PerspectivePulmonary Hypertension

PPARγ Activation: A Potential Treatment For Pulmonary Hypertension

See allHide authors and affiliations

Science Translational Medicine  23 Dec 2009:
Vol. 1, Issue 12, pp. 12ps14
DOI: 10.1126/scitranslmed.3000267


The pathobiology of pulmonary arterial hypertension (PAH) involves multiple molecular pathways and environmental modifiers and is characterized by progressive obliteration of pulmonary arterioles, leading to increased pulmonary vascular resistance (PVR), right heart failure, and death in ≈40 to 60% of patients 5 years after diagnosis. There is emerging evidence that many key genes involved in PAH development are targets of the insulin-sensitizing transcription factor peroxisome proliferator–activated receptor γ (PPARγ), and that pharmacological PPARγ activation would lead to their beneficial induction or repression and subsequent antiproliferative, anti-inflammatory, proapoptotic, and direct vasodilatory effects in the vasculature. PPARγ acts downstream of bone morphogenetic protein receptor II (BMP-RII), which is the cell surface receptor that is mutated or dysfunctional in many forms of PAH. Because our recent clinical observations indicate that insulin resistance may be an environmental risk factor or disease modifier (“second hit”), we suggest that PPARγ-activating agents might be beneficial in the future treatment of both insulin-resistant and insulin-sensitive PAH patients with or without BMP-RII mutations.


Pulmonary arterial hypertension (PAH) is characterized by progressive obliteration of small muscularized pulmonary arteries (arterioles), leading to increased pulmonary vascular resistance (PVR), right heart failure, and death. The pathobiology of PAH is complex and multifactorial: Pulmonary vascular remodeling includes structural and functional changes such as (i) (peri)vascular inflammation; (ii) endothelial cell (EC) damage and dysfunction; (iii) vasoconstriction; (iv) proliferation, migration, resistance to apoptosis, and survival of smooth muscle cells (SMCs), fibroblasts, and related cell types; (v) recruitment of progenitor cells; and (vi) crosstalks among vascular and circulating cells (1, 2). Therefore, it is unlikely that only one factor, pathway, or gene mutation (3) will explain all forms and cases. This underscores the need to explore several (frequently linked) signaling pathways and environmental modifiers of PAH. Because it is uncertain which pathway(s) are particularly active or inactive in an individual patient, tailored PAH therapy is hardly possible. PAH biomarkers and validated imaging techniques indicating disease severity, progression, and prognosis of PAH and associated RV dysfunction would be extremely helpful in guiding established and more-experimental clinical therapies.

Fig. 1. Plexiform lesion in advanced pulmonary arterial hypertension.

The pulmonary artery (arrowheads) is eccentrically expanded by cell proliferation surrounding slit-like and dilated vascular channels. The residual artery shows medial and intimal thickening (hematoxylin and eosin stain; scale bar, 50 μm).


Currently, there are nine approved drugs for the treatment of PAH in adults in the United States, all of which are considered to be pulmonary vasodilators: endothelin-1 (ET-1) receptor antagonists, phosphodiesterase-5 (PDE-5) inhibitors, inhaled nitric oxide, prostanoids (PGI2 analogs), and calcium channel blockers. However, none of the current medical therapies has been shown to be universally effective or able to reverse advanced pulmonary vascular disease. Promising drugs being explored for their potential to reverse PAH include platelet-derived growth factor receptor β (PDGFR-β) and epidermal growth factor receptor (EGFR) blockers; multi-kinase, elastase, and Rho-kinase (ROCK) inhibitors; mitochondrial metabolic modulators (dichloroacetate) (1, 2); and—as we suggest here—peroxisome proliferator–activated receptor γ (PPARγ) agonists (46).


A hallmark of idiopathic (I) and heritable PAH (7) is enhanced signaling by growth factors such as PDGF-BB, EGF, and transforming growth factor β1 (TGF-β1) [reviewed in (1)]. PDGF-BB (through its receptor), PDGFR-β, and EGF (via EGFR) activate mitogen-activated protein kinases (MAPKs) in vascular SMCs. Activated MAPKs such as extracellular-regulated kinase (ERK) induce cell cycle–promoting genes (encoding products such as cyclin D1/D2 and telomerase) and subsequently SMC proliferation, migration, resistance to apoptosis, and survival—all key features of pulmonary vascular remodeling (Figs. 1 and 2A). RhoA/ROCK pathways downstream of the aforementioned growth factors are also involved in the pathogenesis of PAH (2). ROCK is activated by angiotensin II and receptor tyrosine kinases (RTKs) such as PDGFR-β and EGFR and participates in subsequent MAPK phosphorylation and SMC proliferation [reviewed in (8)].

