) 1.10 ± 0.24 0.64 ± 0.12 < 0.01 NADH-dehydrogenase (nmol/min/mg protein) 23.27 ± 3.20 25.93 ± 1.85 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations used are the same as those in Table 11. 100 Table 40. H202-induced GSH depletion of erythrocytes in control and food-deprived rats. GSH Content (% ) H 20 2 (mM) Control Food-deprived P value 0.000 100 100 0.025 70.3 ± 5.8 56.2 ± 5.2 < 0.05 0.050 37.7 ± 3.7 19.0 ± 4.3 < 0.05 0.250 4.0 ± 1.3 4.6 ± 1.1 N.S. 0.500 3.5 ± 1.0 4.6 ± 2.4 N.S. 1.000 3.8 ± 1.3 3.4 ± 1.4 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. GSH levels in red cells incubated in the absence of added H 20 2 (see Methods, 2.8.1) were 4.05+0.33 and 4.03 +0.22 iimol/g Hb for control and food-deprived rats, respectively. 101 Table 41. r^Og-induced MDA production of erythrocytes in control and food-deprived rats. MDA Content7 2°2 (m M) Control Food-deprived P value 0.00 1.0 ± 2.0 1.0 ± 2.0 N.S. 0.50 15.0 ± 8.0 22.0 ± 13.0 N.S. 1.00 243 ± 26 96.0 ± 25.0 < 0.05 1.50 445 ± 33 312 ± 46 < 0.05 2.00 658 ± 88 506 ± 64 < 0.05 3.00 922 ± 59 808 ± 75 < 0.05 Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. VMDA production is expressed as nmol/g Hb. 102 Table 42. Cholesterol and phospholipid contents of erythrocyte membranes in control and food-deprived rats. Control Food-deprived P value Cholesterol (C) 0.52 ± 0.03 0.52 ± 0.05 N.S. (umol/mg protein) Phospholipid P (P) 0.54 ± 0.04 0.55 ± 0.05 N.S. (umol/mg protein) C/P ratio 0.95 ± 0.04 0.93 ± 0.07 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. 103 Table 43. Phospholipid composition of erythrocyte membranes in control and food-deprived rats. nmol Phospholipid Phosphorus/mg protein Control Food-deprived P value SM 42.2 ± 5.73 44.7 ± 8.2 N.S. PI 36.9 ± 4.1 38.4 ± 3.1 N.S. PS 67.6 ± 6.6 75.1 ± 9.2 N.S. PC 255.0 23.0 251.0 ± 31.0 N.S. PE 141.0 ± 17.0 145.0 ± 22.0 N.S. PC/SM 6.1 ± 0.8 5.8 ± 1.3 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations used are the same as those in Table 19. 104 Table 44. Phospholipid composition of plasma in control and food-deprived rats. umol Phospholipid Phosphorus/L Control Food-deprived P value LPC 189 ± 15.5 165 ± 28.3 N.S. SM 55.1 ± 20.8 44.3 ± 14.5 N.S. PI 44.4 ± 9.2 43.2 ± 10.9 - N.S. PC 589 ± 70.0 492 ± 89 N.S. PE 18.9 ± 5.9 16.6 ± 4.0 N.S. PC/SM 11.7 ± 3.9 12.5 ± 5.9 N.S. Values are expressed as mean ± SD, n = 6 in each group. N.S. = non-significant. The abbreviations are the same as those in Table 21. 105 4.5 Spontaneously Diabetic BB Wistar Rats 4.5.1 General features Table 45 shows body weights, plasma biochemical parameters and hemo-globin glycosylation values in ISDBB, their NDLM, and control Wistar rats. The dose of insulin administered to diabetic animals over the 7-12 wk period prevented the loss of body weight and resulted in a growth rate (mean body weight at the time of detection of diabetes was 388 ± 30 g) comparable to that of NDLM. However, control of blood glucose levels was not optimal. Of the eight rats studied, three showed considerable hyperglycemia (28.95 ± 0.57 mM) and hypoinsulinemia (7.33 ± 0.29 yU/ml) [normal value for plasma glucose is 8.64 ± 1.19 and plasma insulin is 45.57 ± 8.10]. The other animals showed more or less normal plasma glucose levels associated -with hyperinsulinemia. Both ISDBB and their NDLM showed s ignif icant ly elevated levels of plasma cholesterol compared to control rats, while plasma phospholipid contents were elevated only in ISDBB. The var iab i l i ty in plasma levels of glucose, tr ig lycerides, and insulin as well as glycosylated Hb levels i l lustrates the extent of variation in achieving optimal control in these insulin-treated diabetic animals. 4.5.2 Tissue antioxidant status Experiments in both ISDBB and their NDLM showed common alterations in tissue antioxidant status when comparison was made with control Wistar rats. These included elevated pancreatic CAT act iv i t ies and lower GSH levels compared to control (Table 47). Hepatic levels of GSH were also lower in BB Wistar rats than in controls (Table 48). On the other hand, the diabetic state in ISDBB was associated with dist inctive changes in antioxi-dant defense mechanisms. The hearts of diabetic rats showed increased CAT and GSSG-RD act iv i t ies and elevated levels of GSH compared to control rats 106 Table 45. General features and plasma levels of glucose, l ipids and insulin in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. Control BB-Wistar ISDBB NDLM Body wt. (g) 434 ± 56 445 ± 67 444 ± 99 Glucose (mM) 8.64 ± 1.19 15.92 ± 10.98 8.43 ± 1.04 Cholesterol (mM) 2.10 ± 0.20 2.57 ± 0.25* 3.01 ± 0.49* + Triglycerides (mM) 1.36 ± 0.27 2.73 ± 1.99 2.23 ± 0.82 Phospholipid P (mM) 1.58 ± 0.14 2.42 ± 0.31* 1.85 ± 0.18 + Insulin (pU/ml) . 45.57 ± 8.10 70.12 ± 65.24 37.6 ± 9.47 HbA l c (%) 3.66 ± 0.18 4.09 ± 0.92 3.73 ± 0.23 Values are expressed as mean ± SD, n = 8 in each group. *p < 0.5, s ignif icantly different from control. + p < 0.5, s igni f icant ly different from diabetic. 107 Table 46. Antioxidant status of heart in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. Control BB-Wistar ISDBB NDLM CAT (K/g wet wt.) 0.09 ± 0.05 0.52 ± 0.34* 0.34 ± 0.17 CuZn-SOD (U/g wet wt.) 727 ± 115 864 ± 131 723 ± 143 GSH-PX (umol NADPH/min/g wet wt.) 5.34 ± 1.11 6.09 ± 1.71 5.22 dz 1.98 GSSG-RD (pmol NADPH/min/g wet wt.) 0.51 ± 0.04 0.69 ± 0.10* 0.50 dz 0.05 + GSH (pmol/g wet wt.) 1.62 ± 0.18 2.01 ± 0.22* 1.50 dz 0.17+ Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icantly different from control. +p < 0.05, s ignif icant ly different from diabetic. The abbreviations used are the same as those in Table 3. 108 Table 47. Antioxidant status of pancreas in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. Control BB-Wistar ISDBB NDLM CAT (K/g wet wt.) 0.16 ± 0.05 0.43 ± 0.17* 0.40 ± 0.12* CuZn-SOD (U/g wet wt.) 685 ± 108 1032 ± 159* 704 ± 101 + GSH-PX (pmol NADPH/min/g wet wt.) 3.84 ± 1.14 5.34 ± 2.16 3.81 ± 1.02+ GSSG-RD (ymol NADPH/min/g wet wt.) 1.44 ± 0.18 1.74 ± 0.26* 1.48 ± 0.23 GSH (ymol/g wet wt.) 1.23 ± 0.18 0.81 ± 0.27* 0.88 ± 0.11* Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icant ly different from diabetic. The abbreviations used are the same as those in Table 3. 109 Table 48. Antioxidant status of l iver in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. Control BB-Wistar ISDBB NDLM CAT (K/g wet wt.) 47.63 ± 5.74 43.07 ± 9.70 47.82 ± 6.61 CuZn-SOD (U/g wet wt.) 3459 ± 388 3445 750 3412 ± 566 GSH-PX (umol NADPH/min/g wet wt.) 24.27 ± 3.96 24.45 ± 6.00 21.81 9.24 GSSG-RD (umol NADPH/min/g wet wt.) 3.45 ± 0.27 3.41 ± 0.41 3.32 ± 0.34 GSH (ymol/g wet wt.) 7.33 ± 0.74 5.52 ± 0.78* 5.98 ± 0.58' Values are expressed as mean * SD, n = 8 in each group • *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 3. 110 (Table 46). The only significant change observed in kidneys of ISDBB rats was an increased act iv i ty of GSH-PX compared to control (Table 49). 4.5.3 Erythrocyte enzymatic changes Erythrocytes of ISDBB rats showed, as in other models of diabetes, higher GSSG-RD act iv i ty but not GSH-PX act iv i ty, while their NDLM had s ign i -f icant ly lower CAT act iv i t ies compared with those of weight-matched controls (Table 50). 4.5.4 Erythrocyte susceptibi l i ty to peroxidative damage No changes were observed in GSH depletion in red cel ls of BB rats compared to control rats in response to H2O2 (Table 51). However, some of the ISDBB rats that showed hyperglycemia did show reduced susceptibi l i ty to GSH depletion typical of poorly controlled diabetic rats. MDA levels were higher in red cel ls of ISDBB rats than those of controls at H2O2 concentrations of 1.5 and 2.0 mM, while MDA levels in NDLM rat erythrocytes were lower than those in diabetic erythrocytes but higher than the control values at 1.5 mM H2O2 concentration (Table 52). 4.5.5 Erythrocyte membrane and plasma l ip id analyses Erythrocyte cholesterol and phospholipid measurements revealed lower membrane cholesterol levels in ISDBB rats, despite the increased plasma cholesterol value, resulting in a s ignif icantly lower C/P ratio (Table 53). Phospholipid profiles of erythrocytes revealed lower SM levels in both groups of BB Wistar animals compared to weight-matched controls and a s igni -ficant increase in PC/SM ratio was seen in ISDBB rats (Table 54). Plasma phospholipid analysis of diabetic rats were, generally, comparable to those observed in uncontrolled chemically-induced diabetes (Table 55). However, plasma of NDLM rats showed higher PC levels compared to controls and s igni -f icant ly lower LPC levels compared to their diabetic counterparts. I l l Table 49. Antioxidant status of kidney in controls, spontaneously diabetic male BB rats and their non-diabetic 1ittermates. Control BB-Wistar ISDBB NDLM CAT (K/g wet wt.) 7.31 ± 1.44 6.85 ± 1.34 7.83 ± 1.55 CuZn-SOD (U/g wet wt.) 1956 ± 162 1983 ± 349 1701 ± 152 GSH-PX (pmol NADPH/min/g wet wt.) 7.29 ± 1.41 10.47 ± 2.52* 8.58 ± 2.43 GSSG-RD (pmol NADPH/min/g wet wt.) 5.07 ± 0.38 5.30 ± 0.37 4.94 ± 0.33 GSH (pmol/g wet wt.) 2.89 ± 0.40 2.83 ± 0.36 2.93 ± 0.31 Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 3. 112 Table 50. Antioxidant enzyme act iv i t ies of erythrocytes in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. Control BB-Wistar ISDBB NDLM CAT (K/g Hb) 40.43 * 1.84 38.63 ± 9.06 30,96 ± 5.00* CuZn-SOD (U/g Hb) 3061 ± 324 3139 ± 451 3080 ± 385 GSH-PX (umol NADPH /min/g Hb) 73.92 ± 14.61 84.51 ± 10.53 81.99 ± 11.31 GSSG-RD (ymol NADPH /min/g Hb) 0.99 ± 0.16 1.30 ± 0.29* 0.94 ± 0.09+ Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s igni f icant ly different from control. +p < 0.05, s ignif icantly different from diabetic. The abbreviations used are the same as those in Table 11. 113 Table 51. H 20 2-induced GSH depletion of erythrocytes in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. H 2 0 2 (mM) GSH Content (; % ) Control BB ISDBB Wistar NDLM 0.000 100 100 100 0.025 83.3 ± 20.5 88.1 ± 19.1 0.050 69.0 ± 8.4 69.7 ± 32.8 64.7 ± 22.7 0.250 11.0 ± 3.1 29.9 ± 25.5 11.9 ± 6.2 0.500 •6.1 ± 2.2 20.7 ± 22.1 8.1 ± 2.6 1.000 7.1 ± 2.3 8.6 ± 4.8 6.6 ± 1.1 Values are expressed as mean ± SD, n = 8 in each group. GSH levels in red cells incubated in the absence of added H 2 0 2 (see Methods, 2.8.1) were 3.47 + 0.78, 3.66 + 0.58 and 3.33 + 0^ .59 ymol/g Hb for control, ISDBB and NDLM rats, respectively. 114 Table 52. H 20 2-induced MDA production of erythrocytes in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. MDA Content7 Control BB Wis tar H2°2 ( m M ) ISDBB NDLM 0.00 2.4 ± 2.6 4.2 ± 4.2 5.4 ± 4.5 0.50 27.2 ± 8.0 26.6 ± 9.6 24.9 ± 10.0 1.00 93.0 ± 59.3 236 ± 234 164 ± 114 1.50 255 ± 40 522 ± 183* 447 ± 89* 2.00 430 ± 70 626 ± 194* 571 48 3.00 648 ± 42 787 ± 220 718 ± 62 Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icantly different from control. VMDA production is expressed as nmol/g Hb 115 Table 53. Cholesterol and phospholipid contents of erythrocyte membranes in controls, spontaneously diabetic male BB rats and their non-diabetic 1ittermates. Control BB Wistar ISDBB NDLM Cholesterol (C) (umol/mg protein) 0.57 ± 0.05 0.49 ± 0.06* 0.58 ± 0.05+ Phospholipid P (P) (pmol/mg protein) 0.62 ± 0.04 0.60 ± 0.04 0.62 ± 0.05 C/P ratio 0.93 ± 0.04 0.82 ± 0.08* 0.93 ± 0.05t Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icantly different from control. + p < 0.05, s ignif icantly different from diabetic. 116 Table 54. Phospholipid composition of erythrocyte membranes in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. nmol Phospholipid Phosphorus/mg protein ISDBB NDLM SM 62.7 ± 13.6 33.2 ± 8.5* 43.2 ± 21.2* PI 38.9 ± 6.1 36.0 ± 5.8 38.9 ± 6.7 PS 80.2 ± 8.6 79.3 ± 13.7 87.0 ± 12.5 PC 286.0 ± 34.0 277.0 ± 25.0 278.0 ± 30.0 PE 165.0 ± 13.0 174.0 ± 17.0 176.0 ± 18.0 PC/SM 5.0 ± 1.4 8.8 ± 2.2* 7.7 ± 3.3 Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icantly different from control. The abbreviations used are the same as those in Table 19. 117 Table 55. Phospholipid composition of plasma in controls, spontaneously diabetic male BB rats and their non-diabetic littermates. pmol Phospholipid Phosphorus/L Control BB Wistar ISDBB NDLM LPC 368 ± 56.0 372 dz 85.0 270 ± 88.0 + SM 118 ± 26.3 67.2 dz 21.7 79.6 ± 60.4 PI 98.9 ± 24.0 131 dz 29.7* 88.8 ± 14.1 + PC 960 75.0 1773 dz 316* 1359 ± 235*+ PE 32.1 ± 15.3 80.2 dz 17.1* 31.5 ± 9.3 + PC/SM 8.5 ± 2.0 26.5 dz 9.2* 19.8 ± 15.5' Values are expressed as mean ± SD, n = 8 in each group. *p < 0.05, s ignif icant ly different from control. + p < 0.05, s ignif icantly different from diabetic. The abbreviations used are the same as those in Table 21. 118 4.6 Diabetic Patients 4.6.1 General features Fasting blood samples were obtained from 39 diabetic patients (17 IDDM and 22 NIDDM) referred to the Diabetic Specialty Centre, Shaughnessy Hospital, and from 19 controls. The general features of both groups are shown in -Table 56. The mean age of diabetic patients was somewhat higher than that of control subjects and i t did attain s tat i s t ica l significance in the NIDDM patients. Fasting plasma glucose levels were s ignif icantly higher in diabetic as compared to control subjects, although HbA^c levels are within normal range for well-control led diabetic patients. However, we were unable to measure HbA-^ levels in control subjects by column chromato-graphy, the c l in ica l method used in measuring HbA^ levels in diabetic patients, and i t was obvious that using the colorimetric method which is available in our laboratory would not allow an appropriate comparison. Therefore, control HbA^ values are not provided. However, the normal range of control values provided by the laboratory at Shaughnessy Hospital obtained using ion-exchange chromatography is between 4-7%. Plasma triglycerides and phospholipid levels were elevated in NIDDM as compared to control subjects. 4.6.2 Erythrocyte enzymatic changes Results of antioxidant enzyme measurements in erythrocytes from diabetic patients showed comparable increases in GSSG-RD act iv i t ies to those seen in the diabetic animals (Table 57). However, unlike the situation in the diabetic animals, CuZn-SOD act iv i t ies were s ignif icantly increased in erythrocytes of diabetic patients. The pattern of changes in these two enzymes was not" influenced by the type of diabetes or the severity of complications in the diabetic patients. The membrane-bound NADH-dehydrogen-ase was s ignif icantly increased only in IDDM (Table 57). 119 Table 56. General features and plasma levels of glucose and l ip ids in control and diabetic subjects. Control Diabetic IDDM NIDDM No. of subjects 19 17 22 Mean Age (yr) 47.9 ± 12.9 52.6 ± 17A 58.0 ± 14.2 P value N.S. < 0.05 Glucose (mM) 4.32 ± 0.53 9.93 ± 4.60 7.95 ± 3.22 P value < 0.05 < 0.05 Cholesterol (mM) 6.43 ± 1.26 5.83 ± 1.92 6.15 ± 1.32 P value N.S. N.S. Triglycerides (mM) 1.52 ± 0.73 2.04 ± 1.83 2.34 ± 1.57 P value N.S. < 0.05 Phospholipid P (mM) 1.92 ± 0.48 2.13 ± 0.60 2.16 ± 0.48 P value N.S. < 0.05 HbA l c (%)* — 7.62 ± 1.51 6.94 ± 1.53 Values are means ± SD. P as compared with the control group. *HbA^c was determined by the Cl in ica l Chemistry Laboratory at Shaughnessy Hospital, by ion-exchange column chromatography. Range of normal values provided by the c l in ica l laboratory, using ion-exchange column chromato-graphy, is between 4-7%. 120 Table 57. Antioxidant enzyme act iv i t ies of erythrocytes in control and diabetic subjects. Control Diabetic IDDM NIDDM CAT (K/g Hb) 96.71 ± 16.51 101.3 ± 16.14 101.7 ± 16.6 P value N.S. N.S. CuZn-SOD (U/g Hb) 1591 ± 202 2057 ± 428 2266 ± 481 P value < 0.05 < 0.05 GSH-PX (umol NADPH/min/g Hb) 14.35 2.06 14.40 ± 2.53 13.38 ± 2.85 P value N.S. N.S. GSSG-RD (umol NADPH/min/g Hb) 3.96 ± 0.53 5.33 ± 0.69 5.47 ± 0.85 P value < 0.05 < 0.05 NADH-dehydrogenase (nmol/min/mg protein) 89.4 ± 11.8 103.1 ± 13.6 97.9 ± 15.4 P value < 0.05 N.S. Values are expressed as mean ± P as compared with the control SD. group. The abbreviations used are the same as those in Table 11. 121 4.6.3 Erythrocyte susceptibi l i ty to peroxidative damage Further analyses were carried out to investigate the response of erythrocytes to H 20 2-induced peroxidative damage. Diabetic erythrocytes showed a s ignif icantly reduced sensit iv ity to GSH depletion with increasing H 20 2 concentrations (Table 58). On the other hand, MDA levels were s ignif icantly higher than those of control subjects at H 2 0 2 concentra-tions of 5.0 and 7.5 mM (Table 59a). While this pattern was generally applicable to diabetic patients as a whole, i t tended to be influenced by the type of diabetes (being more obvious in Type II diabetes) and also by the severity of complications (changes were more obvious in patients with marked complications (Table 59b). 4.6.4 Red cell membrane and plasma l ip id analyses Detailed studies of red cel l and plasma l i p id characteristics were carried out in both diabetic and control subjects. No significant changes were observed either in erythrocyte membrane contents of cholesterol or phospholipids (Table 60) or in individual phospholipid components in either group of diabetic patients (Table 61). However, plasma phospholipid profiles did show significant increases in PC and PE contents in NIDDM and an increase in PC/SM ratio in both IDDM and NIDDM patients (Table 62). 