Nitric oxide (NO) has emerged as an important molecule with diverse regulatory functions. In the blood vessels, NO mediates endothelium-dependent vasodilation, 1,2 and in the central and peripheral nervous system NO is a so-called “unusual neurotransmitter”. 3 It has also been shown that systemic inhibition of NO synthase causes insulin resistance, 4 which can be counteracted by intraportal administration of the nitric oxide donor 3-morpholinosydnonimine (SIN-1). 5 This suggests that at an appropriate dosing schedule, NO donors might serve as drugs that increase insulin sensitivity.
In a very recent study, we have provided evidence for such a possible clinical exploitation; we have found that transdermal nitroglycerin at a dose used for angina prophylaxis suppresses glucose-stimulated insulin release with an increase in insulin sensitivity in healthy male volunteers. 6 This clinical finding seems to indirectly support the concept of nitrergic regulation of peripheral insulin sensitivity 7,8 with the possibility of therapeutic use of nitrates in the management of insulin resistance and/or either type of diabetes. The prolonged use of nitrates, however, is often associated with the development of tolerance with an attenuation of the therapeutic effect. 9 Moreover, endogenous nitrergic mechanisms have also been shown to seriously impair in the state of nitrate tolerance. 10–12 The present work was therefore concerned with the possibility that hemodynamic nitrate tolerance impairs the sensitivity of tissues to the hypoglycemic effect of insulin. To answer the question, we used the hyperinsulinemic euglycemic glucose clamp method in chronically instrumented conscious rabbits, a commonly used methodological approach to study insulin sensitivity in vivo. 13,14 Choosing this model was further supported by our previous experiences with nitrate tolerance in conscious rabbits. 10,15
All experiments performed in this study conform to the European Community guiding principles for the care and use of experimental animals. Moreover, the experiments detailed below were approved by the local ethical boards of our universities.
Adult male New Zealand white rabbits, weighing 3 to 3.2 kg, housed in an animal room (12-hour light/dark periods a day, temperature of 22 to 25°C, humidity of 50 to 70%) with 1 animal per pen, fed commercial laboratory chow and tap water ad libitum, were used throughout. The animals underwent surgery after a 2-week adaptation period.
Surgery was performed under aseptic conditions. The rabbits were anaesthetized with an intravenous bolus of 10 mg · kg−1 diazepam (Sigma, St Louis, MO) and 5 mg · kg−1 ketamine (EGIS Pharmaceuticals Ltd., Budapest, Hungary). Lidocaine (EGIS, Hungary) was given subcutaneously for local pain relief as described previously. 10 Polyethylene catheters were inserted into two major branches of the jugular vein for insulin and glucose infusion and to obtain venous blood samples (maximum volume 0.5 mL) for methemoglobin determinations by means of multiwavelength photometric analysis using an ABL 625 System (Radiometer A/S, Copenhagen NV, Denlarg). A polyethylene tube connected to a Statham P23 DB transducer (Gould, Balainvilliers, France) and an EXPERIMETRIA (EXP-2, Budapest, Hungary) multiscriptor was inserted into the left carotid artery to measure mean arterial blood pressure. The catheters were exteriorized through the back of the neck as described. 10 A small catheter also was inserted into the central artery of the right ear for arterial blood sampling. These lines were kept patent by filling them with sodium heparin solution (100 IU ml−1).