Fig. 2. Model of how PPARγ activation might reverse pulmonary arterial hypertension (PAH).

(A) Heightened PDGF-BB and EGF signaling leading to smooth muscle cell (SMC) proliferation and survival are key features of PAH. Deficiency of both apoE and LRP enhances mitogenic PDGF-BB–MAPK signaling that turns on the cell-cycle machinery, for example, expression of NOR1 and cyclin D1/D2, and induces other growth-promoting genes. PDGF-BB–/PDGFR-β–mediated phosphorylation of MAPKs such as ERK has been shown to lead to N-terminal phosphorylation and thereby inactivation of PPARγ. (B) PPARγ activation induces growth-inhibitory and proapoptotic genes in SMCs and inhibits cell-cycle–promoting genes such as those encoding telomerase, cyclin D1, and retinoblastoma protein. Moreover, PPARγ induces phosphatases that can directly inactivate MAPKs (such as pERK) downstream of PDGFR-β and EGFR. In addition, PPARγ activation can directly inhibit PDGF-BB–mediated pERK activity by blocking its nuclear translocation. Besides gene regulation in SMCs, PPARγ agonists induce the antimitogenic adipocytokine apiponectin, which [in its high molecular weight (HMW) form] sequesters the ligand PDGF-BB (“vasocrine signaling from fat cells”). Thus, decreased levels of PPARγ, apoE, and adiponectin are expected to enhance PDGF-BB/pERK signaling and pulmonary vascular remodeling. By blocking important survival pathways downstream of activated PDGFR-β (such as PI3K), PPARγ agonists might also lead to SMC apoptosis. Therefore, PPARγ agonists have the potential to reverse SMC proliferation and vascular remodeling in PAH patients. TF, transcription factor.


Imatinib (STI571), a compound used to treat chronic myeloid leukemia, blocks the activity of the RTKs PDGFR-β, c-Kit, and Bcr-Abl but hardly affects the EGFR (also an RTK). Hence, a drug target downstream of PDGFR-β and EGFR that when activated inhibits MAPK activity and other growth-promoting pathways might be of additional and even greater benefit than RTK blockers, and be able to arrest or reverse advanced clinical PAH [Fig. 2B and supporting online material (SOM)].


PPARs (-α, -β, -γ, and -δ) are ligand-activated transcription factors belonging to the nuclear receptor superfamily. PPARγ is ubiquitously expressed and plays a major role in adipogenesis, glucose metabolism, and placental and cardiac development. PPARγ is an established vasoprotective transcription factor in systemic atherosclerotic diseases (9); however, its role in pulmonary vascular disease has not been explored in comparable depth. Upon ligand activation, PPARγ heterodimerizes with the retinoid X receptor and regulates multiple target genes, such as those encoding adiponectin, interleukin-6 (IL-6), monocyte chemotactic protein–1 (MCP-1), ET-1, and the endogenous endothelial nitric oxide (NO) synthase (eNOS) inhibitor asymmetric dimethyl-arginine (ADMA), many of which are strongly implicated in the pathobiology of PAH (SOM). The antidiabetic drugs pioglitazone and rosiglitazone, both PPARγ ligands of the thiazolidinedione (TZD) class, have anti-inflammatory properties through PPARγ-dependent and -independent effects in macrophages, adipocytes, and vascular cells (9). In addition, PPARγ ligands exert antiproliferative, proapoptotic, and direct vasodilatory effects (10) in the vasculature and improve endothelial dysfunction and systemic arterial hypertension in vivo (9). PPARγ agonists therefore have therapeutic potential beyond the treatment of insulin resistance (IR).


The unique potential of PPARγ agonists for the treatment of pulmonary vascular disease relates to the fact that many of its downstream targets are positively or negatively associated with the development of PAH.

PPARγ: A drug target downstream of bone morphogenetic protein receptor II. Heritable PAH includes nonfamilial PAH associated with germline mutations in the bone morphogenetic protein receptor II (BMP-RII) gene and familial cases with or without identified germline mutations. Loss-of-function mutations in the BMP-RII gene occur in up to 70% of PAH patients with a positive family history of PAH, in 10 to 40% of PAH patients without a family history of the disease (3, 7), and in 6 to 9% of patients with secondary forms of PAH associated with anorexic drug use or congenital heart defects. However, the inheritance pattern of BMPR-II is that of a dominant gene with low penetrance, in that only ≈10 to 20% of affected family members develop the disease (3). This observation underscores the importance of environmental modifiers such as IR that could potentiate BMP-RII dysfunction, thereby triggering or worsening PAH (4, 6). Abnormalities in downstream effectors of BMP-RII probably contribute to PAH development.

