Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

A decrease in alveolar O₂ tension (PA_O2) causes hypoxia-induced pulmonary vasoconstriction (HPV). The HPV response is a unique regulatory mechanism in the pulmonary circulation that functions to balance perfusion with ventilation. The HPV response has been documented in several animal models, isolated perfused lung preparations, and isolated arterial vessels (1). Hypoxia-induced pulmonary artery contraction is characterized by a slow sustained contractile response that is more pronounced in small resistant vessels than in the larger conduit pulmonary vessels (3). Importantly, hypoxia has been shown to cause myosin light chain (MLC) phosphorylation and shortening of pulmonary arterial smooth muscle cells (PASMCs) (5, 6), indicating the existence of a complete signaling mechanism within PASMCs which senses O₂ changes and transduces the hypoxic signal to smooth muscle contraction. Despite extensive investigation, the exact cellular mechanisms that induce HPV and the mechanisms involved in the sustained phase of HPV are still not well defined. Inhibition of voltage gated potassium ion (K_v) channels and increase of intracellular Ca²⁺ concentration ([Ca²⁺]_i) have been shown to play an important role in the initiation of HPV (7). The increase in intracellular Ca²⁺ has been shown to be linked to hypoxic MLC phosphorylation through the Ca²⁺, calmodulin, MLC-kinase (MLCK) pathway (4, 11, 24). However, the time course of the [Ca²⁺]_i increase does not match with the continuous force development in the sustained phase of HPV (10). Madden and colleagues, Goldman and coworkers, and Zhang and associates reported that, in response to hypoxia, the [Ca²⁺]_i in PASMCs rapidly increases to a peak, then quickly decreases to a constant lower level (11). Robertson and colleagues reported a similar pattern of [Ca²⁺]_i change in intact isolated pulmonary artery (10). Together, these observations suggest that K_v channels and increase of [Ca²⁺]_i play an important role in the initiation of HPV, but other mechanisms are required in the sustained phase of HPV.

The 160 kD Rho-associated serine/threonine kinase (Rho-kinase) has recently been demonstrated to increase Ca²⁺ sensitivity and thus play an important role in smooth muscle contraction. Rho-kinase is activated by binding to active, GTP-bound Rho A (14, 15). Kureishi and coworkers applied recombinant active Rho-kinase to a permeablized portal vein at a constant [Ca²⁺]_i level and found that it directly induced vessel contraction (16). Y-27632, a relatively specific Rho-kinase inhibitor, was demonstrated to inhibit contractions that were induced by diverse stimulators in vascular smooth muscle preparations (17). Rho-kinase also increases MLC phosphorylation level in the cell. One of the pathways is that Rho-kinase phosphorylates the myosin-binding subunit of myosin phosphatase. When phosphorylated by Rho-kinase, the phosphatase has a decreased activity and results in increased MLC phosphorylation levels (18, 19). Other investigations have also shown that Rho-kinase directly phosphorylates MLC at Ser-19 (20, 21). Since Rho-kinase induces MLC phosphorylation, independent of the change in [Ca²⁺]_i, we hypothesized that Rho and Rho-kinase activation play an important role in the sustained phase of HPV. It was recently reported that Y-27632 attenuates hypoxia-induced pulmonary contraction in isolated rat lung and isolated rat pulmonary artery, indicating that Rho-kinase is required for HPV (22, 23). In the present study, experiments were performed on cultured PASMCs and isolated rat pulmonary artery to elucidate, from a cellular mechanistic aspect, the role of the Rho and Rho kinase pathway in HPV. Our results show that hypoxia is sufficient to activate Rho-kinase in PASMCs independent of non-muscle cells; both Rho and Rho-kinase are required in hypoxia-induced sustained MLC phosphorylation in PASMCs.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Anti-MLC antibody and anti-smooth muscle -actin antibody were purchased from Sigma (St. Louis, MO). Anti-Rho-kinase antibody was purchased from Transduction Laboratory (San Diego, CA). Exoenzyme C3 was purchased from List (Campbell, CA). [-³²P]ATP was purchased from Dupont-New England Nuclear (Boston, MA). Protein A/G conjugated beads were purchased from Santa Cruz (Santa Cruz, CA). Lipofectamine was purchased from Life Technologies (Gaithersburg, MD). Y-27632 [(+)-(R)-trans-4-(1-aminoethyl)-N-(4-pyridyl) cyclohexanecarboxamide dihydrochloride] was a kind gift from Yoshitomi Pharmaceutical Ind., Ltd., Osaka, Japan. Clostidium diffile Toxin B was kindly provided by Dr. S. Ganguli (Indiana State University, Terre Haute, IN). MLC was induced and purified from a recombinant MLC construct kindly provided by Dr. P. Gallagher (Indiana University). Other chemicals and cell culture medium were purchased from Sigma.

