INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE LUNG IS A PRIMARY SITE in which oxygen radical generation is linked to inflammation, oxygen toxicity, acute hypertension (7, 9), and ischemia-reperfusion injury (15). Reactive oxygen species have been shown to cause increased pulmonary arterial pressure (5). Isolated pulmonary arterial rings develop active force when exposed to either xanthine oxidase or glucose oxidase, suggesting that H₂O₂ is the specific reactive oxygen species responsible for the contraction (13). In addition, the only reactive oxygen species scavenger found to completely abolish the contractile response is catalase, a specific scavenger of H₂O₂. Superoxide dismutase only reduces the response; mannitol and desferoxamine (a scavenger and an inhibitor of formation of hydroxyl radicals, respectively) have no effect. Therefore, H₂O₂ is implicated as causative of the contractions. H₂O₂ (>10⁴ M) has been reported to cause contractions of pulmonary arterial and venous muscle (13, 28, 29, 37). The cellular mechanism for the contractile response to H₂O₂ is not understood.

H₂O₂-induced contractions are independent of the endothelium and of extracellular calcium (13, 28). H₂O₂ is known to penetrate the cell (9) and could elicit a contraction by changing transmembrane Ca²⁺ flux (27). This is unlikely given the findings that verapamil or Ca²⁺-free and EGTA-containing media had no effect (13). Alternatively, H₂O₂ can stimulate inositol trisphosphate (IP₃) production (10), resulting in intracellular Ca²⁺ release and contraction. However, this is also unlikely, given that ryanodine had no effect on the response (13). Sarcoplasmic reticular Ca²⁺ pumps are known to be damaged by reactive oxygen species in cultured cells (8). However, Ca²⁺-ATPase activity is inhibited by superoxide anion but not by H₂O₂ (33). Given this information, it is probably no surprise that the ₁- and -receptor blockers, phentolamine and propranolol, respectively, also have no effect on the H₂O₂-mediated pulmonary arterial smooth muscle contraction. (It may be recalled that ₁-receptor activation of smooth muscle contraction is transduced primarily through the IP₃-intracellular Ca²⁺ release pathway).

A third possibility is that H₂O₂ stimulates protein kinase C (PKC), which is known to cause phosphorylation of the 20-kDa regulatory myosin light chain (MLC₂₀), although at different sites than those phosphorylated by myosin light chain kinase (MLCK); it may also activate mitogen-activated protein kinase (MAPK) leading to smooth muscle contraction without a significant increase in intracellular free Ca²⁺. Interestingly, 1-(5-isoquinolinylsulfonyl)-2-methylpiperazine (H-7, a PKC inhibitor) significantly reduces the contractile response (13). But one must keep in mind that PKC inhibitors are nonspecific. Indeed, staurosporine (another commonly utilized PKC inhibitor) is more potent in phosphorylating MLCK than in inhibiting PKC activity, and H-7 is also known to inhibit MLCK, as shown by Harold Davis as reported in Packer and Rhoades (24). Phosphorylation of MLCK may render it inactive (3), thus preventing muscle contraction by the currently accepted (and more usual) physiological mechanism of MLCK-catalyzed phosphorylation of myosin. The possibility exists that the H-7 reduction of the reactive oxygen species-induced contraction indicates that H₂O₂ is activating MLCK without raising intracellular Ca²⁺ or activating PKC. A change in the Ca²⁺ sensitivity of the regulatory and/or contractile proteins may be the key.

At least three, if not four, Ca²⁺ binding sites on calmodulin are occupied for the activation of the various calmodulin-dependent kinases (26). Activation associated with an increase in cytoplasmic Ca²⁺ concentration is the result of Ca²⁺ binding first to calmodulin with subsequent binding of the Ca- calmodulin complex to MLCK and consequent activation of MLCK. Phosphorylation of MLC₂₀ by the (Ca²⁺)₄-calmodulin-MLCK complex is the generally accepted mechanism thought to result in the activation of myosin Mg²⁺-ATPase by actin in smooth muscles. Therefore, the purposes of the current study were: 1) to determine whether vasoconstriction is a general response of the vasculature to H₂O₂ or a specific response of pulmonary arterial tissue, 2) to demonstrate that H₂O₂-induced pulmonary arterial muscle contractions are independent of calcium and, finally, 3) to determine whether increased MLC₂₀ phosphorylation correlates with the initiation of H₂O₂-induced arterial smooth muscle contraction.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Young adult (225-275 g) Sprague-Dawley rats were anesthetized (by pentobarbital sodium) and exsanguinated. Several large vessels and several small resistance vessels were excised. The large vessels were defined as conduit vessels and medium-sized muscular arteries and medium-sized intraparenchymal veins. Large vessels included the main pulmonary artery (~1.0-1.5 mm in diameter) and vein (~1.5-2.0 mm in diameter), abdominal aorta (~1.0 mm in diameter), inferior vena cava (~1.5 mm in diameter), and the intraparenchymal vein (~750 µm to 1.0 mm in diameter) from the right middle lobe of the lung. The resistance vessels included fourth generation mesenteric arteries and veins (~200-350 µm diameter) and fifth generation pulmonary arteries (~300-450 µm) from the middle lobe of the right lung.

