Some effects of experimentally-induced diabetes on pituitary-testicular relationships in rats.

The effect of experimentally-induced diabetes mellitus on reproductive organ weights, serum and pituitary gonadotropin levels and serum testosterone levels was studied in 3-month old rats. In experiment 1, intact rats were treated with alloxan monohydrate or streptozotocin. In experiments 2 and 3, intact and castrated rats were rendered diabetic with alloxan (experiment 2) or streptozotocin (experiment 3). The duration of each experiment was 3 weeks. In each experiment diabetes resulted in body weight losses or reduced body weight gain, elevated serum glucose concentrations and reduced assessory sex gland weights (intact rats). Serum levels of testosterone were depressed (P less than 0.05 or P less than 0.01) in diabetic rats. Serum levels of LH were significantly (P less than 0.05) lower in intact diabetics than in controls when pooled data from the three experiments were compared. Serum levels of FSH were not affected by diabetes. Pituitary concentrations of FSH were elevated (P less than 0.05) in diabetics in two of the three experiments, while LH concentrations were elevated (P less than 0.05 or P less than 0.01) in diabetics in all experiments. The hypersecretion of gonadotropins in castrated rats was not affected by diabetes.     

Calcitriol regulates angiotensin-converting enzyme and angiotensin converting-enzyme 2 in diabetic kidney disease

To investigate the effects of calcitriol on angiotensin-converting enzyme (ACE) and ACE2 in diabetic nephropathy. Streptozotocin (STZ) induced diabetic rats were treated with calcitriol for 16 weeks. ACE/ACE2 and mitogen activated protein kinase (MAPK) enzymes were measured in the kidneys of diabetic rats and rat renal tubular epithelial cells exposed to high glucose. Calcitriol reduced proteinuria in diabetic rats without affecting calcium-phosphorus metabolism. ACE and ACE2 levels were significantly elevated in diabetic rats compared to those in control rats. The increase in ACE levels was greater than that of ACE2, leading to an elevated ACE/ACE2 ratio. Calcitriol reduced ACE levels and ACE/ACE2 ratio and increased ACE2 levels in diabetic rats. Similarly, high glucose up-regulated ACE expression in NRK-52E cells, which was blocked by the p38 MAPK inhibitor SB203580, but not the extracellular signal-regulated kinase (ERK) inhibitor FR180204 or the c-Jun N-terminal kinase (JNK) inhibitor SP600125. High glucose down-regulated ACE2 expression, which was blocked by FR180204, but not SB203580 or SP600125. Incubation of cells with calcitriol significantly inhibited p38 MAPK and ERK phosphorylation, but not JNK phosphorylation, and effectively attenuated ACE up-regulation and ACE2 down-regulation in high glucose conditions. The renoprotective effects of calcitriol in diabetic nephropathy were related to the regulation of tubular levels of ACE and ACE2, possibly by p38 MAPK or ERK, but not JNK pathways.     

Activity of angiotensin converting enzyme in women with hypothyroidism]

Serum activity of angiotensin converting enzyme (ACE) has been measured in 13 women (mean age 43.1 years) with primary hypothyroidism by spectrofluorometric method of Friedland and Silverstein. The mean enzyme activity was significantly lower in hypothyroid patients than in healthy persons. There is no significant correlation between ACE activity and thyroid hormones concentration.     

Insulin replacement therapy in diabetic rats using an osmotic pump normalizes expression of enzymes key to hepatic carbohydrate metabolism.

Intensively treating type I diabetics with continuous subcutaneous insulin infusions or multiple daily insulin injections to normalize mean blood glucose concentrations significantly reduces the onset of secondary diabetic complications when compared to conventionally treated diabetics. Our studies focused on characterizing hepatic enzyme expression in intensively and conventionally treated diabetic rats. Alloxan-induced diabetic rats were conventionally treated with insulin injected twice daily or intensively treated with similar daily dosages of insulin administered via a surgically implanted osmotic pump. Our results demonstrate a significant difference in hepatic enzyme expression when these treatment regimes are compared. In conventionally treated diabetic rats, phosphoenolpyruvate carboxykinase (PEPCK) protein and mRNA levels remained slightly elevated when compared to normal animals, glycogen phosphorylase (GP) protein levels were still slightly decreased, and glycogen synthase (GS) protein and mRNA levels remained at the elevated levels observed in untreated diabetics. In contrast, the protein and mRNA levels of all three enzymes were normalized in the insulin pump-treated animals. These results suggest that intensive insulin therapy improves glycemia directly by normalizing hepatic gene expression while conventional insulin therapy normalizes plasma glucose concentrations indirectly.     

