The mechanisms through which diabetic hyperglycemia

Transcription

1 AJH 2001; 14:126S 131S Arterial Pressure Control at the Onset of Type I Diabetes: The Role of Nitric Oxide and the Renin-Angiotensin System Michael W. Brands and Sharyn M. Fitzgerald Little is known about how hyperglycemia in diabetes directly affects renal and cardiovascular function. Therefore, we modified the streptozotocin-model of Type I diabetes in rats to enable chronic cardiovascular study at the earliest stages of diabetes, before there was time for development of vascular structural changes. We showed that the onset of diabetic hyperglycemia increased total peripheral resistance, decreased skeletal muscle blood flow, increased thromboxane production, and caused a transient increase in plasma renin activity (PRA). Mean arterial pressure (MAP) also increased, but the amplitude was modest. Moreover, we measured significant increases in glomerular filtration rate (GFR) and renal plasma flow, and also showed that endothelially mediated vasodilation in skeletal muscle was not impaired. We then tested the hypothesis that nitric oxide (NO) was playing an important role in counteracting a pressor response to the onset of The mechanisms through which diabetic hyperglycemia affects integrative function of the cardiovascular system are poorly understood. Glucose has been shown to directly modify the composition and structure of vascular tissue and thereby lead to secondary development of impaired blood flow and pressure control. In turn, altered control of systemic and regional hemodynamics has been shown to exacerbate the injury process. 1 5 Although the direct cardiovascular actions of hyperglycemia in the diabetic state are not known, deranged control of cardiovascular homeostasis remains a major cause of morbidity and mortality in diabetes. Onset of Type I Diabetes as an Experimental Tool The development of overt hyperglycemia in Type II diabetic patients generally is preceded by many years of diabetes. Our results showed that induction of diabetes in rats with chronic NO synthase inhibition caused a marked and progressive increase in MAP. In addition, PRA increased progressively under those conditions and the increase in GFR was prevented. This suggests that NO may work to keep arterial pressure in control at the onset of hyperglycemia very early in the development of diabetes, possibly by facilitating renal vasodilation and by suppressing activity of the renin-angiotensin system. However, the mechanisms for these interactions and the role of renal vascular resistance and other factors in mediating the hypertensive response remain unknown. Am J Hypertens 2001;14:126S 131S 2001 American Journal of Hypertension, Ltd. Key Words: GFR, renal vascular resistance, renin, hypertension. progressive obesity, insulin resistance, and rising blood pressure; therefore, there are many factors likely to confound assessment of the direct circulatory effects of hyperglycemia. Type I diabetes provides certain advantages relating to the study of cardiovascular actions of hyperglycemia. However, determining the cardiovascular effects of hyperglycemia has been complicated because studies of circulatory function in diabetes generally have focused on the established or later phases, and there is a continually increasing likelihood of being confounded by structural changes in the vasculature. An alternative approach is to study these relationships before there has been time for structural changes in vascular tissues to develop, but it is nearly impossible to study diabetes in human subjects at a stage early enough to be certain that structural changes have not begun, much less to obtain prediabetic cardiovascular control data in those persons. In addition, there is a surprising lack of information in experimental animals Received February 26, Accepted March 6, From the Department of Physiology (MWB), Medical College of Georgia, Augusta, Georgia, and Department of Physiology (SMF), University of Mississippi Medical Center, Jackson, Mississippi. This work was supported by Heart, Lung and Blood Institute Grants HL and HL and conducted during the tenure of an Established Investigator Grant from the American Heart Association (AHA) and Genentech. SMF is the recipient of a Postdoctoral Fellowship award from the Mississippi affiliate of the AHA Southern Research Consortium. Address correspondence and reprint requests to Dr. Michael W. Brands, Department of Physiology, Medical College of Georgia, Augusta, GA ; mbrands@mail.mcg.edu /01/$ by the American Journal of Hypertension, Ltd. PII S (01) Published by Elsevier Science Inc.

