A Myosin Light Chain Kinase Inhibitor, ML-9, Lowers the Intraocular Pressure in Rabbit Eyes

Exp. Eye Res. (2002) 75, 135±142 doi:10.1006/exer.2002.2009, available online at http://www.idealibrary.com on

A Myosin Light Chain Kinase Inhibitor, ML-9, Lowers the Intraocular Pressure in Rabbit Eyes M E G U M I H O N J O *, M A S A R U I N ATA N I , N O R I A K I K ID O , TAT S U YA S AWA M U R A , B E ATR I C E Y. J . T. Y U E , YO S H I H I TO H O N D A A N D H ID E N O B U TA N I H A R A a

Department of Ophthalmology and Visual Sciences, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan, bDepartment of Bioscience, National Cardiovascular Center Research Institute, Suita, Osaka 565-8565, Japan, cDepartment of Molecular Pathophysiology, Graduate School of Pharmaceutical Sciences, Osaka University, Suita, Osaka 565-0871, Japan, dDepartment of Ophthalmology and Visual Sciences, University of Illinois at Chicago, College of Medicine, Chicago, IL, U.S.A. and eDepartment of Ophthalmology, Kumamoto University School of Medicine, 1-1-1 Honjo, Kumamoto, 860-8556, Japan (Received 20 November 2001 and accepted in revised form 5 March 2002) The role of myosin light chain kinase (MLCK) in regulating the intraocular pressure (IOP) and out¯ow facility in rabbit eyes were studied. The IOP and pupil diameter were determined before and after intracameral and intravitreal administration of ML-9, a speci®c MLCK inhibitor. Total out¯ow facility and uveoscleral out¯ow facility was determined 3 hr after intracameral administration of ML-9. Immunoblotting was performed to identify MLCK and the 20-kDa light chain of myosin (MLC) isoforms in human trabecular meshwork (TM) cells. The phosphorylation status of MLC was examined following ML-9 treatment. The effects of ML-9 on the morphology and actin and vinculin distribution in cultured TM cells were also studied. In rabbit eyes, administration of ML-9 resulted in a dose-dependent decrease in IOP. An increase of the out¯ow facility was also observed. Immunoblot analysis revealed the presence of MLCK in human TM cells. Exposure to ML-9 dose-dependently inhibited MLC phosphorylation/activation. The inhibitor caused retraction and dissociation of cells, disruption of actin bundles and impairment of focal adhesion formation in TM cells. ML-9 induces a reduction in IOP and an increase in the out¯ow facility in rabbit eyes. The IOP-lowering effects may be related to alterations in TM cell shapes. Inhibitors of MLCK # 2002 Elsevier Science Ltd. may potentially be developed into novel medications for glaucoma. Key words: MLCK; ML-9; trabecular meshwork cell; intraocular pressure; out¯ow facility.

1. Introduction Phosphorylation of the regulatory light chain of myosin II (MLC) is well known to participate in control of the actomyosin contractility in both smooth muscle and non-muscle cells. MLC phophorylation has been shown to be essential and suf®cient for the formation of stress ®bers and focal adhesions in ®broblastic cells (Kamm and Stull, 1985; Moussavi et al., 1993). The phosphorylation is regulated by the balance of two enzymatic activities, namely, myosin light chain kinase (MLCK) and myosin light chain phosphatase. On the one hand, MLCK phosphorylates MLC in the presence of Ca2‡ and calmodulin (Ca2‡ -CaM), and thereby activates myosin to interact with actin ®laments (Somlyo and Somlyo, 1994; Hartshorne et al., 1998). On the other hand, ROCK/Rho kinase has been shown to phosphorylate the largest subunit of myosin phosphatase in the carboxyl-terminal region, resulting in inhibition of the phosphatase activity (Kimura et al., 1996; Kawano et al., 1999). * Address correspondence to: Megumi Honjo, Department of Ophthalmology and Visual Sciences, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: [email protected]