Genetic Predisposition for Development of Nephropathy in Type 2 Diabetes Mellitus

The aim of the study was to explore the association of the angiotensin-converting enzyme (ACE) gene I/D polymorphism and the methylenetetrahydrofolate reductase (MTHFR) gene C677T polymorphism with development of diabetic nephropathy in type 2 diabetes mellitus. Three groups were recruited during 2007–2011: 232 normal controls, 185 type 2 diabetics without nephropathy, and 407 type 2 diabetics with nephropathy. The ACE I/D and MTHFR C677T polymorphisms were examined using PCR and PCR-RFLP methods. We found no significant association of the ACE I/D polymorphism with diabetic nephropathy in genotype, allele, dominant, and recessive models. We observed a significant association of MTHFR C677T with development of diabetic nephropathy in type 2 diabetics. The MTHFR C677T polymorphism plays a significant role in predisposition of renal insufficiency in diabetic patients.     

Serum and urine malondialdehyde levels in NIDDM patients with and without hyperlipidemia

Malondialdehyde (MDA), a marker of lipid peroxidation, was measured as thiobarbituric acid reactive substance (TBARS) in 78 noninsulin-dependent diabetic patients, 38 hyperlipidemic patients, and 28 healthy subjects. Diabetic patients were divided into groups and subgroups according to the existence of hyperlipidemia and other complications. Serum and urine MDA concentrations were significantly higher in diabetic and nondiabetic patient groups than in the control group. By contrast to urine MDA level, serum MDA level was significantly higher in hyperlipidemic diabetics than that of normolipidemic diabetics. Serum MDA levels in the hyperlipidemic diabetic group and urine MDA levels in both diabetic groups were significantly higher than those in hyperlipidemic nondiabetic group. In both diabetic groups, the existence of complications didn't affect serum and urine MDA levels. No correlation existed between serum and urine MDA levels in both patient groups and control subjects. This study confirmed the existence of lipid peroxidation disorders in diabetic patients.     

Effects of variations in thyroid hormone serum concentrations on serum ACE-activity

Serum angiotensin converting enzyme activities were significantly increased in 26 untreated hyperthyroid patients (20.3 +/- 5.4 U/ml; P less than 0.001) compared with healthy control subjects (13.1 +/- 2.3 U/ml). In 12 patients a significant fall in enzyme activities was observed after treatment compared with pretreatment serum ACE levels (P less than 0.001). Eight patients with hypothyroidism (15.7 +/- 5.1 U/ml) and 11 athyreotic patients, totally thyroidectomized for well-differentiated thyroid cancer, showed no significant differences in serum ACE activities (14.3 +/- 2.2 U/ml) compared with control subjects. After thyroid hormone supplementation a significant increase in serum ACE activity (P less than 0.05) was found in the athyreotic patients. Addition of increasing amounts of L-thyroxine to a serum sample of an athyreotic patient showed no significant effect on ACE activity in vitro. We suggest that the elevated serum ACE activity in hyperthyroidism is not from the thyroid gland, but represents a direct effect of thyroid hormone on ACE synthesis and/or release from endothelial cells.     

Serum selenium levels in diabetic children

The effect of increased selenium uptake on serum selenium in diabetic children was investigated during the first 9 yr of the Finnish nationwide selenium fertilization program, which started in 1984. Serum selenium concentrations were followed in 237 diabetic children (mean age 8.1 yr) and 214 controls from 1984 to 1992. The control group consisted of 107 siblings of the diabetics and of 107 other healthy children of corresponding age groups. Selenium was determined by direct electrothermal atomic absorption spectrophotometry. The effect of the increased uptake was seen in both diabetic and in control persons. Before the autumn of 1985, diabetic patients had significantly higher serum selenium levels than their siblings or the other healthy controls. Toward the end of year 1987, this difference had disappeared. After that, serum selenium levels continued to increase until the year 1990. In 1990 the mean selenium serum level of diabetic patients was 1.36 μmol/L and that of controls 1.33 μmol/L. The duration of diabetes did not have any effect on selenium serum levels. Slightly higher serum selenium in new diabetic patients before the start of therapy was explained by the dehydration state. The patients who were younger than 3 yr had slightly lower selenium serum levels when compared with older age groups. This difference was observed, however, only during the first 3 yr of the study. After that, when the selenium intake increased in general, no age-dependent differences were found anymore. There were no significant differences in serum selenium levels between males and females in either diabetic patients or in controls.     