The large vessels were cleaned of all visible adhering connective tissue and parenchyma under a dissecting microscope and cut into rings. Each vascular ring (3-5 mm in length) was placed in a tissue bath containing 10 ml modified Krebs-Henseleit buffer solution (KHB; in mM:120.7 NaCl, in mM:15.5 NaHCO₃, in mM:5.9 KCl, in mM:1.20 NaH₂PO₄, in mM:1.2 MgCl₂, in mM:2.5 CaCl₂, and in mM:11.5 glucose and bubbled with 95% O₂-5% CO₂). The ring was gently threaded onto a horizontally oriented, fixed-position surgical steel wire (300 µm in diameter, 5 mm in length). Once it was anchored, a second wire of the same dimensions but connected to a force transducer (Grass model FT 03C) was introduced into the lumen above the stationary wire. Isometric tension was recorded as a function of time on a strip-chart recorder (Gould, model 2400). Racking the force transducer upward resulted in stretching the arterial ring along its transverse axis. Such extensions were essentially changes in "circumferential length". Each vascular ring was stretched in this fashion to produce the mean optimal resting tension (RP_o) for maximum active tension development (P_o) and allowed to equilibrate for 1 h at the previously determined RP_o for each specific vessel type (e.g., 0.7 g for rat pulmonary arterial rings). Vascular rings were then maximally contracted with 80 mM KCl to establish P_o. After wash-out, the vascular muscle was allowed to relax to baseline before addition of 10⁴ M H₂O₂. The apparatus and methodology have been described in detail previously (25).

The basic system (Living Systems Instrumentation) for experiments on resistance vessel segments consists of a video dimension analyzer, a monochrome video camera and monitor, a pressure servo control, a flow pump, a pressure transducer, a recorder, a microscope, and a vessel. Each isolated arterial or venous segment was mounted in the vessel chamber. The ends of each segment were drawn onto microcannulas (outer diameter <80 µm). Any small side branches along the vessel were also tied off. The vessels were inflated to desired transmural pressures using KHB within the cannulas. Mounted vessels were equilibrated for 1 h at a steady pressure, which varied with vessel type and which was selected from preliminary experiments of limited length-tension relationships. At the end of the hour, the vessels were supramaximally stimulated with 30 mM KCl. Once a plateau in the isobaric contraction had been achieved, the bath was washed out and the vessel segment relaxed (i.e., dilated to resting dimensions). After complete relaxation, 10² M H₂O₂ was introduced into the bath and the response was recorded.

Actual diameter measurements were made by a single horizontal scan line, which was selected using the scan line knob to intersect the wall and lumen of a vessel of which the longitudinal axis was made perpendicular to the scan line by rotating the TV camera. Two operator-adjustable windows were then created by the start and width controls so that they bracketed the vessel walls and were sufficiently large to accommodate any anticipated movement of the vessel wall due to constriction or dilation of the vessel during an experiment. As the electron beam of the Vidicon tube scans the image of the vessel, the signal voltage obtained varies according to the optical density of the image, which varies between the walls and the lumen or the field outside the vessel preparation. Discriminator adjustment is accomplished by the level controls so that the voltage levels of a ramp signal are sampled at the time in the scan period when a change in contrast, or density, is encountered. These voltages are then appropriately subtracted so that the voltage differences are proportional to the left and right wall thicknesses and to the lumen diameter. By prior calibration using a stage micrometer, these voltages (available as analog output signals and indicated on the left wall, right wall, and diameter digital voltmeters) may be scaled to the actual dimensions in micrometers. A more detailed description of the methodology and apparatus has been published previously (23).

