Relationship between thromboxane/prostacyclin ratio and diabetic vascular complications.

To elucidate the relationship between the thromboxane A2/prostacyclin (TXA2/PGI2) ratio and diabetic complications, the levels of 11-dehydro-thromboxane B2 and 2,3-dinor-6-keto-prostaglandin F1alpha, the urinary metabolites of thromboxane A2 and prostacyclin, were measured in diabetics by gas chromatography/selected ion monitoring. We compared the TXA2/PGI2 ratio in healthy volunteers and diabetics. The TXA2/PGI2 ratio of diabetics was significantly higher than that of healthy volunteers and we could reconfirm the hypercoagulable condition in diabetics. We also investigated the difference of TXA2/PGI2 levels in diabetics with retinopathy and neuropathy. The TXA2/PGI2 ratio of diabetics with retinopathy showed significantly higher level than without retinopathy. However, the TXA2/PGI2 ratio of diabetics with neuropathy was the same as without neuropathy. These results suggest that the TXA2/PGI2 ratio reflects the pathological conditions of diabetes, especially the change of vasculature. The monitoring and improvement of TXA2/PGI2 ratio could be useful for the prevention of diabetic vascular complications.     

Increased thromboxane A2 production but normal prostacyclin by the placenta in hypertensive pregnancies

The production of vasodilatory, antiaggregatory prostacyclin (PGI₂) and vasoconstrictory, proaggregatory thromboxane A₂ (TxA₂) by the placenta was studied in the cases of hypertensive pregnancy complications by superfusing pieces from maternal and fetal sides of placentae of 9 pre-eclamptic, 6 hypertensive and 11 healthy women and measuring the release of 6-keto-prostaglandin F_1α (6-keto-PGF_1α) and thromboxane B₂ (TxB₂), the breakdown products of PGI₂ and TxA₂ respectively, from the superfusate. Both sides of the placentae from the controls produced 6-keto-PGF_1α (maternal side 0.5±0.1 ng/g/min dry weight of tissue, mean±SEM; fetal side 0.7±0.2 ng/g/min) and TxB₂ (maternal side 2.5±0.4 ng/g/min; fetal side 2.7±0.5 ng/g/min with no correlation between the two. The 6-keto-PGF_1α production was normal in hypertensive complications whereas the TxB₂ production was increased on the fetal side of the placentae obtained from the pre-eclamptic (3.7±0.3 ng/g/min: p<0.05) and hypertensive women (4.1±0.4 ng/g/min; p<0.025). This may explain the occurrence of microthrombi and infarctions in placentae of hypertensive women.     

In vivo insulin responsiveness for glucose uptake and production at eu- and hyperglycemic levels in normal and diabetic rats.

Whole body glucose uptake (BGU) and hepatic glucose production (HGP) at maximal plasma insulin concentrations (+/- 5000 microU/ml) were determined by eu- (EC) (6 mM) and hyperglycemic (HC) (20 mM) clamps (120 min), combined with [3-3H]glucose infusion, in normal and streptozotocin-treated (65 mg/kg) 3-day diabetic, conscious rats. In normal rats, during EC, BGU was 12.4 +/- 0.4 mg/min and during HC, when urinary glucose loss was 0.54 +/- 0.09 mg/min, BGU was 25.5 +/- 1.6 mg/min. However, throughout the final 60 min of HC, glucose infusion rate (GIR) was not constant but a linear decline in time (r = -0.99) of 17%, P less than 0.0001, was observed indicating a hyperglycemia-induced desensitization process. In diabetic rats, during EC, BGU was 7.7 +/- 0.3 mg/min and during HC, BGU was 15.5 +/- 1.4 mg/min. Throughout the final 60 min of HC, GIR was constant, suggesting that the hyperglycemia-induced desensitization process was already completed. In normal and diabetic rats, HGP was similar: during EC 0.2 +/- 0.5 mg/min and 0.1 +/- 0.5 mg/min, and during HC 0.4 +/- 0.4 mg/min and 0.5 +/- 0.6 mg/min, respectively. In vitro adipocyte and muscle insulin receptor studies showed normal to increased receptor number and increased receptor autophosphorylation in diabetic compared to normal rats. In conclusion: (i) 3-day diabetic rats show, at maximal plasma insulin concentrations, insulin resistance to BGU, but not to HGP. The resistance to BGU is equally present (reduction of 38%) at eu- and hyperglycemic levels as compared to normal rats. (ii) 3-day diabetic rats reveal no defect in adipocyte and muscle insulin receptor function. These data indicate that the diabetes induced insulin resistance for BGU is at the post-receptor level and due to a decreased maximal capacity (Vmax) for glucose uptake, with no change in affinity, or Km.     