Lack of relationship between Alu repetitive elements in angiotensin converting enzyme and the severity of diabetic retinopathy

BackgroundAngiotensin-converting enzyme (ACE) stimulates angiogenesis that leads to the development of diabetic retinopathy (DR). Alu repetitive elements in ACE gene increase the expression of this enzyme. We investigated the frequency of Alu repetitive elements, insertion/deletion (I/D) polymorphism, in angiotensin-converting enzyme among diabetic retinopathy patients and whether this polymorphism is associated with the severity of retinopathy in Jordanians with type 2 diabetes.MethodsA total of 277 subjects participated in this case/ control study (100 diabetic patients without DR, 82 diabetic patients with DR, and 95 healthy control). Blood samples were withdrawn, followed by DNA extraction. Alu repetitive elements were examined by polymerase chain reaction followed by gel electrophoresis.ResultsThe genotype and allele frequencies among diabetic patients, were close to healthy controls (genotypes, II 44.4 vs. 44.7%, ID 44.4 vs. 42.6%, DD 12.2 vs. 12.8%, P = 0.402 and 0.677 respectively, alleles, I 65.6 vs. 66%, D 34.4 vs. 34%, P=0.863). Complicated diabetics with retinopathy showed similar genotype and allele frequency to those without complications. The severity of diabetic retinopathy in affected individuals was not correlated with I/D polymorphism (P=0.862).ConclusionsWe conclude that the presence of Alu repetitive elements did not increase the development or progression risk to retinopathy in Jordanian type 2 diabetic patients. No association between I or D alleles with the severity of DR was detected.     

Low phospholipid arachidonic acid values in diabetic platelets.

Platelet aggregation is enhanced in diabetes mellitus, and platelets may be implicated in the pathogenesis of diabetic angiopathy. Increased platelet aggregation is probably mediated by the production of the proaggregatory prostaglandin thromboxane, which is synthesised from arachidonic acid (C20:4) by the action of the platelet enzymes cyclo-oxygenase and thromboxane synthetase. The fatty acid composition of platelet phospholipid was measured in 20 normal controls, 10 insulin-treated diabetics with no or minimal retinopathy, and 10 insulin-treated diabetics with proliferative retinopathy. The percentage of arachidonic acid was significantly higher in controls (mean 22.6%) than in the diabetics with no or minimal retinopathy (mean 18.5%; p less than 0.025) and the diabetics with proliferative retinopathy (mean 14.6%; p less than 0.001). The percentage of linoleic acid was lower in controls (mean 8.9%) than in the diabetics with no or minimal retinopathy (mean 12.6%; p less than 0.01) and diabetics with proliferative retinopathy (mean 13.1%; p less than 0.001). The mean percentage of linolenic acid was significantly lower in the diabetics with proliferative retinopathy (2.7%) than in the normal control group (4.4%; p less than 0.01). A significant negative correlation was found between the percentages of arachidonic acid and glycosylated haemoglobin (Rs = -0.58; p less than 0.001). A significant positive correlation was found between linoleic acid and the percentage of glycosylated haemoglobin (Rs = 0.51; p less than 0.01). The reciprocal correlation between percentages of arachidonic acid and glycosylated haemoglobin suggests that diabetic control may influence thromboxane release and platelet activity directly and that low percentages of arachidonic acid reflect the increased degree of in-vivo activation.     

Possible role of VEGF in the progression of kidney disease in streptozotocin (STZ)-induced diabetic rats: effects of an ACE inhibitor and an angiotensin II receptor antagonist.