0014-4835/02/$35.00/0

The ROCK-mediated inhibition of myosin phosphatase may also account for the increased MLC phosphorylation, and the resultant contractility of actomyosin to induce stress ®bers and focal adhesions (ChrzanowskaWodnicka and Burridge, 1996; Kimura et al., 1996; Kawano et al., 1999). Both MLCK activity and myosin phosphatase inhibition are essential for the assembly of actin stress ®bers (Totsukawa et al., 2000). We have recently shown that inhibitors of ROCK, Y-27632 and HA-1077 caused a reduction in the intraocular pressure (IOP) and an increase in the out¯ow facility in rabbit eyes. The effects were possibly exerted through alteration in the contractility and shapes of trabecular meshwork (TM) cells (Honjo et al., 2001a,b). ML-9, 1-(5-chloronaphthalenesulfonyl)1H-hexahydro-1,4-diazepine, is a speci®c MLCK inhibitor. This compound has been previously found to inhibit smooth muscle contraction (Zhao et al., 1996) and alter cellular actions including cell migration and cytoskeletal organization (Saito et al., 1998; Samizo et al., 1999). Here we demonstrate that inhibition of MLCK induced a signi®cant decrease in IOP and an increase in the out¯ow facility in rabbits, possibly through # 2002 Elsevier Science Ltd.

136

M. HONJO ET AL.

regulation of the MLC phosphorylation and in¯uence on TM cells.

animals in each group were randomly chosen and served as controls.

2. Materials and Methods

Total Out¯ow Facility and Uveoscleral Out¯ow

Animals and Drug Application

Total out¯ow facility and uveoscleral out¯ow were measured as previously described (Taniguchi et al., 1996; Honjo et al., 2001a,b). Total out¯ow facility was determined by two-level constant pressure perfusion (25 and 35 mmHg) 3 hr after intracameral administration of 1.2 ml of 100 mM ML-9 and vehicle, according to the method of Barany (1964). Brie¯y, the anterior chambers of rabbits anesthetized with 40 % urethane were perfused with mock aqueous humor (BSS plus, Santen Pharmaceutical Co. Ltd., Osaka, Japan) by a constant pressure of either 25 or 35 mmHg, which was alternately applied at 10 min intervals. During each 10 min period, ¯uid ¯ow was measured for 8 min beginning 2 min after the pressure change was induced. Uveoscleral out¯ow was determined with a perfusion technique using ¯uorescein isothiacyanate-dextran (FITC-dextran, MW ˆ 71 200, Sigma), (Taniguchi et al., 1996; Honjo et al., 2001a,b) beginning at 3-hr time point after the intracameral administration of 1.2 ml of 100 mM ML-9 and vehicle. The IOP level was set to 20 mmHg and the FITC-dextran solution was perfused continuously through the anterior chamber for 30 min, then washed. The amount of tracer in the tissues, anterior uvea, anterior sclera, posterior sclera plus posterior uvea and the posterior segment ¯uid plus vitreous, was measured using a ¯uorophotometer.

Adult Japanese white rabbits weighing 2±2.5 kg were studied. All studies were conducted in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. For IOP and pupil diameter measurements, the rabbit eyes were anesthetized by topical instillation of 2 % lidocaine. For measurements of out¯ow facility or uveoscleral out¯ow, the rabbits were anesthetized by very slow administration of a 40 % urethane (1.0±1.5 ml kg 1) solution into a marginal ear vein. ML-9 (Sigma, St. Louis, MO, U.S.A.) was dissolved in 50 % ethanol and diluted with saline to concentrations ranging from 10±1000 mmol L 1 (mM). The ®nal concentration of ethanol was less than 0.05 % (v/v). Anterior chamber injections and injections to the vitreous cavity were performed as previously described (Honjo et al., 2001a,b). Brie¯y, anterior chamber injections were made under an operating microscope, using a microsyringe (Hamilton, Reno, NV, U.S.A.) with a 30-gauge needle. The needle was threaded through the corneal stroma for approximately 3 mm, then directed into the anterior chamber so that the wound was self-sealing. For intravitreal injections, the needle was inserted 3 mm through the temporal sclera 2 mm posterior to the limbus. 1.2 ml of 10, 100 or 1000 mM of ML-9 or control solutions (as an equal volume of saline diluent of ethanol) was injected into the anterior chamber. Fourteen ml of 10, 100 or 1000 mM of ML-9 was administered intravitreally. Slit Lamp Biomicroscopy The integrity of the corneal epithelium, the presence or absence of anterior chamber ¯are or cells and lens clarity were examined 0.5, 1, 3, 6, 9, 12, 18 and 24 hr after the drug administration. IOP and Pupil Diameter Measurement A calibrated pneumotonometer (Alcon, Forth Worth, TX, U.S.A.) was used to monitor the IOP. It was measured before administration of ML-9, and at 0.5, 1, 3, 6, 9, 12, 18 and 24-hr time points after the intracameral and intravitral administration. The pupil diameter was measured with a millimeter ruler (Digimatic Caliper; Mitutoyo Co. Ltd., Tokyo, Japan) under standard laboratory light at the same time points as the IOP measurements were made. Six animals were used for each different dose in intracameral and intravitreal groups, so that 36 animals were used in this IOP and pupil diameter study. All animals received vehicle in the contralateral eyes, but six animals of 18

