Inhibition of Rho-Kinase Attenuates Hypoxia-Induced Angiogenesis in the Pulmonary Circulation
http://www.100md.com
循环研究杂志 2005年第7期
the Departments of Physiology (J.-M.H., K.H., A.N., R.J.P., P.M.) and Pharmacology (C.M.C.), Conway Institute of Biomolecular and Biomedical Research and the Dublin Molecular Medicine Centre (P.M.), University College, Dublin, Ireland.
Abstract
Pulmonary hypertension (PH) is a common complication of chronic hypoxic lung diseases, which increase morbidity and mortality. Hypoxic PH has previously been attributed to structural changes in the pulmonary vasculature including narrowing of the vascular lumen and loss of vessels, which produce a fixed increase in resistance. Using quantitative stereology, we now show that chronic hypoxia caused PH and remodeling of the blood vessel walls in rats but that this remodeling did not lead to structural narrowing of the vascular lumen. Sustained inhibition of the RhoA/Rho-kinase pathway throughout the period of hypoxic exposure attenuated PH and prevented remodeling in intra-acinar vessels without enlarging the structurally determined lumen diameter. In chronically hypoxic lungs, acute Rho kinase inhibition markedly decreased PVR but did not alter the alveolar to arterial oxygen gap. In addition to increased vascular resistance, chronic hypoxia induced Rho kinaseeCdependent capillary angiogenesis. Thus, hypoxic PH was not caused by fixed structural changes in the vasculature but by sustained vasoconstriction, which was largely Rho kinase dependent. Importantly, this vasoconstriction had no role in ventilation-perfusion matching and optimization of gas exchange. Rho kinase also mediated hypoxia-induced capillary angiogenesis, a previously unrecognized but potentially important adaptive response.
Key Words: pulmonary hypertension angiogenesis RhoA Rho-kinase Y-27632
Introduction
Sustained pulmonary hypertension (PH) is a common complication of chronic hypoxic lung diseases that is strongly associated with increased morbidity and reduced survival. Moreover, the presence of cor pulmonale is an independent predictor of increased mortality, suggesting that PH contributes directly to mortality (reviewed in Hopkins et al1). The increase in pulmonary vascular resistance (PVR) caused by chronic hypoxia has previously been attributed to structural changes in the vasculature, in particular encroachment of the remodeled arteriolar walls into the vascular lumen and loss of blood vessels, although recent reports have cast doubt on this paradigm.1eC3 In particular, we have recently shown for the first time that hypoxia induces angiogenesis in the adult pulmonary circulation, a potentially beneficial adaptation, and does not cause vessel loss as previously believed.2
The small G-protein RhoA and its downstream effector Rho-kinase (ROCK) play a central role in diverse cellular functions including smooth muscle contraction, cytoskeletal rearrangement, cell migration, cell proliferation, and gene expression.4eC8 Given these important functions, it is not surprising that disturbances of this pathway have been identified as important pathogenetic mechanisms in many diseases of the cardiovascular system, including systemic hypertension, arteriosclerosis, and ischemic heart disease.9eC11 Blockade of the RhoA/ROCK pathway effectively corrects blood pressure in a number of animal models of systemic hypertension8,11 and is a key regulator of vascular smooth muscle proliferation and migration in disease-induced systemic vascular remodeling.9,10 This pathway also regulates endothelial cell proliferation and migration. In vitro models and in vivo Matrigel implants show that inhibition of RhoA or ROCK prevents growth factoreCinduced endothelial cell migration and organization into capillary-like structures12,13 and vessel formation.14 Moreover, ROCK inhibitors attenuate tumor growth in vivo in a manner that is compatible with an antiangiogenic effect.15 Recognition of the therapeutic potential of inhibiting this pathway has lead to the development of specific small molecule inhibitors of ROCK.8,11,15
Taken together, these reports suggest that the RhoA/ROCK pathway might contribute to the development of hypoxic PH and the associated pulmonary vascular remodeling. The purpose of this study was to assess the potential role of this pathway in the development of chronic hypoxia-induced PH and the associated structural changes in the pulmonary vasculature. In particular, we wished to test the hypothesis that, although chronic inhibition of ROCK would inhibit the development of hypoxic PH, it would simultaneously inhibit hypoxia-induced pulmonary angiogenesis.
Materials and Methods
An expanded Materials and Methods section can be found in the online supplement available at http://circres.ahajournals.org.
Hypoxic Pulmonary Hypertension
Chronic hypoxic PH was induced by housing adult male rats in a normobaric hypoxic chamber (FiO2 0.10) for 1 or 3 weeks as previously described.16 Control rats were housed in normoxic conditions (FiO2 0.21) in the same room. To assess the effects of chronic ROCK inhibition, rats were randomized into 3 groups; 1 group was maintained in normoxic conditions (control group), while 2 groups were housed in hypoxic conditions for 1 week. Y-27632 was used to inhibit the phosphorylation activity of ROCK.4eC8,17 One hypoxic group (CH-Y27632) received Y-27632 (30 mg · kgeC1 · dayeC1, orally) while the other did not receive Y-27632 (CH group).
Measurement of Vascular Pressures and Gas Exchange In Vivo
Rats were anesthetized, tracheostomised, paralyzed, and mechanically ventilated. Venous and arterial cannulae were placed for drug administration, blood gas sampling, and blood pressure measurement. Right ventricular pressure was measured by introduction of a needle via a subdiaphragmatic approach just before euthanasia.
Isolated Perfused Lung Preparation
Lungs were isolated under anesthesia, mechanically ventilated, and perfused at constant low flow with a mixture of blood and physiological saline of constant hematocrit so that pulmonary arterial pressure (PAP) was a direct index of PVR.17 Capillary pressure was assessed by the double occlusion technique.17
Tissue Collection
At the end of each isolated lung experiment, the left lung (LL) was fixed under standard conditions for stereological quantification of structural vascular changes.2 The right lungs were snap frozen for subsequent protein and mRNA analysis.
Data Analysis
For stereological analyses, total volume of specific compartments, vessel and capillary lengths, and alveolar epithelial and capillary endothelial surface areas were reported per left lung. For normally distributed data, responses were reported as mean±SEM, n refers to the number of rats from which tissue was obtained. For multiple comparisons of means across experimental groups, analysis of variance was performed followed by StudenteCNewman Keuls post-hoc test for pair-wise comparisons. For non-normally distributed data, multiple comparisons of medians across experimental groups were performed using the KruskaleCWallis test followed by ManneCWhitney U with the Bonferonni post-hoc correction; values of P<0.05 were accepted as statistically significant.
Results
Effect of Acute ROCK Inhibition on Established Hypoxic Pulmonary Hypertension
Mean peak right ventricular systolic pressure (RVSPpeak) measured during ventilation with 30% oxygen (53.5±3.5 mm Hg) in rats (n=5) maintained chronically hypoxic for 3 weeks was significantly (P<0.01) greater than that in control nonhypoxic rats (28.8±1.4 mm Hg), demonstrating the presence of fixed PH; ie, increased PVR that was not attributable to acute hypoxic pulmonary vasoconstriction (HPV). After administration of the potent ROCK inhibitor Y27632 (15 mg · kgeC1),11 mean RVSPPeak was reduced significantly (P<0.02) to a value of 34.6 (±5.6) mm Hg.
Changes in RVSPPeak in response to Y27632 in vivo may not have resulted from changes in PVR but may have resulted from changes in venous return, cardiac output, left atrial pressure, neurogenic control, and other factors. To examine the direct effect of ROCK inhibition on PVR in lungs chronically hypoxic for 3 weeks, we used an isolated lung preparation. Mean PAP in chronically hypoxic lungs (n=5) was 15.1 (±0.6) mm Hg and was significantly reduced (P<0.01) to 13.0 (±0.9) mm Hg by Y27632 (3 e.leC1). At the highest Y27632 concentration (100 e.leC1) mean PAP was further reduced (P<0.01) to 10.7 (±0.4) mm Hg, a value that was similar to that in control nonhypoxic lungs (see below and Figure 1).
Thus the dominant mechanism underlying chronic hypoxic PH was sustained vasoconstriction and not structural changes in the pulmonary vasculature. These results suggested that sustained ROCK inhibition in vivo would prevent the development of chronic hypoxic PH.
Chronic ROCK Inhibition In Vivo
To confirm that chronic exposure to a hypoxic environment for 1 week caused fixed PH, we measured RVSPPeak in anesthetized rats (n=5) during ventilation with 30% oxygen, and found that after such exposure RVSPPeak was significantly (P<0.01) elevated (42.5±3.0 mm Hg) when compared with control nonhypoxic rats (28.8±1.4 mm Hg). For the reasons outlined above, further investigation of the role of ROCK in the development of PH was undertaken using the isolated lung preparation.