Two endothelium-derived factors, endothelin (ET), a vasoconstrictor, and vascular endothelial growth factor (VEGF), an angiogenic factor are thought to be involved in the pathogenesis of diabetic vascular complications. The aim of this study was to determine the effects of an angiotensin II type I (AT-1) receptor antagonist and an ACE inhibitor on the pathogenesis of VEGF and ET-1-mediated kidney disease in STZ-induced diabetic rats. Two days after STZ administration, diabetic rats were treated for 8 weeks with enalapril maleate, an ACE inhibitor, candesartan cilexetil, an AT-1 receptor antagonist, or saline. Urinary albumin and N-acetyl beta-D glucosaminidase (NAG) excretion as well as the VEGF protein content in the kidney were all found to be elevated in diabetic rats. Administration of enalapril maleate or candesartan cilexetil decreased the level of microalbuminuria and NAG excretion in diabetic rats. Administration of enalapril maleate also suppressed the elevated renal VEGF protein content in these animals while candesartan cilexetil treatment had no effect. Serum ET-1 and VEGF levels were unchanged by these treatments. These data support a role for AT-1 receptor antagonists and ACE inhibitors in the prevention of diabetic nephropathy, and suggest that the former may work by reducing renal VEGF levels.     

Angiotensin-converting enzyme activity decreases during fasting

Serum angiotensin-converting enzyme (ACE) activity varies directly with thyroid hormone levels in states of altered thyroid function. Because T3 levels decrease during fasting, ACE activity was examined to ascertain if it was reduced in this low T3 condition. Eighteen obese euthyroid subjects were hospitalized and placed on a weight-maintaining diet for 4 days. Nine subjects (Group 1) underwent a fast (50 kcal/day) for 8 days. Nine (Group 2) subjects received T3 (5 micrograms q 3 h) during an identical fast. Weight loss was identical in both groups (-4.4 kg). Serum T3 fell in Group 1 from 104 +/- 8 to 50 +/- 4 ng/d/(p less than .05) but was unchanged in Group 2 (114 +/- 11 ng/dl fed vs. 120 +/- 14 ng/dl fasted). Blood pressures fell significantly in Group 1 (mean systolic: 112----104 mmHg; diastolic: 71----65 mmHg, p less than 0.05), but not in Group 2 subjects. ACE activity fell progressively in Group 1 subjects during fasting (14.4 +/- 1.6 U/ml fed vs. 12.8 +/- 1.4 U/ml fasted p less than 0.05). ACE activity was not decreased significantly early in the fast in patients given T3, but by late fast (days 6-8) was reduced to the same degree as in Group 1 subjects. Glucose and insulin levels fell similarly in both groups. Conclusions: (1) ACE activity is reduced during starvation. This effect is not mediated by T3. (2) Blood pressure reduction during fasting may result from the low T3 levels, but not from decreased ACE activity. Interpretation of serum ACE activity must be viewed in the context of a patient's diet.     

Diurnal plasma free fatty acid profiles in normal and diabetic pregnancies.

Diurnal plasma free fatty acid (FFA) and glucose concentrations were measured in 23 normal women, 13 chemical diabetics, and 11 insulin-dependent diabetics in late pregnancy. The mean diurnal FFA value was lowest in the insulin-requiring diabetics and highest in the chemical diabetics. Although no association was found between percentile birthweight and maternal mean fasting or diurnal FFA, there was a positive correlation between percentile birthweight and both mean maternal fasting plasma glucose and mean diurnal glucose in normal and chemical diabetic women. These observations support the view that glucose is the major substrate used by the fetus, but birthweight may be influenced more by the total substrate crossing the placenta than by maternal levels of either glucose or FFA alone.     

Increased urinary angiotensin-converting enzyme 2 in renal transplant patients with diabetes

Angiotensin-converting enzyme 2 (ACE2) is expressed in the kidney and may be a renoprotective enzyme, since it converts angiotensin (Ang) II to Ang-(1-7). ACE2 has been detected in urine from patients with chronic kidney disease. We measured urinary ACE2 activity and protein levels in renal transplant patients (age 54 yrs, 65% male, 38% diabetes, n?=?100) and healthy controls (age 45 yrs, 26% male, n?=?50), and determined factors associated with elevated urinary ACE2 in the patients. Urine from transplant subjects was also assayed for ACE mRNA and protein. No subjects were taking inhibitors of the renin-angiotensin system. Urinary ACE2 levels were significantly higher in transplant patients compared to controls (p?=?0.003 for ACE2 activity, and p≤0.001 for ACE2 protein by ELISA or western analysis). Transplant patients with diabetes mellitus had significantly increased urinary ACE2 activity and protein levels compared to non-diabetics (p<0.001), while ACE2 mRNA levels did not differ. Urinary ACE activity and protein were significantly increased in diabetic transplant subjects, while ACE mRNA levels did not differ from non-diabetic subjects. After adjusting for confounding variables, diabetes was significantly associated with urinary ACE2 activity (p?=?0.003) and protein levels (p<0.001), while female gender was associated with urinary mRNA levels for both ACE2 and ACE. These data indicate that urinary ACE2 is increased in renal transplant recipients with diabetes, possibly due to increased shedding from tubular cells. Urinary ACE2 could be a marker of renal renin-angiotensin system activation in these patients.     

