cyclic gmp-dependent stimulation reverses g-protein-coupled inhibition of smooth muscle myosin light...

JOBNAME: BBRC 220#3 PAGE: 1 SESS: 7 OUTPUT: Thu May 2 01:54:11 1996/xypage/worksmart/tsp000/69829b/43

Cyclic GMP-Dependent Stimulation Reverses G-Protein-CoupledInhibition of Smooth Muscle Myosin Light Chain Phosphatase

Xuqiong Wu,* Avril V. Somlyo,*,† and Andrew P. Somlyo*,‡,1

*Department of Molecular Physiology and Biological Physics, †Department of Pathology, and ‡Department of InternalMedicine, University of Virginia, 1300 Jefferson Park Avenue, Charlottesville, Virginia 22908

Received February 12, 1996

8-bromo-cyclic guanosine monophosphate (8-br-cGMP), at constant Ca2+, accelerated the dephosphorylationof the regulatory myosin regulatory light chain and relaxation of permeabilized rabbit ileum smooth muscle.These effects were independent of myosin light chain kinase activity. Similarly, 8-br-cGMP relaxedGTPgS-induced force and inhibited the GTPgS-induced increase in myosin light chain phosphorylation atconstant [Ca2+]. We conclude that cyclic GMP-dependent kinase, activated by 8-br-cGMP, increases smoothmuscle myosin light chain phosphatase activity and counteracts the inhibition of the latter enzyme by G-protein(s). © 1996 Academic Press, Inc.

Dephosphorylation of the regulatory light chain (MLC20) of smooth muscle myosin by myosinlight chain phosphatase (SMPP-1M) can be inhibited by a G-protein-coupled process, resulting inincreased MLC20 phosphorylation (1–3). Because the level of force developed by smooth muscleis primarily dependent on the level of MLC20 phosphorylation achieved at a given ratio of theactivities of the Ca2+, calmodulin-dependent myosin light chain kinase (MLCK) and SMPP-1M,inhibition of SMPP-1M can increase force even in the absence of a change in cytoplasmic [Ca2+]i(“Ca2+-sensitization”). We had suggested that the reverse mechanism, Ca2+-independent activationof SMPP-1M, could lead to enhanced dephosphorylation of MLC20 and consequent relaxation ofsmooth muscle (“desensitization to Ca2+”; 2). In view of previous reports that cyclic adenosine(cAMP) and guanosine (cGMP) monophosphate as well as the catalytic subunits of the cAMP- andcGMP-dependent protein kinase can relax permeabilized smooth muscles at constant [Ca2+] (4–6),cGMP was considered to be a likely candidate messenger of SMPP-1M activation. Cyclic GMP isa physiologically significant smooth muscle relaxant second messenger that is produced in re-sponse to agents such as atrial natriuretic peptide, endothelium-derived relaxing factor (nitricoxide), and exogenous nitrovasodilators (rev. in 7–9). Although these agents and stable cGMPanalogs, such as 8-bromo-cGMP (8-br-cGMP), have multiple effects associated with relaxation ofintact smooth muscle, including reduction of cellular Ca2+, their effects on myosin light chainphosphorylation and force can be examined independently of changes in [Ca2+] in permeabilizedsmooth muscle. Using this approach we show that cyclic GMP activates SMPP-1M and that thisactivation counteracts the inhibition of SMPP-1M activity by GTPgS.

MATERIALS AND METHODS

Sheets of longitudinal muscle were peeled from the rabbit ileum and cut into small strips, 200–250mm wide and 3–4 mmlong. Strips or muscle bundles of similar dimensions were cut from rabbit femoral artery and bladder as well as pig bladderfor force measurements. Muscle strips were stretched to 1.3 times resting length and attached to a force transducer (AE801;

1 To whom correspondence should be addressed. Fax: (804) 982-1616. E-mail: [email protected]: MLC20, 20 kDa myosin light chain; SMPP-1M, smooth muscle myosin light chain phosphatase; MLCK,

myosin light chain kinase; cAMP, adenosine 39:59-cyclic monophosphate; cGMP, guanosine 39:59-cyclic monophosphate;8-br-cGMP, 8-bromo-guanosine 39:59-cyclic monophosphate; GTP-gS, guanosine 59-[g-thio]triphosphate; ML-9, 1-(5-chloronaphthalene-1-sulfonyl)-H -hexahydro-1,4-diazepine; ATP-gS, adenosine 59-[g-thio]triphosphate; G-kinase, cGMP-dependent protein kinase, respectively.

BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS220,658–663 (1996)ARTICLE NO. 0460

6580006-291X/96 $18.00Copyright © 1996 by Academic Press, Inc.All rights of reproduction in any form reserved.