Mean right ventricular (RV) weight (see supplemental Table I) in a group maintained hypoxic for 1 week (CH group) and in a hypoxic group chronically treated with Y-27632 (CH-Y27632) for 1 week were both significantly higher than that of the control group. The sustained treatment with ROCK inhibitor significantly reduced the RV hypertrophy compared with hypoxia alone although it did not alter the polycythemic response to hypoxia (supplemental Table I).
We assessed the total PVR in isolated lungs under 3 different conditions, ie, lungs ventilated with 16%, 7%, or 3% O2 (Figure 1). These concentrations give alveolar values that approximate normal sea level alveolar pO2, the alveolar pO2 experienced by the rats during hypoxic exposure in the chamber, and an alveolar pO2 that produces a maximal hypoxic vasoconstrictor response. Lungs from the control group showed normal low PVR when ventilated with 16% O2 and significant increases in resistance in response to reduction in alveolar O2 indicating normal HPV (Figure 1). Chronically hypoxic lungs ventilated with 7% O2 had significantly elevated vascular resistance compared with control lungs ventilated with the same inspired oxygen, and this resistance was not reduced by acute return to normal alveolar oxygen concentration (16% O2). Reduction to 3% O2 in the chronically hypoxic lungs caused a further small increase in resistance to a value similar to that seen in control lungs at the same inspired oxygen (Figure 1). Chronic administration of the ROCK inhibitor Y-27632 prevented the development of PH in response to sustained exposure to hypoxia and also abolished the acute hypoxic vasoconstrictor response (Figure 1).
To evaluate the contribution of ROCK-dependent vasoconstriction to the control of PVR in these chronically hypoxic and control lungs, Y-27632 was added to the perfusate up to 3 eol/L, a concentration that selectively inhibited ROCK.8,11 In normoxic lungs, ventilated with 16%O2, Y-27632 up to 3 eol · leC1 did not change the PAP (Figure 2A), demonstrating that ROCK activity did not play a role in the control of vascular resistance in the normal lung.
In the CH group, acute administration of Y-27632 markedly decreased the PAP in a dose-dependent fashion (Figure 2A). This concentration of ROCK inhibitor abrogated approximately two-thirds of the increase induced by chronic hypoxia through an effect that was mainly localized to the arterial (precapillary) segment (Figure 2B). This demonstrated that a vasoconstrictor activity of ROCK played a key role in maintaining chronic hypoxic PH.
Finally, in the CH-Y27632 group, in which the baseline PAP was similar to that in the control group, acute addition of Y-27632 had no significant effect on PA pressure in the isolated lungs, demonstrating that the chronic treatment with Y-27632 had maximally inhibited the effect of ROCK activity on PVR (Figure 2A).
In a separate series of experiments, we assessed the effect of higher, potentially nonselective8,11 concentrations of Y-27632 (up to 100 eol/L) on the PAP in lungs maintained chronically hypoxic for 1 week and found that PAP was decreased to 10.1 (±0.2) mm Hg (n=3), a value similar to that in control lungs (9.6±0.3 mm Hg). These results were similar to those obtained in the 3-week hypoxic group and supported the conclusion that the dominant mechanism underlying chronic hypoxic PH is sustained vasoconstriction.
In the presence of Y-27632 (3 eol · leC1), reducing the oxygen level to 3% did not produce significant pulmonary vasoconstriction in any group, indicating that ROCK is a key mediator of acute HPV (data not shown).
Effect of Acute ROCK Inhibition on Pulmonary Oxygen Uptake in Hypoxic Rats
In chronically hypoxic rats ventilated with a hypoxic inspirate (FiO2 0.10), there was no change in either the A-aO2 gap or the partial pressure of oxygen in arterial blood (PaO2) after acute administration of Y-27632 (see online supplement). The A-aO2 gap and PaO2 were unchanged after injection of vehicle (data not shown).
Effect of Chronic Inhibition of ROCK on Hypoxia-Induced Angiogenesis and Vascular Remodeling
The mean left lung volume was significantly increased in the CH group (3.28±0.15 mL) when compared with both the control group (2.54±0.12 mL, P<0.001) and the CH-Y-27632 group (2.98±0.05 mL, P<0.05).
In the control lungs, the majority of intra-acinar vessels were thin-walled, with a single elastic lamina and no discernable tunica media (see online supplement results). In the CH group remodeling of the wall of intra-acinar vessels was evident, as shown by the development of separate internal and external elastic laminae and an intervening layer of media (see online supplement results). After chronic inhibition of ROCK, the structure of vessel walls was similar to that observed in the control group suggesting inhibition of hypoxia-induced vascular remodeling.
In the CH group, the mean total length of intra-acinar vessels (9236±630 cm · LLeC1) was significantly higher than in the control group (5327±392 cm · LLeC1, P<0.001). After chronic inhibition of ROCK, the total length of intra-acinar vessels (7864±508 cm · LLeC1) was less than in the CH group although it remained significantly higher than in control group (P<0.01). Analysis of the mean lumen diameter of the intra-acinar vessels showed that there was no difference between the 3 groups (Figure 3A). However, the ratio of the wall thickness to the lumen diameter was significantly increased in the CH-group when compared with control and CH-Y27632 lungs (Figure 3B).
The total capillary volume and total capillary length (Table 1) were significantly increased in the CH group when compared with the control group, whereas after chronic treatment with Y-27632 these hypoxia-induced increases were significantly reduced and not significantly different from those in control group. The total capillary endothelial surface area was also significantly increased in the CH lungs when compared with the control group (Table 2). In the CH-Y-27632 group, capillary endothelial surface area was significantly less than that in the CH group although it remained elevated above the control value (Table 2). Chronic hypoxia also caused a significant increase in the total alveolar epithelial surface area when compared with the control group (Table 2), although in the CH-Y-27632 group this value was significantly less that in the CH group and similar to that in the control group. In both hypoxic groups, the increase in capillary endothelial surface area was proportionately greater than that in the epithelial surface area as indicated by the increase in the mean ratio of capillary endothelium to alveolar epithelium (Table 2).
Effect of Hypoxia on RhoA, RhoA Activation, and ROCK Expression
The total amount of RhoA was similar in each of the different groups of rats; ie, control, CH, and CH-Y-27632 (Figure 4). RhoA activation, expressed as the percentage of total RhoA in GTP-RhoA form, showed no significant difference between the different groups (Figure 4). In CH and CH-Y-27632 groups levels of ROCK I and II isoforms were significantly increased when compared with the control group (Figure 5).
Effect of Chronic ROCK Inhibition on Vascular Growth Factor Gene Expression
Chronic hypoxia increased expression of platelet-derived growth factor- (PDGF-), and this effect was significantly reduced by ROCK inhibition (Figure 6). Chronic ROCK inhibition also significantly reduced transforming growth factor -1 (TGF-1) expression when compared with chronically hypoxic lungs (Figure 6). Interestingly, expression of the proangiogenic genes vascular endothelial growth factor A (VEGF-A) and angiopoeitin-1 (ANG-1) was not significantly altered at the end of one week of hypoxia and was unaffected by ROCK inhibition.
Discussion
It has been generally thought that PH predominantly resulted from hypoxia-induced structural changes in the pulmonary vasculature, which produced a fixed increase in resistance. These structural changes included remodeling of the arteriolar walls leading to encroachment into the vascular lumen, and loss of blood vessels, suggesting that interventions, which successfully ameliorated PH, would act by preventing or reversing such structural changes. The results of the present study cast doubt on this paradigm by demonstrating that chronic hypoxia did not cause a structurally based reduction of pulmonary vascular lumen diameter, nor did it cause a loss of pulmonary vessels. Moreover, we report that chronic inhibition of ROCK abrogated the development of hypoxic PH, not by preventing structural encroachment into the vascular lumen, but by inhibiting sustained pulmonary vasoconstriction. Importantly, we also found hypoxia-induced capillary angiogenesis in the adult lung that was dependent on the RhoA/ROCK pathway. ROCK inhibition reduced the expression of key vascular growth factors suggesting that this was one of the mechanisms by which it attenuated hypoxia-induced vascular remodeling.
Initially, we confirmed that exposure to hypoxia for three weeks induced fixed PH in vivo; ie, increased RVSPPeak that was not immediately reversed on exposure to normal alveolar oxygen and thus was not mediated by acute hypoxic pulmonary vasoconstriction. Previously, this observation, together with the extensive remodeling of pulmonary arterioles, has been taken as evidence that the increased PVR in chronic hypoxic PH was predominantly attributable to structural changes in the vasculature. However, in our experiments administration of the ROCK inhibitor Y-27632 produced a marked fall in PAP suggesting that ROCK-dependent vasoconstriction was the major determinant of hypoxic PH.