2 AJH June 2001 VOL. 14, NO. 6, PART 2 ARTERIAL PRESSURE CONTROL AND TYPE I DIABETES 127S FIG. 1. Daily mean arterial pressure averaged from 15 rats during the pre-streptozotocin (STZ) control period (C), during the insulin replacement control periods (IR) in which normoglycemia was established, and during 4-day diabetic periods (D) that were induced by withdrawing the insulin replacement therapy. earlier than 3 to 4 weeks after induction of diabetes, and previous attempts at early-stage studies in animals have not been able to control for potential side-effects of the inducing agent (eg, streptozotocin [STZ]). To determine the cardiovascular and renal responses to hyperglycemia in diabetes, while addressing these issues we modified the STZ model of Type I diabetes, using chronically instrumented rats. 6,7 Our model enabled studies with chronic, continuous hemodynamic measurements in the same animal during control conditions through the period immediately after induction of diabetes and beyond, while controlling for any effects of STZ not related to its effect to decrease insulin levels. This is accomplished by providing continuous intravenous insulin replacement beginning the morning after STZ administration, and adjusting the dose as needed to maintain normoglycemia. After approximately 1 week, diabetes is initiated by removing insulin from the intravenous infusion. Thus, diabetes can be turned on and off as needed, and the degree of hyperglycemia regulated by adjusting the intravenous insulin infusion, while continuously measuring systemic and regional cardiovascular function. 6,7 Renal and Cardiovascular Responses to Onset of Diabetes With this new experimental model, we reported that induction of poor glycemic control increased mean arterial pressure 6 (Fig. 1). Despite the modest amplitude of the pressure rise, several additional features of the response were noteworthy. First, the increase in pressure was rapid in onset, consistent with peripheral vasoconstriction. Second, the increase occurred despite significant urinary sodium and volume losses: the increase in arterial pressure may have contributed to the natriuresis, but it still could be predicted that elimination of the urinary losses may have yielded a greater increase in arterial pressure. Third, arterial pressure decreased rapidly with restoration of good glycemic control and reversed, rising with equal rapidity, with onset of a second diabetic period. 6 Those results indicated that the onset of diabetes could affect systemic hemodynamics directly, and also suggested there was underlying vasoconstriction. That was supported by a subsequent study in which we measured cardiac output and arterial pressure continuously, 24 h/day. 7 Cardiac output decreased progressively during the first week of diabetes, due in part to the significant decrease in sodium balance. Total peripheral resistance increased markedly, however, and the changes in cardiac output and sodium balance during the diabetic and recovery periods suggested that there was a vasoconstrictor influence associated with the period of poor glycemic control. In another study, 24-h/ day measurement of hindquarter blood flow revealed that skeletal muscle was a primary contributor to the increased peripheral vascular resistance. 8 In addition, we measured significant increases in plasma renin activity (PRA) and urinary thromboxane excretion during the first week of diabetes, providing possible mechanisms for the increased vascular tone. 7 Thus, there was good evidence for vasoconstriction as well as potential humoral mediators identified. However, in light of these changes, it became interesting to question why the increase in mean arterial pressure was so modest. Role of Nitric Oxide There is considerable evidence that endothelially mediated vasodilation in particular, nitric oxide (NO) dependent

3 128S ARTERIAL PRESSURE CONTROL AND TYPE I DIABETES AJH June 2001 VOL. 14, NO. 6, PART 2 FIG. 2. Percent change in hindquarter blood flow from baseline value (first Saline period). The acute flow responses were measured on days 2 and 6 of a 6-day diabetic period and during the insulin replacement periods before (Control) and after (Recovery) the diabetic period. The saline and acetylcholine (ACh) periods were 20 min each and in sequence. The two ACh doses were 1 and 10 g/min, intra-arterially. dilation is impaired in diabetes 9 17 ; but the activity of the NO system at the onset of diabetes, especially in the whole animal, has not been thoroughly addressed. We determined the function of the system on days 2 and 6 after induction of diabetes, by infusing acetylcholine directly into the hindquarters of conscious rats that were chronically instrumented with an arterial infusion line and a flow probe at the iliac bifurcation. 