G Protein-Mediated Ca²⁺-Sensitization of CPI-17 Phosphorylation in Arterial Smooth Muscle

Abstract

CPI-17 is a unique phosphoprotein that specifically inhibits myosin light chain phosphatase in smooth muscle and plays an essential role in agonist-induced contraction. To elucidate the in situ mechanism for G protein-mediated Ca²⁺-sensitization of CPI-17 phosphorylation, α-toxin-permeabilized arterial smooth muscle strips were used to monitor both force development and CPI-17 phosphorylation in response to GTPγS with varying Ca²⁺-concentrations. CPI-17 phosphorylation increased at unphysiologically high Ca²⁺ levels of pCa ≤6. GTPγS markedly enhanced the Ca²⁺ sensitivity of CPI-17 steady-state phosphorylation but had no enhancing effect under Ca²⁺-free conditions, while the potent PKC activator PDBu increased CPI-17 phosphorylation regardless of Ca²⁺ concentration. CPI-17 phosphorylation induced by pCa 4.5 alone was markedly inhibited by the presence of PKC inhibitor but not ROCK inhibitor. In the presence of calyculin A, a potent PP1/PP2A phosphatase inhibitor, CPI-17 phosphorylation increased with time even under Ca²⁺-free conditions. Furthermore, as Ca²⁺ concentration increased, so did CPI-17 phosphorylation rate. GTPγS markedly enhanced the rate of phosphorylation of CPI-17 at a given Ca²⁺. In the absence of calyculin A, either steady-state phosphorylation of CPI-17 under Ca²⁺-free conditions in the presence of GTPγS or at pCa 6.7 in the absence of GTPγS was negligible, suggesting a high intrinsic CPI-17 phosphatase activity. In conclusion, cooperative increases in Ca²⁺ and G protein activation are required for a significant activation of total kinases that phosphorylate CPI-17, which together overcome CPI-17 phosphatase activity and effectively increase the Ca²⁺ sensitivity of CPI-17 phosphorylation and smooth muscle contraction.

1. Introduction

CPI-17, a potent myosin phosphatase (MLCP) inhibitor protein, plays a critical role in regulating smooth muscle contraction [1, 2]. We recently demonstrated that CPI-17 is a remarkable Ca²⁺-dependent messenger that mediates GPCR stimulation to MLCP regulation in intact arterial smooth muscle [3], which acts on MLC phosphorylation in addition to the classical Ca²⁺/calmodulin-mediated regulation of MLC kinase (MLCK) [4]. α₁-Agonist increases CPI-17 phosphorylation levels from negligible values at the resting state to about 0.4 mol/mol protein within seconds. This marked phosphorylation increase is coupled to the rapid onset of both MLC phosphorylation and muscle contraction and is also associated with agonist-induced SR Ca²⁺ release and PKC activation but not Ca²⁺ influx or ROCK activation [3]. Furthermore, the significant increase in [Ca²⁺]_i induced by high K⁺ depolarization does not increase CPI-17 phosphorylation. Together, these results indicate that at least one second messenger, very likely diacylglycerol (DAG), works with Ca²⁺ to stimulate Ca²⁺-dependent CPI-17 phosphorylation through Ca²⁺-dependent PKC. However, little is known about the role of Ca²⁺ sensitivity and G-protein activity in regulating in situ CPI-17 phosphorylation in smooth muscle. Here, the mechanism for Ca²⁺-dependent CPI-17 phosphorylation and its effect of G protein activation is investigated in α-toxin-permeabilized arterial smooth muscle, where the SR Ca²⁺ was depleted with Ca²⁺ ionophore A23187 and the [Ca²⁺]_i concentration was clamped with 10 mM EGTA.

2. Materials and Methods

2.1. Tissue preparation, force measurement, and cell permeabilization

All animal procedures were approved by the Animal Care and Use Committee of the Boston Biomedical Research Institute. Strips of rabbit femoral artery smooth muscle were prepared and mounted for force measurements and quick-freezing using liquid nitrogen-cooled propane, as described previously in detail [3, 5]. Briefly, adventitia-free and de-endothelialized smooth muscle strips (70 μm thick, 0.75 mm wide, and 3 mm long) were dissected from rabbit femoral arteries and mounted on a force transducer assembly. Force levels were monitored throughout the experiments. The compositions of external and intracellular solutions were described previously and Ca²⁺ concentrations in the intracellular solutions were clamped with 10 mM EGTA at pH 7.1 [5, 6]. For cell membrane permeabilization, strips were treated for 30 min at 30 °C with 20 μg/ml purified Staphylococcus aureus α-toxin (List, Campbell, CA) at pCa 6.7 and further treated with 10 μM Ca²⁺-ionophore A23187 for 20 min at 25 °C to deplete the sarcoplasmic reticulum of Ca²⁺ and maintain constant cytoplasmic Ca²⁺ as described previously [6,7]. The pCa is defined as −log(molar concentration of free Ca²⁺). Thereafter, the temperature was maintained at 20°C.

