Upregulation of Phosphodiesterase 1A1 Expression Is Associated With the Development of Nitrate Tolerance
Background The efficacy of nitroglycerin (NTG) as a vasodilator is limited by tolerance, which develops shortly after treatment begins. In vascular smooth muscle cells (VSMCs), NTG is denitrated to form nitric oxide (NO), which activates guanylyl cyclase and generates cGMP. cGMP plays a key role in nitrate-induced vasodilation by reducing intracellular Ca2+ concentration. Therefore, one possible mechanism for development of nitrate tolerance would be increased activity of the cGMP phosphodiesterase (PDE), which decreases cGMP levels.
Methods and Results To test this hypothesis, rats were made tolerant by continuous infusion of NTG for 3 days (10 μg · kg−1 · min−1 SC) with an osmotic pump. Analysis of PDE activities showed an increased function of Ca2+/calmodulin (CaM)–stimulated PDE (PDE1A1), which preferentially hydrolyzes cGMP after NTG treatment. Western blot analysis for the Ca2+/CaM-stimulated PDE revealed that PDE1A1 was increased 2.3-fold in NTG-tolerant rat aortas. Increased PDE1A1 was due to mRNA upregulation as measured by relative quantitative reverse transcription–polymerase chain reaction. The PDE1-specific inhibitor vinpocetine partially restored the sensitivity of the tolerant vasculature to subsequent NTG exposure. In cultured rat aortic VSMCs, angiotensin II (Ang II) increased PDE1A1 activity, and vinpocetine blocked the effect of Ang II on decrease in cGMP accumulation.
Conclusions Induction of PDE1A1 in nitrate-tolerant vessels may be one mechanism by which NO/cGMP-mediated vasodilation is desensitized and Ca2+-mediated vasoconstriction is supersensitized. Inhibiting PDE1A1 expression and/or activity could be a novel therapeutic approach to limit nitrate tolerance.
Received May 11, 2001; revision received August 16, 2001; accepted August 22, 2001.
Nitroglycerin (NTG) remains one of the foremost drugs in the treatment of angina pectoris. When given in the short term, NTG has potent vasodilator capacities on arteries, veins, and coronary collateral vessels.1 NTG induces vasorelaxation by releasing NO. NO activates soluble guanylyl cyclase and subsequently increases cGMP. cGMP in turn activates a cGMP-dependent protein kinase that has been shown to mediate vasorelaxation via phosphorylation of proteins that regulate intracellular Ca2+ levels.2 The efficacy of chronic NTG administration, however, is limited by the rapid development of tolerance,3 involving decreased vascular sensitivity and diminished cGMP elevations in vascular smooth muscle cells (VSMCs) in response to continued nitrate treatment. Nitrate tolerance is also associated with cross-tolerance to other endothelium-dependent and -independent vasodilators.2 Several mechanisms have been proposed to account for this phenomenon, such as neurohumoral counterregulation4 or mechanisms intrinsic to the vascular tissue itself, such as desensitizing NO/cGMP signaling and responses.2 Chronic NTG treatment has also been shown to be associated with an increase in sensitivity to vasoconstrictors such as catecholamines, angiotensin (Ang) II, KCl, and serotonin,5 all of which may compromise the vasodilating capacity of NTG, thereby contributing to tolerance. Another aspect of organic nitrate therapy is the development of rebound ischemia due to coronary rebound constriction after abrupt cessation of long-term NTG therapy.6
See p 2263
cGMP appears to be a key player in NTG-mediated vasorelaxation, and the cGMP level is highly regulated by phosphodiesterase (PDE). At least 4 different major PDE activities have been identified in VSMCs (Table). Different PDEs play distinct roles in controlling vascular tone.7 In this study, we found that the activity and expression of Ca2+/calmodulin (CaM)–stimulated PDE (PDE1A1), which preferentially hydrolyzes cGMP, were induced by NTG treatment. Vinpocetine, a selective inhibitor of the PDE1 family, was able to partially restore the sensitivity of tolerant vessels to subsequent NTG exposure. These findings indicate that increased PDE1A1 activity can decrease cGMP levels, which may contribute at least in part to the attenuation of the NTG-mediated vasodilation as well as the supersensitivity to vasoconstrictors.
