Thapsigargin, A Ca2+-ATPase inhibitor, relaxes rat aorta via nitric oxide formation

Life Sciences, Vol. 54, No. 9, pp. PL 153-158, 1994 Copyright © 1994 Elsevier Science Ltd Printed in the USA. All rights reserved 0024-3205/94 $6.00 + .00

Pergamon

PHARMACOLOGY LETTERS Accelerated Communication

THAPSIGARGIN, A CaZ÷-ATPase INHIBITOR, RELAXES RAT AORTA VIA NITRIC OXIDE FORMATION Hideki Moritoki, Tetsuhiro Hisayama, Wataru Kondoh and Shougo Takeuchi Department of Chemical Pharmacology, Faculty of Pharmaceutical Sciences, University of Tokushima, Shomachi, Tokushima 770, Japan. (Submitted October 13, 1993; accepted October 29, 1993; received in final form December 13, 1993) Abstract. Thapsigargin induced endothelium-dependent relaxation and cGMP production in rat thoracic aorta, and these effects were inhibited by nitric oxide (NO) pathway inhibitors, a calmodulin inhibitor and removal of Ca 2+, suggesting that NO is involved in the thapsigargin-induced relaxation. Thapsigargin may deplete Ca 2÷ stores in the endothelial cells by inhibiting the Ca2+-ATPase, a Ca 2÷ pump, which in turn triggers influx of extracellular Ca 2+, leading to activation of constitutive NO synthase and resultant NO generation. The NO thus formed may activate soluble guanylate cyclase to produce cGMP in the vascular smooth muscle.

The non-phorbol ester-type tumor promotor thapsigargin has been shown to cause a transient increase in intracellular Ca 2÷ in the NG115-401L neuronal cell line in the absence of Ca 2÷ (1). It has been reported that thapsigargin increased intracellular Ca 2+ by potent and highly selective inhibition of the Ca2+-ATPase of inositol trisphosphate-sensitive Ca 2÷ stores (2, 3, 4), and has been used as a probe for studying the processes of storage and release of intracellular Ca z÷. The effect of thapsigargin on Ca 2+ movement in cultured endothelial cells has been examined using the fluorescent indicator fura-2 as a marker (5, 6). However, little is known about the pharmacological effects of thapsigargin on isolated vascular preparations. There are reports that in vascular preparations, thapsigargin causes Ca2+-dependent contraction as a result of inhibition of CaZ+-ATPase in the Ca 2+ stores (7), and increases the amplitude of spontaneous contraction (8), and that it has both contractile and relaxant effects on rat aorta (9). In contrast, thapsigargin selectively blocks agonist-stimulated release of NO and prostaglandin 12 from cultured endothelial cells or perfused artery (10). Since Ca 2+ has been known to play a central role in production of nitric oxide (NO) in the vascular endothelium (11, 12), we examined whether inhibition by thapsigargin of CaZ+-ATPase in the endothelial Ca2+-stores (endoplasmic reticulum) could cause vascular relaxation through activation of NO synthase and resultant NO formation. Materials and m e t h o d s

Oraan bath experiments: Male Wistar rats (8 to 10 weeks old) were used. Ring segments of rat thoracic aorta of 3 mm length were mounted in a 10 ml organ chamber at 34°C containing oxygenated (95 % 02 and 5 % CO2) Krebs solution (pH 7.4) of the following composition (mM): NaC1 115.3, KC14.9, CaC12 2.5, MgSO4 1.2, KH2PO4 1.2, NaHCO3 25.0, glucose 11.1, EGTA disodium salt 0.03 and ascorbic acid 0.11. Ca2+-free solution consisted of Krebs solution without CaC12 and with 2 mM EGTA. The preparations were maintained at 1.0 g resting tension, and were equilibrated for 2 hours before the start of experiments. Responses were measured isometrically. For the measurement of relaxation, the arteries were precontracted with the EC80 concentration of phenylephrine (0.1-4). 3 IIM), and relaxations were plotted as percentages of the phenylephrine-induced contraction. Correspondence: H.Moritoki, Department of Chemical Pharmacology, Faculty of Pharmaceutical Sciences, University of Tokushima, Shomachi, Tokushima 770, Japan.