Pulmonary Arterial Ring Study

The arterial ring preparation was performed according to a method reported by Yuan and associates (3) with some modification. Male Sprague-Dawley rats (200-300 g) were anesthetized with pentobarbital sodium intraperitonially (50 mg/kg body weight) and exsanguinated by cutting the abdominal aorta. Heart and lungs were removed en bloc from the thoracic cavity and placed in warm (37°C) modified Kreb's solution (138 mM NaCl, 4.7 mM KCl, 1.2 mM NaH₂PO₄. 1.2 mM MgSO₂, 5 mM HEPES, 1.8 mM CaCl₂, 10 mM glucose, pH 7.4, gassed with air). Pulmonary arterial branches were cleaned of all visible parenchyma. Fourth-branch intrapulmonary arteries with outer diameter ~ 0.5 mm were isolated for contractile studies. To record isometric force, each ring was cut into segments (2.5 to 3.0 mm in length) and mounted onto two horizontally oriented surgical steel wires in a standard muscle bath (filled with 10 ml Krebs solution, 37°C, gassed initially with air and switched to 100% N₂ when hypoxia treatment starts). The apparatus was connected with a force transducer (Model FT03C; Grass, Quincy, MA). Each arterial ring was equilibrated for 1 h at a mean resting tension of 0.3 g. For active tension development each vessel was contracted with 40 mM KCl. KCl was washed out thoroughly to allow the tension to return to resting level. At the level of resting tension, hypoxia stimulation was initiated by switching the solution to 100% nitrogen pre-bubbled solution, and hypoxia was maintained by continuously bubbling the superfusion solution with 100% nitrogen.

PASMC Culture and Hypoxia Treatment

Cultured smooth muscle cells from intrapulmonary arteries of the Sprague-Dawley rats were prepared using an explant method. The intrapulmonary arteries were excised and cleaned under sterile conditions. The endothelium of pulmonary arteries was removed by a scalpel. The vessels were cut into 2-mm pieces and incubated in Dulbecco's modified Eagle's medium (DMEM) supplemented with 20% fetal bovine serum (FBS), 100 U/ml penicillin, 0.1 mg/ml streptomycin, and 0.25 µg/ml amphotericin B, in a humidified atmosphere with 5%CO₂-95% air at 37°C. After 7 d, the tissue explants were removed and fresh DMEM containing 20% FBS was added until the cells reached confluence. Cells were subcultured after detachment with 0.05% trypsin in phosphate-buffered saline (PBS) and maintained in DMEM media containing 10% FBS. The media were changed every 48 h and cells were subcultured once weekly. Experiments were performed on cells from passages 2 through 5 after 24 h of serum starvation. To test the purity of smooth muscle cells in the cell culture, a sample was simultaneously stained with anti-smooth muscle -actin antibody and 4,6-diamidino-2-phenylindole (DAPI), for smooth muscle -actin and the nucleus, respectively. The purity was determined by viewing the staining with a fluorescence microscope. Studies were performed on smooth muscle cells with purity > 90%. Cells were made hypoxic by adding 5%CO₂-95%N₂ pre-gassed medium. Cells were then placed into a hypoxic incubator gassed with 5%CO₂-95%N_2. The O₂ tension (PO₂) in the culture media was maintained at ~ 25-30 mm Hg. Following hypoxia, cells were quickly frozen in liquid nitrogen.

