Rho kinase regulation of vasopressin-induced calcium entry in vascular smooth muscle cell: Comparison between rat isolated aorta and cultured aortic cells

Anneloes Martinsen, Nicolas Baeyens, Xavier Yerna, Nicole Morel∗

Abstract

In addition to its role in artery contraction, Rho kinase (ROCK) is reported to be involved in the Ca2+ response to vasoconstrictor agonist in rat aorta. However the signaling pathway mediated by ROCK had not been investigated so far and it was not known whether ROCK also contributed to Ca2+ signaling in cultured vascular smooth muscle cells (VSMC), which undergo profound phenotypic changes. Our results showed that in VSMC, ROCK inhibition by Y-27632 or H-1152 had no effect on the Ca2+ response to vasopressin, while in aorta the vasopressin-induced Ca2+ entry was significantly decreased. The inhibition of myosin light chain kinase (MLCK) by ML-7 depressed the vasopressin-induced Ca2+ signal in aorta but not in VSMC. The difference in ROCK sensitivity of vasopressin-induced Ca2+ entry between aorta and VSMC was not related to an alteration of the RhoA/ROCK pathway. However, MLCK expression and activity were depressed in cultured cells compared to aorta. We concluded that the regulation of vasopressininduced Ca2+ entry by ROCK in aorta could involve the myosin cytoskeleton and could be prevented by the downregulation of MLCK in VSMC. These results underline the important differences in Ca2+ regulation between whole tissue and cultured cells.

Keywords:
Rho kinase
Myosin light chain kinase
Calcium
Vascular smooth muscle cell
Aorta

1. Introduction

In smooth muscle, changes in cytosolic calcium concentration ([Ca2+] ) play a key role in regulating diverse cellular processes cyt including cell growth, cell differentiation and muscle contraction. The increase in [Ca2+] by agonist stimulation and thesubsequent formation of the Ca2+–calmodulin complex leads tothe activation of myosin light chain kinase (MLCK) which phosphorylates myosin light chain (LC20) resulting in smooth muscle contraction [1]. Dephosphorylation of LC20 by myosin light chain phosphatase (MLCP) evokes smooth muscle relaxation. The inhibition of MLCP by Rho kinase (ROCK), which is activated by RhoA after agonist stimulation, promotes the phosphorylation ofLC20 [2–4].
It has been demonstrated that ROCK is involved in noradrenaline-activated Ca2+ entry in rat aorta and mesenteric artery [5]. ROCK-sensitive Ca2+ entry has been shown to bedistinct from voltage- or store-operated Ca2+ channels, suggestinga role for non-selective cation channels [5]. However, the exact mechanism between ROCK activation and Ca2+ entry has not yet been elucidated.
ROCK is known to mediate a great number of processes that involve actin cytoskeleton by phosphorylating different protein targets, among which Lim kinase, which controls the actin polymerization through the phosphorylation of cofilin [6], and the ezrin–radixin–moesin (ERM) proteins, which serve as cross-linkers between actin filaments and membrane proteins at the cell surface [7]. It has been observed that actin depolymerization does affect neither the Ca2+ response to noradrenaline in aorta nor its sensitivity to ROCK inhibition [8], suggesting that the actin cytoskeleton can probably not be considered as the main target of ROCK to regulate receptor-activated Ca2+ entry [9].
On the other hand, by inhibiting myosin phosphatase, ROCK is involved in the regulation of myosin activity. It has been shown that, in the renal conducting duct, myosin IIA could be implicated in the trafficking of intracellular aquaporin-2 to the apical plasma membrane after vasopressin stimulation [10]. Furthermore, it has been suggested that MLCK modulates the activation process of TRPC5 in HEK cells [11,12] and in murine ileal myocytes [13]. On the basis of the effect of the naphthalene sulphonamide derivatives ML-7 and ML-9, the contribution of MLCK to the activation of nonselective cation current has been proposed in guinea-pig gastric myocytes [14] and in rabbit portal vein myocytes [15]. Transfection of endothelial cells [16] or monocytes/macrophages [17] with MLCK antisense completely prevents capacitative Ca2+ entry in response to thapsigargin. These observations reveal a potential role of myosin cytoskeleton in plasma membrane channels activation, and allow to postulate that myosin cytoskeleton could mediate the effect of ROCK on Ca2+ signaling.
Vascular smooth muscle cells (VSMC) are characterized by the ability to undergo major modifications in phenotype in response to local disturbances, increasing their rate of cell proliferation and migration, developing their synthetic properties and losing their contractile properties. This phenotypic switching from a contractile to a non-contractile proliferative phenotype plays a major role in a number of diseases like atherosclerosis, cancer and hypertension and is observed when VSMC are grown in culture [18]. The loss of contractility and the alteration in intracellular Ca2+ handling are characteristic features of this switching and are associated with changes in several ion channels expression, like members of TRPC family, and pumps, as sarcoplasmic reticulum Ca2+-ATPase 2b [19].
However,thecontributionofROCKtoreceptor-activatedCa2+ entry in cultured VSMC has not been described. The objective of the present study was to further investigate the role of ROCK in receptor-activated Ca2+ entry and to deter- mine whether ROCK contribution to Ca2+ signaling can be observed in cultured VSMC. To this aim, the role of ROCK in vasopressininduced Ca2+ response was compared in aorta and aortic cultured VSMC. Our results indicate that the regulation of vasopressinstimulated Ca2+ entry by ROCK in isolated aorta could involve myosin phosphorylation. On the opposite, in aortic cultured VSMC, vasopressin-stimulated Ca2+ entry was not affected by ROCK or MLCK inhibition, although ROCK activity was confirmed in cultured cells, supporting the existence of important differences in the Ca2+ signaling pathways between whole artery and cultured cells.