Development of Nitrate Tolerance In Vivo
Nitrate tolerance was induced as described previously.8 Male Sprague-Dawley rats (Charles River Laboratories, Wilmington, Mass; 250 to 300 g) were anesthetized with ketamine 40 mg/kg, xylazine 0.5 mg/kg, and acepromazine 5 mg/kg IP. An osmotic minipump (model 2 ML1, Alza Corp) filled with either NTG (Zeneca Inc) or vehicle (propylene glycol) was implanted subcutaneously at the dorsum of the neck. NTG was infused at a rate of 10 μg · kg−1 · min−1 for 3 days.
To assess tolerance, mean arterial pressure (MAP) was continuously monitored by a catheter in the right common carotid artery on a polygraph recorder (Grass Instruments). Bolus doses of NTG or hydralazine were administered through the right jugular vein.
Three thoracic aortas were pooled and homogenized. The supernatant of the tissue extract was fractionated as described previously.9 Fractions containing PDEs were assayed for PDE activity or subjected to Western blot analysis.
In Vitro PDE Enzyme Assay
PDE assays were carried out according to established procedures.9
Western Blot Analysis
Western blots for PDE1A1 and PDE5A1 with isoform-specific antibodies were carried out as described previously.9,10
Relative Quantitative RT-PCR and Northern Blot Analysis
Tissue RNA was extracted with a Total RNA Isolation Kit (Ambion). First-strand cDNA was synthesized with the SuperScript Preamplification System (Gibco-BRL). Relative quantitative reverse transcription–polymerase chain reaction (RT-PCR) was performed with 18s rRNA as an internal control by use of Ambion’s competimer technology. Isoform-specific primers were used to generate the 281-bp PCR product for PDE1A1 or 335-bp PCR product for PDE5A1. PCR products were separated on a 2% agarose gel, stained with ethidium bromide, and visualized under UV light. The relative intensity of PDE1A1 or PDE5A1 PCR products was determined by densitometry. Northern blot analysis was performed with the Clontech hybridization system according to the manufacturer’s protocol. The PCR product specific to PDE1A1 or PDE5A1, mentioned above, was used as a probe.
Ex vivo evaluation of nitrate tolerance and reversal was performed as described previously.11 Aortic rings were stretched with a preload of 2 g, equilibrated, and then preconstricted with phenylephrine. Preconstricted aortic rings were pretreated with vehicle or 100 μmol/L vinpocetine for 10 minutes, followed by exposure to ascending concentrations of NTG.
Aortic extracts were prepared by homogenization in cold 5% trichloroacetic acid. VSMC extracts were prepared by lysis of cells in cold 100% ethanol. A cGMP 125I radioimmunoassay kit (NEN) was used for measurement of cGMP levels as described in the manufacturer’s protocol. Protein levels were determined by the method of Bradford (Bio-Rad).
Rat aortic VSMCs were isolated and maintained as described previously.12 VSMCs (passages 6 to 9) were growth-arrested for 48 hours before the indicated drug treatment.
Measurement of Ca2+/CaM-Stimulated PDE Activity
The technique involves extraction and assay of enzyme activity at a low temperature and in the presence of trifluoperazine to prevent the dissociation and reassociation of the ternary Ca2+-CaM-PDE complex during cell disruption.13 The percent maximal CaM-stimulated PDE activity was computed as (activity without CaM or EGTA−activity with EGTA)/(activity with CaM−activity with EGTA)×100.
One- or 2-way ANOVA was used to compare differences between treatment means (expressed as mean±SEM). After ANOVA, comparison of 2 populations was made by Student’s unpaired t test. A value of P<0.05 was considered significant.
Induction of In Vivo Nitrate Tolerance
To assess the development of nitrate tolerance in vivo, the hypotensive and vasorelaxant effects of NTG were examined. There was no significant difference in baseline MAP between the vehicle- and NTG-treated groups (104.3±3.9 versus 100.8±4.1 mm Hg). In the vehicle-treated group, acute NTG challenges caused a dose-dependent drop in MAP (Figure 1), but 3-day NTG treatment significantly decreased the NTG-induced drop in MAP (Figure 1), indicating that changes in MAP were significantly blunted in the NTG-treated groups. The hydralazine (1 mg/kg bolus)–induced drop in MAP was similar in both groups (Figure 1), indicating that NO-independent vasodilation was not different between the 2 groups.
Increased Ca2+/CaM-Stimulated PDE Activity in Nitrate-Tolerant Rat Aortas
To examine the change of PDE activity in nitrate-tolerant vessels, we performed PDE assays. cGMP hydrolytic activities in tolerant aortas were significantly higher than in control aortas (Figure 2A), whereas cAMP hydrolytic activities were similar (Figure 2B). In the presence of Ca2+/CaM, cGMP hydrolytic activities were increased much more in tolerant aortas than in control aortas (Figure 2A). This observation suggests that the activity of a Ca2+/CaM-stimulated PDE, which preferentially hydrolyzes cGMP, is induced by NTG treatment.
