Glucose Toler Ated Potassium (Kcn) Channels in Pancreatic /I-Cells
PDF hosted at the Radboud Repository of the Radboud University Nijmegen
The following full text is a publisher's version.
For additional information about this publication click this link. http://hdl.handle.net/2066/21830
Please be advised that this information was generated on 2021-09-27 and may be subject to change. Journal of Human Hypertension (1995) 9, 691 693 © 1995 Stockton Press, All rights reserved 0950-9240 95 $ 12.00 Do KCa channels and carbonic anhydrase play a role in thiazide-induced hyperglycaemia?
P Pickkers, M Schächter, PS Sever and AD Hughes Dapartnumt of Clinical Pharmacology, Queen Elizabeth the Queen Mother Wing, St. Mary's Hospital Medical School, Imperial College of Science Technology and Medicine, South Wharf Road, London W2 I NY, UK
Summary:
Thiazide diuretics are widely used to treat hypertension, zides may be due to their ability to open calcium-activ but their use Is associated with impaired glucose toler ated potassium (KCn) channels in pancreatic /i-cells. ance. We propose that the diabetogenic action of thia
Keywords; thiazich diiuvtica; K chuiuwl; glucose; puncwutic ß~ctrfl; calcium
»*•••. ,,, «. t«r • * ■ ? •» • !< M • * ► i ß « j > * ^ *A- \ t 0Û • I * • A . \ « « , * w ^ k ^
Thiazide diuretics are a widely used therapy for cific class of ATP-dopendent potassium channels hypertension. However, glucose intolerance de (KATP channels) and that the consequent hyper velops in about 3% of hypertensive patients treated polarisation impairs insulin release,*1”11 and prob with thiazides.1*” The mechanism of this adverse ably accounts for the diabetogenic effect of this effect is poorly understood. Glucose intolerance drug. This effect of diazoxide can be inhibited by caused by thiazide diuretics is dose-dependent and agents such as glibenclamide which block KATP usually reversible after discontinuation of treat channels and are used clinically as hypoglycaemic ment.1,2 It is functionally correlated with hypokal- agents. It has also been proposed that diazoxide aemia and has been proposed to be secondary to it, exerts direct vasodilator activity by activating the In some studies prevention of the decline in serum same, or a similar, KATI> channel in vascular smooth potassium prevents hyperglycaemia- and hypokal- muscle cells.12 aemia is associated with a reduced pancreatic /i-cell Recently it has been found that hydrochlorothia response to glucose. However, it has also been sug zide also reduces insulin release. This action of gested that hypokalaemia may be an aggravating hydrochlorothiazide was associated with a fall in factor rather than a primary stimulus.J Besides the glucose-stimulated **r,Ca“+ uptake in /J-cells.^ As effect of thiazides on pancreatic cells, extra-pan discussed above, influx of calcium is one of the creatic; mechanisms have also been proposed to most distal steps in stinnilus-secretion coupling in explain the diabetogenic action of the thiazide pancreatic /i-cells and is crucial to insulin release.7 diuretics;1,5 We wish to propose a novel hypothesis We propose that the action of hydrochlorothiazide regarding the diabetogenic effects of thiazides based on calcium influx into the pancreatic /i-cell may be on their action on a class of potassium (K) channels, by the same mechanism by which the drug inhibits which in turn may be related to their effects on calcium entry into vascular smooth muscle cells. intracellular pH (pHj) through inhibition of car Our group has previously shown that hydrochloro bonic anhydrase. thiazide relaxes vascular smooth muscle cells In the pancreas, depolarisation of the plasma