This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pickkers, P. Articles by Hughes, A. D. Search for Related Content PubMed PubMed Citation Articles by Pickkers, P. Articles by Hughes, A. D. Pubmed/NCBI databases Compound via MeSH Substance via MeSH Hazardous Substances DB HYDROCHLOROTHIAZIDE Related Collections Hypertension - basic studies Ion channels/membrane transport Other Vascular biology

(Hypertension. 1999;33:1043-1048.)
© 1999 American Heart Association, Inc.

Scientific Contributions

Inhibition of Carbonic Anhydrase Accounts for the Direct Vascular Effects of Hydrochlorothiazide

Peter Pickkers; Robinder S. Garcha; Michael Schachter; Paul Smits; Alun D. Hughes

From the Department of Pharmacology, University Hospital Nijmegen, Nijmegen, Netherlands (P.P., P.S.), and Department of Clinical Pharmacology, National Heart and Lung Institute, Queen Elizabeth the Queen Mother Wing, St. Mary's Hospital, Imperial College of Science, Technology, and Medicine, London, UK (R.S.G., M.S., A.D.H.).

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Abstract—Hydrochlorothiazide has been shown to exert direct vasodilator effects by activation of calcium-activated potassium (K_Ca) channels in human and guinea pig isolated resistance arteries. Since hydrochlorothiazide binds to and inhibits the enzyme carbonic anhydrase and because K_Ca channel activation is pH sensitive, we investigated the role of intracellular and extracellular carbonic anhydrase in the vascular effects of thiazide diuretics. Small arteries were isolated from guinea pig mesentery and studied by use of a microvascular myograph technique. In some experiments, tone and intracellular pH (pH_i) were measured simultaneously with 2',7'-bis(2-carboxyethyl)-5(6)'-carboxyfluorescein (BCECF-AM). Bendroflumethiazide, a thiazide diuretic with minimal inhibitory effects on carbonic anhydrase, had little effect on noradrenaline-induced tone (16±8% relaxation) compared with hydrochlorothiazide (74±12% relaxation). In contrast to hydrochlorothiazide, the action of bendroflumethiazide was unaffected by 100 nmol/L charybdotoxin, a selective blocker of K_Ca channels. All inhibitors of carbonic anhydrase relaxed noradrenaline-induced tone in a concentration-dependent manner, and this effect was blocked by charybdotoxin. Hydrochlorothiazide and the inhibitors of carbonic anhydrase failed to relax tone induced by a depolarizing potassium solution. Acetazolamide and hydrochlorothiazide increased pH_i by 0.27±0.07 and 0.21±0.04, respectively, whereas bendroflumethiazide had a much smaller effect: 0.06±0.03. The rise in pH_i induced by any agent was not inhibited by charybdotoxin. The vasorelaxant effect of hydrochlorothiazide is shared by other inhibitors of carbonic anhydrase. Inhibitors of carbonic anhydrase, but not bendroflumethiazide, cause intracellular alkalinization, which is associated with K_Ca channel opening. These data suggest that the vasodilator effect of thiazide diuretics results primarily from inhibition of vascular smooth muscle cell carbonic anhydrase, which results in a rise in pH_I, leading to K_Ca channel activation and vasorelaxation.

Key Words: hydrochlorothiazide • carbonic anhydrase inhibition • muscle, smooth, vascular • pH_i • potassium channels

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Thiazide diuretics were developed in the 1950s by chemical modification of carbonic anhydrase inhibitors. Although their blood pressure–lowering effects have been well documented, their mechanism of action is still not fully resolved. The principal site of action of thiazides is the early segment of the distal nephron, where they inhibit a luminal transmembrane–coupled NaCl transport system. In the long term, thiazides act by reducing peripheral resistance rather than by their diuretic effects,¹ and therefore a direct vascular effect has been proposed.

Previous studies in isolated human and guinea pig resistance arteries have established a direct vasodilator activity of hydrochlorothiazide. This vasorelaxant response to hydrochlorothiazide was abolished by charybdotoxin and iberiotoxin, both selective blockers of large-conductance Ca²⁺-activated potassium (K_Ca) channels, but not by inhibitors of other vascular K⁺ channels.² On the basis of the fact that thiazide-like drugs such as cicletanine and diazoxide lead to hyperpolarization in vascular smooth muscle cells³ and the fact that hydrochlorothiazide increases ⁸⁶Rb efflux as a marker of K⁺ efflux,^{2 4} it was proposed that hydrochlorothiazide opens K_Ca channels, thereby leading to K⁺ efflux and membrane hyperpolarization. The resultant closure of voltage-dependent Ca²⁺ channels leads to a fall in [Ca²⁺]_i and vasorelaxation.⁵

In addition to [Ca²⁺]_i, the open state probability of the K_Ca channel is also modulated by intracellular pH (pH_i). Channel opening is inhibited by intracellular acidosis in carotid body cells,⁶ while in isolated blood vessels intracellular alkalinization leads to relaxation associated with hyperpolarization of the cell membrane and a consequent fall of [Ca²⁺]_i.⁷ At present, how hydrochlorothiazide opens the K_Ca channel is unknown; it could act by direct interaction with the channel or involve an intermediate intracellular biochemical effect.