Culture of Human TM Cells Normal human eyes from donors were obtained from the Illinois eye bank (Chicago, IL, U.S.A.). Trabecular tissues excised from eyes were cultured on Falcon Primaria ¯asks (Becton Dickson, Lincoln Park, NJ, U.S.A.) as previously described (Yue et al., 1984; Sawaguchi et al., 1992). The culture medium included Dulbecco's Eagle's minimum essential medium (DMEM), 10 % fetal bovine serum (FBS) and antibiotics. Cells were maintained in a 95 % air±5 % CO2 atmosphere at 378C and passaged using the trypsin-EDTA method. TM cells from passages three through eight were used for our studies. Chemicals and Antibodies FITC-phalloidin, mouse monoclonal antibody to vinculin, mouse anti-MLC and rabbit anti-MLCK were obtained from Sigma. Mouse anti-phosphorylated MLC was a kind gift from Asahi Chemical Industry Co., Ltd. (Tokyo, Japan). Appropriate secondary antibodies were obtained from Chemicon International (Temecula, CA, U.S.A.) and Amersham (Little Chalfont, U.K.). Chemicals for electrophoresis were purchased from Bio-Rad Laboratories (South

OCULAR EFFECTS OF MLCK INHIBITOR

Richmond, CA, U.S.A.) and Daiichi pure chemicals (Tokyo, Japan). Preparation of Lysates from Whole Cell and Immunoblotting To examine the levels of MLCK and phosphorylated MLC, detergent lysates of TM cells were prepared in Laemmli's SDS±polyacrylamide gel electrophoresis (SDS±PAGE) sample buffer, as previously described (Honjo et al., 2001a,b). Brie¯y, con¯uent cultures of human TM cells were kept in serum-free DMEM overnight and were incubated for 40 min with 0.1±10 mM of ML-9 with serum stimulation. Serum starvation was also studied. The cells were scraped and lysed, and equivalent amounts of protein-containing lysates were subjected to SDS±PAGE using a 10±20 % gradient gel (Daiichi pure chemicals, Tokyo, Japan). The proteins were transferred onto polyvinylidene di¯uoride membranes (Millipore Co., Bedford, MA, U.S.A.). For blotting, the membrane was blocked at 48C with 2 % bovine serum albumin in phosphate buffered saline (PBS) containing 0.005 % Tween 20 (PBST) for 16 hr, and was incubated at 48C with antiMLCK or anti-phospho-MLC for 24 hr. The membrane was further incubated with biotin-conjugated secondary antibody (Amersham, NJ, U.S.A.) and ABC solution (ABC Elite kit, Vector). After extensive washing the blotted protein bands were visualized with Konica immunostain (Konica immunostain HRP1000, Konica, Tokyo, Japan).

137

Statistical Analysis Data were analyzed by Student's unpaired t-test as a post hoc test of time course of IOP and pupil diameter. Mann-Whitney U-test was used for analyses of the aqueous humor dynamics. A P value of 50.05 was considered to be statistically signi®cant. 3. Results Effects of ML-9 on IOP, Pupil Diameter, and Out¯ow Facility Compared with contralateral vehicle-treated controls, the IOP in rabbit eyes was signi®cantly (P 5 0.05) lowered between 0.5 and 18 hr after intracameral administration of 100 and 1000 mM ML-9 (resulting in 1 and 10 mM ®nal concentration in the anterior chamber). The IOP reduction was maximal between 6 and 12 hr with 1000 mM of ML-9 (Fig. 1(A)). After intravitreal administration, signi®cant IOP reductions (P 5 0.05) were noted between 0.5 and 12 hr, and the maximal reductions

Effects of ML-9 on Shape of Human TM Cells Semi-con¯uent cultures were incubated with 0.1±10 mM of ML-9 with serum. The cultures were monitored by phase-contrast microscopy and photographed immediately after drug application, and 10, 30 and 60 min later. The drug solution was removed and replaced with the DMEM containing 10 % FBS. In all cases, recovery of normal morphology was documented 15 hr later. Immunohistochemistry Immunohistochemistry was performed as previously described (Honjo et al., 2001a,b). Brie¯y, human TM cells were incubated with 0.1±10 mM of ML-9 for 40 min, ®xed and permeabilized for 2 min with 3 % paraformaldehyde/PBS and 0.5 % Triton X-100 (Sigma), and were further ®xed at 48C with 3 % paraformaldehyde for 20 min. Filamentous actin (F-actin) was labelled with FITC-phalloidin. The primary antibodies used in this study were anti-vinculin. Fluorescence was visualized under an epi¯uorescence microscope (Zeiss Axioplan, Oberkochen, Germany) and with a confocal laserscanning microscope (Bio-Rad Laboratories).