2. Materials and methods

2.1. Solutions and drugs

Composition of the buffered saline solution (BSS) was (in mM): NaCl 137; KCl 6; MgCl2 1.2; CaCl2 2; glucose 10; HEPES 10 at pH 7.4 with Tris. The physiological solution (PSS) was composed of (in mM): NaCl, 122; KCl, 5.9; NaHCO3, 15; MgCl2, 1.2; CaCl2, 1.25; glucose, 11. The KCl solution (PSS KCl) was composed of (in mM): NaCl, 27; KCl, 100; NaHCO3, 15; MgCl2, 1.2; CaCl2, 1.25; glucose, 11.Fura-2 AM, ionomycin, H-1152, ML-7, RHC-80267, andRo-318220 came from Calbiochem (Bierges, Belgium). 2Mercaptoethanol was from Merck (Overijse, Belgium). Pharmalytes were from GE Healthcare Europe GmBH (Diegem, Belgium). Dulbecco’s modified Eagle’s medium (DMEM, 41965), foetal bovine serum (FBS), l-glutamine, penicillin, streptomycin were from Invitrogen (Merelbeke, Belgium). Vasopressin, N-nitro-l-arginine,OPC-21268, EGTA, EDTA, cremophor-EL, verapamil, mouse monoclonal anti-actin -smooth muscle antibody (clone 1A4), IGEPAL CA-630, urea and glycerol were from Sigma–Aldrich (Bornem, Belgium). Y-27632 and OBAA were from Tocris Bioscience (Bristol, United Kingdom). Primary antibodies: rabbit polyclonal anti-phospho-ezrin (Thr567)/radixin (Thr564)/moesin (Thr558) (anti-phospho-ERM) was from Cell Signaling (Bioké, Leiden, The Netherlands); mouse anti-myosin light chain kinase (anti-MLCK) and rabbit anti-phospho myosin light chain (pSer19) (anti-phospho-LC20) were from Sigma–Aldrich (Bornem, Belgium); goat polyclonal anti-actin came from Santa Cruz Biotechnology, Inc (Heidelberg, Germany); mouse monoclonal anti-RhoA from Cytoskeleton (Tebu-bio, Boechout, Belgium). Secondary antibodies: goat anti-rabbit (Biotium CF770) and goat anti-mouse (Biotium CF770) were from VWR (Leuven, Belgium) and donkey anti-goat (Alexa Fluor 680) came from Invitrogen (Merelbeke, Belgium).

2.2. Tissue preparation

All experiments were in accordance with institutional guidelines for the use of experimental animals. Male Wistar rats (mean body weight of 250 g) were anesthetized and sacrificed by decapitation. Aorta was quickly removed, immersed in BSS, cleaned of all fat and connective tissue. The endothelium was carefully removed by gently rubbing the lumen of the artery with forceps tips.

2.3. Vascular smooth muscle cells culture

VSMC were obtained from rat aorta by the explant method as described previously [20]. Briefly, cleaned aorta was cut into small pieces which were placed in a petri dish filled with DMEM in the presence of 10% FBS, 6 mM glutamine and 1% penicillin/streptomycin (10,000 units/ml and 10 mg/ml respectively), and cultured in an incubator at 37◦C with humidified 5% CO2. After 3–4 days, cells spread out, aorta pieces were removed and cells started to proliferate. We used cultured VSMC up to passage 5. For Ca2+ measurement, VSMC were subcultured on glass coverslip (15 mm × 15 mm) at 100,000 cells/well until confluence and maintained in a serum-free medium for 24 h before use. For Western blot measurement, VSMC were subcultured in 6-well plates at 100,000 cells/well until confluence and maintained in a serum-free medium for 24 h before use. VSMC were stained with an -smooth muscle anti-actin antibody to confirm the identity of VSMC.

2.4. Measurement of isometric contractile tension and [Ca2+] in [Ca2+] was measured in 2mm rings from aorta loaded with 5 M fura-2 acetoxymethyl ester (fura-2 AM) dissolved in PSS containing 0.05% cremophor EL for 3 h at room temperature. After fura-2 loading period, rings were mounted under a tension of 20 mN in a 3 ml cuvette, which is part of a fluorimeter (CAF110, JASCO, Tokyo, Japan), perfused with PSS containingM N-nitro-l- arginine and gassed with 95% O2 and 5% CO2 at 37◦C. Fluorescence signals at 340 nm (F340), 380 nm (F380) and ratio (F340/F380), were simultaneously recorded with contractile tension, measured by an isometric force transducer, by using a data acquisition hardware (MacLab – sampling rate 4s−1) and data recording software (Chart v3.3, AD Instruments Pty Ltd., Castle Hill, Australia). After washing, artery rings were stimulated with 100 mM PSS KCl. To evaluate the different components of the Ca2+ signal, artery rings were first perfused with Ca2+-free solution containing 2mM EGTA for 2min and then stimulated with vasopressin (10nM) in Ca2+-free solution to evoke intracellular Ca2+ release. Ca2+ entry was thereafter induced by adding 1.2mM Ca2+ in the perfusion solution in the presence of the agonist. When required, vessels were incubated for 2–3 min with an inhibitor (OPC-21268 3 M, Y-27632 10 M, H1152 1 M, ML-7 3–10 M, verapamil 1 M) before the addition of Ca2+ and agonist response was evoked in the continuous presence of the inhibitor. Controls were performed in the presence of the same volume of solvent used to dissolve the inhibitor (water or DMSO). Ca2+ signal was calibrated at the end of each experiment. Maximal ratio (Rmax) was obtained by adding ionomycin (10 M) in PSS KCl, while minimal ratio (Rmin) was measured in the presence of EGTA (2.6 mM). After washing, the autofluorescence of the artery was measured at 340 nm and 380 nm by quenching the fura-2 fluorescence with MnCl2 (6.6 mM). Autofluorescence values were subtracted from experimental values measured at 340 nm and 380nm in order to calculate [Ca2+] by the application of the Grynkiewicz equation, as previously described [5,21]. Vasopressininduced Ca2+ entry was expressed as the difference between the [Ca2+] measured in the presence of vasopressin after the addition of Ca2+ in the perfusion and the basal [Ca2+] measured in normal cyt physiological solution.

2.5. Measurement of [Ca2+] in VSMC

VSMC cultured on glass coverslip were washed with BSS and loaded with 2.5 M fura-2 AM for 1 h30 at room temperature. Coverslip was mounted in a fluorimeter (CAF100, JASCO, Japan) and washed with BSS. Ca2+ signal was then measured as described for the artery rings.