To confirm these results and further separate different PDE activities, we resolved the PDE activities of control and tolerant aortas by high-performance liquid chromatography. In fractions 18 to 30, which contain enzymes of both the PDE1 and PDE5 families,9 cGMP hydrolytic activity of tolerant aortas is significantly higher in the presence of Ca2+/CaM than in the presence of EGTA (Figure 2C). In contrast, fractions 18 to 30 from control vessels hydrolyzed cGMP to a similar extent in the presence of EGTA or Ca2+/CaM (Figure 2D). These data indicate that increased cGMP hydrolytic activity in tolerant vessels is mainly the Ca2+/CaM-stimulated PDE activity.
Protein Level of PDE1A1, but Not PDE5A1, Is Increased in Nitrate-Tolerant Aortas
To determine whether the induction of PDE activity in tolerant vessels was due to increases in the enzyme level, Western blot analyses with PDE1A- and PDE5A-specific antibodies were performed (Figure 3). The PDE1A1 protein level was increased 2.3-fold in tolerant vessels (Figure 3A), whereas the PDE5A1 protein level was not significantly changed (Figure 3B). These results indicate that PDE1A1, a Ca2+/CaM-stimulated PDE with much higher affinity for cGMP than for cAMP,14 is selectively upregulated in nitrate-tolerant vessels.
mRNA Level of PDE1A1, but Not PDE5A1, Is Increased in Nitrate-Tolerant Aortas
To determine whether the increase in protein level of PDE1A1 was due to an increase in its mRNA level, relative quantitative RT-PCRs using PDE1A1- and PDE5A1-specific primers were performed (Figure 4). mRNA levels of PDE1A1 in tolerant aortas were increased by 2.4-fold compared with control aortas (Figure 4A). In contrast, the PDE5A1 mRNA level was not significantly altered (Figure 4B). We obtained similar results by Northern blot analysis (Figure 4C).
Effects of PDE1 Family–Selective Inhibitor Vinpocetine on Reversal of Nitrate Tolerance Ex Vivo
NTG-treated aortic rings exhibited a significantly decreased vasorelaxation response to subsequent NTG compared with control. Preincubation of tolerant rings with vinpocetine, however, restored the vasorelaxant effect of NTG. Vinpocetine also potentiated the vasorelaxant effect of NTG in control rings (Figure 5A).
We also tested the effects of vinpocetine on nitrate tolerance–associated attenuation of cGMP elevations in response to subsequent NTG exposure as shown in Figure 5B. The basal cGMP concentrations were not significantly different between tolerant and control aortic rings. After exposure to NTG, an NTG-induced increase in cGMP in tolerant rings was clearly decreased compared with control, and vinpocetine enhanced the NTG-induced increase in cGMP in tolerant rings, suggesting that PDE1A1 inhibitor is able to partially reverse the nitrate tolerance of cGMP response.
Stimulation of PDE1A1 Activity by Ang II and the Role of PDE1A1 in Ang II–Mediated Inhibition of Atrial Natriuretic Peptide–Evoked cGMP Accumulation in VSMCs
To study the mechanism by which induction of PDE1A1 might cause nitrate tolerance and nitrate tolerance–induced supersensitivity to vasoconstrictors, we measured the effects of Ang II on PDE1A1 activity, which represent the extent of activation in vivo. As shown in Figure 6A, PDE1A1 activity in control VSMCs is 47.8% of the maximum. Ang II increased PDE1A1 activity to 95.6% of the maximum. This observation is consistent with the time course of force production and changes in intracellular Ca2+ concentration in response to Ang II treatment15 and suggests that Ang II is able to stimulate the PDE1A1 activity, probably via an Ang II– mediated increase in Ca2+ concentration.
To further determine the functional effects of PDE1A1 activity stimulated by Ang II, we measured the effect of Ang II on cGMP accumulation in response to atrial natriuretic peptide (ANP). The addition of ANP to VSMCs rapidly increased intracellular cGMP. Simultaneous addition of Ang II markedly decreased the ANP-induced cGMP accumulation. Vinpocetine significantly blocked the inhibitory effect of Ang II (Figure 6B). These results suggest that PDE1A1 in VSMCs plays a major role in mediating inhibition of ANP-evoked cGMP accumulation by Ang II.