122 Table 58. ^G^-induced GSH depletion of erythrocytes in control and diabetic subjects. GSH Content {% ) Control Diabetic P value H 20 2 (mM) (n=16) (n=39) 0.000 100 100 0.050 36.4 ± 9.4 59.9 ± 22.5 < 0.05 0.150 18.0 ± 3.7 34.4 ± 14.8 < 0.05 0.250 16.4 ± 3.0 28.2 ± 13.3 < 0.05 0.500 15.8 ± 2.4 23.8 ± 9.8 < 0.05 1.000 15.1 ± 1.6 20.5 ± 6.9 < 0.05 Values are expressed as mean ± SD. GSH levels in red cel ls incubated in the absence of added H^ Og (see Methods, 2.8.1) were 3.43 + 0.36 and 3.29 + 0.54 ymol/g Hb for control and diabetic subjects, respectively. 123 Table 59a. H-pO^-induced MDA production of erythrocytes in control and diabetic subjects. MDA Content7 Control Diabetic P value H 20 2 (mM) (n=19) (n=39) 0.00 6.61 ± 7.42 4.54 ± 5.06 N.S. 2.50 21.91 ± 10.56 26.15 ± 12.33 N.S. 5.00 99.5 ± 58.81 151.3 ± 66.3 < 0.01 7.50 305.3 ± 94.3 366.6 ± 114.1 < 0.001 10.00 522.8 ± 114.4 557.8 ± 106.6 N.S. 15.00 771.6 ± 106.4 802.9 ± 114.8 N.S. Values are expressed as mean ± SD. VMDA values are expressed as nmol MDA/g Hb. Table 59b. H 90 9-induced MDA production of erythrocytes in control and diabetic subjects. Type of Diabetes Complications* H202(mM) Controls Type I Type II marked moderate none (n=19) (n-17) (n=22) (n=10) (n-17) (n-7) 0.00 6.6 ± 7.4 3.8 ± 4.2 5.2 ± 5.6 4.5 ± 6.5 4.8 ± 4.8 4.4 ± 4.4 P value N.S. N.S. N.S. N.S. N.S. 2.50 21.9 ± 10.-6 24.3 ± 9.2 28.1 ± 14.4 28.3 ± 10.8 26.4 ± 10.3 25.4 ± 18.9 P value N.S. N.S. N.S. N.S. N.S. 5.00 99.5 ± 58.8 131.7 ± 56.7 166.4 ± 70.4 185.7 ± 74.9 149.1 ± 68.4 125.8 ± 27.1 P value N.S. < 0.01 < 0.01 < 0.05 N.S. 7.50 305.3 ± 94.3 345.0 ± 116.8 383.3 ± l l i . 8 433.2 ± 98.5 366.2 ± 123.6 301.7 ± 60.1 P value < 0.001 < 0.001 < 0.001 < 0.001 N.S. 10.00 522.8 ± 114.4 526.7 ± 90.8 581.9 ± 113.6 589.8 ± 85.1 569.6 ± 100.1 480.6 ± 88.5 P value N.S. N.S. N.S. N.S. N.S. 15.00 771.6 ± 106.4 764.2 ± 76.6 832.8 ± 131.2 828.5 i 99.2 805.0 ± 103.8 781.0 ± 124.7 P value N.S. N.S. N.S. N.S. N.S. Values are expressed as mean ± SD. MDA values are expressed as nmol MDA/g Hb. P as compared with the control group. N.S. = non-significant. *Regardless of type of diabetes. For further detai ls , see methodology. 125 Table 60. Cholesterol and phospholipid contents of erythrocyte membranes in control and diabetic subjects. Control Diabetic IDDM NIDDM (n=l7) (n=22) Cholesterol (C) (umol/mg protein) 0.50 ± 0.06 0.52 ± 0.05 0.51 ± 0.05 P value N.S. N.S. Phospholipid P (P) (umol/mg protein) 0.58 ± 0.09 0.62 ± 0.08 0.59 ± 0.08 P value N.S. N.S. C/P ratio 0.87 ± 0.06 0.85 ± 0.08 0.86 ± 0.07 P value N.S. N.S. Values are expressed as mean ± SD. P as compared with the control group. 126 Table 61. Phospholipid composition of erythrocyte membranes in control and diabetic subjects. nmol Phospholipid Phosphorus/mg protein Control Diabetic IDDM NIDDM (n=19) (n=17) (n=22) SM 102.9 ± 31.2 119.0 ± 48.6 124.6 ± 39.0 P value N.S. N.S. PI + PS 123.5 ± 32.5 134.7 ± 15.2 124.9 ± 20.7 P value N.S. N.S. PC 180.8 ± 27.2 190.6 ± 22.4 178.8 ± 24.2 P value N.S. N.S. PE 171.5 ± 38.4 175.9 ± 24.5 167.0 ± 22.3 P value N.S. N.S. PC/SM 1.9 ± 0.7 2.0 ± 1.2 1.6 ± 0.6 P value N.S. N.S. Values are expressed as mean ± SD. P as compared with the control group. The abbreviations used are the same as those in Table 19. 127 Table 62. Phospholipid composition of plasma in control and diabetic subjects. pmol Phospholipid Phosphorus/L Control Diabetic IDDM NIDDM (n=19) (n»17) (n=22) LPC + PI 163 ± 48.1 168 ± 52.2 157 ± 33.0 P value N.S. N.S. SM 352 ± 108 343 ± 145 348 ± 141 P value N.S. N.S. PC 1331 ± 320 1522 ± 409 1562 ± 338 P value N.S. < 0.05 PE 78.5 ± 25.7 95.8 ± 28.9 99.8 ± 31.2 P value N.S. < 0.05 PC/SM 3.87 ± 0.51 4.83 ± 1.28 5.16 ± 2.09 P value < 0.05 < 0.05 Values are expressed as mean ± SD. P as compared with the control group. The abbreviations used are the same as those in Table 21. 128 6. DISCUSSION 6.1 Evaluation of the Experimental Data: Coefficient of Variation Analysis of the present data using the coefficient of variation gave values that covered the range 1%-100%, although the majority of them were within the acceptable values (between 5 and 15 percent) (190), indicating the presence of large.var iabi l i ty among the experimental data. Variabi l ity in the data obtained could be caused by two types of factors, those that cause similar variations in measured values for both control and diabetic animals, and factors that could be attributable to differences in the diabetic state and its severity. Common factors that presumably could cause variations in the present observations might include such things as the systematic errors as well as the biological variations in the values of any of the parameters measured. The discrepancy between multiple determinations of the same chemical and biochemical assay never exceeded 5%. While any of the above factors may contribute, they are l ikely to apply to both control and diabetic groups. This is evident from the s imi lar i t ies in coefficients of variation for those parameters that were not altered by the diabetic state. With regard to abnormalities restricted to the diabetic animals (notably emaciation and other overt signs of diabetes), these were such that blind studies were v irtual ly impossible. Moreover, i t was clear from observations in diabetic animals that diabetic symptoms were not classical or uniform but, in fact, varied considerably and covered a range of symptoms from severe diabetes with marked emaciation to a complete and spontaneous recovery from the disease. The presence of such variations in the diabetic animals in the present study is l ikely the major factor responsible for the finding that the coefficient of variation was higher in diabetic and, for that matter, in insulin-treated diabetic animals than in controls. 129 Therefore, although I acknowledge the presence of considerable va r i -ab i l i ty in the present data, this has not prevented the detection of many highly significant differences between diabetic and non-diabetic animals. Moreover, i t was apparent from the experimental data that variations in the severity of diabetes, per se, are a major factor causing increased var iab i l -ity in the data of diabetic animals. However, the var iab i l i ty in severity of diabetic states did serve one useful purpose in showing that a correla-tion did appear to exist between grossly observable estimates of severity of the disease (e.g., emaciation) and certain biochemical variables, suggesting that the latter did seem to reflect the severity of underlying disease state. 6.2 Differentiation Between Diabetes- and Chemical-Induced Changes in Antioxidant Status The complex nature of the changes in antioxidant status observed in the present study might represent an interplay between alterations related in a primary way to the production of the diabetic state, to secondary changes due to body weight loss and f ina l l y to direct effects of the diabe-togenic agent unrelated to its diabetes-inducing actions. With regard to the latter point, although such a poss ib i l i ty cannot be ruled out in our study, we feel i t is unlikely for the following reasons. F i r s t l y , STZ (191) and ALX (192) are rapidly metabolized and would therefore not be expected to exert direct effects over the 12 wk period used in our study. Secondly, we fa i led to detect any alterations in tissue antioxidant enzyme act iv i t ies in the STZ- and ALX-injected animals that fa i led to develop diabetes. Thirdly, the many s imi lar it ies in tissue antioxidant alterations between the ALX-diabetic animals and those with STZ-diabetes would seem to, indicate that these changes are best attributed to the diabetic state produced by these structurally different diabetogenic agents rather than to their direct toxic effects. Fourthly, insulin treatment completely reversed most of the a l ter -130 ations observed in the diabetic tissues. Final ly, changes similar to those seen with ALX or STZ were observed in tissues of spontaneously diabetic BB Wistar rats. In addition, the red cel l measurements reported here were made 12 wk after the injection of ALX or STZ. Rat red cel l s have a life-span .'of approximately 50 days (193) so that any observations regarding red cel ls in these rats would be unlikely due to the direct oxidant effects of ALX or STZ on red ce l l s . Support for this proposal is derived from the similar behavior of red blood cel l s from both ISDBB rats and human diabetic patients. 6.3 Effect of Body Weight Loss on Tissue Antioxidant Status The results obtained demonstrate that in the absence of any pharmaco-logical perturbation, food-deprivation (with resulting body weight loss) per se is associated with alterations in tissue defense mechanisms against oxidative injury. However, the complex nature of the observed changes was not readily explained on the basis of a simple and uniform compensatory response to increased endogenous oxidant stress in the various tissues examined. The effects of food-deprivation on scavenging enzyme act iv i t ies in tissues did not direct ly parallel those in red ce l l s . The increased act iv-i ty of CAT in heart and both CAT and CuZn-SOD in pancreas, as well as increased CuZn-SOD and GSH-PX act iv i t ies in kidney was accompanied by a decreased act iv ity of CuZn-SOD in heart, and of CAT act iv ity and GSH content in l iver of food-deprived rats. The findings in the l iver agree with those of Isaac and Binkley (173) who showed that 24 h of food-deprivation in rats was accompanied by decreases in hepatic CAT act iv i ty and GSH levels but no changes in GSSG-RD act iv i ty. These investigators also demonstrated increased peroxide generation. Xia et a l . (194) also reported that a 24 h fast in rats lowered the hepatic GSH content of l iver without changing its level in the heart. 131 Lammi-Keefe et a l . reported increased CAT act iv i t ies in hearts from rats deprived of food for either 2 days (172) or for 18 days (195). Food-deprivation has been shown to have an inhibitory effect on insulin-dependent glucose transport in rat heart (196). Hypoinsulinemia, which is present in these food-deprived animals, has been suggested to be associated with a stimulation of fatty acyl CoA oxidase (197), which in i t iates the B-oxidation of fatty acids and results in the increased production of hydrogen peroxide (198). This might be related to the observed increases in CAT act iv ity associated with food-deprivation-induced hypoinsulinemia. These results suggest that, in studies of free radical scavenging enzyme alterations in animal models of chemically-induced diabetes (where decreases in body weight are a common feature), i t wil l be important to differentiate between changes due to the decrease in body weight and those due to the disease i t se l f . 6.4 Tissue Antioxidant Status in Chemically-Induced Diabetes We have shown that body weight loss (induced by a 72 h period of food-deprivation) associated with hypoinsulinemia produced a variety of alterations in tissue antioxidant enzyme systems. The alterations in the act iv i t ies of cardiac CAT, pancreatic CAT and CuZn-SOD, hepatic CAT and GSH content, and renal GSH-PX act iv ity seen in diabetic rats qualitatively paralleled those seen following food-deprivation. It should also be pointed out that the magnitude of the alterations in these diabetic rats increased with the degree of body weight loss. If the changes observed in the food-deprived rats are attributable to hypoinsulinemia, body weight loss would be expected to aggravate the hypoinsulinemia in the diabetic animals and thereby increase the severity of tissue antioxidant abnormalities. However, the long-term hypoinsulinemia together with the hyperglycemia and elevated plasma l ipids seen in chronic diabetes are associated with qualitatively 132 different changes to those observed in the short-term effects of hypoin-sulinemia and hypoglycemia seen in the food-deprived animals. Thus, chronic hypoinsulinemia (12 wk of diabetes) was associated with depressed CuZn-SOD act iv i ty in l i ver and kidneys while 72 h of food deprivation-induced hypoin-sulinemia of a lesser degree than that in diabetes not only lacked this effect but instead showed an increase in renal CuZn-SOD act iv i ty. These results suggest the presence of more than one factor in the development of these alterations in tissue antioxidant defense mechanisms in diabetes, and that the duration of hypoinsulinemia ( i .e, diabetes) could be one factor. However, the available information in support of the hypothesis that the duration of diabetes (and the possible increased oxidative stress) could be important in determining tissue antioxidant alterations and susceptibi l ity to oxidative damage, is sparse. CuZn-SOD act iv i ty, for example, has been shown to be depressed in STZ-diabetic l i ver and kidney 10 or 11 days after the induction of diabetes (133) but not 1n diabetes of only 5 days duration (62). The finding in control animals that the heart and pancreas contain relat ively low act iv i t ies of radical scavenging enzymes as compared to l iver and kidney agrees with that of Grankvist et a l . (63) and that of Matkovics et a l . (6 5,66) in mouse and rat tissues, respectively. The diabetes-induced _ increases in the act iv ity of antioxidant enzymes in tissues in which re la -t ively low act iv i t ies are present may be compensatory in nature. Increases, possibly adaptive, in the act iv i t ies of CAT (199), SOD and GSH-PX (200) have been described in some tissues subjected to conditions of oxidative stress. The largest increases in the present study involved the enzyme CAT, the act iv i ty of which was markedly elevated in both heart and pancreas of non-insulin-treated diabetic animals, suggesting that endogenous production of Hp0? may be increased in diabetes. As pointed out ear l ier, insulin 133 lack is known to promote the peroxisomal e—oxidation of fatty acids in diabetic l iver with resulting formation (201). This fact, when taken with our observation that insulin treatment prevents the increases in cardiac and pancreatic CAT act iv i ty, would support the suggestion that these increases may be a compensatory response to an increase in endogenous H2O2 production in heart and pancreas. CAT has been implicated in the detoxification of high H 20 2 concentrations, whereas GSH-PX is more sensitive to lower concentrations of H 20 2. Doroshow et a l . (202) have proposed that in tissues lacking appreciable CAT act iv i ty (such as heart muscle) H 20 2 detoxification is c r i t i c a l l y dependent on the activity of GSH-PX. The results in heart and pancreas, however, suggest that increases in CAT act iv i ty may be an important adaptive response to conditions of increased peroxidative stress in these tissues, and support the suggestion that CAT plays a major role in the detoxification of H 20 2 in the myocar-dium (203). H 20 2 has been reported to act as an inducer of tissue SOD (65) and the increased pancreatic CuZn-SOD act iv ity observed in diabetic animals might be another manifestation of an adaptive response in a tissue where the enzyme act iv ity in question is normally low relative to other tissues. The two diabetogenic agents did d i f fer , however, in their effects on GSH levels in heart and pancreas. Although neither was s ignif icantly altered in STZ-diabetic animals, cardiac GSH content was s ignif icantly higher in ALX-diabetic rats compared to control, while pancreatic levels of GSH in ALX-diabetic rats were decreased, although these did not attain s tat i s t ica l significance. However, they did attain s tat i s t ica l significance using ANOVA and Tukey's test when ALP-treated groups rather than insu l in -treated groups were compared (Table 25). Large intra-group variations in the former groups (Table 6) presumably concealed the between-group d i f f e r -134 ence and account for this discrepancy. With regard to the maintenance of tissue GSH levels, the enzyme GSSG-RD would be expected to play a role since i t is required to regenerate GSH from GSSG. However, the fact that the dissimilar alterations in cardiac and pancreatic GSH levels in ALX-diabetes occurred despite the similar increases in GSSG-RD act iv i ty in both tissues suggests that other factors (e.g., synthesis or degradation) may be involved in determining the levels of tissue GSH. Furthermore, the finding that a reduction in GSH content does not necessarily correlate with corresponding changes in GSSG-RD or GSH-PX act iv i t ies supports the observations of Curello et a l . (204) who demonstrated that heart ischemia and reperfusion are asso-ciated with a reduction in GSH content but GSSG-RD and GSH-PX act iv i t ies were not affected. Although the significance of the differing alterations in cardiac and pancreatic GSH contents in ALX- and STZ-diabetes is not clear at the moment, they may be indicative of differences in GSH metabolism in these two models of chemically-induced diabetes which might have some bearing on the mechanisms of diabetogenesis for these two diabetogenic agents. They also i l lustrate that alterations in the metabolism of GSH may d i f fer for various