Histamine-induced Vasoconstriction Involves Phosphorylation of a Specific Inhibitor Protein for Myosin Phosphatase by Protein Kinase C α and δ Isoforms

Histamine stimulus triggers inhibition of myosin phosphatase-enhanced phosphorylation of myosin and contraction of vascular smooth muscle. In response to histamine stimulation of intact femoral artery, a smooth muscle-specific protein called CPI-17 (for protein kinase C-potentiatedinhibitory protein for heterotrimeric myosin light chain phosphatase of 17 kDa) is phosphorylated and converted to a potent inhibitor for myosin phosphatase. Phosphorylation of CPI-17 is diminished by pretreatment with either Y27632 or GF109203x, suggesting involvement of multiple kinases (Kitazawa, T., Eto, M., Woodsome, T. P., and Brautigan, D. L. (2000) J. Biol. Chem. 275, 9897–9900). Here we purified and identified CPI-17 kinases endogenous to pig artery that phosphorylate CPI-17. DEAE-Toyopearl column chromatography of aorta extracts separated two CPI-17 kinases. One kinase was protein kinase C (PKC) α, and the second kinase was purified to homogeneity as a 45-kDa protein, and identified by sequencing as PKCδ. Purified PKCδ was 3-fold more reactive with CPI-17 compared with myelin basic protein, whereas purified PKCα and recombinant RhoA-activated kinases (Rho-associated coiled-coil forming protein Ser/Thr kinase and protein kinase N) showed equal activity with CPI-17 and myelin basic protein. Y27632 inhibited CPI-17 phosphorylation by purified PKCδ with IC50 of 0.6 μm (in the presence of 0.1 mm ATP) or 14 μm (2.0 mm ATP). Y27632 significantly suppressed CPI-17 phosphorylation in smooth muscle cells, and the contraction of permeabilized rabbit femoral artery induced by stimulation with phorbol ester. GF109203x inhibited phorbol ester-induced contraction of rabbit femoral artery by 80%, whereas a PKCα/β inhibitor, Go6976, reduced contraction by 47%. The results imply that histamine stimulation elicits contraction of vascular smooth muscle through activation of PKCα and especially PKCδ to phosphorylate CPI-17.

In vascular smooth muscle, elevation of cytosolic Ca 2ϩ con-centration leads to activation of Ca 2ϩ /calmodulin-dependent myosin light chain kinase and phosphorylation of smooth muscle myosin (1). Studies using smooth muscle strips permeabilized with ␣-toxin, ␤-escin, and other detergents revealed that myosin phosphorylation is enhanced by stimulus with various agonists, such as histamine, or by treatment with drugs such as phorbol ester, in the presence of constant submaximal Ca 2ϩ concentration (2)(3)(4). These findings indicate that myosin phosphatase activity is suppressed in response to various stimuli to produce Ca 2ϩ sensitization of smooth muscle contractility (reviewed in Ref. 5). Myosin phosphatase (MLCP) 1 is a heterotrimeric holoenzyme, consisting of a 110-kDa myosin targeting subunit (MYPT1), a 21-kDa noncatalytic subunit (M21), and a ␦ isoform of PP1 catalytic subunit (PP1C␦) (6).
Rho-activated kinase (ROCK) and protein kinase C (PKC) have been proposed to mediate the inhibition of MLCP in response to various agonists (5). Pretreatment of ␣-toxin-permeabilized portal vein strips with GTP␥S and ATP␥S reduced MLCP activity in parallel with thiophosphorylation of MYPT1 (7). This focused attention on GTP-activated kinase pathways, i.e. ROCK. The inhibitory site (Thr 695 in chicken MYPT1) is also phosphorylated by a MLCP-associated kinase similar to ZIP kinase, called ZIP-like kinase (8,9). ROCK also phosphorylates Thr 695 of MYPT1 and induces inhibition of MLCP activity (10 -14). Active RhoA, a small G protein, associates on the coiled-coil region of ROCK to enhance kinase activity (reviewed in Ref. 15). Y27632, a pyridine derivative used as a specific ROCK inhibitor, inhibits both phenylepherine-and GTP␥S-induced Ca 2ϩ sensitization of smooth muscle (16,17), suggesting that ROCK-mediated inactivation of MLCP increases Ca 2ϩ sensitivity in response to agonists. Involvement of ROCK in inhibition of MLCP explains RhoA-dependent regulation of Ca 2ϩ sensitization of smooth muscle contraction. Dominant active RhoA elicits vasoconstriction of saponin-or ␤-escin-permeabilized rabbit mesenteric artery (18,19). Stimulation of smooth muscle with phenylepherine potentiates and translocates endogenous RhoA from cytosol to membrane, though localization of active ROCK is obscure (20,21). These results support the view that Rho-dependent mechanisms are the predominant pathway for Ca 2ϩ sensitization.
