Small GTPase Rho and its downstream effector, Rho kinase, have been implicated in agonist-stimulated Ca2+ sensitization of 20-kDa myosin light chain (MLC20) phosphorylation and contraction in smooth muscle. In the present study we demonstrated for the first time that excitatory receptor agonists induce increases in amounts of an active GTP-bound form of RhoA, GTP-RhoA, in rabbit aortic smooth muscle. Using a pull-down assay with a recombinant RhoA-binding protein, Rhotekin, we found that a thromboxane A2 mimetic, U-46619, which induced a sustained contractile response, induced a sustained rise in the amount of GTP-RhoA in a dose-dependent manner with an EC50 value similar to that for the contractile response. U-46619-induced RhoA activation was thromboxane A2 receptor-mediated and reversible. Other agonists including norepinephrine, serotonin, histamine, and endothelin-1 (ET-1) also stimulated RhoA, albeit to lesser extents than U-46619. In contrast, ANG II and phorbol 12,13-dibutyrate failed to increase GTP-RhoA. The tyrosine kinase inhibitor genistein substantially inhibited RhoA activation by these agonists, except for ET-1. Thus excitatory agonists induce Rho activation in an agonist-specific manner, which is thought to contribute to stimulation of MLC20 phosphorylation Ca2+ sensitivity.
- myosin light chain phosphorylation
- Rho kinase
- myosin phosphatase
- tyrosine kinase
rho is a memberof the Ras superfamily of small GTP-binding proteins (40). Like Ras, Rho cycles between a GDP-bound inactive state and a GTP-bound active state. Rho has been implicated in actin reorganization, cell shape change, cell motility, and cell growth and transformation in non-smooth muscle cells (24, 30, 40). Accumulating evidence demonstrates that Rho also functions as a signaling molecule in smooth muscle, in which it activates a mechanism that sensitizes the contractile machinery to Ca2+ (13, 23, 33,36). We (23, 25) and others (6, 11,13, 19, 20) demonstrated, in permeabilized vascular and non-vascular smooth muscle, that excitatory agonists and guanosine 5′-O-(3-thiotriphosphate) (GTPγS) potentiated Ca2+-induced 20-kDa myosin light chain (MLC20) phosphorylation and contraction in a manner sensitive to Rho-inactivating bacterial toxins. It was also demonstrated that the Rho-dependent enhancement of Ca2+-induced MLC20phosphorylation was caused by inhibition of myosin phosphatase activity (19, 20, 23, 25, 36). In intact smooth muscle, cell-permeable Rho-inactivating bacterial toxins have been shown to suppress receptor agonist-induced MLC20 phosphorylation and contraction (5, 22). Subsequent investigations demonstrated that a Rho-activated serine/threonine protein kinase, Rho kinase/ROCK/ROK, mediated Rho-dependent myosin phosphatase inhibition in smooth muscle, through Rho-kinase-catalyzed phosphorylation of a 110-kDa myosin targeting subunit, MYPT/MBS, of myosin phosphatase at Thr-695 (17, 23, 32, 35, 36, 39). Very recently, it was suggested that Rho-kinase-mediated phosphorylation of a smooth muscle-specific myosin phosphatase inhibitor protein, CPI-17, also contributed to myosin phosphatase inhibition in agonist-stimulated smooth muscle (3, 18). Despite these observations, no direct effects of excitatory agonists on Rho activity in smooth muscle have been determined.
Recently, a new biochemical assay for determining amounts of an active form of RhoA, GTP-bound RhoA (GTP-RhoA), in cells has been developed (29). This assay showed, in non-muscle cells, that stimulation of cells with serum or lysophosphatidic acid (LPA) and cell adhesion to extracellular matrices induced an increase in the cellular amount of GTP-RhoA (29). Using this technique, we evaluated whether excitatory agonists had any effects on the amount of GTP-RhoA in vascular smooth muscle. Previous studies (7, 8, 16,30) on non-smooth muscle cells demonstrated the involvement of tyrosine kinases in Rho-induced stress fiber formation and morphological changes such as neurite retraction. These studies suggested that tyrosine kinases were involved in the activation process of Rho. Therefore, we also examined the involvement of tyrosine kinases in agonist-induced RhoA activation in vascular smooth muscle.