Differential effects of megavitamin E on prostacyclin and thromboxane synthesis in streptozotocin-induced diabetic rats

Diabetic subjects tend to develop microvascular complications believed to be due to platelet hyperaggregability. This increased platelet sensitivity is though to be the result of an imbalance of PGI2 and TXA2 production in diabetes. This study sought to determine whether megavitamin E supplementation could restore PGI2/TXA2 balance in streptozotocin-diabetic rats. Endogenous release of PGI2 by isolated aorta, determined via radioimmunoassay of its stable metabolite, 6-keto-PGF1 alpha, was significantly greater (P less than 0.05) in rats receiving 100x the normal vitamin E requirement than in untreated diabetic rats. PGI2 synthesis was negatively correlated with plasma glucose levels (r = -0.87, P less than 0.05) in non-fasted rats at sacrifice. Vitamin E supplementation, at both the 10x and the 100x level, significantly depressed (P less than 0.05) thrombin-stimulated synthesis of TXA2 in washed platelet. PGI2 and TXA2 production were expressed as a ratio. Megavitamin E therapy appears to increase this ratio over that seen in the diabetic animal. The data suggest that vitamin E, at high levels, exerts an ameliorating influence of the PGI2/TXA2 imbalance of diabetes.     

Interactions between oxytocin, glucagon and glucose in normal and streptozotocin-induced diabetic rats

Oxytocin has been suggested to have glucoregulatory functions in rats, man and other mammals. The hyperglycemic actions of oxytocin are believed to be mediated indirectly through changes in pancreatic function. The present study examined the interaction between glucose and oxytocin in normal and streptozotocin (STZ)-induced diabetic rats, under basal conditions and after injections of oxytocin. Plasma glucose and endogenous oxytocin levels were significantly correlated in cannulated lactating rats (r = 0.44, P less than 0.01). To test the hypothesis that oxytocin was acting to elevate plasma glucose, adult male rats were injected with 10 micrograms/kg oxytocin and killed 60 min later. Oxytocin increased plasma glucose from 6.1 +/- 0.1 to 6.8 +/- 0.2 mM (P less than 0.05), and glucagon from 179 +/- 12 to 259 +/- 32 pg/ml (P less than 0.01, n = 18). There was no significant effect of oxytocin on plasma insulin, although the levels were increased by 30%. A lower dose (1 microgram/kg) of oxytocin had no significant effect on plasma glucose or glucagon. To eliminate putative local inhibitory effects of insulin on glucagon secretion, male rats were made diabetic by i.p. injection of 100 mg/kg STZ, which increased glucose to greater than 18 mM and glucagon to 249 +/- 25 pg/ml (P less than 0.05). In these rats, 10 micrograms/kg oxytocin failed to further increase plasma glucose, but caused a much greater increase in glucagon (to 828 +/- 248 pg/ml) and also increased plasma ACTH. A specific oxytocin analog, Thr4,Gly7-oxytocin, mimicked the effect of oxytocin on glucagon secretion in diabetic rats. The lower dose of oxytocin also increased glucagon levels (to 1300 +/- 250 pg/ml), but the effect was not significant. A 3 h i.v. infusion of 1 nmol/kg per h oxytocin in conscious male rats significantly increased glucagon levels by 30 min in normal and STZ-rats; levels returned to baseline by 30 min after stopping the infusion. Plasma glucose increased in the normal, but not STZ-rats. The relative magnitude of the increase in glucagon was identical for normal and diabetic rats, but the absolute levels of glucagon during the infusion were twice as high in the diabetics. To test whether hypoglycemia could elevate plasma levels of oxytocin, male rats were injected i.p. with insulin and killed from 15-180 min later. Plasma glucose levels dropped to less than 2.5 mM by 15 min. Oxytocin levels increased by 150-200% at 30 min; however, the effect was not statistically significant.(ABSTRACT TRUNCATED AT 400 WORDS)     