PL-154

Ca2+-ATPase Inhibitor and NO Formation

Vol. 54, No. 9, 1994

Concentration-response curves were obtained by adding thapsigargin cumulatively. The inhibitors, when used, were added for 20 min before application of thapsigargin. Removal of the endothelium was performed by rubbing the lumen of the artery with cotton thread, and was confirmed by the loss of a relaxant response to 10 ~tM acetylcholine (ACh) at the start and end of the experiments. For study of the effect of removal of Ca 2+, 2 min after replacing normal Krebs solution with Ca2+-free solution, the arteries were incubated with 10 ~tM PGF2ct for 5 min, and then challenged with a single dose of thapsigargin. Assav of cGMP: Arteries were first incubated with 0.3 pM phenylephrine for 5 min, and then challenged with 0.1 I~M thapsigargin for 90 sec. To study the effect of removal of Ca 2÷, 2 min after replacing normal Krebs solution with Ca2+-free EGTA-containing medium, the arteries were incubated with 10 pM PGF2ct for 5 min, and then challenged with thapsigargin. NO pathway inhibitors, when used, were applied before and throughout the incubation with thapsigargin. Amounts of cGMP were measured by radioimmunoassay. Statistical analysis: Values are expressed as mean values + S.E.M. The statistical significance of differences was analyzed by Studenfs unpaired t-test, and p values of less than 0.05 were considered as significant. Materials: Drugs used were thapsigargin, calmidazolium (Sigma Chemical Co., St Louis, MO), LY83583 (6-anilino-5, 8-quinoline-dione, Calbiochem, San Diego, CA), NG-nitro L-arginine and NO-nitro D-arginine (Peptide Institute, Osaka, Japan) and prostaglandin F2et (Ono Pharmaceutical Co. Ltd., Osaka, Japan). Kit for radioimmunoassay of cGMP was obtained from Yamasa Shoyu Co.Ltd. (Choshi, Japan).

Results Relaxant effect of thapsigargin on rat thoracic aorta Cumulative application of thapsigargin at concentrations of above 1 nM caused concentrationdependent relaxation of isolated rat thoracic aorta precontracted with the ECs0 concentration of phenylephrine (0.1-0.3 pM) with an ECs0 value of 5.33 + 1.08 nM, n=5-6 (Fig.l). The maximal relaxation induced by a single dose of 30 nM thapsigargin was attained in 60-90 sec, and the effect was sustained until thapsigargin was washed out. Once thapsigargin had been applied, NNA (v-M)

~

LY 83583 ( ~ M )

100 ~ ~ 30 )

so

CMZ(~M)

1

10

0.3

lOO

9

8 7 6 TG ( - l o g M )

~

7

TG ( - l o g M )

9

8

TG

7

l-logM)

Fig. 1 Thapsigargin (TG)-induced relaxations of rat thoracic aorta, and the effects of NG-nitro L-arginine (NNA), LY83583 (LY) and calmidazolium (CMZ) on the relaxation. The ordinates show relaxations of the arteries expressed as percentages of contraction induced by the EC80 concentrations of phenylephrine (0.1-0.3 ~tM) applied to produce tone. Tissues were exposed to the inhibitors for 20 min before and during application of thapsigargin. Each point represents the means of values (n=5-6) in preparations from 6 to 10 rats; vertical lines indicate S.E.M.

Vol. 54, No. 9, 1994

Ca2+-ATPase Inhibitor and NO Formation

PL-155

phenylephrine no longer induced tone even after washing the preparation for 4 hours. In the arteries removed of the endothelium, thapsigargin failed to induce relaxation even at a concentration as high as 1 glVl (data not shown). Effect of thaDsi~ar~in on cGMP formation In preparations w~th intact endothelium, thapsigargin at a concentration of 0.1 ixM, which was sufficient to induce relaxation, stimulated cGMP production with time to reach a maximum in 90 sec. The effect was sustained for at least up to 20 min. As shown in Fig.2, the cGMP level increased from the basal level of 1.40 + 0.64 pmol/mg protein to 49.63 + 10.83 pmol/mg protein (n=5-6). On the other hand, in the arteries without endothelium, 0.1 txM thapsigargin did not stimulate cGMP production.

60 t-

~

50

[]

Q.

E

0.1 p,M TG

40

'6 E o. 1 0 &.

(.9 o

0 m

m

Endo (- )

NNA 30 I~M

m

m

LY 1 p.M

CMZ 10 ~M

Fig. 2 Thapsigargin (TG)-stimulated production of cGMP, and the effects of NC-nitro Larginine (L-NNA), LY83583 (LY) and calmidazolium (CMZ) and removal of the endothelium (Endo - ) on the cGMP production in the rat thoracic aorta. The ordinate shows amounts of cGMP expressed as pmol/mg protein in 90 sec. Open columns, basal levels of cGMP; closed columns, cGMP levels in the presence of 0.1 ixM thapsigargin. The inhibitors were applied 20 min before and during incubation with thapsigargin. Underlined columns show cGMP formation in the preparations treated with the inhibitors. Each column represents the means of values (n=5~6) in preparations from 10 to 12 rats; vertical lines indicate S.E.M. * p < 0.05, ** p < 0.01, significantly different from the control value without these inhibitors. Effects of comoounds affecting NO pathway on the actions of thapsigargin Treatment of the arteries for 20 min with the NO synthase inhibitor NO-nitro L-arginine (30 IxM) or the soluble guanylate cyclase inhibitor LY83583 (1 laM), inhibited thapsigargin-induced relaxation (Fig. 1) and cGMP production stimulated by 0.1 lard thapsigargin (Fig.2). In contrast to N6-nitro Larginine, its enantiomer NO-nitro D-arginine was without effect (data not shown). The calmodulin inhibitor calmidazolium at a concentration of 10 I~M inhibited the relaxation caused by thapsigargin (Fig. 1) and thapsigargin-stimulated increase in cGMP level (Fig.2), without affecting those induced by nitroprusside (data not shown). Effect of removal of (~a2+ on the actions of thaosi~ar~in In Ca2+-free condition, thapsigargin at concentrations of up to 1 taM did not relax the arteries