MATERIALS AND METHODS
Tissue preparation and tension measurements.
Male Japanese White rabbits weighing 2.0–2.5 kg were killed by an intravenous injection of pentobarbital sodium. The animals were maintained in compliance with the “Guidelines of the Care and Use of Laboratory Animals” at the Takara-machi campus of Kanazawa University. The descending portion of the thoracic aorta was removed immediately and placed into an ice-cold modified Krebs-Henseleit buffer (119 mM NaCl, 4.7 mM KCl, 1.2 mM KH2PO4, 1.5 mM MgSO4, 1.5 mM CaCl2, 25 mM NaHCO3, and 11 mM glucose) aerated with 95% O2-5% CO2(37, 38). The aorta was cleaned of adhering loose connective tissues and cut into rings. Endothelial cells were removed by gently rubbing the intimal surface with a wooden stick. The rings were mounted under 1 g of resting tension in 10-ml static incubation muscle chambers, equilibrated in the modified Krebs-Henseleit buffer at 37°C, and gassed with 95% O2-5% CO2. The rings were stimulated, and the tension generated was measured isometrically with force-displacement transducers (UM-203; Kishimoto Medical Instruments, Kyoto, Japan).
The coding sequences for the Rho-binding domain of Rhotekin (amino acids 7–89) were amplified from mouse brain total RNA by RT-PCR with the sense primer 5′-GGGGATCCATCCTGGAGGACCTCAATATGCTCTACA-3′ and the antisense primer 5′-GGGAATTCTTAGCCTGTCTTCTCCAGCACCTGGGCCTCCT-3′ and cloned into a pGEX-2T vector (Amersham-Pharmacia) at BamHI and EcoRI sites. The sequences were confirmed by DNA sequencing. Human RhoA cDNA obtained from Dr. Pascal Madaule (Institut National de la Santé et de la Recherche Médicale, Paris, France) was ligated into a pGEX-4T-2 vector (Amersham-Pharmacia) atEcoRI-SalI sites.
The GST-Rho binding domain of Rhotekin fusion protein (GST-Rhotekin) and GST-RhoA were produced according to the instruction manual provided by the supplier of the pGEX vectors. RhoA was cleaved from GST-RhoA by thrombin (31), and after the cleavage, thrombin was eliminated by passing the reaction mixture through a benzamidine Sepharose 6B column (Amersham-Pharmacia). The RhoA thus generated was loaded with [35S]GTPγS (Amersham-Pharmacia) (31).
Determination of tissue GTP-RhoA.
Aortic rings contracted isometrically and were quickly frozen by immersing in liquid nitrogen. Frozen tissues were homogenized in 450 μl of a homogenizing buffer comprising 50 mM Tris · HCl (pH 7.2), 1% Triton X-100, 0.5% deoxycholic acid, 0.1% SDS, 500 mM NaCl, 10 mM MgCl2, 20 μg/ml each of leupeptin and aprotinin, and 1 mM phenylmethylsulfonyl fluoride (PMSF) (26, 29). Homogenates were clarified by centrifugation at 14,000 g at 4°C for 10 min. A small portion of supernatants (25 μl) was taken to determine protein concentrations by Lowry's methods and amounts of total RhoA by Western blot analysis. Equal amounts of supernatants (450 μg of proteins) were incubated with GST-Rhotekin (20 μg) immobilized onto GSH-Sepharose 4B beads (Amersham-Pharmacia) at 4°C for 30 min. The beads were then washed twice with a washing buffer comprising 50 mM Tris · HCl (pH 7.2), 1% Triton X-100, 150 mM NaCl, 10 mM MgCl2, 20 μg/ml each of leupeptin and aprotinin, and 1 mM PMSF. RhoA bound to beads was solubilized in Laemmli's SDS sample buffer and boiled for 5 min. Protein (17 μg of protein) from each extract was subjected to Western analysis with an anti-RhoA antibody to evaluate total amounts of RhoA in the extract. Each sample was analyzed by SDS-15% PAGE, followed by Western blotting with a specific mouse monoclonal anti-RhoA antibody (26C4; Santa Cruz Biotechnology) and an alkaline phosphatase-conjugated rabbit anti-mouse IgG1 antibody (Zymed). Specific bands were visualized with nitroblue tetrazolium and 5-bromo-4-chloro-3-indolylphosphate. This procedure specifically detected GTP-RhoA, because other small G proteins, including Rac, Cdc42, and Ras, were not detected in cellular proteins bound to GST-Rhotekin by Western blot analysis with specific antibodies against these small GTPases. Densities of bands corresponding to RhoA were quantitated with a Quantity One bioimage analyzer (Precision Data Imaging, Oceanside, NY). There was a linear relationship between the loaded amount of extracts and the densitometric value of RhoA immunoreactivity over the range of values observed in our assay. The amount of GTP-RhoA was normalized for the total amount of RhoA in each sample. The quantitative data of normalized amounts of GTP-RhoA are expressed as multiples over a value in unstimulated tissues, which is expressed as 1.0.