2.2. Immunoblotting

Permeabilized femoral artery strips were rapidly frozen and treated as previously described [1, 5]. The strips were dried and homogenized in electrophoresis sample buffer and equal amounts of the same tissue extracts were loaded onto two 15% (w/v) polyacrylamide gels, and the separated proteins transferred to the same nitrocellulose membranes. The membranes were blocked in Tris-buffered saline solution containing 0.05% Tween 20 and 5% nonfat milk and incubated with a primary antibody followed by an alkaline phosphatase-conjugated secondary antibody. The immunoblots were developed with an alkaline phosphatase substrate solution to visualize immunoreactive proteins. The alkaline phosphatase product bands were digitized with a color scanner and analyzed with image processing software (Signal Analytics Co., Vienna, VA). Western blotting experiments were always carried out in duplicate. We compared the ratio of phosphorylated CPI-17 at Thr38 to the total amount of CPI-17 in the paired set of Western blots.

2.3. Statistical analysis

Where applicable, results are expressed as the mean ± SEM. Significance was evaluated using one-way ANOVA or Student’s t-test. A level of p<0.05 was considered to be statistically significant.

3. Results

3.1. Ca²⁺ sensitivity of CPI-17 phosphorylation

To investigate the Ca²⁺ sensitivity of CPI-17 phosphorylation, we used α-toxin-permeabilized smooth muscle to control free [Ca²⁺]_i. In contrast to other cell permeabilization methods, endogenous small proteins, including CPI-17, are retained in α-toxin-permeabilized preparations at levels similar to intact tissues while the cytoplasmic concentration of small molecules such as ATP and EGTA can be controlled [8]. The free Ca²⁺ concentration was buffered with 10 mM EGTA and intracellular Ca²⁺ stores were depleted with A23187 [6,7]. When Ca²⁺ was increased from pCa <8 (no added Ca²⁺ in 10 mM EGTA-containing solution) to pCa 6.7, minimal force was detected (Fig. 1A). Upon increasing to pCa 6, force developed to a level near the maximum level induced by pCa 4.5. The G protein activator GTPγS (30 μM) and PKC activator PDBu (3 μM) markedly enhanced contraction at pCa 6.7 while the enhancing effect of both activators was minimal at pCa >8 and pCa 4.5, suggesting that those activators primarily increase the Ca²⁺ sensitivity of smooth muscle contraction.

Fig. 1.

Effect of 30 μM GTPγS and 3 μM PDBu on the Ca²⁺ sensitivity of force development (A) and CPI-17 phosphorylation (B) in α-toxin-permeabilized rabbit femoral artery smooth muscle. A: Force levels are expressed as a percentage of contraction produced at pCa 4.5 under control conditions (n=6). pCa = −log(molar concentration of free Ca²⁺). pCa >8 is the free Ca²⁺ concentration when no calcium is added to relaxation solution containing 10 mM EGTA at pH 7.1. (B) Paired set of representative western blots for CPI-17 (upper panel) and pThr38 CPI-17 (lower panel) and a summary of phosphorylated levels of CPI-17 measured 7.5 min after addition of Ca²⁺ in the presence and absence of GTPγS or PDBu expressed as a percentage of total phosphorylation (phosphorylated CPI-17/total CPI-17) at pCa 6.7 in the presence of GTPγS for 7.5 min (n=3–6). ND, not determined.

Under control conditions, CPI-17 phosphorylation was negligible from pCa >8 to 6.7, and significantly increased upon further increases in Ca²⁺ concentration to pCa 6 and 4.5 (Fig. 1B). Similar to force enhancement, GTPγS at 30 μM, which is known to produce a maximal effect on MLC phosphorylation and contraction [9], strikingly increased CPI-17 phosphorylation at Ca²⁺ concentrations ranging from pCa 6.7 to 4.5 but not at pCa >8, suggesting an increase in Ca²⁺ sensitivity. In contrast, PDBu increased CPI-17 phosphorylation to a maximal level even at pCa >8, suggesting that the activity of an in situ Ca²⁺-independent CPI-17 kinase, such as a novel PKC isoform is increased.

3.2. Effect of inhibitors on Ca²⁺-induced CPI-17 phosphorylation

CPI-17 phosphorylation at pCa 4.5 was strongly reduced by PKC inhibitor GF-109203X (3 μM) and partially by G protein inhibitor GDPβS (1 mM), while ROCK inhibitor Y-27632 (10 μM) did not significantly affect phosphorylation levels (Fig. 2A) suggesting that high Ca²⁺ activates PKC partially through a G protein-dependent pathway. Elimination of CPI-17 phosphorylation by GF-109203X at pCa 4.5 only slightly affected the maximum contraction (Fig. 2B), suggesting that MCLK was near maximally activated at pCa 4.5 [9].

