(Received for publication, March 4, 1997)
From the First Department of Internal Medicine, Mie University
School of Medicine, Tsu, Mie 514, Japan, the First
Department of Physiology, Yamaguchi University School of Medicine, Ube,
Yamaguchi 755, Japan, the ¶ Division of Signal Transduction, Nara
Institute of Science and Technology, Ikoma, Nara 630-01, Japan, and the
§ Division of Molecular Cardiology, Research Institute of
Angiocardiology, Faculty of Medicine, Kyushu University,
Fukuoka 812, Japan
Small GTPase Rho plays pivotal roles in the Ca2+ sensitization of smooth muscle. However, the GTP-bound active form of Rho failed to exert Ca2+-sensitizing effects in extensively Triton X-100-permeabilized smooth muscle preparations, due to the loss of the important diffusible cofactor (Gong, M. C., Iizuka, K., Nixon, G., Browne, J. P., Hall, A., Eccleston, J. F., Sugai, M., Kobayashi, S., Somlyo, A. V., and Somlyo, A. P. (1996) Proc. Natl. Acad. Sci. U. S. A. 93, 1340-1345). Here we demonstrate the contractile effects of Rho-associated kinase (Rho-kinase), recently identified as a putative target of Rho, on the Triton X-100-permeabilized smooth muscle of rabbit portal vein. Introduction of the constitutively active form of Rho-kinase into the cytosol of Triton X-100-permeabilized smooth muscle provoked a contraction and a proportional increase in levels of monophosphorylation of myosin light chain in both the presence and the absence of cytosolic Ca2+. These effects of constitutively active Rho-kinase were wortmannin (a potent myosin light chain kinase inhibitor)-insensitive. Immunoblot analysis revealed that the amount of native Rho-kinase was markedly lower in Triton X-100-permeabilized tissue than in intact tissue. Our results demonstrate that Rho-kinase directly modulates smooth muscle contraction through myosin light chain phosphorylation, independently of the Ca2+-calmodulin-dependent myosin light chain kinase pathway.
Smooth muscle contraction is primarily regulated by the levels of
phosphorylation of myosin light chain
(MLC),1 which has heretofore been
considered to be governed by a Ca2+-calmodulin
(CaM)-dependent MLC kinase pathway (1-4). However, as the use of
Ca2+ indicator revealed that the force/Ca2+
ratio is variable, the Ca2+-CaM-dependent MLC
kinase pathway cannot solely account for the mechanisms of agonist- or
GTPS-induced increases in the force/Ca2+ ratio,
so-called Ca2+ sensitization (1, 5-9). Thus, an additional
mechanism that can regulate Ca2+ sensitization of smooth
muscle has been proposed. Using membrane permeabilization of smooth
muscle, the possibility that monomeric Ras family G-proteins, such as
Rho, contribute to Ca2+ sensitization of smooth muscle was
demonstrated (10-12). Direct activation of G-proteins by the
application of GTP
S (8, 9), agonists (1, 5-8), and GTP-activated
Rho (10-12) could exert Ca2+-sensitizing effects on
saponin- or
-escin-permeabilized smooth muscle. However, the
activated Rho failed to induce Ca2+ sensitization of
extensively Triton X-100-permeabilized smooth muscle (11). Considering
that extensive Triton X-100-permeabilization allows higher molecular
weight compounds to diffuse from the cytosol of smooth muscle of the
rabbit portal vein (13), important diffusible factor(s) for the
Ca2+ sensitization of smooth muscle might be lost during
extensive permeabilization by Triton X-100, an event that would result
in no response to activated Rho.
We have recently reported that Rho-kinase, which is activated by GTP-bound active form of Rho (14-16), phosphorylates not only MLC, thereby activating myosin ATPase (17), but also myosin phosphatase, thus inactivating it in vitro (18). These findings in a cell-free system, plus the previous reports of G-protein-mediating Ca2+ sensitization as described above, suggest that Rho-kinase may induce contraction and concomitant MLC phosphorylation of the smooth muscle. We examined the effects of the constitutively active form of Rho-kinase on smooth muscle extensively permeabilized by Triton X-100 and attempted to determine if Rho-kinase would be the factor lost during extensive Triton X-100 permeabilization.
The catalytic subunit of recombinant Rho-kinase (CAT; molecular mass is about 80 kDa) was expressed as a glutathione S-transferase fusion protein and purified using a baculovirus system and a glutathione-Sepharose column (17). The kinase activity of the elute was determined by phosphorylation assay using S6 peptide as a substrate (15) in buffer containing 50 mM Tris-HCl, pH 7.5, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, and 100 µM ATP (0.5-20 Gbq/mmol). CaM was purified from bovine brain by previously described method (19). Other materials and chemicals were obtained from commercial sources.