In contrast to ROCK, PKC does not phosphorylate MYPT1 directly. However activation of PKC by addition of phorbol ester induces inhibition of MLCP and contraction of vascular smooth muscles (22,23). An inhibitor protein specific for MLCP was isolated from pig aorta smooth muscle extracts and called CPI-17 for PKC-potentiated inhibitory protein for heterotrimeric myosin light chain phosphatase of 17 kDa (24). Expression of CPI-17 is highly restricted to smooth muscle tissues, and it is especially abundant in arterial smooth muscles. For example, CPI-17 is estimated to be 7 M in rabbit femoral artery (25,26). Phosphorylation of Thr 38 in CPI-17 converts it to a potent MLCP inhibitor with an IC 50 of ϳ5 nM (24,27). Phospho-CPI-17 enhances myosin phosphorylation and contraction of both permeabilized arterial smooth muscle and intact fibroblast (28,29). Unlike other well known type 1 phosphatase (PP1) inhibitor proteins, such as inhibitor1, DARPP32, and inhibitor2, which inhibit monomeric PP1C but not MLCP, CPI-17 can inhibit the trimeric form of MLCP, without dissociation of subunits (25). CPI-17 is a soluble and globular protein with molecular mass of 17 kDa (27). Permeabilization of femoral artery strips using ␤-escin or Triton X-100 depletes endogenous CPI-17 with loss of the contractile response to phorbol ester. The PKC-induced contraction of permeabilized artery is reconstituted by addition of recombinant CPI-17 (30). Furthermore, the expression pattern of CPI-17 among six different smooth muscle tissues correlates with their extent of PKC-induced contraction, implying that CPI-17 is key to the PKC-mediated Ca 2ϩ sensitization (26). Assays with purified kinases show that Thr 38 of CPI-17 can be phosphorylated by multiple kinases such as PKC, ROCK, PKN, and ZIP-like kinase (24,(31)(32)(33). Indeed, CPI-17 is phosphorylated in femoral artery strips in response to stimulus with phenylepherine, histamine, GTP␥S, and phorbol ester (34). Importantly, histamine-induced phosphorylation of endogenous CPI-17 and contraction of femoral artery are diminished by treatment with both GF109203x (a PKC inhibitor) and Y27632 (a ROCK inhibitor), suggesting that both PKC and ROCK are involved in CPI-17-induced vasoconstriction (34). Here, using recombinant CPI-17 as a substrate, we purified from pig aorta smooth muscle CPI-17 kinases that are sensitive to GF109203x and Y27632. PKC␣ and PKC␦ were purified from pig aorta smooth muscle using recombinant CPI-17 as a substrate. We measured CPI-17 kinase activity both in vitro and in smooth muscle cells using GF109203x and Y27632 as inhibitors, and found that both inhibitors reduce CPI-17 phosphorylation by PKC and contraction of arterial smooth muscle by phorbol ester.
Active catalytic fragment of PKN was transiently expressed in COS7 cells. The expression vector of PKN is designed to express C-terminal FLAG-tagged protein in mammalian cells (36). COS7 cells in a 150-mm dish are transiently transfected for 48 -72 h with 20 g of the expression vector. Cells were homogenized with 1 ml of 0.1 M NaCl, 1 mM EGTA, 0.8 mM Pefabloc (Roche Molecular Biochemicals), 5 g/ml leupeptin, 1 M microcystin LR, 0.1% 2-mercaptoethanol, and 50 mM MOPS-NaOH, pH 7.0, plus 1% IGEPAL CA-630 (Sigma). The supernatant was obtained by 10-min centrifugation at 20,000 ϫ g. FLAG-tagged PKN was absorbed onto anti-FLAG (M2)-conjugated agarose (Sigma) for 2 h at 4°C. After washing the beads three times with the lysis buffer and then twice with the buffer without IGEPAL CA-630, FLAG-tagged PKN was eluted with five volumes of 0.1 mg/ml FLAG peptide (Sigma) in the buffer without IGEPAL CA-630 and used for kinase assay immediately.
