Effect of Hypercholesterolaemia on Voltage-Operated Calcium Channel Currents in Rabbit Arterial Smooth Muscle Cells

Home , Tetraethylammonium

Journal of Human Hypertension (1999) 13, 849–853  1999 Stockton Press. All rights reserved 0950-9240/99 $15.00 http://www.stockton-press.co.uk/jhh Effect of hypercholesterolaemia on voltage-operated calcium channel currents in rabbit arterial smooth muscle cells

GF Clunn, S Wijetunge and AD Hughes Clinical Pharmacology, NHLI, St. Mary’s Hospital, Imperial College of Science, Technology and Medicine, South Wharf Road, London, W2 1NY, UK

Cholesterol is a major component of cell membranes capacitance was also greater in NZ cells. Consequently, and influences membrane fluidity. Watanabe heritable there was no significant difference in current density hyperpercholesterolaemic rabbits (WHHL) possess between NZ and WHHL cells either in the absence of defective receptors for low density lipoprotein leading drug or in the presence of the calcium channel agonist to increased plasma cholesterol, accumulation of chol- (+)202 791. Current voltage-relationships, kinetics of esterol in the arterial wall and atherosclerosis. In this fast inactivation and steady-state inactivation of IBa also study calcium channel currents (IBa) were compared did not differ significantly between WHHL and NZ. These using conventional whole cell voltage clamp techniques findings suggest that hypercholesterolaemia in WHHL in ear artery cells isolated from control New Zealand has no direct effect on calcium channel current density

White rabbits (NZ) with those from WHHL. IBa were larger or voltage-modulation in arterial smooth muscle cells. in cells isolated from NZ than from WHHL, however cell

Keywords: calcium channel; cholesterol; vascular smooth muscle; Watanabe hypercholesterolaemic rabbit

Introduction atic cholesterol toxicity or other organ damage which develops in cholesterol-fed rabbits.15 WHHL Cholesterol is a major component of cell membranes 1,2 has therefore been proposed to be a model of human and influences membrane structure and fluidity. hereditary hypercholesterolaemia and understand- Changes in membrane fluidity affect the activity of 2 ing pathology in this strain may shed light on a number of enzymes and ion channels, and in human disease processes. addition cholesterol may modulate membrane bound proteins through direct interactions with hydrophobic regions of membrane proteins.3 Materials and methods Incubation with cholesterol has been reported to promote Ca2+ influx into isolated arteries4 and cul- Single arterial smooth muscle cells were freshly dis- 5–8 persed from WHHL and NZ ear arteries by an enzy- tured vascular smooth muscle cells. Voltage- 16 operated calcium channels are a major route of Ca2+ matic digestion technique. Arteries isolated from entry into smooth muscle9 and several studies have WHHL showed no visible evidence of atheromatous implicated this channel in mediating the choles- changes. Vessels from both strains were cut into terol-induced rise in intracellular Ca2+.4,6,10 Elevated short strips (2–3 mm) and incubated in a low- plasma cholesterol may be therefore an important calcium physiological salt solution containing modulator of Ca2+ entry into vascular smooth muscle (mM): NaCl 130, KCl 6, CaCl2 0.01, MgCl2 1.2, glu- and hence influence vascular tone. This action cose 14 and HEPES 10.7 buffered to pH 7.4 with could contribute to the adverse cardiovascular NaOH and 2 mg/ml bovine serum albumin, 1 mg/ml effects of hypercholesterolaemia. We have examined collagenase, 0.5 mg/ml papain and 5 mM dithiothre- itol for 50–60 mins at 37°C. Cells were dispersed by this question further by comparing IBa in single arterial smooth muscle cells isolated from control mild agitation in low-calcium physiological salt sol- New Zealand White rabbits (NZ) with those from ution. After centrifugation the cells were resus- Watanabe heritable hyperpercholesterolaemic