Recently, we demonstrated that BMP-2, an antiproliferative ligand of the crucial cell-surface receptor BMP-RII, induces nuclear shuttling and DNA binding of PPARγ in human pulmonary artery SMCs (PASMCs) (5). Previous work by others showed that PAH patients have reduced pulmonary expression of BMP-2 (11), PPARγ (12), and apolipoprotein E (apoE) (11) mRNAs. We found that all three factors are linked and demonstrated an antiproliferative BMP-2/PPARγ/apoE axis in human and murine PASMCs (5) (Fig. 2B): (i) Using small hairpin RNA interference and transgenic techniques, the antiproliferative effect of BMP-2 was shown to be BMP-RII-, PPARγ-, and apoE-dependent (4, 5); (ii) pharmacological PPARγ activation inhibited PDGF-BB–induced ERK phosphorylation and completely blocked PDGF-BB–induced proliferation of PASMCs isolated from a PAH patient with a frameshift mutation in the BMP-RII gene; and (iii) deletion of PPARγ in vascular SMCs caused PAH in vivo (5).

These findings suggest that BMP-RII dysfunction decreases endogenous PPARγ activity and enhances PDGF-BB/MAPK (e.g., ERK) pathways and associated vascular remodeling. Hence, a strategy aimed at activating PPARγ could reverse the PAH phenotype (4).

PPARγ-mediated cell-cycle regulation and apoptosis. PPARγ activation ultimately inhibits the G1→S–phase transition that is mandatory for cell cycle progression and vascular SMC proliferation, e.g., by stabilizing the cyclin-dependent kinase inhibitor p27KIP1 (13) or inhibiting telomerase (14). By blocking important survival pathways downstream of activated PDGFR-β [such as those regulated by phosphatidylinositol 3-kinase (PI3K)], PPARγ agonists also induce apoptosis of proliferating SMCs (9) (Fig. 2B). In addition, fluid shear stress reduces PPARγ expression in ECV304 ECs. When transfected with a dominant negative version of PPARγ and injected into the tail vein of nude mice, ECV304 cells formed lumen-obliterating pulmonary vascular lesions (12). Hence, decreased PPARγ expression appears to characterize an abnormal, proliferating, apoptosis-resistant EC phenotype that may be reversed by pharmacological PPARγ activation.

PPARγ, apoE, and adiponectin counterbalance growth factor–mediated vascular remodeling. The PPARγ targets apoE and adiponectin inhibit PDGF-BB–induced PASMC proliferation, and deficiency of either target protein has been linked to PAH (4, 5, 15, 16). ApoE promotes the internalization of PDGFR-β (17), and adiponectin sequesters the ligand PDGF-BB (18) (Fig. 2B). Both BMP-2 and a PPARγ agonist also induce apoE expression and secretion, thereby impairing PDGF-BB/MAPK signaling (i.e., ERK phosphorylation) (5) in human PASMCs. Activated PPARγ also blocks PDGF gene expression (19) and induces low-density lipoprotein receptor–related protein (LRP) (20), the receptor necessary for apoE-mediated suppression of PDGF-BB signaling (17) (Fig. 2B). Additional counteractive interactions between PPARγ and phosphoERK exist intracellularly; these include both the activation of phosphatases and the prevention of nuclear translocation in SMCs, and the phosphorylation of PPARγ at its N terminus that leads to its inactivation (Fig. 2; for details see the legend and SOM). Furthermore, transgenic mice with a targeted deletion of PPARγ in ECs and macrophages (Tie 2 Cre PPARγ flox/flox) have significantly higher pulmonary PDGFR-β protein expression than littermate controls (21). Thus, decreased levels of PPARγ, apoE, and adiponectin are expected to enhance PDGF-BB–pERK signaling and pulmonary vascular remodeling (Fig. 2).

Male apoE-deficient mice (apoE−/−), when fed a high-fat diet, do not up-regulate the insulin sensitizers adiponectin and leptin (in contrast to control mice), but develop IR and severe PAH. A 4-week treatment with the PPARγ agonist rosiglitazone led to an eightfold induction of plasma adiponectin, improved insulin sensitivity, and complete regression of PAH (4) (Fig. 2B). Thus, PPARγ target genes other than apoE (e.g., the adiponectin gene) are also implicated in the antiremodeling and anti-inflammatory effects of PPARγ agonists (4, 5, 15, 16). Moreover, adiponectin is an endogenous antithrombotic factor (22). Hence, induction of adiponectin might be beneficial in pulmonary hypertension, even in the secondary forms associated with chronic thromboembolism, or parenchymal lung disease/chronic alveolar hypoxia (23).

