An essential role of myosin light-chain kinase in the regulation of agonist- and fluid flow-stimulated Ca²⁺ influx in endothelial cells

^a Internal Medicine III, Hamamatsu University School of Medicine, Hamamatsu, Japan
^b Department of Biomedical Engineering, Graduate School of Medicine, University of Tokyo, Tokyo, Japan
^c Life Science Center, Asahi Chemical Ind.,Co. Ltd., Ohito-cho, Shizuoka, Japan
^d Department of Pharmacology, Nagoya University School of Medicine, Nagoya, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cytosolic Ca²⁺ ([Ca²⁺]_i) plays an important role in endothelial cell signaling. Although it has been suggested that the influx of Ca²⁺ can be triggered by depletion of intracellular Ca²⁺ stores, the mechanism (or mechanisms) underlying this phenomenon needs further elaboration. In the present study, involvement of myosin light-chain kinase (MLCK) in the regulation of Ca²⁺ signaling was investigated in agonist- and fluid flow-stimulated endothelial cells loaded with Ca²⁺-sensitive dyes. Bradykinin (BK) and thapsigargin caused an increase in [Ca²⁺]_i followed by a sustained rise due to Ca²⁺ influx from extracellular space and shifted total myosin light-chain (MLC) from the unphosphorylated to the diphosphorylated form. ML-9 (100 µM), an inhibitor of MLCK, abolished Ca²⁺ influx and prevented MLC diphosphorylation in BK- and thapsigargin-treated cells, but did not affect Ca²⁺ mobilization from internal stores. Fluid flow stimulation (shear stress=5 dynes/cm²) increased [Ca²⁺]_i and enhanced MLC phosphorylation. ML-9 also inhibited Ca²⁺ response and MLC phosphorylation in fluid flow-stimulated cells. The Ca²⁺ influx in response to BK was linearly correlated with the diphosphorylation of MLC in ML-9 treated cells. Effects of ML-5 and ML-7, analogs of ML-9, to inhibit Ca²⁺ influx paralleled their potencies to inhibit MLCK activity. These findings demonstrate that MLCK plays an essential role in regulating the plasmalemmal Ca²⁺ influx in agonist- and fluid flow-stimulated endothelial cells. This study is the first to report the close relationship between Ca²⁺ influx and MLC diphosphorylation.—Watanabe, H., Takahashi, R., Zhang, X.-X., Goto, Y., Hayashi, H., Ando, J., Isshiki, M., Seto, M., Hidaka, H., Niki, I., Ohno, R. An essential role of myosin light-chain kinase in the regulation of agonist- and fluid flow-stimulated Ca²⁺ influx in endothelial cells. FASEB J. 12, 341–348 (1998)

Key Words: fluid flow • calcium • bradykinin • thapsigargin

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
IMPORTANT FUNCTIONAL properties of vascular endothelial cells are known to be closely coupled to changes in cytosolic Ca²⁺ concentration ([Ca²⁺]_i) (1–3). An increase in [Ca²⁺]_i in endothelial cells has been shown in response not only to several agonists (4, 5), but also to mechanical stimuli (6, 7). Agonists such as bradykinin (BK)² and histamine initiate rapid formation of inositol 1, 4, 5-triphosphate (IP₃) through the activation of phospholipase C (PLC) (8) and result in an increase in [Ca²⁺]_i in endothelial cells (4, 5). BK-stimulated Ca²⁺ response is biphasic, consisting of a rapid initial peak, followed by a plateau phase. The initial peak reflects, at least in part, the release of Ca²⁺ from IP₃-sensitive intracellular stores, whereas the plateau phase is due to the influx of Ca²⁺ from the extracellular space (4, 5). Numerous studies have suggested that the influx of Ca²⁺ can be triggered by the depletion of intracellular Ca²⁺ stores (9, 10). However, the signal that links the depletion of intracellular Ca²⁺ stores to influx of Ca²⁺ through the plasma membrane remains to be identified. Recently, we have reported that myosin light-chain kinase (MLCK) inhibitors almost completely abolish Ca²⁺ influx across the plasma membrane in agonist-stimulated endothelial cells without affecting the mobilization of Ca²⁺ from intracellular stores (11). However, other protein kinase (or kinases) may also be involved in the regulation of Ca²⁺ signaling after the stimulation of endothelial cells with agonist.

Fluid flow and the associated shear stress have also been suggested to induce activation of PLC and a rapid increase in intracellular levels of IP₃ (12, 13). Although fluid flow is considered to be one of the most important physiological stimuli for production and release of nitric oxide (NO) (14), which is enhanced by the increase in [Ca²⁺]_i (6, 7), the mechanism (or mechanisms) of regulating [Ca²⁺]_i in fluid flow-stimulated endothelial cells is not clearly understood. If fluid flow stimulation also produces IP₃ and results in the increase in [Ca²⁺]_i, the signal transduction cascades initiated by agonist and fluid flow stimulation of endothelial cells may then be thought to share a common role.

In the present study, we wished to clarify the role of MLCK in the regulation of Ca²⁺ signaling, elucidating whether MLCK inhibitor is also able to inhibit the increase in [Ca²⁺]_i in response to the physiological stimulus, fluid shear stress, and whether a causal relationship exists between Ca²⁺ entry and the phosphorylation of MLC in agonist- and fluid flow-stimulated endothelial cells.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Cell culture
Porcine aortic endothelial cells were isolated, as previously described (15), by gently scraping the intima of the descending part of porcine aortas. After centrifugation at 250 x g for 10 min in M199 solution, the pellet of endothelial cells was purified from this suspension, resuspended in M199 solution with Earle's salts, supplemented with 100 IU/ml penicillin G, 100 µg/ml streptomycin, and 20% newborn calf serum (NCS), then aliquoted into polybiphenyl dishes fixed on 10 x 10-mm glass coverslips, and incubated at 37°C in 5% CO₂ for 2 days. The medium was renewed every day.