2.6. Determination of ERM and LC20 phosphorylation and MLCK expression

Phospho-ERM, phospho-LC20 and MLCK expression were determined by Western blot. Prior to Western blot assessment, aorta rings (4 mm) or VSMC were stimulated with 10 nM vasopressin for 2 min in the absence or presence of inhibitor (Y-27632 10 M, H-1152 1 M or ML-7 3-10 M). Aorta rings were snapfrozen in liquid nitrogen. In order to avoid dephosphorylation of phospho-proteins, aorta rings were fixed in 10% acetone/TCA for 1 h at −80◦C and 20 min at −20◦C, and washed in cold acetone. Proteins were then extracted with a urea-containing lysis buffer (urea 9.5 M; IGEPAL CA-630 2%; pharmalytes GE Healthcare 1%; 2-mercaptoethanol 5%) supplemented with 1% Halt Protease & Phosphatase inhibitors cocktail (Thermo Scientific). VSMC were directly lysed with lysis buffer (NaCl, 137 mM; Tris, 20 mM; glycerol, 10%, IGEPAL CA-630, 2%; EDTA, 2 mM) supplemented with 1% Halt Protease & Phosphatase inhibitors cocktail and the protein extract was quickly frozen in liquid nitrogen.Proteins were resolved by SDS page on 4–12% Bis–Tris gels NUPAGE (Invitrogen, Merelbeke, Belgium) and transferred to polyvinylidene difluoride (PVDF) membranes (Invitrogen, Merelbeke, Belgium). Membranes were probed with anti-phospho-ERM, anti-phospho-LC20 or anti-MLCK and anti-actin as loading control. Bands were detected with fluorescent secondary antibodies and quantified with an Odyssey infrared imaging system (Li-Cor, Westburg, Leusden, The Netherlands). The expression of phospho-ERM, phospho-LC20 and MLCK was normalized to actin expression for each sample.

2.7. Assessment of RhoA activation

The activation of RhoA was determined by using an absorbance based G-LISA kit following the manufactor instructions (GLISA RhoA Activation Assay Biochem Kit, Cytoskeleton, Denver,USA). Before GTP-RhoA assessment, aorta rings or cultured VSMC were stimulated for 2 min with vasopressin 10 nM and snap-frozen in liquid nitrogen. Total RhoA was determined by Western blot with an anti-RhoA and was used to normalize GTP-RhoA expression. Activated RhoA in stimulated samples was compared to the level of GTP-RhoA in unstimulated samples.

2.8. Statistical analysis

Data are presented as means ± SEM from at least 3 samples of aorta or aortic VSMC. Unpaired Student’s t tests or one-way ANOVA with Bonferroni’s post-test were used to compare control and treated conditions. Statistical analysis and graphs were performed with the GraphPad software from Prism. p-Values lower than 5% indicated a significant difference.

3. Results

3.1. Effect of Rho kinase inhibition on vasopressin-evoked Ca2+ signal in aorta compared to VSMC

As VSMC in culture lose their noradrenergic response [22], we used vasopressin as an agonist to induce the calcium response. To assess the effect of ROCK inhibition on vasopressin-induced Ca2+entry, Ca2+ stores were first emptied by stimulating cells with vasopressin in a Ca2+-free medium before activating Ca2+ entry by the addition of Ca2+ into the solution. The effect of vasopressin stimulation in cultured VSMC was compared to its effect in freshly isolated aorta (Fig. 1) The basal Ca2+ concentration was higher in the whole tissuethan in cultured cells (110 ± 4 nM, n = 12 and 59 ± 11 nM, n = 18, respectively). As shown in Fig. 1, vasopressin (10 nM) induced a Ca2+ response in both aorta and cultured VSMC. In both models,vasopressin induced the release of Ca2+ from an intracellular com-partment and activated Ca2+ entry after the addition of Ca2+ in theperfusionsolution.However,theCa2+ releasecomponentofthesig nal, measured in Ca2+-free solution, was larger in VSMC compared to the aorta (162 ± 19 nM, n = 18 and 23 ± 6 nM, n = 12, respectively). After Ca2+ entry activation by the addition of Ca2+ into the perfusionsolution, [Ca2+] increased by 157 8nM (n =18) in VSMC and by138 ± 15 nM (n = 12) in aorta (Fig. 1). These experiments showed that in VSMC, [Ca2+] measured after vasopressin-induced intra-cellular Ca2+ release and Ca2+ entry reached a similar level, while inaorta,theamplitudeofCa2+ releasefromintracellularstoresevokedby vasopressin was markedly smaller than the increase in [Ca2+]cyt resulting from the activation of Ca2+ entry. In order to determinewhether Ca2+ entry induced by vasopressin was activated by theemptying of intracellular Ca2+ stores or required the occupancy ofthe vasopressin receptor, aorta or VSMC were incubated in the presence of the vasopressin V1 receptor antagonist OPC-21268 (3 M) immediately after Ca2+ store depletion (Fig. 1B and E). In aorta,in the presence of OPC-21268 (3M), the re-addition of Ca2+ inthe perfusion with vasopressin only restored the [Ca2+] to itsbasal level so that vasopressin-evoked Ca2+ entry was completelyblocked, as was also the contraction (Fig. 1B and C). In cultured VSMC, vasopressin-induced Ca2+ entry was also nearly completelyabolished in the presence of the V1 receptor antagonist (Fig. 1E and F). Moreover, in aorta, the Ca2+ entry response was inhibited bygadolinium (Gd3+) (supplemental data 1), as observed in culturedVSMC [20].
The involvement of ROCK in Ca2+ entry was tested by addingROCK inhibitors in the perfusion solution. Y-27632 (10 M) or H-1152 (1M) was applied after the peak of Ca2+ release produced byvasopressin in Ca2+-free solution, in order to avoid any bias due toa potential change in the amplitude of Ca2+ release. As expected,in aorta (Fig. 1A and C), the Ca2+ entry signal induced by vasopressin was significantly inhibited in the presence of Y-27632, from 138 ± 15 nM to 38 ± 12 nM (n = 6, p < 0.0001 versus control). Similar inhibition was observed in the presence of the other ROCK inhibitorH-1152, which reduced Ca2+ entry to 53±7nM (n =5, p <0.001versus control) (Fig. 1C). Contraction induced by vasopressin measured 2min after the addition of Ca2+ in the perfusion, was quitesmall reaching 4.6 ± 0.6 mN (n = 12) or 33 ± 5% of the KCl-induced contraction. The contractile tension was completely blocked by both ROCK inhibitors (Fig. 1A).
In cultured VSMC, ROCK inhibition by either Y-27632 or H-1152 did not significantly change vasopressin-induced Ca2+ entry (Fig. 1D and F), which increased [Ca2+] by 136 12nM (n =12) or157 ± 15 nM (n = 5) in the presence of Y-27632 or H-1152, respectively.As the selectivity of Y-27632 could be questioned [23], we measured the effect of PKC inhibition in the presence of the PKCoperated Ca2+ channels blocker, verapamil, in aorta and cultured VSMC. Cytosolic calcium concentration ([Ca2+] ) was measured in control aorta rings and cultured aorticVSMC stimulated with 10 nM vasopressin, or after incubation in the presence of ROCK inhibitors, Y-27632 (10 M) or H-1152 (1 M), or in the presence of the vasopressin V1 receptor antagonist, OPC-21268 (3M), or in the presence of the voltage-operated Ca2+ channels blocker, verapamil (1M) added immediately after the peak of Ca2+release evoked by vasopressin in Ca2+-free solution. Contraction (mN) was simultaneously recorded in aorta rings. (A., B., D. and E.) Typical recordings of Ca2+ signal andcontraction in aorta (A. and B.) and Ca2+ signal in VSMC (D. and E.). Dotted line represents the basal level of cytosolic Ca2+. (C. and F.) Bar charts of the increase in cytosolic calcium concentration ([Ca2+] , in nM) measured as the difference between the cytoslic Ca2+ concentration measured after the addition of Ca2+ in the perfusion solutionand the basal level of cytosolic Ca2+ in controls and in the presence of Y-27632 10M, H-1152 1M, OPC-21268 3M or verapamil 1M. Data are mean values from 3 to12 different aorta rings (C.) and 4–18 different VSMC populations (F.). inhibitor, Ro-318220 (1M) on the Ca2+ signal in vasopressin-stimulated VSMC. As already shown in noradrenaline-stimulated aorta (Ghisdal et al. [5]), the Ca2+ signal induced by vasopressinwas not affected by PKC inhibition in VSMC (supplemental data 2).

