Reversible or Irreversible Remodeling in Pulmonary Arterial Hypertension

Abstract

Vascular remodeling is an important pathological feature of pulmonary arterial hypertension (PAH), which leads to increased pulmonary vascular resistance, with marked proliferation of pulmonary artery smooth muscle cells (SMC) and/or endothelial cells (EC). Successful treatment of experimental PAH with a platelet-derived growth factor (PDGF) receptor tyrosine kinase inhibitor offers the perspective of “reverse remodeling” (i.e., the regression of established pulmonary vascular lesions). Here we ask the question: which forms of pulmonary vascular remodeling are reversible and can such remodeling caused by angiogenic proliferation of EC be reversed? It is important to emphasize that the report showing reduction of vascular remodeling by PDGF receptor tyrosine kinase inhibitor showed only a reduction of the pulmonary artery muscularization in chronic hypoxia and monocrotaline models, which lack the feature of clustered proliferated EC in the lumen of pulmonary arteries. The regression of vascular muscularization is an important manifestation, whereby proliferative adult SMC convert back to a nonproliferative state. In contrast, in vitro experiments assessing the contribution of EC to the development of PAH demonstrated that phenotypically altered EC generated as a consequence of a vascular endothelial growth factor receptor blockade did not reverse to normal EC. Whereas it is suggested that the proliferative state of SMC may be reversible, it remains unknown whether phenotypically altered EC can switch back to a normal monolayer-forming EC. This article reviews the pathogenetic concepts of severe PAH and explains the many forms in PAH with reversible or irreversible remodeling.

Pulmonary vascular remodeling is an important pathological feature of pulmonary arterial hypertension (PAH), which leads to increased pulmonary vascular resistance and reduced compliance, with marked proliferation of pulmonary artery smooth muscle cells (SMC) and/or endothelial cells (EC) resulting in the obstruction of blood flow in the resistance pulmonary arteries (1, 2). In a recent Perspective article, Rai and coworkers (3) characterized severe PAH as a quasi-neoplastic, angioproliferative disorder, and this concept provides a new framework for antiproliferative, antiangiogenic therapy in severe PAH.

Successful treatment with a platelet-derived growth factor (PDGF) receptor tyrosine kinase inhibitor of experimental pulmonary hypertension offers the prospect of “reverse remodeling” (i.e., the regression of established pulmonary vascular lesions) (4).

This review asks the question: which forms of pulmonary vascular remodeling are reversible and can the remodeling caused by the angiogenic proliferation of EC be reversed (5–8)? It is important to emphasize that the report showing a reduction of vascular remodeling by PDGF receptor tyrosine kinase inhibitor (4) showed only the reduction of pulmonary artery muscularization in chronic hypoxia and monocrotaline models, which lack the features of clustered proliferated EC in the lumen of pulmonary arteries (9, 10).

Further studies are necessary to characterize pulmonary vessel phenotypes to determine the reversibility of vascular remodeling and to stop angiogenesis in PAH.

SMC AND EC IN THE MANY FORMS OF PAH

Pulmonary vascular remodeling is characterized by the thickening of all three layers of the blood vessel wall (i.e., the adventitia, the media, and the intima). Such thickening is due to hypertrophy (cell growth) and/or hyperplasia (proliferation) of the predominant cell type within each of the layers, i.e., fibroblasts, SMC, and EC, as well as increased deposition of extracellular matrix components including collagen, elastin, and fibronectin (11). The thickening of the media occurs consistently in the arteries, at all levels of the pulmonary arterial tree, and less frequently in the veins (12). In addition, there is an extension of new smooth muscle into the partially muscular and nonmuscular peripheral arteries, and this is termed muscularization. This may be due to the differentiation of precursor cells (i.e., pericytes and “intermediate” cells, the latter being intermediate between pericytes and muscle cells in structure into SMC) (9). The other possibilities are the differentiation of fibroblasts and circulating mononuclear mesenchymal precursors into muscle cells (13) and the endothelial to mesenchymal transition (EMT) (14). The proliferation of pulmonary artery SMC in PAH is enhanced, whereas apoptosis is depressed (15–18). Many factors drive SMC proliferation, including bone morphogenetic protein receptor-2 (BMPR-2) mutations (19), the de novo expression of the anti-apoptotic protein survivin (15, 16), the increased expression/activity of the serotonin transporter (SERT) (20, 21) and the increased expression/activity of PDGF receptor. Dysfunctional voltage-sensitive potassium channels have been described in the pulmonary artery SMC from patients with idiopathic PAH (22), and impaired K⁺ channel performance has been linked to pulmonary vasoconstriction, whereas vasorelaxation due to nitric oxide and cyclic guanosine monophosphate has been linked to the activation of the Ca²⁺-dependent K⁺ channels (23). Moreover, it has been suggested that PAH has a cancer aspect because pulmonary artery SMC in PAH and cancer cells are both associated with mitochondrial disorders (24–26).