PL-156

Ca2+-ATPase Inhibitor and NO Formation

Vol. 54, No. 9, 1994

previously contracted with 10 ~tM PGF2a (Fig.3). In addition, thapsigargin failed to stimulate cGMP production (Fig.3). These conditions attenuated relaxation and cGMP production induced by ACh, without affecting those caused by nitroprusside (Fig.3).

lOO 9

8 7 TG (-tog M)

8

o

60.

9 8 NP(-IogM)

ACh (0.1 p.M)

TG (0.1 I~M) A C

7 6 5 ACh(-Iog M)

NP (0.1 ,uM) 15

12

NS

50. 4(

E

5"

IE Q.

5

4

Q,.

0

0

0

m

m

Ca-free

Ca-free

m

Ca-free

Fig.3 Effect of removal of Ca 2÷ on the thapsigargin (TG)-induced relaxation and cGMP production in rat thoracic aorta. For study of the effect of removal of Ca 2+, 2 min after replacing normal Krebs solution with Ca2+-free EGTA-containing medium, the arteries were contracted with 10 pM PGF2ct, and then challenged with a single dose of thapsigargin. ACh and nitroprusside (NP) were used as references. Upper figures: The ordinates show relaxations of the arteries expressed as percentages of contraction induced by 10 IxM PGF2ct. Open symbols represent responses obtained in normal Krebs solution (2.5 mM Ca2+), and closed symbols represent those obtained in Ca2+-free medium. Lower figures: The ordinate shows amounts of cGMP expressed as pmol/mg protein in 90 sec. Open columns, basal level of cGMP; closed columns, cGMP level in the presence of 0.1 gtM thapsigargin, 0.1 ~tM ACh or 0.1 pM NP; underlined columns, cGMP formation in Ca2+-free medium. Other experimental conditions were as for figs. 1 and 2. Each point and column represent the means of values (n=5) in preparations from 6 to 8 rats; vertical lines indicate S.E.M. * p < 0.05, compared with the value obtained in normal Krebs solution containing 2.5 mM Ca 2÷ (unpaired t-test). NS, not significantly different from corresponding control value. Discussion

The present study clearly demonstrates that the Ca2+-ATPase inhibitor thapsigargin induces endothelium-dependent relaxation of rat thoracic aorta and stimulates production of cGMP, an index of NO formation, suggesting that NO mediates the relaxation. This possibility was supported by the finding that the NO synthase inhibitor NG-nitro L-arginine (13) and the soluble guanylate cyclase

