J Mol Cell Cardiol22,
Myosin
Leonard
1017-1023
Light
Chain Muscle
P. Adam,
Department of Medicine,
(1990)
and Caldesmon Phosphorylation Stimulated with Endothelin-1*
Lorraine
Blair Brengle
Miliot,
in Arterial
and David
R. Hathaway
and the Krannert Institute of Cardiology, Indiana University School of Medicine, Indianapolis, Indiana 46202, USA
(Received 13 November 1989, accepted in revisedform 22 March 1990) P. ADAM, LORRAINE MILIO, BLAIR BRENGLE AND DAVID R. HATHAWAY. Myosin Light Chain and Caldesmon Phosphotylation in Arterial Muscle Stimulated with Endothelin-1. 3ounmlof Molecular and Cellular Cardiology (1990) 22, 1017-1023. Endothelin-1 contracts porcine carotid arterial smooth muscle with an ED,,, of 10 ILM. Contraction is associated with phosphorylation of the 20000 dalton-regulatory light chain subunits of vascular myosin. Phosphopeptide mapping oflight chains isolated from 32P0,-loaded muscle strips stimulated by endothelin-1 (5 x 10-s M) and comparison with maps generated from light chains phosphorylated in virro or muscles stimulated with KC1 (110 mM) or angiotensin-II (5 x 10-s M) indicates that Ca’+-calmodulin activation of myosin light chain kinase is a biochemical pathway stimulated by all three agonists. However, a small amount of phosphate (17%) was detected in a light chain peptide phosphorylated by protein kinase C. Endothelin-1 also stimulated phosphorylation of the thin filament protein, caldesmon, (from 0.35 mol PO,/mol caldesmon to 0.52 mol PO,/mol). Collectively, these results provide evidence that the effects of endothelin-1 on force generation and maintenance in vascular muscle may be dependent upon myosin light chain phosphorylation by Gas+ calmodulin - requiring myosin light chain kinase and upon a thin filament mechanism that is modulated by phosphorylation of caldesmon.
LEONARD
KEY
WORDS:
Smooth
muscle;
Myosin
light
chains;
Caldesmon;
Introduction Endothelin1 is a 21 amino acid peptide secreted by many different types of cells. In the vasculature, endothelial cells secrete endothelin1 which in turn stimulates the contraction ofvascular smooth muscle [I]. Recent studies have indicated that endothelin-1 arises from a 203 amino acid precursor, preproendothelin, although the enzymes responsible for maturation of the active peptide and the proximate stimuli for secretion have not yet been identified [2]. Although it is clear that endothelin1 causes an immediate increase in intracellular Ca2+ in vascular cells, the nature of signal transduction is not completely understood [3]. Current data suggest that endothelin-I can either directly or indirectly lead to the opening of voltage sensitive Ca2+ channels as well as stimulate the phosphoinositide pathway gen-
Angiotensin;
Endothelin.
erating inositol 1,4,5 triphosphate [4]. The latter should predictably lead to release of Ca2+ from an intracellular storage site that contains a specific inositol 1,4,5 triphosphate receptor [5j. In cultured vascular smooth muscle cells, endothelin-1 stimulates DNA synthesis and cellular proliferation [S]. Moreover, it enhances the generation of diacylglycerol and activates protein kinase C [7]. Activation of protein kinase C has been implicated in many cellular processes, including modulation of intracellular pH and contractility [a]. Phosphorylation of the 20 000 dalton regulatory light chain of myosin by a specific Ca2+ and calmodulin-dependent myosin light chain kinase is the specific signal that initiates crossbridge cycling and, hence, tension generation, in smooth muscle [9]. On the other hand, sustained tension involves modulation of the
*This work was supported in part by grants HL32947 and HL06308 from the National American Heart Association and the Herman C. Krannert Fund. TPresent Address: Department of Obstetrics and Gynecology, University of California Boulevard, Sacramento, California 95816, USA. 0022-2828/90/091017
+ 07 $03.00/O
6
Institutes Davis,
1990 Academic
of Health, 1621
the
Alhambra
Press Limited
L. P. Adam
1018