Since it is known that most thiazide diuretics bind to and inhibit carbonic anhydrase,⁸ we hypothesized that a rise in pH_i by inhibition of carbonic anhydrase could represent the mechanism of action by which thiazide diuretics open vascular K_Ca channels and relax resistance arteries. We were able to examine the influence of carbonic anhydrase inhibitor activity since different thiazide compounds exert different degrees of carbonic anhydrase inhibitor activity.⁹ It was also possible to distinguish between an effect on intracellular or extracellular membrane–bound forms of carbonic anhydrase by the use of lipophilic and hydrophilic carbonic anhydrase inhibitors.⁹

The vasodilator effects of thiazides may contribute to their antihypertensive properties, and opening of K_Ca channels by these agents may represent a novel mode of action of these drugs in the vasculature.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Tissue and Myograph Procedure
After approval of our ethics committee, guinea pig mesentery was removed from male animals (250 to 300 g) killed by cervical dislocation. Mesenteric resistance arteries (n=44 in total; ID, 320±1.9 µm) were dissected free of surrounding tissue and mounted as ring segments in an isometric microvascular myograph. The myograph contained 10 mL physiological saline solution (PSS) (in mmol/L: NaCl 118, KCl 4.7, CaCl₂ · 6H₂O 2.5, MgSO₄ · 7H₂O 1.17, NaHCO₃ 25.0, NaH₂PO₄ · 2H₂O 1.0, Na₂EDTA 0.03, glucose 5.5) maintained at 37°C and aerated with 95% O₂ and 5% CO₂. The vessels were allowed to equilibrate for 1 hour and then were set at a "normalized" internal circumference of 0.9L₁₀₀, estimated to be 90% of the circumference they would maintain if relaxed and exposed to 100 mm Hg transmural pressure. This was calculated for each individual artery on the basis of the passive length-tension characteristics of the artery and the LaPlace relationship.¹⁰ At this setting near-maximal force development can be obtained, and the internal diameters referred to were derived from this calculation.

Before start of the studies, vessels were tested for viability with the use of a depolarizing potassium solution (KPSS: PSS with equimolar substitution of 118 mmol/L KCl for NaCl) and noradrenaline (10 µmol/L). Those vessels failing to produce a tension equivalent to 90 mm Hg to these stimulants were discarded.

Effects of Carbonic Anhydrase Inhibitors on Vascular Tone
Vessels were precontracted with noradrenaline (10 µmol/L), and once stable tone was attained, concentration-response curves (n=4 for each agent; 10^-9 to 10^-4.5 mol/L) were constructed for acetazolamide, benzolamide (hydrophilic), or ethoxzolamide (lipophilic). Because benzolamide has a lower ether/water partition coefficient and a markedly lower pK_a value (3.2) than acetazolamide (7.4),⁹ it is generally assumed that it permeates into cells very slowly and therefore more or less specifically inhibits the activation of extracellular carbonic anhydrase.

Interaction Between Carbonic Anhydrase Inhibitors and Vascular Potassium Channels
It was previously demonstrated^{5 11} that the vascular action of hydrochlorothiazide was absent in vessels precontracted with a depolarizing high-potassium solution and also was inhibited by charybdotoxin, but not by antagonists of other potassium channels such as glibenclamide (ATP-dependent K⁺ channel) and apamin (small-conductance K_Ca channel).² To demonstrate that carbonic anhydrase inhibitors (n=4 for each agent; 30 µmol/L) exert direct vasoactivity by a mechanism similar to that of hydrochlorothiazide, we precontracted some vessels with noradrenaline in the presence of KPSS. Under these depolarized conditions, potassium channel activation will have a negligible effect on membrane potential and therefore should not reduce calcium entry and vascular tone. If the vasodilation induced by carbonic anhydrase inhibitors is evoked by hyperpolarization of vascular smooth muscle due to increased K⁺ conductance, its action should be inhibited by depolarized conditions. Additionally, the vasorelaxant properties of the carbonic anhydrase inhibitors (30 µmol/L) were compared before and after incubation with charybdotoxin (n=5; 20 minutes; 100 nmol/L) or glibenclamide (n=5; 20 minutes; 100 µmol/L) in noradrenaline-contracted vessels. Charybdotoxin is a selective inhibitor of K_Ca channels,¹² while glibenclamide is a selective blocker of K_ATP channels.¹³

Interaction Between Carbonic Anhydrase Inhibitors, the Eicosanoid System, and the Endothelium
The vasorelaxant effect of acetazolamide (30 µmol/L) was determined with or without 30 minutes of preincubation with 20 µmol/L indomethacin. Indomethacin is a potent NSAID, and it has been well established that NSAIDs inhibit prostaglandin synthesis by blocking the enzyme cyclooxygenase, which is involved in the generation of prostaglandin from arachidonic acid.¹⁴ In addition, the effect of endothelial removal was examined in 4 vessels. Endothelium was removed from vessels mounted in the myograph by passing a hair through the lumen of the vessel.¹⁵ The efficacy of this procedure was confirmed by abolition of relaxation to endothelium-dependent vasodilators acetylcholine (10 µmol/L) or substance P (100 nmol/L).

Effects of Thiazide Diuretics on Vascular Tone and Their Interaction With K_Ca Channels
If the ability of hydrochlorothiazide to activate K_Ca channels and relax resistance arteries is dependent on its carbonic anhydrase–inhibiting activity, any vascular effects of bendroflumethiazide, a thiazide that practically lacks carbonic anhydrase–inhibiting activity,⁹ should not be associated with K_Ca channel activation. To test this hypothesis, we compared the vascular effects of each drug (n=8 to 12; 30 µmol/L) and determined whether these effects were inhibited by charybdotoxin (20 minutes; 100 nmol/L). Because it was previously established that the vasorelaxant effect of hydrochlorothiazide is dose dependent,^{2 11} we used the concentration that elicited the maximal effect.