3.2. Contribution of voltage-operated Ca2+ channels invasopressin-induced responses

As shown in Fig. 1A, the contraction evoked by an agonist in rat aorta involves both intracellular and extracellular Ca2+ mobilization. The extracellular Ca2+ entry can result from the opening ofvoltage-operated Ca2+ channels (VOC) or non-voltage dependentcation channels. To determine whether there could be a relation between the origin of the increase in [Ca2+] induced by vaso-pressin and its sensitivity to ROCK inhibitors, we investigated the sensitivity of the Ca2+ response to a blocker of VOC, verapamil(1 M), in aorta compared to VSMC (Fig. 1C and F). In fura-2 loaded aorta rings, vasopressin-evoked Ca2+ entry measured by the addition of Ca2+ in the Ca2+-free solution was inhibited by verapamilfrom 138 ± 15 nM (n = 12) to 39 ± 4.7 nM (n = 6, p < 0.001) (Fig. 1C).However, in cultured aortic VSMC, the Ca2+ entry induced byvasopressin was unaffected by verapamil: [Ca2+] increased by157 ± 8 nM (n = 18) in the absence of verapamil and 125 ± 22 nM (n = 5) in the presence of verapamil (Fig. 1F).

3.3. RhoA activation in isolated artery and cultured cells

We further investigated whether the difference in sensitivity to ROCK inhibition between aorta and cultured cells could be related to a difference in the upstream activator of ROCK, the small GTPase RhoA [2], in response to vasopressin stimulation. We therefore measured the GTP-RhoA level in aorta and VSMC after 2 min stimulation with vasopressin (Fig. 2). In both models, vasopressin significantly activated RhoA (p < 0.001). The ratio GTP-RhoA/total RhoA significantly increased after vasopressin stimulation by a factor 2 ± 0.2 (n = 4) in aorta and 2.4 ± 0.1 (n = 3) in VSMC. These results suggestthattheabsenceofeffectofROCKinhibitionontheCa2+ signal in cultured cells cannot be explained by the lack of stimulation of RhoA by vasopressin.

3.4. Measurement of Rho kinase activation in vasopressin-stimulated aorta and VSMC

To determine whether the difference in sensitivity to ROCK inhibition of Ca2+ entry signal in aorta and cultured cells could berelated to difference in ROCK activation, we measured the phosphorylation of downstream targets of ROCK, the ERM proteins [4], after vasopressin stimulation in both aorta and VSMC. Vasopressin (10 nM) induced the phosphorylation of ERM proteins in aorta and VSMC (Fig. 3). The level of phospho-ERM expression in response to vasopressin was quite similar in aorta and in cultured cells: phospho-ERM expression increased by a factor 4.1 ± 0.5 (n = 14) in aorta and 5.9 ± 1.4 (n = 5) in VSMC. In agreement with the involvement of ROCK in ERM phosphorylation [24], phosphoERM expression in response to vasopressin in aorta was inhibited by 68 ± 9% (n = 4, p < 0.001) in the presence of Y-27632 (Fig. 3A). In cultured VSMC, Y-27632 inhibited ERM phosphorylation after vasopressin stimulation by 40 ± 5% (n = 8, p < 0.001; Fig. 3B). This effect was confirmed in the presence of the other inhibitor of ROCK, H-1152 (1 M), which inhibited vasopressin-stimulated phosphoERM expression by 40 ± 8.5% (n = 5, p < 0.001; Fig. 3B). These results confirm the expression and activation of ROCK after vasopressin stimulation in both aorta and cultured VSMC.