Measurement of MLC Phosphorylation

The level of phosphorylation of MLC was measured by urea-glycerol gel electrophoresis and Western blot (24). Ice-cold 10% trichloroacetic acid (TCA), 10 mM dithiothreitol (DTT) was added to the frozen cells to denature the proteins. Pellets were harvested, washed with diethyl ether and dissolved in 6M urea, 20 mM Tris, 22 mM glycine, 10 mM DTT. Samples were applied to a 10% acrylamide gel with 40% glycerol for electrophoresis. Western blots were performed and MLC was detected with a monoclonal anti-MLC antibody. The amount of unphosphorylated MLC and phosphorylated MLC was determined by scanning densitometry (GS-670; Bio-Rad, Hercules, CA). The level of phosphorylated MLC was expressed as a percentage of phosphorylated MLC over total MLC (total MLC = phosphorylated MLC + unphosphorylated MLC).

Immunoprecipitation and Rho-Kinase Activity Assay

Confluent cultured PASMCs were treated with lysis buffer (1%Triton X-100, 1 mM EDTA, 1 mM EGTA, 2 µg/ml leupeptin, 2 µg/ml aprotinin, 2 µg/ml pepstatin in PBS). Cells were homogenized by passing the lysate through a 25G needle and cell debris was removed by centrifuging the solution at 16,000 × g for 10 min. Fifty microliters of pre-washed protein A/G conjugated beads were charged with 4 µg of anti-Rho-kinase antibody by mixing the two for 2 h at 4°C. Cell lysate containing 5 µg of protein was added to the charged beads, which were incubated for another 2 h at 4°C. After washing six times, the pellet was suspended in the Rho-kinase assay buffer (40 mM Tris-HCl, pH 7.5, 2 mM EDTA, 1 mM DTT, 6.5 mM MgCl₂, 0.1% CHOAPS, 2 µg/ml leupeptin, 2 µg/ml pepstatin) for the kinase assay. The Rho-kinase assay is a modification of the method described by Amano and colleagues (25). The reactions were performed in 50 µl of Rho-kinase assay buffer with 10 µM ATP, 1.25 µCi [-³²P]ATP, 4 µg MLC, and the immunoprecipitation complex. After 10 min at 30°C the reaction was terminated by adding 4 × sodium dodecyl sulfate sample buffer to the reaction mixture. The kinase activity was determined by autoradiographic analysis.

Statistical Treatment of Data

Student's t test was used to compare results from control and experimental groups. All results were expressed as mean values ± standard error, and P < 0.05 indicated mean values were significantly different.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Rho-Kinase Activity Increases in Response to Hypoxia

To directly demonstrate whether hypoxia activates Rho-kinase in PASMCs, the Rho-kinase activity was measured using MLC as the substrate. Confluent cultured PASMCs were stimulated with hypoxia for 0, 20, 40, 60, 90, and 120 min. Cell lysates were immunoprecipitated with an anti- Rho-kinase antibody. The immunocomplex was then used for the kinase assay with purified MLC as the substrate. Addition of Y-27632 significantly inhibited the kinase activity in the immunocomplex in a concentration-dependent manner, but ML-9, a MLCK inhibitor, was not found to inhibit the kinase activity (data not shown). As shown in Figures 1A and 1B, hypoxia increased Rho-kinase activity increased at 40 and 60 min by 19.7 ± 0.85% and 42.6 ± 14.0%, respectively (n = 4, P < 0.05). The level of Rho- kinase protein did not significantly change during 120 min of hypoxia stimulation (n = 3) (Figure 1C).

