Activation of Myosin Light Chain Phosphatase in Intact Arterial Smooth Muscle During Nitric Oxide-induced Relaxation*

We investigated whether myosin light chain phosphatase activity changes during nitric oxide-induced relaxation of contracted intact carotid media and how changes in phosphatase activity mediate this relaxation. We also investigated one mechanism for regulating this phosphatase. Myosin phosphatase activity, myosin light chain phosphorylation, guanosine 3 (cid:1) ,5 (cid:1) -cyclic mono-phosphate (cGMP) concentration, and phosphorylation of the inhibitory protein CPI-17 were all assayed in homogenates of one carotid media ring at each time point during nitric oxide-induced relaxation. The application of sodium nitroprusside to histamine-contracted media caused rapid declines in light chain phosphorylation and force. These were temporally correlated with a rapid elevation of cGMP and a large transient increase in myosin phosphatase activity. During the early response to nitroprusside, when force declined, increases in myosin phosphatase activity, concurrent with cGMP-mediated decreases in calcium and myosin light chain kinase activity, could accelerate light chain dephosphorylation. CPI-17 was dephosphorylated upon application of nitroprusside at the same time that myosin phosphatase activity increased, suggesting that the removal of inhibition by phospho-CPI-17 contributed to the increase in myosin phosphatase activity. After 20 min of nitroprusside, (Bio-Rad), and a chemiluminescent substrate (ECL RPN 22109). X-ray film (Kodak Bio- Max) was exposed by the luminescent signal for several different dura-tions (30, 60, and 180 s) to record the signals from all the dilutions. The film was scanned to produce digital images using a Molecular Dynamics laser densitometer. Image analysis software (NIH Image) was used to measure the integrated intensity of the bands in digital images. A few immunoblots were visualized using the colorimetric Amplified Opti- 4CN substrate (Bio-Rad) and scanned on a Hewlett Packard ScanJet IIc scanner, and the images were analyzed using UN_SCAN_IT software. Previous comparisons indicated that values for MRLC phosphorylation obtained by the two methods were similar. The integrated intensity of the dephosphorylated band was plotted against the log of the dilution (a measure of the relative amount of protein) for each exposure. Using the linear portion of the resulting curve and the measured intensities of the phosphorylated bands, values for dilution (or relative amount of pro- tein) of the phosphorylated MRLC were obtained by interpolation. p d where d and p are the dilutions of the dephosphorylated and phosphorylated MRLC, respectively, was used to calculate the percentage phosphorylated for a given sample. Several measurements (from the different film exposures) were obtained for each sample that varied by (cid:5) 5%. The homogenates centrifuged at 14,000 (cid:3) and the supernatant was subjected to radioimmunoassay for cGMP content (48) the Diabetes Core Lab the Medicine. The data were expressed as picomol

The phosphorylation of Ser-19 on the 20-kDa myosin regulatory light chains (MRLC) 1 is the primary determinant of cross-bridge attachment and cycling during contraction and relaxation in smooth muscle (1,2). The extent of phosphorylation is determined by the balance of the activities of the Ca 2ϩ -calmodulin-activated myosin light chain kinase (MLCK) and the myosin light chain phosphatase (MLCP) (3). Initially MLCP was presumed to be constitutively active, and MRLC phosphorylation and steady-state force were reported to be proportional to myoplasmic [Ca 2ϩ ] (4). Relaxation occurred when a stimulus was removed, the myoplasmic [Ca 2ϩ ] and MLCK activity were reduced, and MRLC phosphorylation declined. However, relaxation can be induced in the presence of excitatory stimuli without proportional decreases in steadystate MRLC phosphorylation by various treatments (5)(6)(7)(8)(9) including nitrovasodilators. These relaxations could involve a change in MLCP activity.
Nitric oxide released from endothelial cells, nonadrenergic noncholinergic inhibitory innervation, or exogenous nitrovasodilators is an important inhibitory agent in smooth muscle. Nitric oxide activates soluble guanylyl cyclase to generate cGMP (10). Elevated [cGMP] presumably activates cGMP-dependent protein kinase (PKG). The phosphorylation of several intracellular mechanisms by PKG results in a reduction in myoplasmic [Ca 2ϩ ] (11).