3.5. Role of myosin light chain kinase in the calcium response to vasopressin

A role for MLCK has been proposed in the regulation of Ca2+entry through the modulation of TRPC channel activity in HEK cells [11,12]. We, therefore, investigated the effect of a MLCK inhibitor, ML-7, in vasopressin-induced Ca2+ entry in aorta compared toVSMC (Fig. 4). Aortas were pre-incubated in the presence of verapamil (1 M), to eliminate the contribution of VOC in the response to vasopressin. In verapamil-treated aorta, ML-7 at 3 M depressed the Ca2+ entry signal induced by vasopressin from 39±4.7nM(n = 6) to 3.6 ± 1.5 nM (n = 6, p < 0.001; Fig. 4A). Simultaneously,contraction was reduced from 3.2 ± 0.4 mN (n = 6) in the presence of verapamil alone to 0.8 ± 0.3 mN (n = 6, p < 0.01) after incubation with ML-7 and verapamil.However, in cultured VSMC, ML-7, even at a higher concentration of 10M, did not affect the Ca2+ signal induced by vasopressin (Fig. 4B).

3.6. Measurement of MLCK activation in vasopressin-stimulated aorta and cultured VSMC

We further investigated the activity of MLCK in cultured VSMC compared to aorta to determine whether the difference in sensitivity to MLCK inhibition of vasopressin-induced Ca2+ signaling pathway in aorta and VSMC could be related to difference in MLCK activation. To this aim, we measured myosin light chain (LC20) phosphorylation in whole aorta and in cultured VSMC after vasopressin stimulation in the presence of the MLCK inhibitor, ML-7 at 3 M in aorta and 10 M in VSMC. LC20 was phosphorylated by vasopressin in aorta and in cultured VSMC although this phosphorylation was nearly ten times lower in VSMC than in vasopressin-stimulated aorta: phospho-LC20 was increased by a factor 18 ± 1 (n = 12) in aorta versus 2.7 ± 0.3 (n = 10) in VSMC. ML-7 inhibited vasopressin-induced LC20 phosphorylation in aorta by 28 ± 5% (n = 6, p < 0.001) while vasopressin-induced phospho-LC20 was not affected by ML-7 treatment in VSMC, suggesting that the level of MLCK activation by vasopressin in VSMC was very low (Fig. 5). Moreover, we observed that phosphorylation of LC20 by vasopressin was significantly inhibited by Y-27632 in both aorta and VSMC, inhibition reaching 47 ± 13% (n = 7, p < 0.001) and 37 ± 12% (n = 6, p < 0.05) respectively, which confirms the expression and the activation of ROCK in aorta as in VSMC. To explain the low MLCK activity in vasopressin-stimulated VSMC, we measured the MLCK expression by Western blot in VSMC compared to aorta. As shown in Fig. 6, MLCK expression was markedly lower in VSMC compared to aorta, which could explain the absence of phospho-LC20 inhibition by ML-7 in VSMC.