Because the changes in peak right ventricular systolic pressure in response to Y-27632 in vivo might not have been attributable to changes in PVR but might have resulted from changes in other parameters, we used the isolated ventilated lung preparation to examine the direct effect of ROCK inhibition on PVR. Although the PA pressure in such preparations is much lower than in vivo values, it is a direct index of PVR. Our finding that Y-27632 caused PAP to fall immediately to near normal values in lungs made chronically hypoxic for 3 weeks demonstrated that ROCK inhibition normalized PVR by a direct action on pulmonary resistance vessels and strongly suggested that hypoxic PH was not structurally based.
In view of the evidence that Y-27632 could acutely reverse chronic hypoxic PH, we examined the effect of chronic Y-27632 administration during 1 week of exposure to hypoxia. In agreement with previous reports,18,19 we first confirmed that exposure to hypoxia for 1 week produced marked PH in vivo, which persisted when alveolar oxygen was acutely elevated to supra-normal values (see Results), a finding that was confirmed in isolated lung preparations (Figure 1). In addition, we observed extensive vascular wall remodeling in CH lungs (Figure 3), as previously demonstrated after 1 week of hypoxia.18,19 Thus, these lungs demonstrated the features that had previously been interpreted as indicating structurally based PH. However, our stereological analysis demonstrated other hypoxia-induced structural changes in the pulmonary vasculature (angiogenesis and wall remodeling without lumen reduction) that differ radically from those suggested by the conventional paradigm (Figure 3). Why have such structural changes not been previously detected The techniques we used to analyze vascular structure differ in 2 important ways from those previously used to address this problem, as we have previously discussed.2 First, stereology allows quantitative analysis of 3-dimensional structural parameters based on the 2-dimensional information available in histological sections and takes account of the changes in whole lung volume caused by chronic hypoxia.20 Second, vessels were recruited and fixed under no-flow conditions at a defined transmural distending pressure that was constant at all points in the vasculature, was unaffected by changes in vascular resistance, and was identical in both control and hypertensive lungs. Thus the vascular dimensions measured in this study reflect structural change and were unaffected by pre-existing smooth muscle tone in vivo.
Our observation that chronic hypoxia did not cause encroachment of the remodeled vascular wall into the lumen is in good agreement with our previous report but extends those findings in 2 important ways.2 First, we show that this outward remodeling of the vessel wall occurred in lungs in which we had hemodynamically confirmed the development of fixed PH (Figure 1). Second, we found that chronic administration of Y-27632, a selective ROCK blocker,11 inhibited the development of increased PVR in response to chronic hypoxia (Tables 1 and 2) and attenuated right ventricular hypertrophy in vivo (supplemental Table I), without causing structural enlargement of the lumen (Figure 3). We confirmed that the dose used chronically in vivo was effective in blocking the contractile activity of the ROCK pathway by demonstrating that after isolation, these lungs showed no additional vasodilator response to acute Y-27632 administration (Figure 2). The failure of chronic ROCK inhibition to significantly enlarge mean-maximal-lumen diameter was observed, despite the fact that it successfully attenuated wall remodeling, a finding that is supported by the recent report of Fagan et al.21 Taken together, our results show that chronic hypoxic PH develops without structural change in the mean-maximal-vascular-lumen diameter and, furthermore, PH can be prevented without enlarging the structurally determined lumen diameter. This suggests that chronic hypoxic PH is not caused by structurally imposed changes in vascular resistance.
This conclusion is further reinforced by the demonstration that acute administration of a high concentration of Y-27632 normalized PVR in isolated lungs made chronically hypertensive by exposure to 1 and 3 weeks of hypoxia, both in vivo and in isolated lungs (Figure 2 and see Results), results that confirm the recent report of Nagaoka et al.22 Evidence supporting the important role of RhoA/ROCK is provided by the demonstration of a marked reduction in PVR (Figure 2) at concentrations of Y-27632 that are selective for ROCK (3 micromolar or less),8,11 although these concentrations did not return pulmonary arterial pressure to normal values. It has been previously shown that Y-27632 relaxes vascular smooth muscle, preconstricted by either agonists or hypoxia, by blocking ROCK-mediated phosphorylation of the regulatory protein CPI-17 and the MYPT-1 of myosin light chain phosphatase (MLCP), both of which effects activate MLCP, thus reducing myosin light chain phosphorylation and relaxing smooth muscle.15,21 Administration of higher concentrations of Y-27632 returned PVR to control values, although this vasodilator action might not have resulted from specific ROCK inhibition.5,8 Clearly, acute normalization could not occur if vascular resistance was increased by a fixed structural mechanism.
The RhoA/ROCK-dependent sustained vasoconstriction observed in chronically hypoxic lungs is distinct from that mediating acute HPV because acute normalization of alveolar oxygen did not reduce PVR after either 1 or 3 weeks of hypoxic exposure. This suggested that the sustained vasoconstriction of chronic hypoxia did not have a role in ventilation-perfusion matching. To test this directly, we examined the effect of acute ROCK inhibition on the alveolar-arterial oxygen gap, an index of physiological shunting, and found that this was unaltered (see online supplement). This suggests that chronic ROCK-mediated hypoxic PH is a maladaptive response, because it overloads the right ventricle without improving gas exchange and is in marked contrast to the beneficial optimization of ventilation-perfusion matching produced by acute HPV.
In the present study, we found that chronic hypoxia induced capillary angiogenesis in the adult lung in vivo (Tables 1 and 2 ), confirming the results of our previous report.2 In support of this Pascaud et al reported that angiostatin, an inhibitor of angiogenesis, aggravated PH in the hypoxic lung, an observation that suggests new vessel formation in hypoxic lungs.23 More recently, Beppu and colleagues demonstrated hypoxia-induced capillary angiogenesis in the adult mouse lung.24 We now extend those findings by reporting that chronic ROCK inhibition significantly attenuated this hypoxia-induced angiogenesis (Tables 1 and 2). Inhibition of RhoA or ROCK has previously been shown to prevent growth factoreCinduced endothelial cell migration and organization into capillary-like structures in vitro and vessel formation in Matrigel implant.12,13,15,25,26 Our finding that Y-27632 attenuated hypoxia-induced angiogenesis in the lung is, to our knowledge, the first report directly demonstrating that the RhoA/ROCK pathway has a central role in hypoxic angiogenesis within adult tissues in vivo. In the lung, this may be particularly important as hypoxic angiogenesis increases the diffusing capacity for oxygen and may therefore be an important adaptive response to hypoxia.2 The possibility that hypoxia-induced angiogenesis might improve oxygen uptake in the lung awaits direct functional evidence. However, studies in high altitude and burrowing animals adapted to hypoxic environments show that the enlarged pulmonary capillary bed found in such species enhances oxygen uptake during exercise in hypoxia when oxygen uptake becomes diffusion limited, but not at rest when uptake is perfusion limited.27
The action of ROCK in the chronically hypoxic lung was not the result of increased expression of RhoA, a finding in good agreement with the recent report of Jernigan et al.28 Moreover, ROCK activity could not have resulted from increased RhoA activation because GTP-RhoA was reduced (although not significantly) and not increased (Figure 4). This result is similar to that of Bailly et al who showed reduced RhoA activation in chronically hypoxic pulmonary vascular smooth muscle,29 although increased RhoA activation has been reported in chronically hypoxic lungs after 4 weeks.28 The ROCK effects that we observed may have been mediated through the increased expression of both its isoforms (Figure 5), a finding in good agreement with the demonstration that hypoxia augments expression of this enzyme in cultured pulmonary smooth muscle cells.30
ROCK activity is an important regulator of cytoskeletal rearrangement in endothelial cells during migration, and blockade of this action inhibits in vitro organization of endothelial cells into capillary structures.14,26 Such inhibition of cytoskeletal reorganization could have contributed to the attenuated angiogenesis that we observed in the CH-Y-27632 group. We explored the possibility that inhibition of remodeling and angiogenesis might also have been mediated by altered expression of key vascular growth factors. Our finding that chronic Y-27632 treatment reduced PDGF- and TGF-1expression (Figure 6) suggests that alteration of these growth factors might be an important mechanism by which ROCK inhibition attenuated arteriolar wall remodeling. Interestingly, neither chronic hypoxia nor ROCK inhibition changed the expression of VEGF-A or ANG-1 at 1 week (Figure 6). It is possible that expression of these genes might have been altered by hypoxia at earlier time points and that Y-27632 had blocked these changes, thus inhibiting angiogenesis. Alternatively other proangiogenic genes may play a predominant role in the lung.