kinetics of cross-bridge cycling such that detachment rate slows, permitting force maintenance at low crossbridge cycling rates [IO]. The latter has been called the “latch state”. Recent studies have proposed that the actinbinding protein, caldesmon, may modulate actomyosin ATPase kinetics in smooth muscle [II, 121. Moreover, we have shown that agonists which induce the latch state stimulate phosphorylation of caldesmon [13]. In this study we report that endothelin-1 stimulates phosphorylation of the 20000 dalton light chain subunits of myosin in porcine carotid arterial muscle. In addition, it also stimulates phosphorylation of caldesmon. Materials
and Methods
Radiolabeled compounds (Y[~*P]ATP, 32P0, and ‘251-Protein A) were purchased from New England Nuclear. Electrophoresis supplies were obtained from BioRad and EM Science. All other supplies were purchased from the Sigma Chemical Co. Protein pu&ation Calmodulin was purified from bovine testicles by the method of Klee [14]. Myosin regulatory light chains (LC20), caldesmon and myosin light chain kinase were purified from chicken gizzard smooth muscle by methods developed in this laboratory [21,15,16]. Physiological preparations Porcine carotid artery segments (10 mm long and 5 mm wide) were denuded of endothelium, attached to an isometric force transducer (Grass Instrument Co., Fourier transform, 0.03) and bathed in physiological saline solution (PSS) consisting of 140 mM NaCl, 4.7 mM KCI, 1.2 mM Na2HP04, 1.2 mM MgS04, 1.6 mM CaC12, 5.6 mM glucose and 2 mM MOPS’, pH 7.4. Loading of muscles with 32P0, (1.25 mCi/ml) was performed in phosphate-free buffer for 90 minutes at 4g tension and 36°C. Contractile agonists were added as indicated in the text. For KCldepolarization, high potassium buffer ‘Abbreviations: adenosine-5’-triphosphate.
MOPS,
4-morpholine
propane
sulfonic
et al.
(IlOmM) was prepared by substituting KC1 for NaCl in PSS. Incubations were terminated by freeze-clamping muscles with liquid nitrogen-cooled clamps. Protein phosphorylation measurements The stoichiometry of phosphorylation of the 20000 dalton myosin light chains was determined by glycerol-urea gel electrophoresis followed by radioimmunoblotting with an antiserum developed to the chicken gizzard LCzO [17]. Caldesmon phosphorylation was determined as we have recently reported [13]. Briefly, caldesmon was extracted for 5 minutes from 10 mg of ground tissue in 25 ~1 of a buffer consisting of 190 mM NaCI, 6 mM EDTA’, 2oj, sodium dodecylsulfate and 50 mM Tris, pH 7.4. Following dilution with 50 ~1 of the same buffer minus sodium dodecylsulfate, the sample was clarified by sedimentation and the supernatant was collected for immunoprecipitation with rabbit anti-caldesmon antiserum ( 1: 4) and protein-A agarose beads [23]. Following a 2 h incubation, the beads were washed four times with 500 ~1 of buffer containing 500 mM NaCI, 5 mM EDTA, lviO Triton X-100, 0.591, deoxycholate, 0.1 s; sodium dodecylsulfate and 50 mM Tris, pH 7.4. Caldesmon was quantitated by densitometric scanning of Coomassie Blue-stained gels containing immunoprecipitated caldesmon and purified caldesmon standards. Following this, caldesmon bands were excised and counted for 32P0, incorporation. The specific activity of the 32P0, was determined from calculation of the cellular ATP’ specific activity as we have previously described [13]. Peptide mapping Purified L&e was phosphorylated in vitro as we have described [15] and subjected to sodium dodecylsulfate polyacrylamide gel electrophoresis [18] followed by electroblotting onto nitrocellulose [17]. L&e phosphorylated in 32P0,-loaded muscles was collected from urea gels electroblotted onto nitrocellulose. Labeled L&e samples on nitrocellulose were subjected to tryptic digestion (0.1 acid;
EDTA,
ethylene
diamine
tetra
acetic
acid;
ATP,
Emdothelin
and Protein
mg/ml) in 50 mM NHdHC03 for 24 h at room temperature. Samples were lyophilized and resuspended in 20 ~1 of deionized water. Thin layer electrophoresis followed by thin layer chromatography was performed as described by others [19]. Results The comparative dose-response curves for endothelin-1 and angiotensin II are shown in Figure 1. Both agents are potent vasoconstrictors with an EDSo of 10 nM for endothelin-1 and 12 nM for angiotensin-II. Moreover, using porcine carotid arterial muscle, maximal endothelin-1 contractions are 65% that of KC1 contractions, while maximal angiotensinII contractions are 24% of the KC1 maximum. A comparison of the time course of contraction and phosphorylation of the 20 000 dalton myosin light chains is shown in Figure 2. Major differences are evident for the three contractile agonists that were administered at maximal doses (i.e. KCl, 110 mrq endothelin1, 5 x lo-* M; angiotensin, 5 x lo-* M). The typical spike and plateau of L&o phosphorylation observed with KC1 has been reported by several groups [20-221. Endothelin gave a more gradual increase, followed by a decline in LCZo phosphorylation. However, at 60 min, levels of LCZo phosphorylation were still