Experiments to address pharmacological or biochemical parameters involved both rat pulmonary arterial strips as previously described and strips that were cut from second generation porcine pulmonary arteries ranging from 2-4 mm in diameter. In the latter case, helical strips with dimensions of 12 mm in length and 2 mm in width ranged from about 4-18 mg in mass due to variation in arterial wall thickness. H₂O₂ dose-response data were obtained for the porcine arterial strip preparations. (H₂O₂ dose-response curves for rat pulmonary arterial muscle strips had been reported previously; see Ref. 28.) Strips were attached to force transducers in muscle baths of modified KHB solution (in mM:115 NaCl, in mM:25 NaHCO₃, in mM:1.38 NaH₂PO₄, in mM:2.51 KCl, in mM:2.46 MgSO₄, in mM:1.91 CaCl₂, and in mM:5.56 dextrose and bubbled with 95% O₂-5% CO₂). The strips were stretched to optimal resting length by adjusting the passive tension to the RP_o of 3.6 ± 0.1 g for maximum force development, P_o, determined in earlier experiments. The strips were allowed to equilibrate for 1 h at RP_o. The arterial muscle strips were maximally contracted with 120 mM KCl. After the contractions had come to plateaus, the baths were washed out with fresh KHB. After complete relaxation of about 1 h duration, four different concentrations of H₂O₂ (0.1, 0.316, 1.0, and 3.16 mM) were used to generate dose-response curves to H₂O_2.

Porcine and rat pulmonary arterial muscle strips were stimulated with H₂O₂ in the presence or absence of extracellular calcium (Ca²⁺-free KHB in mM:115 NaCl, in mM:25 NaHCO₃, in mM:1.38 NaH₂PO₄, in mM:2.51 KCl, in mM:2.46 MgSO₄, in mM:5.56 D-glucose, and in mM:0.1 EGTA). In another series of experiments, 1 and 10 mM H₂O₂-induced contractions in rat and porcine pulmonary artery were elicited in the presence or absence of 10, 30, 50, or 100 µM ML-9, an MLCK inhibitor.

Additional experiments were performed to provide further evidence that H₂O₂ induces contraction independently of the usual calcium signal in systemic arterial smooth muscle. Smooth muscle strips were cut from caudal arteries (~500 µm in diameter) excised from young adult male spontaneously hypertensive rats (SHR) and Wistar-Kyoto rats (WKY) immediately after death. The arterial strips were attached to force transducers and stretched to RP_o (0.5 g) and were allowed to equilibrate for 1 h. Contraction with 90 mM KCl was used to establish P_o. To deplete both intracellular and extracellular calcium, the ionophore, ionomycin, was used in calcium-free buffer followed by ionomycin with high calcium to confirm contractile ability and, finally, ionomycin calcium-free buffer until the muscle relaxed completely (Ca²⁺-free buffer contained 40 µM ionomycin, 2 mM EGTA, 145 mM KCl, 5 mM MOPS, 1.2 mM MgSO₄, and 5.6 mM glucose, pH 7.4 at 35°C; high-Ca²⁺ buffer contained 40 µM ionomycin, 5 mM CaCl₂, 145 mM KCl, 5 mM MOPS, 1.2 mM M₂SO₄, and 5.6 mM glucose, pH 7.4 at 35°C). The addition of 3 mM H₂O₂ to the relaxed muscle caused contraction even in the absence of calcium.