Measurement of cytosolic calcium concentration
[Ca²⁺]_i was measured as previously described (11) in endothelial cells adhering to glass coverslips. The cells were incubated for 45 min in modified Tyrode's solution (composition in mM: 150.0 NaCl, 2.7 KCl, 1.2 KH₂PO₄, 1.2 MgSO₄, 1.0 CaCl₂, and 10.0 N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid, with pH 7.4 at 25°C) containing 10% NCS and 2µM fura-2/AM, a fluorescent Ca²⁺ indicator. The cells were washed three times with modified Tyrode's solution to remove the fura-2/AM and serum from the extracellular fluid, and then left unincubated for 20 min before measurements were started. All experiments were performed at 25°C. The absorption shift of fura-2 that occurred upon binding was determined by scanning the excitation spectrum between 340 and 380 nm while monitoring the emission at 510 nm. The resultant fluorescent image was analyzed every 30 s from individual cells with an [Ca²⁺]_i analyzer (Argus 50, Hamamatsu Photonics, Hamamatsu, Japan), using an ultra high-sensitivity television camera (CCD). The fluorescence ratio (F340/ F380) was obtained by dividing, after background subtraction, the 340 by 380 nm images on a pixel-by-pixel basis. Intracellular calibration was performed by the method of Li et al. (16). To obtain the maximum R_max or minimum R_min value of the fluorescence ratio after fura-2 loading, endothelial cells were exposed to modified Tyrode's solution containing 10 µM ionomycin and 3 mM of Ca²⁺ or 5 mM ethylene glycol-bis(ß-aminoethyl ether)-N,N,N',N'-tetraacetic acid, respectively. [Ca²⁺]_i was calculated according to the equation of Grynkiewicz et al. (17). BK, thapsigargin, ML-9, ML-7, ML-5, bisindolylmaleimide I, protein kinase A (PKA) inhibitor peptide, and HA1077 had no effect on fura-2 fluorescence itself or on the autofluorescence of unloaded cells when examined at the concentrations used in this study. Ca²⁺ influx was calculated by subtracting the area of the Ca²⁺ response curve in the absence of extracellular Ca²⁺ from that in the presence of extracellular Ca²⁺ after stimulation of BK, and expressed as the percent of BK-stimulated Ca²⁺ influx. For measurement of [Ca²⁺]_i in fluid flow-stimulated cells, fluorescent digital images of indo-1 were obtained with a confocal laser scanning system (MRC-1000UV; Bio-Rad, Hemel Hempstead, U.K.) equipped with an ultraviolet argon ion laser (Enterprise Ion Laser; Coherent, Inc., Santa Clara, Calif.). Light with a 351 nm wavelength passed from the laser into the scanning unit and excited the cells through a 40 x objective (Fluor 40 water immersion, N.A. 1.15, Nikon). The fluorescence emitted by the cells formed 2-dimensional dual images through the objective, a beam splitter, two bandpath filters (405 nm and 480 nm), and photomultipliers. Indo-1 underwent a large shift in peak emission from 480 (F480) to 405 (F405) nm upon binding Ca²⁺, so that semiquantitative Ca²⁺ imaging could be performed by expressing recorded emission fluorescence as a ratio (F405/F480) to eliminate potential artifacts produced by variations in cell thickness, dye distribution in the cells, or photobleaching. Multiple regions were traced to monitor the time course of F405/F480 by accessory time course software of the Bio-Rad 1000UV system. ML-9 had no effect on indo-1 fluorescence itself or on autofluorescence of unloaded cells when examined at concentrations used in this study.

Flow-loading apparatus
To produce a well-defined flow, a parallel plate flow chamber was used for shear stress experiments (6). One side of the chamber was a coverslip (2.6x4.5x0.02 cm) on which cells were cultured; the other was machined from a polymethacrylate plate. These two flat surfaces were held approximately 200 µm apart by a silicone rubber gasket. The flow perfusate containing 500 nM ATP was supplied at a flow rate of 3 ml/min through the flow chamber from upstream via a silicone tube connected with a reservoir and drained downstream by a roller/tube pump (ATTO, Tokyo, Japan) with a depulsator. The intensity of wall shear stress (, dyn/cm²) on the endothelial cell layer was calculated by the formula = 6 µQ/a²b, where µ is the viscosity of the perfusate (0.0094 P at 37°C), Q is flow volume (ml/s), and a (0.02 cm) and b (1.4 cm) are cross-sectional dimensions of the flow path.

Determination of myosin light-chain (MLC) phosphorylation
MLC phosphorylation in endothelial cells was measured by separation of un-, mono- and diphosphorylated forms by glycerol-polyacrylamide gel electrophoresis (PAGE) according to their respective negative charges of phosphate molecules, followed by electrophoretic transfer of the proteins to a nitrocellulose membrane. The relative amounts of each form were quantified by an immunoblotting technique that used an anti-MLC antibody, as previously described (18). This method has been validated in a previous study (18) with [³²P]ATP incorporation into separated MLC bands and phosphomyosin-specific phosphatase. The technique has demonstrated that protein bands separated on gel electrophoresis represents un-, mono-, and diphosphorylated MLC. Briefly, after agonist and fluid flow stimulation in the presence or absence of ML-9, cells were exposed to 5% trichloroacetic acid containing 2 mM dithiothreitol (DTT). After centrifugation at 2500 x g for 3 min, the pellet was washed with 10 mM DTT/acetone, resuspended in urea sample buffer, and processed for urea/glycerol/PAGE and immunoblotting by a modification of the procedure described by Persechini et al. (19). The urea extracts of confluent endothelial cells before and after agonist and fluid flow stimulation contained un-, mono-, and diphosphorylated MLC. Relative quantification of un-, mono-, and diphosphorylated MLC was made from densitometric scans of immunostained nitrocellulose blots.