Phosphorylation
1019
elevated above basal values. On the other hand, angiotesin II gave a brief burst of L&o phosphorylation with a rapid decline to basal values within 15 min. Thus, of the three, only KC1 and endothelin1 produced sustained LCZo phosphorylation. In general, the time course of tension development and maintenance differed among the three agonists. For KC1 and angiotensin, tension developed rapidly but declined to zero for angiotensin while remaining sustained for KCl. Endothelin-1 gave a more gradual increase in tension followed by a slow decline. To more thoroughly compare the three contractile agonists, muscles were loaded with 32P0, and phosphop e p tide maps of LC,, were obtained at peak levels of phosphorylation. The results are summarized in Figure 3. For comparative purposes, L&o phosphorylated in vitro with Y[~~P]-ATP and myosin light chain kinase was mapped as a control. In all cases, a predominant phosphopeptide was identified with essentially identical electrophoretic and chromatographic mobility. This has been shown by others to correspond to phosphorylation of serine-19 on L&o [20]. With endothelin-1, a second phosphopeptide was also detected (see Endo, *). This spot contained approximately 17 f 4% of the total phosphate or a molar stoichiometry of 0.03 mol P04/mol LC20. Using purified LC20 in vitro as substrate we (data not shown) and others have shown this to correspond to a peptide phosphorylated by protein kinase C
PI.
[Ago&]
(nht 1
FIGURE 1. A comparison of the effects of endothelin-1 and angiotensin-II on maximal contractile force in porcine carotid arterial smooth muscle. Muscle strips were prepared and stimulated with agonists as described under Materials and Methods. Each point represents the mean f S.E. of three experiments.
In an earlier study, we demonstrated that the actin-binding protein, caldesmon, was phosphorylated in response to several agonists [13]. In all cases, sustained tension correlated with significant levels of phosphate (i.e. up to 1 mol) incorporated into caldesmon. As shown in Figure 4, all three agonists stimulated phosphorylation of caldesmon at 60 minutes. Moreover, despite basal levels of LC20 phosphorylation and the absence of sustained force, angiotensin II stimulated caldesmon phosphorylation. Discussion Basic investigation regarding the effects and mechanism of action of endothelin-I have progressed rapidly since its discovery 2 years
L. P. Adorn
1020
40
-
et al.
IlOmMKCI
30
0.3
20
0.2 ----..
w..................... 0.1
IO
0
0
20
1
0.4
Endothelin
t
IO
0.4
0.3
5
0.2
0.1
0
0 0
20
40 Time
60
(min)
FIGURE 2. Time course of tension and myosin regulatory arterial smooth muscle. (a) KC1 (110 mM); (b) Endothelin-1 represent the mean + S.E. of at least three experiments.
light chain (LC,,,) phosphorylation in porcine carotid (5 x 10-s M); (c) Angiotensin-II (5 x 10-s M). All data
Endothelin
and Protein
Phosphorylation
MLCK
0.I-
1021 KCI
ELECTROPHORESIS
FIGURE 3. PhosphoReptide maps of my&n regulatory light chain phosphorylated ia Z&O and in intact porcine carotid arterial muscle. For mapping of light chains phosphorylated in intact muscle, muscles were loaded with 32P04 and light chains were separated by glycerol-urea gel electrophoresis as described under Materials and Methods. Following tryptic digestion of light chains electroblotted onto nitrocellulose, peptides were subjected to isoelectric focusing followed by thin layer chromatography [19]. MLCK, purified L&o was phosphorylated in vifro with y3sP-ATP muscles; Endo, phosphopep[16] and subjected to mapping; KCl, phosphopeptide map of LC *,, from KCl-stimulated tide map of LCsc, from endothelin-1 stimulated muscles (asterisk indicates second phosphopeptide); Ang, phosphopeptide map of LCas from angiotensin-II-stimulated muscles.