Vol. 54, No. 9, 1994

Ca2+-ATPase Inhibitor and NO Formation

PL-157

inhibitor LY83583 (14) suppressed thapsigargin-induced relaxation and thapsigargin-stimulated cGMP production. The relaxing effect of thapsigargin was entirely dependent on extracellular Ca2+, because in Ca 2+free solution, thapsigargin failed to induce relaxation and did not elevate the cGMP level, whereas these conditions did not influence the effects of nitroprusside. Furthermore, the finding that the calmodulin inhibitor calmidazolium abolished the relaxation and the cGMP formation suggests that thapsigargin-induced relaxation is dependent on calmodulin. It has been reported that endotheliumdependent ACh- and bradykinin-induced relaxations were mediated by Ca2+/calmodulin-dependent constitutive NO synthase (15, 16) and were inhibited by calmodulin inhibitors (17). Therefore, the NO synthase responsible for the vasorelaxing effect of thapsigargin seems to be of the constitutive type that is dependent on CaZ+/calmodulin. In rat thoracic aorta, thapsigargin may act on the Ca 2÷ stores in the endothelial cells to deplete Ca2+ as a result of inhibition of the Ca2÷-ATPase, a Ca 2÷ pump (3, 4), which in turn triggers influx of extracellular Ca 2÷ in a similar way to that proposed previously as capacitative model for parotid acinar cells (2, 18) and cultured vascular endothelial cells (6). The channels mediating thapsigargintriggered Ca 2÷ entry do not seem to be voltage-dependent Ca2÷ channels, because endothelial cells have been suggested to be devoid of voltage-dependent Ca 2÷ channels (19). The increased level of intracellular Ca 2+ may then activate constitutive NO synthase to generate NO, because activation of NO synthase and resultant NO formation are known to depend on the level of free Ca 2÷ in the endothelial cells (11, 12). The NO thus formed in the endothelial cells may activate soluble guanylate cyclase in the vascular smooth muscle to induce cGMP production and resultant relaxation, The Ca2+ that activates NO synthase mediating thapsigargin-induced relaxation may be Ca 2+ entering from the extracellular space, but not Ca2÷ released from Ca2÷ stores in the endothelial cells as a result of inhibition of the Ca2÷-ATPase, a Ca 2÷ pump, because in Ca2+-free conditions, thapsigargin did not induce relaxation or stimulate cGMP production. The sustained relaxation of the rat thoracic aorta observed on application of thapsigargin probably reflect continuous influx of extracellular Ca2+ due to sustained depletion of the store by thapsigargin, because the influx of Ca 2÷ through plasma membrane is thought to be regulated by the Ca2+ content of the stores (2, 18). In summary, our data provide the evidence that the Ca2+-ATPase inhibitor thapsigargin induces NOmediated relaxation of rat aorta by a mechanism involving activation of a constitutive NO synthase as a result of Ca2÷ entry into the endothelial cells.

Acknowledgements This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science and Culture of Japan (03671054), and by grants from the Fujisawa Foundation and the Suzuken Memorial Foundation.

l~eferences 1. 2. 3. 4. 5. 6. 7.

T.R.JACSON, S.I.PA'Iq'ERSON, O.THASTRUP and M.R.HANLEY, Biochem.J. 253 8186 (1988). H. TAKEMURA, A.R.HUGHES, O.THASTRUP and J.W.PUTNEY, Jr., J. Biol. Chem. 264 12266-12271 (1989). O.THASTRUP, Agents Action 29 8-15 (1990). O.THASTRUP, P.J.CULLEN, B.K.DROBAK, M.R.HANLEY and A.P.DAWSON, Proc. Natl. Acad. Sci. USA. 87 2466-2470 (1990). R.J.DOLOR, L.M.HURWITZ, Z.MIRAZA, H.C.STRAUSS and A.R.WHORTON, Am. J. Physiol. 262 (Cell Physiol.3_!) C171-181 (1992). W.P.SCHILLING, O.A.CABELLO and L.RAJAN, Biochem. J. 284 521- 530 (1992). A.M.LOW, V.GASPAR, C.Y.KWAN, P.J.DARBY, J.P.BOURREAU and E.E.DANIEL, J. Pharmacol. Exp. Ther. 258 1105=1113 (1991).

PL-158

8. 9. 10.

Ca2÷-ATPase Inhibitor and NO Formation

Vol. 54, No. 9, 1994

E.O.MIKKELSEN, S.H.POULSEN and S.E.CHR1STENSEN, Pharmacol. Toxicol. 70 152-156 ( 19921. E.O.MIKKELSEN, O.THASTRUP and S.B.CHRISTENSEN, Pharmacol. Toxicol. 62 711 (19881. H.MACARTHUR, M.HECKER, R.BUSSE and J.R.VANE, Br. J. Pharmacol. 108 100-

105 1!993). 11. 12. 13. 14. 15. 16. 17. 18. 19.

A.MULSCH, E.BASSENGE and R.BUSSE, Naunyn-Schmideberg's Arch. Pharmacol. 340 767-770 (1989). K.SCHMIDT, B.MAYER and W.R.KUKOVETZ, Eur.J.Pharmacol. 17_...00 157-166 (1989). P.K.MOORE, O . A . A L - S W A Y E H , N.W.S.CHONG, R.A.EVANS and A.GIBSON, Br. J. Pharmacol. 99 408-412 (1990). E.MALTA, P,S.MACDONALD and G.J.DUSTING, Naunyn-Schmideberg's Arch. Pharmacol. 337 459-464 ( 19881. R.BUSSEandA.M(_ILSCH, F E B S L e t t . 265 133-136(19901. U.FORSTERMANN, J.S.POI,LOCK, H.H.H.W.SCHMIDT, M.HELLER and F.MURAD, Proc. Natl. Acad. Sci. U.S.A. 88 1788-1792 (19911. V.B.SCHINI and P.M.VANHOUTTE, J. Pharmacol. Exp. Ther. 261 553-559(1992). J.W.PUTNEY Jr., Cell Calcium 1 1 , 6 1 1 - 6 2 4 119901. K.TAKEDA, V.SCHINI and H.STOEDIE, Pfluegers Arch. 410 385-393 119871.