U-46619, AG-1478, and daidzein were obtained from Calbiochem. Endothelin-1 (ET-1) was bought from the Peptide Institute (Osaka, Japan). Norepinephrine (NE), 5-hydroxytryptamine (5-HT), histamine (His), genistein, and phorbol 12,13-dibutyrate (PDBu) were obtained from Sigma. ONO-3708 was kindly donated by Ono Pharmaceuticals (Kyoto, Japan).
Each set of data was expressed as the mean ± SE. One-way or two-way analysis of variance (ANOVA) followed by Dunnet's test was performed to determine statistical significance in differences between mean values. For analysis of the data in Figs. 3 and 4 B, Student-Newman-Keuls multiple tests were employed.
Glutathione S-transferase (GST)-Rhotekin specifically binds to a GTP-loaded form of RhoA (see materials and methods and Ref. 28). We evaluated how qualitatively and efficiently GST-Rhotekin bound to GTP-RhoA. We incubated [35S]GTPγS-loaded recombinant RhoA ([35S]GTPγS-RhoA) in different amounts with GST-Rhotekin immobilized onto beads in the assay buffer and then washed the beads with the washing buffer and determined the 35S radioactivity associated with the beads. As shown in Table1, the bead-associated 35S radioactivity increased linearly in proportion with increases in added amounts of [35S]GTPγS-RhoA. GST-Rhotekin beads bound to 86–95% of [35S]GTPγS-RhoA added in three different amounts. Thus GST-Rhotekin binds to GTP-RhoA qualitatively and efficiently.
The addition of U-46619 (30 nM), a stable analog of thromboxane A2, to aortic smooth muscle induced a sustained contraction, reaching a plateau after 10–15 min (Fig.1 A). We freeze-clamped aortic smooth muscle contracted by U-46619 and determined the amounts of GTP-RhoA in it via a pull-down assay using recombinant GST-Rhotekin. The amount of GTP-Rho, which was at a detectable level in unstimulated aortic smooth muscle, substantially rose within 5 min of the addition of U-46619 and reached a plateau after 10 min (Fig. 1 B,top). The U-46619-induced increase in the amount of GTP-RhoA was sustained for at least 45 min of the observation period. Total amounts of cellular RhoA in a portion of the extracts are shown in Fig.1 B, bottom. The relative values of the GTP-Rho amounts normalized for the amounts of total RhoA are shown in Fig.1 C. The amount of GTP-RhoA was estimated to be ∼0.5% of total RhoA in the extract of nonstimulated smooth muscle. In the sustained plateau phase of RhoA activation, the normalized level of GTP-RhoA was approximately sevenfold higher than that in unstimulated muscle. The stimulatory effects of U-46619 on the amount of GTP-RhoA, as well as on the isometric tension development, were dose dependent: the EC50 value and maximally effective concentration of U-46619 for both effects were similar at 3–5 nM and 100 nM, respectively (compare Fig. 2,A and B). These results clearly show that an excitatory agonist induces the activation of RhoA in vascular smooth muscle.