Fig. 2.

Effect of G protein and protein kinase inhibitors on CPI-17 phosphorylation (A) and maximum contraction (B) at pCa 4.5. Permeabilized arterial smooth muscle strips were pretreated in the presence and absence of 1 mM GDPβS, 3 μM GF-109203X, and 10 μM Y-27632 for 10 min in relaxing solution with 1mM EGTA before Ca²⁺ was increased to pCa 4.5 buffered with 10 mM EGTA for 7.5 min. Panel A shows a paired set of representative western blots for CPI-17 (upper panel) and pThr38 CPI-17 (lower panel) and a summary of CPI-17 phosphorylation levels expressed as the percentage of phosphorylation level at pCa 4.5 for 7.5 min under control conditions (n=4). In Panel B, force levels are expressed as a percentage of maximum contraction produced at pCa 4.5 under control conditions (n=4). * represents statistical significance (p<0.05).

3.3. Rate of CPI-17 phosphorylation when CPI-17 phosphatase is inactivated

CPI-17 phosphatase is a PP1 type phosphatase [10]. To examine the mechanism of the CPI-17 phosphorylation reaction in situ, CPI-17 dephosphorylation was blocked by treatment with the potent PP1/PP2A inhibitor, calyculin A [11]. Permeabilized strips were first incubated in MgATP-free, CP (creatine phosphatase)-free, and Ca²⁺-free solution to dephosphorylate CPI-17 and block kinase activity. Thirty minutes after ATP removal, 1 μM calyculin A was added to inhibit CPI-17 phosphatase activity. Thirty min after calyculin A addition under rigor conditions, 4.5 mM MgATP, 10 mM CP-, and 1 μM calyculin A-containing solution at the indicated Ca²⁺ concentration was exchanged for the rigor solution to initiate the phosphorylation reaction. CPI-17 was time-dependently phosphorylated even at pCa >8 (Fig. 3). Increases in Ca²⁺ to pCa 7 and 6 increased the rate of CPI-17 phosphorylation, suggesting that in situ CPI-17 kinase(s) is active even at near resting Ca²⁺. GTPγS in the presence of calyculin A significantly enhanced CPI-17 phosphorylation even at pCa >8 (Fig. 3) whereas increases in CPI-17 steady-state phosphorylation upon addition of GTPγS were minimal under identical conditions except in the absence of phosphatase inhibitor (Fig. 1B). GTPγS addition resulted in a large increase in the rate of CPI-17 phosphorylation at pCa 7, corresponding to a large enhancement of steady-state phosphorylation at the same pCa in the absence of inhibitor (Fig. 1B). PDBu at 3 μM increased the rate of phosphorylation regardless of Ca²⁺ concentration in the absence of calyculin A (not shown).

Fig. 3.

Effect of Ca²⁺ and 30 μM GTPγS on the CPI-17 phosphorylation rate in the presence of 1 μM calyculin A. MgATP (4.5 mM) and CP (10 mM) were added to initiate the phosphorylation reaction after a 30-min incubation in rigor solution containing calyculin A with (filled symbols) or without (open symbols) GTPγS at the indicated Ca²⁺ concentration (n=3 at pCa 7). Ca²⁺ levels were clamped with 10 mM EGTA.