Protein Sequencing-Partial amino acid sequencing was performed as described previously (25). Briefly, final preparation of CPI-17 kinase (9.3 g) was subjected to SDS-PAGE. The Coomassie-stained 44-kDa band was excised, and the protein in the gel piece was digested with 0.2 g of lysylendopeptidase C (Wako Pure Chemical). Peptides were extracted with 0.1% trifluoroacetic acid and separated by reverse phase high performance liquid chromatography column (Tosoh; ODS 80Ts, 4.6 ϫ 200 mm). Amino acid sequences of isolated peptides were analyzed using a PerkinElmer Applied Biosystems model 492 protein sequencer in the core facility at the Hokkaido University.
Cell Culture-COS7 cells were cultured in Dulbecco's modified Eagle's medium (DMEM; Life Technologies, Inc.) supplemented with 10% newborn calf serum (Life Technologies, Inc.). Rat aorta smooth muscle cells were kindly provided by Dr. Gary Owens (University of Virginia) and cultured in DMEM/F-12 (1:1) with 10% fetal bovine serum (Life Technologies, Inc.). Smooth muscle cells were grown in DMEM/F-12 without serum for 24 h to achieve quiescence. Ectopic DNAs were transiently transfected as described previously (29).
Kinase Assay-Phosphorylation was assayed using either incorporation of 32 P from [␥-32 P]ATP (Figs. 1-3, and Table I) or by immunoblotting with anti-P-CPI-17 antibody (Figs. 4 and 5). Both assays were carried out at 30°C using 0.2 mg/ml substrate proteins and 0.1 or 2 mM ATP, in the presence of 10 mM Mg(OAc) 2 , 1 mM dithiothreitol, 1 M microcystin LR, 50 mM sodium ␤-glycerophosphate, 0.8 mM Pefabloc, and 25 mM MOPS-NaOH, pH 7.0. Reactions were initiated by addition of kinase preparations. 32 P-Labeled CPI-17 and myelin basic protein were absorbed onto a piece of phosphocellulose paper (P-81; Whatman) and free radioactive ATP was removed by repeated washing with 75 mM o-phosphoric acid. Extent of phosphorylation was calculated from radioactivity on phosphocellulose paper by Cerenkov counting. Phospho-rylation of CPI-17 with non-radioactive ATP was detected by immunoblotting as described previously (34).
Phosphorylation of CPI-17 in rat aorta smooth muscle cells were assayed by immunoblotting methods using anti-P-CPI-17 antibody. After serum starvation, 10 5 cells were stimulated with PMA. Kinase inhibitors were added in the medium at 30 min before the stimulation. The stimulation was terminated by addition of 10% trichloroacetic acid. Insoluble proteins were scraped and collected by centrifugation. The protein pellet was washed three times with acetone and then dried. The proteins were dissolved in buffer including 1% SDS and subjected to immunoblotting. The band intensity of phosphorylated CPI-17 on the x-ray film was quantified by use of densitometer and analyzing software (Molecular Dynamics).
Other Procedures-Preparation of rabbit femoral artery strips and force measurement were performed as described previously (34). Protein concentration was determined by the improved method of Bradford (38), using bovine serum albumin as a standard. SDS-PAGE was carried out using methods of Porzio and Pearson or Laemmli (39,40). Immunoblotting was performed as described previously (29).