Measurements of pH_i
In some vessels, measurements of pH_i were obtained as described previously.^{7 16} In brief, vessel segments were set up in a single-channel myograph dedicated to fluorescence measurements and incubated with 10 µmol/L of the acetoxymethyl ester of the pH-sensitive dye 2',7'-bis(carboxyethyl)5(6)'-carboxyfluorescein (BCECF-AM). Fluorescence was measured with a Deltascan spectrofluorimeter (Photon Technology International) connected to an inverted Axiovert 35 fluorescence microscope (Carl Zeiss) using only quartz objectives (Ultrafluor x10). pH_i was assessed on the basis of the ratio of fluorescence emission measured at 510 nm, which was evoked by excitation at 450- and 495-nm light. Emission signals and vascular tone were measured simultaneously at 1 Hz and acquired online with an analog/digital interface (Photon Technology International) connected to an IBM computer. Data were stored on an optical disk and later analyzed offline with commercially available software (Photon Technology International). At the end of each experiment, the ratio was calibrated with 4 solutions (K⁺/HEPES, in mmol/L: KCl 140, MgCl₂ 1.0, CaCl₂ 1.6, EDTA 0.026, glucose 10, HEPES 10.0) in the pH range of 6.8 to 7.4 containing nigericin (10 µmol/L), as described previously.¹⁶ Nigericin is a K⁺/H⁺ ionophore that will equilibrate intracellular and extracellular pH in high-potassium buffers. The first solution was applied to the myograph for 7 minutes, and the subsequent solutions were added for 5 minutes each. With the use of this technique, a linear regression line could be calculated and the other intensity ratios could be evaluated to give true pH readings.

Effect of Acetazolamide, Hydrochlorothiazide, and Bendroflumethiazide on pH_i
Vessels were prepared as described above, and the effect of acetazolamide (n=10; 30 µmol/L), hydrochlorothiazide (n=10; 30 µmol/L), or bendroflumethiazide (n=10; 30 µmol/L) on pH_i was compared. The effect of acetazolamide and hydrochlorothiazide on pH_i was also examined after incubation with charybdotoxin (n=6; 100 nmol/L; 20 minutes).

Drugs and Solutions
Acetazolamide, bendroflumethiazide, hydrochlorothiazide, indomethacin, nigericin, and substance P were obtained from Sigma Chemical Company. Pluronic and preweighed aliquots of BCECF were purchased from Molecular Probes; one fresh aliquot was used for each experiment. Charybdotoxin was purchased from Calbiochem. Benzolamide and ethoxzolamide were a generous gift from Professor Thomas Maren (University of Florida, Gainesville, Fla). Thiazides and carbonic anhydrase inhibitors were dissolved in dimethyl sulfoxide. All serial dilutions were made in distilled water. The final concentration of dimethyl sulfoxide of 0.1% (vol/vol) had no effect on vessel reactivity.

Statistics
All data are expressed as mean±SEM, with the number of observations in parentheses. Statistical significance of values was tested with a 2-tailed paired Student's t test. Concentration-response data were fitted to a logistic function by nonlinear regression, and pD₂, the concentration of drug producing half-maximal response, was calculated. Concentration-response data were compared in terms of -log(pD₂) and maximum response by Student's t test for paired data. P<0.05 was considered significant.

	Results

Top Abstract Introduction Methods Results Discussion References

 
Effects of Carbonic Anhydrase Inhibitors on Vascular Tone
All 3 carbonic anhydrase inhibitors induced a concentration-dependent relaxation of guinea pig vessels (Figure 1). The pD₂ values of acetazolamide, benzolamide, and ethoxzolamide were 5.7±0.1, 6.3±0.3, and 7.0±0.3 (n=4 for all), respectively. The vasorelaxant effect was found not to be dependent on the presence of endothelium, and responses to acetazolamide were unaffected by preincubation with indomethacin (20 µmol/L) (maximum relaxation to acetazolamide, 74±8%; in the presence of indomethacin, 82±2%; n=6). The vasorelaxant effect was probably not due to inhibition of intracellular carbonic anhydrase, since the hydrophilic benzolamide exerted a similar response to the lipophilic ethoxzolamide.

View larger version (18K): [in this window] [in a new window]   Figure 1. Concentration-response curve showing relaxant effects of carbonic anhydrase inhibitors acetazolamide (), ethoxzolamide (lipophilic, ), or benzolamide (hydrophilic, ) after noradrenaline preconstriction. Each value represents the mean±SE percentage of control response to noradrenaline of 4 arteries.

Interaction Between Carbonic Anhydrase Inhibitors and Potassium Channels
All carbonic anhydrase inhibitors failed to relax KPSS-induced tone (n=4). The effect of incubation with charybdotoxin, an inhibitor of K_Ca channels, on the relaxation of the carbonic anhydrase inhibitors is shown in Figure 2. Incubation with charybdotoxin (100 nmol/L) had no effect on the subsequent contraction to noradrenaline. The vasorelaxant effect of all 3 carbonic anhydrase inhibitors was significantly inhibited by charybdotoxin. Substance P (100 nmol/L) was used as a control, and its vasorelaxant effect was not inhibited by charybdotoxin. Acetazolamide-induced relaxation was unaffected by the K_ATP-selective antagonist glibenclamide; relaxation to acetazolamide was 73.8±8.2% and 77.5±7.7% in the absence and presence of glibenclamide, respectively (P=NS).

View larger version (15K): [in this window] [in a new window]   Figure 2. Relaxant effect of carbonic anhydrase inhibitors acetazolamide (ACZ), benzolamide (BZL), and ethoxzolamide (EZL) (30 µmol/L) after preconstriction with noradrenaline in the absence () and presence () of charybdotoxin (CTx) (100 nmol/L). Substance P (Sub P) (100 nmol/L) was used as a control for charybdotoxin. Each value represents the mean±SE percentage of control response to noradrenaline of 4 arteries. *Significantly different from control (P<0.001).