View larger version (33K): [in this window] [in a new window]   Figure 1.   Effect of hypoxia on Rho-kinase activity. (A) A representative experiment showing the change of Rho-kinase activity in PASMCs exposed to hypoxia for different lengths of time, using MLC as substrate. The upper panel shows autoradiographic signals. The lower panel is from Coomassie blue staining of the same gel. (B) Densitometric analysis of the autoradiographic signals normalized with the value from Coomassie blue staining. Data are expressed as means ± SE (n = 4). *P < 0.05. (C) A Western blot showing the level of Rho-kinase protein under hypoxia stimulation for 0, 20, 40, 60, 90, and 120 min.

Effect of Rho-Kinase Inhibitor on Hypoxia-Induced MLC Phosphorylation in PASMCs

The time course of hypoxia-induced MLC phosphorylation in PASMC is shown in Figure 2. The level of MLC phosphorylation increased as early as 10 min after the cells were exposed to hypoxia and the high level of phosphorylation was sustained during the 120-min hypoxia exposure period. The magnitude of MLC phosphorylation level after hypoxia increased by ~ 2 fold from the basal normoxia level of 39.1 ± 2.03% (n = 6).

View larger version (13K): [in this window] [in a new window]   Figure 2.   Time course of hypoxia-induced MLC phosphorylation in PASMCs. PASMCs were exposed to hypoxia for 10, 20, 40, 60, 90, or 120 min. The phosphorylated MLC (MLC-P) and the unphosphorylated MLC (MLC) were separated by urea-glycerol gel electrophoresis and detected by Western blot (n = 6). The data are expressed as a percentage (± SE) of MLC phosphorylation, which was determined by densitometry.

To determine if Rho-kinase is involved in hypoxia-induced MLC phosphorylation, cultured PASMCs were preincubated with 0, 0.1, 1, 5, and 10 µM of Y-27632 for 45 min before being exposed to hypoxia for 60 min. As shown in Figures 3A and 3B, Y-27632 inhibited hypoxia-induced MLC phosphorylation in a concentration-dependent manner (n = 3). In another set of experiments, preincubation with 5 µM of Y-27632 significantly attenuated the effect of hypoxia in increasing MLC phosphorylation (Figures 3C and D). There was also a slight effect on the basal level of MLC phosphorylation (n = 3) (Figures 3C and D), suggesting that Rho-kinase may play a role in maintaining the basal level of MLC phosphorylation. Taken together, these results show that Rho-kinase is involved in hypoxia-induced MLC phosphorylation.

View larger version (16K): [in this window] [in a new window]   Figure 3.   Effect of Rho-kinase inhibitor (Y-27632) on hypoxic MLC phosphorylation in PASMCs. (A) A Western blot demonstrating the concentration-dependent effect of Y-27632 on hypoxic MLC phosphorylation. PASMCs were pre-treated with 0 (lane 1), 0.1 (lane 2), 1 (lane 3), 5 (lane 4), and 10 (lane 5) µM of Y-27632 for 45 min followed by 60 min of hypoxia. (B). Mean data of the effect of Y-27632 on hypoxic MLC phosphorylation level (n = 3). (C) A representative Western blot showing the effect of 5 µM of Y-27632 on hypoxic MLC phosphorylation in PASMCs. Cells were treated with hypoxia (lanes 2 and 4) in the presence (lane 4) and absence (lane 2) of Y-27632; cells incubated under normoxia conditions (lanes 1 and 3) with (lane 3) and without (lane 1) Y-27632 pre-treatment served as control. (D) Mean data from three independent experiments described in C. Open bars represent experiments without Y-27632 treatment. Solid bars represent experiments with Y-27632 treatment. Data are expressed as means ± SE. *P < 0.05.

Rho Is Involved in Hypoxia-Induced MLC Phosphorylation in PASMCs

Because Rho-kinase is activated by binding to active RhoA, we hypothesized that hypoxia activates the Rho/Rho-kinase pathway with Rho as the upstream signaling molecule. To test this hypothesis, toxin B, a Rho protein inhibitor, and exoenzyme C3, a specific Rho inhibitor, were used. Cells were pretreated with 10 µg/ml exoenzyme C3 plus 10 µl/ml lipofectamine in serum-free medium for 60 min, followed by 60 min of hypoxia exposure. As shown in Figure 4A, C3 significantly inhibited hypoxia-induced MLC phosphorylation. Lipofectamine slightly increased MLC phosphorylation level in both normoxic and hypoxic conditions (n = 3).