Force Measurement of the Triton X-100-permeabilized Smooth Muscle of the Rabbit Portal VeinSmall strips of male rabbit (2-2.5 kg) portal veins were manually dissected (50-100 µm wide and 0.5-1 mm long), connected to an isometric force transducer (UL-2GR, Minebea, Japan), and mounted in a well (200 µl) on a bubble plate (6). After recording contractions evoked by 118 mM K+, the strips were incubated in relaxing solution, followed by 0.5% Triton X-100 for 20 min at 25 °C. The solutions have been described in detail elsewhere (6). CaM (0.5 µM) was added to all reactive solutions for experiments with chemical permeabilization.
Immunoblot Analysis of Rho-kinase and MLC0.5% Triton X-100-permeabilized and intact rabbit portal veins were homogenized in extraction buffer containing 50 mM Tris-HCl, pH 7.2, 400 mM NaCl, 2 mM EGTA, 1 mM EDTA, 1 mM DTT, 0.1 µM p-amidinophenylmethanesulfonyl fluoride hydrochloride, 10 µg/ml leupeptin, and 1 mM benzamidine. Each extract was centrifuged at 100,000 × g for 30 min at 4 °C, and the supernatant was subjected to SDS-polyacrylamide gel electrophoresis (20) followed by immunoblotting (21). Anti-Rho-kinase rabbit polyclonal antibodies were generated against the glutathione S-transferase fusion-catalytic domain of Rho-kinase (15), and anti-MLC rabbit polyclonal antibodies were provided by Dr. J. T. Stull (University of Texas Southwestern Medical Center, Dallas, TX). Immunostained proteins were visualized by Supersignal (Pierce). Densities of bands of immunostained proteins were quantitated using a scanning densitometer, Densitograph (Atto, Tokyo, Japan).
Measurement of the Extent of MLC PhosphorylationAfter treatment with 5.1 µg/ml (0.06 µM) CAT and/or 10 µM wortmannin, the fringe-like strips of the rabbit portal veins permeabilized by Triton X-100 were quickly placed in a frozen slurry of acetone containing 10% trichloroacetic acid and 10 mM DTT to terminate the contractile responses. After depletion of trichloroacetic acid, the strips were homogenized in urea sample buffer containing 20 mM Tris base, 22 mM glycine, pH 8.6, 8 M urea, 10 mM DTT, 10% sucrose, and 0.1% bromphenol blue. The extracts were subjected to glycerol-urea polyacrylamide gel electrophoresis following immunoblotting using anti-MLC antibodies as documented (22). Immunostained proteins were visualized colorimetrically with 4-chloro-1-naphthol and subjected to densitometrical quantitation.
Introduction of CAT, which is not only the constitutively active
form of Rho-kinase but also the highly homologous domain among rat
(14), bovine (15), human (16), and mouse Rho-associated kinases (23),
into the cytosol of the extensively Triton X-100-permeabilized rabbit
portal vein smooth muscle provoked a contraction both at a constant
cytosolic Ca2+ (pCa 6.5; Fig.
1a) and at a nominally zero cytosolic
Ca2+ buffered with 10 mM EGTA (pCa
8.0; Fig. 1c). CAT exerted contraction, whereas the
vehicle had no effect on the force (Fig. 1, a and c). These contractions were completely reversed by wash out
of CAT, contrary to those induced by 10 µM microcystin-LR
(24, 25) (Fig. 1b). In the absence of cytosolic
Ca2+ at pCa
8.0, the CAT-induced
contraction was also reversible (data not shown). In neither intact nor
-toxin-permeabilized strips of the portal vein did CAT exert the
contractile effects (data not shown). These observations are
interpreted to mean that constitutively active CAT could be introduced
into the cytosol of the smooth muscle only by extensive membrane
permeabilization to induce a reversible contraction.
MLC phosphorylation mediated by Ca2+-CaM-dependent MLC
kinase pathway plays a primary role in smooth muscle contraction
through myosin-actin-interaction and the consequent activation of
myosin ATPase (2-4). To investigate involvement of
Ca2+-CaM-dependent MLC kinase pathway in the CAT-induced
contraction, we examined the effects of wortmannin (WM), a potent MLC
kinase inhibitor (26) on force development induced by cumulative
application of CAT (Fig. 2). In the presence of
cytosolic Ca2+ at pCa 6.5, in which
Ca2+-CaM-dependent MLC kinase should be active,
10 µM WM shifted the dose-response curve down and to the
right. In the absence of cytosolic Ca2+ at pCa
8.0, in which MLC kinase would be hardly activated, 10 µM WM did not affect CAT-induced force development. In
the absence of CAT, treatment of 10 µM WM completely
inhibited the cytosolic Ca2+-provoked contraction at
pCa 6.5, in the Triton X-100-permeabilized fibers.
Considering our finding that WM did not affect the activity of CAT up
to 100 µM in vitro (data not shown), this
WM-sensitive component of CAT-induced contraction at pCa 6.5 seemed to be due to inhibition of the Ca2+-provoked
contraction through the Ca2+-CaM-dependent MLC kinase
pathway but not related to the CAT-mediated pathway. All these
observations suggest that the CAT-induced contraction of smooth muscle
of rabbit portal vein permeabilized by Triton X-100 is modulated
independently by the Ca2+-CaM-dependent MLC kinase
pathway.