4. Discussion

The present study shows that ROCK participates in the cytosolic Ca2+ increase induced by vasopressin in isolated aorta but not in cultured aortic smooth muscle cells, despite the similar activation of the RhoA-ROCK pathway by vasopressin in isolated artery and cultured cells. Our results suggest that Ca2+ regulation by ROCK could be mediated by myosin phosphorylation in isolated artery, while in VSMC, the lack of sensitivity could be related to the weak phosphorylation of myosin. VSMC are known to change from a native contractile phenotype to a proliferative phenotype when they are cultured and this phenotypic switching is associated with major modifications affecting Ca2+ handling and excitation-response coupling [19]. Among others, cultured VSMC lose their noradrenergic response [22]. This is the reason why we used vasopressin as agonist to compare the receptor-activated Ca2+ response in isolated aorta and VSMC.
In Ca2+-free solution, vasopressin induced a rapid and transient Ca2+ release, which was markedly larger in cultured cells than in aorta. This observation is in agreement with the increased IP3sensitive Ca2+ content of intracellular Ca2+ stores, described in proliferative VSMC compared to freshly isolated cells, which is explained by the increased expression of IP3 receptor and sarcoplasmic reticulum Ca2+-ATPase (SERCA) pump in proliferating cells [19].
The addition of Ca2+ into the perfusion in the presence of vasopressin evoked a rapid Ca2+ entry both in isolated artery and cultured cells. This Ca2+ entry was completely abolished in the presence of the V1 receptor antagonist OPC-21268, despite the previous depletion of Ca2+ stores. This observation suggests that vasopressin-induced Ca2+ entry requires the occupancy of the V1 receptor and that vasopressin-induced store depletion by itself does not activate Ca2+ entry, in isolated aorta as in VSMC.
The vasopressin-induced Ca2+ increase was not sensitive to ROCK inhibitors (Y-27632, H-1152) in VSMC on the contrary to what was observed in isolated aorta. The pyridine derivative ROCK inhibitor, Y-27632 is relatively selective against ROCK as it has a 100 times greater potency for inhibiting ROCK than for conventional PKC isoforms [25]. In addition, most results obtained with Y-27632 were confirmed with H-1152, which is 10 times more potent on ROCK than Y-27632 with a very weak activity against other kinases [23]. The concentration of inhibitors (10 M for Y-27632 and 1 M for H-1152) was chosen on the basis of concentration-effect relations, so that they produced maximum inhibition of the contraction to agonist in isolated artery [26]. In aorta, ROCK inhibition decreased the Ca2+ entry signal induced by vasopressin by 62–72%. This result is in agreement with the inhibition of noradrenaline-stimulated Ca2+ entry previously observed in aorta rings [5]. Ghisdal et al. further demonstrated that in isolated aorta, inhibition of ROCK does not affect either KCl-induced Ca2+ entry or Ca2+ entry activated by intracellular store depletion in response to SERCA pump inhibition by thapsigargin, suggesting that ROCK-dependent Ca2+ entry is distinct from VOC and from store-depletion operated channels [5]. The best candidates for driving receptor-activated non-voltagedependent Ca2+ influx are the non-selective cation channels of the TRPC family, which are reported to be expressed and activated by receptor stimulation in different types of arteries, including the aorta [27–30] and cultured VSMC [19,20,31]. The inhibition of receptor-activated Ca2+ entry by Gd3+ in cultured cells [20] as in isolated aorta (supplemental data 1) is in favor of the participation of TRPC channels [32]. Nonetheless, Gd3+ has been reported to inhibit ORAI-mediated store-dependent Ca2+ entry in proliferative VSMC [33] so that the selectivity of this antagonist for TRPC channels can be questioned. Receptor-operated TRPC channels are reported to be activated by DAG in a PKC-independent manner [34], which is in agreement with the absence of effect of PKC inhibition on noradrenaline-activated calcium entry in isolated artery [5] and on vasopressin-stimulated Ca2+ entry in VSMC (supplemental data 2).
So far, no element seems to indicate that the difference in ROCK sensitivity of receptor-operated Ca2+ entry between isolated aorta and cultured cells could be related to TRPC channels expression. Indeed, the main change associated with proliferation consists in the upregulation of the expression of different TRPC isoforms, among which TRPC1, TRPC3 and TRPC6 [19,20,32,35] and the increase in the whole cell Ca2+ current [36].
On the opposite to non-voltage-dependent Ca2+ channels, L-type VOC expression is affected by cell culture [36], which is in agreement with the absence of effect of the VOC blocker verapamil on the Ca2+ entry signal induced by vasopressin in cultured cells, while verapamil inhibited the increase in cytosolic Ca2+ induced by vasopressin in aorta. As ROCK inhibition still depressed receptor-induced Ca2+ signal in VOC blocker-treated aorta [5], it is unprobable that a difference in the contribution of VOC is responsible for the different sensitivity of the Ca2+ response to ROCK inhibition between aorta and cultured cells.
The difference in ROCK sensitivity of receptor-operated Ca2+ entry between isolated aorta and cultured cells could be related to the expression or activation mechanism of ROCK. The resistance of the Ca2+ entry signal to ROCK inhibition in cultured VSMC cannot be explained by the lack of expression of ROCK. Indeed, ROCK has been reported to be expressed in VSMC [37], in A7r5, a cell line derived from embryonic rat aorta [38], and in several other smooth muscle cells [39], where it is involved in cell migration, proliferation, differentiation, apoptosis [40]. The activation of ROCK by vasopressin was tested by measuring the phosphorylation of ERM proteins in isolated aorta and in cultured aortic cells. The ERM proteins, which serve as cross-linkers between actin filament and membrane proteins at the cell surface [7], are well known targets of ROCK. In intact aorta, noradrenaline induces a ROCK-dependent phosphorylation of ERM proteins [24]. Vasopressin similarly induced the phosphorylation of ERM in both aorta and cultured VSMC, and vasopressin-dependent ERM phosphorylation was inhibited in the presence of a ROCK inhibitor, suggesting that vasopressin induced the activation of the RhoA/ROCK pathway in cultured aortic cells as in aorta. In agreement with these data, the level of GTP-RhoA, the upstream activator of ROCK, was not different in vasopressin-stimulated VSMC and in intact aorta. Furthermore, as the receptor-activated Ca2+ signal is not affected by the inhibition of arachidonic acid production either in VSMC [20] or in intact aorta (supplemental data 3), it is unlikely that direct activation of ROCK by arachidonic acid [41] is involved in vasopressin-induced Ca2+ responses.
These results suggest that the difference in the Ca2+ signal sensitivity to ROCK inhibition between aorta and cultured aortic cells is related to the downstream pathway activated by ROCK. ROCK is known to regulate several cell functions associated with cytoskeleton changes [40]. Actin does not appear to be involved in the ROCK-dependent activation of Ca2+ entry in receptor-activated artery [8]. The contribution of myosin was tested by investigating the effect of LC phosphorylation on receptor-activated Ca2+ signal. We used the MLCK inhibitor, ML-7. ML-7 did not affect the Ca2+ entry signal in cultured VSMC, even at higher concentration (10 M), but this compound significantly inhibited vasopressininduced Ca2+ entry in aorta at a lower concentration (3M). This effect is in agreement with the inhibition of vasopressin-stimulated MLCK activity measured by the phosphorylation of LC20 in aorta. On the opposite, ML-7 did not affect phospho-LC20 expression in cultured cells, where the phosphorylation of LC20 by vasopressin was markedly lower than in aorta. The non-selectivity of ML-9, a homologous compound of ML-7, against MLCK has been pointed out since MLCK-independent inhibition of TRPC6 channel has been reported [42]. High concentration of ML-9 has also been shown to inhibit cytosolic Ca2+ increase in response to KCl in guinea-pig tracheal smooth muscle [43]. However, ML-7 is more than 10-fold more potent in inhibiting MLCK than ML-9 (IC50: 300 nM vs 3.8 M, respectively) [44]. In addition, direct inhibition of VOC by ML-7 cannot explain the effect of this compound on the calcium signal in aorta as it was tested in the presence of the VOC blocker verapamil. Furthermore, comparison of MLCK expression in isolated aorta and cultured cells indicated that MLCK was markedly downregulated in proliferative cells, which can explain the loss of the ML-7 sensitivity of the Ca2+ signal induced by vasopressin. Since in proliferative cells MLCK and VOC are downregulated, further studies should investigate whether there is a link between these two observations.
Taken together, the present results suggest that ROCK involvement in Ca2+ entry activation might require functional MLCK and might involve myosin phosphorylation [12,45, for review]. However, the confirmation of the role of MLCK in the activation of calcium channel will require more selective tools. In addition, it would be interesting to determine whether the ROCK-MLCK-Ca2+pathway contributes to Ca2+ signaling in other arteries, like small resistance arteries and what could be its role in the development of pathological disorders that are associated with vascular smooth muscle cell growth.

5. Conclusion

Although cultured cells are often used to investigate intracellular signaling pathways, culture can introduce major bias in the study of the Ca2+ entry signaling pathway induced by receptor activation. Indeed, the present results show that ROCK is involved in receptor-dependent Ca2+ entry in aorta but not in cultured aortic smooth muscle cells, indicating that transposition of models obtained from cultured cells experiments to the situation in whole artery has to be done with caution.