AME, Horten, Norway) in a well on a temperature-controlled “bubble” plate. All experiments were carried out at 22–24°C unless otherwise indicated. After reproducible contractile responses to 154 mM K+ were observed, the strips wereincubated in normal relaxing solution (G1: Ca2+-free, 1 mM EGTA) for about 5 min and permeabilized by incubation for30–45 min at room temperature with 5,000–10,000 units/ml (based on rabbit red blood cell hemolysis)Staphylococcusaureusa-toxin (List Biological Laboratories, Campbell, CA) in pCa 6.5 or pCa 6.3 solution with 10mM A23187 to depleteintracellular calcium stores (10,11). Details of the Ca2+-buffered (with 10 mM EGTA) solutions used for studies ofpermeabilized strips have been published (11). For measurements of MLC20 phosphorylation the muscles mounted on thebubble plate were rapidly transferred into liquid nitrogen cooled Freon 22 and cut away from the hooks.Frozen samples were freeze-substituted in 10% trichloroacetic acid in acetone, subjected to two-dimensional gel elec-

trophoresis and stained with colloidal gold as described previously (10,11). MLC20 phosphorylation was quantified bydensitometry with a Bio-Rad GS-670 imaging densitometer. The percentage of MLC20 phosphorylation is given as (mono-phosphorylated + diphosphorylated)/total MLC20 (phosphorylated + unphosphorylated).For MLC20 phosphorylation measurements experiments were carried out at lower temperatures (15°C, 12°C or 5°C),

in order to obtain relatively tonic contractile response to calcium and to slow the rate of dephosphorylation of thephasic (12–14) rabbit ileum smooth muscle in order to better time-resolve these measurements. Results are presented asmean ± S.D.

RESULTS

Dephosphorylation of MLC20 and Relaxation of Ca2`-Induced Tension Induced by 8-br-cGMPat Constant [Ca2`]

Dose-response curves of 8-br-cGMP in rabbit ileum smooth muscle indicated its relaxant effectat 1mM (data not shown), with maximal relaxation (80–90% of the contractile response to pCa 6.2)achieved at 100mM (Fig. 1). Similar responses were found with preparations of rabbit femoral

FIG. 1. Acceleration of relaxation and dephosphorylation of MLC20 by 8-br-cGMP ina-toxin permeabilized rabbitileum smooth muscle contracted with pCa 6.2 at 12°C. Exposure to pCa 6.2 resulted in a typical phasic contraction whichdecayed to approximately 50% of peak force. Muscle strips were frozen (arrows) at 10 min after exposure to pCa 6.2 formeasurements of MLC20 phosphorylation levels. (A): Control. (B): 100mM 8-br-cGMP treated. %4 MLC20 phosphor-ylation. The difference in MLC20 phosphorylation between “control” and “8-br-cGMP-treated” groups was highly signifi-cant, p < 0.001. Representative of n4 11 experiments.

Vol. 220, No. 3, 1996 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

659


artery and bladder as well as with pig bladder. To determine whether the relaxation of Ca2+-inducedtension could be attributed to dephosphorylation of MLC20 a-toxin permeabilized rabbit ileumsmooth muscle strips were contracted with pCa 6.2 solution at 12°C with or without subsequentaddition of 100mM 8-br-cGMP (4 min after contraction was induced by pCa 6.2). Muscles werefrozen at identical times (10 min) after the initiation of contraction. In the presence of 8-br-cGMP,phosphorylation of MLC20 decreased to 28 ± 6.9% (n4 11) compared to 37 ± 6.5% (control) (n4 11) (Fig. 1). These results were significantly different (p < 0.01), suggesting that the cGMP-induced relaxation at constant calcium is the result of a decrease in the ratio of MLCK/SMPP-1Mactivities.

Acceleration of MLC20 Dephosphorylation and Relaxation by 8-br-cGMP Independent ofMLCK Activity

To determine whether the 8-br-cGMP-induced desensitization to Ca2+ is mediated by inhibitionof MLCK or stimulation of SMPP-1M activity, we measured its effect on the rate of dephosphory-lation of MLC20 and relaxation at 5°C ina-toxin permeabilized rabbit ileum smooth muscle inwhich MLCK activity was blocked (11,13). Muscles contracted with a pCa 6.2 solution weretransferred to Ca2+-free (10 mM EGTA) relaxing solution containing ML-9, an inhibitor of MLCK,with or without 8-br-cGMP (100mM). Relaxation was significantly faster in the presence of8-br-cGMP: t1/2 4 2.4 min vs. 3.5 min in controls (n4 3 each). MLC20 phosphorylation deter-mined at 2 min and 5 min after transfer into relaxing solution with ML-9 showed that dephos-phorylation was significantly (p < 0.001) faster in the presence of 8-br-cGMP: at 2 min, phos-phorylated MLC20 was 37 ± 6.9% (n4 16) in controls and 22 ± 5.1% (n4 16) with 8-br-cGMP;by 5 min the respective values (19 ± 13.1% vs. 17 ± 6.1%) were not significantly different. SinceMLCK activity was blocked in the Ca2+-free solution (G10) containing 50mM ML-9, this resultindicates that 8-br-cGMP accelerated dephosphorylation of MLC20 by stimulation of SMPP-1Mactivity.

