Smooth Muscle Myosin Phosphatase-associated Kinase Induces Ca2+ Sensitization via Myosin Phosphatase Inhibition*

Smooth muscle calcium sensitization reflects an inhibition of myosin light chain phosphatase (SMPP-1m) activity; however, the underlying mechanisms are not well understood. SMPP-1m activity can be modulated through phosphorylation of the myosin targeting subunit (MYPT1) by the endogenous myosin phosphatase-associated kinase, MYPT1 kinase (MacDonald, J. A., Borman, M. A., Muranyi, A., Somlyo, A. V., Hartshorne, D. J., and Haystead, T. A. (2001)Proc. Natl. Acad. Sci. U. S. A. 98, 2419–2424). Recombinant chicken gizzard MYPT1 (M130) was phosphorylated in vitro by a recombinant MYPT1 kinase, and the sites of phosphorylation were identified as Thr654, Ser808, and Thr675. Introduction of recombinant MYPT1 kinase elicited a calcium-independent contraction in β-escin-permeabilized rabbit ileal smooth muscle. Using an antibody that specifically recognizes MYPT1 phosphorylated at Thr654(M130 numbering), we determined that this calcium-independent contraction was correlated with an increase in MYPT1 phosphorylation. These results indicate that SMPP-1m phosphorylation by MYPT1 kinase is a mechanism of smooth muscle calcium sensitization.

Ca 2ϩ binding to calmodulin and activation of MLCK. MLCK catalyzes the phosphorylation of the regulatory myosin light chains at Ser 19 (reviewed in Refs. 2, 4, and 5), resulting in cross-bridge cycling and force development (6). Relaxation follows a return of [Ca 2ϩ ] i to resting levels, the subsequent inactivation of MLCK, and dephosphorylation of myosin catalyzed by SMPP-1M (3). Smooth muscle contraction may also occur in the absence of changes in [Ca 2ϩ ] i following agonist stimulation or by activation of G-proteins with GTP␥S or AlF 4 (4). This Ca 2ϩ sensitization is thought to reflect an inhibition of SMPP-1M via a G-protein-linked mechanism (4,7,8). It is possible to elicit a Ca 2ϩ -independent contraction by completely inhibiting SMPP-1M activity with phosphatase inhibitors, e.g. microcystin (9 -12). This contraction correlates with an increase in myosin regulatory light chain phosphorylation at Ser 19 and Thr 18 (12)(13)(14). A great deal of attention has recently been focused on identifying the kinase or kinases responsible for this myosin phosphorylation in the absence of changes in [Ca 2ϩ ] i . The results of recent efforts have identified both zipper-interacting protein kinase (ZIPK) (14) and integrin-linked kinase (13) as kinases responsible for Ca 2ϩ -independent myosin phosphorylation and contraction in smooth muscle.
SMPP-1M is a heterotrimeric protein composed of a 37-kDa catalytic subunit (PP-1c␦), a 110 -130-kDa regulatory myosin phosphatase targeting subunit (MYPT1), and a 20-kDa subunit of unknown function (10,15,16). The myosin phosphatase activity of SMPP-1M is believed to be regulated by phosphorylation of the MYPT1 subunit. Several phosphorylation sites have been identified on MYPT1, including an inhibitory site of phosphorylation by an endogenous kinase identified as Thr 695 (in the chicken M133 isoform, equivalent to Thr 654 in the chicken M130 isoform) (17).
The molecular mechanism through which activation of Gproteins inhibits SMPP-1M activity is not well established. Evidence suggests that the recently identified endogenous smooth muscle SMPP-1M-associated kinase (MYPT1 kinase) may be implicated in this pathway (1). In support of this, MYPT1 kinase and MYPT1 are colocalized in smooth muscle, and MYPT1 kinase is phosphorylated and activated in response to treatment with calcium-sensitizing agents (1).