cel l types, and this could be an important factor deter-mining differing susceptibi l i t ies of various tissues to oxidant injury (205). With regard to the alterations in pancreatic tissue, Lazarow (205) has suggested that analysis of the antioxidant status of whole pancreatic tissue may not necessarily reflect the antioxidant status of B-cells since these constitute only about 0.5%of the total weight of the pancreas. However, comparison between the antioxidant status of i s le t cel ls and exocrine pancreas or whole pancreas suggests that although the B-cells constitute a small fraction of the total weight of the pancreas (206 ), they contain an appreciable amount of total tissue enzymes. The enzymatic act iv i ty of CuZn-SOD in the isolated rat pancreatic is lets was found to be more than 135 100-fold greater than that in the whole pancreas (129). Another group found that the CuZn-SOD act iv ity of the i s let cel l s was 68% of that in the exocrine pancreas (63). Isolated mouse pancreatic is lets contained 27% of the CAT activity of the exocrine pancreas and 27% of the GSH-PX act iv i ty of exocrine pancreas (63). Other studies reported isolated rat pancreatic is lets contain 20% of the GSH-PX act iv ity of whole pancreas (64), while GSSG-RD act iv i ty of rat i s lets was 57 % of that in the exocrine pancreas (149). Further, evidence is available indicating that the effect of ALX on the GSH level of isolated pancreatic e-cells is qualitatively comparable to our results on whole pancreatic tissue. For example, Lazarow (207) has suggested that the diabetogenic action of ALX is associated with a loss of free SH groups, particularly in pancreatic i s lets , a finding that was confirmed histochemically by MacDonald (208) and by Malaisse et a l . (64) who demonstrated that ALX decreased the GSH content of pancreatic B-cells. Therefore, i t must be emphasized that, in the present study, although the observed changes in pancreatic antioxidant status are more indicative of a generalized pancreatic tissue change rather than one primarily involving i s let ce l l s , the poss ib i l i ty that these changes may, in part, reflect those of i s let cel l s cannot be ruled out. With the exception of the possibly compensatory increase in GSH-PX act iv i ty in diabetic rat kidney (which in control animals is relat ively low in comparison with other tissues), other antioxidant enzymes whose a c t i v i -t ies are normally relat ively high showed a decrease in chronic diabetes. Thus, while renal and hepatic tissues showed decreases in CAT and CuZn-SOD act iv i t ies , hepatic tissues showed decreases in GSH-PX act iv i ty and GSH levels as well. While the reason for the above-mentioned decreases in renal and hepatic antioxidants is as yet unexplained, i t is known that reactive oxygen radicals can themselves reduce the act iv i ty of these enzymes 136 (209,210). Diabetes, in the present study, was associated with a decreased CAT act iv ity in l i ver and kidney, tissues which are particularly abundant in peroxisomes. CAT is an integral component of the peroxisomal pathway of fatty acid oxidation and a reduction in CAT activity has been reported to cause extensive alterations in l ip id metabolism not only in these 'two t i s -sues, but may also extend well beyond those to other tissues (211). Liver is the only organ which showed, a decrease in GSH-PX act iv i ty in diabetic rats. The function of GSH-PX in animal tissues has been suggested to include the reduction of endogenously formed peroxides of unsaturated fatty acids present in membranous components of subcellular organelles (212), thereby preventing oxidative degradation of phospholipids. Whether the depression of GSH-PX act iv i ty that we observed in diabetic l i ver might favor intracel lular peroxide accumulation is unknown, although diabetic l iver did show a decrease in GSH content. GSH, a major scavenger of oxygen reactive intermediates, protects cel l s against the effects of free radicals and of reactive oxygen intermediates (e.g., peroxides) that are formed endogenously (213) , and a reduction in ce l lu lar GSH level may be associated with an increased susceptibi l i ty to oxidant stress or reflect a response to i t (214) . An association between the observed impairment in hepatic and renal CuZn-SOD enzyme act iv i t ies and GSH depletion is suggested by the observation of Loven et a l . (133) that GSH administration prevented the decreases in hepatic and renal CuZn-SOD act iv i t ies in rats with STZ-induced diabetes. Loven et a l . (133) demonstrated that the decrease in hepatic GSH levels in STZ diabetes was accompanied by an increase in l iver cysteine and glutamate levels, but not glycine levels, while plasma levels of these amino acids were depressed, suggesting an increased demand for these amino acids and/ or an increased rate of glutathione degradation in experimental diabetes. Many adverse effects are observed when intracel lular levels of GSH 137 f a l l below 30-40% of control levels (215). Although GSH depletion (to any level greater than this threshold value) need not in i t se l f result in l iver injury (216), i t is generally believed that i t does increase the sensit ivity of this tissue to the well known damaging processes associated with the metabolism of specific drugs (217). In the present studies, hepatic GSH content in diabetic rats did not decrease to the aforementioned c r i t i c a l level (215). However, given the fact that other enzymatic antioxidants in liver'were also impaired in diabetic rats, this may increase the suscepti-b i l i t y of l iver to oxidative stress. In this regard, i t should be empha-sized that the present alterations in tissue antioxidant defense systems in l i ver and kidney of diabetic animals are suggestive of, but do not provide conclusive evidence of, increased susceptibi l i ty toward oxidative stress. Further, for several reasons, these alterations are by no means indicative of functional impairment in these organs. F i r s t l y , despite the. marked degree of changes in comparison with other tissues these organs s t i l l contain relat ively higher act iv i t ies of antioxidant defenses relative to other tissues, a fact that may be suff icient to prevent any appreciable functional impairment. Secondly, the available l iterature does not provide enough evidence to permit an association of the present alterations with abnormal functions in either l i ver or kidney. Therefore, further studies are required to test the implications of these findings and to determine their consequences, particularly in terms of morphological and functional characteristics of these two organs in diabetic animals. However, i t seems reasonable to hypothesize that the presence of such alterations in renal and hepatic defenses against increased oxidative stress may impair the ab i l i ty of these organs to eliminate circulating toxic meta-bolites, e.g., l ip id peroxides, and probably could increase the burden on other tissues with less antioxidant potential, e.g., heart and pancreas. 138 Further studies are required to test this hypothesis in which tissues, inherently low in antioxidant mechanisms,, are studied for alterations in antioxidant mechanisms and susceptibi l ity to oxidant injury in animals which have been diabetic for longer than the 3 month period used in the present study. We have not examined diabetic tissues for the presence of increased l i p id peroxidation in this particular study; however, evidence is available in the l iterature that demonstrates the presence of increased l ip id peroxide levels in diabetic tissues (66,149-151) and this has been proposed to contribute to diabetic complications (149,150). Moreover, we were able to detect higher levels of plasma peroxides (measured by TBA-reactive material) in ALX- and STZ-diabetic rats, compared to controls, as early as 60 h post-ALX- and STZ-injection (unpublished data). The ab i l i ty of insulin to reverse the above-mentioned changes in tissue enzyme act iv i t ies and GSH contents suggests that this hormone can influence tissue "antioxidant" status. The finding that insulin treatment for 4 wk reverses the alterations in certain kidney and l iver antioxidant enzymes agrees with results of Loven et a l . (133) in STZ-diabetes ut i l i z ing 6 days of insulin treatment. However, the persistence of decreased CuZn-SOD act iv i t ies in both l iver and kidney of diabetic rats after 12 wk of t reat-ment with insulin suggests the presence of a residual def ic i t in tissue antioxidant status in l iver and kidney despite insulin treatment. However, our study cannot rule out the poss ib i l i ty of direct ALX or STZ damage, particularly ALX damage to kidney (218), as another explanation for the decrease in CuZn-SOD act iv i t ies although i t is unlikely, as has been mentioned previously. The finding that allopurinol treatment was able to restore the act iv i t ies of renal CuZn-SOD act iv i t ies deserves further evalu-ation, particularly in combination with insulin therapy in future experi-ments on animal models of diabetes mellitus. The mechanism responsible for 139 this effect of ALP on renal CuZn-SOD in not clear, althought ALP has been reported to stimulate the de novo synthesis of mouse l iver cytosolic CuZn-SOD, thereby increasing i ts act iv ity (219). In addition to its well documented inhibition of xanthine oxidase, ALP has also been shown to be an effective hydroxyl radical scavenger (220) and this latter property may explain the marked reduction in the peroxide-induced increase in MDA forma-tion in red cel l s of ALP-treated animals. 6.5 Tissue Antioxidant Status in Spontaneously Diabetic BB Rats The results obtained demonstrate that two types of alterations in antioxidant status exist in BB rats: strain-related changes (increased CAT activ ity in pancreas and decreased GSH levels in pancreas and l iver of both ISDBB and their NDLM) and diabetes-related changes (manifested by increases in cardiac CAT and GSSG-RD, pancreatic CuZn-SOD and GSSG-RD, and renal GSH-PX). These results also show that the general pattern of alterations in antioxidant enzymes, as in studies of the effects of food deprivation-induced weight loss or chemically-induced diabetes on tissue antioxidant enzyme systems in the rat, was an increase often seen in an enzyme whose act iv i ty in the tissue in question was low relative to that in other tissues. Tissue GSH revealed a complex pattern of changes (an increase in heart of ISDBB and a decrease in pancreas and l iver of both ISDBB and NDLM). There is evidence suggesting parallelism in immune system defects and metabolic abnormalities of both SDBB and NDLM rats (138,140,221). For example, i t has been shown that both SDBB and NDLM pancreata have mononu-clear insul it is (81,139-141,222) and that thymectomy or administration of rat anti-lymphocyte serum (77) or s i l i c a gel (an agent highly specif ic for preventing macrophages from in f i l t ra t ing pancreatic is lets) (223) prevented diabetes in susceptible animals. Nakhooda et a l . (141) observed mild in su l -i t i s and abnormal glucose tolerance in BB rats not showing overt diabetes 140 and interpreted this finding in terms of a pre-diabetic state in which B-cell destruction insufficient to manifest diabetes is present in these animals. These data suggest the presence of endogenous injury (probably cel1-mediated) to the pancreas of BB rats which might contribute to the development of diabetes. In view of these considerations, NDLM are less than ideal controls for studies involving SDBB. Normoglycemic diabetes-prone animals may go on to develop diabetes with or without a demonstrable period of glucose into ler -ance (224) at any time in their l i f e span, although at the age of our animals (mean 7.62 months) such a poss ib i l i ty has been considered unlikely (81). Consistent with this hypothesis is our observation that about half of the NDLM rats showed plasma phospholipid profi les (elevation of PC/SM ratio) comparable to those characteristic of ISDBB plasma. Therefore, caution must be exercised in interpreting the significance of changes in tissue antioxi-dant status in "control" NDLM rats. Although the decrease in pancreatic GSH levels (relative to Wistar controls) was comparable in both groups of BB animals, those with overt diabetes also showed significant increases in pancreatic CuZn-SOD and GSSG-RD act iv i t ies . These changes suggest a greater degree of oxidative stress in the ISDBB as compared with the NDLM animals. The finding that pancreatic GSH levels in both ISDBB and NDLM rats were s ignif icantly lower than those of male Wistar controls deserves further consideration. As mentioned ear l ier , i t has been suggested that the diabetogenic action of ALX is associated with destruction of free SH groups, particularly in the pancreatic is lets (207) and prior administration of glutathione or cysteine, both of which contain a free SH group, prevented the development of ALX-induced diabetes in rats (225). It is necessary to mention that such observations tend to implicate (but do not confirm) a role of sulfhydryl 141 group-containing compounds (e.g., GSH) in the mechanism of diabetogenesis. Further studies wil l be required to more fu l l y assess the relationship between pancreatic SH group content in ISDBB or their NDLM and the develop-ment of diabetes in these animals. One approach would be to administer GSH or other sulfhydryl compounds to these animals at an age before they are l ike ly to become diabetic (usually less than 40 days old) and assess the incidence of diabetes in these animals in comparison with that of non-treated animals. With regard to the situation in tissues other than pancreas, the hearts of ISDBB animals showed increases in GSSG-RD as well as CAT ac t i v i -t ies. In addition to the decreased GSH content of l iver, the kidney showed an increased act iv i ty of GSH-PX seen both in the ISDBB animals and in STZ-and ALX-diabetic rats. The pancreatic and cardiac enzymatic changes in the ISDBB group paralleled those seen in STZ- and ALX-induced diabetes . However, alterations in GSH contents in both pancreas and heart in ISDBB were also seen in ALX-treated animals, but not STZ-treated animals. These results suggest that, as far as GSH status is concerned, the diabetic state produced by ALX approximates more closely than STZ the characteristics of the spontaneously diabetic BB rat model. However, in contrast to the situation in the ISDBB animals (which required daily insulin-treatment), these changes were completely reversed by insulin administration in the case of chemically-induced diabetes. Whether this discrepancy merely reflects sub-optimal insulin therapy in the ISDBB animals or some fundamental mechan-i s t i c difference between spontaneous and chemically-induced diabetes is unclear at the present time. In conclusion, the present data have shown that tissue antioxidant systems are altered in experimental diabetes and the reversal of these changes by insulin treatment seems indicative of their association with the 142 process of diabetogenesis. These studies also have shown that alterations in tissue antioxidant status in both groups of BB rats deserve further study to investigate whether or not they may be a factor determining the increased susceptibi l i ty of the BB Wistar rat strain to developing diabetes. The fact that comparable (although apparently not identical) changes in tissue ant i -oxidant status are present in ISDBB rats and in animals made diabetic by STZ or ALX administration suggests that the changes in the chemically-induced diabetes probably are more l ike ly attributed to hypoinsulinemia rather than direct toxic effects exerted by the diabetogenic agents. 