It has recently been suggested that inhibitors of the RhoA/ROCK pathway might be useful therapeutic agents in the treatment of PH. In this context our findings are particularly noteworthy because they suggest that such inhibitors might act as a "double-edged sword". Successful attenuation of PH would be expected to relieve right ventricular strain and the resultant adverse consequences.1 However, reduction in hypoxia-induced angiogenesis in the lung might simultaneously have an adverse effect on pulmonary diffusing capacity for oxygen, which might be a previously unsuspected adverse consequence of such a therapeutic approach. This possible effect of ROCK inhibition will require further investigation.
In summary, we found that increased expression of ROCK is a key signaling event in hypoxia-induced vascular adaptations in the lung. Activation of this protein kinase leads to chronic hypoxic PH, not by mediating structural reduction of the diameter of the lumen of pulmonary blood vessels, but by causing sustained vasoconstriction. In contrast, the ROCK dependent capillary angiogenesis that we report here may be a beneficial adaptive response, improving oxygen uptake in the lung particularly during exercise.
Acknowledgments
This work was supported by the Health Research Board and the Program for Research in Third Level Institutions of the Higher Education Authority of Ireland.
Dr P. McLoughlin has received an unrestricted research grant from Actelion, a company that manufactures an endothelin receptor antagonist used in the treatment of primary pulmonary arterial hypertension.
Both authors contributed equally to this work.
References
Hopkins N, McLoughlin P. The structural basis of pulmonary hypertension in chronic lung disease: remodelling, rarefaction or angiogenesis J Anat. 2002; 201: 335eC348.
Howell K, Preston RJ, McLoughlin P. Chronic hypoxia causes angiogenesis in addition to remodelling in the adult rat pulmonary circulation. J Physiol. 2003; 547: 133eC145.
Sylvester JT. The tone of pulmonary smooth muscle: ROK and Rho music Am J Physiol Lung Cell Mol Physiol. 2004; 287: L624eC30.
Fu X, Gong MC, Jia T, Somlyo AV, Somlyo AP. The effects of the Rho-kinase inhibitor Y-27632 on arachidonic acid-, GTPgammaS-, and phorbol ester-induced Ca2+-sensitization of smooth muscle. FEBS Lett. 1998; 440: 183eC187.
Hyvelin JM, O’Connor C, McLoughlin P. Effect of changes in pH on wall tension in isolated rat pulmonary artery: role of the RhoA/Rho-kinase pathway. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L673eCL684.
Somlyo AP, Somlyo AV. Signal transduction by G-proteins, rho-kinase and protein phosphatase to smooth muscle and non-muscle myosin II. J Physiol. 2000; 522 Pt 2: 177eC185.
Ridley AJ. Rho family proteins: coordinating cell responses. Trends Cell Biol. 2001; 11: 471eC477.
Somlyo AP, Somlyo AV. Ca2+ sensitivity of smooth muscle and nonmuscle myosin II: modulated by G proteins, kinases, and myosin phosphatase. Physiol Rev. 2003; 83: 1325eC1358.
Mallat Z, Gojova A, Sauzeau V, Brun V, Silvestre JS, Esposito B, Merval R, Groux H, Loirand G, Tedgui A. Rho-associated protein kinase contributes to early atherosclerotic lesion formation in mice. Circ Res. 2003; 93: 884eC888.
Hattori T, Shimokawa H, Higashi M, Hiroki J, Mukai Y, Kaibuchi K, Takeshita A. Long-term treatment with a specific Rho-kinase inhibitor suppresses cardiac allograft vasculopathy in mice. Circ Res. 2004; 94: 46eC52.
Uehata M, Ishizaki T, Satoh H, Ono T, Kawahara T, Morishita T, Tamakawa H, Yamagami K, Inui J, Maekawa M, Narumiya S. Calcium sensitization of smooth muscle mediated by a Rho-associated protein kinase in hypertension. Nature. 1997; 389: 990eC994.
Uchida S, Watanabe G, Shimada Y, Maeda M, Kawabe A, Mori A, Arii S, Uehata M, Kishimoto T, Oikawa T, Imamura M. The suppression of small GTPase rho signal transduction pathway inhibits angiogenesis in vitro and in vivo. Biochem Biophys Res Commun. 2000; 269: 633eC640.
van Nieuw Amerongen GP, Koolwijk P, Versteilen A, van Hinsbergh VW. Involvement of RhoA/Rho kinase signaling in VEGF-induced endothelial cell migration and angiogenesis in vitro. Arterioscler Thromb Vasc Biol. 2003; 23: 211eC217.
Hoang MV, Whelan MC, Senger DR. Rho activity critically and selectively regulates endothelial cell organization during angiogenesis. Proc Natl Acad Sci U S A. 2004; 101: 1874eC1879.
Somlyo AV, Phelps C, Dipierro C, Eto M, Read P, Barrett M, Gibson JJ, Burnitz MC, Myers C, Somlyo AP. Rho kinase and matrix metalloproteinase inhibitors cooperate to inhibit angiogenesis and growth of human prostate cancer xenotransplants. FASEB J. 2003; 17: 223eC234.
Ooi H, Cadogan E, Sweeney M, Howell K, O’Regan RG, McLoughlin P. Chronic hypercapnia inhibits hypoxic pulmonary vascular remodeling. Am J Physiol Heart Circ Physiol. 2000; 278: H331eCH338.
Cadogan E, Hopkins N, Giles S, Bannigan JG, Moynihan J, McLoughlin P. Enhanced expression of inducible nitric oxide synthase without vasodilator effect in chronically infected lungs. Am J Physiol. 1999; 277: L616eCL627.
Meyrick B, Reid L. Hypoxia and incorporation of 3H-thymidine by cells of the rat pulmonary arteries and alveolar wall. Am J Pathol. 1979; 96: 51eC70.
Rabinovitch M, Gamble W, Nadas AS, Miettinen OS, Reid L. Rat pulmonary circulation after chronic hypoxia: hemodynamic and structural features. Am J Physiol. 1979; 236: H818eCH827.
Gundersen HJ, Bagger P, Bendtsen TF, Evans SM, Korbo L, Marcussen N, Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, et al. The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS. 1988; 96: 857eC881.
Fagan KA, Oka M, Bauer NR, Gebb SA, Ivy DD, Morris KG, McMurtry IF. Attenuation of acute hypoxic pulmonary vasoconstriction and hypoxic pulmonary hypertension in mice by inhibition of Rho-kinase. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L656eCL664.
Nagaoka T, Morio Y, Casanova N, Bauer N, Gebb S, McMurtry I, Oka M. Rho/Rho kinase signaling mediates increased basal pulmonary vascular tone in chronically hypoxic rats. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L665eCL672.
Pascaud MA, Griscelli F, Raoul W, Marcos E, Opolon P, Raffestin B, Perricaudet M, Adnot S, Eddahibi S. Lung overexpression of angiostatin aggravates pulmonary hypertension in chronically hypoxic mice. Am J Respir Cell Mol Biol. 2003; 29: 449eC457.
Beppu H, Ichinose F, Kawai N, Jones RC, Yu PB, Zapol WM, Miyazono K, Li E, Bloch KD. BMPR-II heterozygous mice have mild pulmonary hypertension and an impaired pulmonary vascular remodeling response to prolonged hypoxia. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L1241eCL1247.
Somlyo AV, Bradshaw D, Ramos S, Murphy C, Myers CE, Somlyo AP. Rho-kinase inhibitor retards migration and in vivo dissemination of human prostate cancer cells. Biochem Biophys Res Commun. 2000; 269: 652eC659.
Cascone I, Giraudo E, Caccavari F, Napione L, Bertotti E, Collard JG, Serini G, Bussolino F. Temporal and spatial modulation of Rho GTPases during in vitro formation of capillary vascular network. Adherens junctions and myosin light chain as targets of Rac1 and RhoA. J Biol Chem. 2003; 278: 50702eC50713.
Weibel ER. Understanding the limitation of O2 supply through comparative physiology. Respir Physiol. 1999; 118: 85eC93.
Jernigan NL, Walker BR, Resta TC. Chronic hypoxia augments protein kinase G-mediated Ca2+ desensitization in pulmonary vascular smooth muscle through inhibition of RhoA/Rho kinase signaling. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L1220eCL1229.
Bailly K, Ridley AJ, Hall SM, Haworth SG. RhoA activation by hypoxia in pulmonary arterial smooth muscle cells is age and site specific. Circ Res. 2004; 94: 1383eC1391.