Plasma fibrinogen levels in type II diabetics

Plasma fibrinogen levels measured by an immunoassay method on 170 type II diabetic patients exhibited a bimodal distribution with one small population demonstrating levels greater than those of the normal reference range. The mean plasma level of fibrinogen in the type II diabetics was higher than that of the normal population. Spearman's correlations demonstrated statistically significant positive relationships in type II diabetic patients between fibrinogen levels and fasting glucose levels, serum cholesterol, glycosylated hemoglobin and urinary albumin excretion rate. These relationships suggest that increased plasma fibrinogen may be another marker for coronary heart disease complications encountered by diabetics.     

Differential autocrine modulation of atrial and ventricular potassium currents and of oxidative stress in diabetic rats

The autocrine modulation of cardiac K(+) currents was compared in ventricular and atrial cells (V and A cells, respectively) from Type 1 diabetic rats. K(+) currents were measured by using whole cell voltage clamp. ANG II was measured by ELISA and immunofluorescent labeling. Oxidative stress was assessed by immunofluorescent labeling with dihydroethidium, a measure of superoxide ions. In V cells, K(+) currents are attenuated after activation of the renin-angiotensin system (RAS) and the resulting ANG II-mediated oxidative stress. In striking contrast, these currents are not attenuated in A cells. Inhibition of the angiotensin-converting enzyme (ACE) also has no effect, in contrast to current augmentation in V cells. ANG II levels are enhanced in V, but not in A, cells. However, the high basal ANG II levels in A cells suggest that in these cells, ANG II-mediated pathways are suppressed, rather than ANG II formation. Concordantly, superoxide ion levels are lower in diabetic A than in V cells. Several findings indicate that high atrial natriuretic peptide (ANP) levels in A cells inhibit RAS activation. In male diabetic V cells, in vitro ANP (300 nM-1 muM, >5 h) decreases oxidative stress and augments K(+) currents, but not when excess ANG II is present. ANP has no effect on ventricular K(+) currents when the RAS is not activated, as in control males, in diabetic males treated with ACE inhibitor and in diabetic females. In conclusion, the modulation of K(+) currents and oxidative stress is significantly different in A and V cells in diabetic rat hearts. The evidence suggests that this is largely due to inhibition of RAS activation and/or action by ANP in A cells. These results may underlie chamber-specific arrhythmogenic mechanisms.     

Plasma endothelin-1 levels and albumin excretion rate in normotensive,microalbuminuric type 2 diabetic patients

BACKGROUND: Endothelin-1 (ET-1) is able to determine functional and structural renal alterations and plasma levels of this vasoconstrictor peptide are increased in diabetic patients. In a selected group of type 2 normotensive diabetic patients with microalbuminuria, we investigated circulating ET-1 levels compared to a control group and verified whether there is a relationship between ET-1 levels and albumin excretion rate in diabetics. SUBJECTS AND METHODS: Thirty-two microalbuminuric type 2 diabetic patients (12 males and 20 females; mean age 57 +/- 8 years) without hypertension, renal failure, hypercholesterolemia or atherosclerotic damage were selected. The control group was made up of 28 healthy subjects matched for sex and age. Blood pressure, creatinine clearance, serum cholesterol and plasma ET-1 values were determined in diabetic and control group. In diabetic patients, glycosilated hemoglobin and urinary albumin excretion rate were also assayed. Mean ET-1 values in diabetics and controls were compared using Student's t-test. Linear regression test was done to relate two variables. Statistical significance was set at p<0.05. RESULTS: Mean ET-1 values were significantly higher in the diabetic group than in controls (11.77 +/- 1.16 pg/ml vs 8.9 +/- 2.1 pg/ml; p<0.05). No relationship (p>0.05) was found between circulating ET-1 and blood pressure, creatinine clearance, serum cholesterol and metabolic control in diabetics. There was a significant positive correlation (r=0.403; p=0.03) between plasma ET-1 levels and albumin excretion rate in diabetic patients. CONCLUSIONS: Our results showed that circulating ET-1 values were increased in microalbuminuric, normotensive, type 2 diabetic patients and correlated with albumin excretion rate. These findings confirm that endothelial dysfunction, as expressed by ET-1 levels, occurs early in these patients and support the hypothesis of a potential role for this peptide in development of microalbuminuria in diabetic nephropathy.     