F IG . 1. Effects of ML-9 on the intraocular pressure. ML-9 administered into the anterior chamber (A) and vitreous (B) in the rabbit eyes. The contralateral eyes were treated with the same volume of vehicle. W, vehicle alone; w 10 mM; r 100 mM; q 1000 mM of ML-9. The results are presented as mean + S.E.(M.) (n ˆ 6). The signi®cance of the data was evaluated by Student's unpaired t-tests; *P 5 0.05, **P 5 0.01 and {P 5 0.005, compared to controls with vehicle alone.

138

were seen between 0.5 and 3 hr with the 1000 mM concentration (Fig. 1(B)). No anterior chamber, lens or fundus abnormalities in rabbit eyes were detected by slit lamp examination. Compared with contralateral vehicle-treated controls, the pupil diameter (PD) in rabbit eyes after ML-9 treatment was signi®cantly (P 5 0.005) larger from 0.5 to 18 hr following intracameral administration of 1000 mM of ML-9. The maximal PD was observed from 3 to 9 hr with 1000 mM of ML-9 (P 5 0.005) (Fig. 2(A)). Intravitreal administration of ML-9 did not change the PD as appreciably as intracameral administration, with only 1000 mM of ML-9 at the 3-hr time point being statistically signi®cant (P 5 0.05) (Fig. 2(B)). Out¯ow facility was measured 3 hr after intracameral administration of ML-9 when maximal IOP reduction was observed. Results summarized in Table I showed that the average out¯ow facility was approximately 1.9-fold higher in the eyes treated with 1000 mM of ML-9 (0.25 + 0.02 ml min 1 mmHg 1, P 5 0.005) than that in the contralateral vehicletreated control eyes (0.14 + 0.01 ml min 1 mmHg 1).

M. HONJO ET AL.

TABLE I Effects of ML9 on out¯ow facility in the rabbit Out¯ow facility (ml min 1 mmHg 1) ML9 Vehicle Signi®cance*

0.25 + 0.02 0.14 + 0.01 P 5 0.005

Uveoscleral out¯ow (ml min

1

)

0.43 + 0.03 0.37 + 0.02 NS

Values are means + S.E.(M.) for six animals. *Mann-Whitney test.

Uveoscleral out¯ow in the treated eyes (0.43 + 0.03 ml min 1) was within the normal limits of controls (0.37 + 0.02 ml min 1). Phosphorylation of MLC in Cultured Human TM Cells Immunoblot blot analyses using anti-phospho MLC and anti-MLCK antibodies detected a protein band of expected molecular size, about 20 and 160 kDa, respectively, in human TM cells. The total MLCK expression did not change following the ML-9 treatment (data not shown). The phosphorylated MLC, which was detected in normal TM cells in the presence of serum stimulation, vanished however after the ML-9 treatment in a dose-dependent manner (Fig. 3). Effects of ML-9 on Morphology of Cultured Human TM Cells By phase-contrast microscopy, TM cells in semicon¯uent cultures retracted and became thinner when treated with 0.1±10 mM of ML-9 in the presence of serum for 60 min (Fig. 4). To determine whether such changes were related to the serum stimulation, the cells were also incubated in serum-free medium. Retraction and dissociation were seen 30±60 min later (Fig. 4, right panels). With the con¯uent cultures, treatment with 0.1±10 mM of ML-9 resulted in frequent partial

F IG . 2. Effects of ML-9 on the pupil diameter. ML-9 administered into the anterior chamber (A) and vitreous (B) in the rabbit eyes. The contralateral eyes were treated with the same volume of vehicle. W, vehicle alone; w 10 mM; r 100 mM; q 1000 mM of ML-9. The results are presented as mean + S.E.(M.) (n ˆ 6). The signi®cance of the data was evaluated by Student's unpaired t-tests; *P 5 0.05, **P 5 0.01 and {P 5 0.005, compared to controls with vehicle alone.