Intravenous Glucose Disposal
The rabbits were given an intravenous bolus of 0.5 g/kg glucose dissolved in isotonic sodium chloride in 15 mL volume over 2 minutes by means of an infusion pump (Braun, Melsungen, Germany). Arterial blood samples were collected before infusion as well as at 10, 30, 60, 120, and 180 minutes after infusion for blood glucose and plasma insulin determination. Glucose was measured by the glucose oxidase method; plasma insulin immunoreactivity was determined by means of radioimmunoassay (RIA) using IZINTA (Isotope Institute, Budapest, Hungary) insulin RIA kits as described. 6,16
Hyperinsulinemic Euglycemic Glucose Clamp Studies
Human regular insulin was infused at a constant rate (13 mU/kg, NOVO Nordisk, Copenhagen, Denmark) via one of the venous catheters over 120 minutes. This insulin infusion yielded plasma insulin immunoreactivity of 100 ± 5 μU/ml in the steady state (see below). Blood samples (0.3 mL) were taken from the arterial cannula for blood glucose concentration at 10-minute intervals. Blood glucose concentration was maintained constant (5.5 ± 0.5 mmol/l) by a variable rate of glucose infusion via the second venous cannula. When blood glucose had stabilized for at least 30 minutes, we defined this condition as steady state. In the steady state, additional blood samples (0.5 mL) were taken for plasma insulin determination at 10-minute intervals. The glucose infusion rate (mg/kg/min) during steady state was used to characterize insulin sensitivity. 17
Induction of Hemodynamic Nitrate Tolerance
Hemodynamic nitrate tolerance was induced by continuous exposure to transdermal nitroglycerin as described. 15 In brief, exposure of the rabbits to transdermal patches releasing approximately 0.07 mg/kg/h nitroglycerin (Nitroderm TTS 5, Ciba Hungaria, Budapest, Hungary) over 7 days results in the lack of the hypotensive response to an intravenous test dose (30 μg/kg) of nitroglycerin, which produces a more than 20% decrease in mean arterial blood pressure in control animals. The hypotensive response, however, is preserved after a 6-hour treatment with transdermal nitroglycerin. 15 Thus, in the present work, a 6-hour treatment with transdermal nitroglycerin was used to study the effect of nitroglycerin in the non-tolerant state, whereas a continuous 7-day exposure to TTS 5 patches was used to study the effect of nitrate tolerance on the variables investigated.
One week after surgery, the rabbits were randomized to three major experimental groups. Group 1 (n = 12) animals entered the glucose disposal study, whereas Group 2 (n = 24) and Group 3 (n = 24) rabbits were used for hyperinsulinemic euglycemic clamp experiments. Group 1 rabbits were further randomized as to whether they received either transdermal patches releasing approximately 0.07 mg/kg/h nitroglycerin (Nitroderm TTS 5, Ciba Hungaria, Budapest, Hungary) or matching placebo patches continuously over 7 days. (Each patch was replaced daily with a new one.) Six hours after removal of the last patch, the intravenous glucose disposal study was commenced. Group 2 animals were randomized in the same way. In this group of rabbits, the hyperinsulinemic euglycemic glucose clamp studies were started 6 hours after removal of the last patch. Group 3 rabbits were divided into four subgroups; six animals received TTS 5 patches over 6 hours in the third hour of which the hyperinsulinemic euglycemic clamp study was commenced (ie, in the presence of transdermal nitroglycerin in non-tolerant state). The preceding 3-hour exposure to transdermal nitroglycerin was necessary to overcome problems resulting from an initial decrease in blood pressure due to “patch on” that could have influenced insulin sensitivity through a secondary baroreflex activation. 6 Another subgroup of six animals received transdermal nitroglycerin over 7 days and hyperinsulinemic clamp studies occurred on the last day (ie, in the presence of transdermal nitroglycerin in the tolerant state). The control group received matching placebo patches over the corresponding periods. After completion of the glucose disposal or the insulin clamp studies, we confirmed the presence or absence of nitrate tolerance by measuring changes in blood pressure in response to the intravenous test dose of nitroglycerin (30 μg/kg) as described previously. 15 We show the drug study design in Figure 1.
Data are expressed as means ± standard deviation (SD) of the mean and were analyzed with ANOVA followed by a modified t test for paired data. P values were adjusted according to Bonferroni method. 18
Effect of Hemodynamic Nitrate Tolerance on Baseline Hemodynamics and Methemoglobin Formation
Transdermal nitroglycerin produced a transient decrease in mean arterial blood pressure in the first 30 minutes after the patch had been applied with no change in heart rate. A 7-day continuous exposure to transdermal nitroglycerin, however, produced an increase in heart rate irrespective of the presence or absence of the patches. The increase in heart rate was statistically significant from the third day of continuous patch on with 277 ± 11, 280 ± 14, 275 ± 9, and 281 ± 15 b.p.m. on the third, fourth, fifth, and sixth days, respectively (P < 0.05 versus control for each). Baseline mean arterial blood pressure did not change over the 7-day period irrespective of the presence or absence of active or placebo patches (Table 1).