Nitrovasodilators such as sodium nitroprusside (SNP) and nitroglycerin cause a rapid relaxation of activated swine carotid media that is associated with an increase in [cGMP] and decreases in [Ca 2ϩ ] i and MRLC phosphorylation (6). In carotid media submaximally contracted by 3 M histamine, relaxation was attributed to nitrovasodilator-induced decreases in [Ca 2ϩ ] i with proportional decreases in MRLC phosphorylation (6). In media maximally contracted by 10 M histamine, there was a decrease in the Ca 2ϩ sensitivity of MRLC phosphorylation (phosphorylation was disproportionately low for the measured [Ca 2ϩ ] i ) (6,12) early during the NO-induced relaxation. This decrease was not caused by the phosphorylation of site A on MLCK or a change in MLCK activity (12) and therefore could reflect an increase in MLCP activity. The steady-state force measured later during NO-induced relaxation was much lower than would be predicted by the measured MRLC phosphorylation (6). Changing the dependence of force on MRLC phosphorylation could be the result of changes in MLCP activity or caused by some MRLC phosphorylation-independent regulatory mechanism (thin filament-associated) that supersedes regulation by phosphorylation (13).
cGMP and cGMP-activated PKG decrease the Ca 2ϩ sensitivity of contraction in permeabilized smooth muscle at constant [Ca 2ϩ ] i (14,15). Nonhydrolyzable analogs of cGMP accelerate MRLC dephosphorylation in the absence of MLCK activity in permeabilized tissues, implying that cGMP mediates an increase in MLCP activity (16,17). Measurements of MLCP activity in intact tissues are few (18 -20). None have demonstrated activation of MLCP.
Several potential mechanisms for the regulation of MLCP activity have been described. The phosphorylation of Thr-695 on the 130-kDa targeting subunit of MLCP (MYPT1) can reduce MLCP activity (21,22). Rho-associated protein kinase (23) and ZIP-like kinase (24) phosphorylate Thr-695 on MYPT1. Both the GTP␥S-induced decrease in the rate of MRLC dephosphorylation and the GTP␥S-enhanced phosphorylation of MYPT1 appeared to be mediated by Rho-associated protein kinase in permeabilized ileum (25), suggesting that agonists acting through G proteins reduce MLCP activity through the Rho-associated protein kinase-mediated phosphorylation of MYPT1. Arachidonic acid may inhibit MLCP activity by dissociating the holoenzyme (26,27) or activating a kinase that phosphorylates MYPT1 (28). The activation of protein kinase C leads to an inhibition of MLCP activity in intact tissue and Ca 2ϩ sensitization in permeabilized smooth muscle (20,29) that is mediated by the MLCP inhibitory protein, CPI-17 (30,31). Phosphorylation of Thr-38 on CPI-17 converts this smooth muscle-specific (32) protein to a potent and specific inhibitor of MLCP activity (33,34). Phosphorylated CPI-17 inhibits the PP1 catalytic subunit (33). Several kinases including protein kinase C (33), Rho-associated protein kinase (35), protein kinase N (36), and ZIP-like kinase (37) phosphorylate CPI-17 on Thr-38 in vitro. In intact femoral artery, Thr-38 of CPI-17 was phosphorylated in response to vasoconstrictors including histamine (38), suggesting that CPI-17 mediates agonist-induced inhibition of MLCP. There is evidence that PKG activates MLCP (and other phosphatases (39)). PKG was shown to bind directly to MYPT1 and to phosphorylate it in vitro (40). Nakamura et al. (41) reported that the in vitro phosphorylation of MYPT1 had no effect on MLCP activity to isolated myosin light chains and suggested a more indirect mechanism for MLCP activation in vivo. Sauzeau et al. (42) reported that PKG inhibited the membrane anchoring of RhoA, thereby potentially inhibiting RhoA-mediated sensitization of contraction to Ca 2ϩ in permeabilized vascular tissues.
Our objectives in this study were to determine whether MLCP activity changes during the SNP-induced relaxation of maximally contracted intact carotid media and to investigate how the changes in MLCP activity are involved in regulating this relaxation. Adjacent rings from a single artery were frozen at various times during relaxation and multiple assays were performed in each arterial ring to obtain a time course of simultaneous changes in [cGMP], MLCP activity, MRLC phosphorylation, and force. We also compared changes in [Thr-38]CPI-17 phosphorylation and MLCP activity to investigate whether CPI-17 phosphorylation regulates MLCP activity in intact carotid media.