F IG . 3. Immunoblot analyses of phosphorylated-MLC in cultured human TM cells. Homogenate of whole cell lysates from cultured human TM cells were run in SDS±PAGE. Representative immunoblots showing that the level of phosphorylated MLC observed with serum stimulation essentially disappeared upon serum starvation or 0.1±10 mM ML-9 treatment. Total MLCK immunoreactivity remained unchanged either with serum starvation or ML-9 treatment (data not shown). 1, Serum-stimulated; 2, 0.1 mM ML-9; 3, 1 mM ML-9; 4, 10 mM ML-9, 5, serum starvation.

OCULAR EFFECTS OF MLCK INHIBITOR

139

F IG . 4. Effects of ML-9 on morphology of cultured human TM cells. Phase contrast microscopic observation of semi-con¯uent culture of TM cells. (A) Normal morphology of cultured human TM cells with serum. (B) Treatments with ML-9 in concentrations of 0.1±10 mM up to 60 min in the presence of serum resulted in retraction and dissociation of the cells (black arrows). The drug solutions or serum-free DMEM were removed afterwards and replaced with DMEM containing 10 % FBS. Recovery of normal morphology was observed within 2 hr and continued for 15 hr (Original magni®cation 60).

split of the cells, forming intercellular gaps (data not shown). Effect of ML-9 on Actin Filaments and Cellular Adhesions In control cells, actin ®laments were assembled into large radial and circumferential bundles in association with focal adhesions (Fig. 5(A)). After treatment with 0.1±10 mM of ML-9 for 30 min, the distribution of F-actin was altered dramatically in a time- and concentration-dependent manner (Fig. 5(B)). The

intensity of the phalloidin staining was severely reduced, and instead of discrete long ®laments, F-actin distributed in small punctuate structures, mostly located at the ventral face of the cells and in the cortical area under the membrane. The focal adhesion, an important protein complex that links the actin network to the plasma membrane at adhesion sites (Rudiger, 1998), is modulated by the level of actin expression (Schevzov et al., 1995). As expected, vinculin in control cells was predominantly associated with focal adhesions (Fig. 5(A)). After

140

M. HONJO ET AL.

F IG . 5. Distribution of F-actin and vinculin in human TM cells treated with ML-9. (A) Distribution of F-actin (in green) and vinculin (in red) in normal human TM cells. Small white arrows indicate F-actin bundles, and white arrowheads denote vinculin-containing focal adhesions. (a) Confocal images, (b) cells were stained with antibody to vinculin, (c) cells were stained with FITC-phalloidin to visualize F-actin. (B) Distribution of F-actin and vinculin in human TM cells treated with 0.1, 1 and 10 mM of ML-9 for 30 and 60 min. The drug solutions were then removed and replaced with DMEM containing 10 % FBS. Recovery of normal morphology was observed 2 and 15 hr later. White arrows point to F-actin bundles, which disappeared upon ML-9 treatment and recovered after drug removal. Serum starvation was also observed. White arrowheads show vinculin-containing focal contacts, which were decreased by ML-9 and recovered by replacement with DMEM. Bar, bottom right, indicates 10 mm.

ML-9 treatment, deterioration of focal adhesions was evident. These cytoskeletal changes were reversible, and completely recovered by 15 hr after removal of ML-9 (Fig. 5(B)).

4. Discussion The present study demonstrated that a speci®c inhibitor of MLCK, ML-9, when administered either intracamerally or intravitreally, induces a signi®cant decrease in IOP in rabbit eyes. Our out¯ow facility data suggest that ML-9 elicits its pharmacological modulation, at least in part, via regulation of the conventional out¯ow, the main out¯ow route in human eyes that is believed to be regulated by the cellular behavior

and contractility of TM cells (Epstein and Rohen, 1991). Since the vascular anatomy of the out¯ow pathways and orbit in rabbit differs from that of the primates (Poyer et al., 1992), there is also an alternative that the signi®cant IOP-lowering effect of ML-9 in rabbit eyes may be related to changes in the permeability of vasculature in the anterior chamber. However, Tian et al. (2000a) have reported that ML-7, another MLCK inhibitor, increased the out¯ow facility in primates. The data of the two MLCK inhibitors support the notion that inhibition of MLCK seems to target the conventional out¯ow to a large extent. Morphologic change is also documented. After ML-9 treatment, TM cells time- and dose-dependently retracted and dissociated. Disruption of actin micro-