View larger version (21K): [in this window] [in a new window]   Figure 4.   Effect of the Rho protein inhibitor (toxin B) and the specific RhoA inhibitor (exoenzyme C3) on hypoxic MLC phosphorylation in PASMCs. (A) The inhibitory effect of 10 µg/ml C3 on hypoxia-induced MLC phosphorylation in PASMCs. Cells were treated with 10 µl/ml lipofectamine, a vector to facilitate the delivery of C3 into the cells, plus 10 µg/ml exoenzyme C3 for 60 min followed by 60 min of hypoxia (n = 3). (B) Mean data of the concentration-dependent effect of toxin B on hypoxic MLC phosphorylation. PASMCs were treated with different concentrations of toxin B for 10 min followed by 60 min of hypoxia (n = 3). Data are expressed as means ± SE. *P < 0.05.

To test the effect of toxin B, PASMCs were preincubated with 1:50,000, 1:10,000, 1:5,000, and 1:1,000 dilution of purified toxin B (the concentration of the purified toxin B is 72 ng/µl) for 10 min and then treated with hypoxia for 60 min. The stock solution of toxin B was dissolved in ~ 300 mM NaCl solution. Toxin B was diluted 1,000 to 10,000 times in cell culture medium when applied to the cells. The volume of toxin B solution added to the cell culture medium was maintained in the range of 10-20 µl. Hypoxia-treated cells in the absence of toxin B served as the control. As shown in Figure 4B, toxin B inhibited hypoxia-induced MLC phosphorylation in a concentration-dependent manner (n = 3). These data show that Rho is activated and plays a role in hypoxia-induced MLC phosphorylation in cultured PASMCs.

RhoA Activation Is Required for Hypoxia-Induced Rho-Kinase Activation in PASMCs

The following experiments were designed to demonstrate that hypoxia-induced Rho-kinase activation requires RhoA. Cultured PASMCs were pretreated with 10 µg/ml of C3 plus 10 µl/ml lipofectamine in serum free medium for 60 min, followed by 60 min of hypoxia exposure. As shown in Figure 5A, hypoxia-induced Rho-kinase activation was attenuated by C3. Lipofectamine did not significantly influence Rho-kinase activity (n = 3).

View larger version (16K): [in this window] [in a new window]   Figure 5.   Effect of exoenzyme C3 and toxin B on Rho-kinase activity in PASMCs. (A) Cells were treated with hypoxia for 60 min with (solid bar) or without (open bar) pretreatment with 10 µg/ml C3; cells incubated under normoxic conditions with and without C3 served as the control. Rho-kinase activity was measured as described in MATERIALS AND METHODS (n = 3). (B) Cells were treated with hypoxia for 60 min with (solid bar) or without (open bar) 1:5,000 dilution of toxin B; cells incubated under normoxic conditions with and without toxin B served as the control. Rho-kinase activity was measured as described in MATERIALS AND METHODS (n = 3). Data are expressed as means ± SE. *P < 0.05 for the comparison of Rho-kinase activity between normoxia and hypoxia, without toxin B or C3 treatment. +P < 0.05 for the comparison of Rho-kinase activity between hypoxic samples with and without toxin B or C3 treatment.

In another set of experiments, cultured PASMCs were pretreated with a 1:5,000 dilution of toxin B for 10 min and then subjected to 60 min of hypoxia. Cells treated with hypoxia in the absence of toxin B served as the control. As shown in Figure 5B, hypoxia increased Rho-kinase activity and toxin B inhibited Rho-kinase activation induced by hypoxia (n = 3). These results indicate that hypoxia activates RhoA, which subsequently activates Rho-kinase in PASMCs.