Discussion

In intact arterial smooth muscle, steady-state CPI-17 phosphorylation is negligible at rest. Upon stimulation with α₁-agonist, CPI-17 phosphorylation levels rapidly increase in response to SR Ca²⁺-release, leading to MLCP inhibition and a robust increase in MLC phosphorylation [3] mediated by Ca²⁺/calmodulin-dependent MLCK [4]. This study, using permeabilized arterial smooth muscle, presents three critical findings: 1) Elimination of CPI-17 phosphatase activity revealed a hidden CPI-17 kinase activity that persists even under Ca²⁺-free conditions. This Ca²⁺-independent kinase activity was enhanced by G protein activation. However, the Ca²⁺-independent basal and GTPγS-enhanced kinase activities together appear to be insufficient to overcome intrinsic CPI-17 phosphatase activity, and, thus, the steady-state phosphorylation level in the presence of GTPγS under Ca²⁺-free conditions in the absence of calyculin A was still negligible (Fig. 1B); 2) Raising Ca²⁺ concentrations to unphysiologically high levels (pCa <6) markedly increased the rate of phosphorylation in the presence of calyculin A and moderately but significantly enhanced steady-state phosphorylation in the absence of calyculin A, which was strongly inhibited by the PKC inhibitor (Fig. 2B). Physiological [Ca²⁺]_i (from pCa 7 to pCa >6) appreciably stimulated CPI-17 kinase activity compared to that measured for Ca²⁺-free conditions (Fig. 3) although both steady-state CPI-17 phosphorylation levels at pCa 6.7 in permeabilized muscle without GTPγS [1; this study] and during high K⁺-induced contraction in intact muscle [3] was minimal, suggesting that the basal and Ca²⁺-dependent kinase activity within the physiological Ca²⁺ concentration range without G protein activation still could not overcome the intrinsic phosphatase activity. These results suggest that the in situ CPI-17 phosphatase activity is higher than the kinase activity at pCa 6.7 and lower than that at pCa 6 in permeabilized arterial smooth muscle. Free Ca²⁺ concentration at pCa 6.7 in the presence of PDBu can evoke a large contraction that is sensitive not only to Ca²⁺-independent but also Ca²⁺-dependent PKC inhibitors in arterial smooth muscle [12], supporting the idea that physiological Ca²⁺ levels can at least partly stimulate Ca²⁺-dependent PKC; 3) GTPγS markedly increased the rate of CPI-17 phosphorylation by several fold at both pCa >8 (Ca²⁺-free conditions) and 7 (Fig. 3). Incremental GTPγS-induced increases in phosphorylation after 2 min in the presence of pCa 7 were considerably greater than the sum of phosphorylation levels at pCa 7 alone and of GTPγS-induced changes for Ca²⁺-free conditions at the same time point, suggesting that G protein activation enhances not only basal but also Ca²⁺-dependent CPI-17 kinase activities. The steady-state phosphorylation in the presence of GTPγS under Ca²⁺-free conditions, however, was negligible while that at pCa 6.7 was maximal, suggesting that the GTPγS-enhanced rate of phosphorylation is still lower under Ca²⁺-free conditions but much higher at pCa 6.7 than that of intrinsic phosphatase activity. In intact arterial smooth muscle, when agonist-induced Ca²⁺ increases are blocked with ryanodine and nicardipine, CPI-17 phosphorylation only slowly increases to low levels [3], consistent with the present results. This study cannot exclude the possibility that G protein activation inhibits CPI-17 phosphatase activity, resulting in an increase in apparent kinase activity. Without considering the possible regulation of CPI-17 phosphatase activity, however, the present data can be explained by a combination of the GTPγS-induced increase in Ca²⁺-dependent and –independent CPI-17 kinases and the constitutively active phosphatase as discussed above. If the phosphatase activity were reduced by G protein activation without changing the kinase activity, the rate of CPI-17 phosphorylation would not be changed with GTPγS in the presence of calyculin A, but this is not the case (Fig. 3). Phosphorylation of CPI-17 Thr38 is known to occur in response to agonists following activation of PKC and ROCK in arterial [1, 3, 12] and integrin-linked kinase (ILK) in intestinal smooth muscles [13]. PDBu, but not GTPγS, enhanced the steady-state phosphorylation of CPI-17 under Ca²⁺-free conditions, suggesting that the total activation of Ca²⁺-independent CPI-17 kinases (nPKC, ROCK, and possibly ILK) by a maximal activation of G protein cannot overcome the intrinsic phosphatase activity and is considerably lower than PDBu-induced activation of nPKC in arterial smooth muscle. If the Ca²⁺-independent CPI-17 kinase were totally inactivated, the Ca²⁺ sensitivity of CPI-17 phosphorylation would be shifted to much higher than physiological Ca²⁺ concentrations, suggesting a physiological role for Ca²⁺-independent activity. Other Ca²⁺-independent kinases, such as zipper-interacting kinase and p21-activated kinase, can also phosphorylate isolated CPI-17 at Thr38 [14, 15]. Further studies are needed to identify which kinase is responsible for the increased in situ CPI-17 phosphorylation under different conditions.

Research highlights.

• Phosphorylation of smooth muscle-specific CPI-17 increases at only high Ca²⁺.

• G protein activation increases Ca²⁺ sensitivity of CPI-17 phosphorylation and smooth muscle contraction, but has no enhancing effect under Ca²⁺-free conditions.

• When phosphatase is blocked, CPI-17 phosphorylation increases with time even at low Ca²⁺.

• Because of high intrinsic CPI-17 phosphatase activity, cooperative increases in Ca²⁺ and G protein activation are required for a significant activation of CPI-17 kinases and smooth muscle contraction.

Abbreviations

CPI-17

protein kinase C-potentiated phosphatase inhibitor protein 17 kDa

pCa

-log(concentration of free Ca²⁺ in molar)

PKC

protein kinase C

PDBu

phorbol 12,13-dibutyrate

ROCK

Rho-associated kinase

PP1

protein phosphatase type 1

PP2A

protein phosphatase type 2A

MLC

myosin light chain

MLCP

myosin light chain phosphatase

MLCK

myosin light chain kinase

SR

sarcoplasmic reticulum

DAG

diacylglycerol