through an inhibition of calcium influx.H This ef membrane by closure of K channels and opening of fect of hydrochlorothiazide is associated with an cation and chloride channels leads to opening of increase in *mRb efflux, a marker of K* efflux,15 voltage-dependent calcium channels and influx of indicating that hydrochlorothiazide increases mem calcium,r,"a The resultant increase of intracellular brane permeability to K f ions, probably by opening calcium ([Ca2+]|) is crucial to the process of insulin K channels. The effects of hydrochlorothizide on release; drugs which interfere with these mechan vascular tone, [Ca2+li and miRb efflux wore inhibited isms inhibit insulin release. It is known that diazox by charybdotoxin, a blocker of calcium-activated ide, a drug structurally related to the thiazides, potassium (KCJ channels,UATl In contrast, the KAn> hvperpolarises pancreatic /f-cells by opening a spe- channel blocker glibenclamide did not affect hydro
«v -* . -• -• t i A ••-•J*- *. Sv*** SÎSA *• tf' ^ < ^ 1*' “ s • • • ' / V W ' ♦ chlorothiazide-induced vasorelaxation or m’Rb ef Oonïïspoiultîiuut: AD Hughes flux.tn This is consistent with a previous report Kcs, and carbonic anhydrase P Pickkers et al
692 showing that hydrochlorothiazide does not affect KCa channels leads to hyperpolarisation of the cell the opening of single KATP channels in the ß-cell membrane and closure of voltage-operated calcium plasma membrane.16 These data suggest that the channels, hence [Ca“*], decreases (Figure 1). In vas opening of calcium activated K channels (KCû) cular smooth muscle this leads to vasodilation, rather than KATP channels mediates the vasodilator whereas in pancreatic ß-cells this leads to reduced effect of thiazides. Interestingly, electrophysiolo- insulin release and impaired glucose tolerance. gical studies have shown that cicletanine, a thia zide-like drug containing a furopyridine group, causes membrane hyperpolarisation in isolated Acknowledgements canine femoral arteries17 probably by an action on KCa channels. KCa channels are present in pancreatic This work was supported by the British Heart ß-cells5 and by analogy with the potassium channel Foundation and the Dutch Heart Foundation. opening properties of diazoxide it is very likely that the effect of hydrochlorothiazide on KC(l channels hyperpolarises the /3-cell membrane, leading to a References reduced insulin release. This effect could explain the inhibitory action of hydrochlorothiazide on 1 Carlsen JE et al. Relation between dose of bendroflua- insulin production. zide, anti-hypertensive effect and adverse biochemical On-going studies in our department suggest that effects. BrMed ƒ1990; 300: 975-97«. the K channel opening properties of hydrochloro 2 Knauf H. The role of low-dose diuretics in essential thiazide may depend on changes in intracellular pH hypertension. ƒ Cardiovasc Pharmacol 1993; 22 (pHJ. Effects of hydrochlorothiazide on pH{ appear (Suppl. 6); S1-S7. to be associated with the ability of the drug to 3 Furman BL. Impairment of glucose tolerance by inhibit carbonic anhydrase activity.18 Interestingly diuretics and other drugs. Pharmacol Thor 1981; 12: others have shown that pre-treatment with acetazo- 613—649, lamide, an inhibitor of carbonic anhydrase, caused 4 Brass EP, Effects of anti-hypertensive drugs on decreased glucose tolerance in mice and decreased endocrine function. Drugs 1984; 27: 447-458. Moore RD, Fidelman ML, Seeholzer SH, Correlation secretion of insulin by isolated pancreatic islets.lH between insulin action upon glycolysis and change in These findings support the idea that inhibition of intracellular pH. Biochem ƒ 1976; 154: 