Rho-Kinase Is Involved in Hypoxia-Induced Pulmonary Arterial Vasoconstriction

Fourth-order branches of rat intrapulmonary pulmonary arterial rings (outer diameter ~ 400-500 µm) were found to contract in response to hypoxia without priming. As shown in Figure 6A, reduction of oxygen tension from 110 mm Hg to 35 mm Hg in the modified Krebs solution induced a slow but sustained contraction in the small pulmonary arteries from a basal tone of ~ 300 mg. The contraction reached a plateau ~ 90 min after the initiation of hypoxia. To determine if Rho-kinase is involved in HPV, Y-27632 with final cumulative concentrations of 0.1, 1, 5, and 10 µM was applied at the plateau of hypoxia-induced contraction. Y-27632 caused a concentration-dependent inhibition of the sustained phase of hypoxia-induced pulmonary arterial contraction (Figures 6B and 6C) (n = 4).

View larger version (9K): [in this window] [in a new window]   Figure 6.   Effect of Y-27632 on hypoxia-induced contraction in rat intrapulmonary artery. (A) A typical isometric trace recorded from the fourth-order branch of rat intrapulmonary artery. The vessel was contracted with 40 mM KCl and then made hypoxic after KCl washed out. Hypoxia caused a slow sustained contraction. (B) A typical relaxation response with Y-27632. At the plateau of HPV, cumulative concentrations of 0.1 (a), 1 (b), 5 (c), and 10 (d) µM of Y-27632 were applied, with an interval of 10 min. (C) Summary of the cumulative concentration-response from the addition of Y-27632 (n = 4). Data are expressed as means ± SE.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present study, we demonstrated that: (1) Rho-kinase is activated by hypoxia in cultured PASMCs; (2) the activation of Rho-kinase is required in hypoxia-induced MLC phosphorylation in PASMCs; (3) hypoxia-induced Rho- kinase activation in PASMCs requires the activation of Rho; and (4) Rho-kinase activation is required in the sustained phase of hypoxia-induced pulmonary vasoconstriction.

HPV is a contractile response to decreased alveolar O₂ tension in the resistance arteries in the pulmonary circulation. Cultured PASMCs and small intrapulmonary arteries have been used previously to study HPV. In the present study, a decrease in O₂ tension from 110 mm Hg to 30 mm Hg induced contraction in intrapulmonary arteries (Figure 6), consistent with the observation of Yuan and colleagues (3). The data also show for the first time that a decrease of O₂ tension induces a rapid and sustained elevation of MLC phosphorylation in cultured PASMCs (Figure 2). This observation supports our preliminary report with intact pulmonary arterial vessels (24). Sustained elevation of MLC phosphorylation suggests that other mechanisms, independent of a change in [Ca²⁺]_i,, are involved in MLC phosphorylation in hypoxia-treated PASMCs. In particular, the hypoxia-induced activation of Rho-kinase in PASMCs, in the absence of other cell types, lends strong support to the idea that PASMCs have an intrinsic mechanism and a complete signaling pathway to maintain sustained contraction during chronic hypoxia.

The present investigation focused on the role of the newly defined Rho/Rho-kinase pathway in the mechanism of HPV. RhoA and Rho-kinase have been reported to be expressed in the lungs, shown by Western blot (15), and in pulmonary artery, shown by reverse transcriptase-polymerase chain reaction (26). Recently, Yamagata and coworkers reported that phenylephrine-induced pulmonary arterial contraction was profoundly inhibited by the relatively specific Rho-kinase inhibitor, Y-27632 (27). This information shows that the Rho/Rho-kinase pathway exists and plays an important role in the regulation of smooth muscle contraction in pulmonary artery.