ago [I]. It has been described as a “potent” vasoconstrictor due both to the low concentrations at which effects are observed and to the prolonged duration of action [I]. The proximate stimuli leading to endothelin-1 release from endothelial cells remain incompletely characterized, but may include such rheological factors as blood flow [Z]. At the molecular level, endothelin-I appears to stimulate the phosphoinositide pathway [4] and to open Ltype Ca * + channels [ZZ]. Activation of protein kinase C via production of diacylglycerol and an increase in intracellular Ca*+ has been reported by others [7j. We have compared endothelin-1 with regard to three parameters; isometric tension,
myosin regulatory light chain phosphorylation and caldesmon phosphorylation to another vasoactive peptide, angiotensin II, and to KC1 depolarization in porcine carotid arterial muscle. In this muscle, angiotensin II and endothelin-1 are roughly equipotent with an EDSOof approximately 10-12 nM when peak tension alone is compared. On the other hand, the onset and duration of tension are prolonged for endothelin1. KC1 depolarization induced a typical tonic contraction as reported by several groups [lo]. Some of the diEerences in onset and duration of the contractile response might be due to agonist diffusion or to other processes such as receptor down-regulation for the peptide hormones.
1022
L. P. Adam
Angiotensm
Endothelin
KCI
FIGURE 4. Phosphorylation of caldesmon in intact porcine carotid arterial muscle. Muscles were pre-loaded with ‘*PO, as described under Materials and Methods and stimulated with angiotensin-II (5 x IO-* M), endothelin-1 (5 x 10-s M) or KC1 (110 IIIM). Caldesmon was purified from muscle extracts by immunoprecipitation and the molar stoichiometry of phosphorylation determined as we have previously described [I3j.
Endothelin-induced contraction was associated with phosphorylation of the 20000 dalton myosin light chains of myosin. Moreover, the major phosphopeptide was identical to that generated by phosphorylation of purified L&e with myosin light chain kinase in vitro. In addition, a similar pattern was observed for KCl-depolarization and angiotensin II. Thus, all three agonistsappear to stimulate the generation of force through the common pathway of Ca’+-camodulin activation of myosin light chain kinase. It is interesting that about 170,bof the total phosphate in L&c from endothelin-stimulated muscles was incorporated into a peptide phosphorylated by protein kinase C [21]. This finding is consistent with the data of others who have measured protein kinase C activation in vascular cells treated with endothelin-1 [fl and further suggeststhat the phosphoinositide pathway is activated by
et al.
endothelin- 1in vascular smooth muscle. However, the stoichiometry of protein kinase Cmediated phosphorylation of L&e was low (0.03 mol/mol) by our measurements,casting some doubt on the significance of this phosphorylation to force maintenance. A novel finding in this study is phosphorylation of the actin-binding protein, caldesmon, in responseto endothelin-1 stimulation. We have previously shown that several agonists which induce sustained contraction of vascular muscle also stimulate phosphate incorporation into this protein [13]. Biochemical studies by us [11] and others have implied a role for caldesmon in the kinetics of crossbridge affinity or cycling in smooth muscle [I,?]. We have suggestedthat phosphorylation of caldesmonmight serve to activate the latch mechanism in vascular muscle which is characterized by slowing of the rate of crossbridge detachment [IO]. Such a mechanism would promote force maintenance but would not generate force, which requires crossbridge cycling. The latter is dependent upon some level of myosin light chain phosphorylation [IO]. An interesting exception to our earlier report, is that angiotensin II stimulated phosphorylation of caldesmon but did not induce sustainedforce. This may be explained by the absenceof LCzO phosphorylation. If the rate of formation of force-bearing crossbridges, which is dependent upon L&e phosphorylation, is lessthan the rate of crossbridge detachment, muscleswould predictably remain relaxed. Thus, in intact muscle, sustained force may require two biochemical mechanisms: some level of myosin light chain kinase activity to form crossbridges through LC2e phosphorylation and a second mechanism to slow detachment rate that requires phosphorylation of caldesmon. Endothelin- 1 appearsto stimulate both processesand this may account for its prolonged duration of action as a vasoconstrictor.