An additional feature that is seen in some forms of PAH in humans is a complex vascular lesion known as a plexiform lesion (27). These lesions contain a disorganized monoclonal EC proliferation in a stroma of myofibroblasts (28, 29). These so-called plexiform or complex vascular lesions are characterized by apoptosis-resistant (30–35), phenotypically altered EC (36–41). Recently, conclusive evidence that idiopathic PAH pulmonary artery EC have a hyperproliferative apoptosis-resistant phenotype compared with cells from control lungs has been shown using cell proliferation, DNA synthesis, and the evaluation of cell death pathways (42). Likewise, human herpes virus 8 (HHV-8) infection of pulmonary microvascular EC results in an apoptotic-resistant phenotype characteristic of severe PAH (43), although it is unclear whether HHV-8 has a pathogenetic role in idiopathic PAH (44–46). Moreover, dysfunctional endothelial progenitor cells, which are hyperproliferative with impaired ability to form vascular networks, are involved in the vascular remodeling associated with PAH (47).

An impairment of the EC function in patients with PAH leads to local thrombosis (48, 49), the interruption of which may improve the prognosis of patients with PAH. Moreover, the expression of tissue factor (TF), the membrane glycoprotein that initiates coagulation (50), facilitates angiogenesis (50), mediates arterial injury in the systemic circulation (51), and increases in the pulmonary arterioles and plexiform lesions of the rats, thus indicating that the induction of TF may contribute to the progression of PAH (52).

Xu and colleagues used an in vitro experiment with pulmonary artery EC from idiopathic PAH (IPAH EC) and control lungs (control EC) to show that glucose metabolism plays the primary role for energy requirements of IPAH EC based on the 3-fold greater glycolytic rate of IPAH EC compared with control EC, thus indicating that there is mitochondrial disorder in EC in idiopathic PAH, like SMC in PAH and cancer cells (53). A share of a mitochondrial disorder in SMC and EC in PAH may support the EMT hypothesis.

REVERSIBLE AND IRREVERSIBLE MODELS OF PULMONARY HYPERTENSION

In contrast to the human disease, both classical rodent models of mild to moderate pulmonary hypertension—the chronic hypoxia and monocrotaline models—lack clustered proliferated EC in the lumen of pulmonary arteries (9, 10). Pulmonary EC constitute a stable cell population with a very low turnover rate and, apparently, neither severe chronic hypoxia/hypoxemia nor monocrotaline pyrrole causes the emergence of a proliferative, dysfunctional EC phenotype (54). The defining pulmonary vascular alteration in both of these models, medial muscular thickening of proliferating SMC, is potentially reversible upon reexposure to normoxia or with the passage of time after monocrotaline injection (9, 11, 55).

Neointimal pulmonary vascular occlusive lesions that consist of proliferating SMC are found in rats after the combination of pneumonectomy with monocrotaline injection (56–60). A combination of compensatory lung growth after a pneumonectomy, hemodynamic factors, and endothelial injury by monocrotaline pyrrole may combine to produce this neointimal pulmonary vascular disease (60). Neointimal vascular occlusion is reversible with simvastatin in this rat model, through its antiproliferative and proapoptotic effects on vascular SMC (61).

A vascular endothelial growth factor (VEGF) receptor blockade with the VEGF-RI/VEGF-R II antagonist SU5416 combined with chronic hypoxia results in severe angioproliferative PAH in adult rats (39). The defining pulmonary vascular alteration in this model, arterial occlusion by proliferating EC, is not reversible upon reexposure to normoxia and with passage of time after SU5416 injection (39). Together, these rat models offer the perspective that medial muscular thickening due to proliferating SMC may be reversible and arterial occlusion by proliferating EC may conversely be irreversible.

