Impaired Relaxation in Aorta from Streptozotocin-diabetic Rats: Effect of Aminoguanidine (AMNG) Treatment

Aim The effect of 8 weeks′ streptozotocin (STZ)- induced diabetes and aminoguanidine (AMNG), the inhibitor of advanced glycosylation reaction, treatment on arteriolar reactivity to vasoactive substances was investigated in vitro. Materials and Methods Studies were performed in untreated control rats (n = 10), STZ-induced (60 mg/kg i.v.) diabetic rats (n = 10), AMNG-treated (600 mg/l given in drinking water throughout 8 weeks) control rats (n = 10) and AMNG-treated (600 mg/l given in drinking water, beginning at 72h after STZ and throughout 8 weeks of diabetes) diabetic rats (n = 10). Results are expressed as the mean ±s.e. Relaxant responses are expressed as a percentage (%) relaxation of noradrenaline-induced tone. Statistical comparisons were made by one-way analysis of variance (ANOVA) followed by Tukey–Kramer multiple comparisons test. Results 1. The decreased body weights (205 ± 6 g) and increased blood glucose levels (583 ± 8 mg/dl) of diabetic rats were partially restored by treatment of aminoguanidine (253 ± 6 g, p < 0.05 and 480 ± 14 mg/dl, p < 0.001, respectively). 2. Diabetes caused a 71% deficit in maximal endothelium-dependent relaxation to acetylcholine for noradrenaline precontracted aortas (p < 0.001). AMNG treatment prevented the diabetes-induced impairment in endothelium dependent relaxation (58 ± 8%) to acetylcholine, maximum relaxation remaining in the non-diabetic range (78 ± 4%). 3. Neither diabetes nor treatment affected endothelium-independent relaxation (pD2 and max. Relax.) to sodium nitroprusside. 4. Vasoconstrictor responses (pD2 and Max. Contraction) to noradrenaline and KCl were not influenced by the diabetic state and treatment. Conclusion Our data suggest that 8 weeks of experimental diabetes is associated with a decreased endothelium-dependent vasodilatation. AMNG treatment may prevent diabetes-induced endothelial dysfunction. This may be mediated via the prevention of advanced glycosylation end product formation, the enhanced release of vasodilator substances such as prostacyclin, the increased elasticity of blood vessels, the antioxidant activity and inhibitor activity of enzyme aldose-reductase by AMNG.


INTRODUCTION
Hyperglycaemia is the primary cause of diabetic micro and macrovascular complications. I1"2J Several major mechanisms have been proposed for hyperglycaemia-induced tissue damage, including advanced glycation end product *Corresponding author. Tel." +90 212 586 15 52, Fax: +90 212 633 01 31, e-mail: gakkan@istanbul.edu.tr 145 (AGE) formation, altered intracellular redox (iii) AMNG-treated control group (AG-C) state, increased polyol pathway flux, apoptosis (n 10), and increased de novo diacylglycerol synthesis (iv) AMNG-treated diabetic group (AG + D) with resultant activation of protein kinase C (n 10). isoform. [3][4][5][6][7] Studies on vascular reactivity using the STZ-Induction of Experimental Diabetes diabetic rat have reported inconsistent results with increased, decreased and unchanged re-Rats were injected intravenously (into the lateral sponsiveness to noradrenaline, KC1, acetylchotail vein) with STZ (60mg/kg)(Sigma, S-0130, line and sodium nitroprusside. E1-14 Lot. 128H1045). STZ was freshly prepared in Today, in addition to the prevention of 0.25 ml saline. The rats were then maintained for hyperglycaemia by use of insulin and oral 8 weeks with free access to food and water. hypoglycaemic agents, it has been also used Duration of the experimental period, blood and tested some agent such as aminoguanidine glucose levels and body weights were measured for the prevention and treatment of complica-beginning from 8th week. tions in Diabetes Mellitus. Aminoguanidine is a small, hydrazine compound structurally identical to the amino-terminal group of arginine and Aminoguanidine Treatment several laboratories have reported beneficial The rats in group III (AMNG-treated control effects of aminoguanidine in some diabetic group) and group IV (AMNG-treated STZcomplications. The mechanisms for these effects diabetic group) were given 600 mg/1 AMNG in of AMNG were accompanied by reduction of drinking water (in group AG + D, beginning at advanced glycation end product (AGE) forma-72h after STZ) for 8 weeks. The dosage in the tion, the inhibition of an isoform of nitric oxide drinking water was three times lower for the synthase (iNOS) and reactive oxygen species diabetic animals because their water intake is (ROS) formation, lipid peroxidation and oxithree times higher than that of nondiabetic dant-induced apoptosis. E15-18 animals. [45] In the dosage calculation, followed The present study was designed to define formula was used: whether AMNG has beneficial effects on altered [(mg AG in per ml drinking waterxml vascular reactivity of the thoracic aortas in daily water intake)]x[(1000/body weight)]. streptozotocin-diabetic rats.

