Snakes and Ladders: A Potential Therapy of Hepatocyte Growth Factor and Pigment Epithelium-derived Factor in Pulmonary Hypertension

Pulmonary hypertension (PH) is a disorder of the lung vasculature defined by a mean pulmonary arterial pressure >20 mm Hg (1). PH is a severe and progressive disease characterized by marked pulmonary vasculature remodeling and a progressive increase in pulmonary vascular resistance, leading to reduced pulmonary blood flow and, ultimately, right heart failure (2). PH is caused by numerous factors, including genetic mutations, autoimmune diseases, infections, and exposure to toxins (3). The disease can be classified into five groups based on the underlying cause (4).

A central feature of PH is vascular remodeling, characterized by vascular smooth muscle cell hyperplasia, increased extracellular matrix protein deposition, endothelial dysfunction, and perivascular inflammation (5). Despite the grave prognosis, current treatments, including phosphodiesterase inhibitors and prostacyclin analogs, fail to halt pulmonary vascular remodeling and thus have limited capacity to stop disease progression (6).

Hepatocyte growth factor (HGF), which regulates cell morphogenesis and motility by activating a tyrosine kinase receptor, is downregulated in experimental PH (7–9). Supplemental HGF, when used as a treatment in an experimental model of PH, has been shown to reduce vascular remodeling and promote angiogenesis, but it induces harmful vascular leakage in immature vessels (8, 9). Pigment epithelium-derived factor (PEDF) is an antiangiogenic molecule whose effects are mediated through inhibition of VEGF3 (vascular endothelial [VE] growth factor receptor 3) and stimulation of the Fas/Fas-ligand system to induce endothelial cell apoptosis. Furthermore, it can decrease vascular permeability by maintaining adherens and tight junction integrity (10–13). The divergent roles of these mediators in the pulmonary circulation provide an exciting opportunity to evaluate their capacity to treat PH when used in combination.

In this issue of the Journal, Qiu and colleagues (pp. 87–98) demonstrate that a combination of PEDF and HGF overexpression had a more significant therapeutic effect on PH than either factor alone (14). The investigators found that this combination therapy improved pulmonary vascular remodeling, reduced right ventricular hypertrophy, and improved overall survival rates in the monocrotaline (MCT)-induced PH rat model. These studies were performed using conventional approaches such as right ventricular systolic pressure measurements and morphometric analysis for vascular remodeling. They also incorporated novel and clinically relevant studies, such as angiography, which provide important physiological readouts. Echocardiography data were also included, which, together with the other measures, provided a robust physiological readout panel underscoring the effectiveness of dual therapy for PH. In addition, this study provides further evidence demonstrating that, in idiopathic PH, lower levels of PEDF and HGF expression are observed compared with controls. An important limitation of this study is the fact that the effect of the dual therapy with HGF and PEDF appeared to be additive at best; this is somewhat surprising given the distinct pharmacological and physiological effects of these agents on their own.

Importantly, the group demonstrated that PEDF curtailed pulmonary vascular leakage that was evident after HGF treatment. Immunofluorescent fibrinogen staining revealed that HGF-treated rats had the most extensive vascular leakage. This was reversed with PEDF treatment, which significantly restored barrier function. Immunofluorescence and Western blotting were further used to reveal that VE-cadherins, a vital component for maintaining cell–cell contact (15), were significantly upregulated and localized to endothelial cell gap junctions. In addition to the important effects on vascular leakage, this study also demonstrated that PEDF was able to block the angiogenic effects of HGF and improve features of PH. These results are intriguing because the role of neovascularization in PH is controversial. Studies have identified vascular pruning, particularly in PAH and PH associated with chronic obstructive pulmonary disease (16, 17). However, in PH associated with lung fibrosis, neovascularization has been identified as part of the pathological response (18). Thus, these studies demonstrate that, in a model of MCT-induced PH, inhibition of neovascularization in combination with reduced vascular remodeling is beneficial for the treatment of PH, highlighting aberrant neovascularization as an important pathological response in PH, at least in this model. These findings are significant because they demonstrate how HGF and PEDF work together to treat PH and reduce the potential side effects of each agent alone. Taken together, these results point to dual therapy with HGF and PEDF as an effective treatment to target vascular remodeling in PH. However, the authors uncovered a potentially serious side effect of PEDF: enhancing vascular remodeling on its own, an effect that was reduced by combination with HGF. This effect can be attributed to the effect of PEDF in inhibiting angiogenesis. This raises some important concerns in the treatment of PH because vascular remodeling is an important pathophysiological feature that is closely associated with higher pulmonary vascular pressures. The exact mechanisms that lead to PEDF-induced vascular remodeling are not fully known and represent an important gap in knowledge; are they a response to its angiogenic effects or is there a direct molecular link that leads to increased smooth muscle cells?