Preparation of proteins and enzyme assay
To qualify the activities of agents to inhibit MLCK, MLCK and MLC from chicken gizzard and calmodulin from bovine brain were purified, as described previously, according to the method of Adelstein and Klee (20). MLCK activity was assayed with MLC as described previously (21). MLC (200 µg) was phosphorylated in a reaction mixture containing 50 mM Tris-HCl (pH 7.0), 10 mM magnesium acetate, 0.2 mM CaCl₂, 100 ng calmodulin, 0.3 µg smooth muscle MLCK, and 30 µM [-³²P]ATP with a specific activity of 50–250 cpm/pmol in a final volume of 0.2 ml. Assays were performed for 5 min at 30°C in duplicate and resulted in a linear incorporation of [³²P]phosphate over the 5-min assay period. The reaction was terminated by the addition of 1 ml of ice-cold 20% trichloroacetic acid after the addition of 500 µg of bovine serum albumin as a carrier protein. The sample was centrifuged at 3000 rpm for 15 min, the pellet resuspended in ice-cold 10% trichloroacetic acid solutions, and the centrifugation-resuspension cycle was repeated three times. The final pellet was dissolved in 3 ml of 1N NaOH and the radioactivity was measured by a liquid scintillation counter.

Materials
Medium 199 was purchased from Boehringer-Mannheim (Mannheim, Germany), newborn calf serum and penicillin-streptomycin from GIBCO (New York, N.Y.), and fura-2/AM and indo-1 were from Dojindo (Kumamoto, Japan) and Wako (Osaka, Japan). Bradykinin and thapsigargin were from Sigma (St. Louis, Mo.); ML-9, bisindolylmaleimide I, PKA inhibitor peptide, and HA1077 were from Calbiochem (La Jolla, Calif.). [-³²P]ATP was purchased from Amersham Japan (Tokyo). All other chemicals were of the best available quality, mostly at analytical grades.

Statistical analysis
Data are expressed as mean ±SD. Statistical evaluation was performed using the Student's t test for unpaired data, one-way analysis of variance (ANOVA) followed by a Bonferroni t test, or ANOVA for repeated measures where appropriate. Values of P < 0.05 were considered statistically significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Effects of protein kinase inhibitors on Ca²⁺ response in agonist-stimulated cells
The effect of various protein kinase inhibitors on Ca²⁺ signaling was investigated in porcine aortic endothelial cells loaded with Ca²⁺-sensitive dye fura-2/AM. Treatment of endothelial cells with BK (10 nM) resulted in a rapid increase in the fura-2 ratio from basal levels of 1.25 ± 0.13 (mean±SD, n=14) to a maximum of 4.16 ± 0.46 for 90 s, followed by a sustained increase (3.53±0.47, 5 min after the addition of BK) in the presence of 1 mM extracellular Ca²⁺ ( Fig. 1A). Treatment with ML-9 (100 µM), a selective MLCK inhibitor that acts by competing with ATP for binding to the kinase (21), attenuated the peak and completely abolished the plateau increase in BK-stimulated cells (peak: 1.99±0.31; plateau: 1.22±0.10, n=14). HA1077 (100 µM), another inhibitor of MLCK (22), also inhibited BK-stimulated Ca²⁺ response (peak: 2.49±0.24; plateau: 1.65±0.21, n=7). On the other hand, neither bisindolylmaleimide I (10 µM), a protein kinase C (PKC) inhibitor (23), nor PKA inhibitor peptide (0.3 µM) (24) affected the BK-stimulated Ca²⁺ response (bisindolylmaleimide I, peak: 4.67±0.60, plateau: 3.60±0.44; PKA inhibitor peptide, peak: 4.40±0.47, plateau: 3.58±0.40; n=7).

View larger version (20K): [in this window] [in a new window]   Figure 1. Effects of various protein kinase inhibitors on bradykinin (BK) (A) and thapsigargin (B) -induced Ca2+ response (expressed as fluorescence ratio F340/F380). Cells were loaded with fura-2/AM and then pretreated with various protein kinase inhibitors for 10 min, except for wortmannin, which required 30 min. Extracellular Ca2+ concentration was 1 mM. Closed bars represent peak increases in Ca2+ response; open bars represent plateau increases 5 and 10 min after application of BK (A) and thapsigargin (B), respectively. Concentrations used: BK, 10 nM; ML-9, 100 µM; BIM (bisindolylmaleimide I), 10 µM; PKAI (protein kinase A inhibitor), 0.3 µM; HA1077, 100 µM; TG (thapsigargin), 1 µM; Wort (wortmannin), 100 µM. Values are means ±SD; n = 14 cells from three separate experiments. (P< 0.01).