Purification and Characterization of Dominant CPI-17 Kinases in Pig Aorta Smooth
Muscle-A soluble extract of pig aorta smooth muscle was subjected to DEAE-Toyopearl column chromatography, and CPI-17 kinase activity in each fraction was assayed using recombinant CPI-17 as a substrate (Fig. 1). Ca 2ϩ /phosphatidylserine/PMA-dependent CPI-17 kinase activ-ity (called P1) was detected in breakthrough fractions (Fig. 1,  closed circles). A spontaneously active CPI-17 kinase activity (called P2) was eluted at about 0.15 M NaCl (Fig. 1, open  circles), that corresponds to the endogenous CPI-17 kinase described previously (24). Minor activities were detected in fractions eluted with [NaCl] ϭ 0.08 -0.25 M, around P2. Overall P1 and P2 activity were 40% and 60% of the total CPI-17 kinase, respectively. The CPI-17 kinase activity from P1 was subjected to sequential chromatographies using phenyl-Sepharose and hydroxylapatite columns. The CPI-17 kinase activity from P1 was dependent on both Ca 2ϩ and phospholipid; the activities (in pmol/min/l) were 0.18 Ϯ 0.06 (with 1 mM EGTA as control), 0.23 Ϯ 0.05 (with 1 mM EGTA and 50 g/ml phosphatidylserine), and 2.5 Ϯ 0.5 (with 0.5 mM CaCl 2 and 50 g/ml phosphatidylserine). An 80-kDa protein was detected by immunoblotting using an ␣ isoform-specific antibody (data not shown). The results indicate that PKC␣ is one of the dominant CPI-17 kinases in pig aortic extracts.
The purification of CPI-17 kinase in P2 fraction is summarized in Table I. An ammonium sulfate fractionation after extraction removed actomyosin which interferes with column chromatography. The spontaneously active CPI-17 kinase was monitored by kinase assay in the presence of 1 mM EGTA. Overall, 50 g of the CPI-17 kinase was isolated from 500 g of pig aorta (30 pieces), representing over 8,000-fold higher specific activity. Protamine-Sepharose chromatography was a critical step and produced 150-fold purification, with a CPI-17 kinase activity in fractions 24 -28 ( Fig. 2A, closed circle). These fractions contained a 44-kDa polypeptide detected by silver staining after SDS-PAGE (Fig. 2B). The 44-kDa protein phosphorylated recombinant CPI-17 in an "in-gel" kinase assay (Fig. 2, A (triangles) and C), indicating that this 44-kDa protein was a spontaneously active CPI-17 kinase. Lysylendopeptidase fragments of the 44-kDa kinase were sequenced by Edman degradation, and all amino acid sequences determined are shown in upper lines in Table II. A total of 84 residues were sequenced, and all peptide sequences matched to human PKC␦ sequence (GenBank NM_006254) with 85% identity. The results show that the 44-kDa CPI-17 kinase isolated from P2 is an active fragment of pig PKC␦.
This PKC␦ purified from pig aorta phosphorylated recombinant CPI-17 3-fold more than myelin basic protein as substrate (Fig. 3). By comparison, under the same conditions, PKC␣ phosphorylated myelin basic protein twice as much as CPI-17 (Fig. 3). ROCK and PKN Rho-activated kinases have been reported to phosphorylate CPI-17 (31,32). The activity of recombinant ROCK and PKN against CPI-17 was 50% and 100% of myelin basic protein kinase activities, respectively (Fig. 3), although neither ROCK nor PKN was detected as a CPI-17 kinase in pig aorta extracts (Fig. 1).
Inhibition of PKC␦ by Y27632-Histamine stimulation enhances phosphorylation of CPI-17 in intact rabbit femoral artery, and this response to histamine was inhibited by preincubation with 3 M GF109203x or 10 M Y27632 (34). Those results led us to conclude that both PKC and ROCK might be involved in phosphorylation of CPI-17 induced by histamine (34). We tested the inhibition of PKC␣ and PKC␦ purified from pig aorta by GF109203x and Y27632 (Fig. 4). In the presence of a physiological ATP concentration (2 mM), 10 M Y27632 suppressed the CPI-17 kinase activity of PKC␦ by 40% (Fig. 4A). Y27632 inhibited the activity of PKC␦ with an IC 50 of 14 M, with 50 M Y27632 fully blocking the kinase activity (Fig. 4B). Compared with PKC␦, ROCK was slightly more sensitive to Y27632, with IC 50 of 6 M, which is consistent with the IC 50 value of 3 M calculated from K i value (16). In the presence of 0.1 mM ATP, Y27632 potently inhibited PKC␦ activity, with IC 50 of 0.6 M, and complete inhibition at 10 M (Fig. 4B). The CPI-17 kinase activity of PKC␣ was completely inhibited with 3 M GF109203x (Fig. 4A). At this concentration GF109203x inhibited PKC␦ over 80%. Ro 31-8220, another inhibitor that blocks multiple PKC isoforms, fully inhibited CPI-17 kinase activities of both PKC␣ and PKC␦. In these assays ROCK was resistant to GF109203x or Ro 31-8220. Therefore inhibition of the purified PKC␦ by Y27632 was not due to activity of ROCK in the PKC␦ preparation. Even though the catalytic domain of PKN resembles the PKC families, GF109203x had essentially no effect on its CPI-17 kinase activity.