Effects of Thiazide Diuretics on Vascular Tone and Their Interaction With K_Ca Channels
In agreement with previous reports,^{2 5 11} , hydrochlorothiazide (30 µmol/L) relaxed guinea pig vessels (74±12%; P<0.001), and this effect was almost totally abolished by charybdotoxin (P<0.001). In contrast, bendroflumethiazide had little effect on vascular tone (relaxation 16±8%; n=12). The small relaxation seen in response to bendroflumethiazide was not significantly inhibited by charybdotoxin (Figure 3).

View larger version (12K): [in this window] [in a new window]   Figure 3. Relaxant effect of thiazide diuretics hydrochlorothiazide (HCT) (30 µmol/L) and bendroflumethiazide (BFM) (30 µmol/L) after preconstriction with noradrenaline in the absence () and presence () of charybdotoxin (CTx) (100 nmol/L). Each value represents the mean±SE percentage of control response to noradrenaline of 8 to 12 arteries. *Significantly different from control (P<0.05).

Effects of Carbonic Anhydrase Inhibitors and Thiazides on pH_i
Resting pH_i in isolated guinea pig mesenteric arteries was 7.18±0.19 (n=15). As shown in Figure 4, the vasorelaxant effect of acetazolamide and hydrochlorothiazide was associated with a rise in pH_i. Bendroflumethiazide caused a small rise in pH_i, but this was not statistically significant. In vessels incubated in charybdotoxin, the acetazolamide- and hydrochlorothiazide-induced rises in pH_i were not significantly affected (Figure 5).

View larger version (16K): [in this window] [in a new window]   Figure 4. Effect of acetazolamide (ACZ) (30 µmol/L) and hydrochlorothiazide (HCT) (30 µmol/L) on pHi. Also shown for reference is the effect of aeration with 100% O2/0% CO2 (0% CO2). Representative tracings are shown of measured pHi in relaxed arteries mounted in a microvascular myograph and loaded with BCECF-AM, as described in Methods. Periods of drug exposure are indicated by horizontal bars; drug was washed out as indicated (w/o). Calibration bar is shown. Tracings are representative of 10 similar experiments.

View larger version (13K): [in this window] [in a new window]   Figure 5. Peak effect of acetazolamide (ACZ) (30 µmol/L), hydrochlorothiazide (HCT) (30 µmol/L), and bendroflumethiazide (BFM) (30 µmol/L) on pHi in the absence (n=10) and presence (n=5) of charybdotoxin (CTx) (100 nmol/L). Each value represents the mean±SEM of n observations. *P<0.01 by Student's t test.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Our experiments were designed to determine to what extent inhibition of carbonic anhydrase by thiazide diuretics accounts for their direct vasodilator effects. We have shown that at clinically relevant concentrations, the vasorelaxant effect of hydrochlorothiazide attributable to K_Ca channel opening is shared by other agents that inhibit carbonic anhydrase. In contrast, even a high concentration of bendroflumethiazide, a thiazide that practically lacks carbonic anhydrase–inhibiting activity, only minimally affected vascular tone. Furthermore, both hydrochlorothiazide and acetazolamide increased pH_i in association with opening the K_Ca channel, while bendroflumethiazide had minimal effects on either pH_i or tone. Since the effect of hydrochlorothiazide and acetazolamide on pH_i was also present in vessels preincubated with charybdotoxin, a blocker of the K_Ca channel, we conclude that the rise in pH_i is likely to be a cause and not a consequence of K_Ca channel activation and/or vasorelaxation. Since we have previously reported that charybdotoxin also inhibits the fall in intracellular calcium induced by hydrochlorothiazide,⁵ it is also unlikely that changes in intracellular calcium account for the rise in pH_i. Therefore, we conclude that inhibition of carbonic anhydrase and the associated rise in pH_i may play a primary role in the vasodilator effect of hydrochlorothiazide. Although our findings concentrate on the carbonic anhydrase–inhibiting properties and not on inhibition of the NaCl cotransporter by thiazide diuretics, these mechanisms may be interrelated, since studies in rat distal colon demonstrated that thiazide-induced inhibition of NaCl absorption is directly due to inhibition of mucosal carbonic anhydrase.¹⁷ However, to our knowledge NaCl cotransport has not been demonstrated in vascular smooth muscle cells.

Carbonic Anhydrase Inhibition and pH_i
Relatively few studies have focused on the effects of inhibition of carbonic anhydrase on pH_i and tone in vascular smooth muscle cells. In agreement with our findings, acetazolamide has also been reported to increase pH_i in turtle bladder cells,¹⁸ kidney cells,¹⁹ choroid plexus epithelial cells,²⁰ and the mandibular gland.²¹ Under different conditions and in different cells, acetazolamide has also been reported to decrease²² or have no effect²³ on pH_i. The mechanism of action of the acetazolamide-induced rise in pH_i is not completely understood, but most reports focus on an intracellular accumulation of HCO₃^{24 25} due to inhibition of Cl/HCO₃ exchange. In addition, an acetazolamide-sensitive inward chloride pump, different from the Cl/HCO₃ exchange and NaKCl cotransporter, has been reported in rat arterial vascular smooth muscle cells.²⁶ Others found the same acetazolamide-induced inhibition of renal Cl/HCO₃ exchange in vivo and suggested that in the presence of acetazolamide, H⁺ extrusion continues, but the rate of reaction of OH^- with CO₂ is diminished as a result of carbonic anhydrase inhibition.^{19 27} In our experiments the hydrophilic benzolamide was approximately as effective as acetazolamide and ethoxzolamide, indicating that inhibition of the extracellular membrane–bound form of carbonic anhydrase is responsible for the intracellular alkalinization, K_Ca channel activation, and vasorelaxation. It is unclear how inhibition of this enzyme can mediate changes in pH_i, but it seems possible that this extracellular enzyme might modulate Cl/HCO₃ exchange, resulting in an attenuated HCO₃ extrusion and increase in pH_i. The present study does not allow definite conclusions on the mechanism of the acetazolamide-induced increase in pH_i, but our observation that the vascular action and the pH_i effect are shared by a thiazide diuretic that also exerts carbonic anhydrase–inhibiting activity and not by a thiazide that lacks this effect suggests that both effects may be due to inhibition of carbonic anhydrase.