To clarify whether CAT induces contraction with a concomitant increase
in levels of MLC phosphorylation, we examined the effects of CAT on MLC
phosphorylation, using immunoblotting with anti-MLC polyclonal antibody
(Fig. 3). As shown in lanes 1-3 of Fig.
3a, at pCa 8.0, monophosphorylation of MLC
was detected only in the presence of CAT and was insensitive to 10 µM WM (42.77 ± 9.22% of the total amount of
immunostained MLC (n = 4) in the absence of WM,
35.95 ± 3.39% (n = 4; p > 0.05)
in the presence of WM, respectively). At pCa 6.5, shown in
lanes 4-6 of Fig. 3a, CAT potentiated the level
of monophosphorylation of MLC (60.33 ± 1.42%, n = 4), which was partially inhibited by 10 µM WM
(26.05 ± 5.18%, n = 4, p < 0.01). Based on the statistical analysis and the results in Fig. 2,
this WM-sensitive component of CAT-induced MLC phosphorylation at
pCa 6.5 also seemed to be due to inhibition of
Ca2+-CaM-dependent MLC kinase activity. These
results are consistent with that of counterparts of the effects of CAT
on the contractile responses (Fig. 3b). It was concluded
that CAT potentiates the contractile response by increasing the extent
of monophosphorylation of MLC.
To determine if native Rho-kinase is one of the cofactors diffusible
during permeabilization by Triton X-100, we examined the amounts of
native Rho-kinase in intact and permeabilized fibers by immunoblot
analysis using rabbit polyclonal antibodies against Rho-kinase. To
standardize the densitometrical value, the ratio of densitometrical
quantification of immunostaining of Rho-kinase to that of MLC was
calculated in both intact and permeabilized fibers. As shown in Fig.
4, the amounts of native Rho-kinase in the Triton
X-100-permeabilized rabbit portal vein were markedly lower than those
in intact tissue (0.06 ± 0.01 (n = 4) for
permeabilized sample, 0.95 ± 0.02 (n = 4) for
intact sample, respectively), whereas the amounts of the possible
cytoskeletal proteins, such as MLC and myosin heavy chain in
permeabilized fibers were similar to the counterparts of intact fibers.
These results confirm that extensive permeabilization by Triton X-100
allows for the loss of cytosolic proteins, including Rho-kinase,
whereas cytoskeletal proteins such as myosin are stable. Based on all
of these findings taken together plus evidence that the direct
activation of G-proteins did not exert contractile effects on the
extensively Triton X-100-permeabilized smooth muscle (11), we consider
that Rho-kinase may be a valid candidate for the key molecule in
G-protein-mediating smooth muscle contraction and may be the molecule
lost during extensive permeabilization by Triton X-100.
We demonstrate here what seems to be the first evidence that Rho-kinase is a direct effector on the contractile apparatus of smooth muscle, independently of the Ca2+-CaM-dependent MLC kinase pathway. Except for Ca2+-independent MLC kinase (9, 13), we find no documentation that the exogenous addition of kinases to the cytosol of permeabilized smooth muscle directly exerts contractile responses comparable with findings with CAT. Because the inhibition of myosin phosphatase may possibly be the main mechanism of the G-protein-mediating Ca2+ sensitization of smooth muscle contraction (1, 9, 24, 25, 27), the CAT-induced contraction of G-protein-uncoupled smooth muscle permeabilized by Triton X-100 (Fig. 1) may be also mediated by the inhibition of myosin phosphatase. This notion is supported by our previous finding that Rho-kinase inhibited the activity of myosin phosphatase through thiophosphorylation of its myosin-binding subunit in vitro (18). However, at cytosolic zero Ca2+, microcystin-LR-induced contraction of the permeabilized smooth muscle was reduced by an MLC kinase inhibitor (25), whereas the CAT-induced contraction was insensitive to it (Fig. 2). Such differential sensitivities of the MLC kinase inhibitor to CAT- and myosin phosphatase inhibitior-induced contractions support the idea that myosin phosphatase inhibition alone cannot account for the CAT-induced contraction at the cytosolic zero Ca2+. Taking this together with our report that Rho-kinase directly provokes the phosphorylation of MLC and activates myosin in vitro (17), we suggest that the mechanism(s) of CAT-induced contraction of Triton X-100-permeabilized rabbit portal vein might be a concomitant monophosphorylation of MLC directly induced by CAT independently of a Ca2+-CaM-dependent MLC kinase pathway. We propose that Rho-kinase is considered a valid key molecule in G-protein-mediating Ca2+ sensitization of smooth muscle contraction.
We thank Dr. J. T. Stull for providing the anti-myosin antibody, Drs. D. J. Hartshorne and M. P. Walsh for helpful discussions, and M. Ohara for critical comments on the manuscript.