Since it is not known whether ionomycin with EGTA depletes all intracellular calcium stores, another series of experiments using Triton X-100 freeze-glycerinated permeabilized porcine pulmonary arterial smooth muscle was performed according to the method of McMahon and Paul (18). Fourth generation pulmonary arterial strips were isolated and placed under a low resting tension in an equilibrating relaxing solution (in mM:150 sucrose, in mM:50 KCl, in mM:5 EGTA, and in mM:20 imidazole, pH 7.4). A 4-h permeabilization step at 0°C with 1% Triton X-100 [1% Triton X-100, 0.5 mM dithioerythritol (DTE), 150 mM sucrose, 50 mM KCl, 5 mM EGTA, and 20 mM imidazole, pH 7.4] was followed by freeze/glycerol protein stabilization (50% glycerol, 0.1 mM DTE, 4 mM EGTA, 10 mM MgCl₂, 7.5 mM ATP, 5 mM phosphocreatine, and 20 mM imidazole, pH 6.7). The tissue was stored in a subzero freezer and was used within 3 days. Permeabilized porcine pulmonary arterial strips were attached to an isometric force transducer, stretched to RP_o, and equilibrated at room temperature in a 0 Ca²⁺/EGTA relaxing solution (4 mM EGTA, 10 mM MgCl₂, 0.1 µM calmodulin, 10 U/ml creatine phosphokinase, 5 mM phosphocreatine, 7.5 mM ATP, and 20 mM imidazole, pH 6.7). The experiment started with a brief switch to a high-Ca²⁺ solution (4 mM EGTA, 4 mM CaCl₂, 10 mM MgCl₂, 0.1 µM calmodulin, 10 U/ml creatine phosphokinase, 5 mM phosphocreatine, 7.5 mM ATP, and 20 mM imidazole, pH 6.7) to induce active tension development to establish viability. Strips were then switched back to the 0 Ca²⁺/EGTA solution, and 3 mM H₂O₂ was added.

Experiments to address the role of MLC phosphorylation in H₂O₂-induced contractions were carried out. Pulmonary arterial smooth muscle strips were stimulated with 1 mM H₂O₂ and were freeze clamped with tongs cooled to the temperature of liquid N₂ (195.79°C) at varying time points during contraction. The frozen tissue was stored at 70°C until processed for measurement of the levels of phosphorylation of the MLC₂₀ at the various time points, utilizing a similar protocol to that previously reported (38). Once weighed, the frozen samples were placed in a frozen slurry of the denaturation solution [90% acetone, 10% trichloroacetic acid (TCA), and 10 mM dithiothreitol (DTT) frozen in 4-ml aliquots in liquid N₂]. The solution aliquots with the immersed samples were allowed to melt at 20°C. The denatured tissues were then homogenized in 60× µl vol/mg wet wt cold (4°C) homogenizing solution (10% TCA and 10 mM DTT). The homogenized samples were centrifuged at 7,000 rpm for 60 s, the supernatant was discarded, and the remaining pellet washed three times (5 min/wash) with ether. After the third wash, the remaining ether was evaporated in a hood for 20 min. The pellets were resuspended in 60 µl vol/mg wt urea sample buffer (91.5% 8 M urea, 0.5 M DTT, 100 µl saturated sucrose, 0.2% bromophenol blue, and 8.35% urea gel buffer: 241 mM Tris, 266 mM glycine in H₂O) on a multiple tube vortex mixer for 60 min. The resuspended samples were applied (20-µl and 15-µl samples that consisted of ~0.3 mg protein/well) to wells in urea-glycerol minigels and subjected to electrophoresis at 400-V constant voltage until 20 min after the dye front ran off. The resultant protein bands were transferred to nitrocellulose blots by Western blot technique at 1.5 A for 1 h at 20°C. The blots were blocked in 0.3% milk in 15 mM NaCl and 10 mM Tris buffer (pH 7.4) for 1 h, and then washed in 1:2,000 dilution of antiserum raised to recombinant human umbilical arterial MLC₂₀ (antigen provided by Dr. James T. Stull, University of Texas Southwestern Medical Center at Dallas) overnight. The blots were then incubated in goat anti-rabbit IgG-peroxidase (1:500) conjugated by mixing for a minimum of 1 h. Finally, the blots were developed in a 4-chloro-1-naphthol/H₂O₂ solution, and the relative amounts of nonphosphorylated and phosphorylated MLC₂₀ were measured densitometrically.

Results are presented as means ± SE. Statistically significant differences were established when P < 0.05 with Student's t-test for any two mean values and with ANOVA followed by Newman-Keuls test when comparing multiple means.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Typical isometric responses of rat pulmonary arterial and venous rings and aortic and inferior vena cavae rings to high K⁺ and to H₂O₂ are shown in Fig. 1. The rings of muscle from the large arteries and veins all responded to H₂O₂ with contractile force that was 20% of their respective responses to 80 mM KCl.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Typical tension vs. time tracings of the responses of rat large vessel rings to high K+ and to 104-102 M H2O2. The large pulmonary arterial and venous rings (A and B), the aortic and inferior vena cavae rings (C and D), and the intraparenchymal pulmonary venous ring (E) all responded to H2O2 with contractions that were 20% of their respective responses to 80 mM KCl.