Methemoglobin formation was below 1% over the whole experimental period and did not change with the development of hemodynamic nitrate tolerance (data not shown).
Effect of Nitrate Tolerance on Glucose Disposal
It is seen from the data in Table 2 that an intravenous glucose load (0.5 mg/kg) increased blood glucose level to a similar degree in tolerant and control animals. However, both pre-load or post-load plasma insulin levels were significantly higher in tolerant than in control rabbits at each time point. It is also shown that pre-load blood glucose levels did not differ from each other in control (placebo-treated) and tolerant (treated with active nitroglycerin patches) animals.
Effect of Nitrate Tolerance on Insulin Sensitivity
The glucose infusion rate to maintain euglycemia (5.5 mmol/l) at clamped hyperinsulinemia (100 μU/ml) was significantly lower in tolerant than in control (placebo-treated) rabbits (Fig. 2).
Effect of Transdermal Nitroglycerin on Insulin Sensitivity in the Presence or Absence of Nitrate Tolerance
Acute treatment with transdermal nitroglycerin significantly increased insulin sensitivity reflected in an increase in glucose infusion rate to maintain euglycemia in animals with active “patch on” as compared with that in rabbits with placebo patches (Fig. 3). However, a 7-day exposure to active patches producing hemodynamic nitrate tolerance produced resistance to the hypoglycemic effect of insulin irrespective of the presence or the absence of nitroglycerin patches on the seventh day (Fig. 3).
The results confirm our previous finding that a 7-day continuous exposure to transdermal patches releasing approximately 0.07 mg.kg.hour-1 nitroglycerin produces hemodynamic nitrate tolerance in conscious rabbits. 15 The `tolerant' state is characterized by the lack of hypotensive response to an intravenous bolus of nitroglycerin, which is known to produce a marked decrease in blood pressure in `non-tolerant' rabbits. The results also show that nitrate tolerance is not confined to an attenuation of the vasodilatory effect of nitroglycerin but also disturbs the regulation of carbohydrate metabolism by decreasing insulin sensitivity. This is suggested by results from experiments at two different paradigms: (1) plasma insulin immunoreactivity increased substantially in the `tolerant' animals during glucose disposal and (2) during hyperinsulinemic euglycemic glucose clamp studies, a much lower glucose infusion rate maintained euglycemia at a clamped supraphysiological plasma insulin level in `tolerant' than in `non-tolerant' animals. Since this latter method serves as the `gold standard' for determining insulin sensitivity, in whole animals, we concluded that hemodynamic nitrate tolerance reduced insulin sensitivity at least in rabbits. In addition, the present results are in accordance with those of Kovacs et al, 6 who reported that transdermal nitroglycerin decreased insulin release and increased insulin-stimulated glucose uptake in healthy male volunteers. Nevertheless, this metabolic effect of nitroglycerin was also blocked in the `tolerant' state.
That some endogenous nitrergic mechanisms impair in nitrate tolerance has been demonstrated by results from our laboratory in the past decade. Preconditioning, the most potent cardioprotective mechanism described to date, 19 was found to be blocked by nitrate tolerance in rabbits 10,15 and in rats. 11 In these cases, the preconditioning phenomenon was triggered by rapid ventricular pacing, the protective mechanism of which was then shown to be underpinned by the sensory-effector function of nitrergic nerves. 20,21 Sari et al 12 observed that non-adrenergic, non-cholinergic relaxation of the rabbit sphincter of Oddi, a phenomenon known to be essentially nitrergic, also seriously impaired in nitrate tolerance. Further evidence for the deterioration in nitrergic sensory nerve function in nitrate tolerance was provided by Oroszi et al 22 who found that coronary vasodilation by capsaicin decreased significantly in hearts of guinea pigs made tolerant to the hypotensive effect of nitroglycerin. Similar results were obtained after inhibition of NO synthesis using the same model. 23 Thus, the consequences of hemodynamic nitrate tolerance resembled those seen after NO synthase inhibition using four different experimental paradigms in three different species.