Animals
Albino Wistar rats were bred in our laboratory and nourished ad libitum with standart pellet diet and had free access to tap water.
Adult male and female Wistar rats (n=40), weighing 200-300g, were divided into four groups: Rats were sacrified by stunning followed by decapitation. A segment (3-5cm) of thoracic aorta was removed and placed in an ice-cold Krebs-Ringer solution (KRS) of the following composition (mmol/L): NaC1 (118), KC1 (4.7), CaC12 (2.5), MgSO4.7H20 (1.2), KH2PO4 (1.2), NaHCO3 and glucose (11.1) and then trimmed free of adhering fat and connective tissue and cut into rings of 3 mm width. The rings were opened by cutting the vessels longitudinally. Subsequently, they were fixed with stainless steel clips at both ends and then placed in 20 ml organ baths containing KRS, gassed with carbogen (95% 02+5% CO2) providing pH 7.4 at 37C. The preparation were connected to isometric force displacement transducer (MAY, FDT 10-A isometric transducer) connected to MAY, TDA 97 polygraph and were equilibrated for 90min at optimal resting tension of 2g. During this time, the KRS in the organ bath was replaced every 20 min.

Concentration-response Curves
After equilibration, the thoracic aorta strips were exposed to 10 -5 mol/L ( EC90) noradrenaline until the contraction reached the plateau (approximately 15 min) in order to measure the fast and slow components of vascular response to noradrenaline. The fast component was measured from the baseline to the point at which the rate of contraction started to decrease abruptly. The slow component was measured from this point to the top of the contraction. The total response was the sum of these two components. [16] Concentration-response curves were obtained with noradrenaline. Noradrenaline (10-9/3 10 -9-10-4/3 10-4mol/L) was added in a cumulative manner until a maximal response was achieved. After the addition of each dose, a plateau response was obtained before the addition of a subsequent dose. Cumulative relaxation curves to acetylcholine (Ach) (10-8-10-4 mol/L) and sodium nitroprusside (SNP) (10-8-104mol/L) were obtained in each strip precontracted submaximally (approximately EC90, 10-5mol/L) by addition of noradrenaline. Concentration-response curves were obtained with KC1. KC1 (20,40,60 and 80 mmol/L) was added in a cumulative manner until a maximal response was achieved. After the addition of each dose, a plateau response was obtained before the addition of a subsequent dose.
At the end of each experiment, tissue was blotted dry, measured and weighed.

Data Analysis
Contractile responses to noradrenaline, KC1 and relaxations to Ach (Inh.%) and SNP were calculated as the increase (and decrease for Ach and SNP) in tension (mg) in response to the agonist per mg of aorta. Agonist pD2 value (=-log EC50) was calculated from each agonist concentration-response curve by linear regression analysis of the linear portion of the curve and taken as a measure of the sensitivity of the tissues to each agonist. All values are expressed as (mean 4-SE). Statistical analysis of the data was performed using one-way analysis of variance (ANOVA) followed by Tukey-Kramer Multiple comparisons test. p K 0.05 was considered as statistically significant.

General Characteristics
Data showing changes in body weight and final blood glucose concentration for all groups are summarized in Table I.
Eight weeks after injection, all rats treated with STZ exhibited severe hyperglycaemia and their blood glucose levels (583 4-8mg/dl) were significantly higher than those of NC (p < 0.001), AG-C (p < 0.001) and AG+D (p < 0.001)rats.
Water intakes and AG dosages were also shown in Table I.

Agonist-induced Contractions
The contractile responses of aortic strips from all groups are shown in Tables II and III Table IV, either fast or slow components of responses to noradrenaline were not significantly different among all the experimental groups.