Typical isobaric responses of the pulmonary and mesenteric resistance arterial and venous segments to H₂O₂ are shown in Fig. 2. All of the small vessels investigated, whether arterial or venous or pulmonary or systemic, responded with active constriction to H₂O₂.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Typical transmural pressure (top) and diameter changes (bottom) vs. time trace recordings of the rat small mesenteric arterial (A) and venous segments (B). C: typical tracings of an experiment on a rat pulmonary resistance arterial segment, with transmural pressure vs. time (top) and diameter changes vs. time (bottom) shown. All of the small vessels investigated responded with active vasoconstriction to H2O2.

Mean data in Fig. 3 show the concentration dependence for the time course of porcine pulmonary arterial contraction in response to H₂O₂ (n = 3 for each concentration at each measured time point). The maximum contraction of about 65% P_o in 40 min resulted with 3 mM H₂O₂. Lower concentrations gave slower rates of tension development, and plateaus were reached later in the time courses studied. Actual concentrations of H₂O₂ are likely changing during the time course of the experiment due to the scavenging actions of endogenous catalase. The higher concentration of H₂O₂ was used subsequently for experimental purposes to maximize the response despite the presence of endogenous catalase.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Mean data (n = 3 for each concentration at each measured time point) shows the concentration dependence for the time course of contraction of porcine intraparenchymal pulmonary arterial (1.5-2.0 mm in diameter) muscle in response to H2O2. The maximum contraction of ~65% maximum active tension (Po) in 40 min resulted from stimulation with 3 mM H2O2. Lower concentrations gave a slower rate of tension development, and plateaus were reached later in the time courses of the contractions (P < 0.05). Actual concentrations of H2O2 are likely diminishing during the time courses of the contractions due to the scavenging actions of endogenous catalase. The 3 and 1 mM concentrations of H2O2 were used in subsequent experiments to maximize the responses in spite of the activity of endogenous catalase.

Mean isometric tension vs. time tracings for porcine pulmonary arterial muscle in response to 5 × 10⁵ M norepinephrine (NE) and 1 mM H₂O₂ in the presence (n = 10 and n = 6, respectively) and absence (n = 11 and n = 5, respectively) of extracellular calcium and ML-9 (n = 3 for both NE and H₂O₂) are shown in Fig. 4. Both calcium depletion and ML-9 blocked the contractile response to NE but had no effect on the contractile response to H₂O₂. In isolated rat pulmonary artery, ML-9 had a concentration-dependent blocking effect on contraction to 80 mM KCl (P < 0.05) but had no effect on the contractile response to either 1 or 10 mM H₂O₂ (P > 0.05).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Repeated stimulation with 5 × 105 M norepinephrine (NE) and washout with zero calcium Krebs-Henseleit buffer solution (KHB) resulted in calcium depletion of porcine intraparenchymal pulmonary arterial (1.5-2.0 mm in diameter) muscle strips. Three NE stimulation and wash-out cycles nearly abolished the NE-induced contraction in the zero Ca2+ buffer. ML-9, a myosin light chain kinase (MLCK) blocker, had a similar inhibitory effect on tension development to NE stimulation (A; P < 0.05). However, neither calcium depletion nor treatment with ML-9 had any effect on contraction to subsequent stimulation with 1 mM H2O2 (B; P > 0.05).

Figure 5 shows that rat caudal arterial muscle depleted of calcium using ionomycin in the presence of EGTA (n = 12) responded to 3 mM H₂O₂ with tension development that was significantly greater in magnitude than the H₂O₂ response in normal calcium KHB (n = 5; P < 0.051) and similar in magnitude to the contractile response to KCl in normal calcium KHB (n = 12; P > 0.05). The KCl response in KHB was abolished in the absence of calcium (n = 5; P < 0.001). Since there was no difference in the response of the SHR and WKY muscle to H₂O₂ (P > 0.05), data from the two strains were pooled.

View larger version (25K): [in this window] [in a new window]   Fig. 5.   Magnitudes of contractile responses of arterial muscle to high K+ and to H2O2 in normal and in zero Ca2+ solutions. Isolated rat caudal arterial muscle was contracted with 90 mM KCl to establish Po. To deplete both intracellular and extracellular calcium, the ionophore ionomycin was used with 0.01 mM EGTA in a zero calcium buffer, followed by ionomycin with calcium to confirm contractile ability, and finally ionomycin with 0 Ca2+/EGTA buffer until the muscle relaxed completely. In the absence of calcium, plateau levels of tension in response to 3 mM H2O2 were similar in magnitude to the contractile response to KCl depolarization with normal calcium KHB (P > 0.05). The fact that the H2O2 response in 0 Ca2+/EGTA/ionomycin was significantly greater than the H2O2 response in normal Ca2+ KHB (P < 0.051) is likely explained by the fact that ionomycin effectively permeabilizes the membrane, making exogenously applied H2O2 more readily available to the intracellular components such as the contractile apparatus.