PPARγ, detrimental adipocytokines, endothelin-1, and the NO pathway. Yudkin and colleagues (24) previously proposed that detrimental adipocytokines, such as tumor necrosis factor–α and IL-6, are secreted from perivascular fat cells and inhibit the eNOS pathway of insulin signaling, leaving unopposed vasoconstriction mediated by endothelium-derived ET-1, a key player in PAH. PPARγ agonists would counteract these deleterious effects as they induce eNOS, and impair the expression of thrombin, ET-1, and the endogenous NO synthase inhibitor ADMA, thereby potentially reversing endothelial dysfunction. Increased pulmonary ET-1 expression and elevation of the ADMA plasma concentration were observed in PAH patients, and increased ADMA concentrations negatively correlated with cardiac index and human survival rates (25).

Endothelial progenitor cells, PDE inhibitors, and PPARγ agonists. Adult IPAH patients have reduced numbers of endothelial progenitor cells (EPCs) as compared with healthy controls; a reduced number of circulating EPCs correlated with poor hemodynamics and abnormal concentrations of inflammatory markers and ADMA. Treatment with sildenafil, a vasodilatory PDE-5 inhibitor commonly used for PAH therapy, led to a dose-dependent rise in EPC numbers that was greater than those found with other therapies (26). Intriguingly, rosiglitazone inhibited the negative effects of C-reactive protein (CRP) on EPC survival, differentiation, and function in a separate study (27), and transplantation of autologous EPCs improved mean pulmonary arterial pressure, PVR, cardiac output, and the distance walked in 6 min in patients with IPAH (28). Hence, sildenafil and PPARγ agonists might represent a pharmacological means of increasing circulating EPCs in PAH patients that probably can improve clinical outcome.

PPARγ activation suppresses vascular and perivascular inflammation. Circulating proinflammatory cytokines (e.g., IL-6 and MCP-1) that are mainly derived from monocytes and macrophages and regulated by nuclear factor kappa B (NF-κB), nuclear factor of activated T cells, and other transcription factors, substantially contribute to the development and/or progression of PAH (1, 2). TZDs reduce the activation and inflammation of ECs by suppressing the activity of NF-κB and activator protein–1 via PPARγ-dependent and -independent mechanisms. The overall result is an impaired production of chemokines, adhesion molecules, reactive oxygen species, and major histocompatibility complex class II proteins, which are crucial for T cell activation and initiation of an immune response (9). PPARγ agonists inhibit DNA binding of NF-κB (29) and NF-κB–dependent matrix metalloproteinase (MMP)–9 activation in vascular SMCs (30), which are both implicated in PAH development (31). Thus, PPARγ agonists have the potential to counteract proinflammatory NF-κB pathways and associated MMP-9 hyperactivity in PAH patients.

PPARγ-mediated inhibition of extracellular matrix proteinases. PPARγ ligands impair MMPs that are activated by elastase. Elastase inhibitors not only prevent but also reverse advanced fatal PAH in rats (32). Rosiglitazone has been shown to activate glycogen synthase kinase–3β in vascular SMCs, which inhibits proliferation and migration by blocking NF-κB–dependent MMP-9 activation (30). Hence, along with elastase inhibitors such as elafin, PPARγ agonists could block the detrimental PAH-promoting effects of heightened elastase and MMP activity.

Rho kinase inhibition by PPARγ. Besides the induction of SMC proliferation (see above), RhoA/ROCK activation leads to endothelial dysfunction, induction of vasoactive cytokines via the NF-κB pathway, and chronic pulmonary remodeling that may involve up-regulation of TGF-β1 and down-regulation of the cell-cycle inhibitor p27KIP1. Pioglitazone improves endothelial dysfunction in the presence or absence of inflammation and inhibits ROCK pathways in vascular SMCs—the latter through up-regulation of the cytosolic protein tyrosine phosphatase SHP-2 (33). Thus, PPARγ agonists might provide antimitogenic ROCK inhibition, which has been effective in experimental and clinical PAH [reviewed in (8)].