Takemoto M, Sun J, Hiroki J, Shimokawa H, Liao JK. Rho-kinase mediates hypoxia-induced downregulation of endothelial nitric oxide synthase. Circulation. 2002; 106: 57eC62.(Jean-Marc Hyvelin, Kather)
Abstract
Pulmonary hypertension (PH) is a common complication of chronic hypoxic lung diseases, which increase morbidity and mortality. Hypoxic PH has previously been attributed to structural changes in the pulmonary vasculature including narrowing of the vascular lumen and loss of vessels, which produce a fixed increase in resistance. Using quantitative stereology, we now show that chronic hypoxia caused PH and remodeling of the blood vessel walls in rats but that this remodeling did not lead to structural narrowing of the vascular lumen. Sustained inhibition of the RhoA/Rho-kinase pathway throughout the period of hypoxic exposure attenuated PH and prevented remodeling in intra-acinar vessels without enlarging the structurally determined lumen diameter. In chronically hypoxic lungs, acute Rho kinase inhibition markedly decreased PVR but did not alter the alveolar to arterial oxygen gap. In addition to increased vascular resistance, chronic hypoxia induced Rho kinaseeCdependent capillary angiogenesis. Thus, hypoxic PH was not caused by fixed structural changes in the vasculature but by sustained vasoconstriction, which was largely Rho kinase dependent. Importantly, this vasoconstriction had no role in ventilation-perfusion matching and optimization of gas exchange. Rho kinase also mediated hypoxia-induced capillary angiogenesis, a previously unrecognized but potentially important adaptive response.
Key Words: pulmonary hypertension angiogenesis RhoA Rho-kinase Y-27632
Introduction
Sustained pulmonary hypertension (PH) is a common complication of chronic hypoxic lung diseases that is strongly associated with increased morbidity and reduced survival. Moreover, the presence of cor pulmonale is an independent predictor of increased mortality, suggesting that PH contributes directly to mortality (reviewed in Hopkins et al1). The increase in pulmonary vascular resistance (PVR) caused by chronic hypoxia has previously been attributed to structural changes in the vasculature, in particular encroachment of the remodeled arteriolar walls into the vascular lumen and loss of blood vessels, although recent reports have cast doubt on this paradigm.1eC3 In particular, we have recently shown for the first time that hypoxia induces angiogenesis in the adult pulmonary circulation, a potentially beneficial adaptation, and does not cause vessel loss as previously believed.2
The small G-protein RhoA and its downstream effector Rho-kinase (ROCK) play a central role in diverse cellular functions including smooth muscle contraction, cytoskeletal rearrangement, cell migration, cell proliferation, and gene expression.4eC8 Given these important functions, it is not surprising that disturbances of this pathway have been identified as important pathogenetic mechanisms in many diseases of the cardiovascular system, including systemic hypertension, arteriosclerosis, and ischemic heart disease.9eC11 Blockade of the RhoA/ROCK pathway effectively corrects blood pressure in a number of animal models of systemic hypertension8,11 and is a key regulator of vascular smooth muscle proliferation and migration in disease-induced systemic vascular remodeling.9,10 This pathway also regulates endothelial cell proliferation and migration. In vitro models and in vivo Matrigel implants show that inhibition of RhoA or ROCK prevents growth factoreCinduced endothelial cell migration and organization into capillary-like structures12,13 and vessel formation.14 Moreover, ROCK inhibitors attenuate tumor growth in vivo in a manner that is compatible with an antiangiogenic effect.15 Recognition of the therapeutic potential of inhibiting this pathway has lead to the development of specific small molecule inhibitors of ROCK.8,11,15
Taken together, these reports suggest that the RhoA/ROCK pathway might contribute to the development of hypoxic PH and the associated pulmonary vascular remodeling. The purpose of this study was to assess the potential role of this pathway in the development of chronic hypoxia-induced PH and the associated structural changes in the pulmonary vasculature. In particular, we wished to test the hypothesis that, although chronic inhibition of ROCK would inhibit the development of hypoxic PH, it would simultaneously inhibit hypoxia-induced pulmonary angiogenesis.
Materials and Methods
An expanded Materials and Methods section can be found in the online supplement available at http://circres.ahajournals.org.
Hypoxic Pulmonary Hypertension
Chronic hypoxic PH was induced by housing adult male rats in a normobaric hypoxic chamber (FiO2 0.10) for 1 or 3 weeks as previously described.16 Control rats were housed in normoxic conditions (FiO2 0.21) in the same room. To assess the effects of chronic ROCK inhibition, rats were randomized into 3 groups; 1 group was maintained in normoxic conditions (control group), while 2 groups were housed in hypoxic conditions for 1 week. Y-27632 was used to inhibit the phosphorylation activity of ROCK.4eC8,17 One hypoxic group (CH-Y27632) received Y-27632 (30 mg · kgeC1 · dayeC1, orally) while the other did not receive Y-27632 (CH group).
Measurement of Vascular Pressures and Gas Exchange In Vivo
Rats were anesthetized, tracheostomised, paralyzed, and mechanically ventilated. Venous and arterial cannulae were placed for drug administration, blood gas sampling, and blood pressure measurement. Right ventricular pressure was measured by introduction of a needle via a subdiaphragmatic approach just before euthanasia.
Isolated Perfused Lung Preparation
Lungs were isolated under anesthesia, mechanically ventilated, and perfused at constant low flow with a mixture of blood and physiological saline of constant hematocrit so that pulmonary arterial pressure (PAP) was a direct index of PVR.17 Capillary pressure was assessed by the double occlusion technique.17
Tissue Collection
At the end of each isolated lung experiment, the left lung (LL) was fixed under standard conditions for stereological quantification of structural vascular changes.2 The right lungs were snap frozen for subsequent protein and mRNA analysis.
Data Analysis
For stereological analyses, total volume of specific compartments, vessel and capillary lengths, and alveolar epithelial and capillary endothelial surface areas were reported per left lung. For normally distributed data, responses were reported as mean±SEM, n refers to the number of rats from which tissue was obtained. For multiple comparisons of means across experimental groups, analysis of variance was performed followed by StudenteCNewman Keuls post-hoc test for pair-wise comparisons. For non-normally distributed data, multiple comparisons of medians across experimental groups were performed using the KruskaleCWallis test followed by ManneCWhitney U with the Bonferonni post-hoc correction; values of P<0.05 were accepted as statistically significant.
Results
Effect of Acute ROCK Inhibition on Established Hypoxic Pulmonary Hypertension
Mean peak right ventricular systolic pressure (RVSPpeak) measured during ventilation with 30% oxygen (53.5±3.5 mm Hg) in rats (n=5) maintained chronically hypoxic for 3 weeks was significantly (P<0.01) greater than that in control nonhypoxic rats (28.8±1.4 mm Hg), demonstrating the presence of fixed PH; ie, increased PVR that was not attributable to acute hypoxic pulmonary vasoconstriction (HPV). After administration of the potent ROCK inhibitor Y27632 (15 mg · kgeC1),11 mean RVSPPeak was reduced significantly (P<0.02) to a value of 34.6 (±5.6) mm Hg.
Changes in RVSPPeak in response to Y27632 in vivo may not have resulted from changes in PVR but may have resulted from changes in venous return, cardiac output, left atrial pressure, neurogenic control, and other factors. To examine the direct effect of ROCK inhibition on PVR in lungs chronically hypoxic for 3 weeks, we used an isolated lung preparation. Mean PAP in chronically hypoxic lungs (n=5) was 15.1 (±0.6) mm Hg and was significantly reduced (P<0.01) to 13.0 (±0.9) mm Hg by Y27632 (3 e.leC1). At the highest Y27632 concentration (100 e.leC1) mean PAP was further reduced (P<0.01) to 10.7 (±0.4) mm Hg, a value that was similar to that in control nonhypoxic lungs (see below and Figure 1).
Thus the dominant mechanism underlying chronic hypoxic PH was sustained vasoconstriction and not structural changes in the pulmonary vasculature. These results suggested that sustained ROCK inhibition in vivo would prevent the development of chronic hypoxic PH.
Chronic ROCK Inhibition In Vivo
To confirm that chronic exposure to a hypoxic environment for 1 week caused fixed PH, we measured RVSPPeak in anesthetized rats (n=5) during ventilation with 30% oxygen, and found that after such exposure RVSPPeak was significantly (P<0.01) elevated (42.5±3.0 mm Hg) when compared with control nonhypoxic rats (28.8±1.4 mm Hg). For the reasons outlined above, further investigation of the role of ROCK in the development of PH was undertaken using the isolated lung preparation.