EXPERIMENTAL PROCEDURES
Tissues-Swine common carotid arteries were obtained from a slaughterhouse and transported at 4°C in physiological salt solution, and experiments were performed within 2 days. Media tubes were dissected from intact arteries by peeling connective tissue and adventitia away from the media (43). A series of 8 or 16 adjacent rings (3-mm wide) cut from one media tube and de-endothelialized were mounted on Grass (Quincy, MA) FT.03 force transducers and maintained at 37°C in physiological saline solution containing: 140 mM NaCl, 4.7 mM KCl, 1.2 mM Na 2 HPO 4 , 1.6 mM CaCl 2 , 5.6 mM D-glucose, 1.2 mM MgCl 2 , 2.0 mM MOPS buffer (pH 7.4 at 37°C), and 0.02 EDTA. Isometric force was recorded continuously. Each ring was stretched to its optimum length for force development and then maximally contracted for 10 min with 109 mM K ϩ physiological saline solution to assess viability (43). Tissues that did not produce a stress greater than 1.0 ϫ 10 5 N/m 2 during maximal contraction were discarded.
Experimental Protocol-Carotid media rings were stimulated to contract by 10 M histamine. When steady-state contraction was achieved after 10 min, 30 M SNP was applied in the continued presence of histamine. Each point in the time course of MLCP activity or MRLC phosphorylation was measured in a different ring from the same artery.
Individual rings were quick-frozen at the appropriate time during contraction or relaxation by rapid submersion in liquid nitrogen or an acetone/dry ice slurry. One half of each frozen ring was used to assay MLCP activity. The remaining half was homogenized to measure MRLC phosphorylation and CPI-17 phosphorylation or alternatively homogenized to assay [cGMP]. In some experiments two rings were frozen at each time point so that all four assays could be performed in the same artery.
Isolation of Myosin-Native smooth muscle myosin was isolated from frozen turkey gizzard (PelFreez) using the method of Sellers et al. (44) with the following modifications. The ionic strength of the actomyosin extract of washed turkey gizzard myofibrils was elevated slowly by adding 0.6 M KCl, 20 mM MgSO 4 , and 5 mM ATP (pH 7.0). After the two precipitations with (NH 4 ) 2 SO 4 , the pellet was dissolved in and then dialyzed overnight against 5 mM KH 2 PO 4 , 0.6 M KCl, 1 mM EGTA, 1 mM NaN 3 , and 0.5 mM DTT at pH 7.0. The dialysate was centrifuged at 27,000 ϫ g for 45 min. The solubilized myosin was dialyzed two times for 3 h in 5 mM KH 2 PO 4 , 20 mM KCl, 1 mM EGTA, 1 mM NaN 3 , and 0.5 mM DTT (pH 6.5) and then dialyzed overnight in the same solution with 20 mM MgCl 2 at pH 6.5. The precipitated myosin was centrifuged at 30,000 ϫ g for 30 min. The pellet was reconstituted in and then dialyzed twice against 20 mM KH 2 PO 4 , 0.6 M KCl, 2 mM EGTA, 2 mM NaN 3 , and 0.5 mM DTT (pH 7.5). Glycerol (50%) was added, and the aliquots were stored at Ϫ20°C.
Phosphatase Activity Assay-MLCP activity in intact smooth muscle was assayed using the method of Gong et al. (19) with modifications. Liquid nitrogen-frozen carotid media rings were weighed, pulverized at Ϫ60°C, and homogenized in a glass homogenizer on ice in 50 mM Tris-HCl (pH 7.5), 0.1 mM EGTA, 0.1 mM EDTA, 0.1% ␤-mercaptoethanol, 1 mM benzamidine, and 25 g/ml leupeptin and aprotinin in the proportion of 1 mg of tissue (wet weight)/10 l of buffer. This slurry was further homogenized with a Teflon pestle in an Eppendorf tube on ice. Homogenates were diluted 25-fold in ice-cold 25 mM Tris-HCl (pH 7.5), 0.1 mM EDTA, and 0.1% ␤-mercaptoethanol. After 3 min of warming to 25°C in the water bath, reactions were initiated by adding 10 l of 32 P-labeled myosin substrate per 40 l of diluted homogenate. The reaction was stopped at 0, 1, 3, 6, 9, 12, and 15 min after the addition of substrate by pipetting a 50-l aliquot of reactants into 100 l of cold 25% trichloroacetic acid and 25 l of 5 mg/ml bovine serum albumin. The stopped reactions were incubated 10 min on ice and then centrifuged for 3 min at 13,000 ϫ g. One hundred microliters of the supernatant were counted by the Cerenkov method. The 32 P i released due to phosphatase activity was expressed as a percentage of the 32 P-labeled myosin present to correct for the possibility that different amounts of somewhat insoluble myosin substrate were pipetted into each reaction tube; the cpm for each sample was divided by the cpm for the total myosin present in 50 l of reactants. Series of measurements at different times were fitted with rising single exponentials (which often collapsed to straight lines) to determine the initial reaction rate (Fig. 1). The rates were normalized by the protein concentration (measured by the Bradford method (46)) of the homogenate. We expressed MLCP activity as the percentage of P i released/(min⅐g homogenate protein). The activity for each artery was normalized by the stoichiometry of the substrate to enable comparison between arteries. This method was validated by results showing that incremented additions of unphosphorylated myosin to reactions caused classic competitive inhibition of MLCP activity. Daily assays of "blank" reactions of 32 P-labeled myosin substrate without homogenate exhibited undetectable phosphatase activity.