strain pended in similar physiological salt solution, but 11 containing 1.7 mM CaCl2. The cells were stored on (WHHL). WHHL rabbits possess defective recep- ° tors for low density lipoprotein (LDL). This results cover slips at 4 C and used within 6 to 8 hours. in a marked elevation of plasma cholesterol,12–14 Experiments were performed using the whole-cell accumulation of cholesterol in the arterial wall12 configuration of the patch-clamp technique with a and marked atherosclerosis,14 but without the hep- List EPC-7 patch-clamp amplifier. Patch pipettes were pulled from borosilicate glass and had resist- ances of 3 to 5M⍀. The junction potential (Ͻ5 mV) Correspondence: Dr AD Hughes, Clinical Pharmacology, NHLI, between the electrode and the bath solution was St. Mary’s Hospital, Imperial College of Science, Technology and subtracted using the Vp offset on the amplifier and Medicine, South Wharf Road, London, W2 1NY, UK electrode capacitance was compensated elec- Hypercholesterolaemia and calcium channels GF Clunn et al 850 tronically. No compensation was made for series inactivating current is half inactivated, k is a slope ⍀ resistance which was less than 5M . The pipette factor and Inon is the fraction of essentially non- 18 solution contained (mM): NaCl 126, MgCl2 1.2, ethy- inactivating current which is present in these cells. ␤ Ј Ј lene glycol-bis ( -amino-ethyl ether) N,N,N ,N - Kinetics of the fast inactivation process of IBa were tetraacetic acid (EGTA) 2, adenosine 5Ј-triphosphate analysed by fitting data to a mono-exponential func- (Mg salt) 2, tetraethylammonium chloride 10, and tion: HEPES 11 buffered to pH 7.2 with NaOH. The − ␶ I = I ·et/ + I experiments were carried out in ‘high-barium sol- t fast non = = ution’ containing (mM): BaCl2 110 and HEPES 10 Where It current at time t, Ifast amplitude of fast- = buffered to pH 7.2 with TEA-OH to increase the size inactivating current, Inon amplitude of non-inactiv- of the inward current elicited by depolarization and ating current and ␶ = time constant of the fast inacti- to minimise calcium-dependent inactivation of cur- vating current. rents.16 Fits were performed by non-linear regression Voltage-clamp command pulses were provided by using Prism 2.01 (GraphPad Software, USA) or cus- a PC via a Labmaster A/D interface board using com- tom macros written for Excel (Microsoft, USA). Data mercially available software (PClamp 5.5, Axon are means ± s.e.m. of n observations. Comparison of Instruments, CA, USA). Data were recorded on-line results was made using a Student’s t-test for after analogue-to-digital conversion at between 2.5– unpaired data. P Ͻ 0.05 was considered significant. 10 kHz depending on the voltage protocol used. The currents were digitally filtered at 2 kHz and leak currents were subtracted digitally using average values Results

of steady leakage currents elicited by a 10 mV hyper- Comparison of IBa in NZ and WHHL polarizing pulse. Cell capacitance was estimated as described for fast whole cell recording by Lindau IBa evoked by depolarization from a holding poten- − and Neher,17 assuming a specific membrane capaci- tial of 60 mV were larger in cells isolated from NZ tance of 1 ␮F/cm2. All recordings were made at than from WHHL across a range of activating poten- room temperature. tials (Figure 1a). However cells derived from WHHL had smaller capacitance compared with those derived from NZ (WHHL = 28±2pF, NZ = 40±3pF; Drugs and reagents n = 6 for both; P = 0.02). Thus in terms of current + density there was no significant difference between ( )202791 (isopropyl-4-(2,11,3-benzoxadiazol-4-yl)- the I-V relationship obtained in cells isolated from 1,4-dihydro-2,6-dimethyl-5-nitro-3-pyridine-carbox- the two strains of rabbit (Figure 1b) and voltage- ylate was a gift from Sandoz AG (Basel, dependence of activation did not differ significantly Switzerland). All other chemicals were obtained between the two cell