Pretreatment with ML-9 also inhibited the rise in [Ca²⁺]_i caused by thapsigargin, which is an inhibitor of endoplasmic reticulum Ca²⁺-ATPase (25) and has been shown to stimulate Ca²⁺ influx in endothelial cells without activation of the IP₃ pathway. Thapsigargin (1 µM) induced a gradual increase in [Ca²⁺]_i, attaining the peak in 3 min, followed by a sustained increase in the presence of 1 mM extracellular Ca²⁺ (peak: 5.59±0.47; plateau: 5.31±0.46, n=14). ML-9 inhibited the peak and almost completely abolished the plateau increase in [Ca²⁺]_i caused by thapsigargin (peak: 1.92±0.19; plateau: 1.35±0.13, n=14) ( Fig. 1B). Pretreatment with HA1077 also inhibited the thapsigargin-stimulated Ca²⁺ response (peak: 2.55±0.35; plateau: 1.80±0.15, n=7). Wortmannin is a fungal metabolite that has been shown to be a potent selective inhibitor of phosphatidylinositol 3 (PI₃)-kinase at low concentrations and of MLCK at high concentrations (26). Pretreatment of the cells for 30 min with 100 µM wortmannin significantly attenuated the thapsigargin-stimulated Ca²⁺ response (peak: 2.03±0.28; plateau: 1.18 ± 0.07, n=7), an effect qualitatively similar to that of ML-9. This effect should be due to the effect of wortmannin to inhibit MLCK, because thapsigargin activates Ca²⁺ influx without activating PI₃ kinase. On the other hand, neither the PKC inhibitor bisindolylmaleimide I (10 µM) nor PKA inhibitor peptide (0.3 µM) showed any effect on the thapsigargin-stimulated Ca²⁺ response (bisindolylmaleimide I, peak: 5.78±0.57, plateau: 5.66±0.52; PKA inhibitor peptide, peak: 5.65±0.52, plateau: 5.56±0.48; n=7).

In the absence of extracellular Ca²⁺, BK caused only a small and transient increase in [Ca²⁺]_i (basal levels of 0.83±0.03 to a maximum of 2.04±0.37 for 30 s, n=14), which appeared to be due to the release of Ca²⁺ from intracellular stores and was not affected by ML-9 (1.91±0.39 for 30 s after the addition of BK). The thapsigargin-induced rise had a smaller peak than that observed after treatment with bradykinin in the absence of extracellular Ca²⁺, and was not affected by ML-9, either (data not shown). These findings indicate that ML-9 inhibits Ca²⁺ entry from the extracellular space, but does not affect the mobilization of Ca²⁺ from intracellular stores in agonist-stimulated cells.

Effect of ML-9 on Ca²⁺ response in fluid flow-stimulated cells
To examine the effect of ML-9 on fluid flow-stimulated Ca²⁺ response, aortic endothelial cells were loaded with indo-1, and the fluorescence ratio (F405/F480) was continuously monitored as an index of relative [Ca²⁺]_i. Typical examples of the recordings from the experiments are shown in Fig. 2. Fluid flow stimulation was associated with a biphasic elevation in [Ca²⁺]_i characterized by a rapid initial peak, followed by a steady or oscillating plateau phase in the presence of ATP (500 nM). The initial transient component should reflect, at least in part, the release of Ca²⁺ from intracellular stores, and the second phase should reflect the more prolonged transmembranous Ca²⁺ influx (6). Seventy to eighty percent of the cells revealed the Ca²⁺ response, which consisted of an initial transient peak, followed by a stable plateau phase that was maintained as long as the cells were exposed to fluid flow ( Fig. 2A). Twenty to thirty percent of the cells showed an oscillatory increase in [Ca²⁺]_i ( Fig. 2B). Treatment with ML-9 (100 µM) decreased the peak of the initial transients and almost completely inhibited both the oscillation and the plateau phases of Ca²⁺ response in the fluid flow-stimulated cells (basal level: 0.64±0.11; levels at 2 min of fluid flow stimulation: 1.12±0.25 without ML-9 and 0.68±0.19 with ML-9, n=10). The effect of ML-9 was immediately reversed upon removal of the inhibitor from the luminal perfusate (data not shown), which is consistent with our previous findings in agonist-stimulated cells.

View larger version (15K): [in this window] [in a new window]   Figure 2. Representative traces showing the effect of ML-9 (100 µM) on fluid flow-stimulated Ca2+ response in endothelial cells in the presence of 1 mM extracellular Ca2+ and 500 nM ATP. Indo-1-loaded cells were exposed to fluid flow (shear stress=5 dynes/cm2) with and without ML-9. Fluid flow stimulation caused peak and plateau (A) or oscillatory (B) Ca2+ increases. Treatment with ML-9 blocked both the oscillatory and plateau phases of the Ca2+ response to fluid flow stimulation.

Effects of ML-9 on agonist- and fluid flow-stimulated MLC phosphorylation
The role of MLCK is to phosphorylate MLC from nonphosphorylated to monophosphorylated and diphosphorylated forms. Immunoblotting of protein from confluent primary cultures of porcine aortic endothelial cells using a polyclonal anti-MLC antibody revealed a relatively high basal level of monophosphorylated MLC compared to that of diphosphorylated MLC (MLC-P: 29.8±5.8%; MLC-PP: 8.1±4.1% in control cells, from three separate experiments) ( Fig. 3). A significant increase in MLC diphosphorylation was detected after the addition of either thapsigargin (1 µM) or BK (10 nM) (MLC-PP: 64.5±18.5% with thapsigargin, 42.7±11.7% with BK). Pretreatment with ML-9 (100 µM) almost completely inhibited the formation of diphosphorylated MLC in thapsigargin- and bradykinin-stimulated cells. The effect of ML-9 to inhibit the formation of diphosphorylated MLC after the stimulation by BK was dose dependent. It is interesting that the content of monophosphorylated MLC was not influenced under these conditions. Figure 4 shows the relation between the diphosphorylation of MLC and Ca²⁺ influx in BK-stimulated cells. The effect of ML-9 to inhibit the formation of diphosphorylated MLC was linearly correlated with its ability to prevent Ca²⁺ entry in BK-stimulated cells.