References 1 YANAGISAWA, M., KURIHARA, H., KIMURA, S., etal. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature 332, 41 l-+15 (1988). 2 YANAGISAWA, M., MASAKI, T., Endothelin, a novel endothelium-derived peptide. Biochem Pharmacol 38, 1877-1883 (1989). 3 MASA~I, T. The discovery, the present state, and the future prospects of endothelin. J Cardiovasc Pharmacol 13. Sl-S4 (1989)
Endothelia 4
6 7 8 9 10 II 12 13 14 15 16 17 18 19 20
21 22
Phosphorylation
1023
T., MORRIS, R. C., JR., IVES, H. Endothelin-induced increases in vascular smooth muscle Ca’+ do not on dihydropyridine-sensitive Cazf channels. J Clin Invest 84,635639 (1989). EHRLICH, B. E., WATRAS, J. Inositol 1,4,5-triphosphate activates a channel from smooth muscle sarcoplasmic reticulum. Nature 336,538-586 (1988). KOMURO, I., KURIHARA, H., SUGIYAMA, T., TAKAKU, F., YAZAKI, Y. Endothelin stimulates c-fos and c-myc expression and proliferation of vascular smooth muscle cells. FEBS Lett 2, 249-252 (1988). GRIENDLING, K., TSUDA, T., ALEXANDER, R. W. Endothelin stimulates diacylglycerol accumulation and activates protein kinase C in cultured vascular smooth muscle cells. J Bioi Chem 264,8237-8240 (1989). RASMINEN, H., TAKAWA, Y. PARK, S. Protein kinase C in the regulation ofsmooth muscle contraction. FASEB J 1, 177-185 (1987). ADELSTEIN, R. S., SELLERS, J. R. Effects of calcium on vascular smooth muscle contraction. Am J Cardiol 59, 4B-10B (1987). HAI, C.-M., MURPHY, R. A. Ca’ +, crossbridge phosphorylation, and contraction. Annu Rev Physiol51, 285-298 (1989). LASH, J., SELLERS, J., HATHAWAY, D. The effects of caldesmon on smooth muscle heavy actomeromyosin ATPase activity and binding of heavy meromyosin to actin. J Biol Chem 261, 16155-16160 (1986). SOBUE, K., KANDA, K., TANAKA, T., UEKI, N. Caldesmon: a common actin-linked regulatory protein in the smooth muscle and non-muscle contractile system. J Cell Biochem 37, 3 17-325 ( 1988). ADAM, L., HAEBERLE, J. HATHAWAY, D. Phosphorylation ofcaldesmon in arterial smooth muscle. J Biol Chem 264, 769&7703 (1989). KLEE, C. B. Conformational transition accompanying the binding of Ca’+ to the protein activator of 3’, 5’-monophosphate-dependent phosphodiesterases. Biochemistry 16, 1017-1024 (1977). HATHAWAY, D., HAEBERLE, J. Selective purification of the 20,000-Da light chains of smooth muscle myosin. Anal Biochem 135, 37-43 ( 1983). HATHAWAY, D., KONICKI, M., COOLICAN, S. Phosphorylation of myosin light chain kinase from vascular smooth muscle by CAMP- and cGMP-dependent protein kinases. J Mol Cell Cardiol 17, 841-850 (1985). HATHAWAY, D., HAEBERLE, J. A radioimmunoblotting method for measuring myosin light chain phosphorylation levels in smooth muscle. Am J Physio1249, C345351 (1985). PORZIO, M., PEARSON, A. Improved resolution of myofibrillar proteins with sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Biochim Biophys Acta 490,27-34 (1977). HAEBERLE, J., SUTTON, T., TROCKMAN, B. Phosphorylation of two sites on smooth musce myosin. J Biol Chem 263, 44244429 ( 1988). IKEBE, M., HARTSHORNE, D., ELZING, M. Identification, phosphorylation and dephosphorylation of a second site for myosin light chain kinase on the 20,000 dalton light chain of smooth muscle myosin. J Biol Chem 261, 36-39 (1986). BENGUR, A., ROBINSON, E., APELLA, E., SELLERS, J. Sequence of the sites phosphorylated by protein kinase C in the smooth muscle myosin light chain. J Biol Chem 262, 7613-7617 (1987). SILBERBERG, S., PODER, T., LACERDA, A. Endothelin increases single-channel calcium currents in coronary arterial smooth muscle cells. FEBS Lett 247, 68-72 (1989). MITSUHASKI,
depend
5
and Protein