Pulmonary arterial hypertension (PAH) is a fatal disease characterized by an increase in pulmonary arterial pressure with subsequent right ventricular failure and death (Rosenkranz, 2015). Generally, the major pathological changes of PAH are vasoconstriction and vascular remodelling of the small pulmonary arteries (Tuder et al., 2009) involving the thickening of the intima-media, smooth muscle cell proliferation, endothelial cell proliferative lesions formation, vascular inflammation and immune dysregulation (Tuder et al., 2007; Price et al., 2012; Humbert et al., 2019). There are different categories in PAH including idiopathic PAH, heritable PAH, drug or toxin-induced PAH and PAH associated with connective tissue disease, human immunodeficiency virus infection, portal hypertension, congenital heart disease and schistosomiasis (Galiè et al., 2016). Idiopathic PAH is the most common type of PAH in western countries with congenital heart disease-related PAH being the most prevalent in Asia (Humbert et al., 2006; Lim et al., 2019).

Currently approved PAH drugs such as prostacyclin analogues, endothelin receptor antagonists and phosphodiesterase-5 (PDE-5) inhibitors aim to control pulmonary vascular tone (Olsson and Hoeper, 2009). Enormous progress has been made in the therapeutic practice of PAH in the past decades but there is still a long way to go because none of these present drugs are curative. They reduce morbidity and only slightly improve survival with no effect on PAH mortality rate.

Many molecular pathways have been linked to the development and treatment of PAH.It has been revealed that the nitric oxide (NO) pathway plays a very essential and central role in regulating vascular tone in the pulmonary circulatory system by interacting with other crucial signaling pathways. Phosphodiesterase-5 (PDE-5) inhibitors and soluble guanylate cyclase stimulators are the current drug classes acting on this pathway. Meanwhile, novel compounds targeting on NO pathway exhibits great potential in development of next generation therapy medications. This review seeks to summarize the research progress of PAH regarding to NO pathway, discuss the critical role of NO pathway in understanding the pathophysiology of PAH and new drug development.

Current PAH therapy focuses on three main pathways. These are the prostacyclin, nitric oxide, and endothelin pathways. Drugs acting on these pathways are summarized in Table 1.

Nitric oxide (NO) is a biological molecule that regulates many physiological and pathological processes in the body. It was proposed that the release of a vasodilating factor by the endothelial cells as one of the mechanisms behind acetylcholine-induced vasodilation in vivo (Furchgott and Zarwodski, 1980). The identity of nitric oxide as the endothelial-derived relaxing factor was not known until 1987 when two different studies confirmed it (Ignarro et al., 1987; Palmer et al., 1987).

Nitric oxide is formed from the oxidation of L-arginine to form citrulline and NO in the presence of nitric oxide synthases (NOS) and molecular oxygen (Moncada and Higgs, 1993) as shown in Figure 1. Arginase competes with eNOS for L-arginine by converting L-arginine to urea and L-ornithine (Luiking et al., 2012). N ^G , N ^G -dimethyl-L-arginine (ADMA) is known to inhibit this process by competing with L-arginine for eNOS (Zakrzewicz and Eickelberg, 2009). N ^G, N ^G-dimethylarginine dimethylaminohydrolases (DDAH1 and DDAH2) are responsible for the degradation of ADMA (Tain and Hsu, 2017). Three isoforms of nitric oxide synthases have been described in mammals. These are neuronal NOS (nNOS, NOS1), inducible NOS (iNOS, NOS2), and endothelial NOS (eNOS, NOS3) (Nathan and Xie, 1994).

Endothelium-derived NO diffuses into vascular smooth muscle cells and stimulates soluble guanylate cyclase (sGC) to produce cGMP, further activating associated protein kinases such as protein kinase G I (PKGI) which causes vasorelaxation (Sausbier et al., 2000; Surks, 2007). cGMP is broken down mainly by phosphodiesterase 5 (PDE5) (Francis et al., 2000). See Figure 1 for more details.