View larger version (70K): [in this window] [in a new window]   Figure 3. Immunostaining of myosin light-chain (MLC) in cultured porcine aortic endothelial cells. The cells were exposed to an agonist (either 1 µM thapsigargin or 10 nM BK) or the agonist with ML-9 for 2 min. Agonist stimulation was terminated by a reaction with 6.5% trichloroacetic acid. ML-9-treated cells were incubated for 10 min with various concentrations (1–100 µM) of ML-9 before the addition of either BK or thapsigargin. Upper, middle, and lower bands represent unphosphorylated (UP), monophosphorylated (P), and diphosphorylated (PP) MLC, respectively. The degree of phosphorylation is expressed as percent of the total extracted MLC.

View larger version (18K): [in this window] [in a new window]   Figure 4. Relationship between phosphorylation of MLC and inhibition of Ca2+ influx in cells stimulated with 10 nM BK. Open and closed circles represent percent of mono- and diphosphorylated MLC (corresponding concentrations of ML-9 in brackets), respectively, in total MLC. Diphosphorylation of MLC was inversely correlated with the inhibition of Ca2+ influx (R=0.99625).

To examine the effect of shear stress on the phosphorylation of MLC, endothelial cells were exposed to fluid flow conditions for 2 min. Fluid flow induced MLC phosphorylation, as observed in agonist-stimulated cells ( Fig. 5), and increased the formation of monophosphorylated and diphosphorylated MLC. ML-9 also completely abolished this process.

View larger version (62K): [in this window] [in a new window]   Figure 5. Immunostaining of MLC in fluid flow-stimulated endothelial cells. The cells were exposed to fluid flow for 2 min with or without 10 min pretreatment with 100 µM ML-9. Fluid flow stimulation was terminated by the reaction with 6.5% trichloroacetic acid. Upper, middle, and lower doublet bands represent unphosphorylated (UP), monophosphorylated (P), and diphosphorylated (PP) MLC, respectively. Fluid flow stimulation (FF) enhanced MLC phosphorylation and ML-9 inhibited it.

Effects of ML-5, ML-7, and ML-9 on BK-stimulated Ca²⁺-response
Although a selective inhibitor of MLCK, ML-9 at high concentrations can also inhibit PKC and PKA (21). To further evaluate the likely involvement of MLCK in the regulation of Ca²⁺ entry, the effects of ML-5, ML-7, and ML-9, synthetic analogs of naphthalene sulfonyl compounds ( Fig. 6) that have different selectivities for MLCK (21), were contrasted in BK-stimulated cells ( Fig. 7). The IC₅₀ values of ML-9, ML-7, and ML-5 for inhibition of MLCK were 6.8, 0.7, and 80µM, respectively. At the concentrations used in the present study, ML-7 appeared to inhibit BK-stimulated Ca²⁺ influx more potently than did ML-9 and ML-5. The effects of these agents to inhibit BK-stimulated Ca²⁺ influx seem to be well correlated with their potencies to inhibit MLCK. Such a linear correlation, however, is not seen between their ability to inhibit BK-stimulated Ca²⁺ influx and their potencies to inhibit PKC and PKA (data not shown). Although ML-5 has been shown to be the most potent inhibitor of PKC and PKA, its activity to inhibit Ca²⁺ influx is less potent than that of ML-7 and ML-9. All these results indicate that the effect of ML-9, ML-7 and ML-5 to inhibit BK-stimulated Ca²⁺ entry in endothelial cells is due to their inhibitory effect on MLCK.

View larger version (16K): [in this window] [in a new window]   Figure 6. Chemical structures of ML-5, ML-7, and ML-9.

View larger version (22K): [in this window] [in a new window]   Figure 7. Dose-dependent inhibition of BK-stimulated Ca2+ influx in endothelial cells by ML-5, ML-7, and ML-9. Values are means ±SD; Each value consists of n = 10 cells from two separate experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this study, involvement of MLCK in the regulation of Ca²⁺ signaling was investigated in agonist- and fluid flow-stimulated endothelial cells. Agonist and fluid flow stimulation caused an increase in [Ca²⁺]_i and resulted in a shift of total MLC from the unphosphorylated to the diphosphorylated form. ML-9, an inhibitor of MLCK, abolished the influx of Ca²⁺ from the extracellular space and prevented formation of diphosphorylated MLC not only in agonist-, but also in fluid flow-stimulated endothelial cells. The effect of ML-9 to inhibit Ca²⁺ influx was dose dependent and correlated well with the inhibition of the formation of diphosphorylated MLC. ML-5 and ML-7, synthetic analogs of ML-9, inhibited Ca²⁺ influx in accordance with their activities to inhibit MLCK.

ML-9 is a kinase inhibitor that acts by competing with ATP for binding to the kinase (21). MLCK is 1–1.5 logs more sensitive to ML-9 than are PKC and PKA, and high doses of ML-9 can also inhibit these two kinases. In the present study, we also demonstrated that 1) inhibition of neither PKC nor PKA affected the agonist-stimulated Ca²⁺ response; 2) HA1077 and wortmannin, other inhibitors of MLCK, also blocked the agonist-stimulated Ca²⁺ response. At the concentrations used, bisindolylmaleimide I has been shown to inhibit both PKA and PKC more selectively than does ML-9, and PKA inhibitor peptide to be one or two logs more selective to PKA than is ML-9 (23, 24). Since bisindolylmaleimide I and PKA inhibitor peptide both failed to attenuate the BK- and thapsigargin-stimulated Ca²⁺ response, it is unlikely that inhibition of either PKC or PKA by ML-9 accounts for the ability of ML-9 to eliminate Ca²⁺ influx in agonist-treated endothelial cells. Although a nonspecific effect of ML-9 on the Ca²⁺ influx pathway cannot be ruled out, the fact that the MLCK inhibitors HA1077 and wortmannin also inhibited Ca²⁺ influx strongly suggests that the effects of ML-9, HA1077, and wortmannin under the conditions of our experiments are specific for MLCK. In addition, ML-5 and ML-7, synthetic analogs of ML-9, inhibited Ca²⁺ influx in accordance with their ability to inhibit MLCK. All in all, these observations imply that MLCK is involved in the regulation of Ca²⁺ influx in endothelial cells.