REVERSAL OF MUSCULARIZATION

Although less well studied, the shift of vascular SMC, whereby proliferative adult vascular SMC convert back to a nonproliferative state, is an important observation. This particular phenotype switch is essential for limiting SMC accumulation and for the termination of vascular remodeling. The regulatory factors that drive proliferative SMC into a nonproliferative state, and hold them in that state, are critical for effective vascular remodeling and for limiting vascular disease (62). Li and coworkers generated unique lines of nonimmortalized human vascular SMC that are capable of a conversion from proliferative to nonproliferative states (63). In the presence of serum, these SMC proliferate, migrate, and elaborate extracellular matrix similar to primary SMC. Upon withdrawal of the serum, however, they undergo a reproducible program of cellular maturation whereby they exit the cell cycle, migrate into multilayered aggregates, and then acquire the ability to rapidly contract. Li and colleagues (63) concluded that this shift to a nonproliferative and mature phenotype may be attributed to cellular plasticity, rather than the selective expansion (or loss) of distinct cell subpopulations, because the SMC are clonal. Although this plasticity of SMC has been shown in systemic arteries, the same plasticity may therefore be essential for reversing the vascular remodeling caused by proliferating SMC in PAH.

In vitro experiments conducted to assess the contribution of EC to the development of PAH have demonstrated that a VEGF receptor blockade with the VEGF-RI/VEGF-R II antagonist SU5416, under conditions of increased fluid shear stress, causes initial apoptosis followed by exuberant proliferation of the surviving EC (40). These EC are hyperproliferative and apoptosis-resistant and express the anti-apoptotic protein survivin (40). Moreover, human pulmonary microvascular EC transdifferentiated with SU5416 in vitro demonstrate that the shift to a transdifferentiated phenotype could be attributed to selection of distinct cell subpopulations (i.e., stem-like cells), and suggest that endothelial–mesenchymal transdifferentiation might be an important contributor to pathophysiological vascular remodeling in complex vascular lesion of PAH (41), because, although bone marrow–derived cells could participate in arterial neointimal formation after mechanical injury, they do not contribute substantially to pulmonary arterial remodeling in an experimental PAH model (64). Moreover, those EC are not capable of returning to normal EC states over 10 passages from the withdrawal of SU5416 (41). Likewise, Arciniegas and coworkers suggested that EMT was an important contributor to vascular remodeling in PAH, but might be reversible (65).

DRUGS AND “REVERSIBLE OR IRREVERSIBLE REMODELING”

The drugs that are currently used for the treatment of PAH act not only by opposing any abnormal vasoconstriction, but also by inhibiting the growth of normal SMC: endothelin-receptor antagonists, phosphodiesterase type 5 inhibitors, and prostacyclin derivatives (66). Some patients with IPAH have been treated successfully with vasodilators with normal or nearly normal hemodynamics. It is unclear whether they have been diagnosed very early before remodeling occurred or they are a different phenotype. However, since vascular remodeling likely becomes progressively as the disease advances in patients who are resistant to these agents (67), new drugs may be developed to specifically target pulmonary vascular remodeling for these patients.

A few cases of clinical and hemodynamic improvements with the PDGF receptor tyrosine kinase inhibitors have also been reported in human PAH (68–70). Therefore, agents like these drugs may perhaps achieve reversal of remodeling, that is, the regression of established muscularized pulmonary arteries (4). A recently completed phase II clinical trial evaluating the safety and efficacy of imatinib mesylate, a tyrosine kinase inhibitor with antineoplastic activity, in PAH failed to meet the primary efficacy end point of improvement in exercise capacity; however, many secondary end points, including pulmonary hemodynamics, were significantly improved (71). However, the effects of these drugs in human PAH are currently missing and there is no data on reverse remodeling with this class of drugs in humans. It is probable that most patients with severe PAH at the time of their diagnosis have irreversible structural alterations of their microscopically small pulmonary arterioles, that is, irreversible pulmonary vascular remodeling believed to be caused by angiogenic proliferation of phenotypically altered and transdifferentiated EC (5–8, 40, 41).

Rapamycin, an antiproliferative immunosuppressor that arrests the cells in the G1 phase of the cell cycle (72), was examined in randomized, vehicle-controlled trials using a rodent model of severe PAH (57). Because of this approval for clinical practice, the potential of rapamycin for rapid translation to human PAH therapies is offered. Rapamycin is used clinically in cardiovascular medicine, as an antiproliferative agent applied to coronary stents to reduce local restenosis (73). Rapamycin inhibits hypoxia-induced activation of S6 kinase in PA adventitial fibroblasts (74), suggesting a possibility for therapeutic benefit in PAH. Rapamycin has also been shown to attenuate experimental PAH and suppress neointimal SMC proliferation in a pneumonectomy with monocrotaline-PAH model (57). In that study, however, rapamycin failed to reverse established PAH, thus suggesting that antiproliferative (rapamycin) agents were not sufficient to induce apoptosis of neointimal SMC and that additional proapoptotic agents might be needed.