597-004, carbonic anhydrase activity may modulate pan 6 Sehlin J, Interrelationship between chloride fluxes in creatic ß-cell funçtion and glucose tolerance, In pancreatic islets and insulin release. Am ƒ Phvsioj addition, there are data indicating that bendro- 1978; 235: 501-508. flumethiazide, which unlike other thiazide diuret 7 Wollheim C, Sharp GWG. Regulation of insulin re ics has minimal inhibitory effects on carbonic anhy lease by calcium. Physiol Rev 1981; 61: 914-973. drase at therapeutic levels, does not impair glucose 8 Pederson OH, Findlay I. Electrophysiology of the pan tolerance in hypertensive patients20,21 and even im creas. Physiol Rev 1987; 67; 1054-1110, ~ proves glucose tolerance in the long term,22 9 Malaisse WJ, Malaisse-Lagae F. Effect of thiazides As a result of our studies we postulate that thia upon insulin secretion in vitro. Arch Int Pharmaco- dynam 1968; 171: 235-239. zide-induced hyperglucaemia is, at least in part, 10 Henquin JG, Meissner HP. Opposite «fleets of tolbuta due to hyperpolarisation of the /3-cell via an action mide and diazoxide on nüRb fluxes and membrane on KCa channels. This effect may be caused by potential in pancreatic /k:ells. Bioclwm Pharmacol inhibition of carbonic anhydrase by hydrochloro 1982; 31:1407-1415. thiazide resulting in an increase in pHi( Opening of 11 Trübe G, Roosman P, Ohno-Shusaku T. Opposite effects of tolbutamide and diazoxide on the ATP Thiazide dependent K+ channel in mouse pancreatic jU\ells. Ca2 Pfiiiger's Arch ' A K* 1986; 407: 493-499. carbonic © '12 Edwards G, Weston AH. The pharmacology of ATP-
9* anhydrase sensitive potassium channels. /Inmi Rav Pharmacol Toxicol 1993; 33: 597-637. 13 Sandstrom PE. Inhibition by hydrochlorothiazide of insulin release and calcium influx in mouse pan creatic /j-cells. BrJ Pharmacol 1993; 110: 1359-1362. PANCREATIC 14 Pickkers P, Hughes AD. Relaxation and decrease in ß-CELL [Ca +]i by hydrochlorothiazide in guinea pig isolated mesenteric arteries. Br ƒ Pharmacol 1995; 114: 703-707. 15 Calder JA, Schächter M, Sever PS, Potassium channel opening properties of thiazide diuretics in isolated guinea pig resistance arteries, J Cardiovasc Pharmacol 1994; 24: 158-164. Diazoxide 16 Gilles KD et al. Effect of sulfonamides on metabolite- V regulated ATP*-sensitive K* channel in rat pancreatic ------sulphonylurea 0-cells. Am J Physiol 1989; 257: 1119-1127. 17 Siegel G et al. K+ channel opening and vascular tone. Figure 1 Diagrammatic scheme illustrating the proposed In: Mulvany MJ et al (eds). Resistance Arteries» Struc mechanism of inhibition of insulin release by thiazides. KCil- ture and Function, pp. 130-139. Elsevier: Amster calcium-activated K channel, KATP-ATP-dependent K channel. dam, 1991. VOC: voltage-operated calcium channel 18 Beyer KH, Baer JE. Physiological basis for the action Kç* and carbonic anhydrase P Pickkers et al
693 of newer diuretic agents, Pharmacol Hcv 1961; 13: 21 Corcoran JS at al. Hypertension in rxon-insulin 517—5(32, dependent diabetes; a comparison of atenolol, nifedi 10 Bosfquist L, Backman AM, St ramberg G. Effects of pine and captopril combined with bendro flume thia acetazolamide on insulin release, serum glucose and zide. Diabet Med 1987; 4: 164-168, insulin, glucose tolerance and alloxan sensitivity of 22 Berglnmd G, Andersson O, Widgren B, Low-dose mice. Med Mol 1980; 58: 169-173. anti-hypertensive treatment with a thiazide diuretic 20 Derglnnd G at a I. Anti-hypertensive effect and si de is not diabetogenic. A 10-year controlled trial with effects of bondro flume thiazide and propranolol. Acta bendro flume thiazide. Acta Mod Scand 1986; 220: Mvd Scand 1976; 199: 499-506, 419—424,