The present study showed, first, that in vivo NTG treatment to induce hemodynamic nitrate tolerance increased PDE1A1 enzyme activity, protein levels, and mRNA expression; second, that inhibition of PDE1A1 in tolerant vessels restored vasorelaxation and cGMP response to subsequent NTG exposure; and third, that PDE1A1 plays an important role in regulation of intracellular cGMP levels in response to vasodilators and vasoconstrictors. The upregulation of PDE1A1 may therefore provide a new mechanism to explain, at least in part, the decreased sensitivity of the vasculature to NTG and the phenomenon of enhanced vasoconstriction observed after chronic NTG treatment (Figure 7).
Role of PDE1A1 in Nitrate Tolerance
PDE1A1 was previously shown to be important for the regulation of vascular cGMP levels and reactivity.7,16 The findings of the present study concur with this concept. We found that NTG treatment significantly increases PDE1A1 activity and expression. An increase in PDE1A1 activity will reduce cGMP accumulation, thereby causing the decreased sensitivity of the vasculature to subsequent NTG treatment, because cGMP is a key player in regulation of vascular relaxation by decreasing intracellular Ca2+ concentration via activation of downstream cGMP-dependent protein kinase-I.
Changes in PDE1A1 expression in intact aortas after NTG treatment most likely occurs in VSMCs, because PDE1A1 immunoreactivity was restricted to the VSMC layer in vessels, as shown by immunocytochemical analysis.10 Endothelium has been believed to be important in the genesis of tolerance, because removal of endothelium reduces nitrate tolerance significantly, although not completely.17 The fact that PDE1A1 is stimulated by vasoactive reagents released from endothelium, such as endothelin-1 and local Ang II, suggests that the effect of PDE1A1 is much greater in the presence of endothelium than in the absence of endothelium, which is consistent with the observation that nitrate tolerance is much more severe with endothelium than without and demonstrates the importance of PDE1A1 in tolerance. In addition, many previously published studies in which NTG was administered in vivo have demonstrated cross-tolerance to endothelium-dependent vasodilators and other nitrovasodilators.17–19 Induction of PDE1A1 activity also fits the presence of cross-tolerance in the setting of NTG tolerance.
An induction of PDE1A1 would also explain, at least in part, the phenomenon of supersensitivity to vasoconstrictors in NTG-tolerant vessels observed in experimental animals and human studies.5,18 Most vasoconstrictors, such as norepinephrine, Ang II, and endothelin-1, increase the intracellular Ca2+. PDE1A1 activity can be stimulated by Ca2+ >10-fold in vitro.14 Our observations that Ang II stimulated PDE1A1 activity and PDE1A1 inhibitor blocked Ang II–mediated attenuation of ANP-evoked cGMP accumulation support the role of PDE1A1 in vasoconstrictor-mediated regulation of vascular cGMP levels. Such an attenuation of ANP- or NTG-evoked cGMP accumulation has also been observed with many other Ca2+-raising vasoconstrictors.18,20 These observations together suggest that the vasoconstriction effects of these vasoconstrictors are at least partially mediated by blockade of the vasodilatory pathway of cGMP and support the idea that induction of PDE1A1 increases the magnitude of vasoconstrictor-mediated attenuation of cGMP accumulation, which may explain the phenomenon of supersensitivity to vasoconstrictors in NTG-tolerant vessels.
Rebound angina, associated with an intermittent-treatment regimen for the prevention of nitrate tolerance, can be a serious problem and may result in death.21 This rebound effect is believed to be associated with increased sensitivity to vasoconstrictors.6 Thus, supersensitivity to vasoconstrictors mediated by induction of PDE1A1 and the subsequent reduced vasodilatory forces on removal of nitrates together may explain the rebound phenomenon. Inhibition of PDE1A1 may be a novel therapeutic tool for the prevention of the rebound phenomenon.
Mechanisms Underlying Nitrate Tolerance
Nitrate tolerance seems to be multifactorial. Several mechanisms have been proposed to explain the phenomenon of nitrate tolerance. As a vasodilator, NTG lowers blood pressure, which in turn activates neurohumoral counterregulatory mechanisms, such as an activation of the renin-angiotensin system, increases in vasopressin levels, intravascular volume expansion, and increases in catecholamine release. These circulating vasoconstrictor forces may limit NTG-mediated vasodilation (pseudotolerance). Induction of PDE1A1 would enhance the vasoconstrictor forces.