6.6 Allopurinol Nand Experimental Diabetes Allopurinol (ALP) is able to inhibit xanthine oxidase and consequently the generation of superoxide. Recently, Fridovich (226) examined the l i t e r -ature concerning superoxide and suggested that the superoxide radical can i t se l f exert deleterious effects in l iving cel ls independently of its ab i l i ty to interact with Wftz with the resulting formation of the powerful oxidant, the hydroxyl radical. On the other hand, i t has been suggested that B-cells may be particularly vulnerable to ALX because they are highly effective in reducing ALX (leading to the formation of superoxide anion and ^ 2 ) or in catalysing the iron-dependent formation of the highly reactive hydroxyl radical from and superoxide radical (227). Therefore, inhibition of superoxide generation should be beneficial in reducing the formation of the highly toxic hydroxyl radical. It has become clear over the past few years that free-radical reaction products may play a role in the pathogenesis of various pancreatic diseases (228-231). Data from experimental models of acute pancreatitis (e.g., involving free fatty acid infusion and ischemia) have suggested the enzyme xanthine oxidase is important in generating oxygen-derived free radicals, whose important role in the pathogenesis of this disease was inferred from 143 the ab i l i ty of ALP to block the damaging effect caused by these free radicals (229), thereby eliminating the manifestations of pancreatitis. Sanfey et a l . (228,231) have demonstrated that the combination of SOD and CAT (but neither agent alone) wil l substantially ameliorate the manifesta-tions of acute pancreatitis that develops in response to FFA infusion into canine pancreas, suggesting a role of both superoxide and H2O2 in this model of pancreatitis. In this regard, Sanfey et a l . (229) have in fact suggested that the pancreas may be prone to the development of free radical-mediated ' injury because the endogenous pancreatic enzymes are capable of free radical generation via xanthine oxidase activation. Although we cannot, of course, compare acute pancreatitis (regardless of its pathogenesis) with the pancreatic damage in experimental diabetes, the fact that the involvement of free radical-related processes has been proposed in both acute pancreatitis (229) and in ALX-induced damage in 6-islet cel l s raised the poss ib i l i ty of whether or not xanthine oxidase may be involved in the pathogenesis of ALX diabetes. The present study was therefore undertaken to investigate whether or not xanthine oxidase-derived oxygen free radicals may play a role in the pathogenesis of diabetes using the specific xanthine oxidase inhibitor ALP. If the mechanism of free-radical production in experimental diabetes is mediated, at least in part, through the enzyme xanthine oxidase, then ALP should either prevent or at least ameliorate the diabetic state. In the present study, the prior administration of ALP (50 mg/kg body wt., intraperitoneally) lowered ketonuria in 7 out of 8 ALX-diabetic rats but fa i led in one diabetic animal which died in ketosis. The 8 other ALX-diabetic rats which did not receive ALP died within the f i r s t wk of ALX injection with marked ketosis. Although this finding is based on ketonuria without a systematic analysis of blood ketone bodies and plasma FFA, the 144 results are suggestive of an effect of ALP on ketone body metabolism. In a recent series of experiments (data not shown) we have been able to demonstrate the presence of increased levels of l ip id peroxides as well as increased levels of plasma ketone bodies following the injection of diabetogenic doses of ALX or STZ. A single dose of ALP (100 mg/kg body wt., intravenous injection given at various times (up to 1 h)) prior to or following ALX or STZ administration (50 mg/kg body wt. each) was able not only to lower ketonuria, but also ketonemia as well as prevent the increase in plasma triglycerides and l ip id peroxides observed in„diabet ic rats within 58-60 h after the induction of diabetes. Despite the protective effects produced by this dose of ALP, diabetes was not prevented. It is not clear whether these protective effects of ALP involve interdependent mechanisms or are merely independent effects. We have chosen this time period (58-60 h) of diabetes for two reasons. F i r s t , mortality rate increases a few hours after this time in non-ALP-treated diabetic animals and therefore a repre-sentative comparison would be impossible. Second and more importantly, by approximately 84 h of diabetes no significant difference could be observed between ALP-treated and non-ALP-treated diabetic animals, a finding that suggests the involvement of ALP and/or its metabolite oxypurinol in these protective effects, given that the ha l f - l i f e of ALP is 2-3 h and that of oxipurinol is 18-30 h (232). Experiments involving repeated administration of ALP to diabetic animals wil l be required to c l a r i f y this point. As yet, the mechanism(s) responsible for the protective effects of ALP on the increased levels of ketone bodies, triglycerides and l ip id peroxides remain(s) to be investigated in further studies. However, i t is possible that ALP exerts its effects through an inhibition of xanthine oxidase (with particular reference to its effect on tr ig lyceride levels) since a recent study (233) has demonstrated that elevated serum triglyceride levels did 145 correlate with increased serum xanthine oxidase act iv i ty in patients with hyperlipoproteinemia type IV. If this is in fact the case, there would seem to be a discrepancy between the effect of acute (100 mg/kg. body wt., intra-venously) ALP administration (which lowers plasma triglyceride levels) and the increased levels of triglycerides observed following chronic (1.9 mg/day in drinking water for 12 wk) ALP treatment. Whether or not the route of administration, dosage or duration of treatment contribute to this discrep-ancy is unknown. Furthermore, the poss ib i l i ty of a direct effect of ALP on ketone body metabolism, triglycerides and l ip id peroxidation mechanisms independently of inhibition of xanthine oxidase or possibly involving the scavenging of hydroxyl radicals (220), cannot be ruled out at present. It is important to point out that although alkali treatment has been recom-mended to correct diabetic ketoacidosis (234) the poss ib i l i ty that the ab i l i ty of ALP to lower ketonuria is in fact attributable to the high a lkal in i ty of the dissolving vehicle (pH 12.0) rather than to a direct effect of ALP per se has been ruled out in recent experiments in our labora-tory. Diabetic animals receiving the same alkaline vehicle (pH 12.0) fa i led to demonstrate similar protective effects. Ketonuria in uncontrolled diabetes is well known to result from increased l iver synthesis of ketones (235). Urinary ketones most l ike ly arise from fat breakdown and a simple decarboxylation of keto acids, although the exact metabolic origin of these and other carbonyl compounds during the early stages of diabetes has not been def initely established (236). It has been hypothesized that the abnormally high levels of carbonyl compounds found in urine may arise, at least in part, from l ip id peroxidation (236). Furthermore, rats placed on a diet deficient in both vitamin E and selenium and showing evidence of increased l ip id peroxidation (in association with increased plasma levels of MDA and decreased act iv ity 146 of GSH-PX), also showed increased levels of urinary ketones (237). In the present experiments, the ab i l i ty of ALP to lower both ketonuria and l ip id peroxidation supports such a possible relationship between these two para-meters (236,237). Much more work must be done to assess the role of increased l ip id peroxidative act iv i ty in the production of ketonuria and to further investigate the protective effects of ALP in experimental diabetes. The inabi l i ty of ALP to prevent diabetes raises several poss ib i l i t ies . F i r s t l y , the dose may have been insuff icient to produce complete inhibition of xanthine oxidase due to insuff icient duration of pretreatment and higher doses or longer periods of treatment might be effective. Secondly, xanthine oxidase-generated superoxide radicals may, not play a major role in the induction of diabetes. 6.7 Erythrocyte Antioxidant Status The three main conditions which favor peroxidation are a high degree of unsaturation in the l ip id substrate, a rich supply of oxygen, and the presence of transition-metal catalysts. Although these conditions are to varying degrees f u l l f i l l e d by most l iving systems, perhaps none meets the requirements as well as the red ce l l (238). The l ipids in these cel ls are highly unsaturated, the cel l s are exposed to a higher oxygen tension than any other tissue, and they are loaded with Hb-iron, one of the most powerful peroxidation catalysts known (238). Red ce l l s , which contain the biochem-ical components of the oxidation defence mechanisms present in tissues throughout the body, may provide a convenient model system with which to explore the possible role of oxygen free radical processes in determining the nature and severity of diabetic complications in man. Evidence has been found which indicates the presence of alterations in reactive oxygen radical scavenging processes and in the susceptibi l i ty of red cel l s from diabetic patients and animals to peroxidative damage. The 147 increased activites of GSH-PX in animals with chemically-induced diabetes and GSSG-RD in erythrocytes of a l l diabetic animals and patients as well as the maintenance of basal GSH levels and the reduction in GSH depletion in response to H202~induced damage in diabetic erythrocytes may be suggest-ive of an adaptive response to an increased oxidative stress. There is evidence to suggest that hyperglycemia may be involved in this protective response in erythrocytes by way of increased intracel lu lar generation of NADPH from the HMPS (239). NADPH is used as a hydrogen donor by GSSG-RD to regenerate GSH from GSSG, thereby maintaining the level of GSH which is required for the detoxification of peroxides by GSH-PX. In this regard, i t should be mentioned that red ce l l s , unlike B-cells, can modulate the HMPS act iv i ty by a rapidly effective glucose-dependent.regulation (121). B-cells can increase the shunt act iv i ty by de novo synthesis (over days) of G5PD and phosphogluconate dehydrogenase (121). Such "coarse" control over HMPS act iv i ty by B-cells may contribute to the susceptibi l i ty of these cel l s to oxidative stress (82). Such HMPS-related protection by glucose has been implicated in the prevention of red cel l CuZn-SOD and CAT inhibition by the hemolytic agent napthoquinone sulfonate (240). GSSG-RD elevation has been reported in erythrocytes from diabetic patients (90), while GSH-PX has been reported to be either decreased (51,89) or increased (66). Matkovics et a l . (66) suggested that the increased GSH-PX in diabetic erythrocytes may be a response to an increased oxidative stress to these ce l l s . Erythrocytes of NDLM rats, unlike those of other diabetic animals or humans, showed s ignif icantly reduced CAT act iv i ty. The fact that the GSH-PX act iv i ty and GSH content in these erythrocytes were within the normal range tends to exclude the poss ib i l i ty that the relative deficiency in CAT act iv-i ty may render red cel ls more susceptible to endogenous oxidative stress. 148 Furthermore, because CAT has a low af f in i ty for H^G ,^ i t has been suggested that GSH-PX is the main mechanism for H2O2 detoxification within the erythrocyte and that red cel l CAT is of l i t t l e importance in this regard (239). SOD act iv i t ies in red blood cel ls were s ignif icantly increased in erythrocytes of diabetic patients as compared with controls, but were unchanged in diabetic animals. Two reports have appeared concerning the CuZn-SOD act iv i ty in erythrocytes from human diabetics. There was a small but significant decrease in erythrocyte CuZn-SOD act iv i ty in subjects with IDDM maintained on insulin therapy for at least 2 yr but this decrease has been suggested to be of no biological significance (89). In contrast, subjects with NIDDM maintained on oral hypoglycemic agents have a marked decrease in erythrocyte CuZn-SOD act iv ity (66). These differences may be attributable to different etiologies of IDDM and NIDDM, as well as to the differences in metabolic control achieved by long-acting insulin injections vs. oral hypoglycemic agents (82). However, the present study indicates that CuZn-SOD act iv i ty in erythrocytes of human diabetics was increased regardless of the type of diabetes or, for that matter, the type of treat-ment. This discrepancy is unlikely to be attributable to differences in metabolic control, since erythrocytes of ALX-diabetic rats maintained on insulin therapy for a 12 wk period did not show increased SOD act iv ity despite the hyperinsulinemia observed in these animals. Red cel l membranes of IDDM patients, but not those of NIDDM patients or those of diabetic animals, showed an increased act iv i ty of NADH dehydro-genase compared to control. It seems unlikely that this increase can be attributed to a lack of insulin which is reported (241) to exert an inh ib i -tory effect on this enzyme, since diabetic animals with marked hypoinsulin-emia fai led to show a comparable increase in the act iv i ty of this enzyme. 149 6.8 Erythrocyte Susceptibil ity to Oxidative Stress The enhanced susceptibi l ity of red cel ls to GSH depletion in food-deprivation might have resulted from a decreased supply of NADPH (due to hypoglycemia) in combination with a decreased GSSG-RD act iv i ty. Regarding H202~induced increases in red cel l MDA levels, 3 days of food-depriva-tion (associated with decreased plasma glucose and insulin levels) resulted in lower levels of MDA than controls following H2O2 challenge, despite the fact that GSH depletion was enhanced, indicating that despite the potentially adverse effects of food-deprivation (and hypoglycemia) on GSH depletion in erythrocytes, there is no evidence of a corresponding increase in susceptibi l i ty toward oxidative stress. The reduced susceptibi l i ty of red cel l s to peroxide-induced GSH depletion was present in al l diabetic animals and subjects with hyperglycemia and this seems to be rather more dependent on plasma glucose levels than on the diabetic state per se. As discussed previously, erythrocytes have the machinery to metabolize glucose through the HMPS to generate NADPH, which is ut i l ized by GSSG-RD to regener-ate GSH from GSSG, thereby maintaining levels of GSH. In in vitro experi-ments, we have incubated erythrocytes with various concentrations of glucose for up to 60 min and have demonstrated protection against peroxide-induced GSH depletion within 5 min of incubation similar to that observed in diabetes (hyperglycemia) but no effect on MDA levels (up to a maximal incubation time of 60 min) was detectable (data not shown). The reduced susceptibi l i ty of diabetic erythrocytes to GSH depletion which was also predicted on the basis of the observed increases in GSSG-RD act iv i ty would have been expected to prevent MDA production (by virtue of its ab i l i ty to decompose H 20 2 and prevent l ip id peroxidation). However, the present study revealed that erythrocytes from diabetic animals and patients showed higher MDA levels in response to exogenous H o 0 o challenge, suggesting 150 the presence of increased susceptibi l ity to oxidative stress. The molecular basis of this abnormality is as yet unknown. The present study also demonstrated that erythrocytes from NIDDM patients produced higher MDA levels in response to the ^ 2 than those from IDDM patients, although in both cases MDA levels were higher than those of controls. Comparable alterations have been reported in erythrocytes of NIDDM patients (66,51), but IDDM patients were not included in these studies. Furthermore, although the number of patients investigated here was too small to draw firm conclusions, i t is interesting to note that red cel ls from patients with severe complications showed s ignif icantly increased levels of MDA while those from patients with no complications were ind i s t in -guishable from those of control subjects. Very recently, Uzel et a l . (51) observed that erythrocyte susceptibi l i ty to H202-induced MDA production in NIDDM patients was higher in diabetic individuals with retinopathy than those without retinopathy, although in both cases the levels of MDA were s ignif icantly higher than controls. Clearly more studies in a larger number of patients are required to elucidate the possible mechanism(s) responsible for the as yet unexplained increased MDA production in these patients and its relationship to the nature and severity of diabetic complications. Trotta et a l . (242) demonstrated that, in red cel ls exposed to t-butyl hydroperoxide, glucose protection of Hb decomposition by hydroperoxides (and the generation of TBA-reactive materials) was only partly mediated by GSH. In the present study, i t has been shown that alterations in the suscept ib i l -ity of red cel ls to peroxide-induced GSH depletion did not appear related to the capacity of erythrocytes to generate MDA in response to H2O2, and increased levels of MDA in diabetic red cel ls were observed despite the protective effect of hyperglycemia on red cel l GSH. The poss ib i l i ty that the measured MDA levels in red cel l s could have resulted, at least in part, 151 from an interaction between the l ip id hydroperoxide (a product of a series of reactions involving I^ O^ and membrane l ipids) and Hb, resulting in their decomposition to various end-products, including MDA (243), cannot be ruled out in the present study. Therefore, our results demonstrate that while erythrocytes from poorly controlled diabetic animals maintained normal GSH levels (via HMPS, as described previously) they showed an increased susceptibi l i ty to l ip id peroxidation, suggesting that GSH depletion and MDA production in red cel ls exposed to in vitro peroxide challenge seem to be independent of each other. On the other hand, vigorous insulin treatment (as reflected by relative hypoglycemia and by hyperinsulinemia) not only fa i led to normalize MDA levels in reponse to H2O2 challenge but, rather, markedly lowered them, suggesting a decreased susceptibi l i ty to oxidative stress. Further studies are required to investigate whether or not this altered susceptibi l -i ty of diabetic red ce l l membrane l ipids to in vitro peroxidation may contribute, at least partly, to the many changes (e.g., alterations in membrane f lu id i t y (244) and the decreased deformabi1ity (33)) of erythro-cytes which have both been reported in diabetic erythrocytes. 