Comparison of umbilical vein models for measurement of relative prostacyclin and thromboxane production

There is growing evidence that blood vessels generate TXA₂ in addition to PGI₂. We examined effluents from continously perfused human umbilical vein and supernatants from umbilical vein rings for TXB₂ and 6-keto-PGF_1α measurements (stable metabolites of TXA₂ and PGI₂, respectively). TXB₂ and 6-keto-PGF_1α were identified in all samples. 6-keto-PGF_1α to TXB₂ ratio was higher in intact vein effluents than in the venous ring supernatants (112:1 and 28:1, respectively, P<0.01). Arachidonate stimulation increased 6-keto-PGF_1α and TXB₂ levels similarly in the intact vein effluent. In contrast, stimulation of the venous rings resulted in a relatively larger increase in TXB₂ than in 6-keto-PGF_1α. This caused 6-keto-PGF_1α to TXB₂ ratio to decline (p<0.01). The identity of TXB₂ was confirmed in several different ways. These data suggest that 1) human umbilical veins produce TXA₂ in addition to PGI₂, 2) TXA₂ release is more by venous rings than by the intact vein probably reflecting contribution from non-endothelial layers, and 3) arachidonate stimulation causes relatively greater release of TXA₂ than of PGI₂ from the venous rings, whereas release of PGI₂ and TXA₂ is similar from the intact vein.     

Plasma thromboxane and prostacyclin: comparison during normal pregnancy and pregnancy complicated by hypertension

Plasma levels of thromboxane B2 (TXB2) and 6-keto-prostaglandin F1 alpha (6-keto-PGF1 alpha), stable metabolites of two prostanoids with opposing biological effects, TXA2 and prostacyclin, were measured by radioimmunoassay in normal pregnancy (controls) and pregnancy complicated by hypertension (PIH) from 32 to 36 (Period 1; P1) and from 36 to 40 (Period 2; P2) weeks of gestation. The plasma concentration of each compound in the control subjects was 265.6 +/- 58.4 (TXB2), 132.4 +/- 16.5 (6-keto-PGF1 alpha) for P1 (n = 10) and 142.6 +/- 11.8 (TXB2), 68.5 +/- 5.2 (6-keto-PGF1 alpha) for P2 (n = 10) respectively (pg/ml, mean +/- s.e). In the patients with PIH, TXB2 concentrations increased moderately for P1 (419.2 +/- 21.2; n = 7) and significantly (p less than 0.005) for P2 (452.8 +/- 31.0; n = 7) respectively (pg/ml, mean +/- s.e), while the plasma levels of 6-keto-PGF1 alpha revealed a slight to moderate decrease both for P1 (84.5 +/- 4.0; n = 7) and P2 (59.7 +/- 8.1; n = 7) respectively (pg/ml, mean +/- s.e). The physiological balance of TXB2 to 6-keto-PGF1 alpha was significantly greater (p less than 0.005) in the patients with PIH, where the TXB2/6-keto-PGF1 alpha ratio was 5.2 +/- 0.7 for P1 and 9.4 +/- 2.3 for P2 respectively (mean +/- s.e) compared with that of the controls, where it was 2.4 +/- 0.4 for P1 and 2.0 +/- 0.2 for P2 respectively (mean +/- s.e).(ABSTRACT TRUNCATED AT 250 WORDS)     

Spider silk fibroin enhances insulin secretion and reduces blood glucose levels in diabetic mice

Fibroin silk proteins make up the cocoons of silkworms and spider webs and are rich in glycine and alanine residues. Recent studies have shown that silk fibroin hydrolysate from silkworms improves blood glucose and lipid metabolism. In the present study, we investigated the anti-diabetic effects of spider silk fibroin protein in type 2 diabetic mice. Recombinant AvMaSp-R, which consists of the 240 amino acid repetitive domain of major ampullate silk protein (AvMaSp) from the spider Araneus ventricosus, was produced in baculovirus-infected insect cells. We tested the effects of oral AvMaSp-R administration on serum insulin and blood glucose levels in diabetic mice and found that AvMaSp-R increases serum insulin levels and reduces blood glucose levels in diabetic mice. Consequently, our results are the first to provide evidence that silk fibroin protein from spiders enhances insulin secretion, which leads to reduced blood glucose levels in type 2 diabetic mice.     