To elucidate the mechanism through which MYPT1 kinase acts, the following study was initiated. The following data show that introduction of a constitutively active recombinant MYPT1 kinase (rMYPT1K) into ␤-escin-permeabilized smooth muscle provokes a Ca 2ϩ -independent contraction. We report that this Ca 2ϩ -independent contraction is associated with an inhibition of myosin phosphatase activity caused by phosphorylation at the Thr 695 inhibition site.
Identification of a Calcium-independent Kinase from Rabbit Bladder-Rabbit bladder was cleaned of connective tissue and frozen in liquid nitrogen. Approximately 5 g of tissue was thawed at 4°C in the presence of buffer A (50 mM HEPES, pH 7.4, 50 mM MgCl 2 , 1 mM dithiothreitol, 1 mM CaCl 2 , 1 g/ml leupeptin, 1 g/ml aprotinin, 1 g/ml pepstatin, and 100 g/ml Pefabloc) and homogenized in 2.5 volumes (w/v) of this buffer. The homogenate was centrifuged (6000 ϫ g) for 1 h, and the supernatant was filtered over glass wool. The extract was clarified by centrifugation (100,000 ϫ g) for 1 h and then applied to a 5.0 ϫ 10-cm column of calmodulin-Sepharose followed by a 5.0 ϫ 10-cm column of ethylenediamine and phenylenediamine ␥-linked ATP-Sepharose equilibrated in buffer A. The column was washed sequentially with buffer A containing 0.5 M NaCl to reduce any nonspecific binding, with buffer A plus 5 mM EGTA to eliminate any calcium-dependent kinases, and with buffer A containing 100 M geldanamycin to minimize the recovery of the 90-kDa heat shock protein. 2 Kinase activity was eluted in 2-ml fractions with 10 mM ATP in buffer A. Fractions containing RLC kinase activity were pooled, dialyzed against buffer B (25 mM Tris, pH 7.5, 1 mM dithiothreitol, and 1 mM benzamadine), and applied to an AP-1Q anion exchange column (1.5 ϫ 10-cm) equilibrated in buffer B. The column was washed with buffer B and then developed with a 0 -1 M salt gradient. The fractions (1 ml) were assayed for RLC kinase activity. Fractions containing RLC kinase activity were pooled, dialyzed against buffer B, and applied to a SMART MonoQ PC 1.6/5 column. The column was developed with a 0 -1 M salt gradient, and the fractions (25 l) were assayed for RLC kinase activity.
Kinase Assays-Kinase assays included 1 l of enzyme (SMART fraction) diluted in 25 mM HEPES, pH 7.4, and 100 M RLC substrate peptide to a final volume of 30 l. The reactions were started with the addition of 20 l of ATP solution (6.25 mM MgCl 2 and 0.75 mM ATP (20,000 cpm/nmol)) and carried out at 25°C. The reactions were terminated after 20 min upon addition of 100 l of 20 mM H 3 PO 4 . Aliquots (100 l) of the reaction mixture were spotted on to P81 paper. The P81 paper was washed three times with 20 mM H 3 PO 4 and placed into 1.5-ml Eppendorf tubes, and 32 P incorporation was determined by scintillation counting.
Western Blot Analysis-For Western blotting, SMART fractions (5 l) were subjected to SDS-PAGE (12% acrylamide) and transferred to polyvinylidene difluoride membranes. Nonspecific binding sites were blocked with 10% nonfat dry milk in Tris-buffered saline containing 0.5% Tween 20. Following washing, the membranes were incubated with primary polyclonal antibody to human ZIPK (1:1000) overnight at 4°C. The blots were washed and incubated with anti-rabbit antibody conjugated to horseradish peroxidase (1:2000) for 1 h. The proteins were visualized by chemiluminescence (ECL; Amersham Biosciences).