8 The results provided no evidence for impaired vasodilation; in fact, the vasodilatory response tended to increase over the 6-day diabetic period (Fig. 2). This suggests the NO system is active and perhaps even stimulated during the initial stages of diabetes, consistent with other reports in the literature. 15,16,18 25 One possibility, therefore, is that the actions of the NO system at the onset of diabetes are important in blunting the tendency for blood pressure to increase. Thus, in the face of the skeletal muscle vasoconstriction and the increases in PRA and thromboxane production, NO may be essential to keep blood pressure in relative control. One mechanism may be through increasing renal sodium excretory capability by facilitating an increase in the glomerular filtration rate (GFR). We measured significant increases in GFR and renal plasma flow during the onset of diabetes that promptly returned to control levels upon restoration of normoglycemia during the recovery period. 7 This change would shift the renal pressure natriuresis relationship to lower blood pressures and would tend to counteract other pressor stimuli, thus providing a potential mechanism through which NO could exert its protective action. Induction of Diabetes Causes Marked Hypertension When Nitric Oxide Synthesis Is Inhibited To test the hypothesis that the NO system serves to attenuate a pressor response to hyperglycemia at the onset of diabetes, we induced diabetes in rats in which NO synthesis was inhibited chronically with N- -nitro-l-arginine methyl ester (L-NAME). 26 As shown in Fig. 3, mean arterial pressure (MAP) in rats measured 24 hr/day increased progressively during a 3-week diabetic period. The increase was approximately 60 mm Hg greater than the mild change measured in the control diabetic rats and 20 mm Hg greater than the effect of L-NAME alone in normal rats. Thus, diabetes and impaired NO synthesis combined to cause a potentiated increase in MAP. This suggests that the NO system is even more important at the onset of diabetes than it is in the normal state for maintaining blood pressure. In addition, an important role for GFR was supported by the observation that the hyperfiltration that occurred in the normal diabetic rats was prevented in the diabetic rats with chronic L-NAME treatment. 26 Important Role for Angiotensin II In addition to the correlation between the control of GFR and the changes in arterial pressure in diabetes, under conditions of freely functioning and chronically inhibited NO synthesis, very interesting changes in the activity of the renin-angiotensin system occurred that also tracked closely with the blood pressure responses. 26 Figure 4 shows, in the vehicle-treated rats, that the induction of

4 AJH June 2001 VOL. 14, NO. 6, PART 2 ARTERIAL PRESSURE CONTROL AND TYPE I DIABETES 129S FIG. 3. Mean arterial pressure response to induction of diabetes in normal rats (filled circles) and in rats treated chronically with L-NAME intravenously (open squares). Plot with closed squares represents effect of chronic L-NAME treatment in normal nondiabetic rats. diabetes caused a biphasic response in plasma renin activity (PRA). This confirms an earlier study at the onset of diabetes, 27 and is interesting for several reasons. First, it confirms our earlier report that the first week of diabetes causes an increase in PRA. In addition, the measurement of PRA back at control levels by week 3 is consistent with the vast majority of studies that report renin-angiotensin system activity is normal or suppressed in diabetes. It is important to note, in that regard, that most measurements in STZ-treated rats are not begun until 2 to 3 weeks after administration of STZ; and measurements in human subjects, of course, are made much longer after onset. Thus, these serial measurements over a control period and the first 3 weeks of diabetes show that there is a biphasic response to the induction of poor glycemic control, 26 and help to explain why there often is a discrepancy in reporting of PRA in this model of diabetes. The increase in PRA when diabetes was induced under conditions of chronic NO synthesis inhibition also suggests that the system is important in controlling blood pressure in diabetes. Evidence that the changes in PRA FIG. 4. Plasma renin activity measured during control period and once weekly during the 3-week diabetic period in the same rats depicted in Fig. 3. Note that the control period was the period in which the L-NAME rats had already begun chronic L-NAME treatment but diabetes had not yet been induced in any group.