Myosin Light Chain Phosphorylation-Frozen pieces of media rings were slowly thawed in acetone over a 2.5-h period, air-dried, weighed, and homogenized in 1% (w/v) SDS, 10% (v/v) glycerol, and 20 mM DTT on ice. The phospho and dephospho forms of MRLC were separated by isoelectric focusing in glycerol urea gels with a pI ampholyte range of 4.0 -6.5. To provide an internal calibration that covers the potential range of variations in the amounts of the two MRLC species, a series of homogenate dilutions (1, 1/2, 1/4, 1/8, 1/16, and 1/32) of each homogenate was loaded onto each gel (47). The protein was transferred to nitrocellulose and immunodetected using a polyclonal smooth muscle light chain-specific primary antibody (Dr. James Stull), a horseradish peroxidase-conjugated anti-rabbit secondary antibody (Bio-Rad), and a chemiluminescent substrate (ECL RPN 22109). X-ray film (Kodak Bio-Max) was exposed by the luminescent signal for several different durations (30, 60, and 180 s) to record the signals from all the dilutions. The film was scanned to produce digital images using a Molecular Dynamics laser densitometer. Image analysis software (NIH Image) was used to measure the integrated intensity of the bands in digital images. A few immunoblots were visualized using the colorimetric Amplified Opti-4CN substrate (Bio-Rad) and scanned on a Hewlett Packard ScanJet IIc scanner, and the images were analyzed using UN_SCAN_IT software. Previous comparisons indicated that values for MRLC phosphorylation obtained by the two methods were similar. The integrated intensity of the dephosphorylated band was plotted against the log of the dilution (a measure of the relative amount of protein) for each exposure. Using the linear portion of the resulting curve and the measured intensities of the phosphorylated bands, values for dilution (or relative amount of protein) of the phosphorylated MRLC were obtained by interpolation. The ratio, p/(d ϩ p), where d and p are the dilutions of the dephosphorylated and phosphorylated MRLC, respectively, was used to calculate the percentage phosphorylated for a given sample. Several measurements (from the different film exposures) were obtained for each sample that varied by Ͻ 5%.
Cyclic GMP Concentration-Pieces of frozen rings were thawed in acetone, air-dried, weighed, and homogenized in 0.1 N HCl at 0°C. The homogenates were centrifuged at 14,000 ϫ g for 10 min, and the resulting supernatant was subjected to radioimmunoassay for cGMP content (48) by the Diabetes Core Lab at the University of Virginia School of Medicine. The data were expressed as picomol of cGMP/mg of tissue dry weight.
[Thr-38]CPI-17 Phosphorylation-Phosphorylation of [Thr-38]CPI-17 was measured as described by Kitazawa et al. (38). Aliquots of homogenates used to measure MRLC phosphorylation that contained 400 M PEFABLOC (protease inhibitors) and 1 M microcystin-LR were subjected to SDS-polyacrylamide gel electrophoresis (49). The proteins were transferred to nitrocellulose. Phospho-[Thr-38]CPI-17 was immunolabeled using an anti-phospho-[Thr-38]CPI-17 antibody. Then the blot was stripped for 30 min with 50 mM imidazole-HCl containing 1% SDS and 1% ␤-mercaptoethanol at 55°C and was reprobed with an anti-CPI-17 antibody to determine the total amount of CPI-17 present. The relative phosphorylation of [Thr-38]CPI-17 was obtained from the ratio of the intensities of phospho and total CPI-17 labeling on the blots.
Statistics-The data are expressed as means Ϯ S.E. Statistical significance was determined by Student's t test. The number of measurements is given by n. A p value less than 0.05 was considered significant.