Inhibition of PMA-induced Phosphorylation of CPI-17 by Y27632 in Rat Aorta Smooth
Muscle Cells-Phosphorylation of endogenous CPI-17 in cultured rat aorta smooth muscle cells was measured by immunoblotting methods using anti-P-CPI-17 antibody (Fig. 5). After stimulation for 5 min with 0.1 M PMA, phosphorylation of CPI-17 was enhanced 5-fold compared with quiescent conditions. The phosphorylation of CPI-17 was decreased in a dose-dependent fashion by 30 min of preincubation with Y27632. GF109203x was a more effective inhibitor than Y27632, and reduced the CPI-17 phosphorylation to 40%. Ro 31-8220 potently inhibited the phosphorylation of CPI-17 to below basal level, indicating that PKC isoforms activated by PMA are responsible for essentially all the CPI-17 phosphorylation in smooth muscle cells. Ro 31-8220 is relatively more potent for PKC␦/⑀ isoforms compared with GF109203x (41,42). The different potency between GF109203x and Ro 31-8220 against CPI-17 phosphorylation suggests the involvement of at least two isoforms of PKC. The results also show that Y27632 inhibits the CPI-17 kinase activity of PKC in living cells.    3. Phosphorylation of CPI-17 and myelin basic protein by purified kinases. The purified pig aorta PKC␣ and PKC␦ and recombinant human ROCK2 and PKN active fragments were used in kinase assays using recombinant CPI-17 (0.2 mg/ml) and myelin basic protein (0.2 mg/ml) as substrate. Relative activity with CPI-17 was calculated and plotted. Activity with myelin basic protein was assigned as 100%, and triplicate assays were performed and mean values Ϯ S.E. shown.

Inhibition of PKC-induced Contraction of Smooth Muscle by Y27632, GF109203x
, and Go6976 -We tested effects of kinase inhibitors on PDBu-induced contraction of ␣-toxin-permeabilized rabbit femoral artery that retains endogenous CPI-17. Fig. 6A shows force traces of ␣-toxin-permeabilized rabbit femoral artery stimulated with 0.1 M PDBu. Stimulation with 0.1 M PDBu provoked contraction (57% of maximum at pCa 4.5) of ␣-toxin permeabilized rabbit femoral artery at constant Ca 2ϩ concentration (pCa ϭ 6.7) (Fig. 6A, left). Preincubation for 30 min with 10 M Y27632 suppressed both the initial rate and maximum level of contraction induced with PDBu (Fig. 6A,  right). PDBu-induced contraction was reduced to 65% with Y27632, compared with control. Consistent with these results, 10 M Y27632 reduced histamine-induced contraction of rabbit femoral artery to 60% (34). Thus, Y27632 equivalently inhibited both histamine-and PDBu-induced contraction. The inhibition of PDBu-induced contraction by 10 M Y27632 diminished with increasing PDBu concentrations from 0.01 to 3.0 M (Fig. 6B). Fu et al. (17) also reported that 10 M Y27632 does not suppress the contraction of rabbit pulmonary artery induced by addition of 20 M PDBu. The results suggest that contraction induced with low concentrations of phorbol ester is sensitive to Y27632, similar to vasoconstriction in response to histamine stimulation.