pH_i and Vascular Tone
One of the various ways (for review, see Reference 28²⁸ ) in which changes of pH_i could alter the force development in smooth muscle cells is through potassium channel modulation, since marked effects of pH_i on the K_Ca channel have been reported in various tissues.^{29 30 31} Since the vasorelaxant action of hydrochlorothiazide is inhibited by charybdotoxin, we hypothesized that a pH_i change, due to the carbonic anhydrase–inhibiting activity of the drug, was the trigger for K_Ca channel activation. In isolated type I cells of the neonatal rat carotid body, the K⁺ current that was inhibited by intracellular acidosis was also inhibited by charybdotoxin and not by apamin,⁶ suggesting that the K⁺ current is carried through large-conductance K_Ca channels. In accordance, the vascular effects of hydrochlorothiazide have also been reported to be inhibited by charybdotoxin and iberiotoxin^{2 5 11} but not by apamin.^{2 11}

Direct Vasoactivity of Carbonic Anhydrase Inhibitors
Although we used the carbonic anhydrase inhibitors in the present study as a tool to elucidate the mechanism of action of thiazide diuretics, the observation that acetazolamide is a direct vasodilator at clinically relevant concentrations in isolated resistance arteries is an interesting finding in itself. The vascular effects of acetazolamide have been well studied, especially on the cerebral vasculature,³² but our finding that the vasorelaxant effect of acetazolamide is associated with a rise in pH_i and K_Ca channel activation is completely novel. We found that the dose-dependent vasorelaxant effect of acetazolamide is caused by opening of K_Ca channels and not mediated by other K⁺ channels, the eicosanoid system, or the endothelium.

In vivo, systemic administration of acetazolamide can produce pronounced hypercapnia. Because hypercapnia is a potent dilator of cerebral blood vessels,³³ it is possible that the direct vasodilator mechanism that we describe does not account for the cerebral vasodilation in response to acetazolamide. In our experiments in isolated arteries, the maximal response to acetazolamide is reached within minutes, whereas the maximal vasodilator response in the carotid vascular bed after systemic administration of acetazolamide takes up to 1 hour.³² Furthermore, the acetazolamide-induced vasodilation of cerebral vessels appears to be dependent on prostaglandin synthesis but not on nitric oxide release, since it was found to be inhibited by indomethacin³⁴ but not by N^G-nitro-L-arginine,³⁵ an inhibitor of nitric oxide synthase. In contrast, our results indicate that the direct vasorelaxant effect of acetazolamide is independent of both local prostaglandin synthesis and the endothelium. It is assumed that cerebral vessels are sensitive to a fall in extracellular pH,³⁶ whereas our data concentrate on the rise in pH_i.

In contrast to the well-studied effects of acetazolamide on the cerebral vasculature, not much is known about its direct vascular effects in other vascular beds. Since vascular effects appear to depend on inhibition of carbonic anhydrase activity, this could account for contradictory reports regarding the vasoactivity of acetazolamide. Vascular carbonic anhydrase activity varies from organ to organ and also between species.³⁷ It has been reported that rabbit aorta does not contain carbonic anhydrase activity,³⁸ and, in agreement, thiazide diuretics that possess carbonic anhydrase–inhibiting activity do not relax isolated rabbit arteries (A.D.H., unpublished data, 1996). It was reported that the direct vasorelaxant effects of hydrochlorothiazide are present in human and guinea pig vessels but not in rat resistance arteries,³⁹ and it is of interest that acetazolamide fails to change pH_i in rat mesenteric resistance arteries (Aalkjær C., written communication, 1997).

It is difficult to speculate whether the aforementioned properties of acetazolamide and hydrochlorothiazide play a role during long-term administration of the drugs in humans. In the present study we used rather high concentrations of the thiazide diuretics and carbonic anhydrase inhibitors; however, the concentration-response curves show that direct vascular effects are seen at clinically relevant concentrations.³⁹ Furthermore, the antihypertensive effects of thiazide diuretics take several weeks to reach their maximum and also wear off slowly after termination of therapy. This may suggest that slow accumulation in the target organ takes place, especially since the thiazide-like agent indapamide was found at a 9-fold higher concentration in vascular smooth muscle cells than in the plasma.⁴⁰ Consequently, despite the high concentrations used in this study, it is conceivable that the mechanism described may be relevant to the actions of these agents in vivo after long-term administration.

Conclusion
We have previously shown that hydrochlorothiazide relaxes human³⁹ and guinea pig^{2 5 11 39} isolated arteries by opening K_Ca channels. This effect of hydrochlorothiazide is not shared by bendroflumethiazide and seems related to its activity as an inhibitor of carbonic anhydrase. At clinically relevant concentrations, other inhibitors of carbonic anhydrase also relax vascular smooth muscle by activation of K_Ca channels associated with an increase in pH_i. In view of the efficacy of the hydrophilic inhibitor benzolamide, this effect probably does not involve an effect on intracellular carbonic anhydrase. As a result of our studies, we propose that acetazolamide and thiazide diuretics that inhibit carbonic anhydrase activity produce intracellular alkalosis of vascular smooth muscle cells. The rise in pH_i as a consequence of inhibition of carbonic anhydrase appears to activate the K_Ca channel, resulting in hyperpolarization of the vascular smooth muscle cell, reduction of voltage-dependent calcium channel activity, fall in [Ca²⁺]_i, and vasorelaxation. It is possible that this novel mechanism of vasodilation contributes to the antihypertensive action of thiazides in vivo.