An important observation seen in the present study is that hypoxia significantly increases Rho-kinase activity in serum-deprived PASMCs (Figure 1). Smooth muscle cells incubated in the presence of serum are in a secretory phenotype. Removing the serum converts PASMCs to their contractile function by transforming them from a secretory to a contractile phenotype (44, 45). The 24-h serum deprivation used in this study provides a good model for the investigation of Rho/Rho-kinase pathway in PASMc (18, 25). The Rho-kinase activity assay used in the present study was a relatively specific method, because in vitro addition of Y-27632, a specific Rho-kinase inhibitor, significantly inhibited the kinase activity in the immunocomplex in a concentration-dependent manner (data not shown). In contrast, ML-9, a MLCK inhibitor, was not found to inhibit the kinase activity (data not shown). A similar assay using an immunocomplex to measure Rho-kinase activity has also been reported in canine basilar artery (28). The activation of Rho-kinase induced by hypoxia is indeed required by hypoxic pulmonary arterial contraction, since Y-27632 relaxed hypoxia-induced intrapulmonary arterial contraction (Figure 6), and Y-27632 also inhibited hypoxia-induced increase of MLC phosphorylation in PASMCs (Figure 3). Y-27632 is a relatively specific Rho-kinase inhibitor that is widely used to study the role of Rho-kinase. In the range of 1 to 10 µM, Y-27632 selectively inhibits Rho-kinase, but has little effect on MLCK, the other important kinase that phosphorylates MLC and leads to smooth muscle contraction. The K_i of Y-27632 to Rho- kinase is 0.14 µM, while the K_i to MLCK is > 250 µM (29). It has been reported that Y-27632 significantly relaxes Rho-kinase-mediated GTPS-induced contraction (29), but does not inhibit MLCK-mediated Ca²⁺-induced smooth muscle contraction (17). Together, the data in the present study show that hypoxia is sufficient to activate Rho-kinase and the activation of Rho-kinase is required in HPV. The data provide biochemical evidence for the recent observation made by our laboratory (22) and Robertson and associates (23) showing that Rho-kinase is linked to the sustained contraction in HPV. The Rho-kinase signaling pathway is thus strongly implicated as the mechanism, which together with the Ca²⁺/MLCK pathway, contributes to the sustained contraction and MLC phosphorylation seen in HPV. Ward and colleagues proposed that endothelium-dependent mechanism(s) contribute to the Ca²⁺ sensitization in PASMCs (30). However, our results in cultured PASMCs, with smooth muscle cell purity > 90%, suggest that hypoxia is sufficient to activate Rho-kinase, and this effect does not require endothelial cells.

We observed that the elevation of MLC phosphorylation occurs 10 min after hypoxia (Figure 2), while Rho-kinase activity significantly increases at 40 min after hypoxia (Figure 1). The results strongly suggest that Rho-kinase does not play a role in the initiation of MLC phosphorylation. Because [Ca²⁺]_i increases in seconds after hypoxia (12, 13), we postulate that Ca²⁺/MLCK plays a major role in the initiation of MLC phosphorylation, while the main function of Rho-kinase is to promote MLC phosphorylation in the sustained phase of HPV when [Ca²⁺]_i decreases. We have also preincubated small intrapulmonary arteries with 5 µM of Y-27632 for 15 min, and found that a contractile response could not be observed in response to hypoxia (data not shown). These preliminary data support the idea that a certain level of activation of MLCK is a prerequisite for Rho-kinase to increase MLC phosphorylation. In other words, the increase of MLC phosphorylation requires both the MLCK pathway and the Rho-kinase pathway to be functional; this is in agreement with the observations that HPV is inhibited by Ca²⁺ blockers (2, 4), ML-9 (24), and Y-27632.

Because the measurement of Rho-kinase involved an immunoprecipitation step, a potential question arises as to whether this procedure could interfere with the activity. The immunoprecipitation procedure could possibly modify the magnitude of peak activity. However, the procedure would not affect or shift the time in which peak activity occurs, and therefore would not modify the interpretation of the data.