RESULTS
Assay of MLCP Activity-Native whole myosin phosphorylated by MLCK was used as substrate to selectively measure myosin light chain phosphatase activity. Although several types of phosphatase can dephosphorylate isolated 20-kDa myosin light chains, only MLCP dephosphorylates intact myosin at appreciable rates (50 -52).
The homogenates assayed in this study were diluted 25-fold during the assay reaction. This was determined by comparing the time course of dephosphorylation of 32 P-labeled myosin by four different dilutions of one homogenate (Fig. 1A). The 25-fold dilution of this homogenate is shown by filled circles (2 l/50 l). The plot of the initial rate versus homogenate concentration (Fig. 1B) shows that at this dilution phosphatase activity was linearly proportional to the amount of homogenate (phosphatase) present. The reaction was not substrate limited at the phosphatase activities measured in this study. The measurement of phosphatase activity in a given homogenate was reproducible; on average, the difference between duplicate measurements of a given homogenate was 15% of the measurement (n ϭ 10).
The phosphatase inhibitor okadaic acid has been used to distinguish the activities of different phosphatases. Concentrations near 1 nM completely inhibit protein phosphatase type 2A, whereas concentrations near 1 M are required to inhibit PP1 (53). MLCP is considered to be a PP1 phosphatase (50). Myosin phosphatase activity measured in carotid artery homo-genates was unaffected by concentrations of okadaic acid up to 1 nM, indicating that protein phosphatase type 2A contributed little to the MLCP activity. For comparison, the phosphorylase phosphatase activity of carotid homogenates, which would include both PP1 and protein phosphatase type 2A, was inhibited 22-40% by 1 nM okadaic acid. Okadaic acid at 1 M inhibited 88% of myosin phosphatase activity, which is consistent with a predominantly PP1-type activity.
SNP-induced Relaxation and Activation of MLCP Activity-The application of 30 M SNP to histamine-activated media rings caused a 66 Ϯ 5.8% (n ϭ 10) reduction in force. The force declined rapidly and reached the lowest level at about 4 min ( Fig. 2A). Over the next 10 min a slight rebound was seen in the force traces for individual arteries before the force leveled off at a low final steady state.
MRLC phosphorylation decreased rapidly during the first 2 min of SNP treatment (Fig. 2B). The average decrease was 10.2 Ϯ 1.8% MRLC phosphorylation (n ϭ 9) or 50% of suprabasal phosphorylation. In some individual arteries, MRLC phosphorylation increased after 5 min in the continued presence of SNP.
MLCP activity increased 3-4-fold during the SNP-induced relaxation (Fig. 2C). MLCP activity rose within 1 or 2 min, reached a peak value by 5 min, and then declined over the next 10 -15 min to the level measured in the presence of histamine alone. Variations in the early time course and in the time to peak among different arteries underlie the irregular rise to peak in the average activity shown in Fig. 2. MLCP activity measured at 5 min of SNP was significantly different from the activity measured after 10 min of histamine alone (p ϭ 0.001, n ϭ 10). The cpm for each measurement was normalized by the cpm for the total 32 P-labeled myosin present in 50 l of reactants to obtain the value for the percentage of P i released. The time courses were fitted using a rising exponential to determine the initial rate. B, the initial rates are plotted as a function of homogenate concentration (l/50 l).
The concentration of cGMP was low and did not change during 40 min of 10 M histamine alone (data not shown). As expected, SNP caused a rapid increase in [cGMP] within 2 min (Fig. 2D). The [cGMP] level remained elevated 2-4-fold for at least 30 min.
Dephosphorylation of CPI-17-[Thr-38]CPI-17 phosphorylation (Fig. 3) was measured as well as MLCP activity, [cGMP], and MRLC phosphorylation in five arteries. When SNP was applied in the presence of histamine, CPI-17 phosphorylation rapidly declined during the first 5 min. The decrease in CPI-17 phosphorylation coincided in time with the rise in [cGMP] in response to SNP (Fig. 2D) and the rise in MLCP activity. At later times, although [cGMP] remained significantly elevated, CPI-17 phosphorylation rebounded at the same time that MLCP activity declined to basal levels. There was a close temporal correlation between the reciprocal changes in CPI-17 phosphorylation and MLCP activity, particularly during the early response to SNP in the presence of histamine.