Agonist-induced Relaxations
The acetylcholine-mediated relaxation (pD2 and %Max. Relax.) of aortic strips precontracted with NA is shown in Tables II and III. Either the diabetic state or AMNG treatment had no effect upon the pD2 pattern of relaxations in all groups. In contrast, the maximum relaxations generated by aortic strips from diabetic rats to Ach were markedly smaller (p < 0.001) than those generated by corresponding control tissues (NC) from age-matched rats. Treatment of diabetic animals with AMNG prevented the diabetes-induced depression of relaxations to Ach (DC vs. AG + D, p < 0.01).  [10-14, 19, 20] response.
It has been shown that hyperglycaemia which is one of the most important markers of diabetes, caused tissue damage with several mechanisms, including advanced glycation end product (AGE) formation, increased polyol pathway flux, apoptosis and reactive oxygen species (ROS) formation. 3,8,9,21] Endothelial dysfunction and impaired endothelium-dependent relaxations have been consistently demonstrated with the histologic and experimental studies in animal models of diabetes mellitus. [19,[22][23][24][25][26][27] The present results also demonstrated that in aortas from STZ-diabetic rats, the endothelium-dependent relaxant responses to Ach (25 4-5%) was significantly decreased compared with NC (744-4%, p < 0.001) rats. Relaxation in the AMNG treated-diabetic group (AG+D) was significantly greater than that seen in the diabetic control (DC) group and % Max. Relax.
values returned to near-control values (NC) (58 4-8%). Our results are in agreement with the majority of previous studies [22,[23][24][25][26][27] although others have reported no difference in vascular responsiveness to Ach in diabetic rats. [28,29] The endothelium-dependent relaxant responses to agents such as acetylcholine are largely due to release of endothelium derived relaxing factor (EDRF). I301 EDRF is now considered to be identical to nitric oxide which is a key transducer of the vasodilator signalling. I30, 31] Impaired endothelium-dependent relaxation in STZ-induced diabetic rat might be due to increased blood glucose level, decreased blood insulin level, decreased influx of Ca 2+ into endothelium or decreased release of Ca 2+ from its storage sites, a decreased content or inactivation of NO syntase and decreased diffusion of NO into the smooth muscle. 581 In addition, diabetes is believed to cause endothelial damage via oxidative stress which induces ROS and lipid peroxidation and a diabetes-induced functional change in vascular endothelial cells could be a key event in the development of the altered endothelium-dependent vasoreactivity. [32,33] AMNG used in this study is a potential therapeutic agent for preventing the generation of advanced glycation end products in diabetes mellitus. [34] Although Crinjns et al. (1998) have reported no beneficial effect of AMNG-treatment, I451 Bucala et al. (1991) previously demonstrated that acceleration of the advanced glycosylation process in vivo results in a timedependent impairment in endothelium-dependent relaxation and inhibition of advanced glycosylation with aminoguanidine prevents nitric oxide quenching, and ameliorates the vasodilatory impairment. 35 These results agree with our findings. In addition, a number have suggested that AGEs decrease NO and cGMP levels and vascular mechanical properties and AMNG-treatment ameliorates this disturbances. I35, 36,38] On the other hand, AMNG could increase the release of vasodilator substances such as prostacylin, [37] and inhibit ROS formation [171 and oxidant-induced apoptosis. 181 An alternative explanation for beneficial effect of AMNG is its role on the polyol pathway. It has been shown that polyol pathway is related with the deficit for endothelium-dependent relaxation and aldose reductase inhibitors can prevent this deficit in aorta from STZ-diabetic rats [131 and Kumari et al., have recently demonstrated that AMNG is an aldose reductase inhibitor. [39,40] The majority of previous studies showed that the responsiveness to the endothelium-independent vasodilator, SNP, is not impaired in diabetics versus control [33'43-46] whilst some [41,42] has demonstrated an impaired response. This study have also showed an unchanged sensitivity and a maximum relaxation to SNP in diabetics. Our results agree with the majority of previous studies. [33,[43][44][45][46] The contractile response of the rat aorta to noradrenaline, a nonselective alpha agonist, is biphasic, consisting of a fast and a slow component. 47 The fast component of the contraction induced by noradrenaline is a consequence of activation of alpha-1 adrenoceptors and has been demonstrated to be due to mobilisation of intracellular calcium whereas the slow component reflects activation of alpha-2 adrenoceptors and is directly dependent upon an influx of extracellular calcium. [48,49] Despite the controversy concerning the effects of diabetes on the maximal response, sensitivity, fast and slow components of response to the alpha-adrenoceptor agonist, such as noradrenaline, [49][50][51][52] most studies agree that agonist potency is not altered by diabetes. [20,[53][54][55] The present study has also demonstrated an unchanged sensitivity, a maximum contraction and its components to noradrenaline in diabetes. Our results are in agreement with some of the previous studies, t20, 53, 84, 58 The discrepancy between our results and those of other groups may be due to the different experiment protocols. In the present study, there were also no significant differences between the diabetic tissues and control tissues in the responsiveness to KC1. However, although some authors found conflicting results, 56'57 the present results [28,54] agree with those of other groups.
In conclusion, our study demonstrates that the chronic AMNG-treatment reversed the diminished vascular relaxation. However, its exact mechanism of action remain unclear and requires further investigation.