When the specific thromboxane A2 receptor antagonist ONO-3708 (10 μM) was added to aortic smooth muscle precontracted with U-46619 (30 nM) for 15 min, it induced a gradual relaxation of aortic muscle with a decline in tension to ∼20% of the value in the absence of the antagonist after 45 min (Fig.3 A, top). Concomitant with this, ONO-3708 induced a marked decline in the amount of GTP-RhoA (6.0- vs. 2.3-fold over the basal level in the absence and presence of ONO-3708 after 45 min) (Fig. 3 A,bottom). When aortic smooth muscle was pretreated with the thromboxane A2 receptor antagonist, both contraction and RhoA activation by subsequent addition of U-46619 were nearly totally abolished (Fig. 3 B). Thus U-46619-induced activation of RhoA was thromboxane A2 receptor mediated and reversible.
We next examined the effects of several other excitatory agonists on RhoA activation in aortic smooth muscle (Fig.4). NE, 5-HT, His, ET-1, and PDBu all induced a sustained contraction of aortic smooth muscle. In contrast to these agonists, ANG II induced only a transient contraction. Aortic smooth muscle was contracted by these agonists at doses that induced peak tensions of the same amplitudes caused by U-46619 (30 nM) and was freeze-clamped at the time points when the tension reached a peak. NE, 5-HT, His, and ET-1 all induced increases in the amounts of GTP-RhoA. Interestingly, the stimulation of RhoA by these agonists was less than that caused by U-46619 and was not uniform; the stimulating effects on RhoA of NE and ET-1 were relatively stronger compared with that of 5-HT. In contrast, ANG II and PDBu failed to increase the amount of GTP-RhoA.
Previous studies in non-muscle cells, including neuronal (16) and non-neuronal cells (7, 8), demonstrated the involvement of tyrosine kinases, especially epidermal growth factor (EGF) receptor tyrosine kinase, in agonist-induced, RhoA-mediated neurite retraction and stress fiber formation. We tested the possibility that tyrosine kinases are involved in agonist-induced contraction and RhoA activation in smooth muscle by using the nonselective tyrosine kinase inhibitor genistein and the specific EGF receptor tyrosine kinase inhibitor AG-1478. Pretreatment of aortic smooth muscle with genistein, but not its inactive analog, daidzein, inhibited U-46619-induced contraction in a dose-dependent manner with a maximal 90% inhibition at 100 μM (Fig.5 A). The contraction induced by NE, 5-HT, and His was less sensitive to genistein compared with that induced by U-46619. Interestingly, ET-1-induced contraction was essentially insensitive to genistein (Fig. 5 B). Consistent with these observations on agonist-induced contraction, genistein inhibited RhoA activation induced by U-46619 most strongly and that induced by NE, 5-HT, and His to a lesser extent. On the other hand, genistein failed to inhibit ET-1-induced RhoA activation (Fig.5 C). Daidzein was ineffective in inhibiting U-46619-induced RhoA activation, indicating a specific inhibitory effect of genistein on RhoA activation (Fig. 5 D). Because a previous study (27) reported that genistein inhibited a voltage-dependent Ca2+ current, we examined the involvement of the voltage-dependent L-type Ca2+ channels in U-46619-induced RhoA activation. The dihydropyridine Ca2+-channel antagonist did not inhibit U-46619-induced RhoA activation up to 1 μM (Fig. 5 D). AG-1478, which has been demonstrated to suppress EGF-mediated responses at several tens of nanomolar concentrations in various cell types (7, 8, 16), did not inhibit contraction or the RhoA activation induced by any of these agonists in aortic smooth muscle at concentrations up to 10 μM.