6.9 Erythrocyte and Plasma Lipid Components In view of the known dynamic equilibrium between l ip id components of plasma and erythrocyte membranes (245) and the marked plasma l ip id abnormal-i t ies in diabetes, we have undertaken a detailed study of erythrocyte membrane l ipids to investigate, among other things, the possible molecular basis of their increased susceptibi l i ty to peroxidative damage in diabetes. Despite the significant increase in plasma levels of cholesterol in STZ-diabetes and of phospholipid in both STZ- and ALX-induced diabetes, neither total phospholipid nor cholesterol contents were s ignif icantly altered in diabetic erythrocytes. No significant changes were noted in 152 erythrocyte C/P ratio, an important determinant of membrane f lu id i t y , in agreement with the results of van Doormal et a l . (246) in diabetic patients, but not those those of Bryszewska et a l . (247), who reported an increase in this ratio in type I diabetics. Chandramouli and Carter (248) studied cholesterol and phospholipid composition of erythrocyte ghosts in 4-8 wk STZ-diabetic rats and reported a decrease in cholesterol but not phospho-l ip id levels. In contrast, Sosin et a l . (249) reported an increased choles-terol content of erythrocytes from 10 day STZ-diabetic rats, while Kamada and Otsuji (244) reported no significant changes in erythrocyte C/P ratio in we11-controlled type II diabetic patients. Clearly, variations in the duration of the disease, species and/or the degree of diabetic control may in part explain these discrepancies. BB-diabetic erythrocyte membranes did show a significant decrease in cholesterol content and in the molar ratio of C/P. The observed inverse correlation between plasma and membrane choles-terol levels noted in these animals was somewhat unexpected in view of the tendency of red cel ls to accumulate cholesterol from lipoproteins under certain pathological conditions (250). However, the reason for the lack of a comparable decrease in cholesterol content of red ce l l membranes of other diabetic animals and patients is not clear at present. A study of experi-mentally-induced hyperlipidemia (produced by feeding with a high fat-high cholesterol diet) in STZ-diabetic rats showed that the marked elevations of plasma cholesterol produced in these animals were associated with corres-ponding increases in erythrocyte membrane cholesterol (251). It is of interest that a signif icant increase in PE content was observed only in ALX-diabetic erythrocyte membranes. PE contains a high concentration of polyunsaturated fatty acids which are susceptible to peroxidative injury (158). The major renewal pathway for PE fatty acids through PC (252) has been shown to be stimulated by peroxidant injury 153 (253). The fact that ALX-diabetic erythrocytes showed signif icantly greater increases in MDA than STZ-diabetic erythrocytes at high concentrations of H2O2 could reflect a more stimulatable renewal pathway, and this could be the reason for the observed increase in PE. Another possible reason for the observed increase in the content of PE could be a decrease in methyl transferase act iv i ty which converts PE to PC. Such a decrease has been reported in ALX-diabetic rats (254). Increases in PE content have also been reported both in l iver microsomes of STZ-diabetic rats (255) and in plate-lets from poorly-controlled diabetic patients (256). On the other hand, Chandramouli and Carter (248) showed no significant changes in any of the major phospholipid classes in STZ-diabetic rat erythrocyte ghosts. The present finding of a decrease in erythrocyte membrane SM content and the lack of change in PC concentration contrasts with that of Kamada et a l . (244) who found a relative decrease in PC and a relative increase in SM contents in erythrocytes of diabetic patients. PC and SM are the major phospholipid components of plasma lipoproteins, whose exchange with erythro-cytes results in the relative enrichment of PC and SM at the external mem-brane surface of the red cel l (257). An alteration in PC/SM would be expected to perturb the structural and functional characteristics of membranes (258) because of the inherent structural differences between these two phospholipid species as reflected, for example, in their ' opposite effects on membrane f l u id i t y . SM tends to confer increased r ig id i ty to biomembrane structures while PC produces an increased f l u id i t y . Studies of diabetic red cel l s have revealed a reduction in membrane f l u id i t y which was greatest in poorly-controlled patients (259). Our finding that the PC/SM ratio is increased in both plasma and red cel l s of diabetic animals suggests that the red ce l l defect may be an acquired one. However, such an increase in PC/SM ratio would predict an increased membrane f l u id i t y , a finding which 154 is inconsistent with the reduction reported in diabetic patients (244). While this discrepancy cannot be readily explained, i t may reflect the involvement of additional as yet unidentified membrane compositional a l tera-tions which are related to the degree of metabolic control as well as to the severity of c l in ica l diabetes. In contrast to the observed increases in total phospholipids and also the PC, PE and PI contents of diabetic plasma, LPC was s ignif icantly reduced in STZ-diabetes. The reason for the decrease in LPC is unknown; however, this may have resulted from an inhibition of lecithin:cholesterol acyltransferase (LCAT). Misra et a l . (260) found that LCAT act iv i ty was decreased in ketotic diabetic patients but returned to normal with improved diabetic control. Kamada and Otsuji (244) reported a slight but non-significant increase in LCAT act iv i ty in we11-controlled NIDDM patients. Thus, the above-mentioned phospholipid changes may represent an altered dynamic exchange between plasma and erythrocyte membranes in diabetes, possibly related to the we11-documented plasma lipoprotein abnor-malities present in this disease, particularly when patients are poorly controlled (261). Alternatively, these may be a manifestation of a compen-satory adaptation, possibly to conditions of increased oxidative stress. These data do, however, show that erythrocytes in diabetes are influenced by alterations in plasma to some degree but pathological changes in plasma are by no means direct ly or simply predictive of the situation in red ce l l s . 6.10 Similarit ies Between Erythrocyte Alterations in Animal and Human Diabetes Despite the fact that erythrocytes from various models of diabetes showed variable changes in antioxidant properties, three common alterations were evident. The reduced susceptibi l i ty to peroxide-induced GSH depletion was present both in diabetic animals and in diabetic patients and this seems 155 to be rather more dependent on plasma glucose levels than on the diabetic state per se. Another common abnormality was an increased production of MDA in poorly controlled diabetic animals and, to a lesser extent, in diabetic patients following exposure of red cel l s to H 20 2 in vitro. Also, the increased act iv i ty of GSSG-RD in red cel ls is common to a l l diabetic animals and subjects we have studied. The reason for the inabi l i ty of insulin treatment or oral hypoglycemic therapy to normalize this enzyme act iv i ty is not clear at present. 6.11 Summary and Conclusions Free radicals are chemical species characterised by the presence of an unpaired electron in their molecular structure. Because they have short, half l ives, their concentration in most biological materials (including pathologically altered tissues) is too low for direct detection by electron-spin-resonance (ESR) spectroscopy (28), so that the evidence for their act iv i ty is largely indirect (262). However, in recent years the term "oxidative stress" has come to be used to denote a disturbance in the peroxidant-antioxidant balance in favor of the former (263). The fact that body weight loss was associated with alterations in tissue antioxidant mechanisms suggests that in studies of free radical scavenging enzyme alterations in animal models of diabetes (where decreases in body weight are a common feature) i t wil l be important to differentiate between changes due to the decrease in body weight and those due to the disease i t se l f . The alterations in the act iv i t ies of cardiac CAT, pancreatic CAT and CuZn-SOD, hepatic CAT and GSH, and renal GSH-PX seen in ALX- and STZ-diabe-t i c rats qualitatively parallel those seen following food-deprivation. It should also be pointed out that the magnitude of the alterations in diabetic rats increased with the degree of body weight loss. If the changes observed 156 in the food-deprived rats are attributable to hypoinsulinemia, body weight loss would be expected to aggravate the hypoinsulinemia in the diabetic animals and thereby increase the severity of tissue antioxidant abnormali-t ies . However, the long-term hypoinsulinemia, together with hyperglycemia and elevated plasma l ipids seen in chronic diabetes, are associated with qualitatively different changes to those observed in the short-term effects of hypoinsulinemia and hypoglycemia seen in food-deprived animals. These results suggest the presence of more than one factor in the development of these alterations in tissue antioxidant status in diabetes, and that the duration of hypoinsulinemia ( i .e, diabetes) could be one factor. Therefore, in the l iver (and to less extent the kidney) which exhibited high act iv ity of tissue antioxidant enzymes, chronic diabetes (or prolonged hypoinsulin-emia) may possibly have resulted in a depression of antioxidant mechanisms that might lead to increased susceptibi l ity to oxidative stress in these organs. The ab i l i ty of insulin to reverse the above-mentioned changes suggests that they are hypoinsulinemia-related and that this hormone can influence tissue "antioxidant" status. However, the persistence of decreased CuZn-SOD act iv i ty in both l iver and kidney of ALX-diabetic rats after 12 wk of t reat-ment with insulin is not clear at present, and requires further investiga-tion to test whether i t reflects the presence of a residual def ic i t in tissue antioxidant capacity of l iver and kidney despite insulin treatment, or is merely an outcome of a direct effect exerted by ALX. The two diabetogenic agents ALX and STZ differed in their effects on GSH levels in heart and pancreas. Although neither was s ignif icantly altered in STZ-diabetic animals, cardiac GSH content was s ignif icantly higher in ALX-diabetic rats compared to controls, while pancreatic levels of GSH in ALX-diabetic rats were decreased. However, i t should be mentioned 157 that these dissimilar responses in cardiac and pancreatic. GSH levels in ALX-diabetes occurred despite the fact that similar increases in GSSG-RD act iv i ty were present in both tissues. Therefore, these differences in GSH levels suggest that other factors (e.g., relating to synthesis or degrad-ation) may be involved in determining the levels of tissue GSH in experimen-tal diabetes. Although the significance of these differing alterations in cardiac and pancreatic GSH contents in ALX- and STZ-diabetes is not clear at the moment, they are presumbly indicative of differences in GSH metabolism in these two models of chemically-induced diabetes which might have some bearing on the mechanisms of diabetogenesis of these two diabetogenic agents. The results obtained from BB rats demonstrate that two types of a l ter -ations in antioxidant status exist: strain-related changes (increased CAT activ ity in pancreas and decreased GSH levels in pancreas and l iver of both ISDBB and their NDLM) and diabetes-related changes (manifested by increases in act iv i t ies of cardiac CAT and GSSG-RD, pancreatic CuZn-SOD and GSSG-RD, and renal GSH-PX). The tissue antioxidant GSH revealed a complex pattern of changes (an increase in heart of ISDBB and decrease in pancreas and l iver of both ISDBB and NDLM). In view of these considerations, NDLM are less than ideal controls for studies involving ISDBB. Therefore, caution must be exercised in interpreting the significance of changes in tissue antioxidant status in "control" NDLM rats. Whether or not these "strain-related" a l ter -ations in antioxidant status increase the susceptibi l i ty of these animals to developing diabetes remains unknown. The fact that comparable (although not identical) enzymatic changes are present in BB Wistar rats with overt diabetes and in animals made diabetic by STZ or ALX administration indicates that these models of chemically-induced diabetes are suggestive of a l tera-tions more l ikely attributable to hypoinsulinemia than to direct chemical effects exerted by the diabetogenic agents. Our data also demonstate that 158 the alterations in GSH antioxidant status characterizing ALX-diabetes more closely paralleled changes seen in the ISDBB rat than did those in the diabetic state induced by STZ. Certain alterations were observed in red cel l s from diabetic patients and animals with experimental diabetes suggesting that these alterations are more l ikely to be diabetes-related than species-dependent. Despite the apparent protective effect of hyperglycemia on red cel l GSH depletion (via the supply of NADPH through HMPS) the susceptibi l i ty of red cel l s from diabetic patients and animals to peroxidative damage is increased, suggest-ing that these two parameters are independent of each other. In addition, insulin treatment did not normalize MDA production in red cel l s subjected to oxidative challenge since vigorous insulin treatment to both ALX- and STZ-diabetic rats resulted in a markedly decreased MDA production in response to H2O2. Moreover, GSSG-RD act iv i ty of red cel l s was increased in both uncontrolled and insulin-treated diabetic animals as well as in diabetic patients maintained on various types of therapy ( insul in, oral hypoglycemic drugs, or diet). However, dissimilar alterations in erythrocyte antioxidant enzymes were also observed in erythrocytes from diabetic subjects and animals. For example, diabetic patients showed an increased act iv i ty of CuZn-SOD, while erythrocytes from diabetic animals showed no alterations in the act iv i ty of this enzyme. Erythrocyte membrane NADH-dehydrogenase act iv i ty was increased only in diabetic patients with Type I diabetes, but not in Type II diabetes or in diabetic animals. Erythrocytes from ALX- and STZ-diabetic animals showed an increase in the act iv i ty of GSH-PX and those from NDLM BB rats showed a decrease in CAT act iv i ty, conditions that were not observed in human diabetes. One of the objectives for studying red cel l s was to investigate their 159 predictive value in assessing changes observed in other tissues. The finding that GSH-PX act iv i t ies in erythrocytes and kidneys both are elevated and both are normalized by insulin treatment deserves further investigation to determine its possible relevance to renal complications in diabetes. Thus, our data have shown that tissue antioxidant systems are altered in experimental diabetes and the reversal of these changes by insulin treat-ment seems indicative of their association with hypoinsulinemia. The complex patterns observed vary from one tissue to another but may be the result of compensatory changes, usually involving enzymes whose act iv i ty in the particular tissue may be l imit ing, and direct inhibitory effects of endogenous oxidants on the enzymatic components of tissue antioxidant systems. Furthermore, the duration of hypoinsulinemia might contribute to the nature of alterations in antioxidant mechanisms. However, the present study cannot conclusively specify the source of the increased oxidative stress in diabetes, although preliminary studies in our laboratory suggest that alterations in l ip id metabolism may be relevant in this regard. Although the possible relationship of such changes in tissue antioxidant status to the development of secondary diabetic complica-tions is at present unknown, our findings do suggest approaches to exploring this important question by the use of pharmacological interventions known or expected to alter tissue antioxidant status. Based on our finding that insulin treatment did not normalize a l l the changes in tissue antioxidant defense mechanisms and in view of.the abi l i ty of ALP to lower plasma ketone and l ip id peroxide levels as well as to augment CuZn-SOD act iv i t ies in diabetic kidney, a target tissue for diabetic complications, future studies in our laboratory wil l focus on the effects of pharmacological interventions affecting tissue antioxidant status on the course of experimental diabetes, as an approach to devising rational therapeutic measures for the management 160 of the many and varied complications of c l in ica l diabetes. One obvious possibl ity could be a combination of insulin or hypoglycemic drugs with an antioxidant or ALP. In regard to chronic ALP treatment in drinking water, i t must be emphasized that the resultant effect of ALP to normalize renal CuZn-SOD levels must be interpreted with caution since the same treatment also increased plasma triglyceride and phospholipid levels in diabetic animals and resulted in a significant reduction in pancreatic GSH content, a characteristic feature of BB rats. F inal ly, as far as the antioxidant defense mechanisms are concerned, our results suggest that diabetes is associated with some common alterations in these mechanisms regardless of the model (chemically-induced versus the spontaneous type of diabetes (Appendices I—IV)) or species used (animals versus human diabetes (Appendix V). Furthermore, for comparison and in order to test the val idity of animal models of diabetes, rabbits made diabetic with ALX ( 100 mg/kg body wt.) for other ongoing experiments in our laboratory were also examined and the above-mentioned parameters measured. Results comparable to those observed in the rat models were obtained (data not shown) providing further support for the suggestion that alterations in antioxidant defense mechanisms are more diabetes-related than species-dependent. 