Lack of Effect of 8-br-cGMP on MLC20 Thiophosphorylation

To further verify that the reduction of MLC20 phosphorylation was due to activation of SMPP-1M, rather than inhibition of MLCK, we used ATP-gS as substrate, because MLC thiophosphory-lation resists dephosphorylation (11). Ileum smooth muscle strips (a-toxin permeabilized) wereincubated in Ca2+-free, ATP-free solution at 15°C for 20 min to deplete endogenous ATP, followedby ATP-free pCa 6.2 with or without 100mM 8-br-cGMP for five min. Subsequently contractionand MLC20phosphorylation were initiated by addition of ATPgS. MLC20 thiophosphorylation wasnot affected by 8-br-cGMP: it was in, respectively, control and 8-br-cGMP-treated muscles 56 ±2.7% vs. 56 ± 2.8% at 2 min, and 62 ± 1.5 vs. 60 ± 0.5 at 5 min (n4 3 each), suggesting that8-br-cGMP has no effect on MLCK activity. Similarly, pre-incubation with 8-br-cGMP (100mM)had no effect on the time course of the contractions initiated by addition of 2 mM ATP-gS (datanot shown).

Relaxation of GTP-gS-Sensitized Force by 8-br-cGMP

A 7- to 8-fold increase in the steady state tension induced by GTP-gS (50mM) at pCa 6.2 wasrelaxed to 40–50% of its original level by 5mM 8-br-cGMP, and maximally relaxed to 20–30% ofthe original GTP-gS-potentiated tension by 100mM 8-br-cGMP (Fig. 2).

Inhibition of GTP-gS-Induced Increase in MLC20 Phosphorylation by Pretreatmentwith 8-br-cGMP

To further determine the effect of 8-br-cGMP on the GTP-gS-induced increase in MLC20phosphorylation at constant [Ca2+], we pre-incubateda-toxin-permeabilized rabbit ileum smooth


660


muscle strips with 8-br-cGMP (100mM) at 15°C for 5 min before addition of GTP-gS (50mM).This pre-treatment with 8-br-cGMP inhibited the subsequent GTP-gS-induced calcium-sensitization of force (Fig. 3) and significantly (p < 0.001) reduced the GTP-gS-induced increasein MLC20 phosphorylation (27 ± 6.1%, n4 21), compared to that (40 ± 6.3%, n4 21) of controls(Fig. 3).

DISCUSSION

The major findings of our study are that 8-br-cGMP can stimulate SMPP-1M activity and thatthis activation can counteract the inhibition of the enzyme by the G-protein-coupled pathwayactivated by GTPgS. Our results, obtained in permeabilized smooth muscles in which any inter-ference by changes in [Ca2+] was eliminated by depleting intracellular Ca2+-stores with A23187and buffering Ca2+ with 10 mM EGTA, are consistent with previous studies showing cyclicnucleotide-induced relaxation of smooth muscle independently of changes in [Ca2+] (4–6). Wehave excluded the possibility that the effects of 8-br-cGMP were due to inhibition of MLCK,because they persisted when MLCK activity was blocked by ML-9 and Ca2+-free conditions.Furthermore, 8-br-cGMP had no effect on the rate of thiophosphorylation by ATPgS that is notsensitive to phosphatase activity, nor did it slow or reduce the amplitude of the contractionassociated with thiophosphorylation. These results are also consistent with reports indicating thatphosphorylation of MLCK by cGMP-dependent protein kinase is confined to a single, “physi-ologically irrelevant” site and does not inhibit MLCK activity (15,16).The activation of MLC20 phosphatase activity by 8-br-cGMP was sufficient to not only relax

permeabilized smooth muscle contracted with Ca2+, but also relaxed the GTPgS-induced contrac-tion and inhibited the GTPgS-induced increase in MLC20 phosphorylation that is due to inhibition

FIG. 2. Reversal of GTP-gS-induced calcium-sensitization of force by 8-br-cGMP. Rabbit ileum smooth muscle waspermeabilized witha-toxin in the presence of 10mM A23187 at room temperature. Force records show initial phasiccontraction induced by pCa 6.2, followed by GTP-gS-induced contraction (Ca2+-sensitization) at constant [Ca2+] andsubsequent relaxation of latter by 8-br-cGMP (100mM). Representative of n4 20 experiments.