Besides inhibition by Y27632, the contraction induced by PDBu stimulation was diminished over 80% by 30-min preincubation with 3 M GF109203x (Fig. 7A). In contrast to GF109203x, preincubation with 30 M Go6976, a PKC␣/␤ isoform-specific inhibitor (42), reduced PDBu-induced contraction of femoral artery by only 47% (Fig. 7B). The results show that PKC␣/␤ isoform plus PKC isoform(s) insensitive to 30 M Go6976 and sensitive to 3 M GF109203x, such as PKC␦/⑀,  (22,23). Involvement of CPI-17 in this PKC-mediated inhibition of MLCP was suggested by permeabilization using ␤-escin and Triton X-100 that diminished PDBu-induced contraction of femoral artery and depleted endogenous CPI-17. Reconstitution of CPI-17 into the skinned muscle restored the response to PDBu (30). Further, the expression level of CPI-17 correlates with PKC-induced contraction in various smooth muscle tissues (26). However, other kinases such as ROCK, PKN, and ZIP-like kinase are reported to phosphorylate CPI-17 (31)(32)(33). Thus, the question remained: which kinases phosphorylate CPI-17 in vascular smooth muscle? Here we purified the major CPI-17 kinases from pig aorta smooth muscle and identified them as PKC␣ and PKC␦. The PKC␦ was purified as an active fragment. Another peak of kinase that retained PMA-dependent activity eluted after the 44-kDa PKC␦, and it might be the intact PKC␦ (Fig. 1). In rabbit femoral artery, PDBu-induced contraction is potently inhibited by treatment with GF109203x, a general PKC inhibitor, and is partially suppressed with Go6976, a specific PKC␣/␤ inhibitor (42,43). The PKC isoform composition is varied among tissues and animals (reviewed in Refs. 44 and 45). For example, in rat mesenteric arterial smooth muscle and in contractile smooth muscle cells from rabbit aorta and human renal artery, PKC␦ is predominantly expressed compared with PKC⑀ (46,47). On the other hand, in ferret aorta smooth muscle, PKC␣ and PKC⑀ are expressed, but expression level of PKC␦ is below detection limit of immunoblotting (48). Indeed, PDBu-induced contraction of rabbit femoral artery is insensitive to rottlerin, a selective PKC␦ inhibitor (49) (data not shown), whereas PDBu-induced CPI-17 phosphorylation in permeabilized femoral artery is insensitive to Ca 2ϩ concentration (34). Thus, various isoforms of PKC might phosphorylate CPI-17 in different tissues of animals.
Inhibition of PKC␦ by Y27632-A pyridine derivative, Y27632, has been widely used for studies of the RhoA/ROCK pathway as a ROCK specific inhibitor (16,50). Inhibition by Y27632 is reported to be specific to ROCK with inhibitory constant values (K i ) with 0.14 -0.22 M, which are Ͼ100-fold lower compared with PKA, Ca 2ϩ -dependent PKC, and MLCK (16,50). Agonist-and GTP␥S-induced Ca 2ϩ sensitization were blocked by treatment with Y27632, suggesting involvement of ROCK in smooth muscle contraction (14,16). Recently, our results with Y27632 suggested involvement of RhoA/ROCK in CPI-17 phosphorylation in response to histamine stimulation (34). However, we found out that the PKC␦ purified from pig aorta smooth muscle was inhibited with 10 M Y27632. PKC⑀ also was reported to be more sensitive to Y27632 than Ca 2ϩdependent PKC (16). In the ATP binding pocket, the putative Y27632 binding site, only a couple of side chains differ between ROCK and PKC␦. This suggests that direct inhibition of PKC␦/⑀ by Y27632 is possible. It has been reported that Y27632 does not inhibit contraction of rabbit pulmonary artery induced by addition of 20 M PDBu (17). Presumably, other kinases and/or downstream kinases of PKC, such as protein kinase D (51), could phosphorylate CPI-17 in response to higher concentrations of phorbol ester, in an indirect manner. A bisindolylmaleimide derivative, GF109203x, potently inhibits PKC isoforms, but not ROCK. From these observations we conclude that GF109203x reveals the contribution of both PKC␣/␤ and PKC␦/⑀ but not ROCK, whereas Y27632 inhibits the activity of ROCK and PKC␦/⑀ in smooth muscle cells. Y35526, a derivative of Y27632, affinity-labeled ROCK1 of smooth muscle (16). However, a pyridine and a cyclohexane ring of Y27632 are replaced with a pyrrolopyridine and a benzene ring in Y35526, in addition to modification with an iodo-and an azido-group. Therefore, it is quite possible that these substitutions impaired affinity of Y35526 for PKC␦.