Low adiponectin and high CRP and IL-6 levels are associated with an increased risk for IR, metabolic syndrome, and systemic and possibly also pulmonary vascular disease in humans. ET-1 might be a key component of this pathobiology, because it inhibits adiponectin secretion and insulin sensitivity in healthy humans. We recently demonstrated that IR and dyslipidemia are more common in female PAH patients than in the general population (45.7% versus 21.5%; P < 0.001) and may be environmental risk factors or disease modifiers that might increase the incidence of PAH in the upcoming years. The presence of IR was associated with poorer combined 6-month event-free survival from RV failure, transplantation, or death (58% versus 79%, hazard ratio 2.57, 95% confidence interval 1.03-6.06; P < 0.05) (6). Based on our preliminary experimental (4) and clinical data (6), we hypothesize that IR, dyslipidemia, and dysregulation of ET-1– and PPARγ-dependent factors such as adiponectin may increase the susceptibility to PAH that can be reversed by PPARγ activation. However, it is important to note that the beneficial effects of PPARγ agonists in the pulmonary (4, 23, 34) and systemic circulation (9) do not depend on the presence of IR. Recently, Walcher et al. demonstrated a very rapid effect of single-dose rosiglitazone treatment on endothelial function in nondiabetic healthy men, underscoring the fact that PPARγ agonists exhibit direct vascular (vasodilatory) effects (10).


Prospective studies are needed to address whether IR, dyslipidemia, impaired PPARγ and apoE function, and low adiponectin levels are risk factors for the development or progression of PAH in humans. Most importantly, randomized controlled trials (RCTs) should investigate the potential of Food and Drug Administration–approved PPARγ agonists in the treatment of PAH in the presence or absence of IR. [For a discussion of the safety of PPARγ agonists (TZDs), see the SOM.] The molecular links and mechanisms that underlie IR, apoE, and adiponectin levels, PPARγ function, and known PAH pathways should also be explored. By applying advanced technologies such as proteomics and metabolomics, biomarkers might be found that will aid in (i) monitoring PAH progression and response to therapy, (ii) tailoring pharmacotherapy, and (iii) PAH screening, potentially, in certain at-risk populations. In addition, it will be important to study the prevalence and clinical relevance of genetic polymorphisms of PPARG 1/2 and PPARγ target genes in PAH patients.


Several RCTs currently investigate the therapeutic potential of pioglitazone in proliferative-inflammatory lung diseases other than PAH, such as non–small-cell lung cancer and asthma (35). The magnitude of these very recent research efforts underlines the potential of PPARγ-activating drugs beyond diabetes therapy. The antiproliferative BMP-2/PPARγ/apoE axis in SMCs (5) probably exists in other cell types and very likely has additional proapoptotic and anti-inflammatory properties, given that the key player PPARγ affects multiple genes besides apoE (9). Thus, the aforementioned translational studies on the protective role of PPARγ in PAH will have substantial impact on our understanding and future treatment of other emerging lung and collagen vascular diseases, as well as cancer, all of which are characterized by uncontrolled proliferation, inflammation, and resistance to apoptosis. Moreover, a connection between PPARγ and apoE has been made in patients with Alzheimer’s disease, in that the improvement of cognitive function with rosiglitazone is not apparent in patients who carry the APOE epsilon 4 allele (36). We speculate that disruption of BMP-RII–PPARγ and PPARγ-apoE signaling might underlie many different pathologic processes.


Taken together, emerging evidence indicates that multiple key genes involved in the pathobiology of PAH are targets of PPARγ and that pharmacological activation of PPARγ would lead to their beneficial induction and stabilization (genes encoding eNOS, p27KIP1, adiponectin, apoE, and LRP) or repression (genes encoding PDGF, MAPK, cyclin D1, telomerase, ET-1, ADMA, IL-6, NF-κB, elastase/MMP, and RhoA/ROCK). We suggest that PPARγ agonists might reverse pulmonary vascular remodeling in PAH patients with or without BMP-RII dysfunction and that IR might be a risk factor or disease modifier (so-called “second hit”). Hence, future treatment of both insulin-resistant and insulin-sensitive PAH patients might include PPARγ agonists or selective PPARγ modulators. RCTs on the use of these agents in pulmonary vascular disease are urgently needed.

Supporting Online Material

SOM Text



    References and Notes

    1. We are indebted to G. Berry of Stanford University for his help with the histology shown in Fig. 1. G.H. was supported by the American Heart Association/Pulmonary Hypertension Association (grant 0425943H). R.T.Z. is a recipient of an Actelion Entelligence Career Development Award. The authors declare that they have no conflict of interest related to the topic of this article. R.T.Z. has served on the speaker’s bureau for Actelion (2005–2007) and Co-Therix (2006), is a consultant to Gilead Pharmaceuticals (2009), and serves on the CHARTER advisory board to United Therapeutics (2008–2009).
    • Citation: G. Hansmann, R. T. Zamanian, PPARγ activation: A potential treatment for pulmonary hypertension. Sci. Transl. Med. 1, 12ps14 (2009).

    View Abstract

    Navigate This Article