161 Appendix I. Alterations in antioxidant systems of heart in various models of diabetes. CAT CuZn-SOD GSH-PX GSSG-RD GSH Female Rats STZ-DM i N.S. N.S. N.S. STZ-Ins. 4 wk N.S. N.S. N.S. N.S. N.S. ALX-DM + N.S. N.S. + + ALX-Ins. 6 wk N.S. N.S. N.S. N.S. N.S. ALX-Ins. 12 wk N.S. N.S. N.S. N.S. N.S. ALP-ALX-DM + N.S. N.S. t Food-deprivation + + N.S. N.S. N.S. . Male Rats ISDBB t N.S. N.S. + T NDLM N.S. N.S. N.S. N.S. N.S. N.S., no significant difference. + , s ignif icantly increased compared with control. +, s ignif icantly decreased compared with control. CAT, catalase; CuZn-SOD, CuZn-superoxide dismutase; GSH-PX, glutathione peroxidase; GSSG-RD, glutathione reductase; GSH, reduced glutathione. STZ, streptozotocin; ALX, alloxan; Ins., insulin-treated; ALP, allopurinol-treated; DM, diabetes mellitus; ISDBB, insulin-treated spontaneously diabetic BB rat; NDLM, non-diabetic littermates of BB rat. 162 Appendix II. Alterations in antioxidant systems of pancreas in various models of diabetes. CAT CuZn-SOD GSH-PX GSSG-RD GSH Female Rats STZ-DM + + N.S. t N.S. STZ-Ins. 4 wk N.S. N.S. N.S. N.S. N.S. ALX-DM + N.S.* N.S. -t- N.S.* ALX-Ins. 6 wk N.S. N.S. N.S. N.S. N.S. ALX-Ins. 12 wk N.S. N.S. N.S. N.S. N.S. ALP-ALX-DM + N.S. N.S. t N.S. Food-deprivation + + N.S. N.S. N.S. Male Rats ISDBB + N,S. + 4-NDLM + N.S. N.S. N.S. 4-N.S., no significant difference, t , s ignif icantly increased compared with control. +• , s ignif icantly decreased compared with control. * , indicates significance ( + for CuZn-SOD and +< for GSH) when compared with ALP-treated groups. The abbreviations used are the same of those in Appendix I. 163 Appendix III. Alterations in antioxidant systems of l iver in various models of diabetes. CAT CuZn-SOD GSH-PX GSSG-RD GSH Female Rats STZ-DM 4 4 4 N.S. 4-STZ-Ins. 4 wk N.S. N.S. N.S. N.S. N.S. ALX-DM 4 4 4 N.S. 4 ALX-Ins. 6 wk N.S. N.S. N.S. N.S. N.S. ALX-Ins. 12 wk N.S. 4 N.S. N.S. N.S. ALP-ALX-DM N.S. N.S. 4 N.S. N.S. Food-deprivation 4 N.S. N.S. N.S. 4 Male Rats ISDBB N.S. N.S. N.S. N.S. 4 NDLM N.S. N.S. N.S. N.S. 4 N.S., no significant difference, t , s ignif icantly increased compared with control. + , s ignif icantly decreased compared with control. The abbreviations used are the same of those in Appendix I. 164 Appendix IV. Alterations in antioxidant systems of kidney in various models of diabetes. CAT CuZn-SOD GSH-PX GSSG-RD GSH Female Rats STZ-DM 4- 4- + N.S. N.S. STZ-Ins. 4 wk N.S. N.S. N.S. N.S. N.S. ALX-DM 4- 4- t N.S. N.S. ALX-Ins. 6 wk N.S. N.S. N.S. N.S. N.S. ALX-Ins. 12 wk N.S. 4- N.S. N.S. N.S. ALP-ALX-DM 4- N.S. + N.S. N.S. Food-deprivation N.S. N.S. + N.S. N.S. Male Rats ISDBB N.S. N.S. + N.S. N.S. NDLM N.S. N.S. N.S. N.S. N.S. N.S., no significant difference • t , s ignif icantly increased compared with control • + , s ignif icantly decreased compared with control • The abbreviations used are the same of those in Appendix I. 165 Appendix V. Alterations in antioxidant enzymes of erythrocytes in various models of diabetes. CAT CuZn-SOD GSH-PX GSSG-RD Female Rats STZ-DM N.S. N.S. + + STZ-Ins. 4 wk N.S. N.S. N.S. ALX-DM N.S. N.S. + ALX-Ins. 6 wk N.S. N.S. N.S. + ALX-Ins. 12 wk N.S. N.S. N.S. ALP-ALX-DM N.S. N.S. N.S. + Food-deprivation N.S. N.S. N.S. 4-Male Rats ISDBB N.S. N.S. N.S. t NDLM 4- N.S. N.S. N.S. Human DM N.S. + N.S. + N.S., no significant difference. , s ignif icantly increased compared with control. 4-, s ignif icantly decreased compared with control. The abbreviations used are the same of those in Appendix I. 166 REFERENCES 1. Arky, R.A. Prevention and therapy of diabetes mellitus. Nutr. Rev. 41: 165-173, 1983. 2. Toniolo, A., Onodera, T., Yoon, J.-W. and Notkins, A.L. Induction of diabetes by cumulative environmental insults from viruses and chemicals. Nature 288; 383-385, 1980. 3. Assal, J.Ph. Hyperglycemics et complications du diabete. Schweiz.  med. Wschr. 109: 1791-1801, 1979. 4. Effect of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. VIII. Evaluation of insulin therapy: Final report. Diabetes 31 (Suppl. 5): 1-26, 1982. 5. Stout, R.W. Insulin and atheroma—an update. Lancet I: 1077-1079, 1987. 6. Young, C.W. Rationale for glycemic control. Am. J. Med. 79 (Suppl. 3B): 8-11, 1985. 7. Frier, B.M. and Hilsted, J. Does hypoglycemia aggravate the complica-tions of diabetes? Lancet I I ; 1175-1176, 1985. 8. Levin, M.E., Boniuk, I., Anderson, CB. and Avi o l i , L.V. Prevention and treatment of diabetic complications. Arch. Intern. Med. 140: 691-696, 1980. 9. Bennet, P.H. Diabetes and heart disease - the magnitude of the problem. In: Clinical Cardiology and Diabetes (part I) , Scott, R.C. (ed.), Futura Publishing Co., New York (1982), pp. 3-12. 10. Smith, J.W. Heart disease in diabetes mellitus. In: Management of  Diabetes Mellitus, Bressler, R. and Johnson, D.G.( eds.), John Wright P.S.G. Inc., Boston, MA , pp. 285-298, 1982. 11. Study Group, European Atherosclerosis Society. Strategies for the prevention of coronary heart disease: A policy statement of the European Atherosclerosis Society. Eur. Heart. J. _8: 77-88, 1987. 12. Brunzell, J.D., Chait, A. and Bierman, E.L. Plasma lipoproteins in human diabetes mellitus. In: The Diabetes Annual /l. Alberti, K.G.M.M. and K r a l l , L.P. (eds.), Elsevier Science Publishers B.V., pp. 463-479, 1985. 13. Chait, A., Bierman, E.L. and Brunzell, J.D. Diabetic macroangiopathy. In: The Diabetes Annual /l. Alberti, K.G.M.M. and Kr a l l , L.P. (eds.), Elsevier Science Publishers B.V., pp. 323-348, 1985. 14. Mark, J.L. Oxygen free radicals linked to many diseases. Science 235 : 529-531, 1987. 167 15. Greene, D.A., Lattimer, S.A. and Sima, A.A.F. Sorbitol, phosphoinositides, ,and sodium-potassium-ATPase in the pathogenesis of diabetic complications. New Engl. J . Med. 316: 599-606, 1987. 16. Clements, R.S., Jr. and Be l l , D.S.H. Complications of diabetes. Prevalence, detection, current treatment, and prognosis. Am. J . Med. 79 (Suppl.5A): 2-7, 1985. 17. Greene, D.A. Acute and chronic complications of diabetes mellitus in older patients. Am. J . Med. 80 (Suppl.5A): 39-53, 1986. 18. Minaker, K.L. Aging and diabetes mellitus as risk factors for vascular disease. Am. J . Med. 82 (Suppl.IB): 47-53, 1987. 19. Dormandy, T.L. An approach to free radicals. Lancet II: 1010-1014, 1983. 20. Katz, M.A. The expanding role of oxygen free radicals in c l in ica l medicine. West. J . Med. 144: 441-446, 1986. 21. Monnier, V.M., Vishwanath, V., Frank, K.E., Elmets, C.A., Dauchot, P. and Kohn, R.R. Relation between complications of type I diabetes mellitus and collagen-linked fluorescence. New Engl. J . Med. 314: 403-408, 1985. . 22. Edi tor ia l . Browning and diabetic complications. Lancet J_: 1192-1193, 1986. 23. Jones, A.F., Jennings, P.E., Wakefield, A., Winkles, J.W., Lunec, J . and Barnett, A.H. Protein browning and diabetic complications. Lancet II: 459, 1986. 24. Simmons, K.J. Defence against free radicals has therapeutic implica-tions. J . Amer. Med. Assoc. 251: 2187-2192, 1984. 25. Mimnaugh, E.G., Trush, M.A., and Gram, T.E. Stimulation by adriamycin of rat heart and l iver microsomal NADPH-dependent l ip id peroxidation. Biochem. Pharmacol. 30: 2792-2804, 1981. 26. Doroshow, J.H. Effect of anthracycline antibiotics on oxygen radical formation in rat heart. Cancer Res. 43: 460-472, 1983. 27. Hall iwell, B. and Gutteridge, J.M.C. Lipid peroxidation, oxygen radicals, cel l damage and antioxidant therapy. Lancet _I; 1396-1397, 1984. 28. Slater, T.F., Cheeseman, K.H., Davies, M.J., Proudfoot, K. and Xin, W. Free radical mechanisms in relation to tissue injury. Proc. Nutr.  Soc. 46: 1-12, 1987.. 29. Buckley, G.B. The role of oxygen free radicals in human disease processes. Surgery 94: 407-411, 1983. 30. McCord, J.M. Oxygen-derived free radicals in post-ischemic tissue injury. N. Eng. J . Med. 312: 159-163, 1985. 168 31. Brownlee, M. and Cerami, A. The biochemistry of the complications of diabetes mellitus. Ann. Rev. Biochem. 50: 385-432, 1981. 32. Baba, Y., Kai, M., Kamada, T., Setoyama, S., and Otsuji, S. Higher levels of erythrocyte membrane microviscosity in diabetes. Diabetes 28: 1138-1140, 1979. 33. Schmid-Schonbein, H and Volger, E. Red cel l aggregation and red cel l deformability in diabetes. Diabetes 25(Supp1. 2): 897-902, 1976. 34. Di tzel , J . , Anderson, H. and Peter, N.D. Oxygen af f in i ty of hemo-globin and red cel l 2,3-diphosphoglycerate in childhood diabetes. Acta Pediatr. Scand. 64: 355-361, 1975. 35. T r i v e l l i , L.A., Ranney, H.M. and Lai, H.T. Hemoglobin components in patients with diabetes mellitus. New Engl. J . Med. 284: 353-357, 1971. 36. Jones, R.L. and Peterson, CM. Hematologic alterations in diabetes mellitus. Am. J . Med. 70: 339-352,1981. 37. Lowe, G.D.O., Lowe, J.M., Drummond, M.M., Reith, S., Belch, J . J . F . , Kesson, CM. , Wylie, A., Foulds, W.S., Forbes, C D . , MacCuish, A .C and Manderson, W.G. Blood viscosity in young male diabetics with or without retinopathy. Diabetologia 18: .359-363, 1980. 38. Paisey, R.B., Harkness, J . , Hartog, M. and Chadwick, T. The effect of improvement in diabetic control on plasma and whole blood viscosity. Diabetologia 19: 345-349, 1980. 39. Krahenbuhl, B. and Mossaz, A. On vascular non-disease of the foot in diabetes. New Engl. J . Med. 312: 1190,1985. 40. Coffman, J . and LoGerfo, F.W. On vascular non-disease of the foot in diabetes. New Engl. J . Med. 312: 1191,1985. 41. Yoshikawa, T., Furukawa, Y., Wakamatsu, Y., Takemura, S., Tanaka, H. and Konda, M. Experimental hypoxia and l ip id peroxide in rats. Biochem. Med. 27: 207-213,1982. 42. Schneiderman, G. and Goldstick, T. Carbon monoxide-induced arterial wall hypoxia and atherosclerosis. Atherosclerosis 30: 1-15, 1978. 43. Kjeldsen, K., Astrup, P. and Wanstrup, J . Reversal of rabbit atheromatosis by hyperoxia. J . Atherosc. Res. 10: 173-178, 1969. 44. Moore, J.A., Noiva, R. and Wells, I.C Selenium concentrations in plasma of patients with arteriographically defined coronary athero-sclerosis. C l in. Chem. 30: 1171-1173,1984. 45. Glavind, J . Hartmann, S., Clemmeson, J . , Hessen, K.E., and Dam, H. Studies on the role of 1 ipoperoxides in human pathology. II. The presence of peroxidized l ipids in the atherosclerotic aorta. Acta Pathol. Microbiol. Scand. 30: 1-6, 1952. 169 46. Fbgelman, A.M., Schechter, I., Seager, J . , Holom, M., Child, J.S., and Edwards, P.A. Malondialdehyde alteration of low density lipoproteins leads to cholesteryl ester accumulation in human monocyte-macro-phages. Proc. Natl. Acad. Sci. (USA) 77: 2214-2218, 1980. 47. Ledwozyw, A., Michalak, J . , Stepien, A., and Kadziolka, A. The relationship between plasma triglycerides, cholesterol, total l ip id and l ip id peroxidation products during human atherosclerosis. Cl in.  Chim. Acta 155: 275-284, 1986. 48. Jeny, C , ,Goy, J . , Fechner, J . , Loeper, J . and Emerit, J . Study of l ip id peroxidation in human and experimental atheroma by measurements of plasma malondialdehyde. Presse Med. 15: 1131-1133, 1986. 49. Goto, Y. Lipid peroxides as a cause of vascular disease. In: Lipid  Peroxides in Biology and Medicine. Yagi, K. (ed.), Academic Press, New York, pp. 295-303, 1982. 50. Sato, Y., Hotta, N., Sakamoto, N., Matsuoka, S., Omishi, N. and Yagi, K. Lipid peroxide level in plasma of diabetic patients. Biochem.  Med. 21: 104-107, 1979. 51. Uzel, N., Sivas, A., Uysal, M. and Oz, H. Erythrocyte l ip id peroxida-tion and glutathione peroxidase act iv i t ies in patients with diabetes mellitus. Horm. Metab. Res: 19: 89-90, 1987. 52. Nishigaki, I., Hagihara, M., Tsunekawa, H., Maseki, M. and Yagi, K. Lipid peroxide levels of serum lipoprotein fractions of diabetic patients. Biochem. Med. 25: 373-378, 1981. 53. Kessler, I. Mortality experience of diabetic patients. Amer. J.. Med. 51: 715-724, 1971. 54. Kannell, J.W. and McGee, D.L. Diabetes and cardiovascular risk factors: The Framingham study. Circ. 59: 8-13, 1979. 55. Hammond, B. and Hess, M.L. The oxygen free radical system: potential mediator of myocardial injury. J . Am. Coll. Cardiol. 6: 215-220, 1985. 56. Meerson, F.Z., Kagan, V.E., Kazlov, Yu. P., Belinka, L.M. and Arktripenko, Yu.V. The role of l ip id peroxidation in the pathogenesis of ischemic damage and the antioxidant protection of the heart. Basic  Res. Cardiol. 77: 465-485, 1982. 57. McCord, J.M. and Roy, R.S. The pathophysiology of superoxide: role in inflammation' and ischemia. Can. J . Physiol. Pharmacol. 60: 1346-1352, 1982. 58. Guarnieri, C , Flamigni, F. and Caldarera, CM. Role of oxygen in the ce l lu lar damage induced by reoxygenation of hypoxic heart. J . Mol.  Ce l l . Cardiol. 12: 797-808, 1980. 170 59. Aznar, J . , Santos, M.T., Valles, J . and Sala, J . Serum malondial-dehyde-1ike material (MDA-LM) in acute myocardial infarction. J . Cl in. Path. 36: 712-715, 1983. 60. Ross, W.M., Creighton, M.O., Stewart-Dehaan, P.J., Sanwal, M., Hirst, M. and Trevithick, J.R. Modelling cortical cataractogenesis: III. In vivo effects of vitamin E on cataractogenesis in diabetic rats. Can. J . Ophthalmol. 17: 61-66, 1982. 61. Ross, W.M., Creighton, M.O., Trevithick, J.R., Stewart-Dehaan, P.J. and Sanwal, M. Modelling cortical cataractogenesis: VI. Induction by glucose in vitro or in diabetic rats: prevention and reversal by glutathione. Exp. Eye Res. 37: 559-573, 1983. 62. Crouch, R., Kimsey, G., Priest, D.G., Sarda, A. and Buse, M.G. Effect of streptozotocin on erythrocyte and retinal superoxide dismutase. Diabetologia 15: 53-57, 1978. 6 3. Grankvist, K., Marklund, S.L. and Taljedal, I.B. CuZn-superoxide dismutase, Mn-superoxide dismutase, catalase and glutathione peroxi-dase in pancreatic is lets and other tissues in the mouse. Biochem. J . 199: 393-398, 1981. 64. Malaisse, W.J., Maiaisse-Lagae, F., Sener, A. and Pipeleers, D.G. Determinants of the selective toxicity of alloxan to the pancreatic 8-cel l . Proc. Natl. Acad. Sci. USA 79: 927-930, 1982. 65. Matkovics, B. Effects of plant and animal tissue lesions on super-oxide dismutase act iv i t ies . In: Superoxide and Superoxide  Dismutases. Michelson, A.M., McCord, J.MT, ancl Fridovich, TT, eds., Academic Press, New York (1977),pp. 501-515. 66. Matkovics, B., Varga, S. I., Szabo, L. and Witas, H. The effect of diabetes on the activity of the peroxide metabolism enzymes. Hor.  Metab. Res. 14: 77-79, 1982. 67. Fischer, L.J. Drugs and chemicals that produce diabetes. TIPS 6: 72-75, 1985. 68. Asayama, K., Kooy, N.W. and Burr, I.M. Effect of vitamin E deficiency and selenium deficiency on insulin secretory reserve and free radical scavenging systems in i s le t s : decrease of i s let manganosuperoxide dismutase. J . Lab. Clin. Med. 107: 459-46 4, 1986. 69. Malaisse,. W. Alloxan toxic ity to the pancreatic e-cel ls. A new hypothesis. Biochem. Pharmacol. 31: 3527-3534, 1982. 70. Taljedal, I.-B. On insulin secretion. Diabetologia 21: 1-17, 1981. 71. Grankvist, K.,. Holmgren, A., Luthman, M. and Taljedal, I.-B. Thiore-doxin reductase in pancreatic is lets may participate in diabetogenic free radical production. Biochem. Biophys. Res. Commun. 107: 1412-1418, 1982. r 171 72. Ammon, H.P.T. and Mark, M. Thiols and pancreatic B-cell function: a review. Cell Biochem. Function 3: 157-171, 1985. 73. Di l lard, C .J . , Kunert, K.J. and Tappel, A.L. Effect of vitamin E, ascorbic acid and mannitol on alloxan-induced l ip id peroxidation in rats. Arch Biochem. Biophys. 216: 204-212, 1982. 74. Slonim, A.E., Surber, M.L., Page, D.L., Sharp, R. and Burr, I.M. Modification of chemical-induced diabetes in rats by vitamin E. Supplementation minimizes and depletion enhances development of diabetes. J . Cl in. Invest. 71: 1282-1288,1983. 75. May, J.M. and De Haen, C. Insulin-stimulated intracel lular hydrogen peroxide production in rat epididymal fat c e l l . J . Biol . Chem. 254: 2214-2220, 1979. 76. Kataoka, S., Fujiya, H., Toyota, T., Suzuki, R., Itoh, K. and Kumagai, K. Immunologic aspects of the nonobese diabetic (NOD) mouse: abnormalities of cel lular immunity. Diabetes 32: 247-253, 1983. 