types. Current density in NZ- from Sigma (Dorset, UK). and WHHL-derived cells also did not differ significantly in the presence of a near maximal concen- Statistics and data analysis tration (100 nM) of the dihydropyridine calcium channel agonist, (+)202 791 (Figure 1c). Current-voltage (I–V) relationships were obtained by Steady-state inactivation (hinf) of voltage-operated repeated, progressive depolarization to various test calcium channels was examined by using a 6 sec potentials for 200 ms from a holding potential of pre-conditioning protocol which has been pre- −60 mV. The effect of a drug on I–V relationship was viously shown to induce steady-state inactivation of examined after any response to the drug had stabil- voltage-operated calcium channels in NZ cells.18 ised. The peak inward current at each test potential Normalized hinf relationships were similar in cells was measured. Voltage-dependence of activation from both strains (Figure 1d). was derived from the I–V relationships. I–V data was fitted to a modified Boltzmann function: Kinetics of inactivation (g·(V − Vrev) I = The kinetics of the rapid inactivation of I was mea- (V − Vh) Ba 1 + exp(1 − sured from currents evoked by a 250 ms depolariz- k ation to +20 mV from a holding potential of −60 mV. ␶ where Vrev is the reversal potential, Vh is the poten- The time constant of fast inactivation ( fast) was cal- tial required for half activation g is normalized con- culated from a mono-exponential fit of data as ductance and k is the slope factor. described in Methods assuming an effectively non- Steady-state inactivation data were fitted to a inactivating component of the current, as previously 18 ␶ ± = Boltzmann function: described in NZ cells. fast was 77 19 ms (n 5) in NZ-derived cells and 76 ± 17 ms (n = 5) in WHHL- = Imax derived (NS). I + Inon V − V 1 + expͩ hͪ k Discussion where I is the normalized peak current at any poten- Elevated plasma LDL-cholesterol is a recognised risk 19 tial, Imax is the maximum peak current evoked by factor for cardiovascular disease and atheroscler- such a step, Vh is the holding potential at which osis is reported to increase coronary vasoconstric- Hypercholesterolaemia and calcium channels GF Clunn et al 851

᭺ ᭹ Figure 1 Current-voltage (I–V) relationships for IBa in ear artery cells derived from WHHL ( ) and NZ ( ). Currents were evoked by 200 ms depolarizations from −50 mV to +60 mV from a holding potential of −60 mV. (a) Currents expressed in absolute values (pA) = = ± (n 6). The solid lines represent the best fit to a modified Boltzman function as described in Methods. For WHHL Vh 8 5 mV and = ± = ± = ± k 15 62 pA/mV; for NZ Vh 2 3mVk 14 12 pA/mV (NS). (b) Currents expressed in terms of current density (IBa/cell capacitance) (n = 6). Currents were evoked by 200 ms depolarizations from −50 mV to +60 mV from a holding potential of −60 mV after exposure to (+)202 791. (c) Currents in the presence of a dihydropyridine calcium channel agonist, (+)202 791 (100 nM) expressed as current density = + (n 3). (d) Steady-state inactivation (hInf) data with IBa normalized with respect to IBa evoked by a depolariazation to 10 mV following − = ± = ± a 6 sec pre-conditioning period at 80 mV (I/Imax) and pre-conditioning potential (Em). For WHHL, Vh 22 2 mV, k 12 2 pA/mV, ratio = ± = ± = ± = ± of non-inactivating current to total current (Inon/Itotal) 0.15 0.08, and for NZ Vh 25 3 mV, k 10 3 pA/mV, Inon/Itotal) 0.11 0.05 for NZ (n = 4; NS). tion in man.20 Several previous studies in animals have not.24–27 A study of WHHL rabbits did not find have reported that experimental hypercholesterolae- an increase in response to KCl.28 mia and atherosclerosis increase vascular reactivity. These findings contrast with the consistent Hypercholesterolaemia has been reported to increase in reactivity and Ca2+ influx seen in in vitro increase coronary vasoconstrictor responses to nora- studies where isolated arteries or cells