References

[1] J.H. Exton, Phosphoinositide phospholipases and G proteins in hormone action, Annual Review of Physiology 56 (1994) 349–369.
[2] A.P. Somlyo, A.V. Somlyo, Ca2+ sensitivity of smooth muscle and nonmuscle myosin II: modulated by G proteins, kinases, and myosin phosphatase, Physiological Reviews 83 (2003) 1325–1358.
[3] S. Sakurada, H. Okamoto, N. Takuwa, N. Sugimoto, Y. Takuwa, Rho activation in excitatory agonist-stimulated vascular smooth muscle, American Journal of Physiology. Cell Physiology 281 (2001) C571–C578.
[4] A. Gohla, G. Schultz, S. Offermanns, Role for G12/G13 in agonist-induced vascular smooth muscle cell contraction, Circulation Research 87 (2000) 221–227.
[5] P. Ghisdal, G. Vandenberg, N. Morel, Rho-dependent kinase is involved in agonist-activated calcium entry in rat arteries, The Journal of Physiology 551 (2003) 855–867.
[6] K. Riento, A.J. Ridley, ROCKS: multifunctional kinases in cell behaviour, Nature Reviews Molecular Cell Biology 4 (2003) 446–456.
[7] T. Matsui, M. Maeda, Y. Doi, S. Yonemura, M. Amano, K. Kaibuchi, S. Tsukita, Rhokinase phosphorylates COOH-terminal threonines of ezrin/radixin/moesin (ERM) proteins and regulates their head-to-tail association, Journal of Cell Biology 140 (1998) 647–657.
[8] A. Martinsen, X. Yerna, G. Rath, E. Leon Gomez, C. Dessy, N. Morel, Different effect of Rho kinase inhibition on calcium signaling in rat isolated large and small artery, Journal of Vascular Research, http://dx.doi.org/10.1159/ 000341230, in press.
[9] S.J. Gunst, W. Zhang, Actin cytoskeletal dynamics in smooth muscle: a new paradigm for the regulation of smooth muscle contraction, American Journal of Physiology. Cell Physiology 295 (2008) C576–C587.
[10] C.L. Chou, B.M. Christensen, S. Frische, H. Vorum, R.A. Desai, J.D. Hoffert, P. de Lanerolle, S. Nielsen, M.A. Knepper, Non-muscle myosin II and myosin light chain kinase are downstream targets for vasopressin signaling in the renal collecting duct, Journal of Biological Chemistry 279 (2004) 49026–49035.
[11] M.T. Kim, B.J. Kim, J.H. Lee, S.C. Kwon, D.S. Yeon, D.K. Yang, I. So, K.W. Kim, Involvement of calmodulin and myosin light chain kinase in activation of mTRPC5 expressed in HEK cells, American Journal of Physiology. Cell Physiology 290 (2006) C1031–C1040.
[12] S. Shimizu, T. Yoshida, M. Wakamori, M. Ishii, T. Okada, M. Takahashi, M. Seto, K. Sakurada, Y. Kiuchi, Y. Mori, Ca2+-calmodulin-dependent myosin light chain kinase is essential for activation of TRPC5 channels expressed in HEK293 cells, The Journal of Physiology 570 (2006) 219–235.
[13] B.J. Kim, J.H. Jeon, S.J. Kim, I. So, Role of calmodulin and myosin light chain kinase in the activation of carbachol-activated cationic current in murine ileal myocytes, Canadian Journal of Physiology and Pharmacology 85 (2007) 1254–1262.
[14] Y.C. Kim, S.J. Kim, T.M. Kang, S.H. Suh, I. So, K.W. Kim, Effects of myosin light chain kinase inhibitors on carbachol-activated nonselective cationic current in guinea-pig gastric myocytes, Pflugers Archiv: European Journal of Physiology 434 (1997) 346–353.
[15] A.S. Aromolaran, A.P. Albert, W.A. Large, Evidence for myosin light chain kinase mediating noradrenaline-evoked cation current in rabbit portal vein myocytes, The Journal of Physiology 524 (Pt 3) (2000) 853–863.
[16] H. Watanabe, Q.K. Tran, K. Takeuchi, M. Fukao, M.Y. Liu, M. Kanno, T. Hayashi, A. Iguchi, M. Seto, K. Ohashi, Myosin light-chain kinase regulates endothelial calcium entry and endothelium-dependent vasodilation, FASEB Journal 15 (2001) 282–284.
[17] Q.K. Tran, H. Watanabe, H.Y. Le, L. Pan, M. Seto, K. Takeuchi, K. Ohashi, Myosin light chain kinase regulates capacitative Ca2+ entry in human monocytes/macrophages, Arteriosclerosis, Thrombosis, and Vascular Biology 21 (2001) 509–515.
[18] G.K. Owens, M.S. Kumar, B.R. Wamhoff, Molecular regulation of vascular smooth muscle cell differentiation in development and disease, Physiological Reviews 84 (2004) 767–801.
[19] R. Berra-Romani, A. Mazzocco-Spezzia, M.V. Pulina, V.A. Golovina, Ca2+ handling is altered when arterial myocytes progress from a contractile to a proliferative phenotype in culture, American Journal of Physiology. Cell Physiology 295 (2008) C779–C790.
[20] K. Tai, M.C. Hamaide, H. Debaix, P. Gailly, M. Wibo, N. Morel, Agonist-evoked calcium entry in vascular smooth muscle cells requires IP3 receptor-mediated activation of TRPC1, European Journal of Pharmacology 583 (2008) 135–147.
[21] G. Grynkiewicz, M. Poenie, R.Y. Tsien, A new generation of Ca2+ indicators withgreatly improved fluorescence properties, Journal of Biological Chemistry 260 (1985) 3440–3450.
[22] J. Chamley-Campbell, G.R. Campbell, R. Ross, The smooth muscle cell in culture, Physiological Reviews 59 (1979) 1–61.
[23] M. Tamura, H. Nakao, H. Yoshizaki, M. Shiratsuchi, H. Shigyo, H. Yamada, T. Ozawa, J. Totsuka, H. Hidaka, Development of specific Rho-kinase inhibitors and their clinical application, Biochimica et Biophysica Acta 1754 (2005) 245–252.