	Acknowledgments

 
This study was supported by a grant from the Dutch Heart Foundation (grant 94.059) (Dr Pickkers).

	Footnotes

 
Reprint requests to Alun D. Hughes, MBBS, PhD, Department of Clinical Pharmacology, National Heart and Lung Institute, Queen Elizabeth the Queen Mother Wing, St. Mary's Hospital, Imperial College of Science, Technology, and Medicine, London W2 1NY, UK.

Received July 21, 1998; first decision August 21, 1998; accepted December 3, 1998.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Van Brummelen P, Man in `t Veld AJ, Schalekamp MADH. Haemodynamic changes during long term thiazide treatment of essential hypertension in responders and non-responders. Clin Pharmacol Ther. 1980;27:328–336.[Medline] [Order article via Infotrieve]
2. Calder JA, Schachter M, Sever PS. Potassium channel opening properties of thiazide diuretics in isolated guinea-pig resistance arteries. J Cardiovasc Pharmacol. 1994;24:158–164.[Medline] [Order article via Infotrieve]
3. Siegel G, Schnalke F, Schultz G, Stock G. K⁺ channel opening and vascular tone. In: Mulvany MJ, Aalkjær C, Heagerty AM, Nyborg NCB, Standgaard S, eds. Resistance Arteries, Structure and Function. Amsterdam, Netherlands: Elsevier; 1991:130–139.
4. Moura A-M, Worcel M. Mode of action of cyclothiazide and triamterene: ex vivo effect of ²²Na and ⁸⁶Rb efflux from arterial smooth muscle. Eur J Pharmacol. 1983;86:129–133.
5. Pickkers P, Hughes AD. Relaxation and decrease in [Ca²⁺]_i by hydrochlorothiazide in guinea-pig isolated mesenteric arteries. Br J Pharmacol. 1995;114:703–707.[Medline] [Order article via Infotrieve]
6. Peers C, Green FK. Inhibition of Ca²⁺-activated K⁺ currents by intracellular acidosis in isolated type I cells of the neonatal rat carotid body. J Physiol. 1991;437:589–602.[Abstract/Free Full Text]
7. Jensen PE, Hughes AD, Boonen HCM, Aalkjær C. Force, membrane potential, and [Ca²⁺]_i during activation of rat mesenteric small arteries with norepinephrine, potassium, aluminum fluoride, and phorbol ester: effects of changes in pH_i. Circ Res. 1993;73:314–324.[Abstract/Free Full Text]
8. Matsumoto K, Miyazaki H, Fujii T, Hashimoto M. Binding of sulfonamides to erythrocytes and components. Chem Pharm Bull (Tokyo). 1989;37:1913–1915.[Medline] [Order article via Infotrieve]
9. Maren TH. Direct measurements of the rate constants of sulfonamides with carbonic anhydrase. J Pharmacol Exp Ther. 1991;41:419–426.
10. Mulvany MJ, Halpern W. Contractile properties of small arterial resistance vessels in spontaneously hypertensive and normotensive rats. Circ Res. 1977;41:19–26.[Free Full Text]
11. Calder JA, Schachter M, Sever PS. Ion channel involvement in the acute vascular effects of thiazide diuretics and related compounds. J Pharmacol Exp Ther. 1993;265:1175–1180.[Abstract/Free Full Text]
12. Gimenez-Gallego G, Navia MA, Reuben JP, Katz GM, Kaczorowski GJ, Garcia ML. Purification, sequence and model structure of charybdotoxin, a potent selective inhibitor of Ca²⁺-activated potassium channels. Proc Natl Acad Sci U S A. 1988;85:3329–3333.[Abstract/Free Full Text]
13. Longmore J, Newgreen DT, Weston AH. Effect of cromakalim, RF49356, diazoxide, glibenclamide and galanin in rat portal vein. Eur J Pharmacol. 1990;190:75–84.[Medline] [Order article via Infotrieve]
14. Vane JR. Inhibition of prostaglandin synthesis as a mechanism of action for aspirin like drugs. Nat New Biol. 1971;231:232–235.[Medline] [Order article via Infotrieve]
15. Prieto D, Benedito S, Nyborg NCB. Heterogeneous involvement of endothelium in calcitonin gene-related peptide-induced relaxation in coronary arteries from rat. Br J Pharmacol. 1991;103:1764–1768.[Medline] [Order article via Infotrieve]
16. Thomas JA, Buchsbaum RN, Zimniak A, Racker E. Intracellular pH measurements in Ehrlich Ascites tumor cells utilizing spectroscopic probes generated in situ. Biochemistry. 1979;18:2210–2218.[Medline] [Order article via Infotrieve]
17. Goldfarb DS, Chan AJ, Hernandez D, Charney AN. Effect of thiazides on colonic NaCl absorption: role of carbonic anhydrase. Am J Physiol. 1991;261:F452–F458.[Abstract/Free Full Text]
18. Graber M, Dixon T, Coachman D, Devine P. Acetazolamide inhibits acidification by the turtle bladder independent of cell pH. Am J Physiol. 1989;256:F923–F931.[Abstract/Free Full Text]
19. Henderson RM, Bell PB, Cohen RD, Browning C, Iles RA. Measurement of intracellular pH with microelectrodes in rat kidney in vivo. Am J Physiol. 1986;250:F203–F209.