Mean right ventricular (RV) weight (see supplemental Table I) in a group maintained hypoxic for 1 week (CH group) and in a hypoxic group chronically treated with Y-27632 (CH-Y27632) for 1 week were both significantly higher than that of the control group. The sustained treatment with ROCK inhibitor significantly reduced the RV hypertrophy compared with hypoxia alone although it did not alter the polycythemic response to hypoxia (supplemental Table I).
We assessed the total PVR in isolated lungs under 3 different conditions, ie, lungs ventilated with 16%, 7%, or 3% O2 (Figure 1). These concentrations give alveolar values that approximate normal sea level alveolar pO2, the alveolar pO2 experienced by the rats during hypoxic exposure in the chamber, and an alveolar pO2 that produces a maximal hypoxic vasoconstrictor response. Lungs from the control group showed normal low PVR when ventilated with 16% O2 and significant increases in resistance in response to reduction in alveolar O2 indicating normal HPV (Figure 1). Chronically hypoxic lungs ventilated with 7% O2 had significantly elevated vascular resistance compared with control lungs ventilated with the same inspired oxygen, and this resistance was not reduced by acute return to normal alveolar oxygen concentration (16% O2). Reduction to 3% O2 in the chronically hypoxic lungs caused a further small increase in resistance to a value similar to that seen in control lungs at the same inspired oxygen (Figure 1). Chronic administration of the ROCK inhibitor Y-27632 prevented the development of PH in response to sustained exposure to hypoxia and also abolished the acute hypoxic vasoconstrictor response (Figure 1).
To evaluate the contribution of ROCK-dependent vasoconstriction to the control of PVR in these chronically hypoxic and control lungs, Y-27632 was added to the perfusate up to 3 eol/L, a concentration that selectively inhibited ROCK.8,11 In normoxic lungs, ventilated with 16%O2, Y-27632 up to 3 eol · leC1 did not change the PAP (Figure 2A), demonstrating that ROCK activity did not play a role in the control of vascular resistance in the normal lung.
In the CH group, acute administration of Y-27632 markedly decreased the PAP in a dose-dependent fashion (Figure 2A). This concentration of ROCK inhibitor abrogated approximately two-thirds of the increase induced by chronic hypoxia through an effect that was mainly localized to the arterial (precapillary) segment (Figure 2B). This demonstrated that a vasoconstrictor activity of ROCK played a key role in maintaining chronic hypoxic PH.
Finally, in the CH-Y27632 group, in which the baseline PAP was similar to that in the control group, acute addition of Y-27632 had no significant effect on PA pressure in the isolated lungs, demonstrating that the chronic treatment with Y-27632 had maximally inhibited the effect of ROCK activity on PVR (Figure 2A).
In a separate series of experiments, we assessed the effect of higher, potentially nonselective8,11 concentrations of Y-27632 (up to 100 eol/L) on the PAP in lungs maintained chronically hypoxic for 1 week and found that PAP was decreased to 10.1 (±0.2) mm Hg (n=3), a value similar to that in control lungs (9.6±0.3 mm Hg). These results were similar to those obtained in the 3-week hypoxic group and supported the conclusion that the dominant mechanism underlying chronic hypoxic PH is sustained vasoconstriction.
In the presence of Y-27632 (3 eol · leC1), reducing the oxygen level to 3% did not produce significant pulmonary vasoconstriction in any group, indicating that ROCK is a key mediator of acute HPV (data not shown).
Effect of Acute ROCK Inhibition on Pulmonary Oxygen Uptake in Hypoxic Rats
In chronically hypoxic rats ventilated with a hypoxic inspirate (FiO2 0.10), there was no change in either the A-aO2 gap or the partial pressure of oxygen in arterial blood (PaO2) after acute administration of Y-27632 (see online supplement). The A-aO2 gap and PaO2 were unchanged after injection of vehicle (data not shown).
Effect of Chronic Inhibition of ROCK on Hypoxia-Induced Angiogenesis and Vascular Remodeling
The mean left lung volume was significantly increased in the CH group (3.28±0.15 mL) when compared with both the control group (2.54±0.12 mL, P<0.001) and the CH-Y-27632 group (2.98±0.05 mL, P<0.05).
In the control lungs, the majority of intra-acinar vessels were thin-walled, with a single elastic lamina and no discernable tunica media (see online supplement results). In the CH group remodeling of the wall of intra-acinar vessels was evident, as shown by the development of separate internal and external elastic laminae and an intervening layer of media (see online supplement results). After chronic inhibition of ROCK, the structure of vessel walls was similar to that observed in the control group suggesting inhibition of hypoxia-induced vascular remodeling.
In the CH group, the mean total length of intra-acinar vessels (9236±630 cm · LLeC1) was significantly higher than in the control group (5327±392 cm · LLeC1, P<0.001). After chronic inhibition of ROCK, the total length of intra-acinar vessels (7864±508 cm · LLeC1) was less than in the CH group although it remained significantly higher than in control group (P<0.01). Analysis of the mean lumen diameter of the intra-acinar vessels showed that there was no difference between the 3 groups (Figure 3A). However, the ratio of the wall thickness to the lumen diameter was significantly increased in the CH-group when compared with control and CH-Y27632 lungs (Figure 3B).
The total capillary volume and total capillary length (Table 1) were significantly increased in the CH group when compared with the control group, whereas after chronic treatment with Y-27632 these hypoxia-induced increases were significantly reduced and not significantly different from those in control group. The total capillary endothelial surface area was also significantly increased in the CH lungs when compared with the control group (Table 2). In the CH-Y-27632 group, capillary endothelial surface area was significantly less than that in the CH group although it remained elevated above the control value (Table 2). Chronic hypoxia also caused a significant increase in the total alveolar epithelial surface area when compared with the control group (Table 2), although in the CH-Y-27632 group this value was significantly less that in the CH group and similar to that in the control group. In both hypoxic groups, the increase in capillary endothelial surface area was proportionately greater than that in the epithelial surface area as indicated by the increase in the mean ratio of capillary endothelium to alveolar epithelium (Table 2).
Effect of Hypoxia on RhoA, RhoA Activation, and ROCK Expression
The total amount of RhoA was similar in each of the different groups of rats; ie, control, CH, and CH-Y-27632 (Figure 4). RhoA activation, expressed as the percentage of total RhoA in GTP-RhoA form, showed no significant difference between the different groups (Figure 4). In CH and CH-Y-27632 groups levels of ROCK I and II isoforms were significantly increased when compared with the control group (Figure 5).
Effect of Chronic ROCK Inhibition on Vascular Growth Factor Gene Expression
Chronic hypoxia increased expression of platelet-derived growth factor- (PDGF-), and this effect was significantly reduced by ROCK inhibition (Figure 6). Chronic ROCK inhibition also significantly reduced transforming growth factor -1 (TGF-1) expression when compared with chronically hypoxic lungs (Figure 6). Interestingly, expression of the proangiogenic genes vascular endothelial growth factor A (VEGF-A) and angiopoeitin-1 (ANG-1) was not significantly altered at the end of one week of hypoxia and was unaffected by ROCK inhibition.
Discussion
It has been generally thought that PH predominantly resulted from hypoxia-induced structural changes in the pulmonary vasculature, which produced a fixed increase in resistance. These structural changes included remodeling of the arteriolar walls leading to encroachment into the vascular lumen, and loss of blood vessels, suggesting that interventions, which successfully ameliorated PH, would act by preventing or reversing such structural changes. The results of the present study cast doubt on this paradigm by demonstrating that chronic hypoxia did not cause a structurally based reduction of pulmonary vascular lumen diameter, nor did it cause a loss of pulmonary vessels. Moreover, we report that chronic inhibition of ROCK abrogated the development of hypoxic PH, not by preventing structural encroachment into the vascular lumen, but by inhibiting sustained pulmonary vasoconstriction. Importantly, we also found hypoxia-induced capillary angiogenesis in the adult lung that was dependent on the RhoA/ROCK pathway. ROCK inhibition reduced the expression of key vascular growth factors suggesting that this was one of the mechanisms by which it attenuated hypoxia-induced vascular remodeling.
Initially, we confirmed that exposure to hypoxia for three weeks induced fixed PH in vivo; ie, increased RVSPPeak that was not immediately reversed on exposure to normal alveolar oxygen and thus was not mediated by acute hypoxic pulmonary vasoconstriction. Previously, this observation, together with the extensive remodeling of pulmonary arterioles, has been taken as evidence that the increased PVR in chronic hypoxic PH was predominantly attributable to structural changes in the vasculature. However, in our experiments administration of the ROCK inhibitor Y-27632 produced a marked fall in PAP suggesting that ROCK-dependent vasoconstriction was the major determinant of hypoxic PH.