are exposed drenaline in vivo in dogs21 and to increase constric- to cholesterol either in the form of liposomes4,6,7 or tor responses to ergonovine, serotonin and nor- as LDL-cholesterol.6 The effect of depletion of cell adrenaline in the perfused hindlimb of cholesterol by inhibitors of hydroxymethyl glutaryl hypercholesterolaemic and atherosclerotic monk- CoA reductase on voltage-activated calcium channel eys.22 Similarly an enhanced vasoconstrictor currents appears not to have been examined in vas- response to ergonovine has been reported in the per- cular smooth muscle cells, but in chick cardiac myo- fused superior mesenteric vascular bed of the cytes neither lipoprotein-depleted serum (which WHHL rabbit.23 increased cell cholesterol) nor mevinolin, a hyd- Hypercholesterolaemia has also been shown to roxymethyl glutaryl CoA reductase inhibitor, affec- increase the reactivity of isolated arterial smooth ted calcium channel current density.32 Our findings muscle in vitro. In both cholesterol fed animals and suggest that calcium channel current densities in WHHL rabbits responses to serotonin and serotonin WHHL cells are not significantly altered, despite the receptor agonists have been consistently reported to prolonged hypercholesterolaemia seen in this be increased24–28 and may be attributable to an model. In fact, in absolute terms calcium channel 29 increase in 5HT2 receptor number. In contrast currents were smaller in WHHL cells although this responses to phenylephrine24,26,28 have generally effect was offset by a reduction in cell capacitance. been reported to be unaffected or reduced. In the Since cell capacitance in NZ-derived cells was simi- case of a depolarizing potassium solution (KCl), lar to that reported by others using this tissue33 this which would be expected to act largely or wholly presumably indicates a reduced cell surface area in via activation of voltage-operated calcium channels, WHHL cells. The reason for this finding is not the literature is contradictory. Some studies of chol- known. The size of vascular smooth muscle cells esterol-fed rabbits have reported increased isolated from WHHL rabbits appears not to have responses of isolated vessels to KCl30,31 while others been measured in previous studies, however Hypercholesterolaemia and calcium channels GF Clunn et al 852 cultured cells isolated from the aorta of Yoshida 9 Hughes AD. Calcium channels in vascular smooth (YOS) spontaneously hypercholesterolaemic rats muscle cells. J Vasc Res 1995; 32: 353–370. have been reported to be smaller than those from 10 Sen L et al. Cholesterol increases the L-type voltage- normocholesterolaemic controls.34 Whether this sensitive calcium channel current in arterial smooth phenomenon is related to defective LDL receptor muscle cells. Circ Res 1992; 71: 1008–1014. expression and the impaired growth characteristics 11 Watanabe Y. Serial inbreeding of rabbits with hereditary hyperlipidemia (WHHL-rabbit). Atherosclerosis of vascular smooth muscle cells isolated from 35,36 1980; 36: 261–268. WHHL animals reported in some studies 12 Tanzawa K et al. WHHL-rabbit: a low density lipopro- remains speculative. tein receptor-deficient animal model for familial hy- Our findings are not consistent with a direct effect percholesterolemia. FEBS Lett 1980; 118: 81–84. of hypercholesterolaemia on voltage-operated cal- 13 Dietschy JM et al. Cholesterol synthesis in vivo and in cium channels in vivo. It should be noted that mem- vitro in the WHHL rabbit, an animal with defective brane fluidity or cholesterol: phospholipid ratios in low density lipoprotein receptors. J Lipid Res 1983; 24: the cell membrane were not measured in this study, 469–480. but the levels of LDL-cholesterol in WHHL would 14 Buja LM et al. Cellular pathology of progressive be anticipated to affect both these parameters on the atherosclerosis in the WHHL rabbit. An animal model basis of previous studies in cholesterol-fed