CPI-17-mediated Regulation of MLCP in Arterial Smooth
Muscle-Activation of PKC pathway in response to histamine stimulation has been documented well (reviewed in Ref. 52). Fig. 8 illustrates a proposed model of signaling from histamine H1 receptor (H1R) and regulation of MLCP mediated by CPI-17. H1R is coupled with the G␣ q/11 family of trimeric G-protein that mediates activation of phospholipase C (PLC). PLC action produces inositol trisphosphate (IP 3 ) and diacylglycerol (DAG) FIG. 7. Inhibition of PDBu-induced contraction of rabbit femoral artery by Go6976. Contraction of ␣-toxin-permeabilized rabbit femoral artery was recorded as described in Fig. 6. Maximum contraction was measured at pCa 4.5, and relative value was obtained at pCa 6.7. BLK shows contraction prior to PMA stimulation. After stimulation FIG. 8. Signaling pathways for histamine-induced vasoconstriction. Histamine binds to H1 receptor (H1-R) of artery smooth muscle and stimulates a heterotrimeric G protein including G␣ q/11 . G␣ q/11 activates PLC to release IP 3 and DAG from phospholipid. IP 3 induces Ca 2ϩ release from sarcoplasmic reticulum and activation of Ca 2ϩ /CaM-dependent MLCK. Released Ca 2ϩ and DAG activate PKC␣ and PKC␦, which phosphorylate Thr 38 of CPI-17. [Thr 38 ]Phospho-CPI-17 inhibits MLCP and enhances Ca 2ϩ sensitivity of myosin phosphorylation. Y27632 inhibits PKC␦ and GF109203x blocks both PKC␣ and PKC␦. RhoA activation is coupled with potentiation of thromboxane A 2 receptor (TxA2-R). Although ROCK (31) and PKN (32) phosphorylate CPI-17 in vitro, based on sensitivity to different inhibitors, these kinases contribute little to histamine-induced CPI-17 phosphorylation and contraction of vascular smooth muscle. ZIP-like kinase (ZIP-K) was recently found to bind to MYPT1 and phosphorylates both MYPT1 and CPI-17 (9,33).
to activate kinases, such as Ca 2ϩ /CaM-dependent MLCK via elevation of [Ca 2ϩ ] (1), plus PKC␣ and PKC␦. PKC␣/␦ phosphorylate CPI-17 to inhibit MLCP activity and enhance the apparent Ca 2ϩ sensitivity of myosin phosphorylation. The physiological function of PKC on MLCP regulation, however, is still controversial, and only a minor contribution of PKC was suggested using rabbit trachea smooth muscle (53). The CPI-17/ PKC pathway may mediate histamine-induced contraction of vascular smooth muscle, in which CPI-17 is highly expressed (26).
In ferret aorta smooth muscle, both PKC␣ and PKC⑀ translocate to plasma membrane in response to stimulation by phenylepherine (48,54). We propose that CPI-17 is phosphorylated at the plasma membrane, where active PKC locates in response to histamine stimulation, and then the phospho-CPI-17 diffuses to inhibit MLCP on myofilaments. In this way, CPI-17 might be a messenger that delivers the histamine signal from plasma membrane to myofilaments. Agonists such as thromboxane A 2 activate RhoA and also induce translocation of RhoA to plasma membrane (20,21,55) (Fig. 8). Active ROCK probably also locates at plasma membrane, leaving it some distance from MLCP on myofilaments. RhoA/ROCK seems to generate little contribution to CPI-17 phosphorylation in histamine-induced signaling. This is consistent with the V max value of ROCK against CPI-17, which is 6 times lower compared with MYPT1 (31,56). A kinase, called ZIP-like kinase, has been identified as a MYPT1-associated kinase (9) that phosphorylates the Thr 695 inhibitory site of MYPT1 to inhibit MLCP activity. The ZIP-like kinase activity is insensitive to Y27632 but agonist-induced activation of ZIP-like kinase is suppressed with Y27632, suggesting that this kinase might receive a signal from RhoA/ROCK (9). Interestingly, CPI-17 is phosphorylated by ZIP-like kinase in vitro (33). Therefore, RhoA/ROCK could activate ZIP-like kinase on MLCP, which in turn would phosphorylate both MYPT1 and CPI-17 to reduce MLCP activity. CPI-17 has emerged as a hub for convergence of signaling pathways to reduce MLCP activity in response to vasoconstrictors.