Because the changes in peak right ventricular systolic pressure in response to Y-27632 in vivo might not have been attributable to changes in PVR but might have resulted from changes in other parameters, we used the isolated ventilated lung preparation to examine the direct effect of ROCK inhibition on PVR. Although the PA pressure in such preparations is much lower than in vivo values, it is a direct index of PVR. Our finding that Y-27632 caused PAP to fall immediately to near normal values in lungs made chronically hypoxic for 3 weeks demonstrated that ROCK inhibition normalized PVR by a direct action on pulmonary resistance vessels and strongly suggested that hypoxic PH was not structurally based.
In view of the evidence that Y-27632 could acutely reverse chronic hypoxic PH, we examined the effect of chronic Y-27632 administration during 1 week of exposure to hypoxia. In agreement with previous reports,18,19 we first confirmed that exposure to hypoxia for 1 week produced marked PH in vivo, which persisted when alveolar oxygen was acutely elevated to supra-normal values (see Results), a finding that was confirmed in isolated lung preparations (Figure 1). In addition, we observed extensive vascular wall remodeling in CH lungs (Figure 3), as previously demonstrated after 1 week of hypoxia.18,19 Thus, these lungs demonstrated the features that had previously been interpreted as indicating structurally based PH. However, our stereological analysis demonstrated other hypoxia-induced structural changes in the pulmonary vasculature (angiogenesis and wall remodeling without lumen reduction) that differ radically from those suggested by the conventional paradigm (Figure 3). Why have such structural changes not been previously detected The techniques we used to analyze vascular structure differ in 2 important ways from those previously used to address this problem, as we have previously discussed.2 First, stereology allows quantitative analysis of 3-dimensional structural parameters based on the 2-dimensional information available in histological sections and takes account of the changes in whole lung volume caused by chronic hypoxia.20 Second, vessels were recruited and fixed under no-flow conditions at a defined transmural distending pressure that was constant at all points in the vasculature, was unaffected by changes in vascular resistance, and was identical in both control and hypertensive lungs. Thus the vascular dimensions measured in this study reflect structural change and were unaffected by pre-existing smooth muscle tone in vivo.
Our observation that chronic hypoxia did not cause encroachment of the remodeled vascular wall into the lumen is in good agreement with our previous report but extends those findings in 2 important ways.2 First, we show that this outward remodeling of the vessel wall occurred in lungs in which we had hemodynamically confirmed the development of fixed PH (Figure 1). Second, we found that chronic administration of Y-27632, a selective ROCK blocker,11 inhibited the development of increased PVR in response to chronic hypoxia (Tables 1 and 2) and attenuated right ventricular hypertrophy in vivo (supplemental Table I), without causing structural enlargement of the lumen (Figure 3). We confirmed that the dose used chronically in vivo was effective in blocking the contractile activity of the ROCK pathway by demonstrating that after isolation, these lungs showed no additional vasodilator response to acute Y-27632 administration (Figure 2). The failure of chronic ROCK inhibition to significantly enlarge mean-maximal-lumen diameter was observed, despite the fact that it successfully attenuated wall remodeling, a finding that is supported by the recent report of Fagan et al.21 Taken together, our results show that chronic hypoxic PH develops without structural change in the mean-maximal-vascular-lumen diameter and, furthermore, PH can be prevented without enlarging the structurally determined lumen diameter. This suggests that chronic hypoxic PH is not caused by structurally imposed changes in vascular resistance.
This conclusion is further reinforced by the demonstration that acute administration of a high concentration of Y-27632 normalized PVR in isolated lungs made chronically hypertensive by exposure to 1 and 3 weeks of hypoxia, both in vivo and in isolated lungs (Figure 2 and see Results), results that confirm the recent report of Nagaoka et al.22 Evidence supporting the important role of RhoA/ROCK is provided by the demonstration of a marked reduction in PVR (Figure 2) at concentrations of Y-27632 that are selective for ROCK (3 micromolar or less),8,11 although these concentrations did not return pulmonary arterial pressure to normal values. It has been previously shown that Y-27632 relaxes vascular smooth muscle, preconstricted by either agonists or hypoxia, by blocking ROCK-mediated phosphorylation of the regulatory protein CPI-17 and the MYPT-1 of myosin light chain phosphatase (MLCP), both of which effects activate MLCP, thus reducing myosin light chain phosphorylation and relaxing smooth muscle.15,21 Administration of higher concentrations of Y-27632 returned PVR to control values, although this vasodilator action might not have resulted from specific ROCK inhibition.5,8 Clearly, acute normalization could not occur if vascular resistance was increased by a fixed structural mechanism.
The RhoA/ROCK-dependent sustained vasoconstriction observed in chronically hypoxic lungs is distinct from that mediating acute HPV because acute normalization of alveolar oxygen did not reduce PVR after either 1 or 3 weeks of hypoxic exposure. This suggested that the sustained vasoconstriction of chronic hypoxia did not have a role in ventilation-perfusion matching. To test this directly, we examined the effect of acute ROCK inhibition on the alveolar-arterial oxygen gap, an index of physiological shunting, and found that this was unaltered (see online supplement). This suggests that chronic ROCK-mediated hypoxic PH is a maladaptive response, because it overloads the right ventricle without improving gas exchange and is in marked contrast to the beneficial optimization of ventilation-perfusion matching produced by acute HPV.
In the present study, we found that chronic hypoxia induced capillary angiogenesis in the adult lung in vivo (Tables 1 and 2 ), confirming the results of our previous report.2 In support of this Pascaud et al reported that angiostatin, an inhibitor of angiogenesis, aggravated PH in the hypoxic lung, an observation that suggests new vessel formation in hypoxic lungs.23 More recently, Beppu and colleagues demonstrated hypoxia-induced capillary angiogenesis in the adult mouse lung.24 We now extend those findings by reporting that chronic ROCK inhibition significantly attenuated this hypoxia-induced angiogenesis (Tables 1 and 2). Inhibition of RhoA or ROCK has previously been shown to prevent growth factoreCinduced endothelial cell migration and organization into capillary-like structures in vitro and vessel formation in Matrigel implant.12,13,15,25,26 Our finding that Y-27632 attenuated hypoxia-induced angiogenesis in the lung is, to our knowledge, the first report directly demonstrating that the RhoA/ROCK pathway has a central role in hypoxic angiogenesis within adult tissues in vivo. In the lung, this may be particularly important as hypoxic angiogenesis increases the diffusing capacity for oxygen and may therefore be an important adaptive response to hypoxia.2 The possibility that hypoxia-induced angiogenesis might improve oxygen uptake in the lung awaits direct functional evidence. However, studies in high altitude and burrowing animals adapted to hypoxic environments show that the enlarged pulmonary capillary bed found in such species enhances oxygen uptake during exercise in hypoxia when oxygen uptake becomes diffusion limited, but not at rest when uptake is perfusion limited.27
The action of ROCK in the chronically hypoxic lung was not the result of increased expression of RhoA, a finding in good agreement with the recent report of Jernigan et al.28 Moreover, ROCK activity could not have resulted from increased RhoA activation because GTP-RhoA was reduced (although not significantly) and not increased (Figure 4). This result is similar to that of Bailly et al who showed reduced RhoA activation in chronically hypoxic pulmonary vascular smooth muscle,29 although increased RhoA activation has been reported in chronically hypoxic lungs after 4 weeks.28 The ROCK effects that we observed may have been mediated through the increased expression of both its isoforms (Figure 5), a finding in good agreement with the demonstration that hypoxia augments expression of this enzyme in cultured pulmonary smooth muscle cells.30
ROCK activity is an important regulator of cytoskeletal rearrangement in endothelial cells during migration, and blockade of this action inhibits in vitro organization of endothelial cells into capillary structures.14,26 Such inhibition of cytoskeletal reorganization could have contributed to the attenuated angiogenesis that we observed in the CH-Y-27632 group. We explored the possibility that inhibition of remodeling and angiogenesis might also have been mediated by altered expression of key vascular growth factors. Our finding that chronic Y-27632 treatment reduced PDGF- and TGF-1expression (Figure 6) suggests that alteration of these growth factors might be an important mechanism by which ROCK inhibition attenuated arteriolar wall remodeling. Interestingly, neither chronic hypoxia nor ROCK inhibition changed the expression of VEGF-A or ANG-1 at 1 week (Figure 6). It is possible that expression of these genes might have been altered by hypoxia at earlier time points and that Y-27632 had blocked these changes, thus inhibiting angiogenesis. Alternatively other proangiogenic genes may play a predominant role in the lung.