Conflicting levels of exhaled NO have been reported in PAH patients. Data from studies with patients suffering from idiopathic PAH and scleroderma-/drug-associated PAH indicated a reduction in exhaled NO as compared to healthy patients (Kharitonov et al., 1997; Kaneko et al., 1998; Archer et al., 2012). Other studies also found no significant change in exhaled NO levels in idiopathic PAH and scleroderma-associated PAH when compared to healthy subjects (Riley et al., 1997; Olivieri et al., 2006; Malekmohammad et al., 2019). There was no difference in exhaled NO after 3 months of treatment with endothelin receptor antagonists, guanylate cyclase stimulants, phosphodiesterase type 5 (PDE5) inhibitors and prostanoids in comparison with healthy subjects. However, other diagnostic markers such as 6MWD and N-terminal prohormone of brain natriuretic peptide (NT-proBNP) were significantly correlated to disease severity and treatment response (Malekmohammad et al., 2019). Confounding factors such as exhalation flow rate, measurement technique, the NO analyzer used, nasal NO contamination, age, height and smoking (Borrill et al., 2006; Dweik et al., 2012) makes exhaled NO an unreliable marker that needs standardization to be of diagnostic value in PAH.

The use of NO Plasma metabolites (NO_X) as a biomarker and prognostic indicator of PAH is being studied. A study found that patients (age range: 20–57 years) with IPAH had reduced levels of plasma NOx, which correlated inversely with mPAP and patient survival (Zhang et al., 2016). Contradictory studies found elevated levels of NOx in IPAH and congenital heart disease-associated PAH patients with age ranges of 31–77 and 5 days -12 years (Ikemoto et al., 2002; Malinovschi et al., 2011). The patients in the extreme age groups (age averages of 6 and 54 years) seem to show increased levels of NOx. The different results from the various studies could be attributed to the PAH type and ages of the patients studied (Zhang et al., 2016).

Reduced eNOS levels have been reported in PAH patients (Giaid and Saleh, 1995). Another study found an increased expression of eNOS in the plexiform lesions of PAH patients (Berger et al., 2012). The eNOS that is elevated in some PAH patients is likely to be in the uncoupled state causing it to produce more superoxides (NO scavengers) than NO (Klinger et al., 2013). eNOS uncoupling can occur as a result of a decrease in amounts of tetrahydrobiopterin (BH4), a cofactor for eNOS in NO synthesis. There are studies to confirm eNOS uncoupling in pulmonary hypertension models that are BH4-deficient (Khoo et al., 2005; Nandi et al., 2005).

Also, several studies have described elevated levels of ADMA in plasma and serum of IPAH and connective tissue disease (CTD)-associated pulmonary arterial hypertension patients (Kielstein et al., 2005; Fang et al., 2015; Liu et al., 2019). Monocotaline- and hypoxia-induced PAH rat model have been found to have increased ADMA levels with a corresponding decrease in DDAH levels (Millatt et al., 2003; Li et al., 2010). This makes ADMA/DDAH possible diagnostic indicators in PAH

NO signaling remains a very relevant pathway in understanding and treating PAH. Upregulation of NO signaling is crucial in the management of PAH.

The nitric oxide pathway interacts with several other pathways reiterating its complex role in PAH. Nitric oxide interaction with some of these pathways is illustrated in Figure 2. The nitric oxide pathway has been shown to interact with the bone morphogenetic protein (BMP) pathway. Mutations in the bone morphogenetic protein receptor 2 (BMPR2) gene possess the strongest risk factor in the development of PAH, particularly heritable PAH (Machado et al., 2006). Bone Morphogenetic Protein Receptor II (BMPRII) and its ligands, BMP2 and BMP4 were found to be mediators of endothelial nitric oxide synthase activation (Gangopahyay et al., 2011). BMP2 and BMP4 activates eNOS by enhancing its phosphorylation by protein kinase A in pulmonary arterial endothelial cells (PAEC). Interestingly, knocking down BMPRII in PAEC reduced the ability of BMP2 and BMP4 to stimulate eNOS phosphorylation suggesting a crucial role of BMPRII in eNOS activation (Gangopahyay et al., 2011). Plasma nitric oxide metabolites have also been found to be markedly reduced in idiopathic pulmonary arterial hypertension patients with BMPR2 mutations (Zhang et al., 2016).Protein Kinase G I (PKGI) which is one of the downstream targets in the nitric oxide pathway has been found to crosstalk with the BMP pathway. PKGI contributes to the activation of BMP signaling by BMP4 in Human PASMCs (Yang et al., 2013). PKGI upregulates BMP signaling by phosphorylating the BMPRII receptor and smad proteins. Furthermore, PKGI binds to the phosphorylated smad proteins to form a complex and translocates into the nucleus where it upregulates the transcription of Id1 mRNA (Schwappacher et al., 2013). This then facilitates apoptosis and inhibits PASMC proliferation (Schwappacher et al., 2013; Yang et al., 2013).