OCULAR EFFECTS OF MLCK INHIBITOR

®lament bundles and impairment of focal adhesion formation, indicative of perturbation of cellular anchorage and overall architectures, were in addition observed. These data establish that inhibition of MLCK causes disturbance in actin cytoskeleton, affecting consequently the actin-dependent intercellular adherens junctions and cell-matrix focal adhesions, as shown in Fig. 5. Such effects are in line with the recognized role of MLCK in mediating actin organization such as cellular relaxation of actin ®lament and then cellular disorganization (Saito et al., 1998; Samizo et al., 1999). The morphologic changes observed in TM cells with ML-9 are also similar to those reported with cytoskeletal drugs (Epstein et al., 1999; Cai et al., 2000; Tian et al., 2000b; Honjo et al., 2001a,b). However, cytoskeletal drugs such as cytochalasin, latrunculins and ethacrynic acid, which increase out¯ow facility act directly on the cytoskeleton by different mechanisms without involvement of kinases or signal transduction pathways (O'Brien et al., 1997). ML-9 has profound impact on cultured human TM cells. Immunoblot analysis revealed that MLCK is present in cultured human TM cells, and MLC is phosphorylated in normal TM cells with serum stimulation. Treatment with ML-9, as shown in Fig. 3, inhibited the phosphorylation of MLC in a dose-dependent manner. MLCK has two different regulatory roles. Due to its actin-binding activity, MLCK can inhibit the actin±myosin interaction without changing the level of MLC phosphorylation (Kohama et al., 1992). It has also been shown to inhibit MLC phosphorylation directly via interaction of the catalytic domain. In previous studies, ML-9 is reported to act by the second mechanism, interacting with the catalytic domain of MLCK to inhibit MLC phosphorylation (Okagaki et al., 1999; Samizo et al., 1999). The current study reiterates the inhibitory effect of ML-9 against MLC phosphorylation. The same dose-effect relationship was found between the inhibition of MLC phosphorylation and the perturbation of actin cytoskeleton in cultured TM cells. The kinetics of such alterations in cultured cells also paralleled that of the IOP and out¯ow facility changes observed in rabbit eyes after administration of 10±1000 mM of ML-9. These correlations may support the conclusion that inhibiting the MLC phosphorylation by ML-9 abrogates the actin cytoskeleton in TM cells, alters cell morphology, and in turn, has an impact on the out¯ow facility. TM cells on the corneoscleral trabecular beams and in the juxtacanalicular region play a critical role in overall meshwork architecture and the conductance of aqueous humor (Epstein and Rohen, 1991). Quite a few investigations have demonstrated that alteration in the contractility and cellular behaviors of TM cells can affect the IOP and the aqueous out¯ow (Epstein et al., 1999; Cai et al., 2000; Tian et al., 2000a,b; Honjo et al., 2001a,b). In our previous study,

141

inhibition of Rho/ROCK signaling pathways effectively lowered the IOP possibly through an effect on the actin cytoskeleton by interfering with the actomyosin system (Nobes and Hall, 1994; Honjo et al., 2001a,b). Results in the present study with MLCK inhibitor ML9 are similar to those of a speci®c ROCK inhibitor, Y27632 (Honjo et al., 2001a,b). The cytoskeletal effects and the relaxation of the cells produced by both inhibitors are most likely due secondarily to the regulation of actomyosin interaction. Nevertheless, MLCK is reported to be involved in micro®lament assembly in the periphery, while ROCK is involved in micro®lament assembly in the center of the cells, indicating that these two kinases may have distinct roles in spatial regulation of the MLC phosphorylation (Totsukawa et al., 2000). Previous studies have identi®ed that modulation of myosin light chain phosphorylation can cause a secondary loss of actin organizational structure, and may have a potential regulatory role in out¯ow function. Rao et al. (2001) have shown that ROCK inhibitor, Y-27632 treatment decreased myosin light chain phosphorylation and actomyosin organization in TM cells and Schlemm's canal cells, thereby lowering resistance to out¯ow. It has been shown that phosphorylation of the myosinbinding subunit (MBS) of myosin phosphatase was observed in the contracted TM or ciliary muscle, and the phosphorylation was well correlated with contraction of TM or ciliary muscle (Fukiage et al., 2001). The result in the present study is in good accordance with these previous studies and it is reasonable to conclude that cellular relaxation and loss of cellsubstratum adhesions in TM cells via modulation of myosin light chain phosphorylation could result in alteration of resistance to out¯ow. In summary, the present study demonstrates that ML-9, a selective MLCK inhibitor, reduces IOP and increases the out¯ow facility. Such effects may be related to altered cellular behavior of TM cells. Inhibition of MLC phosphorylation may be developed into a novel strategy for the treatment of glaucoma. Acknowledgements This study was supported in part by a Grant-in-Aid for Scienti®c Research from the Ministry of Education, Science, Sports and Culture, Japan from the Ministry of Health and Welfare, Japan; and grants EY 05628 and EY 01792 from the National Eye Institute, Bethesda, MD, U.S.A. M. H. is recipient of a Fellowship of the Japan Society for the Promotion of Sciences for Young Scientists.