Protein Preparations-Pig bladder MYPT1 kinase was purified over ␥-linked ATP-Sepharose as described previously (1). Recombinant chicken MYPT1 isoforms were prepared as described: the full-length MYPT1 rM133 as a GST fusion protein (19), a hexahistidine-tagged C-terminal fragment of the rM130 isoform, residues 514 -963 (20), a C-terminal fragment rM130 T654A mutant isoform, and a GST-rM133 T850A mutant isoform. Recombinant MYPT1 kinase was produced from cDNA clone AI660136 encoding the N-terminal (1-320) portion of human ZIP kinase (Genome Systems Inc., St Louis, MO) The cDNA clone was inserted in-frame into vector pGEX-4T-1 (Amersham Biosciences) to express the GST fusion protein. Because the catalytic characteristics of GST-rMYPT1K are essentially the same as those of the purified native MYPT1 kinase (1), we used GST-rMYPT1K to analyze the phosphorylation of MYPT1. For in situ experiments, The MYPT1 kinase recombinant protein was cleaved by treatment with thrombin as described by the manufacturer (Amersham Biosciences).

Permeabilized Muscle Preparation and Tension
Measurement-Ileum was removed from rabbits anesthetized with halothane and exsanguinated according to approved animal protocol. Sheets of longitudinal muscle were peeled and cut into small strips (200 -250 m wide and 3-4 mm long). The muscle strips were stretched to 1.3 times resting length and attached to a force transducer (SensorOne AE801, Sausalito, CA) in a "bubble" chamber. The muscle strips were permeabilized by incubation for 30 min at room temperature with 50 M ␤-escin in an intracellular solution containing 1 mM EGTA and no added Ca 2ϩ (G1) with 10 M A23187 added for the final 10 min to deplete intracellular calcium stores.
Myosin Phosphatase Phosphorylation Measurements-The permeabilized muscle strips were incubated either with relaxing solution containing 10 mM EGTA (G10) and filtrate or with G10 and rMYPT1K (10 M) for predetermined times. Individual muscle strips were rapidly frozen in liquid nitrogen cooled Freon-22 and carefully removed from the transducers. The strips were transferred to 10% trichloroacetic acid in acetone (Ϫ80°C), stored overnight, and then slowly brought to room temperature, washed extensively with acetone, and dried. The tissues were homogenized in a buffer containing 25 mM Tris, pH 7.0, 2 mM EGTA, 1 mM dithiothreitol, 600 mM NaCl, 0.5% Triton, 1 g/ml leupeptin, and 1 g/ml aprotinin. The tissue homogenates were separated by 10% SDS-PAGE and transferred to polyvinylidene difluoride membranes. Nonspecific binding sites were blocked with 10% nonfat dry milk in Tris-buffered saline containing 0.5% Tween 20. The membranes were incubated with primary monoclonal antibody to MYPT1 (1:2000) or primary polyclonal antibody to MYPT1 phosphorylated specifically at Thr 695 (pM133 T695 ) (1:10,000) overnight at 4°C. The blots were washed and incubated with anti-rabbit antibody conjugated to horseradish peroxidase (1:2000) for 1 h. The proteins were visualized by ECL.
Phosphorylation Site Analysis of 32 P-Labeled Recombinant MYPT1 Isoforms-Phosphorylation of MYPT1 isoforms by native MYPT1 kinase and GST-rMYPT1K were carried out at 37°C in 20 mM HEPES, pH 7.2, 1 mM MgCl 2 , 0.1 mM dithiothreitol, 0.1 mM ATP (500 cpm/mol) with 100 g of substrate in a final volume of 50 l. 32 P-MYPT1 protein samples from in vitro phosphorylation reactions were incubated overnight at 37°C with endoproteinase Lys-C (0.5 g). The digests were acidified by the addition of trifluoroacetic acid and applied to a reverse phase column (Zorbax SB-C18, 4.6 ϫ 250 mm) that had been equilibrated in 0.1% trifluoroacetic acid (Buffer A). The column was washed with Buffer A before the peptides were eluted with a linear gradient of acetonitrile (0 -60% in 100 min). The fractions (200 l) were collected, and the phosphopeptides were identified by Cerenkov radiation. A portion of the fractions containing phosphopeptide was used for phosphoamino acid analysis (21).