rabbits37–40 of familial hypercholesterolemia. Arteriosclerosis and cultured smooth muscle cells from WHHL have 1983; 3: 87–101. been reported to show increased cholesterol con- 15 Donnelly TM, Kelsey SF, Levine DM, Parker TS. Con- 41 trol of variance in experimental studies of hyperlipide- tent. mia using the WHHL rabbit. J Lipid Res 1991; 32: It is interesting that calcium channel antagonists 1089–1098. have been reported to be ineffective in reducing 16 Hughes AD, Wijetunge S. The action of amlodipine on 39,42–44 atherosclerosis in WHHL rabbits, in contrast voltage-operated calcium channels in vascular smooth to their beneficial effects in cholesterol-fed ani- muscle. Br J Pharmacol 1993; 109: 120–125. mals.39,44,45 This may imply that voltage-operated 17 Lindau M, Neher E. Patch-clamp techniques for time- calcium channels may play variable roles in the gen- resolved capacitiance measurements in single cells. eration of atherosclerosis depending on the model Pflugers Arch 1988; 411: 137–146. of hypercholesterolaemia. Further studies will 18 Hering S, Hughes AD, Timin EN, Bolton TB. Modu- necessary to explore this issue and better define the lation of calcium channels in arterial smooth muscle role of membrane fluidity as a (patho) physiological cells by dihydropyridine enantiomers. J Gen Physiol influence on voltage-operated calcium channels in 1993; 101: 393–410. 19 Steinberg D. The cholesterol controversy is over. Why vascular smooth muscle. did it take so long? Circulation 1989; 80: 1070–1078. 20 Ginsburg R et al. Quantitative pharmacologic Acknowledgements responses of normal and atherosclerotic isolated human epicardial coronary arteries. Circulation 1984; This work was supported by a grant from the British 69: 430–440. Heart Foundation. We are grateful to Dr M Jacobs for 21 Rosendorff CJ et al. Cholesterol potentiates the coron- the Watanabe rabbits used in this study. ary artery response to norepinephrine in anesthetised and conscious dogs. Circ Res 1981; 48: 320–329. References 22 Heistad DD et al. Augmented responses to vasoconstrictor stimuli in hypercholesterolemic and atheros- 1 Stubbs CD. Membrane fluidity: structure and dynamics clerotic monkeys. Circ Res 1984; 54: 711–718. of membrane lipids. Essays in Biochemistry 1983; 19: 23 Dalessandri KM et al. Superior mesenteric artery 1–39. vasoactivity in hyperlipidemic Watanabe rabbits ver- 2 Yeagle PL. Modulation of membrane function by chol- sus normal lipidemic New Zealand controls. J Invest esterol. Biochimie 1991; 73: 1303–1310. Surg 1989; 2: 471–477. 3 Ortega A, Mas Oliva J. Cholesterol effect on enzyme + + 24 Henry PD, Yokayama M. Supersensitivity of atheros- activity of the sarcolemmal (Ca2 + Mg2 )-ATPase from clerotic rabbit aorta to ergonovine: mediation by a ser- cardiac muscle. Biochim Biophys Acta 1984; 773: otonergic mechanism. J Clin Invest 1980; 66: 306–313. 231–236. 25 Yokoyama M et al. Hyperreactivity of coronary arterial 4 Bialecki RA, Tulenko TN. Excess membrane cholesterol alters calcium channels in arterial smooth mus- smooth muscles in response to ergonovine from rab- cle. Am J Physiol 1989; 257: 306–314. bits with hereditary hyperlipidemia. Circ Res 1983; 53: 5 Stepp DW, Tulenko TN. Alterations in basal and sero- 63–71. tonin-stimulated calcium permeability and vasoconstr- 26 Verbeuren TJ et al. Effect of hypercholesterolemia on iction in atherosclerotic aorta. Arterioscler Thromb vascular reactivity in the rabbit. I. Endothelium- 1994; 14: 1854–1859. dependent and endothelium-independent contrac- 6 Gleason MM, Medow MS, Tulenko TN. Excess mem- tions and relaxations in isolated arteries of control and brane cholesterol alters calcium movements, cytosolic hypercholesterolemic rabbits. Circ Res 1986; 58: calcium levels, and membrane fluidity in arterial 552–564. smooth muscle cells. Circ Res 1991; 69: 216–227. 