References Barany, E. H. (1964). Simultaneous measurement of changing intraocular pressure and out¯ow facility in vervet monkeys by constant pressure perfusion. Invest. Ophthalmol. 2, 135±43. Cai, S., Liu, X., Glasser, A., Volberg, T., Filla, M., Geiger, B., Polansky, J. R. and Kaufman, P. L. (2000). Effect of latrunculin-A on morphology and actin-associated

142

adhesions of cultured human trabecular meshwork cells. Mol. Vis. 6, 132±43. Chrzanowska-Wodnicka, M. and Burridge, K. (1996). Rhostimulated contractility drives the formation of stress ®bers and focal adhesions. J. Cell Biol. 133, 1403±15. Epstein, D. L. and Rohen, J. W. (1991). Morphology of the trabecular meshwork and inner-wall endothelium after cationized ferritin perfusion in the monkey eye. Invest. Ophthalmol. Vis. Sci. 32, 160±71. Epstein, D. L., Rowlette, L. L. and Roberts, B. C. (1999). Acto-myosin drug effects and aqueous out¯ow function. Invest. Ophthalmol. Vis. Sci. 40, 74±81. Fukiage, C., Mizutani, K., Kawamoto, Y., Azuma, M. and Shearer, T. R. (2001). Involvement of phosphorylation of myosin phosphatase by ROCK in trabecular meshwork and ciliary muscle contraction. Biochem. Biophys. Res. Commun. 288, 296±300. Hartshorne, D. J., Ito, M. and Erdodi, F. (1998). Myosin light chain phosphatase: subunit composition, interactions and regulation. J. Muscle Res. Cell Motil. 19, 325±41. Honjo, M., Inatani, M., Kido, N., Sawamura, T., Yue, B. Y., Honda, Y. and Tanihara, H. (2001a). Effects of protein kinase inhibitor, HA1077, on intraocular pressure and out¯ow facility in rabbit eyes. Arch. Ophthalmol. 119, 1171±8. Honjo, M., Tanihara, H., Inatani, M., Kido, N., Sawamura, T., Yue, B. Y., Narumiya, S. and Honda, Y. (2001b). Effects of rho-associated protein kinase inhibitor Y-27632 on intraocular pressure and out¯ow facility. Invest. Ophthalmol. Vis. Sci. 42, 137±44. Kamm, K. E. and Stull, J. T. (1985). The function of myosin and myosin light chain kinase phosphorylation in smooth muscle. Ann. Rev. Pharmacol. Toxicol. 25, 593±620. Kawano, Y., Fukata, Y., Oshiro, N., Amano, M., Nakamura, T., Ito, M., Matsumura, F., Inagaki, M. and Kaibuchi, K. (1999). Phosphorylation of myosin-binding subunit (MBS) of myosin phosphatase by Rho-kinase in vivo. J. Cell Biol. 147, 1023±38. Kimura, K., Ito, M., Amano, M., Chihara, K., Fukata, Y., Nakafuku, M., Yamamori, B., Feng, J., Nakano, T., Okawa, K., Iwamatsu, A. and Kaibuchi, K. (1996). Regulation of myosin phosphatase by Rho and Rhoassociated kinase (Rho-kinase) [see comments]. Science 273, 245±8. Kohama, K., Okagaki, T., Hayakawa, K., Lin, Y., Ishikawa, R., Shimmen, T. and Inoue, A. (1992). A novel regulatory effect of myosin light chain kinase from smooth muscle on the ATP-dependent interaction between actin and myosin. Biochem. Biophys. Res. Commun. 184, 1204±11. Mobley, P. L., Hedberg, K., Bonin, L., Chen, B. and Grif®th, O. H. (1994). Decreased phosphorylation of four 20-kDa proteins precedes staurosporine-induced disruption of the actin/myosin cytoskeleton in rat astrocytes. Exp. Cell Res. 214, 55±66. Moussavi, R. S., Kelley, C. A. and Adelstein, R. S. (1993). Phosphorylation of vertebrate nonmuscle and smooth muscle myosin heavy chains and light chains. Mol. Cell. Biochem. 128, 219±27. Nobes, C. and Hall, A. (1994). Regulation and function of the Rho subfamily of small GTPases. Curr. Opin. Genet. Dev. 4, 77±81. O'Brien, E. T., Kinch, M., Harding, T. W. and Epstein, D. L. (1997). A mechanism for trabecular meshwork cell