661


of the SMPP-1M activity (11). The previously reported (5) relaxant effect of G-kinase itselfindicates that the effects of 8-br-cGMP observed by us were due to activation of this kinase,through a mechanism that is currently under investigation. However, we note that G-kinase phos-phorylates a substrate in smooth muscle that migrates on SDS gels as az132 kDa protein and isnot MLCK or caldesmon (17). The regulatory, M110subunit of SMPP-1M also migrates as az130kDa protein (18–20), and we are currently exploring the possibility of an activating phosphoryla-tion site on SMPP-1M. Alternatively, such regulation may be the result of inhibitory phosphory-lation of an SMPP-1C inhibitor (1).

ACKNOWLEDGMENTS

We thank Barbara Nordin for preparation of the manuscript. Supported by NIH grant HL19242.

REFERENCES

1. Somlyo, A. P., Kitazawa, T., Himpens, B., Matthijs, G., Horiuti, K., Kobayashi, S., Goldman, Y. E., and Somlyo, A. V.(1989)Adv. Prot. Phosphatases5, 181–195.

2. Somlyo, A. P., and Somlyo, A. V. (1994)Nature372,231–236.3. Gong, M. C., Iizuka, K., Nixon, G., Browne, J. P., Hall, A., Eccleston, J. F., Sugai, M., Kobayashi, S., Somlyo, A. V.,

and Somlyo, A. P. (1996)Proc. Natl. Acad. Sci. USA93, 1340–1345.4. Pfitzer, G., Hofmann, F., DiSalvo, J., and Ruegg, J. C. (1984)Pfluger Arch.401,277–280.5. Pfitzer, G., Merkel, L., Ruegg, J. C., and Hofmann, F. (1986)Pfluger Arch.407,87–91.6. Nishimura, J., and van Breemen, C. (1989)Biochem. Biophys. Res. Commun.163,929–935.7. Warner, T. D., Mitchell, J. A., Sheng, H., and Murad, F. (1994)Advances in Pharmacology26, 171–194.

FIG. 3. Inhibition of GTP-gS-induced calcium-sensitization ofa-toxin permeabilized rabbit ileum smooth muscle bypre-treatment with 8-br-cGMP (100mM). A phasic contraction was induced by pCa 6.2 at 15 °C. (A): 8-br-cGMP (100mM)accelerated relaxation and inhibited the subsequent contraction induced by the addition of 50mM GTP-gS. Representativeof n > 10independent experiments. Muscle strips were frozen at the arrows for measurements of MLC20 phosphorylation.(B): control. MLC20 phosphorylation % was significantly depressed in the 8-br-cGMP-treated muscles compared withcontrols. p!0.001.


662


8. Francis, S. H., and Corbin, J. D. (1994)Annu. Rev. Physiol.56, 237–72.9. McDaniel, N. L., Rembold, C. M., and Murphy, R. A. (1994)Can. J. Physiol. Pharmacol.72, 1380–1385.10. Kitazawa, T., Kobayashi, S., Horiuti, K., Somlyo, A. V., and Somlyo, A. P. (1989)J. Biol. Chem.264,5339–5342.11. Kitazawa, T., Masuo, M., and Somlyo, A. P. (1991)Proc. Natl. Acd. Sci. USA88, 9307–9310.12. Himpens, B., Matthijs, G., Somlyo, A. V., Butler, T. M., and Somlyo, A. P. (1988)J. Gen. Physiol.92, 713–729.13. Gong, M. C., Cohen, P., Kitazawa, T., Ikebe, M., Masuo, M., Somlyo, A. P., and Somlyo, A. V. (1992)J. Biol. Chem.

267,14662–14668.14. Kitazawa, T., and Somlyo, A. P. (1990)Biochem. Biophys. Res. Comm.172,1291–1297.15. Hathaway, D. R., Konicki, M. V., and Coolican, S. A. (1985)J. Mol. Cell Cardiol.17, 841–850.16. Nishikawa, M., de Lanerolle, P., Lincoln, T. M., and Adelstein, R. S. (1984)J. Biol. Chem.250,8429–8436.17. Sarcevic, B., Robinson, P. J., Pearson, R. B., and Kemp, B. E. (1990)Biochem. J.271,493–499.18. Alessi, D., MacDougall, L. K., Sola, M. M., Ikebe, M., and Cohen, P. (1992)Eur. J. Biochem.210,1023–1035.19. Shirazi, A., Iizuka, K., Mosse, C., Fadden, P., Somlyo, A. P., Somlyo, A. V., and Haystead, T. A. J. (1994)J. Biol.

Chem.269,31598–31606.20. Shimizu, H., Ito, M., Miyahara, M., Ichikawa, K., Okubo, S., Konishi, T., Naka, M., Tanaka, T., Hirano, K., Hartshorne,

D. J., and Nakano, T. (1994)J. Biol. Chem.269,30407–30411.


663

cyclic gmp-dependent stimulation reverses g-protein-coupled inhibition of smooth muscle myosin light...

Documents