It has recently been suggested that inhibitors of the RhoA/ROCK pathway might be useful therapeutic agents in the treatment of PH. In this context our findings are particularly noteworthy because they suggest that such inhibitors might act as a "double-edged sword". Successful attenuation of PH would be expected to relieve right ventricular strain and the resultant adverse consequences.1 However, reduction in hypoxia-induced angiogenesis in the lung might simultaneously have an adverse effect on pulmonary diffusing capacity for oxygen, which might be a previously unsuspected adverse consequence of such a therapeutic approach. This possible effect of ROCK inhibition will require further investigation.
In summary, we found that increased expression of ROCK is a key signaling event in hypoxia-induced vascular adaptations in the lung. Activation of this protein kinase leads to chronic hypoxic PH, not by mediating structural reduction of the diameter of the lumen of pulmonary blood vessels, but by causing sustained vasoconstriction. In contrast, the ROCK dependent capillary angiogenesis that we report here may be a beneficial adaptive response, improving oxygen uptake in the lung particularly during exercise.
Acknowledgments
This work was supported by the Health Research Board and the Program for Research in Third Level Institutions of the Higher Education Authority of Ireland.
Dr P. McLoughlin has received an unrestricted research grant from Actelion, a company that manufactures an endothelin receptor antagonist used in the treatment of primary pulmonary arterial hypertension.
Both authors contributed equally to this work.
References
Hopkins N, McLoughlin P. The structural basis of pulmonary hypertension in chronic lung disease: remodelling, rarefaction or angiogenesis J Anat. 2002; 201: 335eC348.
Howell K, Preston RJ, McLoughlin P. Chronic hypoxia causes angiogenesis in addition to remodelling in the adult rat pulmonary circulation. J Physiol. 2003; 547: 133eC145.
Sylvester JT. The tone of pulmonary smooth muscle: ROK and Rho music Am J Physiol Lung Cell Mol Physiol. 2004; 287: L624eC30.
Fu X, Gong MC, Jia T, Somlyo AV, Somlyo AP. The effects of the Rho-kinase inhibitor Y-27632 on arachidonic acid-, GTPgammaS-, and phorbol ester-induced Ca2+-sensitization of smooth muscle. FEBS Lett. 1998; 440: 183eC187.
Hyvelin JM, O’Connor C, McLoughlin P. Effect of changes in pH on wall tension in isolated rat pulmonary artery: role of the RhoA/Rho-kinase pathway. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L673eCL684.
Somlyo AP, Somlyo AV. Signal transduction by G-proteins, rho-kinase and protein phosphatase to smooth muscle and non-muscle myosin II. J Physiol. 2000; 522 Pt 2: 177eC185.
Ridley AJ. Rho family proteins: coordinating cell responses. Trends Cell Biol. 2001; 11: 471eC477.
Somlyo AP, Somlyo AV. Ca2+ sensitivity of smooth muscle and nonmuscle myosin II: modulated by G proteins, kinases, and myosin phosphatase. Physiol Rev. 2003; 83: 1325eC1358.
Mallat Z, Gojova A, Sauzeau V, Brun V, Silvestre JS, Esposito B, Merval R, Groux H, Loirand G, Tedgui A. Rho-associated protein kinase contributes to early atherosclerotic lesion formation in mice. Circ Res. 2003; 93: 884eC888.
Hattori T, Shimokawa H, Higashi M, Hiroki J, Mukai Y, Kaibuchi K, Takeshita A. Long-term treatment with a specific Rho-kinase inhibitor suppresses cardiac allograft vasculopathy in mice. Circ Res. 2004; 94: 46eC52.
Uehata M, Ishizaki T, Satoh H, Ono T, Kawahara T, Morishita T, Tamakawa H, Yamagami K, Inui J, Maekawa M, Narumiya S. Calcium sensitization of smooth muscle mediated by a Rho-associated protein kinase in hypertension. Nature. 1997; 389: 990eC994.
Uchida S, Watanabe G, Shimada Y, Maeda M, Kawabe A, Mori A, Arii S, Uehata M, Kishimoto T, Oikawa T, Imamura M. The suppression of small GTPase rho signal transduction pathway inhibits angiogenesis in vitro and in vivo. Biochem Biophys Res Commun. 2000; 269: 633eC640.
van Nieuw Amerongen GP, Koolwijk P, Versteilen A, van Hinsbergh VW. Involvement of RhoA/Rho kinase signaling in VEGF-induced endothelial cell migration and angiogenesis in vitro. Arterioscler Thromb Vasc Biol. 2003; 23: 211eC217.
Hoang MV, Whelan MC, Senger DR. Rho activity critically and selectively regulates endothelial cell organization during angiogenesis. Proc Natl Acad Sci U S A. 2004; 101: 1874eC1879.
Somlyo AV, Phelps C, Dipierro C, Eto M, Read P, Barrett M, Gibson JJ, Burnitz MC, Myers C, Somlyo AP. Rho kinase and matrix metalloproteinase inhibitors cooperate to inhibit angiogenesis and growth of human prostate cancer xenotransplants. FASEB J. 2003; 17: 223eC234.
Ooi H, Cadogan E, Sweeney M, Howell K, O’Regan RG, McLoughlin P. Chronic hypercapnia inhibits hypoxic pulmonary vascular remodeling. Am J Physiol Heart Circ Physiol. 2000; 278: H331eCH338.
Cadogan E, Hopkins N, Giles S, Bannigan JG, Moynihan J, McLoughlin P. Enhanced expression of inducible nitric oxide synthase without vasodilator effect in chronically infected lungs. Am J Physiol. 1999; 277: L616eCL627.
Meyrick B, Reid L. Hypoxia and incorporation of 3H-thymidine by cells of the rat pulmonary arteries and alveolar wall. Am J Pathol. 1979; 96: 51eC70.
Rabinovitch M, Gamble W, Nadas AS, Miettinen OS, Reid L. Rat pulmonary circulation after chronic hypoxia: hemodynamic and structural features. Am J Physiol. 1979; 236: H818eCH827.
Gundersen HJ, Bagger P, Bendtsen TF, Evans SM, Korbo L, Marcussen N, Moller A, Nielsen K, Nyengaard JR, Pakkenberg B, et al. The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS. 1988; 96: 857eC881.
Fagan KA, Oka M, Bauer NR, Gebb SA, Ivy DD, Morris KG, McMurtry IF. Attenuation of acute hypoxic pulmonary vasoconstriction and hypoxic pulmonary hypertension in mice by inhibition of Rho-kinase. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L656eCL664.
Nagaoka T, Morio Y, Casanova N, Bauer N, Gebb S, McMurtry I, Oka M. Rho/Rho kinase signaling mediates increased basal pulmonary vascular tone in chronically hypoxic rats. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L665eCL672.
Pascaud MA, Griscelli F, Raoul W, Marcos E, Opolon P, Raffestin B, Perricaudet M, Adnot S, Eddahibi S. Lung overexpression of angiostatin aggravates pulmonary hypertension in chronically hypoxic mice. Am J Respir Cell Mol Biol. 2003; 29: 449eC457.
Beppu H, Ichinose F, Kawai N, Jones RC, Yu PB, Zapol WM, Miyazono K, Li E, Bloch KD. BMPR-II heterozygous mice have mild pulmonary hypertension and an impaired pulmonary vascular remodeling response to prolonged hypoxia. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L1241eCL1247.
Somlyo AV, Bradshaw D, Ramos S, Murphy C, Myers CE, Somlyo AP. Rho-kinase inhibitor retards migration and in vivo dissemination of human prostate cancer cells. Biochem Biophys Res Commun. 2000; 269: 652eC659.
Cascone I, Giraudo E, Caccavari F, Napione L, Bertotti E, Collard JG, Serini G, Bussolino F. Temporal and spatial modulation of Rho GTPases during in vitro formation of capillary vascular network. Adherens junctions and myosin light chain as targets of Rac1 and RhoA. J Biol Chem. 2003; 278: 50702eC50713.
Weibel ER. Understanding the limitation of O2 supply through comparative physiology. Respir Physiol. 1999; 118: 85eC93.
Jernigan NL, Walker BR, Resta TC. Chronic hypoxia augments protein kinase G-mediated Ca2+ desensitization in pulmonary vascular smooth muscle through inhibition of RhoA/Rho kinase signaling. Am J Physiol Lung Cell Mol Physiol. 2004; 287: L1220eCL1229.
Bailly K, Ridley AJ, Hall SM, Haworth SG. RhoA activation by hypoxia in pulmonary arterial smooth muscle cells is age and site specific. Circ Res. 2004; 94: 1383eC1391.
Takemoto M, Sun J, Hiroki J, Shimokawa H, Liao JK. Rho-kinase mediates hypoxia-induced downregulation of endothelial nitric oxide synthase. Circulation. 2002; 106: 57eC62.(Jean-Marc Hyvelin, Kather)