The extracellular signal-regulated kinase 1/2 (ERK 1/2) pathway plays a significant role in the pathogenesis of PAH. Activation of the ERK pathway is known to be associated with increased proliferation and reduced apoptosis of PASMCs (Yang et al., 2008; Zhou et al., 2019). NO has been found to induce S-nitrosylation of ERK1/2 leading to reduced ERK phosphorylation and improved MCF-7 cell apoptosis (Feng et al., 2013). This crosstalk could be one of the ways with which NO inhibits proliferation and enhances apoptosis in PASMCs. S-nitrosylation of ERK1/2 by NO in PASMCs is yet to be explored.

The phosphatidylinositol 3-kinase/protein kinase B (PI3K/Akt) pathway is another major pathway that is implicated PAH. There have been contradictory reports on its role in PAH (Fang et al., 2016; Huang et al., 2017; Zheng et al., 2017). The PI3K/Akt pathway is known to phosphorylate eNOS and increase NO levels in many cell types including human pulmonary arterial endothelial cells (HPAECs) (Kim et al., 2008; Liang et al., 2009; Li et al., 2014). This helps to attenuate monocrotaline-induced pulmonary artery endothelial dysfunction and PAH in rats (Li et al., 2014; Zheng et al., 2017). Relaxation of rat pulmonary artery rings is enhanced through the activation of eNOS by the PI3K/Akt pathway (Zhang et al., 2010).

There is evidence that Peroxisome proliferator-activated receptor gamma (PPARγ) activity is reduced in PASMCs from idiopathic PAH patients (Falcetti et al., 2012). Activation of the Peroxisome proliferator-activated receptor gamma (PPARγ) pathway is known to reduce monocrotaline-and hypoxia-induced PAH in rats (Liu et al., 2012; Legchenko et al., 2018). Increased PPARγ expression has been found to enhance the phosphorylation and activation of eNOS in human pulmonary arterial endothelial cells (Li et al., 2014). Enhanced interaction between heat shock protein 90 and eNOS is one of the ways with which PPARγ facilitates eNOS phosphorylation. PPARγ agonists such as rosiglitazone, ciglitazone and pioglitazone have been found to stimulate eNOS activation through various mechanisms. Rosiglitazone was found to activate eNOS by stimulating heat shock protein (HSP)-90–eNOS interaction, with a subsequent eNOS phosphorylation at Ser¹¹⁷⁷ (Polikandriotis et al., 2005). Ciglitazone also caused a reduction in human umbilical vein endothelial cells (HUVEC) membrane NADPH-dependent superoxide anion generation leading to a decrease in endothelial superoxide anion generation and oxidative stress with a resulting increase in NO bioavailability (Hwang et al., 2005; Polikandriotis et al., 2005). Rho-kinase, a serine–threonine kinase is known to cause endothelial dysfunction by inactivating eNOS and reducing NO generation (Nohria et al., 2006). Pioglitazone enhances the dephosphorylation on tyrosine residues of the Vav protein, a group of guanosine nucleotide exchange factors (GEFs) involved in the activation of Rho Kinase. The deactivation of Rho Kinase leads to an increase in eNOS activation with a consequent increase in NO production (Wakino et al., 2004). The PPARγ agonist, telmisartan was found to increase the expression of DDAH II thereby enhancing the degradation of ADMA and increasing NO levels in human umbilical vein endothelial cells (Scalera et al., 2008).

There are two groups of drugs currently used in PAH that act on the nitric oxide pathway, the phosphodiesterase 5 inhibitors (sildenafil and tadalafil) and soluble guanylate cyclase inhibitors (riociguat).

The NO pathway offers various molecular targets that can be exploited to help develop novel drugs to ameliorate the pathophysiological changes that occurs during PAH development. Some of the possible therapeutic targets offered by the NO pathway include arginase, ADMA, DDAH1, eNOS, PDE-5, cGMP, sGC and NO. This section highlights key drug molecules and target that focus mainly on the NO pathway and their implications in PAH.