Tension vs. time tracings for a permeabilized porcine pulmonary arterial muscle control strip and for a strip in response to H₂O₂ in 0 Ca²⁺/EGTA solution are shown in Fig. 6. H₂O₂ resulted in contraction of the permeabilized muscle even in the absence of calcium. Rates of force development in response to H₂O₂ in 0 Ca²⁺-free solution and in response to high Ca²⁺ were measured. Force developed at the significantly slower rate of 0.05 ± 0.02 mN/min (n = 5) in response to H₂O₂ compared with the 0.56 ± 0.10 mN/min (n = 5) rate of force development in response to high Ca²⁺ (P < 0.001).

View larger version (9K): [in this window] [in a new window]   Fig. 6.   Typical tension vs. time tracings of Triton X-100, freeze-glycerinated permeabilized arterial muscle preparations in a zero Ca2+ solution. The permeabilized porcine pulmonary arterial smooth muscle contracted in response to H2O2 even in the 0 Ca2+/EGTA relaxing solution.

Examples of Western blots of MLC₂₀ from porcine pulmonary arterial muscle are shown in Fig. 7. The time courses of mean active tension and mean changes in MLC₂₀ phosphorylation in response to 80 mM K⁺ and 1 mM H₂O₂ stimulation are shown in Fig. 8. MLC₂₀ phosphorylation levels (n = 3-6) actually decreased below resting levels while force (n = 5) increased in response to H₂O₂ (P < 0.01).

View larger version (22K): [in this window] [in a new window]   Fig. 7.   A Western blot of smooth muscle 20-kDa myosin light chain (MLC20) from intact porcine arteries that had been freeze clamped at various time points during contraction. The first lane in each of these blots is from tissue frozen at resting tension. Basal levels of phosphorylated MLC20 were about 8% of total MLC20. In agreement with others, maximal KCl-induced contraction maintained a level of phosphorylation above resting levels throughout the time course of contraction. Surprisingly, levels of phosphorylated MLC20 were substantially below resting levels throughout the time course of H2O2-induced contraction.

View larger version (26K): [in this window] [in a new window]   Fig. 8.   Mean tension developed (%Po) and phosphorylated MLC20 (MLC20-P) levels (%total MLC20) are plotted as functions of time during contraction of porcine pulmonary arterial smooth muscle (n = 3-6 for each point). Phosphorylation of MLC20 peaked at 0.5 min, which was during the maximal rate of tension development in response to KCl stimulation (A). Surprisingly, with H2O2 stimulation (B), phosphorylation of MLC20 decreased significantly within 2 min (P < 0.01) and continued to decline to almost zero throughout the time course of contraction. Substantial active tension developed despite the significant decrease in MLC20 phosphorylation from resting levels in response to H2O2.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Hypoxanthine accumulates during ischemic episodes resulting in superoxide and H₂O₂ bursts during ischemia and reperfusion in various organs including the heart, kidney, lung, brain, and gut. The increased H₂O₂ may restrict blood flow and contribute to ischemia-reperfusion injury. Levels of H₂O₂ within the range used in the current study have been reported in vivo in a variety of pathogenic conditions. Patients with exacerbated chronic obstructive pulmonary disease (6) and asthmatic children (14) exhale micromolar concentrations of H₂O₂. In another study, nonsurvivors of adult respiratory distress syndrome were reported to have higher levels of oxidative stress than did survivors (39). H₂O₂ production was detected directly in the granulocytes that adhered to the pulmonary endothelium in a rat sepsis model (20). In rats subjected to forebrain ischemia, H₂O₂ levels peaked at 0.1 M above baseline during reperfusion (12). Vasoconstriction due to high H₂O₂ levels likely contributes to impaired perfusion.