Recent studies have shown that wall shear stress generated by fluid flow modulates endothelial morphology and function (14). Although shear stress exerts these effects, at least in part, by increasing [Ca²⁺]_i, the mechanisms underlying the shear stress-stimulated Ca²⁺ rise in endothelial cells remain obscure. Consistent with the observations showing that fluid shear stress induces receptor activation-like events in endothelial cells by stimulating PLC with resultant increased IP₃ formation (12, 13), the present study demonstrates that MLCK inhibitors block fluid flow- as well as agonist-stimulated Ca²⁺ response, suggesting that the signal transduction cascades initiated by agonist and fluid flow stimulation of endothelial cells share a common role. Although some differences might be seen in the formation of mono- and diphosphorylated MLC when BK and thapsigargin were compared with fluid flow stimulation, these are probably due to differences in the various degrees of Ca²⁺ influx in response to these stimuli. The Ca²⁺ response stimulated by fluid flow has been reported in previous studies by us and others to be smaller than that by BK and thapsigargin. Thus, the fluid flow-stimulated Ca²⁺ response was reported to be at the 100 nanomolar level (6), whereas the BK- and thapsigargin-stimulated Ca²⁺ response reached the micromolar level (11).

Our results showed that ML-9 prevented the formation of diphosphorylated MLC in response to agonist and fluid flow stimulation. Furthermore, ML-9 inhibited the diphosphorylation of MLC in proportion to its inhibition of Ca²⁺ influx. Formation of monophosphorylated MLC, however, was not influenced under these conditions. Although it is not yet clarified which is more important in regulating [Ca²⁺]_i—the formation of diphosphorylated MLC or the amount of total phosphorylated (mono- plus diphosphorylated) MLC—it seems quite reasonable from these observations that MLCK plays a crucial role. However, it has been open to question as to whether MLCK directly regulates Ca²⁺ influx or indirectly influences the influx through the phosphorylation of MLC. MLCK has been shown to phosphorylate MLC, mediate the organization of cytoskeletal structure, and regulate a variety of contractile events in both smooth muscle and nonmuscle cells, including endothelial cells (27, 28). It is possible, therefore, that MLCK inhibitors prevent the reorganization in cytoskeleton surrounding the plasmalemmal Ca²⁺ influx pathways and secondarily modulate Ca²⁺ influx in agonist-stimulated cells. Alternatively, it is also possible that MLCK regulates Ca²⁺ influx independently of the phosphorylation of MLC, having another target to phosphorylate than MLC, linking directly to the plasmalemmal Ca²⁺ influx pathways, or eliciting production of a second messenger that signals opening of specific Ca²⁺ channels when intracellular stores are depleted. Consequently, the inhibitors of MLCK might prevent the phosphorylation of MLC in part by inhibiting Ca²⁺ influx in agonist-stimulated cells. The results showing the inhibition of MLC phosphorylation by ML-9 may thus only indicate the activity of ML-9 to inhibit MLCK. In other words, the level of phosphorylated MLC may only be a marker of MLCK activity in endothelial cells.

In endothelial cells, as in many other cell types, it is generally accepted that the filling of intracellular Ca²⁺ stores controls the Ca²⁺ influx pathway (4, 5, 9, 10). Activation of the Ca²⁺ influx pathway appears to involve a soluble messenger that is generated at the endoplasmic reticulum and is capable of modulating plasma membrane Ca²⁺ permeability (29). This putative mediator appears to require phosphorylation in order to maintain its activity (30). Recently, it has been shown that inhibition of tyrosine phosphorylation by different tyrosine kinase inhibitors such as herbimycin A, genistein, and piceatannol attenuate agonist-stimulated Ca²⁺ entry into endothelial cells (31). These findings suggest that activation of tyrosine kinases is involved in regulating Ca²⁺ influx in endothelial cells. However, since the inhibitory effect of tyrosine kinase inhibitors on Ca²⁺ influx, even at high concentrations, was only partial, another mechanism (or mechanisms) could also contribute to the regulation of Ca²⁺ influx after agonist stimulation. We have recently reported that inhibition of MLCK almost completely abolishes Ca²⁺ influx in BK-stimulated endothelial cells, including the tyrosine kinase inhibitor-insensitive component (32). This study now demonstrates that inhibition of MLCK is also effective in blocking Ca²⁺ influx in physiologically stimulated cells. These observations strongly suggest that MLCK plays a more critical role than does tyrosine kinase in the regulation of Ca²⁺ influx in endothelial cells. Endothelial cells have been reported to contain a modest level of MLCK (33), and the involvement of MLCK in the intracellular signal transduction cascade has been thought to be restricted mainly to events coupled to morphological changes and cell contraction (27, 28). This concept would now appear to be an underestimation of the role of MLCK in endothelial cell Ca²⁺ signaling. Rather, MLCK is a likely candidate for the main factor regulating Ca²⁺ influx in endothelial cells.