The HPLC fraction containing the major peak of radioactivity was subjected to nanospray mass spectrometry on a QSTAR-Pulsar mass spectrometer (Applied Biosystems, Foster City, CA). Negative mode time of flight was used to measure peptide whole masses (MS), and a precursor ion scan in negative mode was used to identify those parent ions that liberated the characteristic m/z 78.997 corresponding to the phosphate ion. Positive mode time of flight was then used to measure the masses of peptide fragments produced by collision-induced dissociation (MS/MS) and, ultimately, to predict the sequence of the unknown phosphopeptide.
The remainder of the phosphorylation sites were determined using differential proteolysis and cleaved radioactive protein (CRP) analysis. The phosphopeptide samples were immobilized to Immobilon membrane (Millipore) following the manufacturer's instructions. The phosphorylated residues within phosphopeptides were located by determining the cycles in which 32 P was released when the samples were subjected to sequential Edman degradation with a vapor phase amino acid sequencer (Applied Biosystems Procise 494) under conditions that optimize recovery of 32 P (22). The CRP analysis program (23) was used to assign a phosphorylation site to the 32 P released in a specific cycle. The CRP analysis program is available at fasta.bioch.virginia.edu/crp/.

Identification of a Calcium-independent Kinase That Phosphorylates Myosin Light Chains-To identify any additional
Ca 2ϩ -independent kinase or kinases that phosphorylates myosin light chains, ␥-linked ATP-Sepharose affinity chromatography was utilized. A substrate peptide sequence containing the Ser 19 and Thr 18 phosphorylation sites of RLC was synthesized. Kinase activity was isolated from the cytosolic fraction of rabbit 2 P. Fadden and T. A. J. Haystead, unpublished data. bladder, and two distinct peaks of activity toward the RLC peptide were identified (Fig. 1). The behavior of the second peak of kinase activity (655 mM NaCl) through the anion exchange purification steps was similar to that reported for a recently identified SMPP-1M-associated kinase (1). Subsequent Western blot analysis identified this Ca 2ϩ -independent kinase as the 34-kDa ZIPK-like kinase isoform, now referred to as MYPT1 kinase. In a recent report, MYPT1 kinase was shown to have significant sequence homology to ZIPK, confirming that this newly identified Ca 2ϩ -independent light chain kinase and the previously identified MYPT1 kinase are in fact the same protein. The kinase corresponding to the first peak of activity (460 mM NaCl) remains under investigation.
Expression and Characterization of MYPT1 Kinase-MYPT1 kinase was originally identified as a kinase that phosphorylates MYPT1 (1); subsequently this same kinase was shown to phosphorylate a RLC substrate peptide (present data). The phosphorylation of RLC by HeLa ZIPK has been previously demonstrated (24). To determine the cellular function of MYPT1 kinase, recombinant MYPT1 kinase (rMYPT1K) was expressed in Escherichia coli, and its enzymatic properties were examined in vitro. We have analyzed the specificity of rMYPT1K toward two known protein substrates, RLC and MYPT1, by comparing enzyme affinities. Michaelis constants (K m ) were determined for RLC protein and C130, a C-terminal fragment of chicken gizzard M130 that contains the known inhibitory phosphorylation sites. K m values were 53 Ϯ 6 M (n ϭ 4) for the RLC protein and 3 Ϯ 0.04 M (n ϭ 4) for the C130 protein. Catalytic constants (k cat ) were also measured for RLC and C130 and were 1.9 ϫ 10 Ϫ2 s Ϫ1 Ϯ 9.6 ϫ 10 Ϫ4 (n ϭ 4) and 5.0 ϫ 10 Ϫ2 s Ϫ1 Ϯ 1.8 ϫ 10 Ϫ2 (n ϭ 4), respectively. These results suggest that MYPT1 may be a preferred substrate in vivo.