DISCUSSION
The NO-induced relaxation in the presence of excitatory stimuli has been attributed to cGMP-mediated decreases in [Ca 2ϩ ] i and MLCK activity (11). There is also a decrease in the Ca 2ϩ sensitivity of MRLC phosphorylation early during the relaxation (12). Our results demonstrate that the SNP-induced relaxation involves a large transient increase in MLCP activity. During the initial response to SNP, the large increase in MLCP activity at the same time as a decrease in Ca 2ϩ -dependent MLCK activity (resulting in a decrease in the Ca 2ϩ sensitivity of MRLC phosphorylation) could accelerate MRLC dephosphorylation and the decline in force. Decreases in MRLC phosphorylation resulting from changes in both MLCK and MLCP activity determine the decline in force during this early period of relaxation.
At later times during the NO-induced relaxation, maintenance of the low force may involve additional MRLC phosphorylation-independent mechanisms. In the continued presence of SNP and elevated [cGMP], MLCP activity declined to levels measured before SNP was applied. Other studies (6,9) have shown that the dependence of force on MRLC phosphorylation is altered or uncoupled during sustained relaxation in response to nitrovasodilators. During the nitroglycerin-induced relaxation of maximally activated carotid media, [Ca 2ϩ ] i and MRLC phosphorylation rebounded toward levels measured in the absence of the nitrovasodilator, and force remained low (6). The decline in MLCP activity we report, in conjunction with changes in Ca 2ϩ -dependent MLCK activity, could account for an elevated steady-state MRLC phosphorylation. However, the uncoupling of force from phosphorylation suggests that some additional regulatory mechanism is determining the steadystate force. Cyclic GMP-mediated processes such as phosphorylation of telokin (54,55) or phosphorylation of the thin filament-associated protein HSP20 (56, 57) may contribute to an inhibition of cross-bridge interaction with actin thin filaments that results in sustained low force.
The rise in MLCP activity and decline in CPI-17 phosphorylation coincided with the rise in [cGMP], suggesting that the initial change in MLCP activity was cGMP-mediated and could involve PKG. After the first 5 min of SNP, [cGMP] remained elevated, however there was a decline in MLCP activity and a rebound in CPI-17 phosphorylation.
CPI-17 phosphorylation was highest after 10 min in histamine when MLCP activity was lowest. The time course of dephosphorylation of CPI-17 in response to SNP coincided reciprocally with the rise in MLCP activity, suggesting that de- Histamine (10 M) was applied 10 min before SNP and remained present when SNP was present. The rings from each artery were frozen after 10 min of histamine activation (0 min) and at various times after the application of SNP. Values for the ratio of the signal intensity of anti-phospho-[Thr-38]CPI-17 labeling to anti-total CPI-17 labeling were normalized by the value measured during histamine activation in that artery. Thus the changes in CPI-17 phosphorylation are relative. phosphorylation of CPI-17 resulted in disinhibition of MLCP activity. The physiological phosphatase(s) for phospho-CPI-17 is yet to be determined. Kinetic analysis indicated competitive and noncompetitive-mixed inhibition of MLCP activity by phospho-CPI-17 (58). Therefore it is possible that phospho-CPI-17 bound to the catalytic site of MLCP is quite slowly dephosphorylated by MLCP itself in situ, and dephospho-CPI-17 dissociates from MLCP. Surks et al. (40) reported that PKG directly associates with the MYPT1 subunit of MLCP and suggested that this interaction was necessary for a cGMP-mediated dephosphorylation of MRLC. The direct binding of PKG on MYPT1 might activate MLCP to dephosphorylate phospho-CPI-17. Thus, the cGMP-associated increase in MLCP activity might be caused, in part, by a removal of inhibition by CPI-17. The rebound in CPI-17 phosphorylation and decline in MLCP activity suggest that the disinhibition was overcome in the continuous presence of histamine.
We directly measured an increase in MLCP activity in homogenates of intact carotid media. Our results indicate that relaxation induced by a physiologically important pathway mediated by NO and cGMP involves the activation of MLCP activity in intact arterial smooth muscle. The increase in MLCP activity early during the relaxation would accelerate MRLC dephosphorylation attributable to the decline in MLCK activity and enhance the drop in force. The previously described decrease in Ca 2ϩ sensitivity of MRLC phosphorylation is a correlate of the increase in MLCP activity. When MLCP activity returns to basal levels, the sustained low force seems to be determined by other (cGMP-mediated) MRLC phosphorylation-independent mechanisms. We also provide evidence that the phosphorylation/dephosphorylation of CPI-17 plays a role in the activation of MLCP activity during cGMP-mediated relaxation.