It is possible that the activation of MLCK after agonist and fluid flow stimulation induces Ca²⁺ influx into endothelial cells and stimulates the production of NO, which results in vascular relaxation (1, 2, 14). In contrast to its effects in endothelial cells, MLCK activation in smooth muscle cells has been shown to cause vascular contraction (27). It is well accepted that endothelial cells maintain vascular homeostasis by producing vascular relaxing as well as constricting factors (1, 2). MLCK probably exerts these contradictory effects at different sites of action (endothelial cells and vascular smooth muscle cells), which helps maintain appropriate vascular tone. It now seems clear that the vascular system at least has a self-regulating mechanism, MLCK, that exerts counterbalancing effects on endothelial cells and vascular smooth muscle cells. However, further investigation is required to clarify the physiological role (or roles) of MLCK in endothelial cells.

In conclusion, we have demonstrated that agonist- and fluid flow-stimulated Ca²⁺ signaling cascades share a common role in endothelial cells. In this signal transduction cascade, MLCK plays an essential role to regulate the influx of Ca²⁺ across the plasma membrane. This study is the first to report the effect of MLCK inhibition on Ca²⁺ signaling in fluid flow-stimulated endothelial cells and to demonstrate the close relationship between Ca²⁺ influx and MLC phosphorylation. Further studies are now under way to elucidate the mechanisms responsible for the effects of MLCK inhibitors in the signaling cascade.

	ACKNOWLEDGMENTS

 
The authors are grateful to Prof. H. Nakaya (Chiba University) for many helpful insights and to Dr. T. Q. Kim for proofreading. This study was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture and Science of Japan, and Japan Foundation of Cardiovascular Research to H.W.

	FOOTNOTES

 
¹ Correspondence: Internal Medicine III, Hamamatsu University School of Medicine, 3600 Handa-cho, Hamamatsu 431–31, Japan. E-mail: hwat{at}hama-med.ac.jp

² Abbreviations: MLCK, myosin light-chain kinase; MLC, myosin light-chain; BK, bradykinin; PLC, phospholipase C; IP₃, inositol 1, 4, 5-triphosphate; NO, nitric oxide; PKC, protein kinase C; PKA, protein kinase A; NCS, newborn calf serum; DTT, dithiothreitol.