Multiple defects in the renin-angiotensin system in alloxan-diabetic kidney

A defect in the renin-angiotensin system has been shown in diabetic patients and experimental animals, in particular with nephropathy or autonomic neuropathy. The mechanism for this low plasma renin activity (PRA) is poorly understood. In order to clarify this defect, the renin-angiotensin system was studied in alloxan-induced diabetic and age-match control mice. In diabetic animals, kidney renin activity (KRA) was significantly lower than that of the controls, while plasma renin substrate (PRS) concentration was slightly higher and PRA was normal. The amount of injected radiolabeled renin extracted by the kidney was normal, but the amount extracted by the liver was significantly decreased in diabetic animals. On the other hand, the degradation of the extracted renin by both the kidney and the liver was elevated as compared to the controls. This high degradation rate was accompanied by a slight increase in lysosomal protease activity in the kidneys. In in vivo studies, isoproterenol-induced PRA was 20-fold in control animals. In diabetics, isoproterenol-induced PRA was attenuated and rose only four- to fivefold over basal level. The angiotensin converting enzyme (ACE) activity in the kidney was significantly decreased in the diabetic state. It is concluded that there were multiple defects in the renin-angiotensin system in this diabetic model, namely, a depletion of renin storage with subsequent loss of maximal responsiveness to the adrenergic agonist in renin release, an elevation of intrarenal renin degradation together with a deficiency in ACE which would possibly lead to a decrease in intrarenal formation of angiotensin II.     

Reduction of baclofen-, but not sodium valproate-induced growth hormone release in type I diabetic men.

The effects of sodium valproate (a drug enhancing endogenous gamma aminobutyric acid (GABA)-ergic activity) and of the GABA analog baclofen (a GABA B receptor agonist) on serum GH levels was tested in 8 type I diabetic men and 8 normal controls. Sodium valproate (800 mg) or baclofen (10 mg) were given by mouth at 08.30 h on the experimental day. Control tests with a placebo were performed on different occasions. Basal GH levels were similar in controls and diabetic patients. Sodium valproate induced a 7 fold increase in serum GH concentrations in both groups. In contrast, baclofen-induced GH rise was significantly higher in normal controls (mean peak was 3.4 times higher than baseline) than in diabetic patients (mean peak was only 2.1 times higher than basal value). Serum GH levels did not change after placebo administration in any groups. These data suggest the presence of diabetes-induced alterations of a GABAergic pathway mediated by B receptors in the control of GH secretion. Alternatively, the data might indicate a change in diabetic men of other baclofen-sensitive neurotransmissions, different from GABA.     

Serum levels of type III procollagen peptide in diabetes mellitus

Serum levels of type III procollagen peptide (P-III-P) were investigated in 19 patients with type 1 (insulin-dependent) and in 48 (25 orally treated, 23 insulinized) patients with type 2 (non insulin-dependent) diabetes mellitus. Among patients with type 2 diabetes, 16 orally treated and 14 insulin-treated subjects had macrovascular complications. P-III-P levels were not correlated with the duration of diabetes and with glucose control, nor were there any significant sex and age differences in the levels. P-III-P values were significantly higher in the sera of insulin-treated non insulin-dependent diabetic patients with macroangiopathy. These high values (18.5 +/- 10.8 ng/ml) were in contrast with normal values in healthy subjects (8.5 +/- 2.5, P less than 0.001), insulin-dependent diabetics (9.9 +/- 3.4 ng/ml, P less than 0.01), non insulin-dependent diabetics treated with oral agents (8.2 +/- 2.6 ng/ml, P less than 0.001) and insulin-treated non insulin-dependent patients without macroangiopathy (8.2 +/- 4.9 ng/ml, P less than 0.001). Although this study does not demonstrate that an increase in type III collagen synthesis is responsible for the pathogenesis of macroangiopathy, it suggests that insulin-dependent fibroblast sensitization may play a role in the acceleration and progression of macroangiopathy.