The team then attempted to better understand the mechanism behind PEDF’s antiangiogenic and vascular remodeling properties. Using in vitro techniques, they found that knocking down VEGFR2 (VE growth factor receptor 2), a receptor known to interact with PEDF, promoted VE-cadherin junction disintegration. Furthermore, PEDF treatment decreased the extent of VEGFR2 phosphorylated at the Y951 and Y1214 positions. Y951 has been shown to bind to a T cell–specific adaptor and subsequently to increase vascular permeability (19). Y1214 is also known to induce Akt and Erk1/2 activation, both of which are involved in cell proliferation (20, 21). Overexpression of PEDF resulted in substantial binding to VEGFR2 and restricted T cell–specific adaptor anchoring. Because VEPTP (VE protein tyrosine phosphatase) also acts on VEGFR2 and VE-cadherin at endothelial junctions to repress VEGFR2 phosphorylation, the researchers evaluated the effect of PEDF on the binding of VEPTP and VE-cadherin to VEGFR2. PEDF significantly promoted the formation of the VEGFR2/VEPTP complex and VEGFR2/VE-cadherin complex with or without HGF treatment. Therefore, PEDF may play a role in regulating VEGFR2 signaling pathways. The link between PEDF and VEGFR2 has been shown earlier, but this article elucidates a previously unknown link between VEPTP and VE-cadherin that is independent of HGF. This finding provides an important mechanistic link between PEDF and its beneficial physiological effects in PH and identifies these complexes as novel targets for disease therapy that can be exploited to limit vascular leakage in diseases such as acute respiratory distress syndrome, in which loss of barrier function is highly significant and focal adhesion proteins play an important therapeutic role (22). An important gap in knowledge is whether the effects of PEDF on barrier function can be dissociated from its capacity to promote vascular remodeling. Dissociating these responses can help direct the development of novel pharmacological approaches that can limit aberrant angiogenesis and reduce vascular remodeling by HGF and limit its capacity to promote vascular leakage.

This article by Qiu and colleagues demonstrates that a combined treatment with HGF and PEDF has the potential to work as an effective clinical treatment for PH. However, the significance of these findings is limited because the treatment has been tested in only a single model of PH. Further studies in different models, such as the Sugen/hypoxia rat model of PH, would better validate the effectiveness of this dual treatment. The use of a single model of PH, particularly the MCT model of PH, represents an important limitation because this model has been shown to have restricted translational value. Furthermore, strategies aimed at overexpressing mediators in disease are often much harder to translate to the clinic because of the enhanced potential for side effects. In this study, key concerns include the limitation of angiogenic responses by PEDF upregulation and the risk of promoting vascular remodeling in the lung and in other organs, which could lead to important vascular side effects. Further, gene-therapy approaches have several drawbacks, including the difficulty of regulating expression levels and potential off-target effects that may result in overexpression in other organs (23). Thus, comprehensive pharmacological and toxicological studies that assess the effect of overexpression of HGF and PEDF alone and in combination in multiple organs are warranted. Given the difficulty of regulating expression at the gene level, a potential approach would be the use of pharmacological agents that could enhance HGF and PEDF levels, allowing for greater control of HGF/PEDF activity and the identification of doses with potential synergistic effects.

Intriguingly, HGF has been shown to modulate antifibrotic responses in the lung (24–26). These antifibrotic responses, coupled with HGF’s capacity to attenuate vascular remodeling in combination with PEDF, strongly point at this dual therapy as a potential therapeutic option for patients with lung fibrosis and PH, in whom few treatment options other than lung transplantation are available (27). Only prostacyclin, a vasoactive mediator, has been approved for these patients (28), so new agents that are able to target fibrosis and vascular remodeling such as HGF + PEDF have the potential to significantly improve patient care in PH with lung fibrosis.