Received for publication September 1, 1997. Accepted for publication November 20, 1997.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Furchgott, R. F., and Vanhoutte, P. M. (1989) Endothelium-derived relaxing and contracting factors. FASEB J. 3, 2007–2018[Abstract]
2. Ignarro, L. J. (1993) Nitric oxide-mediated vasorelaxation. Thromb. Haemost. 70, 148–151[Medline]
3. Curry, F. E. (1992) Modulation of venular microvessel permeability by calcium influx into endothelial cells. FASEB J. 6, 2456–2466[Abstract]
4. Byron, K. L., Babnigg, G., and Villereal, M. L. (1992) Bradykinin-induced Ca²⁺ entry, release, and refilling of intracellular Ca²⁺ stores. J. Biol. Chem.. 267, 108–118[Abstract/Free Full Text]
5. Schilling, W. P., Cabello, O. A., and Rajan, L. (1992) Depletion of the inositol 1,4,5-triphosphate-sensitive intracellular Ca²⁺ store in vascular endothelial cells activates the agonist-sensitive Ca²⁺ -influx pathway. Biochem. J. 284, 521–530
6. Ando, J., Komatsuda, T., and Kamiya, A. (1988) Cytoplasmic calcium response to fluid shear stress in cultured vascular endothelial cells. In Vitro Cell Dev. Biol. 24, 871–877[Medline]
7. Shen, J., Luscinskas, F. W., Connolly, A., Dewey, C. F., Jr., and Gimbrone, M. A,. Jr. (1992) Fluid shear stress modulates cytosolic free calcium in vascular endothelial cells. Am. J. Physiol. 262, C384–390[Abstract/Free Full Text]
8. Lambert, T. L., Kent, R. S., and Whorton, A. R. (1986) Bradykinin stimulation of inositol polyphosphate production in porcine aortic endothelial cells. J. Biol. Chem. 261, 15288–15293[Abstract/Free Full Text]
9. Adams, D. J., Barakeh, J., Laskey, R., and van Breemen, C. (1989) Ion channels and regulation of intracellular calcium in vascular endothelial cells. FASEB J. 3, 2389–2400[Abstract]
10. Putney, J. W., and Bird, G. J. (1993) The inositol phosphate-calcium signaling system in nonexcitable cells. Endocrin. Rev. 14, 610–631[Medline]
11. Watanabe, H., Takahashi, R., Zhang, X. X., Kakizawa, H., Hayashi, H., and Ohno, R. (1996) Inhibition of agonist-induced entry in endothelial cells by myosin light-chain kinase inhibitor. Biochem. Biophys. Res. Commun. 225, 777–784[Medline]
12. Nollert, M. U., Eskin, S. G., and McIntyre, L. V. (1990) Shear stress increases inositol triphosphate levels in human endothelial cells. Biochem. Biophys. Res. Commun. 170, 281–287[Medline]
13. Prasad, A. R. S., Logan, S. A., Nerem, R. M., Schwartz, C. J., and Sprague, E. A. (1993) Flow-related responses of intracellular inositol phosphate levels in cultured aortic endothelial cells. Circ. Res. 72, 827–836[Abstract/Free Full Text]
14. Noris, M., Morigi, M., Donadelli, R., Aiello, S., Foppolo, M., Todeschini, M. S., Remuzzi, G., and Remuzzi, A. (1995) Nitric oxide synthesis by cultured endothelial cells is modulated by flow conditions. Circ. Res. 76, 536–543[Abstract/Free Full Text]
15. Watanabe, H., Kuhne, W., Spahr, R., Schwartz, P., and Piper, H. M. (1991) Macromolecule permeability of coronary and aortic endothelial monolayers under energy depletion. Am. J. Physiol. 260, H1344–1352[Abstract/Free Full Text]
16. Li, Q., Altschuld, R. A., and Stokes, B. T. (1987) Quantitation of intracellular free calcium in single adult cardiomyocytes by fura-2 fluorescence microscopy: calibration of fura-2 ratios. Biochem. Biophys. Res. Commun. 147, 120–126[Medline]
17. Grynkiewicz, G., Poenie, M., and Tsien, R. Y. (1985) A new generation of Ca²⁺ indicators with greatly improved fluorescence properties. J. Biol. Chem. 260, 3440–3450[Abstract/Free Full Text]
18. Seto, M., Sakurada, K., Kamm, K. E., Stull, J.T., and Sasaki, Y. (1996) Myosin light chain diphosphorylation is enhanced by growth promotion of cultured smooth muscle cells. Pfluegers Arch. 432, 7–13[Medline]
19. Persechini, A., Kamm, K. E., and Stull, J. T. (1986) Different phosphorylated forms of myosin in contracting tracheal smooth muscle. J. Biol. Chem. 261, 6293–6299[Abstract/Free Full Text]
20. Adelstein, R. S., and Klee, C. B. (1981) Purification and characterization of smooth muscle myosin light chain kinase. J. Biol. Chem. 256, 7501–7509[Abstract/Free Full Text]
21. Saitoh, M., Ishikawa, T., Matsushima, S., Naka, M., and Hidaka, H. (1987) Selective inhibition of catalytic activity of smooth muscle myosin light chain kinase. J. Biol. Chem. 262, 7796–7801[Abstract/Free Full Text]
22. Seto, M., Sasaki, Y., Hidaka, H., and Sasaki, Y. (1991) Effects of HA1077, a protein kinase inhibitor, on myosin phosphorylation and tension in smooth muscle. Eur. J. Pharmacol. 195, 267–272[Medline]
23. Toullec, D., Pianetti, P., Coste, H., Bellevergue, P., Grand-Perret, T., Ajakane, M., Baudet, V., Boissin, P., Boursier, E., Loriolle, F., Duhamel, L., Charon, D., and Kirilovsky, J. (1991) The bisindolylmaleimide GF 109203X is a potent and selective inhibitor of protein kinase C. J. Biol. Chem. 266, 15771–15781[Abstract/Free Full Text]
24. Fletcher, W. H., Van Patten, S. M., Cheng, H. C., and Walsh, D. A. (1986) Cytochemical identification of the regulatory subunit of cAMP-dependent protein kinase by use of fluorescently labeled catalytic subunit. Examination of protein kinase dissociation in hepatoma cells responding to 8-Br-cAMP stimulation. J. Biol. Chem. 261, 5504–5513[Abstract/Free Full Text]
25. Thastrup, O., Cullen, P. J., Drobak, B. K., Hanley, M. R., and Dawson, A. P. (1990) Thapsigargin, a tumor promoter, discharges intracellular Ca²⁺ stores by specific inhibition of the endoplasmic reticulum Ca²⁺-ATPase. Proc. Natl. Acad. Sci. USA 87, 2466–2470[Abstract/Free Full Text]
26. Nakanishi, S., Kakita, S., Takahashi, I., Kawahara, K., Tsukuda, E., Sano, T., Yamada, K., Yoshida, M., Kase, H., Matsuda, Y., Hashimoto, Y., and Nonomura, Y. (1992) Wortmannin, a microbial product inhibitor of myosin light chain kinase. J. Biol. Chem. 267, 2157–2163[Abstract/Free Full Text]
27. Kamm, K. E., and Stull, J. T. (1985) The function of myosin and myosin light chain kinase phosphorylation smooth muscle. Annu. Rev. Pharmacol. Toxicol. 25, 593–620[Medline]
28. Wysolmerski, R. B., and Langunoff, D. (1991) Regulation of permeabilized endothelial cell retraction by myosin phosphorylation. Am. J. Physiol. 261, C32–C40[Abstract/Free Full Text]
29. Randriamampita, C., and Tsien, R. Y. (1993) Emptying of intracellular Ca²⁺ stores releases a novel small messenger that stimulates Ca²⁺ influx. Nature (London) 364, 809–814[Medline]
30. Parekh, A. B., Terlau, H., and Stühmer, W. (1993) Depletion of InsP₃ stores activates a Ca²⁺ and K⁺ current by means of a phosphatase and a diffusible messenger. Nature (London) 364, 814–818[Medline]
31. Fleming, I., Fisslthaler, B., and Busse, R. (1995) Calcium signaling in endothelial cells involves activation of tyrosine kinases and leads to activation of mitogen-activated protein kinases. Circ. Res. 76, 522–529[Abstract/Free Full Text]
32. Takahashi, R., Watanabe, H., Zhang, X. X., Kakizawa, H., Hayashi, H., and Ohno, R. (1997) Roles of inhibitors of myosin light chain kinase and tyrosine kinase on cation influx in agonist-stimulated endothelial cells. Biochem. Biophys. Res. Commun. 235, 657–662[Medline]
33. Garcia, J. G. N., Lazar, V., Gilbert-McClain, L. I., Gallagher, P. J., and Verin, A. D. (1997) Myosin light chain kinase in endothelium: Molecular cloning and regulation. Am. J. Respir. Cell Mol. Biol. 16, 489–494[Abstract]