Comparison of umbilical vein models for measurement of relative prostacyclin and thromboxane production

There is growing evidence that blood vessels generate TXA2 in addition to PGI2. We examined effluents from continuously perfused human umbilical vein and supernatants from umbilical vein rings for TXB2 and 6-keto-PGF1 alpha measurements (stable metabolites of TXA2 and PGI2, respectively). TXB2 and 6-keto-PGF1 alpha were identified in all samples. 6-keto-PGF1 alpha to TXB2 ratio was higher in intact vein effluents than in the venous ring supernatants (112:1 and 28:1, respectively, P less than 0.01). Arachidonate stimulation increased 6-keto-PGF1 alpha and TXB2 levels similarly in the intact vein effluent. In contrast, stimulation of the venous rings resulted in a relatively larger increase in TXB2 than in 6-keto-PGF1 alpha. This caused 6-keto-PGF1 alpha to TXB2 ratio to decline (p less than 0.01). The identity of TXB2 was confirmed in several different ways. These data suggest that 1) human umbilical veins produce TXA2 in addition to PGI2, 2) TXA2 release is more by venous rings than by the intact vein probably reflecting contribution from non-endothelial layers, and 3) arachidonate stimulation causes relatively greater release of TXA2 than of PGI2 from the venous rings, whereas release of PGI2 and TXA2 is similar from the intact vein.     

Reevaluation of circulating prostacyclin and thromboxane in diabetes

Although several investigators have attempted to measure the plasma levels of prostacyclin (PGI₂) and thromboxane A₂ (TXA₂) in diabetes and normal subjects, their results have been controversial. In this study, we measured plasma PGI₂ and TXA₂ levels in diabetic patients and normal subjects. The plasma PGI₂ and TXA₂ were determined by RIA as 6-keto-PGF_1a and TXB₂, respectively. The plasma levels of 6-keto-PGF_1a were significantly reduced in diabetics with microangiopathy (52.5 ± 18.9 pg/ml, mean ± SE, p<0.05) compared with those of normal subjects. Diabetics as a whole also showed lower levels of 6-keto-PGF_1a than normal subjects (57.8 ± 26.1 vs. 70.2 ± 20.7 pg/ml), though this was not significant statistically. The plasma 6-keto-PGF_1a levels did not significantly correlate with either age of the patients or duration of diabetes in diabetics. Interestingly, however, hemoglobin Alc significantly correlated inversely with 6-keto-PGF_1a levels in diabetics without microangiopathy (r=−0.60, p<0.05). The plasma levels of TXB₂ in diabetics were significantly higher than those of normal subjects (155.2 ± 69.5 vs. 108.0 ± 30.0 pg/ml, p<0.05). These data suggest that an imbalance of circulating PGI₂ and TXA₂ may contribute to the development of diabetic microangiopathy.     

The effect of exercise on bleeding time and local production of prostacyclin and thromboxane

Exercise at a heart rate corresponding to 30% VO2max for 15 min was associated with an increase in the volume of bleeding time blood from a mean of 133 microliters before exercise to a mean of 218 microliters during and immediately after the exercise. There was similarly an increase in thromboxane B2 production from 6.40 nmol.l-1 before to 11.50 nmol.l-1. Most subjects also showed an increase in the length of the bleeding time and in the production of bleeding time 6-keto-PGF1 alpha. The extent of increase in the bleeding time and in production of 6-keto-PGF1 alpha was quite variable, with subjects showing the largest increases in bleeding time also demonstrating the greatest increases in 6-keto-PGF1 alpha (r = 0.76, P = 0.004). The ingestion of aspirin before exercise markedly inhibited basal bleeding time thromboxane B2 production and blocked the exercise-associated increments in thromboxane B2 and 6-keto-PGF1 alpha production. While the aspirin itself increased the length of the bleeding time, there was not any further increase associated with exercise. In contrast to the effects of acute short-term exercise, long-distance running was associated with a significant decrease in bleeding time, but no change in bleeding time blood volume, bleeding time thromboxane B2, or bleeding time 6-keto-PGF1 alpha. The results show that acute low-level exercise can be associated with significant changes in the volume of blood oozing from a bleeding time incision and in the amount of thromboxane production stimulated at the incisional site. Following exhaustive exercise of long duration, the above changes are no longer seen.     