To the best of our knowledge, this report is the first to describe the interaction between hemodynamic nitrate tolerance and insulin sensitivity. Therefore, that hemodynamic nitrate tolerance induces insulin resistance is the main original observation of the work. Nonetheless, the decrease in glucose-stimulated insulin release by transdermal nitroglycerin with no hyperglycemia in experimental animals is also delineated for the first time in the present article. These results seem to be in accordance with the pioneer findings of Lautt's group on the hepatic insulin sensitizing substance (HISS) mechanism. Based on results with rats they propose that post-prandial insulin release activates a neural mechanism sensitive to NO synthase inhibition linked to the anterior hepatic plexus that increases the sensitivity of peripheral tissues to the hypoglycemic effect of insulin. 5 Lautt's group also suggests that this mechanism can be activated by nitrates. Alternatively, in a very recent study, Reaven, the father of the concept of the insulin resistance syndrome, 24 along with his colleagues 25 found that alterations of the NO/cyclic GMP pathway seem to be an early event in non-diabetic individuals with a family history of type 2 diabetes and that these changes correlate with the degree of insulin resistance. Regarding the significance of the role of NO in the regulation of peripheral insulin sensitivity, 5 NO supplementation may serve as a novel therapeutic approach to managing insulin resistance, for example by using organic nitrates. On the other hand, recent results by McGowder et al 26 do not seem to support this assumption since they found that S-nitrosoglutathione, another NO donor, inhibited insulin release with an increase in blood glucose level in healthy normoglycemic dogs. In their study, however, the non-enzymatic NO donor at the dose applied produced a significant decrease in blood pressure; thus, besides species differences, and the difference between the NO donors used, the pronounced hypotensive response could mask the metabolic effects of the NO donor through counter-regulatory sympathetic reflex activation in dogs, which may explain the virtual contradiction between the results.
In summary, the results provide further evidence for the novel metabolic effects of nitroglycerin with a promise of being a useful drug in the management of insulin resistance especially when combined with stable angina or congestive heart failure. However, it is also shown that the development of hemodynamic nitrate tolerance may per se yield insulin resistance (ie, an independent risk factor of coronary heart disease).
Our aim was to determine whether development of tolerance to the hypotensive effect of nitroglycerin also resulted in tolerance to the insulin-sensitizing effect in conscious rabbits. We conclude that acutely nitrate patches improve insulin sensitivity whereas a 7-day chronic treatment schedule that results in hemodynamic nitrate tolerance also produces insulin resistance.
1. Palmer RM, Ferrige AG, Moncada S. Nitric oxide accounts for the biological activity of endothelium-derived relaxing factor. Nature. 1987; 327:524–526.
2. Ignarro LJ, Buga GM, Wood KS, et al. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A. 1987; 84:9265–9269.
3. Garthwaite J. Neural nitric oxide signalling. Trends Neurosci. 1995; 18:51–52.
4. Sadri P, Legare DJ, Lautt WW. Insulin resistance caused by nitric oxide synthase inhibition. Proc West Pharmacol Soc. 1997; 40:19–20.
5. Lautt WW. The HISS story overview: a novel hepatic neurohumoral regulation of peripheral insulin sensitivity in health and diabetes. Can J Physiol Pharmacol. 1999; 77:553–562.
6. Kovacs P, Szilvassy Z, Hegyi P, et al. Effect of transdermal nitroglycerin on glucose-stimulated insulin release in healthy male volunteers. Eur J Clin Invest. 2000; 30:41–44.
7. Shankar RR, Zhu JS, Ladd B, et al. Central nervous system nitric oxide synthase activity regulates insulin secretion and insulin action. J Clin Invest. 1998; 102:1403–1412.