In Situ Phosphorylation of MYPT1 by rMYPT1K-Because the in vitro kinetics results for rMYPT1K against the two putative protein substrates suggested that MYPT1 was a better substrate, we investigated the in situ phosphorylation of MYPT1 by rMYPT1K in permeabilized smooth muscle. The effect of 10 M rMYPT1K on the extent of MYPT1 phosphorylation was determined in calcium-free conditions (G10). As shown in Fig. 2A, incubation of rabbit ileal strips with rMYPT1K provoked a Ca 2ϩ -independent contraction equivalent to 37.5% Ϯ 2.0% (n ϭ 7) of the maximal calcium (CaG) contraction. To determine whether the contractile effect of rMYPT1K was associated with phosphorylation of MYPT1, rabbit ileal strips were mounted in a bubble chamber and permeabilized with ␤-escin. Following maximal calcium contraction, the muscles were washed in calcium-free solution containing 10 mM EGTA and then incubated for 20 min in the same solution in the absence or presence of 10 M rMYPT1K. After 20 min, the strips were rapidly frozen, and MYPT1 phosphorylation at Thr 695 was measured (see "Experimental Procedures"). Fig. 2 (B and C) shows that incubation of muscle strips FIG. 1. Identification of a calcium-independent light chain ki-nase. Calcium-independent RLC kinase activity was eluted from a SMART MonoQ (1.6/5) anion exchange column with a 0 -1 M NaCl gradient and identified using an in vitro kinase assay. The Ca 2ϩindependent kinase corresponding to the second major peak of activity was identified through Western blot analysis (inset) as a 34-kDa ZIPlike kinase. This same protein has been previously identified as the endogenous myosin phosphatase-associated kinase MYPT1 kinase (1).

FIG. 2. Recombinant MYPT1 kinase induces Ca 2؉ -independent contraction correlated with an increase in MYPT1 phosphorylation.
A, following ␤-escin permeabilization, rabbit ileal strips were contracted once with CaG, relaxed, and incubated for 15 min in the absence (left panel) or presence (right panel) of 10 M rMYPT1K in calcium-free solution containing 10 mM EGTA. Incubation with rMYPT1K elicited a Ca 2ϩ -independent contraction. B, ileum strips were rapidly frozen following the 15-min incubation, and tissue homogenates were blotted for both total myosin phosphatase and myosin phosphatase specifically phosphorylated at Thr 695 . C, the Ca 2ϩindependent contraction induced by rMYPT1K treatment was correlated with a 5-fold increase in myosin phosphatase phosphorylation. The results are expressed relative to the control treatment and are the averages Ϯ S.E. of 12 experiments (control) or 9 experiments (rMYPT1K). with 10 M rMYPT1K causes a 5-fold increase in the level of Thr 695 phosphorylation over control muscle strips (5.0 Ϯ 0.8 in rMYPT1K-treated strips (n ϭ 9); 1.0 Ϯ 0.2 in control strips (n ϭ 12)).
Phosphorylation of MYPT1 in Vitro by Native MYPT1 Kinase and Recombinant GST-rMYPT1K-Initial studies using native MYPT1 kinase suggested phosphorylation of multiple sites on M133. Both phosphoserine and phosphothreonine residues were detected by phosphoamino acid analysis (Fig. 3, inset). To identify the individual phosphorylation sites, we used recombinant MYPT1 proteins phosphorylated by GST-rMYPT1K. As shown in Table I, GST-rMYPT1K phosphorylated multiple sites of rM130 in vitro. The reverse phase chromatography profile showed a major peak of 32 P in fraction 40 of the recombinant wild-type C130 and rM133 T850A protein digests. This peak was absent in the chromatography profile of the recombinant C130 T654A protein digest. Three additional peaks (I, II, and IV) of 32 P were observed in the reverse phase chromatography profile.