The present study demonstrates that H₂O₂ directly causes contraction of both porcine and rat pulmonary and systemic arteries and veins. The magnitude of the responses of the various vascular muscle preparations differed significantly. For example, a similar dose of H₂O₂ resulted in about a 20% P_o tension development in rat pulmonary artery and about a 65% P_o tension development in porcine pulmonary artery. However, one cannot conclude whether such quantitative differences are due to species differences in reactivity or to differences in the exact anatomical source of the specific muscle preparation. Although the large pulmonary arterial preparations from the rat and pig were of about the same size (diameter), the preparations were from different generations. Such quantitative comparisons would require a rigorous study in which many variables would have to be controlled. Variables include, but may not be limited to, species, vessel generation, vessel size, extra- or intraparenchymal location, gender, morphometrics of the vessel walls, and the type of preparation (i.e., ring vs. strip) required by the available techniques to generate data for a given vessel size.

Surprisingly, the contractile response of vascular smooth muscle to H₂O₂ is independent of the usual Ca²⁺- and MLC₂₀ phosphorylation-dependent signal transduction cascade. Indeed, results of experiments on permeabilized arterial muscle show that ion fluxes across smooth muscle cell membranes are not required for the H₂O₂-induced contraction. H₂O₂ has previously been reported to contract rabbit carotid arteries (39) and rat pulmonary arteries even in Ca²⁺-free solution and in the absence of endothelium (28). The possibility remained that H₂O₂ caused Ca²⁺ release from intracellular stores such as mitochondria. However, results of the current study show that H₂O₂ causes a contraction that is greater in magnitude in permeabilized muscle in the absence of Ca²⁺ than occurs in intact muscle in the presence of calcium and that MLC phosphorylation actually declines to near zero levels with H₂O₂ stimulation. (The fact that the H₂O₂-induced contraction is greater in permeabilized than in intact muscle is likely due to the leakiness of the membrane. Exogenous H₂O₂ would have greater and faster accessibility to the internal cell structures such as the contractile apparatus in the ionomycin-treated tissue than in the intact tissue. In addition, endogenous catalase may have leaked out of ionomycin-treated cells resulting in higher intracellular H₂O₂ concentrations than in the intact preparation.) Clearly, an alternative regulatory mechanism must be responsible for the H₂O₂-induced contractile response.

Evidence from both plant and animal cells suggests that H₂O₂ may act as an intracellular second messenger (35). Endogenous H₂O₂ production was reported in Balb/3T3 cells in response to platelet-derived growth factor (PDGF) (30), in pulmonary neuroepithelial cells stimulated with a phorbol ester (36), in mouse osteoblastic cells stimulated with transforming growth factor 1 (22), in human fibroblasts in response to interleukin-1 or tumor necrosis factor- (19), and in human fat cells in response to insulin (16). The response by rat vascular smooth muscle cells to PDGF was reported to require generation of H₂O₂ (32).

Since rises in intracellular Ca²⁺ and MLC phosphorylation are not involved in H₂O₂-induced vasoconstriction, alternative signaling pathways must be explored. Phorbol esters that activate PKC also cause arterial muscle contraction independent of increases in intracellular Ca²⁺ and MLC phosphorylation (2, 34). PKC has been implicated in putative smooth muscle cell signal transduction cascades that activate muscle contraction by way of thin filament-linked proteins such as caldesmon and calponin (1, 21). A study in isolated pulmonary arterial muscle showed that H-7, a PKC inhibitor, reduced the active contractile response to phorbol esters and to oxidants (13). Another possible mechanism includes cytoskeletal actin reorganization, which may increase basal actomyosin ATPase activity and result in contraction. Increased amounts of polymerized actin have been detected in oxidant-injured cells (11). Finally, direct oxidation of contractile proteins may result in actomyosin ATPase activity and vascular smooth muscle contraction (4, 17). These studies suggest that modification of sulfhydryl groups on the myosin head or other contractile or regulatory proteins results in force development independent of the Ca²⁺ concentration or myosin phosphorylation status.

In conclusion, isolated vascular smooth muscle contracts to H₂O₂ independently of the usual Ca²⁺ and MLC₂₀ phosphorylation signal transduction cascade and independent of ion fluxes across smooth muscle cell membranes. H₂O₂-induced contraction appears to be a general response of vascular muscles and is apparently conserved across evolution. Elucidation of the mechanism of H₂O₂-induced vasoconstriction may prove important in developing new therapeutic agents for treating various vascular disorders such as ischemia-reperfusion injury and hypertension. Alternative signal transduction mechanisms such as thin filament linked regulation need to be explored.