8. Shankar RR, Wu Y, Shen H, et al. Mice with gene disruption of both endothelial and neuronal nitric oxide synthase exhibit insulin resistance. Diabetes. 2000; 49:684–687.
9. Parker JD, Parker JO. Nitrate therapy for stable angina pectoris. N Engl J Med. 1998; 338:520–531.
10. Szilvássy Z, Ferdinandy P, Bor P, et al. Loss of preconditioning in rabbits with vascular tolerance to nitroglycerin. Br J Pharmacol. 1994; 112:999–1001.
11. Ferdinandy P, Szilvassy Z, Csont T, et al. Nitroglycerin-induced direct protection of the ischaemic myocardium in isolated working hearts of rats with vascular tolerance to nitroglycerin. Br J Pharmacol. 1995; 115:1129–1131.
12. Sari R, Szilvassy Z, Jakab I, et al. Cross tolerance between nitroglycerin and neural relaxation of the sphincter of Oddi. Pharmacol Res. 1998; 37:505–512.
13. Gilbert M, Basile S, Baudelin A, et al. Lowering plasma free fatty acid levels improve ionsulin action in conscious pregnant rabbits. Am J Physiol. 1993; 264:E576–E582.
14. Mossberg KA, Taegtmeyer H. Time course of skeletal muscle glucose uptake during euglycaemic hyperinsulinaemia in anaesthetized rabbit: a fluorine-18-2-deoxy-2-fluoro-D-glucose study. J Nucl Med. 1992; 33:1523–1529.
15. Szilvassy Z, Ferdinandy P, Nagy I, et al. The effect of continuous versus intermittent treatment with transdermal nitroglycerin on pacing-induced preconditioning in conscious rabbits. Br J Pharmacol. 1997; 121:491–496.
16. Szilvassy Z, Ferdinandy P, Szilvassy J, et al. Loss of preconditioning in atherosclerotic rabbits: The role of hypercholesterolaemia. J Mol Cell Cardiol. 1995; 27:2559–2569.
17. Defronzo RA, Tobin JD, Andres RA. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol. 1979; 237:E214–E223.
18. Wallenstein S, Zucker CL, Fleiss JL. Some statistical methods useful in circulation research. Circ Res. 1980; 47:1–9.
19. Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium. Circulation. 1986; 74:1124–1136.
20. Csont T, Szilvassy Z, Fulop F, et al. Direct myocardial anti-ischaemic effect of GTN in both nitrate-tolerant and nontolerant rats: a cyclic GMP-independent activation of KATP. Br J Pharmacol. 1999; 128:1427–1434.
21. Ferdinandy P, Csont T, Csonka C, et al. Capsaicin-sensitive local sensory innervation is involved in pacing-induced preconditioning in rat hearts: role of nitric oxide and CGRP? Naunyn Schmiedebergs Arch Pharmacol. 1997; 356:356–363.
22. Oroszi G, Szilvassy Z, Nemeth J, et al. Interplay between nitric oxide and CGRP by capsaicin in isolated guinea-pig heart. Pharmacol Res. 1999; 40:125–128.
23. Oroszi G, Szilvassy Z, Nemeth J, et al. Interaction between capsaicin and nitrate tolerance in isolated guinea pig heart. Eur J Pharmacol. 1999; 368:R1–R3.
24. Reaven GM. Insulin resistance, hyperinsulinemia, hypertriglyceridemia and hypertension: parallels between human disease and rodent models. Diabetes Care. 1991; 14:195–202.
25. Piatti PM, Monti LD, Zavaroni I, et al. Alterations in nitric oxide/cyclic GMP pathway in nondiabetic siblings of patients with type 2 diabetes. J Clin Endocrinol Metab. 2000; 85:2416–2420.
26. McGowder D, Ragoobirsingh D, Dasgupta T. The hyperglycemic effect of S-nitrosoglutathione in the dog. Nitric Oxide. 1999; 3:481–491.