We used mass spectrometry to identify the major phosphorylation site (peak III) in the C-terminal fragment of the wildtype rM130 protein (Fig. 4). Phosphopeptides in fraction 40 were treated with trypsin overnight and subjected to nanospray mass spectrometry. The additional minor sites of phosphorylation were identified by determining the cycles in which 32 P was released when 32 P-labeled fractions of the protein digests were subjected to sequential Edman degradation under conditions that optimized recovery of 32 P. Solid phase Edman sequencing of peak I and CRP analysis of MYPT1 (Fig. 5) revealed that Thr 675 was the only residue that could yield the release of 32 P in the eighth cycle following digestion with endoproteinase Lys-C and in the fourth cycle following digestion with endoproteinase Arg-C. From a similar analysis of peak II, we have identified Ser 849 , which corresponds to Ser 854 in Rat3 MBS and Ser 808 in chicken M130, as a minor phosphorylation site. In addition, phosphorylation of the peptide in peak II was reduced but not eliminated in the rM133 T850A protein digest; this is also consistent with the phosphorylation of Thr 850 on the wild-type M130 protein. A phosphoamino acid analysis of peak II phosphopep-tides from wild-type M130 protein confirmed the presence of both Ser(P) and Thr(P) (data not shown). The phosphorylation of Ser 854 was reported to be specific to Rho-associated protein kinase (ROK), and Ser 854 phosphorylation has been used in previous studies to identify ROK-specific phosphorylation of MBS in vivo (25). DISCUSSION Substrate specificity in signal transduction is often conferred to a common catalytic subunit through a unique targeting subunit (reviewed in Ref. 26). The targeting subunits themselves may be targets of signaling pathways that can modulate the activity of the enzyme toward its specific substrate. The smooth muscle myosin phosphatase targeting subunit (MYPT1) is regulated by multiple signaling pathways. In the The phosphorylated residue in each peptide is underlined. c The values were determined from the radioactivity in the HPLC analysis of wild-type M130 as shown in Fig. 3. d ND, not determined. current study, we have provided evidence for a mechanism regulating the activity of MYPT1. There is substantial evidence that smooth muscle Ca 2ϩ sensitization reflects an inhibition of myosin phosphatase activity (reviewed in Refs. 3 and 4). However, the mechanism of SMPP-1M inhibition remains unclear. Several theories have now been described: inhibition by CPI-17, a 17-kDa phosphatase inhibitor and protein kinase C substrate protein that becomes a potent inhibitor of SMPP-1M following phosphorylation by protein kinase C (27,28); dissociation of the SMPP-1M holoenzyme through arachidonic acid interaction (29); and phosphorylation of MYPT1 by an endogenous kinase (1,17) or by ROK (30).
Previous studies reported that phosphorylation of MYPT1 at Thr 695 by an endogenous kinase caused inhibition of SMPP-1M activity (17). We subsequently identified this endogenous kinase that copurifies with the holoenzyme of myosin phosphatase as MYPT1 kinase (1), which shows significant sequence homology to the previously described ZIPK (31). The data presented above indicate that MYPT1 kinase phosphorylates MYPT1 in situ in rabbit ileal smooth muscle and that this MYPT1 phosphorylation is correlated with a Ca 2ϩ -independent contraction. Our data suggest that MYPT1 is indeed a target of smooth muscle MYPT1 kinase contributing to calcium sensitization.
We previously reported that MYPT1 kinase phosphorylates the inhibitory Thr 695 site of MYPT1 (1). In the present study, we report the presence of three additional MYPT1 phosphoryl-ation sites, two of which we have identified. Ser 854 in Rat3 MBS was previously reported to be a phosphorylation site specific to ROK and was proposed to be an effective indicator of ROK activation in vivo (25). Phosphorylation of Ser 849 , equivalent to Ser 854 in Rat3 MBS, did not have an inhibitory effect on SMPP-1M activity (18). In the present study, we have identified Ser 808 , which corresponds to Ser 854 in Rat3 MBS and Ser 849 in chicken M133, as a minor MYPT1 kinase phosphorylation site. Our data indicate that Ser 854 phosphorylation is not unique to ROK and suggest a renewed examination of the role of ROK in SMPP-1M regulation. We also identify Thr 675 as a minor MYPT1 phosphorylation site. The role of this newly identified phosphorylation site in the regulation of SMPP-1M remains unknown.