Studies with isolated vessels from NTG-treated animals also show decreased sensitivity to NTG in the absence of the neurohumoral environment, pointing to intrinsic abnormalities of the tolerant vasculature itself. In particular, multiple steps in NO/cGMP signaling have been found to be affected. These include impaired nitrate biotransformation22; overproduction of reactive oxygen species, which reduces NO bioavailability17; decrease and increase of cGMP metabolic enzymes, such as guanylyl cyclase and PDE, respectively19,23; and attenuation of the downstream cGMP-dependent protein kinase activity.24 Our observation that inhibition of PDE1A1 partially restores the cGMP elevation and vasorelaxation to subsequent NTG exposure in tolerant vessels is consistent with the proposal that nitrate tolerance is contributed to by multiple factors.
Effects of PDE Inhibition on Tolerance
PDE1 isoforms are structurally closer to PDE5 isoforms than any other cAMP-hydrolyzing PDE isoforms in VSMCs. Zaprinast, which inhibits both PDE1 and PDE5 isoforms with higher sensitivity to PDE5, has been shown to be able to reverse NTG tolerance in vitro and in vivo.11,25 The effect of zaprinast on the reversal of nitrate tolerance may be due to the inhibition of both PDE1A1 and PDE5A1. Inhibition of PDE5A1 activity is able to diminish nitrate tolerance because of augmentation of the response to organic nitrates, even though PDE5A1 expression is not altered in the setting of nitrate tolerance.
Vinpocetine is the most selective inhibitor identified to date for PDE1 isozymes. Using recombinant PDE1A1 and PDE5A1 expressed in COS-7 cells, we found that vinpocetine at 100 μmol/L inhibited PDE1A1 activity up to 90% but PDE5A1 activity <5% (data not shown). This is consistent with the previous observation that the IC50 for vinpocetine to inhibit the partially purified PDE5A1 was >1 mmol/L.16 Thus, vinpocetine at a concentration of 100 μmol/L appears to be specific to PDE1A1 in VSMCs. In this study, we have demonstrated the ability of vinpocetine to limit nitrate tolerance in both restoration of vasorelaxation and cGMP response to NTG exposure in tolerant vessels, which suggests that PDE1A1 plays an important role in regulation of vascular reactivity, and also that upregulation of PDE1A1 due to chronic NTG treatment at least partially contributes to nitrate tolerance.
This work was supported in part by National Institutes of Health grant HL-63462 (to Dr Berk), American Heart Association Research Grant 0030302T (to Dr Yan), and a medical school grant from Merck to Dr Berk. Dr Kim was partially supported by Yonsei University, Seoul, Korea. We thank Dr Matthew G. Melaragno and Amy M. Mohan for technical support on animal surgeries.
Munzel T, Giaid A, Kurz S, et al. Evidence for a role of endothelin 1 and protein kinase C in nitroglycerin tolerance. Proc Natl Acad Sci U S A. 1995; 92: 5244–5248.
Munzel T, Li H, Mollnau H, et al. Effects of long-term nitroglycerin treatment on endothelial nitric oxide synthase (NOS III) gene expression, NOS III–mediated superoxide production, and vascular NO bioavailability. Circ Res. 2000; 86: E7–E12.
Sonnenburg WK, Seger D, Beavo JA. Molecular cloning of a cDNA encoding the “61-kDa” calmodulin-stimulated cyclic nucleotide phosphodiesterase: tissue-specific expression of structurally related isoforms. J Biol Chem. 1993; 268: 645–652.
Munzel T, Sayegh H, Freeman BA, et al. Evidence for enhanced vascular superoxide anion production in nitrate tolerance: a novel mechanism underlying tolerance and cross-tolerance. J Clin Invest. 1995; 95: 187–194.
Lange RL, Reid MS, Tresch DD, et al. Nonatheromatous ischemic heart disease following withdrawal from chronic industrial nitroglycerin exposure. Circulation. 1972; 46: 666–678.
Brien JF, McLaughlin BE, Breedon TH, et al. Biotransformation of glyceryl trinitrate occurs concurrently with relaxation of rabbit aorta. J Pharmacol Exp Ther. 1986; 237: 608–614.
Axelsson KL, Ahlner J. Nitrate tolerance from a biochemical point of view. Drugs. 1987; 33: 63–68.
Mulsch A, Oelze M, Kloss S, et al. Effects of in vivo nitroglycerin treatment on activity and expression of the guanylyl cyclase and cGMP-dependent protein kinase and their downstream target vasodilator-stimulated phosphoprotein in aorta. Circulation. 2001; 103: 2188–2194.