L-arginine serves as a substrate not just for eNOS but arginase as well. This means an increase in arginase activity leads to a reduction in the availability of L-arginine to be converted to NO by eNOS (Pernow and Jung, 2013). Arginase metabolizes L-arginine to urea and L-ornithine (Luiking et al., 2012) as shown in Figure 1. Arginase exists in two isoforms namely arginase I and arginase II but arginase II has been found to be much more involved in the development of PAH in humans and animals (Jin et al., 2010; Cho et al., 2013 ). Deletion of arginase II in interleukin-13 overexpressing transgenic mice showed a significant decrease in medial wall thickness suggesting an important role of arginase II in PAH (Cho et al., 2013). PAH patients are known to have high arginase II activity compared to healthy controls with a consequent reduction in NO synthesis (Kao et al., 2015). A recent study found selective arginase II inhibitors L207-0525 and L327-0346 demonstrate a dose-dependent protective activity in monocrotaline-induced PAH rats. Furthermore, combining L207-0525 and L327-0346 with low dose (0.1 mg/kg) tadalafil produced a better protective effect in monocrotakine-induced PAH rats. Therefore, L207-0525 and L327-0346 are potential compounds needing further exploration for PAH treatment (Koklin and Danilenko, 2019).

Nitrates have shown promising signs in the amelioration of PAH in hypoxia- and monocrotaline-induced PAH animals. Both Intraperitoneal and nebulized sodium nitrite reduced pulmonary arterial pressure, right ventricular hypertrophy and vascular remodeling in MCT-induced PAH rats (Zuckerbraun et al., 2010; Pankey et al., 2012). Orally administered sodium nitrate have also shown protective effects in hypoxia-induced PAH mice (Baliga et al., 2012). However, a recent study found oral sodium nitrate does not substantially reduce established MCT-induced PAH in rats (Malikova et al., 2020). The different outcomes in the above studies could be attributed to the difference in disease severity in the PAH models as well as the routes and timing of drug administration (Malikova et al., 2020). Considering the poor response of certain patients to PDE5 inhibitors (PDE5i) due to reduced endogenous nitric oxide production, nitrates could serve as alternative sources of NO in such patients. Also, the impracticability of inhaled nitric oxide therapy further necessitates the need for alternate NO sources. More research has to be done to further explore the potential benefits of nitrates in PAH therapy especially in PDE5i-irresponsive patients.

Oxymatrine, an active alkaloid derived from the traditional Chinese herb Sophora alopecuroides was found to protect against hypoxia- and monocrotaline-induced PAH (Zhang et al., 2014). The mechanism behind the protective effect of Oxymatrine in monocrotaline-induced PAH is likely to be through the reduction of pulmonary ADMA levels even though it did not affect the level of DDAH1 (Dai et al., 2019).

Despite the fact that DDAH1 levels have been found to be reduced in hypoxia-induced PAH, it was uncertain how important of a role DDAH1 dysfunction plays in PAH. Recently, a study on a novel DDAH1 knockout (DDAH1^−/−) rat strain model found DDAH1 dysfunction to significantly worsen RVSP and RVHI in DDAH1^−/− MCT model compared to the wild type MCT model (Wang et al., 2019). Although no in vitro experiment was carried out to target specific cells that are responsible for the progress of PAH, this study still shows how important DDAH1 dysfunction is in the development of PAH making it a possible PAH treatment target.

Apelin signaling is known to regulate endothelial NOS (eNOS) (Chandra et al., 2011). It has been discovered that the novel cyclic biased agonist of the apelin receptor, MM07 significantly reduces the elevation of right ventricular systolic pressure and hypertrophy induced by monocrotaline (Yang et al., 2019). They also found the elevation of eNOS and its mRNA as one of the mechanisms behind its protective effect in monocrotaline-induced PAH.

Udenafil, an oral phosphodiesterase-5 inhibitor approved for the treatment of erectile dysfunction has been found by a recent double-blind, placebo-controlled phase IIb clinical trial to improve 6-MWD in patients with PAH, especially those with a history of ERA therapy (Chang et al., 2019). The improvement in the 6-MWD with udenafil and ERA combination therapy group was found to be better than that of sildenafil and tadalafil in previous trials (Galiè et al., 2009a; Galiè et al., 2009b; Benza et al., 2018). Adverse side effects in the udenafil were also found to be mild and in the expected range (Chang et al., 2019).

TPN171, a new compound that inhibits PDE5 has been found to significantly reduce mPAP in a rat MCT-PAH model. Its efficacy was comparable to that of the sildenafil control group. It has a long half-life making once-daily dosing possible. The effective dose of TPN171 used in the animal experiment was 1 mg/kg which is lower than that of sildenafil (25 mg/kg). This property could make the occurrence of side-effects less likely if TPN171 is used in a clinical setting (Wang et al., 2019). TPN171 is currently in phase II clinical trial.