THE ROLE OF IMMUNITY

There is strong circumstantial evidence for an immune pathogenesis of PAH, i.e., PAH is associated with rheumatoid arthritis, systemic lupus erythematosus, collagen diseases (e.g., scleroderma and mixed connective tissue disease), hypothyroidism, hypersensitivity pneumonitis, and infection with HIV (7, 75, 76). Barst and colleagues (77) and Morse and coworkers (78, 79) detected an association of MHC class II alleles with PAH.

Recently, Daley and colleagues demonstrated that activation of the immune system can induce impressive pulmonary arterial remodeling, that is, pulmonary arterial muscularization without intimal EC proliferation. This pulmonary arterial muscularization in the mouse did not cause pulmonary hypertension (80). Novel treatment strategies for PAH should target the apoptosis-resistant phenotypically altered vascular cells and the disruption of the extracellular matrix between the obliterating cells. Drugs that turn the in situ local vascular immune response off may be essential to the treatment strategies for PAH because immune responses are likely to regulate signals that control cell proliferation, or the differentiation of smooth muscle actin–positive cells, and the rearrangement of the cellular organization which thus results in a severely remodeled arterial wall (80).

The use of the immunosuppressants methotrexate and prednisone to treat pulmonary hypertension with nonspecific inflammatory signs has shown promising beneficial effects in some patients (81). Ogawa and coworkers reported an immunosuppressant, prednisolone, to have an antiproliferative effect on cultured SMC of pulmonary arteries from patients with idiopathic PAH, thus suggesting that prednisolone may be potentially useful therapeutically in patients with idiopathic PAH (82). Moreover, they have shown that PDGF-induced nuclear translocation of NF-κB may play an important role in stimulating pulmonary arterial SMC proliferation (and/or enhancing pulmonary arterial SMC survival), whereas prednisolone may exert both anti-inflammatory and antiproliferative effects on PASMC by inhibiting NF-κB nuclear translocation (83).

However, it should be acknowledged that, to date, the treatment of PAH with immunosuppressive drugs has been disappointing with the exception of less severe PAH associated with connective tissue disease (84).

CONCLUSIONS

The treatment strategies for PAH should consider that there are at least two types of remodeling: pulmonary vascular disease that develops predominantly because of increased muscularization of vessel walls, and pulmonary vascular disease that develops because of EC proliferation (85). The SMC shift between a proliferative and nonproliferative phenotype may be attributed to cellular plasticity, rather than selective expansion of distinct cell subpopulations (63), thus suggesting that this form of vascular remodeling may be reversible. However, phenotypically altered and hyperproliferative SMC in PAH cannot easily convert back to a normal SMC. A recent study has demonstrated irreversible PAH in congenital heart disease (CHD) to be strongly associated with impaired apoptotic regulation of EC (35), with endothelial damage and with increased circulating EC counts (86), thus suggesting that vascular remodeling that develops because of phenotypically altered EC may be irreversible.

Moreover, after inducing apoptosis of phenotypically altered vascular cells, it may be necessary to allow normal EC to spread out and again form a monolayer. Whether phenotypically altered EC can switch back to a normal monolayer-forming EC therefore remains to be elucidated (Figure 1).

Figure 1.

Reversible or irreversible remodeling in pulmonary arterial hypertension (PAH): a hypothetical mechanism. This is a hypothetical figure of a cross-section of pulmonary arteries. Pulmonary vascular remodeling in PAH is characterized by the thickening of the media, a plexiform lesion, and muscularization. Such thickening is due to hyperplasia (proliferation) of phenotypically altered smooth muscle cells (SMC). A plexiform lesion is composed of apoptosis-resistant phenotypically altered endothelial cells (EC). Muscularization is the extension of new smooth muscle into the partially muscular and nonmuscular peripheral arteries. The SMC shift between a proliferative and nonproliferative phenotype may be attributed to cellular plasticity, rather than selective expansion of distinct cell subpopulations, suggesting that this form of vascular remodeling (i.e., likely medial thickening and muscularization) may be reversible. However, irreversible PAH in congenital heart disease is strongly associated with the impaired apoptotic regulation of EC and with endothelial damage, thus suggesting that vascular remodeling, which develops because of phenotypically altered EC (i.e., likely a plexiform lesion) may therefore be irreversible.

In conclusion, the disobliteration or reopening of occluded small pulmonary vessels is increasingly recognized as a treatment goal in PAH. The reopening of even a fraction of the obliterated arterioles would reduce pulmonary vascular resistance and thus unload the stressed right ventricle. Further studies are necessary to characterize the pulmonary vessel phenotypes to determine the reversibility of vascular remodeling in PAH.