In vivo biosynthesis of thromboxane and prostacyclin during exposure to physiological levels of epinephrine

The effects of 20-min epinephrine infusion (0.025 and 0.3 nmol/kg/min) on the in vivo synthesis of thromboxane A2 and prostacyclin were studied in ten healthy male volunteers. We assessed the in vivo biosynthesis of thromboxane A2 and prostacyclin by measurement of the urinary metabolites 2,3-dinor-TxB2 and 2,3-dinor-6-keto-PGF1 alpha, respectively. Epinephrine infusion did not cause any significant changes in the urinary excretion of the two metabolites. Thus, we conclude that physiological levels of epinephrine do not affect the in vivo biosynthesis of thromboxane A2 and prostacyclin.     

Double-label expression studies of prostacyclin synthase, thromboxane synthase and COX isoforms in normal aortic endothelium

We have performed double-label immunofluorescence microscopy studies to evaluate the extent of co-localization of prostacyclin synthase (PGIS) and thromboxane synthase (TXS) with cyclooxygenase (COX)-1 and COX-2 in normal aortic endothelium. In dogs, COX-2 expression was found to be restricted to small foci of endothelial cells while COX-1, PGIS and TXS were widely distributed throughout the endothelium. Quantification of the total cross-sectioned aortic endothelium revealed a 6- to 7-fold greater expression of COX-1 relative to COX-2 (55 vs. 8%) and greater co-distribution of PGIS with COX-1 compared to COX-2 (19 vs. 3%). These results are in contrast to the extensive co-localization of PGIS and COX-2 in bronchiolar epithelium. In rat and human aortas, immunofluorescence studies also showed significant COX-1 and PGIS co-localization in the endothelium. Only minor focal COX-2 expression was detected in rat endothelium, similar to the dog, while COX-2 was not detected in human specimens. Inhibition studies in rats showed that selective COX-1 inhibition caused a marked reduction of 6-keto-PGF(1alpha) and TXB(2) aortic tissue levels, while COX-2 inhibition had no significant effect, providing further evidence for a functionally larger contribution of COX-1 to the synthesis of prostacyclin and thromboxane in aortic tissue. The data suggest a major role for COX-1 in the production of both prostacyclin and thromboxane in normal aortic tissue. The extensive co-localization of PGIS and COX-2 in the lung also indicates significant tissue differences in the co-expression patterns of these two enzymes.     

Urinary thromboxane A2/prostacyclin balance reflects the pathological state of a diabetic

Levels of the stable urinary metabolites of thromboxane A2 and prostacyclin, 11-dehydro-thromboxane B2 (11-dehydro-TXB2) and 2,3-dinor-6-keto-prostaglandin F1alpha (2,3-dinor-6-keto-PGF1alpha) were measured in diabetics to elucidate the relation between the thromboxane A2/prostacyclin (TX/PGI) balance and pathological states of diabetes mellitus. 11-Dehydro-TXB2 and 2,3-dinor-6-keto-PGF1alpha were derivatized to methyl ester-propylamide-dimethylisopropylsilyl ether and methyl ester-methoxime-dimethylisopropylsilyl ether derivatives, respectively, and applied to a gas chromatography/selected ion monitoring. The TX/PGI ratios of diabetics were higher than those of healthy volunteers, suggesting the hypercoagulative states of this disease. The ratios showed positive correlations with the levels of blood glucose. The levels of hemoglobin A1c and triglyceride were correlated weakly with the ratio. Some of the patients who had relatively low levels of blood glucose also showed high TX/PGI ratios. Furthermore, the ratio increased in the order of the groups 1, 2, and 3; group 1 contained patients who did not take medicine for diabetes, group 2 contained those who took oral hypoglycemic agents, and group 3 contained those who received insulin therapy. These observations indicate that the TX/PGI ratio reflects the pathological conditions of diabetes and is a useful marker, having few different features from other markers that are presently used.