20. Johanson CE, Murphy VA. Acetazolamide and insulin alter choroid plexus epithelial cell [Na+], pH, and volume. Am J Physiol.. 1990;258:F1538–F1546.[Abstract/Free Full Text]
21. Pirani D, Evans LAR, Cook DI, Young JA. Intracellular pH in the rat mandibular salivary gland: the role of Na-H and Cl-HCO3 antiports in secretion. Eur J Physiol. 1987;408:178–184.[Medline] [Order article via Infotrieve]
22. Bonanno JA, Srinivas SP, Brown M. Effect of acetazolamide on intracellular pH and bicarbonate transport in bovine corneal endothelium. Exp Eye Res. 1995;60:425–434.[Medline] [Order article via Infotrieve]
23. Matsumoto T, Tomita T. Intracellular alkalinization caused by chloride removal in the smooth muscle of guinea-pig vena cava. Jpn J Physiol. 1993;43:67–73.[Medline] [Order article via Infotrieve]
24. Johanson CE. Differential effects of acetazolamide, benzolamide and systemic acidosis on hydrogen and bicarbonate gradients across the apical and basolateral membranes of the choroid plexus. J Pharmacol Exp Ther. 1984;231:502–511.[Abstract/Free Full Text]
25. Johanson CE, Parandoosh Z, Dyas ML. Maturational differences in acetazolamide-altered pH and HCO₃ of choroid plexus, cerebrospinal fluid, and brain. Am J Physiol.. 1992;262:R909–R914.[Abstract/Free Full Text]
26. Chippenfield AR, Davis JP, Harper AA. An acetazolamide-sensitive inward chloride pump in vascular smooth muscle. Biochem Biophys Res Commun. 1993;194:407–412.[Medline] [Order article via Infotrieve]
27. Sasaki S, Marumo F. Effects of carbonic anhydrase inhibitors on basolateral base transport of rabbit proximal straight tubule. Am J Physiol. 1989;257:F947–F952.[Abstract/Free Full Text]
28. Wray S. Smooth muscle intracellular pH: measurement, regulation and function. Am J Physiol. 1988;254:C213–C225.[Abstract/Free Full Text]
29. Christensen O, Zeuthen T. Maxi K⁺ channels in leaky epithelia are regulated by intracellular Ca²⁺, pH and membrane potential. Pflugers Arch. 1987;408:249–259.[Medline] [Order article via Infotrieve]
30. Cook DL, M. Ikeuchi M, Fujimoto WY. Lowering of pH_i inhibits Ca²⁺-activated K⁺ channels in pancreatic cells. Nature. 1987;311:269–271.
31. Kume H, Takagi K, Satake T, Tokuno H, Tomita T. Effects of intracellular pH upon the Ca inward current and isometrical contractile force in mammalian ventricular myocardium. Pflugers Arch. 1976;366:31–38.[Medline] [Order article via Infotrieve]
32. Démolis P, Chalon S, Giudicelli J-F. Acetazolamide-induced vasodilation in the carotid vascular bed in healthy volunteers. J Cardiovasc Pharmacol. 1995;26:841–844.[Medline] [Order article via Infotrieve]
33. Heistad DD, Kontos HA. Cerebral circulation. In: Shepherd JT, Abboud FM, eds. Handbook of Physiology, Section 2: The Cardiovascular System, Volume III, Peripheral and Organ Blood Flow, Part I. Washington, DC: American Physiological Society; 1983:137–182.
34. Wang Q, Paulson OB, Lassen NA. Indomethacin abolishes cerebral blood flow increase in response to acetazolamide-induced extracellular acidosis: a mechanism for its effect on hypercapnia? J Cereb Blood Flow Metab. 1993;13:724–727.[Medline] [Order article via Infotrieve]
35. Wang Q, Paulson OB, Lassen NA. Effect of nitric oxide blockade by N^G-nitro-l-arginine on cerebral blood flow response to changes in carbon dioxide tension. J Cereb Blood Flow Metab. 1992;12:947–953.[Medline] [Order article via Infotrieve]
36. Vorstrup S, Jensen KE, Thomsen C, Hendriksen O, Lassen NA, Paulson OB. Neuronal pH regulation: constant normal intracellular pH is maintained in brain during low extracellular pH induced by acetazolamide-³¹P NMR study. J Cereb Blood Flow Metab. 1989;9:417–421.[Medline] [Order article via Infotrieve]
37. O'Brasky J, Crandall ED. Organ and species differences in tissue vascular carbonic anhydrase activity. J Appl Physiol. 1980;49:211–217.[Free Full Text]
38. Muhleisen M, Kreye VA. Lack of soluble carbonic anhydrase in aortic smooth muscle of the rabbit. Pflugers Arch. 1985;405:234–236.[Medline] [Order article via Infotrieve]
39. Calder JA, M. Schachter M, Sever PS. Direct vascular actions of hydrochlorothiazide and indapamide in isolated small vessels. Eur J Pharmacol. 1992;220:19–26.[Medline] [Order article via Infotrieve]
40. Campbell DB, Taylor AD, Hopkins YW, Williams JRB. Pharmacokinetics and metabolism of indapamide: a review. Curr Med Res Opin. 1977;5(suppl 1):13–24.