Native MYPT1 kinase was recovered from the cytosolic fraction of bladder smooth muscle following homogenization in the absence of any detergent. Native MYPT1 kinase was previously isolated from a Triton-solubilized fraction consisting of myofilament and cytoskeletal components (1), suggesting the presence of two distinct subcellular fractions of MYPT1 kinase in smooth muscle. Presumably, it is the myofilament-associated pool of MYPT1 kinase that would associate with and phosphorylate myosin phosphatase in vivo. An additional substrate for MYPT1 kinase that may be important in the regulation of smooth muscle contraction is myosin. It was previously reported that a full-length smooth muscle ZIPK phosphorylated myosin at both Ser 19 and Thr 18 (14). We have also observed FIG. 5. Phosphorylation site analysis of MYPT1. Theoretical results obtained from the CRP program are displayed. The full-length protein sequence of M130 was processed by the CRP program with proteolysis selected for lysine (A) or arginine (B). The CRP analysis data include: the Edman sequence cycle number (Cycle #) and any Ser, Thr, or Tyr residues present in the protein (Potential Phosphorylation Sites). Edman sequencing cycle data of C18 chromatography fractions containing 32 P-labeled phosphopeptides are also presented. Phosphopeptides I-IV were coupled to Sequelon AA disks following the manufacturer's instructions. The results represent the amount of 32 P released in each cycle when the fractions were subjected to sequential Edman degradation after treatment with endoproteinase Lys-C or Arg-C. that rMYPT1K phosphorylates RLC and whole myosin on its regulatory light chains in vitro (data not shown). The change in calcium sensitivity associated with rMYPT1K treatment may therefore reflect two distinct phosphorylation mechanisms, i.e. MYPT1 and myosin. This possibility will require further investigation. Current procedures for investigating the contribution of MYPT1 phosphorylation in the calcium sensitization of force development involve the indiscriminate use of MLCK inhibitors (i.e. ML-9). It should be noted that this commonly used MLCK inhibitor also inhibits the activity of MYPT1 kinase in vitro at levels below those required for full MLCK inactivation (IC 50 for MYPT1 kinase: ϳ150 M ML-9). Researchers who use MLCK inhibitors to investigate GTP␥S-or ATP␥S-induced force enhancement should be aware that they may be inadvertently influencing different signaling pathways.
It is unclear why the phosphorylation results obtained with native MYPT1 kinase and rMYPT1K in our studies (this paper and Ref. 1) are different from those recently reported by Niiro and Ikebe (14). Indeed, they have suggested that MYPT1 phosphorylation by native MYPT1 kinase may be physiologically irrelevant because in their hands MYPT1 is a poor substrate for recombinant ZIPK. It is possible, as suggested by Richards et al. (32), that ZIPK (and native MYPT1 kinase) only phosphorylates the M133 isoform of MYPT1 (i.e. Thr 695 site) and not the shorter, spliced-out M130 isoform (i.e. Thr 654 site). The expression of MYPT1 isoforms is both developmentally regulated and tissue-specific (3), thereby raising the possibility that MYPT1 isoforms could modulate the magnitude of agonistinduced Ca 2ϩ sensitization (32). Although we have not completed a comprehensive examination of the phosphorylation kinetics for the M130/M133 isoforms, our preliminary studies (Fig. 3) have shown that both the M130 and M133 isoforms are phosphorylated by MYPT1 kinase.
We have now identified three substrates for the smooth muscle MYPT1 kinase that could be involved in regulating calcium sensitization: MYPT1 (Ref. 1 and present data), myosin (present data), and the protein kinase C-activated peptide inhibitor (CPI-17) (33). We have established the sites of phosphorylation of MYPT1, which include the known inhibitory site (Thr 654 ), a site previously thought to be unique to ROK (Ser 808 ), and a newly identified phosphorylation site whose role in SMPP-1M regulation has not been established (Thr 675 ). We have also demonstrated that a constitutively active fragment of MYPT1 kinase induces contraction in ␤-escin permeabilized ileal smooth muscle and that this Ca 2ϩ sensitization is correlated with an increase in phosphorylation of SMPP-1M at Thr 654 . If this occurs in vivo, it would likely be under conditions where MLCK activity is reduced, i.e. at low calcium concentrations. This would clearly necessitate a signal other than calcium to activate MYPT1 kinase.