The present study successfully applied a pull-down assay specific for GTP-RhoA to intact vascular smooth muscle tissues with the use of a recombinant GST-Rhotekin fusion protein. To evaluate cellular RhoA activity, determination of membrane-bound RhoA was previously performed (5, 10) because it was generally thought that membrane-bound RhoA represented a GTP-bound, active form of RhoA (36). However, it was recently shown that translocation of RhoA to membranes is also affected by the phosphorylation state of RhoA (21). A direct assay for measuring GTP loading of cellular RhoA by immunoprecipitating RhoA and determining RhoA-bound GTP was also reported (28). However, this technique could not detect an increase in RhoA-bound GTP because the available anti-RhoA antibodies did not inhibit Rho GTPase activity, and, hence, only GDP was detected after immunoprecipitation. Determination of an amount of GTP-RhoA with the pull-down assay using GST-Rhotekin gives a direct measure of cellular RhoA activity. The whole procedure is not complicated, and the assay has quite good reproducibility. This method would be useful for clarifying the roles for Rho in smooth muscle.
In the present study, we demonstrated for the first time that a variety of excitatory receptor agonists stimulate RhoA in vascular smooth muscle. RhoA has been shown to interact with a number of effector proteins and to relay its signal downstream (24, 40). In smooth muscle, Rho kinase is now recognized as an important Rho effector, which acts to bring about inhibition of myosin phosphatase activity (17, 23, 33, 35, 36, 39). The present results are consistent with the notion that excitatory receptor agonists activate Rho kinase through stimulation of RhoA activity, resulting in inhibition of myosin phosphatase. It is well established that excitatory agonists also activate the phospholipase C signaling pathway via heterotrimeric Gq/11, leading to an increase in the intracellular Ca2+ concentration and, consequently, myosin light chain kinase activation (28, 33, 36, 38). In excitatory agonist-stimulated smooth muscle, Rho- and Rho-kinase-mediated myosin phosphatase inhibition constitute the molecular basis for Ca2+ sensitization, i.e., potentiation of MLC20 phosphorylation and, consequently, contraction at a given Ca2+ concentration (9, 33, 35, 36,39). Consistent with this notion, the temporal profiles of RhoA activation and contraction caused by the thromboxane A2mimetic U-46619 were both sustained with similar kinetics (Fig. 1), and the U-46619 dose-response curves for these responses were quite similar (Fig. 2).
In the present study, all the excitatory agonists examined, except for ANG II, activated RhoA. The capacities of excitatory agonists to activate RhoA were varied; U-46619 was the strongest in activating RhoA among the agonists (Fig. 4). This result is consistent with a previous report (12) showing that U-46619, which was a relatively weak agonist in terms of Ca2+ mobilization, was the most effective in inducing Ca2+ sensitization of MLC20 phosphorylation and contraction in α-toxin-permeabilized vascular smooth muscle. In contrast to U-46619 and several other agonists, ANG II at the optimal dose failed to stimulate RhoA (Fig. 4). It is of note that those agonists that induced RhoA activation elicited sustained contractile responses, whereas ANG II induced only a transient contraction (37). It has also been demonstrated that ANG II induces a transient increase in MLC20 phosphorylation (37). It is therefore suggested that the inability of ANG II to activate RhoA accounts, at least in part, for the transient nature of ANG II-induced MLC20 phosphorylation and contraction. Thus agonist-induced RhoA activation appears to be receptor type- specific in vascular smooth muscle. This observation is consistent with recent findings in non-smooth muscle cells showing that many of the heptahelical receptors that couple to Gq/11, including vasopressin V1, bradykinin B2, muscarinic M1, endothelin ETA, and thrombin receptors, are linked to stress fiber formation, a hallmark of cellular RhoA activation, whereas others, including ETB and angiotensin AT1 receptors, fail to do so (7, 8). These studies in non-smooth muscle cells also demonstrated that G12 and G13 are the major heterotrimeric G proteins that transduce receptor-derived signals into RhoA activation and stress fiber formation (8). Therefore, it is suggested that an angiotensin receptor, most likely AT1, in vascular smooth muscle is not coupled to G12 or G13 as efficiently as the thromboxane A2 receptor, α-adrenergic receptor, histamine H1 receptor, and 5-HT receptor. While this manuscript was being prepared after the experiments had been finished, it was demonstrated that ANG II receptors were not coupled to G12or G13 as efficiently as vasopressin or endothelin receptors in cultured vascular smooth muscle cells (9).