M. HONJO ET AL.

retraction: ethacrynic acid initiates the dephosphorylation of focal adhesion proteins. Exp. Eye Res. 65, 471±83. Okagaki, T., Ye, L. H., Samizo, K., Tanaka, T. and Kohama, K. (1999). Inhibitory effect of the catalytic domain of myosin light chain kinase on actin±myosin interaction: insight into the mode of inhibition. J. Biochem. (Tokyo) 125, 1055±60. Poyer, J. F., Gabelt, B. and Kaufman, P. L. (1992). The effect of topical PGF2 alpha on uveoscleral out¯ow and out¯ow facility in the rabbit eye. Exp. Eye Res. 54, 277±83. Rao, P. V., Deng, P. F., Kumar, J. and Epstein, D. L. (2001). Modulation of aqueous humor out¯ow facility by the Rho kinase-speci®c inhibitor Y-27632. Invest. Ophthalmol. Vis. Sci. 42, 1029±37. Rudiger, M. (1998). Vinculin and alpha-catenin: shared and unique functions in adherens junctions. BioEssays 20, 733±40. Saito, H., Minamiya, Y., Kitamura, M., Saito, S., Enomoto, K., Terada, K. and Ogawa, J. (1998). Endothelial myosin light chain kinase regulates neutrophil migration across human umbilical vein endothelial cell monolayer. J. Immunol. 161, 1533±40. Samizo, K., Okagaki, T. and Kohama, K. (1999). Inhibitory effect of phosphorylated myosin light chain kinase on the ATP-dependent actin±myosin interaction. Biochem. Biophys. Res. Commun. 261, 95±9. Sawaguchi, S., Yue, B. Y., Chang, I. L., Wong, F. and Higginbotham, E. J. (1992). Ascorbic acid modulates collagen type I gene expression by cells from an eye tissue±trabecular meshwork. Cell. Mol. Biol. 38, 587±604. Schevzov, G., Lloyd, C. and Gunning, P. (1995). Impact of altered actin gene expression on vinculin, talin, cell spreading, and motility. DNA Cell Biol. 14, 689±700. Somlyo, A. P. and Somlyo, A. V. (1994). Signal transduction and regulation in smooth muscle [published erratum appears in Nature 1994 Dec 22±29;372(6508):812]. Nature 372, 231±6. Taniguchi, T., Haque, M. S. R., Sugiyama, K., Hori, N. and Kitazawa, Y. (1996). Ocular hypotensive mechanism of topical isopropyl unoprostone, a novel prostaglandin metabolite-regulated drug, in rabbits. J. Ocul. Pharmacol. Ther. 4, 489±98. Tian, B., Brumback, L. C. and Kaufman, P. L. (2000a). ML-7, chelerythrine and phorbol ester increase out¯ow facility in the monkey eye. Exp. Eye Res. 71, 551±66. Tian, B., Geiger, B., Epstein, D. L. and Kaufman, P. L. (2000b). Cytoskeletal involvement in the regulation of aqueous humor ¯ow. Invest. Ophthalmol. Vis. Sci. 41, 619±23. Totsukawa, G., Yamakita, Y., Yamashiro, S., Hartshorne, D. J., Sasaki, Y. and Matsumura, F. (2000). Distinct roles of ROCK (Rho-kinase) and MLCK in spatial regulation of MLC phosphorylation for assembly of stress ®bers and focal adhesions in 3T3 ®broblasts. J. Cell Biol. 150, 797±806. Yue, B. Y., Lin, C. C., Fei, P. F. and Tso, M. O. (1984). Effects of chondroitin sulfate on metabolism of trabecular meshwork. Exp. Eye Res. 38, 35±44. Zhao, Y., Rhoades, R. A. and Packer, C. S. (1996). Hypoxiainduced pulmonary arterial contraction appears to be dependent on myosin light chain phosphorylation. Am. J. Physiol. 271, L768±L774.