This article has been cited by other articles:

				N. Schliebe, R. Strotmann, K. Busse, D. Mitschke, H. Biebermann, L. Schomburg, J. Kohrle, J. Bar, H. Rompler, J. Wess, et al. V2 vasopressin receptor deficiency causes changes in expression and function of renal and hypothalamic components involved in electrolyte and water homeostasis Am J Physiol Renal Physiol, October 1, 2008; 295(4): F1177 - F1190. [Abstract] [Full Text] [PDF]

				A. K. Kehler, K. Holmgaard, A. Hessellund, C. Aalkjaer, and T. Bek Variable Involvement of the Perivascular Retinal Tissue in Carbonic Anhydrase Inhibitor Induced Relaxation of Porcine Retinal Arterioles In Vitro Invest. Ophthalmol. Vis. Sci., October 1, 2007; 48(10): 4688 - 4693. [Abstract] [Full Text] [PDF]

				L. A. Shimoda, T. Luke, J. T. Sylvester, H.-W. Shih, A. Jain, and E. R. Swenson Inhibition of hypoxia-induced calcium responses in pulmonary arterial smooth muscle by acetazolamide is independent of carbonic anhydrase inhibition Am J Physiol Lung Cell Mol Physiol, April 1, 2007; 292(4): L1002 - L1012. [Abstract] [Full Text] [PDF]

				V. Pech, Y. H. Kim, A. M. Weinstein, L. A. Everett, T. D. Pham, and S. M. Wall Angiotensin II increases chloride absorption in the cortical collecting duct in mice through a pendrin-dependent mechanism Am J Physiol Renal Physiol, March 1, 2007; 292(3): F914 - F920. [Abstract] [Full Text] [PDF]

				C. Hohne, P. A. Pickerodt, R. C. Francis, W. Boemke, and E. R. Swenson Pulmonary vasodilation by acetazolamide during hypoxia is unrelated to carbonic anhydrase inhibition Am J Physiol Lung Cell Mol Physiol, January 1, 2007; 292(1): L178 - L184. [Abstract] [Full Text] [PDF]

				R. A. Phillips New-onset diabetes mellitus less deadly than elevated blood pressure?: following the evidence in the administration of thiazide diuretics. Arch Intern Med, November 13, 2006; 166(20): 2174 - 2176. [Full Text] [PDF]

				L. J. Teppema, H. Bijl, R. R. Romberg, and A. Dahan Antioxidants reverse depression of the hypoxic ventilatory response by acetazolamide in man J. Physiol., May 1, 2006; 572(3): 849 - 856. [Abstract] [Full Text] [PDF]

				G. Dell'Omo, G. Penno, S. Del Prato, and R. Pedrinelli Chlorthalidone Improves Endothelial-Mediated Vascular Responses in Hypertension Complicated by Nondiabetic Metabolic Syndrome Journal of Cardiovascular Pharmacology and Therapeutics, October 1, 2005; 10(4): 265 - 272. [Abstract] [PDF]

				R. Essalihi, H. H. Dao, L.-A. Gilbert, C. Bouvet, Y. Semerjian, M. D. McKee, and P. Moreau Regression of Medial Elastocalcinosis in Rat Aorta: A New Vascular Function for Carbonic Anhydrase Circulation, September 13, 2005; 112(11): 1628 - 1635. [Abstract] [Full Text] [PDF]

				C. Bazzini, V. Vezzoli, C. Sironi, S. Dossena, A. Ravasio, S. De Biasi, M. Garavaglia, S. Rodighiero, G. Meyer, U. Fascio, et al. Thiazide-sensitive NaCl-cotransporter in the Intestine: POSSIBLE ROLE OF HYDROCHLOROTHIAZIDE IN THE INTESTINAL Ca2+ UPTAKE J. Biol. Chem., May 20, 2005; 280(20): 19902 - 19910. [Abstract] [Full Text] [PDF]

				G. Gamba Molecular Physiology and Pathophysiology of Electroneutral Cation-Chloride Cotransporters Physiol Rev, April 1, 2005; 85(2): 423 - 493. [Abstract] [Full Text] [PDF]

				Z. Zhu, S. Zhu, D. Liu, T. Cao, L. Wang, and M. Tepel Thiazide-Like Diuretics Attenuate Agonist-Induced Vasoconstriction by Calcium Desensitization Linked to Rho Kinase Hypertension, February 1, 2005; 45(2): 233 - 239. [Abstract] [Full Text] [PDF]

				A. D Hughes How do thiazide and thiazide-like diuretics lower blood pressure? Journal of Renin-Angiotensin-Aldosterone System, December 1, 2004; 5(4): 155 - 160. [Abstract] [PDF]

				G.-H. Kim, J. W. Lee, Y. K. Oh, H. R. Chang, K. W. Joo, K. Y. Na, J.-H. Earm, M. A. Knepper, and J. S. Han Antidiuretic Effect of Hydrochlorothiazide in Lithium-Induced Nephrogenic Diabetes Insipidus Is Associated with Upregulation of Aquaporin-2, Na-Cl Co-transporter, and Epithelial Sodium Channel J. Am. Soc. Nephrol., November 1, 2004; 15(11): 2836 - 2843. [Abstract] [Full Text] [PDF]

				J. Loffing Paradoxical Antidiuretic Effect of Thiazides in Diabetes Insipidus: Another Piece in the Puzzle J. Am. Soc. Nephrol., November 1, 2004; 15(11): 2948 - 2950. [Full Text] [PDF]

				C. Hohne, M. O. Krebs, M. Seiferheld, W. Boemke, G. Kaczmarczyk, and E. R. Swenson Acetazolamide prevents hypoxic pulmonary vasoconstriction in conscious dogs J Appl Physiol, August 1, 2004; 97(2): 515 - 521. [Abstract] [Full Text] [PDF]

				J. T. Berg, S. Ramanathan, M. G. Gabrielli, and E. R. Swenson Carbonic Anhydrase in Mammalian Vascular Smooth Muscle J. Histochem. Cytochem., August 1, 2004; 52(8): 1101 - 1106. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services