77. Like, A.A., Butler, L., Williams, R.M., Appel, M.C, Weringer, E.J. and Rossini, A.A. Spontaneous autoimmune diabetes mellitus in the BB rat. Diabetes 31(Suppl. 1): 7-13, 1982. 78. Eisenbarth, G.S. Type I diabetes mellitus. A chronic autoimmune disease. New Engl. J . Med. 314: 1360-1368, 1986. 79. Paik, S.-G., Fleischer, N. and Shin, S. Insulin-dependent diabetes mellitus induced by subdiabetogenic doses of streptozotocin: obligatory role of cell-mediated autoimmune processes. Proc. Natl.  Acad. Sci. USA 77: 6129-6133,1980. 80. Like, A.A. and Rossini, A.A. Streptozotocin-induced pancreatic insu l i t i s : new model of diabetes mellitus. Science 193: 415-417, 1976. 81. Mordes, J.P. and Rossini, A.A. Animal models of diabetes. Am. J .  Med. 70: 353-360, 1981. 82. Loven, D.P and Oberley, L.W. Free radicals, insulin action, and diabetes. In: Superoxide Dismutase. Vol. III. Disease States. Oberley, L.W. (ed.), CRC Press, Boca Raton, pp. 152-189, 1985. 83. Gepts, W. Pathologic anatomy of the pancreas in juvenile diabetes mellitus. Diabetes 14: 619-633, 1965. 84. Yoon, J.W., Austin, M., Onodera, T. and Notkins, A.L. Virus-induced diabetes mellitus. Isolation of a virus from the pancreas of a child with diabetic ketoacidosis. New Engl. J . Med. 276: 1173-1179, 1979. 85. Baehner, R.L. Boxer, L.A. and Ingraham, L.M. Reduced oxygen by-products and white blood ce l l s . In: Free Radicals in Biology. Vol. V. Pyror, W.A. (ed.), Academic Press, New York, pp. 91-113, 1982. 172 86. Freeman, B.A. and Crapo, J.D. Biology of disease. Free radicals and tissue injury. Lab. Invest. 47: 412-426, 1982. 87. Metz, S.A. Altered arachidonic acid synthesis and l ip id peroxidation in diabetes mellitus: possible roles in leukocyte dysfunction and other ce l lu lar defects. Med. Hypotheses 12: 341-357, 1983. 88. Marklund, S.L. and Hagglof, B. Plasma EC-superoxide dismutase act iv ity in insulin-dependent diabetic children. Clin. Chim. Acta. 142: 299-305, 1984. 89. Hagglof, B., Marklund, S.L. and Holmgren, G. CuZn-superoxide dismu-tase, Mn-superoxide dismutase, catalase, glutathione peroxidase in lymphocytes and erythrocytes in insulin-dependent diabetic children. 'Acta Endocrinol. 102: 235-239, 1983. 90. Long, W.K. and Carson, P.E. Increased erythrocyte glutathione reductase act iv i ty in diabetes mellitus. Biochem. Biophys. Res.  Commun. 5: 394-399, 1961. : 91. Stock, R. and Currence, J.D. The uric acid and glutathione content of the blood in diabetes mellitus. J . Cl in. Endocrinol. 10: 313-317, 1950. 92. II l ing, K.B., Gray, C.H. and Lawrence, R.D. Blood glutathione and non-glucose reducing substances in diabetes. Biochem. J . 48: 637-640, 1951. " 93. Nath, N., Chari, S.N. and Rathi, A.B. Superoxide dismutase in diabetic polymorphonuclear leukocytes. Diabetes 33: 586-589, 1984. 94. Shah, S.V., Wallin, J.D. and Eilen, S.D. Chemi1uminescence and super-oxide anion production by leukocytes from diabetic patients. J . Clin.  Endocrinol. Metab. 57: 402-409, 1983. 95. Krinsky, N.I., Sladdin, D.G. and Levine, H. Effects of oxygen radicals on platelet functions. In: Oxygen and Oxy-radicals in  Chemistry and Biology, Rogers, M.J. and Powers, E.C. (eds.), Academic Press, New York, pp. 153-160, 1981. 96. Weiss, S.J., Young, J . , Lo Buglio, A.F., Slivka, A. and Nimeh, N.F. Role of hydrogen peroxide in neutrophi1-mediated destruction of the cultured endothelial ce l l s . J . C l in. Invest. 68: 714-721, 1981. 97. Rowe, G.T., Eaton, L.R. and Hess, M.L. Neutrophi1-derived, oxygen free radical-mediated cardiovascular dysfunction. J . Mol. Cel 1.  Cardiol. 16: 1075-1079, 1984. 98. Romsen, J .L . , Hooke, B.G., R.igot, V.H., Schurk, M.A., Swanson, P.P. and Lucchesi, B.R. The effect of ibuprofen on accumulation of Indium-ill-labeled platelets and leukocytes in experimental myocardial infarction. Circ. 66: .1002-1011, 1982. 173 99. Romsen, J .L . , Hooke, B.G., Kunkel, S.L., Abrams, G.D., Schurk, M.A. and Lucchesi, B.R. Reduction of the extent of ischemic myocardial injury by neutrophil depletion in the dog. Circ. 67: 1016-1023, 1983. 100. Halushka, P.V., Rogers, R.C, Loadholt, C B . and Colwell, J.A. Increased platelet thromboxane synthesis in diabetes mellitus. J .  Lab. C l in. Med. 97: 87-96,1981. 101. Colwell, J.A., Halushka, P.V., Sarj i , K.; Levine, J . , Sagel, J . and Nair, R.M.G. Altered platelet function in diabetes mellitus. Diabetes 25: 826-831, 1976. 102. Helmler, M.E. and Lands, W.E.M. Evidence for a peroxide-initiated free radical mechanism of prostaglandin biosynthesis. J . Biol. Chem. 255: 6253-6261, 1980. 103. Thomas, G., Skrinska, V., Lucas, F.V. and Schumacher, O.P. Platelet glutathione and thromboxane synthesis in diabetes. Di abetes 34: 951-954, 1985. 104. Butkus, A., Shirey, E.K. and Schumacher, D.P. Thromboxane biosynthe-sis in platelets of diabetic and coronary artery diseased patients. Artery 11: 238-251, 1982. 105. Ylikorkala, 0., Kaila, J . and Viinikka, L. Prostacyclin and throm-boxane in diabetes. Brit. Med. J . 283: 1148-1150, 1981. 106. Colwell, J.A., Bingham, S.F., Abraira, C , Anderson, J.W., Comstock, J.P., Kwaan, H.C, Nuttall, F. and the cooperative study group. Veterans Administration cooperative study on antiplatelet agents in diabetic patients after amputation for gangrene: II. Effects of aspirin and dipyridamole on atherosclerotic vascular disease rates. Diab. Care 9: 140-148,1986. 107. Dulin, W.E. and Soret, M.G. Chemically and hormonally induced diabetes. In: The Diabetic Pancreas. Volk, B.W. and Wellman, K.F., (eds.), Plenum Press, New York , pp. 425-465, 1977. 108. Dunn, J.S., Sheehan, H.L. and McLetchi, N.G.B. Necrosis of is lets of Langerhans produced experimentally. Lancet I: 484-487, 1943. 109. Rakieten, N., Rakieten, M.L. and Nadkarni, M.V. Studies on the diabetogenic action of streptozotocin. Cancer Chemother. Rep. 29: 91-98, 1963. . 110. Cohen, G. Oxy-radical production in alloxan-induced diabetes: an example of an in vivo metal-catalysed Haber-Weiss reaction. In: Free  Radicals in Molecular Biology, Aging, and Disease. Armstrong, D. et a l (eds.), Raven Press, New York, pp. 307-316, 1984. 111. Heikkila, R.E. and Cabbat, F.S. The prevention of alloxan-induced diabetes in mice by the iron-chelator detapac: suggestion of a role for iron in the cytotoxic process. Experientia 38: 378-379, 1982. 174 112. Fischer, L.J. and Hamburger, S.A. Inhibition of alloxan action in isolated pancreatic is lets by superoxide dismutase, catalase and a metal chelator. Diabetes 29: 213-216, 1980. 113. Grankvist, K., Marklund, S., Sehlin, J . arid Taljedal, I.B. Superoxide dismutase, catalase and scavengers of hydroxyl radicals protect against the toxic action of alloxan on pancreatic i s let cel ls in v i tro. Biochem. J . 182: 17-25, 1979. 114. Cowden, W.B., Lewis-Heghes, P.H. and Clark, I.A. Protection against alloxan-induced diabetes in mice by the free radical scavenger buty-lated hydroxyanisole. Biochem. Pharmacol. 34: 3601-3603, 1985. 115. Thaete, L.G., Crouch, R.K., Buse, M.G. and Spicer, S.S. The protec-tive role of copper-zinc superoxide dismutase against alloxan-induced diabetes: morphological aspects. Diabetologia 28: 677-682, 1985. 116. Robbins, M.J., Sharp, R.A., Slonim, A.E. and Burr, I.M. Protection against streptzotocin-induced diabetes by superoxide dismutase. Diabetologia ,18: 55-58, 1980. 117. Sandler, S., Welsh, M. and Andersson, A. Streptozotocin-induced impairment of i s let e-cell metabolism and its prevention by a hydroxyl radical scavenger and inhibitors of poly (ADP-ribose) synthetase. Acta. Pharmacol. Toxicol. 53: 392-400, 1983. 118. Be l l , R.H., Jr. and Hye, R.J. Animal model of diabetes mellitus: physiology and pathology. J . Surg. Res. 35: 433-460, 1983. 119. Rerup, C C . Drugs producing diabetes through damage of the insulin secreting ce l l s . Pharmacol. Rev. 22: 485-518, 1970. 120. Cooperstein, S.J. and Watkins, D. Effects of sulfhydryl-binding reagents on i s let tissue permeability: protection and reversal by D-glucose and phlorizin. J . Pharmacol. Exp. Ther. 204: 230-239, 1978. 121. Slonim, A.E., Fletcher, T., Burke, T. and Burr, I.M. Effect of streptozotocin on red-blood-cel1-reduced glutathione: modification by glucose, nicotinamide and epinephrine. Diabetes 25: 216-222, 1976. 122. Uchigata, Y., Yamamoto, H., Kawamura, A. and Okamoto, H. Protection by superoxide dismutase, catalase, and poly (ADP-ribose) synthetase inhibitors against alloxan- and streptozotocin-induced i s le t DNA strand breaks and against the inhibition of proinsulin synthesis. J .  B io l . Chem. 257: 6084-6088,1982. 123. Schraufstatter, I.U., Hinshaw, D.B., Hyslop, P.A., Spragg, R.G. and Cochrane, C.G. Oxidant injury of ce l l s : DNA strand breaks activate poly-adenosine diphosphate-ribose polymerase and lead to depletion of nicotinamide adenine dinucleotide. J . C l in. Invest. 77: 1312-1320, 1986. ~ 124. Fischer, L .J . , Falany, J . and Fisher, R. Characteristics of n icot in-amide and N^-methylnicotinamide protection from alloxan diabetes in mice. Toxicol. Appl. Pharmacol. 70: 148-155, 1983. 175 125. Stauffacher, W., Burr, I.M., Gutzeit, A., Beaven, D., Veleminsky, J . and Renold, A.E. Streptozotocin diabetes. Time course of irrevers-ible 3-cell damage: further observations on prevention by nicotin-amide. Proc. Soc. Exp. B io l . Med. 133: 194-200, 1970. 126. Vague, Ph., Vialettes, B., Lassmann-Vague. and Vallo, J . J . Nicotin-amide may extend remission phase in insulin-dependent diabetes. Lancet J_: 619-620, 1987. 127. Asayama, K., Nyfeler, F., English, D., P i l k i s , S.J., and Burr, I.M. Alloxan-induced free radical production in isolated cel l s : selective effect on i s let ce l l s . Diabetes 33: 1008-1011, 1984. 128. Asayama, K., English, D., Slonim, A.E., and Burr, I.M. Chemilumine-scence as an index of drug-induced free radical production in pancre-atic i s lets . Diabetes 33: 160-163, 1984. 129. Crouch, R.K., Gandy, S.E., Kimsey, G., Galbraith, R.A., Galbraith, G.M.P. and Buse, M.G. The inhibition of i s let superoxide dismutase by diabetogenic drugs. Diabetes 30: 235-241, 1981. 130. Gold, G., Manning, M., Heldt, A., Nowlain, R., Pett i t , J.R. and Grodsky, G.M. Diabetes induced with multiple subdiabetogenic doses of streptozotocin. Lack of protection by exogenous superoxide dismutase. Diabetes 30: 634-638, 1981. 131. Gandy, S.E., Buse, M.G., and Crouch, R.K. Protective role of super-oxide dismutase against diabetogenic drugs. J . C l in. Invest. 70: 650-658, 1982. ~~ 132. Papaccio, G., Pisanti, F.A. and Frascatore, S. Acetyl-homocysteine-thiolactone-induced increase of superoxide dismutase counteracts the effect of subdiabetogenic doses of streptozotocin. Diabetes 35: 470-474, 1986. ~' ~~ 133. Loven, D., Schedl, H., Wilson, H., Daabees, T.T., Stegink, L.D., Diekus, M. and Oberley, L. Effect of insulin and oral glutathione on glutathione levels and superoxide dismutase act iv i t ies in organs of rats with streptozotocin-induced diabetes. Diabetes 35: 503-507, 1986. 134. Bitar, M. and Weiner, M. Heme and hemoproteins in streptozotocin-diabetic female rats. , Biochem. Pharmacol. 32: 1921-1928, 1983. 135. Hassing, J.M., Rosenberg, H. and Stohs, S.L. Acetaminophen-induced glutathione depletion in diabetic rats. Res. Commun. Chem. Pathol.  Pharmacol. 25: 3-11, 1979. 136. Price, V.F. and Yollow, D.J. Increased resistance of diabetic rats to acetaminophen-induced hepatotoxicity. J . Pharmacol. Exp. Therap. 220: 504-513, 1982. 137. Nakhooda, A.F., Like, A.A., Chappel, C.I., Murray, F.T. and Marliss, E.B. The spontaneously diabetic Wistar rat. Diabetes 26: 100-112, 1977. 176 138. Patel, S.T., Newman, H.A., Yates, A.J.,. Thilbert, P. and Falko, J.M. Serum l ipids and lipoprotein composition in spontaneously diabetic BB Wistar rats. J . Lipid Res. 25: 1072-1083, 1984. 139. Wright Jr, J.R., Yates, A .J . , Sharma, H.M., and Thibert, P. Pathological lesions in the spontaneously diabetic BB Wistar rat: a comprehensive autopsy study. Metabolism 32 (Suppl. 1): 101-106, 1983. 140. Marliss, E.B. Workshop on the BB rat. Metabolism 32(Suppl. 1): 1-3, 1983. 141. Nakhooda, A.F., Like, A.A., Chappel, C.I., Wei, C.N. and Marliss, E.B. The spontaneously diabetic Wistar rat (the "BB" rat). Studies prior to and during development of the overt syndrome. Diabetologia 14: 199-207, 1978. 142. Behrens, W.A., Scott, F.W., Madere, R. and Trick, K.D. Increased plasma and tissue levels of vitamin E in the spontaneously diabetic BB rat. Life Sci. 35: 199-206, 1984. 143. Karpen, C.W., Cataland, S., 0'Dorisio, T.M. and Panganamala, R.V. Production of 12-hydroxy-eicosatetraenoic acid and vitamin E status in platelets from type I human diabetic subjects. Diabetes 34: 526-531, 1985. ~~ ~ 144. Karpen, C.W., Cataland, S., 0'Dorisio, T.M. and Panganamala, P.V. Interrelation of platelet vitamin E and thromboxane synthesis in type I diabetes mellitus. Diabetes 33: 239-243, 1984. 145. Sedar, A.W., Siver, M.J., .Kocsis, J . J . and Smith, J.B. Fatty acids and the i n i t i a l events of endothelial damage seen by scanning and transmission elctron microscopy. Atherosclerosis 30: 273-284, 1978. 146. Moncada, S., Gryglewski, R.J., Bunting, S. and Vane, R. A l ip id peroxide inhibits the enzyme in blood vessel microsomes that generates from prostaglandin endoperoxides the substance (prostaglandin X) which prevents platelet aggregation. Prostaglandins 12: 715-737, 1976. 147. Harris, R.E., Boxer, L.A. and Baehner, R.L. Consequences of vitamin E deficiency on the phagocytic and oxidative functions of the rat polymorphonuclear leukocyte. Blood 55: 338-343, 1980. 148. Higuchi, Y. Lipid peroxides and a-tocopherol in rat streptozotocin-induced diabetes mellitus. Acta. Med. Okayama 36: 165-175, 1982. 149. Nakakimura, H. and Mizuno, K. Studies on l ip id peroxidation in biological systems. II. Hyperlipoperoxidemia in mice induced by alloxan. Chem. Pharm. Bul l . 28: 2207-2211, 1980. 150. Nishimura, _._C. .and .Kuriyama, K. Alteration of l ip id peroxide and endogenous antioxidant content in retina of streptozotocin-induced diabetic rats: effect of vitamin A administration. Jpn. J .  Pharmacol. 37: 365-372, 1985. 177 151. Pritchard, K.A., J r . , Patel, S.T., Karpen, C.W., Newman, H.A.I, and Panganamala, R.V. Triglyceride-lowering effect of dietary vitamin E in streptozotocin-induced diabetic rats. Increased lipoprotein lipase act iv ity in l ivers of diabetic rats fed high dietary vitamin E. Diabetes 35: 278-281, 1986. 152. Capel, D. and Dorrell, H.M. Abnormal antioxidant defence in some tissues of congenitally obese mice. Biochem. J . 219: 41-49, 1984. 153. Kosower, N.S. and Kosower, E.M. The glutathione status of ce l l s . Int. Rev. Cytol. 54: 109-160, 1978. 154. Erin, A.N., Spirin, M.M., Tabidze, L.V. and Kagan, V.E. Formation of a-tocopherol complexes with fatty acids. A hypothetical mechanism of stabi l ization of biomembranes by vitamin E. Biochim. Biophys. Acta.  7 74 : 96-102, 1984. 155. Nishikimi,• M., Yamada, H. and Yagi, K. Oxidation by superoxide of tocopherols dispersed in aqueous media with deoxycholate. Biochem.  Biophys. Acta. 627: 101-108, 1980. 156. Vatassery, G.T., Morley, J.E. and Kuskowski, M.A. Vitamin E in plasma and platelets of human diabetic patients and control subjects. Am. J .  Cl in. Nutr. 37: 641-644, 1983. 157. Lewis, J.S., Pian, A.K., Baer, M.T., Acosta, P.B. and Emerson, G.A. Effect of long-term ingestion of polyunsaturated fat, age, plasma cholesterol, diabetes mellitus, and supplemental tocopherol upon plasma tocopherol. Am. J . Cl in. Nutr. 26: 136-143, 197 3. 158. Chiu, D., Lubin, B., and Shohet, S.B. Peroxidative reactions in red cel l biology. In: Free Radicals in Biology. Vol. V., Pryor, W.A. ed., New York, Academic Press (1982) pp. 115-160. 159. Jain, S.K. In vivo externalization of phosphatidylserine and phosphatidylethanolamine in the membrane bilayer and hypercoagul-ab i l i ty by the l ip id peroxidation of erythrocytes in rats. J . Cl in.  Invest. 76: 281-286, 1985. : 160. Nishikimi, M. Oxidation of ascorbic acid with superoxide anion generated by the xanthine-xanthine, oxidase system. Biochem. Biophys.  Res. Commun. 63: 463-468, 1975. 161. Redpath, J.L. and Willson, R.L. Reducing compounds in radioprotection and radiosensitization: model experiments using ascorbic acid. Int.  J . Radiat. Biol. 23: 51-63, 1973. 162. Packer, J .E . , Slater, T.F. and Wilson, R.L. Direct observation of a free radical interaction between vitamin E and vitamin C. Nature  (London) 278: 737-738, 1979. 163. Yew, M.S. Effet of streptozotocin diabetes on tissue ascorbic acid and dehydroascorbic acid. Horm. Metab. Res. 15: 158, 1983. 178 164. Rose, R.C. Metabolism and transport of dehydroascorbic acid in erythrocytes of "spontaneous diabetic BB/W" Wistar rats. Metabolism 35: 619-621, 1986. 165. Som, S., Basu, S., Mukherjee, D., Deb, S., Roy Choudhury, P., Mukherjee, S., Chatterjee, S.N. and Chatterjee, I.B. Ascorbic acid metabolism in diabetes mellitus. Metabolism 30: 572-577, 1981. 166. Owens, L.B., Wright, J . and Brown, E. Vitemin C scurvy in diabetic' patients. New Engl. J . Med. 224: 319-323, 1941. 167. Verlangieri, A.J. and Sestito, J . Effect of insulin on ascorbic acid uptake by heart endothelial ce l l s : possible relationship to retinal atherosclerosis. Life Sci. 29: 5-9, 1981. 