Evodiamine is a traditional Chinese medicine for the treatment of cancers (Dong et al., 2012). Evodiamine derivatives (S)-7e and (S)-7 days are newly discovered PDE5 inhibitors with very high selectivity (Zhang et al., 2020). Evodiamine derivative (S)-7 days significantly reduced mPAP and wall thickness in rat MCT-PAH model. Its efficacy is similar to that of sildenafil which was used as a positive control. Additionally, the study found a unique allosteric pocket of PDE5 using evodiamine derivative (S)-7e. Currently approved PDE5 inhibitors only bind to the substrate-binding pocket. This novel allosteric pocket regulates both enzymatic activity and pulmonary hemodynamic function of PDE5 thereby serving as a new therapeutic target for PAH treatment (Zhang et al., 2020).

A new study found PDE10 to be a novel therapeutic target for treating PAH. This study used a highly selective PDE10 inhibitor, 2b to explore the role of PDE10 in PAH. Compound 2b significantly reduced mPAP and RVHI in PAH rats making PDE10 a potential therapeutic target for PAH (Huang et al., 2019).

Studies have found natriuretic peptides to increase cGMP levels (Egom, 2015; Preston et al., 2016). Natriuretic peptides (NP) are inactivated by neprilysin. Natriuretic peptides are known to have antiproliferative effects on PASMCs and also attenuate the development of hypoxia-induced PH (Arjona at al., 1997; Zhao et al., 1999). Furthermore, the beneficial effect of sildenafil was found to be affected by natriuretic peptide activity in mice (Zhao et al., 2003). Also, the infusion of natriuretic peptides in the presence of sildenafil synergistically increased cGMP and reduced RVSP in hypoxia-induced PAH rats (Preston et al., 2016). A recent clinical trial demonstrated that the combination of a neprilysin inhibitor (racecadotril) with a PDE5 inhibitor (sildenafil or tadalafil) acutely increases NP and cGMP levels and improves pulmonary hemodynamics (Hobbs et al., 2019). This indicates that neprilysin inhibitors could have therapeutic use in PAH.

The mechanism of NO/cGMP-induced vasodilation is partly mediated by the activation of voltage-gated K⁺ (K_v) channels (Cogolludo et al., 2001; Jackson, 2018). K_v1.5 is particularly known to contribute the most to K_v current in PASMCs. K_v7 channels have now been found to play a very significant role in generating K_v current in rat PASMCs (Mondéjar-Parreño et al., 2019). K_v7 channel activation is now thought to be imperative to the electrophysiological and relaxant effects of NO donors and riociguat (Mondéjar-Parreño et al., 2019). This makes activation of K_v7 channels a novel mechanism of action of vasodilators used in managing pulmonary arterial hypertension.

A new sGC stimulator, compound 13a (a pyrazolo [3,4-b] pyridine-3-yl pyrimidine derivative) has recently been found to exhibit similar in vitro vasorelaxation potential as riociguat on rat thoracic aorta rings and rat heart Langendorff preparation. Compound 13a also exhibited good oral bioavailability in male Beagle dogs which could make it a possible treatment candidate for PAH (Li et al., 2019). Compound 2 (a pyrazolo [3,4-b] pyridine derivative) is another novel sGC stimulator found to attenuate PAH (Hu et al., 2020). It reduced the migration of HPASMCs under hypoxic conditions significantly. Furthermore, it significantly improved RVSP, myocardial and vascular remodelling in hypoxia-induced PAH rats (Hu et al., 2020). We should notice that the PDE5 inhibitors block the breakdown of cGMP but these effects are dependent on NO availability and sGC activity. So the sGC stimulator or sGC activator may be effective in patients who have not sufficiently responded to a PDE5 inhibitor. Furthermore, sGC stimulator and sGC activator also show some difference, sGC stimulator acts on mature sGC and keep the activity of sGC, while sGC activator help to activate the damaged sGC.

Great strides have been made in PAH research in recent years but searching for new therapeutical agents remains a big challenge in this field. Current therapy only slows disease progression but does not cure the disease. So far the current clinical practice has strongly suggested that targeting the NO pathway is the strategy with the most potential. Critical molecules in the NO pathway such as DDAH1, PDE10 and cGKI show potential for becoming the next new therapeutic targets in PAH treatment. Compounds or derivatives from plant extracts also have very bright prospect. Furthermore, medicines harboring the ability to enhance NO signal and other key signaling pathways at the same time exhibit better hope to finally cure this thorny disease.