5 130S ARTERIAL PRESSURE CONTROL AND TYPE I DIABETES AJH June 2001 VOL. 14, NO. 6, PART 2 FIG. 5. Relationships between arterial pressure, angiotensin II, and nitric oxide during the first 3 weeks of hyperglycemia in the streptozotocin (STZ) model of Type I diabetes in rats. Dashed lines represents responses measured when diabetes is induced in rats treated chronically with the nitric oxide synthase inhibitor L-NAME. Blood pressure and angiotensin II show increased dependence on nitric oxide to remain within normal limits after onset of hyperglycemia, but it is not known yet whether nitric oxide actually is increased initially. were not compensatory for the changes in blood pressure is that the group with the highest blood pressure, ie, the L-NAME plus diabetes group, had the greatest increase in PRA. This suggests the renin-angiotensin system may play a role in mediating the marked hypertension. Thus, when the NO system was inhibited, PRA continued to increase throughout the first 3 weeks of diabetes, and MAP increased progressively as well. On the other hand, when we did not interfere with NO synthesis, ie, in the vehicletreated rats, the induction of diabetes caused a biphasic change in PRA and only a modest and stable increase in MAP (Fig. 5). This raises the question of whether the NO system at the onset of diabetes is required to suppress a stimulatory influence on the renin-angiotensin system and thus to keep arterial pressure within relatively normal limits. Further study is needed to address that question, but we have preliminary evidence that the sympathetic nervous system may be involved in stimulating the increase in PRA when NO synthesis is inhibited. It also is not known how changes in renin-angiotensin system activity are involved in the arterial pressure response. One potential mechanism could be through the control of renal vascular resistance, with the lack of hyperfiltration in the L-NAME treated diabetic rats, possibly because of afferent arteriolar constriction caused by the combination of increased angiotensin II levels and decreased NO production. This mechanism and others, however, require further study. peripheral resistance, and skeletal muscle vascular resistance, and that there is no decrease in endothelially dependent vasodilation in the hindquarters (ie, skeletal muscle). These changes are associated with a modest increase in mean arterial pressure; and when normoglycemia is restored, blood pressure and renin-angiotensin system activity return to control levels. However, the onset of diabetic hyperglycemia in rats with chronic NO synthesis inhibition causes a progressive increase in mean arterial pressure to malignant hypertensive levels; the hypertension is more than the additive effects of L-NAME and diabetes alone. The mechanisms mediating the hypertensive response are not known, but the diabetic rats with chronic NO synthesis inhibition had a significantly lower GFR than did the normal diabetic rats, and also had a progressive increase in PRA. This suggests that the control of renal vascular resistance, angiotensin II levels, and blood pressure depends critically on nitric oxide during hyperglycemia early in diabetes, but the mechanisms for those interactions and how they change as endothelial damage develops further along in the disease process remain unknown. Acknowledgment We thank Allison Hailman for technical assistance. References Summary Our recent studies have shown that hyperglycemia early in diabetes increases renin-angiotensin system activity, total 1. Tuck ML, Stern N: Diabetes and hypertension. J Cardio Pharmacol 1992;19(suppl 6):S8 S Steffes MW, Brown DM, Mauer SM: Diabetic glomerulopathy following unilateral nephrectomy in the rat. Diabetes 1978;27:35 41.