168. Ames, B.W., Cathcart, R., Schwiers, E. and Hochstein, P. Uric acid provides an antioxidant defense in human against oxidant-and radica l -caused aging and cancer: a hypothesis. Proc. Natl. Acad. Sci. USA 78 : 6 8 58-6 86 2, 19 82. 169. Berkowitz, D. Gout, hyperlipidemia and diabetes: Interrelationship. JAMA 197: 77-80, 1966. 170. Herman, J.B. and Goldbourt, U. Uric acid and diabetes. Observations in a population study. Lancet II: 240-243, 1982. 171. McCord, J.M., ' Roy, R.S. and Schaffer, S.W. Free radicals and myocardial ischemia. The role of of xanthine oxidase. Adv.  Myocardial. 5: 183-189, 1985. 172. Lammi-keefe, C.J, Swan, P.B., and Hegarthy, P.V.J. Effect of level of dietary protein and total or partial starvation on catalase and superoxide dismutase act iv ity in cardiac and skeletal muscles in young rats. J . Nutr. 114: 2235-2240, 1984. 173. Isaac, J.T. and Binkley, F. Cyclic AMP-dependent control of the rat hepatic glutathione disulfide-sulfhydryl rat io. Biochim. Biophys.  Acta 498: 29-38, 1977. 174. Bhimji, S., Godin, D.V. and McNeill, J.H. Insulin reversal of biochemical changes in hearts from diabetic rats. Am. J . Physiol. 251: 6 7 0-6 7 5, 1986. ~ ~ 175. Cross, C.E., Watanabe, T.T., Hasegawa, G.K., Goralnik, G.N., Roertgen, K.E., Kaizu, T., Reiser, K.M., Gorin, A.B. and Last, J.A. Biochemical assays in lung homogenates: artifacts caused by trapped blood after perfusion. Toxicol. Appl. Pharmacol. 48: 99-109, 1979. 176. Drabkin, D.L. and Austin, J.H. Spectrophotometric studies II. Preparation from washed blood ce l l s : n i t r i c oxide hemoglobin and sulfhemoglobin. J . Biol . Chem. 112; 51-65, 1935. 177. Aebi, H. Catalase. In: Methods of Enzymatic Analysis. Bergmeyer, H.V., ed., Verlag Chemie., Weinheim. pp 673-684, 1974. 179 178. Winterbourne, C . C , Hawkins, R.E., Brian, M., and Carrel l , R.W. The estimation of red cel l superoxide dismutase act iv i ty. J . Lab. Cl in.  Med. 85: 337-341, 1975. ; : 179. Paglia, D.E. and Valentine, N.W. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidase. J . Lab. C l in. Med. 70: 158-169, 1967. 180. Moron, M.S., Depierre, J.W., and.Mannervik, B. Levels of glutathione, glutathione reductase and glutathione S-transferase act iv i t ies in rat lung and l iver. Biochim. Biophys. Acta 582: 67-78, 1979. 181. Godin, D.V. and Schrier, S.L. Mechanisms of inactivation of erythro-cyte membrane adenosine triphosphatase by carbodiimides. Biochem. 9: 4086-4077, 1970. 182. Howland, J .L . , Osvin, D., Donatelli, M., and Theofrastous, J.P. Inhibition of erythrocyte plasma membrane NADH dehydrogenase by nucleotides and uncouplers. Biochem. Biophys. Acta 778: 400-404, 1984. 183. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. Protein measurement with the Folin phenol reagent. J . B io l . Chem. 193: 265-275, 1951. ; ~~ 184. Higgins, J.A. and Dawson, R.M.C. Asymmetry of the phospholipid bilayer of rat l iver endoplasmic reticulum. Biochim. Biophys. Acta  470: 342-356, 1977. '. 185. Godin, D.V. and Garnett, M. Perturbational effects of inorganic cations on human erythrocyte membranes. J . Memb. B io l . 28: 143-16 8, 1976. ~ — 186. Hostetler, K.Y., Zenner, B.D., and Morris, H.P. Abnormal membrane phospholipid content in subcellular fractions from the Morris 7777 hepatoma. Biochim. Biophys. Acta 441: 231-238, 1976. 187. Stocks, J . and Dormandy, T.L. The autoxidation of human red ce l l l ipids induced by hydrogen peroxide. Br i t . J . Haematol. 20: 95-111, 1971. 188. Gilbert, H.S., Stump, D.D. and Roth, E.F., Jr. A method to correct for errors caused by generation of interfering compounds during erythrocyte l ip id peroxidation. Anal. Biochem. 137: 282-285, 1984. 189. Winterhalter, K.H. Determination of glycosylated hemoglobins. In: Methods in Enzymology. Vol 76. Antonini,E., Rossi-Bernardi, L., and Chiancone, E. Ed. Academic Press. New York p 732-739, 1981. 190. Snedecor, G.W. and Cochran, W.G. Stat ist ical Methods. Iowa State University Press, Ames, p. 63, 1976. 191. Bhuyan, B.K., Kuentzel, S.L., Gray, L.G., Fraser, T . J . , Wallach, D. and Nei1j G.L. Tissue distribution of streptozotocin (NSC-85998). Cancer Chemother. Rep. 58: 157-165, 1974. 180 192. Janes, R.G. and Winnick, T. Distribution of C -labelled alloxan in the tissues of the rat and its mode of elimination. Proc. Soc. Exp.  B io l . Med. 81: 225-229, 1952. 193. Belcher, E.H. and Harris, E.B. Studies on red ce l l l i f e span in the rat. J . Physiol. 146: 217-234, 1959. 194. Xia, Y., H i l l , K.E., and Burk, R.F. Effect .of selenium deficiency on hydroperoxide-induced glutathione release from the isolated perfused rat heart. J . Nutr. 115: 733-742, 1985. 195. Lammi-Keefe, C.L., Hegarty, P.V.J., and Swan, P.B. Effect of starvation and refeeding on catalase and superoxide dismutase act iv i t ies in skeletal and cardiac muscles from 12-month-old rats. Experientia 37_: 25-27, 1981. 1%. Randle, P.J., Newsholme, E.A. and Garland, P.B. Regulation of glucose uptake by muscle. 8. Effects of fatty acids, ketone bodies and pyruvate, and of alloxan diabetes and starvation, on the uptake and metabolic fate of glucose in rat heart and diaphragm muscles. Biochem. J . 93 : 6 5 2-6 6 5, 196 4. 197. Lammi-Keefe, C.J., Swan, P.B., and Hegarty, P.V.J. Evidence for increased peroxidative act iv i ty in muscles from streptozotocin-diabetic rats. Proc. Soc. Exp. Biol . Med. 176: 27-31, 1984. 198. Osumi, T. and Hashimoto, T. Acyl-CoA oxidase of rat l iver : a new enzyme for fatty acid oxidation. Biochem. Biophys. Res. Commun. 83: 479-485, 1978. . 199. Crapo, J.D., Sjostrom, K., and Drew, R.T. Tolerance and cross tolerance using NO2 and O2 I. Toxicology and biochemistry. J . Appl. Physiol. 44: 364-369, 1978. — 200. Yam, J . , Frank, L. and Roberts, R.J. Oxygen-toxicity: comparison of lung biochemical responses in neonatal and adult rats. Pediat. Res. 12: 115-119, 1978. 201. Horie, S., I shi i , H. and Suga, T. Changes in peroxisomal fatty acid oxidation in the diabetic rat l iver. J . Biochem. 90: 1691-1696, 1981. 202. Dorsoshow, J.H., Locker, G.Y. and Myer, C E . Enzymatic defenses of the mouse heart against reactive oxygen metabolites. J . Cl in. Invest. 65: 128-135, 1980. 203. Thayer, W.S. Role of catalase in metabolism of hydrogen peroxide by the perfused rat heart. FEBS. Lett. 202: 137-140, 1986. 204. Curello, S., Ceconi, C , B igol i , C , Ferrar i , R., A lbert in i , A. and Guarnieri, C. Changes in the cardiac glutathione status after ischemia and reperfusion. Experientia 41: 42-43, 1985. 205. Andreoli, S.P., Mallett, C P . and Bergstein, J.M. Role of glutathione in protecting endothelial cel l s against hydrogen peroxide oxidant injury. J . Lab. Clin. Med. 108: 190-198, 1986. 181 206. Lazarow, A. Factors controlling the development and progression of diabetes. Physiol. Rev. 29: 48-74, 1949. 207. Lazarow, A. Protective effect of glutathione and cysteine against alloxan diabetes in the rat. Proc. Soc. Exp. B io l . N.Y. 61: 441-447, 1946. ~ " 208. MacDonald, M.K. Histochemical observations on the sulphydryl content of pancreatic is lets after administration of diabetogenic substances. Quart. J . Exp. Physiol. 44: 177-182, 1959. 209. Hodgson, E.K. and Fridovich, I. The interaction of bovine superoxide dismutase with hydrogen peroxide: inactivation of the enzyme. Biochemistry 14: 5294-5299, 1975. 210. Searle, A.J. and Wilson, R.L. Glutathione peroxidase: effect of superoxide, hydroxyl and bromine free radicals on enzyme act iv ity. J . Radiat. B io l . 37: 213-217, 1980. 211. Crane, D. and Masters, C. On the role of catalase in the oxidation of tissue fatty acids. Arch. Biochem. Biophys. 229: . 104-111, 1984. 212. Flohe, L. and Zimmermann, R. The role of GSH peroxidase in protecting the membrane of rat l iver mitochondria. Biochim. Biophys. Acta. 223: 210-213, 1970. 213. Meister, A. Selective modification of glutathione metabolism. Science 220: 472-477, 1983. 214. Chance, B., Sies, H. and Boveris, A. Hydroperoxide metabolism in mammalian organs. Physiol. Rev. 59: 527-605, 1979. 215. Lauterburg, B.H., Smith, C V . and Mitchell, J.R. Regulation of hepatic glutathione homeostasis. In: Drug Metabolism and Drug  Toxicity. Mitchell, J.R. and Horning, M.G~ (eds.), Raven Press, New Torkl pp. 321-330, 1984. 216. Sun, J.D., Ragsdale, S.S., Benson, J.M. and Henderson, R.F. Effects of the long-term depletion of reduced glutathione in mice administered L-buthionine-S, R-sulfoximine. Fundam. Appl. Toxicol. 5_: 913-919, 1985. 217. Reed, D.J. and Fariss, M.W. Glutathione depletion and suscept ibi l -i ty. Pharmacol. Rev. 36 (Suppl.): 25S-33S, 1984. 218. Evan, A.P., Mong, S.A., Connors, B.A., Aronoff, G.R. and Luft, F .C The effect of alloxan, and alloxan-induced diabetes on the kidney. Anat. Rec. 208: 33-47, 1984. 219. Vartanyan, L.S. and Gurevich, S.M. Xanthine oxidase as a source of superoxide radicals in vivo. Repression and induction of the xanthine oxidase and superoxide dismutase act iv i t ies in mouse l iver upon administration of al lopurinol. Studia Biophys. 113: 95-104, 1986. 182 220. Moorhouse, C P . , Grootveld, M., Halliwel 1,'. B., Quinlan, J.G. and Gutteridge, J.M.C. Allopurinol and oxypurinol are hydroxyl radical scavengers. FEBS Letters 213: 23-28, 1987. 221. Scott, F.W., Trick, K.D., Hynie, I., Braaten, J.T. and Nera, E.A. Time course of serum glucoregulatory variables and l ipids following neonatal thymectomy of diabetes-prone and control BB Wistar rats fed a defined diet. Ann.- Nutr. Metab. 28: 207-219, 1984. 222. Seemayer, T.A., Colle, E., Tannenbaum, G.S., Oligny, L.L., Guttmann, R.D. and Goldman, H. Spontaneous diabetes mellitus syndrome in the rat. III. Pancreatic alterations in aglycosuric and untreated BB Wistar-derived rats. Metabolism 32 (Suppl. I): 26-32, 1983. 223. Oschilewski, U., Kiessel, U., and Kolb, H. Administration of S i l i ca prevents diabetes in BB-rats. Diabetes 34: 197-199, 1985. 224. Marliss, E.B., Nakhooda, A.F., and Poussier, P. Cl inical forms and natural history of the diabetic syndrome and insulin and glucagon secretion in the BB rat. Metabolism 32(Suppl. 1): 11-16, 1983. 225. Palay, S.L. and Lazarow, A. Protection against diabetes and production of hepatic necrosis by use of cysteine and alloxan in the rat. Anat. Rec. 96: 55-75, 1946. 226. Fridovich, I. Biological effects of the superoxide radical. Arch.  Biochem. Biophys. 247: 1-11, 1986. 227. Grankvist, K., Marklund, S. and Taljedal, I.-B. Superoxide dismutase is a prophylactic against alloxan diabetes. Nature 294: 158-160, 1981. 228. Sanfey, H., Buckley, G.B. and Cameron, J.L. Pathogenesis of acute pancreatitis: role of oxygen free radicals. Surg. Forum 34: 222-224, 1983. 229. Sanfey, H., Buckley, G.B. and Cameron, J.L. The pathogenesis of acute pancreatitis. The source and role of oxygen-derived free radicals in three different experimental models. Ann. Surg. 201: 633-638, 1985. 230. Braganza, J.M., Wickens, D.G., Cawood, P. and Dormandy, T.L. L ip id -peroxidation (free-radical-oxidation) products in bi le from patients with pancreatic disease. Lancet 11: 375-378, 1983. 231. Sanfey, H., Buckley, G.B. and Cameron, J.L. The role of oxygen-derived free radicals in the pathogenesis of acute pancreatitis. Ann.  Surg. 200: 405-412, 1984. 232. Flower, R.J., Moncada, S. and Vane, J.R. Analgesic-antipyretics and anti-inflammatory agents: drugs employed in the treatment of. gout. In: Goodman and Gilman's The Pharmacologic Basis of Therapeutics. 7th. ed. Gilman, A.G., Goodman, L.S., Rai l , T.W. and Murad, F. (eds.), MacMillan Publishing Company, New York, p. 711, 1985. 183 233. Povoa, H., J r . , Sa, L.D. and Lessa, V.M. Xanthine oxidase and triglycerides in serum of patients with hyperlipoproteinemia, type IV. Biomed. Biochim. Acta 43: 1201-1203, 1984. 234. Keller, U. Diabetic ketoacidosis: current views on pathogenesis and treatment. Diabetologia 29: 71-77, 1986. 235. Robinson, A.M. and Williamson, D.H. Physiological roles of ketone bodies as substrates and signals in mammalian tissue. Physiol. Rev. 60: 143-187, 1980. 236. Holland, M.L., Rhodes, G.R., Weister, D. and Novotny, M. Chromato-graphic prof i l ing of urinary volat i le and organic acid metabolites of normal and diabetic C57BL/Ks mice. J . Chromatogr. 306: 23-37, 1984. 237. Yancey, M., Stuart, R., Weisler, D. and Novotny, M. Quantitative alterations in the metabolism of carbonyl compounds due to diet-induced l ip id peroxidation in rats. J . Chromatogr. 382: 47-56, 1986. 238. Dormandy, T.L. The autoxidation of red ce l l s . Br it. J . Haematol. 20: 457-461, 1971. 239. Cohen, G. and Hochstein, P. Glutathione peroxidase: the primary agent for the elimination of hydrogen peroxides in erythrocytes. Biochemistry 2: 1420-1428, 1963. 240. McMahon, S. and Stern, A. The interrelationship of superoxide dismutase and peroxidative enzymes in the red c e l l . Biochim. Biophys.  Acta. 566: 253-258, 1979. 241. Crane, F.L., Crane, H.E., Sun, L., Mackeller, W.C., Grebing, C. and Low, H. Insulin control of a transplasma membrane NADH dehydrogenase in erythrocyte membrane. J . Biomembr. Bioenerg. 14: 425-433, 1982. 242. Trotta, R.J., Sullivan, S.G. and Stern, A. Lipid peroxidation and hemoglobin degradation in red blood cel l s exposed to t-butyl hydro-peroxide. Biochem. J . 204: 405-415, 1982. 243. Barber, A.A. and Bernheim, F. Lipid peroxidation: its measurements, occurrence, and significance in animal tissues. Adv. Gerontol. Res. 2: 355-403, 1967. : 244. Kamada, T and Otsuji, S. Lower levels of erythrocyte membrane f lu id i t y in diabetic patients. Diabetes 32: 585-591, 1983. 245. Bagadade, J.D. and Ways, P.O. Erythrocyte membrane composition in exogenous hypertriglyceridemia. J . Lab. Cl in. Med. 75: 53-60, 1970. 246. van Doormaal, J . J . , Muskiet, F.A.J., van Ballegooie, E., Sluiter, W.J. and Doorenbos, H. The plasma and erythrocyte fatty acid composition of poorly controlled, insulin-dependent (type I) diabetic patients and the effect of improved metabolic control. Cl in. Chim. Acta 144: 203-212, 1984. \ 184 247. Bryszewska, M., Watala, C. and Torzecka, W. Changes in f l u id i t y and composition of erythrocyte membranes and in composition of plasma l ipids in Type I diabetes. Brit. J . Haematol. 62: 111-116, 1986. 248. Chandramouli, V., Carter, J.R. Cell membrane changes in chronically diabetic rats. Diabetes 24: 257-262,1975. 249. Sosin, D., Kordowiak, A. and Sarnecka-Keller, M. Osmotic resistance of rat erythrocytes in hyperinsulinism and streptozotocin diabetes as compared with cholesterol and s i a l i c acid contents of erythrocyte ghosts. Hor. Metab. Res. 16 (Suppl.): 89-91, 1984. 250. Owen, J.S., Bruckdorfer, K.R., Day, R.C. and Mclntyre, N. Decreased erythrocyte membrane f l u id i t y and altered l i p id composition in human l iver disease. J . Lipid Res. 23: 124-132, 1982. 251. Bhandara, R., Srinivasan, S.R., Radhakrisnamurthy, B. and Berenson, G.S. Effects of diabetes and high fat-high cholesterol diet on plasma l ip id levels and on erythrocyte membrane composition. Atherosclerosis 42: 263-272, 1982. 252. Shohet, S.B., Nathan D.G. and Karnovsky, M.L. Stages of the incorpor-ation of fatty acids into red blood ce l l s . J . Cl in. Invest. 47: 1096-1108, 1968. ~ 253. Lubin, B.H., Shohet, S.B. and Nathan, D.G. Changes in fatty acid metabolism after erythrocyte peroxidation: stimulation of a membrane repair process. J . C l in . Invest. 51: 338-344, 1972. 254. Hoffman, D.R., Haning, J.A., and Cornatzer, W.E. Effect of alloxan diabetes on phosphatidyl choline biosynthetic enzymes. Proc. Soc.  Exp. Biol. Med. 167: 143-146,1981. ~ 255. Faas, F.H. and Carter, W.J. Altered microsomal phospholipid composi-tion in the streptozotocin diabetic rat. Lipids 18: 339-342, 1983. 256. Kalofoutis, A. and Lekakis, J . Changes of platelet phospholipids in diabetes mellitus. Diabetologia 21: 540-543, 1981. 257. Crabbe, M.J.C., Peckar, C O . , Haider, A.B., and Cheng, H. Erythrocyte glyceraldehyde reductase levels in diabetics with retinopathy and cataract. Lancet 11: 1268-1270, 1980. 258. Borochov, H., Zahler, P. Wilbrandt, W. and Shinitzky, M. The effect of phosphatidylcholine to sphingomyelin mole ratio on the dynamic properties of sheep erythrocyte membrane. Biochim. Biophys. Acta. 470: 382-388, 1977. 259. Baba, Y., Kai, M., Kamada, T., Setoyama, S. and Otsuji, S. Higher levels of erythrocyte membrane microviscosity in diabetes. Diabetes 28: 1138-1140, 1980. " 260. Misra, D.P., Staddon, G., Powell, N., Misra, J . and Cook, D. Lecithin: cholesterol acyltransferase in diabetes mellitus and the effect of insulin in these cases. C l in. Chim. Acta 56: 83-89, 1974. 185 261. P ie t r i , A., Dunn, F.L.,. and Raskin, P. The effect of improved diabetic control on plasma l ip id and lipoprotein levels. A comparison of conventional therapy and continuous subcutaneous insulin infusion. Diabetes 29: 1001-1005, 1980. 262. Dormandy, T.L. Free-radical reactions in biological systems. Ann. R.  Col l . Surg. 80: 188-194, 197 9. 263. Sies, H. Oxidative stress: Introductory remarks. In: Oxidative  Stress. Sies, H. (ed.), Academic Press, London, pp. 1-8, 1985"! 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0097524/manifest

Comment

Related Items