Rho-kinase is activated by binding to active GTP-bound Rho. We utilized exoenzyme C3 and toxin B to determine if Rho is necessary in both hypoxic MLC phosphorylation and hypoxic Rho-kinase activation in PASMCs. Exoenzyme C3 specifically ADP-ribosylates Rho and prevents Rho from being recruited to the cell membrane (31). Garcia and colleagues used lipofectamine to deliver C3 into cultured endothelium cells and found that 5 µg/ml C3 significantly inhibited diperoxovanadate-induced MLC phosphorylation (32). Using the same strategy, our results show that 10 µg/ml C3 profoundly inhibited hypoxia-induced MLC phosphorylation (Figure 4A). C3 was found to attenuate hypoxia-induced Rho-kinase activation as well (Figure 5A). Toxin B is a cell membrane-permeable inhibitor of Rho proteins. Unlike exoenzyme C3 that requires a long incubation time, or requires lipofectamine delivery, toxin B enters cells through receptor-mediated phagocytosis and blocks Rho proteins after a brief incubation. Toxin B glycosylates Rho proteins and inhibits Rho through a different mechanism from exoenzyme C3 (31). Pfitzer and coworkers reported that toxin B significantly inhibited carbachol-induced tonic contraction and MLC phosphorylation in guinea-pig small intestine smooth muscle preparation, without changing [Ca²⁺]_i (33). Our results show that toxin B inhibited hypoxia-induced MLC phosphorylation in PASMCs (Figure 4B), indicating that Rho protein(s) is required in HPV. Toxin B was found to attenuate hypoxia-induced Rho-kinase activation as well (Figure 5B). These data indicate that hypoxia activates Rho and subsequently activates Rho-kinase, contributing to hypoxic MLC phosphorylation and HPV. Rho-kinase can directly phosphorylate MLC in vitro. However, a study done by Iizuka and associates recently demonstrated that direct phosphorylation of MLC by Rho-kinase does not play a role during smooth muscle contraction (46). In other words, Rho-kinase-related MLC phosphorylation is the result of primarily MLC phosphatase inhibition. In the present hypoxia study, we cannot exclude the role of direct Rho- kinase phosphorylations of MLC, but preliminary data from our laboratory suggest that the main mechanism is through myosin phosphatase inhibition. We have recently found that hypoxia causes phosphorylation of the myosin binding subunit (MBS) of myosin phosphatase in PASMCs (43).

The activity of Rho is primarily regulated via an unknown mechanism probably involving guanine nucleotide dissociation inhibitor (GDI[s])), guanine nucleotide exchange factor (GEF[s]), and GTPase activating protein (GAP[s]) (34). In aortic smooth muscle cells, heterotrimeric G-protein coupled receptors involving the G12/13 family are linked to agonist-induced Rho activation (35). Recently it was suggested that cGMP-dependent protein kinase also regulates Rho activity through phosphorylation of RhoA, thus inhibiting Ca²⁺-sensitization in vascular smooth muscle (36). Tyrosine phosphorylation, PKC, or Ca²⁺ were found to be required for the action of Rho under certain circumstances (33, 37). However, the mechanism by which hypoxia activates Rho in PASMCs is unknown. To our knowledge, the direct effect of hypoxia on the function of G-proteins, cGMP pathway, GDI, GAP, and GEF has not been investigated. Immediate consequences of hypoxia, such as reduced redox status change (38), reactive oxygen species production (39), K⁺ channel blockage (7, 8) and membrane depolarization (40), are characteristic responses in PASMCs. Whether any of these responses are linked to Rho activation, with or without the involvement of G-proteins, cGMP pathway, GDI, GAP, or GEF, also needs further investigation. HPV has been shown to be inhibited by tyrosine kinase inhibitors (41) and PKC inhibitors (42). Whether these effects are mediated through the inhibition of Rho activation will also require further investigation.

In summary, our study suggests that the Rho/Rho- kinase pathway plays an important role in hypoxia-induced pulmonary vasoconstriction. Our data support the idea that hypoxia activates multiple signaling pathways that function in a coordinate manner and lead to smooth muscle contraction. Data from this study may provide clues for future directions in understanding the mechanisms of HPV to develop more effective strategies in the intervention of pulmonary hypertension.