27 Verbeuren TJ et al. Effect of hypercholesterolemia on 7 Bialecki RA, Tulenko TN, Colucci WS. Cholesterol vascular reactivity in the rabbit. II. Influence of treat- enrichment increases basal and agonist-stimulated cal- ment with dipyridamole on endothelium-dependent cium influx in rat vascular smooth muscle cells. J Clin and endothelium-independent responses in isolated Invest 1991; 88: 1894–1900. aortas of control and hypercholesterolemic rabbits. 8 Zhou Q, Jimi S, Smith TL, Kummerow FA. The effect Circ Res 1986; 59: 496–504. of cholesterol on the accumulation of intracellular cal- 28 Kolodgie FD, Virmani R, Rice HE, Mergner WJ. Vascu- cium. Biochim Biophys Acta 1991; 1085: 1–6. lar reactivity during the progression of atherosclerotic Hypercholesterolaemia and calcium channels GF Clunn et al 853 plaque. A study in Watanabe heritable hyperlipidemic 38 Gillies P, Robinson C. Decreased plasma membrane rabbits. Circ Res 1990; 66: 1112–1126. fluidity in the development of atherosclerosis in chol- 29 Kalkman HO, Neumann V, Brauner V. Supersensi- esterol-fed rabbits. Atherosclerosis 1988; 70: 161–164. tivity of atherosclerotic rabbit aorta to ergometrine in 39 Hara A, Taketomi T. Characterization and change of mediated by 5-HT2 receptors. J Pharm Pharmacol phospholipids in the aorta of Watanabe hereditable 1989; 41: 876–878. hyperlipidemic rabbit. Jpn J Exp Med 1990; 60: 311– 30 Yokoyama M, Henry PD. Sensitization of isolated can- 318. ine coronary arteries to calcium ions after exposure to 40 Rosenfeld ME et al. Lipid composition of aorta of Wat- cholesterol. Circ Res 1979; 45: 479–486. anabe heritable hyperlipemic and comparably hyper- 31 Ibengwe JK, Suzuki H. Changes in mechanical cholesterolemic fat-fed rabbits. Plasma lipid compo- responses of vascular smooth muscle to acetylcholine, sition determines aortic lipid composition of noradrenaline and high potassium solution in hyper- hypercholesterolemic rabbits Arteriosclerosis 1988; 8: cholesterolaemic rabbits. Br J Pharmacol 1986; 87: 338–347. 395–402. 41 Ecsedi GG, Virag S. Accumulation of cholesterol ester 32 Tan WT et al. Effect of low-denity lipoproteins, mevin- + in cultured smooth muscle cells of congenital hyper- olin, and G proteins on Ca2 response in cultured cholesterolemic (WHHL) rabbits treated with normal chick atrial cells. Am J Physiol 1993; 265: H191–H197. and WHHL rabbit plasma LDL. Methods Find Exp Clin 33 Benham CD, Hess P, Tsien RW. Two types of calcium Pharmacol 1986; 8: 535–542. channels in single smooth muscle cells from rabbit ear 42 Boissonneault GA, Heiniger HJ. 25-Hydroxycholes- artery studied with whole-cell and single-channel re- 45 cordings. Circ Res 1987; 61: I10–I16. terol-induced elevations in Ca uptake: permeability 34 Kolpakov V et al. Reduced smooth muscle cell regen- changes in P815 cells. J Cell Physiol 1985; 125: 471– eration in Yoshida (YOS) spontaneously hypercholes- 475. terolemic rats. Atherosclerosis 1994; 111: 227–236. 43 Tilton GD et al. Failure of a slow channel calcium 35 Oguchi A et al. Vascular smooth muscle cells from antagonist, verapamil, to retard atherosclerosis in the genetically hyperlipidemic rabbit (WHHL rabbit) exhi- Watanabe heritable hyperlipidemic rabbit: an animal bit decreased growth response. Atherosclerosis 1991; model of familial hypercholesterolemia. J Am Coll Car- 90: 101–108. diol 1985; 6: 141–144. 36 Ikeda U et al. Involvement of LDL receptor in the pro- 44 Ishikawa Y et al. Nifedipine-suppressed atheroscler- liferation of vascular smooth muscle cells. Athero- osis in cholesterol-fed rabbits but not in Watanabe sclerosis 1994; 110: 87–94. heritable hyperlipidemic rabbits. Atherosclerosis 37 Gillies P, Robinson C, Chapple R. Effects of hypercho- 1987; 64: 79–80. lesterolemia on the microsomal membrane fluidity of 45 Jackson CL, Bush RC, Bowyer DE. Mechanism of anti- intimal-medial versus medial layers of swine aorta: atherogenic action of calcium antagonists. implications for the pathogenesis of vasospasm. Exp Atherosclerosis 1989; 80: 17–26. Mol Pathol 1987; 47: 90–97.