Tyrosine kinases have been implicated in receptor-mediated stress fiber formation in non-muscle cells (7, 8, 16, 30). Because the effect of a constitutively active RhoA mutant was not affected by a tyrosine kinase inhibitor in these cells, the tyrosine kinase inhibitor-sensitive site was suggested to be involved in the process of receptor agonist-induced RhoA activation (16). Interestingly, it was previously demonstrated in intact smooth muscle that a tyrosine kinase inhibitor attenuated agonist-induced contraction with a reduction in MLC20 phosphorylation (1,15). In permeabilized smooth muscle, genistein inhibited excitatory agonist-induced enhancement of Ca2+-induced contraction, suggesting the involvement of a tyrosine kinase in the Ca2+ sensitization (34). Despite this, direct effects of tyrosine kinase inhibitors on RhoA activation have not yet been demonstrated in smooth muscle or non-smooth muscle cells. In the present study, we demonstrated for the first time that tyrosine kinase(s) are involved in the regulation of RhoA activity. Genistein, a nonspecific tyrosine kinase inhibitor, strongly inhibited U-46619-induced RhoA activation and, to a lesser extent, that induced by other agonists (Fig. 5). In contrast, a specific EGF receptor kinase inhibitor, AG-1478, had no effect on agonist-induced RhoA activation. The latter observation was discordant with previous observations (8) that AG-1478 effectively inhibited stress fiber formation induced by several different receptor agonists in non-muscle cells. Thus molecular species of tyrosine kinases that are responsible for receptor agonist-induced RhoA activation appear to be different between vascular smooth muscle and non-muscle cells. We observed that ET-1-induced RhoA activation was totally insensitive to genistein, unlike the other agonists (Fig. 5). It was previously demonstrated in non-muscle cells that G13- but not G12-mediated stress fiber formation (8) and neurite retraction (16) were sensitive to tyrosine kinase inhibitors. Therefore, it is an interesting possibility that thromboxane A2 and endothelin receptors in smooth muscle are coupled via different heterotrimeric G proteins to RhoA activation, with different sensitivities to genistein. In this regard, recent studies (2, 7) have demonstrated in non-muscle cells that thromboxane A2 receptor is coupled to G13, whereas the endothelin ETA receptor, which is a dominant isoform expressed in vascular smooth muscle, is coupled to G12.
The protein kinase C activator phorbol ester was previously demonstrated to potentiate Ca2+-induced MLC20phosphorylation and contraction in intact and permeabilized smooth muscle (14, 33). However, we failed to detect an increase in GTP-RhoA during a sustained contraction induced by PDBu (Fig. 4). Our results suggest that phorbol ester sensitizes MLC20phosphorylation and contraction to Ca2+ through a mechanism independent of RhoA. The myosin phosphatase inhibitor protein CPI-17 is an interesting protein kinase C target protein in smooth muscle (3). A very recent study (18) demonstrated that excitatory agonists induce phosphorylation of CPI-17 in a manner partially sensitive to both specific protein kinase C and Rho kinase inhibitors.
In conclusion, the present study demonstrated that excitatory agonists activate RhoA in smooth muscle. The magnitude and mode of agonist-induced RhoA activation does not appear to be uniform among agonists. The results provide further evidence for the role of the RhoA signaling pathway in excitatory agonist-induced sensitization of MLC20 phosphorylation to Ca2+, which acts in concert with phospholipase C signaling to result in a full contractile response.
We thank Nobuko Yamaguchi and Yasuhiro Hiratsuka for preparing the manuscript and for technical assistance, respectively.
This work was supported by grants from the Ministry of Education, Science and Culture of Japan, the Japan Society for the Promotion of Science Research for the Future Program, and the Hoh-Ansha Foundation.
Address for reprint requests and other correspondence: Y. Takuwa, Dept. of Physiology, Kanazawa Univ. School of Medicine, 13-1 Takara-machi, Kanazawa, Ishikawa 920-8640, Japan (E-mail:).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2001 the American Physiological Society