[24] N. Baeyens, S. Horman, D. Vertommen, M. Rider, N. Morel, Identification H-1152 and functional implication of a Rho kinase-dependent moesin-EBP50 interaction in noradrenaline-stimulated artery, American Journal of Physiology. Cell Physiology 299 (2010) C1530–C1540.
[25] M. Uehata, T. Ishizaki, H. Satoh, T. Ono, T. Kawahara, T. Morishita, H. Tamakawa, K. Yamagami, J. Inui, M. Maekawa, S. Narumiya, Calcium sensitization of smooth muscle mediated by a Rho-associated protein kinase in hypertension, Nature 389 (1997) 990–994.
[26] W.W. Shum, G.Y. Le, R.L. Jones, A.M. Gurney, Y. Sasaki, Involvement of Rhokinase in contraction of guinea-pig aorta induced by prostanoid EP3 receptor agonists, British Journal of Pharmacology 139 (2003) 1449–1461.
[27] K. Tai, G. Vandenberg, M.C. Hamaide, M. Wibo, N. Morel, Effect of organ culture on noradrenaline-evoked contraction, calcium signalling and TRPC expression in rat mesenteric artery, Journal of Vascular Research 46 (2009) 353–364.
[28] R. Inoue, T. Okada, H. Onoue, Y. Hara, S. Shimizu, S. Naitoh, Y. Ito, Y. Mori, The transient receptor potential protein homologue TRP6 is the essential component of vascular alpha 1-adrenoceptor-activated Ca2+-permeable cationchannel, Circulation Research 88 (2001) 325–332.
[29] A. Dietrich, Y.S.M. Mederos, M. Gollasch, V. Gross, U. Storch, G. Dubrovska, M. Obst, E. Yildirim, B. Salanova, H. Kalwa, K. Essin, O. Pinkenburg, F.C. Luft, T. Gudermann, L. Birnbaumer, Increased vascular smooth muscle contractility in TRPC6−/− mice, Molecular and Cellular Biology 25 (2005) 6980–6989.
[30] S.N. Saleh, A.P. Albert, C.M. Peppiatt, W.A. Large, Angiotensin II activates two cation conductances with distinct TRPC1 and TRPC6 channel properties in rabbit mesenteric artery myocytes, The Journal of Physiology 577 (2006) 479–495.
[31] S. Jung, R. Strotmann, G. Schultz, T.D. Plant, TRPC6 is a candidate channel involved in receptor-stimulated cation currents in A7r5 smooth muscle cells, American Journal of Physiology. Cell Physiology 282 (2002) C347–C359.
[32] A.J. Hill, J.M. Hinton, H. Cheng, Z. Gao, D.O. Bates, J.C. Hancox, P.D. Langton, A.F. James, A TRPC-like non-selective cation current activated by alpha 1adrenoceptors in rat mesenteric artery smooth muscle cells, Cell Calcium 40 (2006) 29–40.
[33] M. Potier, J.C. Gonzalez, R.K. Motiani, I.F. Abdullaev, J.M. Bisaillon, H.A. Singer,M. Trebak, Evidence for STIM1- and Orai1-dependent store-operated calcium influx through ICRAC in vascular smooth muscle cells: role in proliferation and migration, FASEB Journal 23 (2009) 2425–2437.
[34] A.P. Albert, Gating mechanisms of canonical transient receptor potential channel proteins: role of phosphoinositols and diacylglycerol, Advances in Experimental Medicine and Biology 704 (2011) 391–411.
[35] A. Bergdahl, M.F. Gomez, A.K. Wihlborg, D. Erlinge, A. Eyjolfson, S.Z. Xu, D.J. Beech, K. Dreja, P. Hellstrand, Plasticity of TRPC expression in arterial smooth muscle: correlation with store-operated Ca2+ entry, American Journal of Physology. Cell Physiology 288 (2005) C872–C880.
[36] S.J. House, M. Potier, J. Bisaillon, H.A. Singer, M. Trebak, The non-excitable smooth muscle: calcium signaling and phenotypic switching during vascular disease, Pflugers Archiv: European Journal of Physiology 456 (2008) 769–785.
[37] T.P. Woodsome, A. Polzin, K. Kitazawa, M. Eto, T. Kitazawa, Agonist- and depolarization-induced signals for myosin light chain phosphorylation and force generation of cultured vascular smooth muscle cells, Journal of Cell Science 119 (2006) 1769–1780.
[38] R.W. Rees, T. Ziessen, D.J. Ralph, P. Kell, S. Moncada, S. Cellek, Human and rabbit cavernosal smooth muscle cells express Rho-kinase, International Journal of Impotence Research 14 (2002) 1–7.
[39] O. Nakagawa, K. Fujisawa, T. Ishizaki, Y. Saito, K. Nakao, S. Narumiya, ROCK-I and ROCK-II, two isoforms of Rho-associated coiled-coil forming protein serine/threonine kinase in mice, FEBS Letters 392 (1996) 189–193.
[40] G. Loirand, P. Guerin, P. Pacaud, Rho kinases in cardiovascular physiology and pathophysiology, Circulation Research 98 (2006) 322–334.
[41] J. Feng, M. Ito, Y. Kureishi, K. Ichikawa, M. Amano, N. Isaka, K. Okawa, A. Iwamatsu, K. Kaibuchi, D.J. Hartshorne, T. Nakano, Rho-associated kinase of chicken gizzard smooth muscle, Journal of Biological Chemistry 274 (1999) 3744–3752.
[42] J. Shi, S. Takahashi, X.H. Jin, Y.Q. Li, Y. Ito, Y. Mori, R. Inoue, Myosin light chain kinase-independent inhibition by ML-9 of murine TRPC6 channels expressed in HEK293 cells, British Journal of Pharmacology 152 (2007) 122–131.
[43] S. Ito, H. Kume, H. Honjo, I. Kodama, H. Katoh, H. Hayashi, K. Shimokata, ML-9, a myosin light chain kinase inhibitor, reduces intracellular Ca2+ concentration inguinea pig trachealis, European Journal of Pharmacology 486 (2004) 325–333.
[44] M. Saitoh, T. Ishikawa, S. Matsushima, M. Naka, H. Hidaka, Selective inhibition of catalytic activity of smooth muscle myosin light chain kinase, Journal of Biological Chemistry 262 (1987) 7796–7801.
[45] I.S. Ambudkar, Trafficking of TRP channels: determinants of channel function, Handbook of Experimental Pharmacology (2007) 541–557.