6 AJH June 2001 VOL. 14, NO. 6, PART 2 ARTERIAL PRESSURE CONTROL AND TYPE I DIABETES 131S 3. Ritz E, Fliser D, Nowicki M: Hypertension and vascular disease as complications of diabetes, in Laragh JH, Brenner BM (eds): Hypertension: Pathophysiology, Diagnosis, and Management. Raven Press, New York, 1995, pp Mogensen CE: Management of the diabetic patient with elevated blood pressure or renal disease, in Laragh JH, Brenner BM (eds): Hypertension: Pathophysiology, Diagnosis, and Management. Raven Press, New York, 1995, pp Hostetter TH: Diabetic nephropathy, in Brenner BM, Rector FC Jr (eds): The Kidney. WB Saunders, Philadelphia, 1991, pp Brands MW, Hopkins TE: Poor glycemic control induces hypertension in diabetes mellitus. Hypertension 1996;27: Brands MW, Fitzgerald SM, Hewitt WH, Hailman AE: Decreased cardiac output at the onset of diabetes: renal mechanisms and peripheral vasoconstriction. Am J Physiol 2000;278:E917 E Brands MW, Fitzgerald SM: Acute endothelium-mediated vasodilation is not impaired at the onset of diabetes. Hypertension 1998; 32: Poston L, Taylor PD: Endothelium-mediated vascular function in insulin-dependent diabetes mellitus. Clin Sci 1995;88: Bohlen HG, Lash JM: Topical hyperglycemia rapidly suppresses EDRF-mediated vaosdilation of normal rat arterioles. Am J Physiol 1993;265:H219 H Vlassara H, Fuh H, Makita Z, Krungkrai S, Cerami A, Bucala R: Exogenous advanced glycosylation end products induce complex vascular dysfunction in normal animals: a model for diabetic and aging complications. Proc Natl Acad Sci USA 1992;89: Craven PA, Studer RK, DeRubertis FR: Impaired nitric oxidedependent cyclic guanosine monophosphate generation in glomeruli from diabetic rats. J Clin Invest 1994;93: Ting HH, Timimi FK, Boles KS, Creager SJ, Ganz P, Creager MA: Vitamin C improves endothelium-dependent vasodilation in patients with non-insulin-dependent diabetes mellitus. J Clin Invest 1996; 97: Pieper GM: Review of alterations in endothelial nitric oxide production in diabetes. Hypertension 1998;31: Graier WF, Posch K, Wascher TC, Kostner GM: Role of superoxide anions in changes of endothelial vasoactive response during acute hyperglycemia. Hormone Metab Res 1997;29: Cohen RA: Dysfunction of vascular endothelium. Circulation 1993; 87(suppl V):V-67 V Kiff RJ, Gardiner SM, Compton AM, Bennett T: Selective impairment of hindquarters vasodilator reponses to bradykinin in conscious Wistar rats with streptozotocin-induced diabetes mellitus. Br J Pharmacol 1991;103: Cosentino F, Hishikawa K, Katusic ZS, Luscher TF: High glucose increases nitric oxide synthase expression and superoxide anion generation in human aortic endothelial cells. Circulation 1997;96: Craven PA, DeRubertis FR, Melhem M: Nitric oxide in diabetic nephropathy. Kidney Int 1997;52(suppl 60):S46 S Graier WF, Simecek S, Hoebel BG, Wascher TC, Dittrich P, Kostner GM: Antioxidants prevent high-d-glucose-enhanced endothelial Ca2 /cgmp response by scavenging superoxide anions. Eur J Pharmacol 1997;322: Graier WF, Simecek S, Kukovetz WR, Kostner GM: High D- glucose-induced changes in endothelial Ca2 /EDRF signaling are due to generation of superoxide anions. Diabetes 1996;45: Graier WF, Wascher TC, Lackner L, Toplak H, Krejs GJ, Kukovetz WR: Exposure to elevated D-glucose concentrations modulates vascular endothelial cell vasodilatory response. Diabetes 1993;42: Wascher TC, Toplak H, Krejs GJ, Simecek S, Kukovetz WR, Graier WF: Intracellular mechanisms involved in D-glucose-mediated amplification of agonist-induced Ca2 response and EDRF formation in vascular endothelial cells. Diabetes 1994;43: Cohen RA: The potential clinical impact of 20 years of nitric oxide research. Am J Physiol 1999;276:H1404 H Mattar AL, Fujihara CK, Ribeiro MO, De Nucci G, Zatz R: Renal effects of acute and chronic nitric oxide inhibition in experimental diabetes. Nephron 1996;74: Fitzgerald SM, Brands MW: Nitric oxide may be required to prevent hypertension at the onset of diabetes. Am J Physiol Endocrinol Metab 2000;279:E762 E Kikkawa R, Kitamura E, Haneda M, Shigeta Y: Biphasic alteration of the renin-angiotensin-aldosterone system in streptozotocin-diabetic rats. Ren Physiol 1986;9: