Pulmonary vascular dysfunction in metabolic syndrome

Abstract

Metabolic syndrome is a critically important precursor to the onset of many diseases, such as cardiovascular disease, and cardiovascular disease is the leading cause of death worldwide. The primary risk factors of metabolic syndrome include hyperglycaemia, abdominal obesity, dyslipidaemia, and high blood pressure. It has been well documented that metabolic syndrome alters vascular endothelial and smooth muscle cell functions in the heart, brain, kidney and peripheral vessels. However, there is less information available regarding how metabolic syndrome can affect pulmonary vascular function and ultimately increase an individual's risk of developing various pulmonary vascular diseases, such as pulmonary hypertension. Here, we review in detail how metabolic syndrome affects pulmonary vascular function.

Introduction

Metabolic syndrome affects nearly 100 million people in the United States (Roberts et al. 2013), and it is defined as an accumulation of multiple disorders that greatly increase a patient's risk of developing cardiovascular disease, neurological diseases, and other end‐organ complications. The potential aetiological factors for developing metabolic syndrome are obesity, high blood pressure, insulin resistance and glucose intolerance (Grundy et al. 2004). Additional factors such as aging and hormonal changes have also been found to increase a patient's risk for developing metabolic syndrome (Kassi et al. 2011). The role of metabolic syndrome in the development and progression of vascular complications is well documented in the mesenteric artery, heart, kidney, brain and retina (Yuan et al. 2007; Fowler, 2008; Li et al. 2017a; Sifuentes‐Franco et al. 2017); however, there is less information available regarding the effects of metabolic syndrome on pulmonary vascular function (Pugh & Hemnes, 2010). In this review, we will discuss the alteration of pulmonary vascular functions in diabetes, insulin resistance, obesity and dyslipidaemia. Because the change in systemic blood pressure does not correlate to the development and/or severity of pulmonary vascular disease, we will not discuss pulmonary vascular functions in systemic hypertension.

Animal models of metabolic syndrome

The diabetic animal models often overlap with insulin resistance, obesity and/or dyslipidaemia animal models. Therefore, in the section ‘Vascular dysfunction in hyperglycaemia’, we will discuss the literature citing the use of the following animal models. The animal models for type 1 diabetes (T1D) are streptozotocin (STZ)‐treated rats or mice, non‐obese diabetic (NOD) mice, or insulin receptor knockout mice. The animal models for types 2 diabetes (T2D) are mice given a low dose of STZ with high fat (HF) diet, db/db mice, KK mice, and Goto‐Kakizaki (GK) rats. For T1D mice, STZ is given by a i.v. single injection at a dose over 100 mg/kg or over 5 days, with consecutive intraperitoneal injections of STZ at 50 mg/kg. T1D rats were generated by a single i.v. injection of STZ at 60 mg/kg. T2D mice are induced by a single injection of STZ (75 mg/kg, i.p.) with a HF diet. NOD mice (T1D) and Goto‐Kakizaki rats (T2D) are derived from an inbred stock which exhibit diabetes spontaneously. The db/db mouse (T2D) has a mutation in the leptin receptor and exhibits hyperglycaemia, hyperinsulinaemia, obesity and dyslipidaemia. KK mice (T2D) carry the A^y mutation with hyperglycaemia, hyperinsulinaemia and mild obesity.

The HF diet‐induced obese mice, the ob/ob mice, and Zucker fatty rats are discussed in ‘Vascular function in obesity’ section (instead of in the ‘Dyslipidaemia section’). The ob/ob mouse is a strain which has a mutation of its leptin gene and exhibits extreme obesity, dyslipidaemia and insulin resistance. The Zucker fatty rat, characterized by the fa mutation in the leptin receptor, is a model of obesity and it exhibits hyperlipidaemia. Zucker diabetic fatty rats are considered separate from Zucker fatty rats because Zucker diabetic fatty rats also develop diabetes.

The spontaneous hypertensive rat (SHR) is a well‐known rat model for hypertensive studies; however, it becomes obese, dyslipidaemic and exhibits hyperinsulinaemia. Since the increase in systemic blood pressure is not associated with the development of pulmonary hypertension, pulmonary vascular dysfunction seen in the SHR might be more likely due to dyslipidaemia, obesity or hyperinsulinaemia. We will, therefore, discuss SHRs in ‘Vascular function in obesity.’

Regarding mouse models, apolipoprotein E knockout (ApoE^−/− KO) mice and low density lipoprotein receptor KO (LDLR^−/−) mouse are most commonly used for hyperlipidaemia studies. The increase in cholesterol levels and body weight are moderate without feeding a HF diet, because a HF diet exacerbates dyslipidaemia and leads to atherosclerosis. Pulmonary function in these strains will be discussed in the ‘Vascular function in dyslipidaemia’ section. More information regarding cardiovascular diseases (except pulmonary disease) in rodent models with obesity or dyslipidaemia can be found in review articles by Xiangdong et al. (2011) and Emini Veseli et al. (2017).

Vascular function in hyperglycaemia

More than 30 million people (9.4% of population in the United States) suffer from diabetes, and diabetes is the seventh leading cause of death in the United States. Type 1 diabetes (T1D) results from absolute deficiency in insulin secretion, while the cause of type 2 diabetes (T2D) is a combination of resistance to the action of insulin and eventual inadequate compensatory response to insulin secretion. Both the lack of insulin secretion and insulin resistance lead to hyperglycaemia.

Effect of hyperglycaemia on endothelial and smooth muscle cell function (systemic data)

Chronic hyperglycaemia induces vascular complications resulting in vision loss, cardiovascular disease, kidney disorders, and nerve damage (McCance et al. 1997; Hviid et al. 2004; Kolluru et al. 2012). Hyperglycaemia instigates endothelial cell (EC) dysfunction leading to macro‐ and microvascular diseases, which result in morbidity and mortality in patients with diabetes mellitus (DM) (Watkins, 2003; Yuan et al. 2007; Fowler, 2008; Hamilton & Watts, 2013). ECs function as (1) a barrier to prevent migration of circulating cells and extravasation of solutes, (2) a regulator of vascular tone, and (3) a key player in angiogenesis (Vita & Keaney, 2002; Esper et al. 2006; Hamilton & Watts, 2013). We will first discuss the effect of hyperglycaemia on these three EC functions, followed by smooth muscle cell (SMC) functions.

Hyperglycaemia enhances EC permeability in the kidney (Craven et al. 2001; Fowler, 2008), eyes (Grammas & Riden, 2003; Frank, 2015), brain (Sifuentes‐Franco et al. 2017), and heart (Yuan et al. 2000, 2007). Hyperglycaemia upregulates cell adhesion molecules such as VCAM‐1 (Koga et al. 1998; Ina et al. 1999; Jenkins et al. 2008; Elmarakby & Sullivan, 2012) and ICAM‐1 (Booth et al. 2002; Leone et al. 2004; Elmarakby & Sullivan, 2012); therefore, leukocyte adhesion and migration into the vascular wall is accelerated and vascular injury is exacerbated in the different organs of diabetic patients (Hirata et al. 1998; Rubio‐Guerra et al. 2009). One of the triggers to change the expression levels of adhesion molecules in diabetes is the excess production of reactive oxygen species (ROS), such as superoxide anion (O₂ ⁻) (Shi & Vanhoutte, 2009; Fatehi‐Hassanabad et al. 2010; Giacco & Brownlee, 2010; Pangare & Makino, 2012; Stadler, 2012; Joshi et al. 2014; Shaw et al. 2014; Asmat et al. 2016; Robson et al. 2017; Sifuentes‐Franco et al. 2017). In diabetes, it has been reported that ROS production is increased in vascular ECs due to increased activation of NADPH oxidases (Guzik et al. 2002; Ding et al. 2007; Kizub et al. 2014) and augmented mitochondrial ROS generation (Cheng et al. 2011; Pangare & Makino, 2012).

Attenuated endothelium‐dependent relaxation (EDR) and increased vasoconstriction are common features of diabetes (Giacco & Brownlee, 2010; Stadler, 2012; Hamilton & Watts, 2013; Kizub et al. 2014). Impaired EDR and augmented vascular contractility in diabetes are preceded by decreased nitric oxide (NO) bioavailability (Chew & Watts, 2004; Hamilton et al. 2007; Cho et al. 2013; Hamilton & Watts, 2013), attenuated endothelium‐derived hyperpolarization (EDH)‐dependent relaxation (Makino & Kamata, 2000; Fitzgerald et al. 2005; Makino et al. 2008, 2015; Salheen et al. 2015), and increased production of endothelium‐derived contracting factors (EDCFs), such as endothelin‐1 (ET‐1) and thromboxane A2 (Makino & Kamata, 1998, 2000; Elgebaly et al. 2007; Reriani et al. 2010; Kizub et al. 2014).

It has been reported that hyperglycaemia leads to an imbalance in endothelial angiogenesis and apoptosis (Martin et al. 2003; Tahergorabi & Khazaei, 2012; Okonkwo & DiPietro, 2017). In diabetic retinopathy, angiogenesis is significantly increased, and it is a major cause of blindness in diabetes (Kota et al. 2012; Behl & Kotwani, 2015; Ashraf et al. 2016; Okonkwo & DiPietro, 2017). Diabetic nephropathy is also characterized by increased angiogenesis (McGinn et al. 2003; Mironidou‐Tzouveleki et al. 2011; Nakagawa et al. 2013; Okonkwo & DiPietro, 2017). However, the pro‐angiogenic property is not a universal phenomenon in other organs. Endothelial cells are more apoptotic and/or less angiogenic in the heart (Yoon et al. 2005; Chung et al. 2006; Cortigiani et al. 2007; Makino et al. 2008, 2015; Messaoudi et al. 2009), mesenteric artery (Givvimani et al. 2011), and hind limb (Desposito et al. 2015; Lozeron et al. 2015; Tekabe et al. 2015; Ali et al. 2017; You et al. 2017) in diabetes.

Hyperglycaemia has also been shown to instigate vascular smooth muscle cell (SMC) dysfunction and vascular remodelling. Vascular SMCs isolated from diabetic patients or diabetic animals are more proliferative (Oikawa et al. 1996; Sasaki et al. 1999; Wang et al. 2006), resulting in increased media thickness (Wang et al. 2006). In addition, vascular SMCs are more contractile because of an augmented Ca²⁺ influx and increased EDCF production in diabetes (McIntyre et al. 2001; Alabadi et al. 2004; Tong et al. 2008; Navedo et al. 2010; El‐Najjar et al. 2017). In summary, chronic hyperglycaemia is sufficient enough to cause pathophysiological and functional changes in many organs including ECs and SMCs. These characteristic changes in vascular cells are implicated in the development of systemic hypertension, stroke, cardiac ischaemia and amputation in diabetes.

Pulmonary endothelial and smooth muscle cell function in hyperglycaemia

The limitation of this review article is that, to the best of our knowledge, there is no publication that demonstrates the function of pulmonary ECs or SMCs isolated from patients with metabolic syndrome. Therefore, we will discuss EC or SMC functions in animal models with metabolic syndrome and pulmonary arterial function (e.g. pulmonary arterial pressure (PAP)) in animal models and patients with metabolic syndrome. Pulmonary arterial function is separated from this section and discussed in ‘1.3 Hyperglycaemia and pulmonary hypertension’.

Pulmonary endothelial permeability and vascular inflammation

Several groups have shown increased macrophage and leukocyte adhesion in the lung capillary in diabetic rats (Popov et al. 1996; Vianna et al. 2003; Moral‐Sanz et al. 2012; Jagadapillai et al. 2016; Papinska et al. 2016), suggesting that inflammation might be occurring in the pulmonary vasculature. Cayir et al. (2015) demonstrated that ET‐1, ET‐A receptor and ET‐B receptor are overexpressed in the lung of STZ‐induced T1D rats, and the administration of bosentan (an ET‐A/B receptor blocker) in diabetic rats not only decreased the expression levels of ET‐1 and ET receptors, but also downregulated TGFα and TGFβ and decreased interstitial inflammation in the lung. Although the molecular mechanisms by which bosentan decreased TGFα and TGFβ mRNA levels are not discussed in this manuscript, the attenuation of inflammation by bosentan treatment could be expected because ET‐1 could increase endothelial membrane permeability (Porter et al. 2000) and enhance leukocyte adhesion (Helset et al. 1996). Increased ROS production in diabetic mice is also implicated in increased pulmonary EC permeability (Lu et al. 2014; Clemmer et al. 2016).

Vasoreactivity in pulmonary artery

Lopez‐Lopez et al. (2008) were the first group to demonstrate that pulmonary arteries (PAs) in STZ‐induced T1D rats exhibit attenuated EDR due to increased ROS production in the lung. The excess production of ROS resulted from the increased expression of p47phox (a regulatory subunit of NADPH) in diabetic rats. After this finding, they published another manuscript investigating the response of PAs to vasoconstrictors in T1D rats. They demonstrated that 5‐HT‐induced contraction was significantly increased in PAs of T1D rats, partially due to increased 5‐HT_2A receptor expression and cyclooxygenase metabolites (Lopez‐Lopez et al. 2011). In contrast, we found that both endothelium‐dependent and independent relaxation in PAs was not altered in T2D mice compared to the control (Pan et al. 2017a). These contrasting results might be due to different diabetic models (i.e. T1D or T2D), different species (i.e. rat or mouse) or different branches of pulmonary arteries used in the experiment (i.e. main artery or 4th–5th branch of PA).

Pulmonary vascular remodelling

The first report demonstrating the structural change in lung vasculature in diabetes might be the article from Dr Brody's laboratory in 1984 (Grant et al. 1984). They used a unique model of rat fetuses from a control or STZ‐induced T1D mother to assess the capillary growth in the lung. They indicated that the development of capillaries is significantly less in the fetuses of diabetic mothers than in fetuses of control mothers. A reduced number of capillaries would affect the efficiency of gas exchange in the lung; therefore, it would eventually affect the development of fetus. Romani‐Perez et al. (2015) showed decreased vascular density in the lung of adult STZ‐induced T1D rats, whereas we have demonstrated a significant increase in vascular density in the lungs of T2D mice (both inducible T2D mice and genetically modified T2D model, KK mice; Pan et al. 2017a). Although we could not find a report on pulmonary vascular density in diabetic patients, there is a report indicating increased pulmonary blood volume and flow in the lung of diabetic patients (Kuziemski et al. 2011). The increase in pulmonary flow might be a compensatory mechanism of attenuated alveolar‐capillary barrier function in diabetes (Ljubic et al. 1998; Goldman, 2003; Villa et al. 2004). To increase blood volume, the vessels have to be able to dilate effectively and the capillaries should be fully developed. Increased vascular density, which is seen in T2D mice, might be due to the need for increased pulmonary blood flow after the destruction of alveolar‐capillary barrier in diabetes.

It has been demonstrated that the pulmonary arteriolar walls are thickened in diabetic patients (Matsubara & Hara, 1991; Weynand et al. 1999), diabetic rats (Sui et al. 2017) and in infants of diabetic mothers (Colpaert et al. 1995). In an ex vivo study, high‐glucose treatment enhanced cell proliferation in human pulmonary arterial SMCs via increasing ROS generation (Sun et al. 2016), implying that pulmonary vascular resistance might be increased in diabetes. This will be discussed further in the ‘Hyperglycaemia and pulmonary hypertension’ section.

Another structural change seen in the pulmonary artery is that the caveolae number in pulmonary ECs are higher in T1D diabetic animals than in the control, which results from increased caveolin‐1 expression in diabetic ECs (Pascariu et al. 2004; Uyy et al. 2010). The increase in caveolin‐1 expression, followed by the increased number of caveolae, enhances the endothelium permeability (Mehta & Malik, 2006; Sun et al. 2011) and decreases endothelial NO synthase (eNOS) activation (Mineo & Shaul, 2012); therefore, the upregulation of caveolin‐1 in pulmonary ECs in diabetes might be a potential cause of augmented endothelial membrane permeability and/or attenuated EDR in the lungs of diabetic animals.

Hyperglycaemia and pulmonary hypertension

Pulmonary hypertension (PH) is characterized by increased pulmonary vascular resistance (PVR) and pulmonary arterial pressure (PAP). A sustained increase in PVR and PAP leads to right heart failure and eventual death. PAP is a function of cardiac output, and PAP and PVR are increased by concentric pulmonary vascular wall thickening, increased pulmonary arterial contractility and wall stiffness, and in situ thrombosis. Despite many publications demonstrating the adverse effect of hyperglycaemia on pulmonary endothelial or SMC function, there are a few clinical reports/articles discussing PH in diabetes (Table 1; Grinnan et al. 2016). Movahed et al. (2005) demonstrated that patients with DM exhibited significantly higher prevalence of PH than non‐diabetic patients. Makarevich et al. (2007) showed that mean PAP (mPAP) was significantly increased in patients with DM and chronic obstructive pulmonary disease (COPD) compared to patients with COPD alone. Several groups reported that PH patients with higher HbA1c (>6.0%) exhibited shorter ‘6 min walk distance’ (6MWD; a method to assess the disease condition and therapeutic outcome in PH patients) compared to PH patients with < 6.0% HbA1c (Pugh et al. 2011; Poms et al. 2013). Richter et al. (2016) also presented a positive correlation between HbA1c concentration and mPAP and a negative correlation between HbA1c levels and 6 min walk distance and cardiac index in chronic thromboembolic pulmonary hypertension (CTEPH) patients. The survival rate in PH patients with DM was worse than in PH patients without DM (Belly et al. 2012; Benson et al. 2014; Abernethy et al. 2015). Recently, we also reported that pulmonary arterial stiffness and elasticity are significantly higher in PH patients with DM than in PH patients without DM (Whitaker et al. 2018). These data suggest that there is a strong association between DM and the progression or severity of PH. It has to be noted that clinical reports related to PH and diabetes are mostly from patients with T2D, not from T1D patients.

Table 1.

Relationship between hyperglycemia and the progression or severity of PAH in patients

Patients	Progression or severity of PH	Follow‐up (years)	References
845,748 hypertension patient patients: 293,124 with DM 552,624 without DM	Hypertension patients with DM exhibit higher prevalence of PH than non‐diabetic hypertension patients	10	Movahed et al. 2005
158 patients: 55 with COPD and type 2 DM 40 with COPD 30 with type 2 DM 15 with COPD and type 1 DM 18 type 1 DM	COPD patients with type 2 DM show higher mPAP than patients with COPD without DM	NA	Makarevich et al. 2007
41 PAH patients: 23 (56%) with HbA1c > 6.0% 6 patients (15%) with HbA1c > 6.5% 12 (29%) with HbA1c < 6.0%	Shorter mean 6MWD in PAH patients with elevated HbA1c than PAH patients with normal HbA1c	1	Pugh et al. 2011
2,959 PAH patients: 789 without any other disease, 324 with DM 1846 with other common comorbid conditions except obesity	PAH patients with DM show worse 6MWD than PAH patients without any other disease	3	Poms et al. 2013
115 PAH patients: 50 with HbA1c < 5.7% 65 with HbA1c ≥ 5.7%	PAH patients with HbA1c ≥ 5.7% had significantly higher PH‐related death than PAH patients with HbA1c < 5.7%	11	Belly et al. 2012
113 PAH patients: 29 with DM 84 without DM	Lower 10‐year survival rate in PAH patients with DM than PAH patients without DM	6	Benson et al. 2014
261 PAH patients: 55 with DM 206 without DM.	Worse 5‐year survival in PAH patients with DM than in PAH patients without DM	5	Abernethy et al. 2015
45 CTEPH patients: 5 (11%) with HbA1c > 47.5 mmol/mol, 40 (89%) with HbA1c ≤ 47.5 mmol/mol	CTEPH patients with higher HbA1c levels exhibit decreased 6MWD compared to CTEPH patients with lower HbA1c	1	Richter et al. 2016
162 PAH patients: 27 with DM 135 without DM	PAH patients with DM show increased right ventricular afterload compared to PAH patient without DM	4	Whitaker et al. 2018

COPD, chronic obstructive pulmonary disease; CTEPH, chronic thromboembolic pulmonary hypertension; DM, diabetes mellitus; HbA1c, glycosylated haemoglobin; PAH, pulmonary arterial hypertension; 6MWD, 6 min walking distance.

Pulmonary artery systolic pressure is equivalent to right ventricular systolic pressure (RVSP) in the absence of pulmonary outflow obstruction; therefore, RVSP is commonly used to estimate the degree of PH in animal models. There are three methods generally used to induce PH in rodent models; monocrotaline‐induced PH in rat, hypoxia‐induced PH in mouse and rat, and Sugen/hypoxia‐induced PH in mouse and rat. Monocrotaline (an alkaloid in plants, 60–80 mg/kg) is injected to generate PH in rats, but does not work well in mouse. Hypoxia‐induced PH is developed by placing mice or rats in a hypoxic chamber (10% O₂) for over 3 weeks. The Sugen‐hypoxia‐induced PH (SuHx) model is made by subcutaneously injecting animals with Sugen 5416 (SU5416, a vascular endothelial growth factor receptor 2 blocker, 20 mg/kg body weight) once a week during 3 weeks of hypoxic exposure. The rat SuHx model exhibits severe PH with plexiform lesions as seen in patients with idiopathic pulmonary arterial hypertension. The plexiform lesion is, however, not obvious in the mouse SuHx model.

There is no significant difference in RVSP between control and T1D rats (Lopez‐Lopez et al. 2011; Moral‐Sanz et al. 2012); however, there is a slightly higher sensitivity to hypoxia in T1D rats than in control rats (Moral‐Sanz et al. 2012). We demonstrated that RVSP was slightly, but significantly, increased in T2D mice (both inducible T2D and KK mice) compared to the controls. Furthermore, 4 weeks of hypoxic exposure increased RVSP in T2D mice more than in the controls exposed to hypoxia, suggesting that diabetic mice are more susceptible to hypoxia (Pan et al. 2017a). The increased sensitivity to hypoxia was due, at least in part, to increased mitochondrial ROS generation, decreased superoxide dismutase subtype 2 (SOD2, a SOD expressed in the mitochondria) in diabetic pulmonary ECs, and subsequent attenuation of EDR in pulmonary artery of diabetic mice.

There is increasing evidence showing that the development and/or progression of PH is influenced by the status of metabolic syndrome; however, there is less information available regarding PH and diabetes compared to PH and obesity. Studies to reveal the molecular mechanisms underlying the diabetic priming effects would help develop personalized treatment for diabetic patients with PH.

Vascular function in obesity

Effect of obesity on endothelial and smooth muscle cell function (systemic data)

Obesity is a worldwide epidemic that affects millions of individuals ranging from adolescents to the elderly. It is estimated that over 30% of the adult population in the United States are obese (Flegal et al. 2002; Medoff et al. 2009). Obesity is caused by the overconsumption of fats and sugars combined especially with a sedimentary lifestyle. However, various genetic and environmental factors have also been found to be associated with obesity. Fatty tissue, or adipose tissue, is mainly composed of adipocytes and some other cell types including vascular cells. Adipocytes in the healthy body store energy, generate heat and serve as an endocrine organ releasing anti‐inflammatory adipokines such as adiponectin, omentin, CTRP and SFRP5. At obese status, adipocytes start secreting vast pro‐inflammatory adipokines including leptin, visfatin, resistin, TNF‐α and interleukins. Fat accumulation in the organs and increased circulating pro‐inflammatory adipokines lead to many cardiovascular diseases such as cardiac ischaemia, stroke, peripheral artery disease and PH (Fuster et al. 2016; Ortega et al. 2016; Sorop et al. 2017). It is also known that obesity is intimately associated with the development of insulin resistance, type 2 diabetes and dyslipidaemia (Kahn & Flier, 2000).

The ob/ob mouse is an obese mouse with insulin resistance and exhibits decreased coronary flow reserve (an indication of coronary microvascular function) (Westergren et al. 2015) and increased infarct size after myocardium ischaemia‐reperfusion (I‐R) (Pons et al. 2013). There are some contradictory reports on coronary function in Zucker fatty rats (Picchi et al. 2006; Prakash et al. 2006; La Bonte et al. 2008; Lekli et al. 2008; Mourmoura et al. 2013; Bender et al. 2015), but the inconsistencies might be solely due to differences in age and/or circumstances used to house the obese animals.

It has been shown that high‐fat (HF) diet‐induced obesity is associated with attenuated EDR in rat aorta (Zecchin et al. 2007; Kleinschmidt & Oltman, 2014; Liu et al. 2017), rat mesenteric artery (Kleinschmidt & Oltman, 2014), rat cerebral artery (Howitt et al. 2014), rat coronary artery (Kleinschmidt & Oltman, 2014), rat renal afferent arteriolar (Elmarakby & Imig, 2010), mouse aorta (Du et al. 2013; Schafer et al. 2013; Cheang et al. 2014; da Silva Franco et al. 2017; Santos et al. 2017) and mouse mesenteric artery (Wang et al. 2012). HF‐induced obese mice also exhibit enhanced agonist‐induced contraction in rat aorta (Liu et al. 2015), rat basilar artery (Lima et al. 2016) and mouse aorta (Mundy et al. 2007). Note that young obese mice (<10 weeks old) seem to exhibit enhanced EDR in some studies; therefore, the age and duration of the HF diet are a critical factor in determining endothelial function.

The increase in pro‐inflammatory adipokines not only leads to T2D in obesity via altered insulin sensitivity in adipocytes and muscle cells, but it also augments EC permeability, vascular inflammation and SMC proliferation, attenuates vascular relaxation and increases in vasocontractility (Yamagishi et al. 2001; Schafer et al. 2004; Beltowski, 2006; Maenhaut & Van de Voorde, 2011; Gil‐Ortega et al. 2016). Anti‐inflammatory adipokines show many beneficial effects on the cardiovascular system (Whitehead et al. 2006; Hui et al. 2012; Ohashi et al. 2014), including a decrease in vascular tone and vascular inflammation (Maenhaut & Van de Voorde, 2011). However, recent studies have pointed out that plasma adiponectin levels are positively correlated with increased mortality rates in diabetic patients (Cavusoglu et al. 2006; Pilz et al. 2006; Menzaghi & Trischitta, 2018). These unexpected and controversial outcomes must be explained with more case studies.

Pulmonary endothelial and smooth muscle cell function in obesity

Obesity is one of the risk factors of asthma, chronic obstructive pulmonary disease (COPD) and PH (Rubenfeld et al. 2005; Sood, 2010; Friedman & Andrus, 2012; Wei & Wu, 2012; Konter et al. 2013; Aytekin et al. 2014). However, there is much less information available regarding the direct effects of obesity on pulmonary endothelial and smooth muscle cell functions. This is partially due to the coexistence of obesity with other metabolic diseases in vivo, and there is no ex vivo experimental model to induce obesity or mimic the condition seen in obesity in the tissue culture dish or organ bath.

Pulmonary endothelial permeability and vascular inflammation

Clemmer et al. (2016) showed that pulmonary vascular permeability is significantly increased in Zucker‐fatty rats compared to the control and this increase was led by excess ROS production. HF‐induced obese mice did not exhibit increased lung permeability at the resting level; however, they are more susceptible to LPS‐induced acute lung injury (Shah et al. 2015). As described above, adipocytes release many kinds of pro‐inflammatory adipokines during obesity, whereas the secretion of anti‐inflammatory adipokines (e.g. adiponectin) is dramatically reduced in obesity. Medoff has published a review article about the beneficial effect of adiponectin on pulmonary vasculature (Medoff, 2013). As in other vasculature, adiponectin seems to inhibit inflammation, increase NO production in ECs, and decreases SMC proliferation and migration in the pulmonary vasculature. Konter et al. (2013) have reviewed the role of adiponectin in the prevention of acute lung injury. Adiponectin overexpression in HF‐induced obese mice decreases LPS‐induced acute lung injury (Shah et al. 2015). CTRP9 (C1q‐TNF‐related protein‐9) is the closest adiponectin paralogue, and it has been reported that 3 weeks of hypoxia exposure decreased CTRP9 expression levels in PAECs and PASMCs (Li et al. 2017b). Omentin, another anti‐inflammatory adipokine that has decreased secretion in obesity, also reduces the damage of acute lung injury (Qi et al. 2016).

Vasoreactivity in pulmonary artery

Moral‐Sanz et al. (2011) demonstrated that EDR was not altered, but vascular contraction was significantly attenuated in resistant PAs of Zucker fatty rats compared to Zucker lean rats. The reduced response to the vasoconstrictors was due to increased inducible NO synthase (iNOS) expression in PAs of Zucker fatty rats. Leptin is an adipokine that is secreted by food intake to reduce appetite. HF diet‐induced obese mice show an increase in plasma leptin concentration (Ahren et al. 1997), and oversecretion of leptin leads to cardiovascular diseases (Yang & Barouch, 2007; Koh et al. 2008). Leptin also regulates vascular tone in PAs. Leptin does not affect vascular tone in PAs of control rats, whereas it induces vasoconstriction in PAs of SHRs in a dose‐dependent manner (Gomart et al. 2017). Leptin‐induced contraction is independent of endothelium, but depends on excess Ca²⁺ influx via upregulation of TRPC6 and voltage‐gated calcium channels in pulmonary SMCs of SHR.

Pulmonary vascular remodelling

Pulmonary arterial stiffness is increased in children at risk of obesity (Mahfouz et al. 2012), implying that it would increase the probability of raising mPAP and lead to PH in this group of children. In an ex vivo study, hypoxia exposure (2% O₂ for 24 h) increased PASMC proliferation and migration, and CTRP9 (an adiponectin paralogue) treatment attenuated hypoxia‐induced proliferation and migration in PASMCs via inhibition of the TGF‐β1‐ERK1/2 pathway. Leptin treatment significantly increases pulmonary SMC proliferation and expression of adhesion molecules in pulmonary ECs (Huertas et al. 2015), implying that increased plasma leptin levels might be related to vascular remodelling of PAs seen in obesity.

Obesity and pulmonary hypertension

It is a well‐accepted concept that obesity is one of the risk factors in the development or progression of PH (Friedman & Andrus, 2012). Obesity hypoventilation syndrome (OHS) is characterized by obesity (BMI ≥ 30 kg/m²) with chronic daytime hypercapnia ($P_{\mathrm{aC}{\mathrm{O}}_2}$ ≥ 45 mmHg) or an obstructive sleep apnea. These patients have a high risk of developing PH and right heart failure. However, we will not discuss OHS in this review article (see instead: Kauppert et al. 2013; Almeneessier et al. 2017; Kaw, 2017). In addition, we will not discuss details about aminorex fumarate, fenfluramine and dexfenfluramine in this review. They are a potent anorexigen and are prescribed to reduce the patient's body weight. However, they have been claimed to instigate the development of primary PH and have been withdrawn from the market (Michelakis & Weir, 2001; Maclean & Dempsie, 2010).

The potential causes of development of PH in obesity (except in the above cases) include chronic elevation of left ventricular filling pressure and pulmonary endothelial dysfunction. It is well known that obesity is a risk factor of cardiomyopathy, which is characterized by an eccentric ventricular hypertrophy and diastolic heart failure in severely obese patients (Wong et al. 2004). Left ventricular diastolic dysfunction increases left ventricle filling pressure and pulmonary venous pressure, leading to a passive increase in PAP. In some cases, the patients will further develop ‘reactive’ increases in PAP due to a sustained increase in pulmonary venous pressure after structural change of PAs (i.e. pulmonary vascular remodelling via excess proliferation of SMCs and ECs; Segers et al. 2012).

There is contradictory evidence regarding the effect of obesity on the development of PH patients (Table 2). Several groups have found a protective effect of obesity on survival rate in PH patients (Zafrir et al. 2013; Agarwal et al. 2017; Mazimba et al. 2017), whereas Poms et al. (2013) demonstrated that PH patients with obesity showed a decreased 6 min walk distance compared to non‐obese patients. The most recent study has claimed that there is no significant correlation between obesity and survival rate in PAH patients; however, PAH patients with obesity exhibit shorter 6 min walk distances than PAH patients without obesity (Weatherald et al. 2018). Wu et al. (2007) demonstrated that a higher BMI is related to worse mPAP after chronic exposure to high altitude (= chronic hypoxia). More information and follow‐up data are required to fully understand the effect of obesity on the development, progression and survival of PH patients.

Table 2.

Relationship between obesity and the progression or severity of PAH in patients

Patients	Progression or severity of PH	Follow‐up (years)	References
30 normal subjects (BMI = 22–24) 28 obese subjects (BMI ≥ 25)	Staying at high altitude for 6 days increases mPAP in obese subjects more than in normal subjects	4.5	Wu et al. 2007
105 PAH patients: 43 with obesity (BMI ≥ 30 kg/m2) 62 without obesity (BMI ≤ 30 kg/m2).	PAH patients with obesity exhibit better survival rates than PAH patients without obesity	1.4	Zafrir et al. 2013
2,959 PAH patients: 789 without any other disease 956 with obesity (BMI ≥ 30 kg/m2) 1214 with other common comorbid conditions except obesity	PAH patients with obesity show shorter 6MWD than PAH patients without any other disease	3	Poms et al. 2013
267 PAH patients: 28 underweight (BMI < 18.9 kg/m2) 141 normal weight (BMI = 18.9–24.9 kg/m2), 71 overweight (BMI = 25.0–29.9 kg/m2) 19 obese (BMI = 30.0–34.9 kg/m2) 8 morbidly obese (BMI ≥ 35.0 kg/m2)	PAH patients with obesity exhibit better survival rates than PAH patients without obesity	1, 3, 5	Mazimba et al. 2017
18,450 PAH patients: 15,735 without obesity 2,715 with obesity	PAH patients with obesity have a lower hospital mortality than non‐obese PAH patients	9	Agarwal et al. 2017
1,255 PAH patients: 874 without obesity (BMI ≥ 35 kg/m2) 381 with obesity (BMI ≥ 30 kg/m2)	PAH patients with obesity show shorter 6MWD than PAH patients without any other disease	3	Weatherald et al. 2018

BMI, body mass index; 6MWD, 6 min walking distance.

Zucker fatty rats exhibit PH phenotypes with increased mPAP, Fulton index and wall thickness (Irwin et al. 2014). Meng et al. (2017) investigated the effect of a HF diet (20 weeks) on RVSP in 36 different mouse strains and found that 2/3 of strains were responsive to the HF diet and increased RVSP. Kelley et al. (2014) demonstrated that the HF‐induced increase in RVSP was prevented by the reduction of excess ROS production via nitrooctadecenoic acid administration. Some investigators examined whether obesity increases susceptibility to PH in rodent models. Lai et al. (2016) compared the effect of SU5416 (Sugen, 100 mg/kg) administration on the development of PH in Zucker fatty rats. To develop the PH rat model, a single injection of Sugen (20 mg/kg) with exposure to hypoxia for 3 weeks is commonly used. In this study, the authors did not use hypoxic exposure, and therefore the Sugen injection did not lead to severe PH in Zucker lean rats. On the other hand, Sugen administration significantly increased RVSP in Zucker fatty rats, suggesting that the precondition of obesity increased susceptibility to developing PH.

Adiponectin is a multifunctional hormone that has a beneficial effect on vascular functions (Hopkins et al. 2007; Xu & Vanhoutte, 2012; Achari & Jain, 2017). The effect of adiponectin on pulmonary vascular function has been extensively investigated ex vivo and described in review articles (Summer et al. 2011; Medoff, 2013). In spite of the many reports that have investigated adiponectin in pulmonary ECs or SMCs ex vivo, there is limited information about the effect of adiponectin on PH or pulmonary vascular function in vivo. Adiponectin KO mice appear to have increased inflammation in the lung and exhibit a slight, but significant, increase in RVSP (Summer et al. 2009). Weng et al. (2011) used an ovalbumin‐induced PH model and showed that the overexpression of adiponectin attenuated RVSP in ovalbumin‐induced PH mice.

Huertas et al. (2015) demonstrated that PAECs from idiopathic pulmonary arterial hypertension (IPAH) patients exhibit increased leptin production, and the PASMCs from IPAH patients expressed higher leptin receptor levels compared to the control cells. They also showed that chronic leptin administration (0.3 μg/g/day, i.p. for 3 weeks) increased susceptibility to hypoxia‐induced PH (HPH). Interestingly, hypoxia exposure increased leptin expression through HIF1α (Ambrosini et al. 2002). Chai et al. (2015) reported that not only hypoxia‐induced PH mice, but also monocrotaline‐induced PH mice, exhibited an increase in plasma leptin concentration. They further demonstrated that the hypoxia‐induced RVSP increase is significantly attenuated in ob/ob mice compared to wild type (WT) mice. Because ob/ob mice is the leptin KO model, the authors concluded that leptin plays a key role in the development of PH. Leptin itself exhibits a maladaptive effect on vascular functions including increasing proliferation and migration of SMCs, contracting SMCs, and altering EC functions, such as excess production of adhesion molecules, which would lead to undesirable inflammation. Therefore, blocking these phenomena by inhibiting leptin seems to be a promising option for the prevention of PH (Chai et al. 2015). However, inhibition of leptin signalling would not be the preferred treatment for PH patients because it would lead to obesity. What about inducing an overexpression of adiponectin or other anti‐inflammatory adipokines in PH patients? This has potential, but is still not ready to be used as a treatment. These adipokines have autocrine and endocrine effects on physiological and pathophysiological functions in the body. Changing the balance of adipokines might have unexpected adverse effects on the body. It would, therefore, be better to target specific signalling cascades without affecting the autocrine/endocrine system. However, further investigation is required.

Vascular function in dyslipidaemia

Effect of dyslipidaemia on endothelial and smooth muscle cell function (systemic data)

Nearly 37% of adults in the United States exhibit dyslipidaemia and 7% of children and adolescents aged 6–19 in the United States are dyslipidaemic. Dyslipidaemia is often referred to as a common comorbidity of diabetes and obesity. Dyslipidaemia itself does not cause any symptoms, but it leads to cardiovascular complications. Atherogenic dyslipidaemia is characterized by increased blood concentrations of LDL and triglycerides and a decreased plasma HDL concentration. Risk factors for developing atherogenic dyslipidaemia include the consumption of a high‐fat or high‐carbohydrate diet with a sedimentary lifestyle, or having a strong genetic predisposition (Musunuru, 2010).

It is known that circulating LDLs are oxidized through interaction with ROS (Zmyslowski & Szterk, 2017) and become ox‐LDL. OxLDLs are one of the harmful lipids that can alter vascular functions. The development of atherosclerotic plaque is primarily triggered by the binding of oxLDL to a lectin‐like ox‐LDL receptor‐1 (LOX‐1) of ECs. The binding subsequently stimulates the transcription of many genes which are linked to endothelial dysfunction (Mattaliano et al. 2009). EC dysfunction includes decreased barrier functions, increased adhesion molecules, excess production of cytokines and growth factors, and overproduction of ROS. These events lead to macrophage infiltration, LDL transportation and the generation of foam cells. The accumulation of foam cells stimulates SMC proliferation and migration, increases inflammatory cell infiltration, matrix degradation and the build‐up of plaque. The rupture of plaque results in thrombosis and vascular occlusion (Choudhury et al. 2004; Mitra et al. 2011; Di Pietro et al. 2016; Chistiakov et al. 2017).

Studies have found that pretreatment of oxLDL significantly attenuates EDR via the decrease in NO bioavailability in porcine coronary arteries (Tanner et al. 1991) and increases endothelin‐mediated contraction in porcine ciliary arteries (Zhu et al. 1999). Although the concentration used in these studies was high (30–300 μg/ml) compared to plasma concentration seen in patients with dyslipidaemia, local accumulation of oxLDL in the vascular wall during dyslipidaemic condition might reach an oxLDL level close to 30 μg/ml in vivo and affect EDR. OxLDL treatment reduced tube formation in human brain microvascular ECs (Pan et al. 2017b), human coronary ECs (Khaidakov et al. 2012), Human umbilical vein ECs (Ma et al. 2006) and endothelial progenitor cells (Di Santo et al. 2008; Ma et al. 2009). However, there are studies showing an increase in tube formation with oxLDL treatment (Dandapat et al. 2007; Yu et al. 2011). The reason for these controversial results is not known and demonstrates the need for further investigation. It is well documented that the accumulation of oxLDL stimulates SMC proliferation and migration, resulting in vascular remodelling in hyperlipidaemia. For the details, see the review articles by Chatterjee (1992), Gouni‐Berthold & Sachinidis (2002) and Barlic & Murphy (2007).

ApoE^−/− mice on a HF diet or double KO mice for ApoE and LDL receptors are ideal mouse models for atherosclerosis (Ewart et al. 2014; Getz & Reardon, 2016). ApoE plays a role in the clearance of lipids from the blood stream; therefore, ApoE^−/− mice on a HF diet exhibit extreme hyperlipidaemia. These mice also exhibit increased media thickness (Bonthu et al. 1997; Wang et al. 2000), enhanced endothelial permeability (Phinikaridou et al. 2012; Zhu et al. 2014) and attenuated EDR in aorta (Bonthu et al. 1997; Barton et al. 1998; Deckert et al. 1999; Wang et al. 2000), cerebral artery (Yamashiro et al. 2010) and carotid artery (d'Uscio et al. 2001; Ohashi et al. 2006). The causes of decreased EDR include excess ET‐1 production (Barton et al. 1998) and decreased NO bioavailability (d'Uscio et al. 2001; Ohashi et al. 2006; Yamashiro et al. 2010).

Pulmonary endothelial and smooth muscle cell function in dyslipidaemia

Pulmonary endothelial permeability and vascular inflammation

To mimic the condition of hyperlipidaemia ex vivo, the treatment of palmitic acid is often used in cultured cells. So far, there has been no report of a study investigating the effect of palmitic acid on the function of pulmonary vascular cells. However, we found a study investigating the effect of oxLDL on pulmonary vascular cells. OxLDL treatment in PAECs increases the expression of adhesion molecules and LOX‐1 (a receptor of oxLDL) in a dose‐dependent manner (Jiang et al. 2014), implying that PAECs become more permeable and more sensitive to oxLDL. ApoE^−/− mice are more sensitive to acute lung injury because they produce more cytokines than WT mice (Yamashita et al. 2014). ApoE^−/− mice also exhibit increased pro‐inflammatory adipokines, decreased anti‐inflammatory adipokines (Pereira et al. 2012; Kim et al. 2015; Lasrich et al. 2015; Zhu et al. 2015), and overexpress LOX‐1 in ECs (Zhao et al. 2016). This pro‐inflammatory predisposition might exaggerate the damage of an acute lung injury.

Pulmonary vascular remodelling

It has been demonstrated that hypoxia exposure to PASMCs increases LOX‐1 protein expression, and the inhibition of LOX‐1 by siRNA decreases hypoxia‐induced proliferation in PASMCs (Zhang et al. 2018), suggesting that excess production of oxLDL (the agonist of LOX‐1) in obesity would increase PASMC proliferation.

Dyslipidaemia and pulmonary hypertension

Accumulating evidence demonstrates that there are decreased levels of HDLs in patients with PAH, and plasma HDL level is positively correlated with right ventricular function and survival rate in PAH patients (Table 3; Heresi et al. 2010; Zhao et al. 2012; Larsen et al. 2016; Khirfan et al. 2018). Furthermore, Kopec et al. (2017) reported that the serum concentration of LDLs is associated with increased mortality. These data suggest that the serum HDL or LDL levels in PAH patients is indicative of the severity and survivability of PAH.

Table 3.

Relationship between dyslipidaemia and the progression or severity of PAH in patients

Patients	Progression or severity of PH	Follow‐up (years)	References
69 PAH patients 229 control patients	HDL concentration is lower in PAH patients than in control. Higher HDL is associated with decreased mortality and clinical worsening in PAH patients.	1.6	Heresi et al. 2010
76 PAH patients 45 control patients	HDL concentration is lower in PAH patients than controls. Patients with lower HDL exhibit more severe PAH and decreased 6MWD.	2	Zhao et al. 2012
227 PAH patients	Higher HDL negatively correlates with survival rate in PAH patients.	4.4	Larsen et al. 2016
90 CTEPH patients 69 PAH patients 254 control patients	CTEPH patients show lower HDL than control. Higher HDL is associated with lowered PVR in CTEPH patients.	2.8	Khirfan et al. 2018

CTEPH, chronic thromboembolic pulmonary hypertension; PAH, pulmonary arterial hypertension; PVR, pulmonary vascular resistance; HDL, high density lipoprotein; 6MWD, 6 min walking distance.

The results from gene microarray data indicate that the expression level of ApoE is decreased in the lungs with primary PH patients (Geraci et al. 2001). ApoE^−/− mice on a HF diet exhibit severe hyperlipidaemia with atherosclerosis, and they also develop PH (Hansmann et al. 2007; Lawrie et al. 2011; Renshall et al. 2018). The increase in RVSP in HF‐fed ApoE^−/− mice was reduced towards the level found in WT mice by chronic administration of PPARγ agonist (Hansmann et al. 2007), IL‐1 receptor antagonist (Lawrie et al. 2011), and a dual ET‐A/ET‐B receptor antagonist (Renshall et al. 2018). The lipid profile was not presented in any of these studies; therefore, it is not clear whether lowering of the LDL/triglyceride concentration in the blood contributes to a decrease in RVSP in these mice (especially in the case of PPARγ agonist). Hansmann et al. (2008) further investigated the molecular mechanisms of PPARγ agonist‐mediated prevention of PH. Bone morphogenetic protein receptor 2 (BMPR2) is an important molecule to regulate pulmonary SMC remodelling, and some patients with PAH have a mutation of the BMPR2 gene (Morrell, 2006; Orriols et al. 2017). They found that ApoE is essentially protecting BMP2‐induced PPARγ signalling cascade through the inhibition of the PDGF‐MAPK pathway. In the absence of ApoE, BMP2‐induced PPARγ activation is attenuated. PPARγ negatively regulates SMC proliferation; therefore, lower PPARγ activation increases SMC proliferation. Since SNPs of human BMPR2 are associated with LDL cholesterol concentration (Talmud et al. 2009), it is possible that PPARγ agonist might work to prevent the development/progression of PH in patients with hyperlipidaemia.

Although ECs express LOX‐1 more than other cell types (Sawamura et al. 1997), the level of LOX‐1 expression is relatively low under physiological conditions. It is, however, increased by any stresses or stimulations (Ogura et al. 2009; Lee et al. 2018). Zhang et al. (2017) demonstrated that hypoxia exposure (10%, 3 weeks) significantly increased LOX‐1 protein expression in PAs via the downregulation of miRNA let‐7g. Transgenic mice overexpressing LOX‐1 did not show any pulmonary vascular remodelling and phenotypes; however, these mice were more susceptible to developing HPH due to excess ROS production in PAs (Ogura et al. 2013).

Treatment of PH using medications for metabolic syndromes

There are many drugs that treat metabolic syndromes, and some of them have been tested clinically or preclinically in PH patients or animal models with PH (Rhodes et al. 2009). Note that we will not discuss medications that treat both systemic hypertension and PH.

Metformin is an anti‐diabetic drug which decreases hepatic glycogenolysis. Chronic metformin administration in hypoxia‐induced PH rats significantly decreased media thickness in PAs, enhanced EDR in PAs, and reduced mPAP via decreasing Rho kinase and MAPK activities and increasing eNOS activation (Agard et al. 2009). Metformin administration also decreased RVSP in Sugen/hypoxic‐induced PH rats (Dean et al. 2016) and Sugen‐treated Zucker fatty rats (Lai et al. 2016). There is an ongoing clinical trial to test metformin in PAH patients at Vanderbilt University (ClinicalTrials.gov Identifier: NCT01884051), and the valuable information from that study is expected to appear in the near future.

PPARγ agonist is a class of thiazolidinediones (TZDs) used in anti‐diabetic/insulin resistance drugs to increase insulin sensitivity. It is also known to improve dyslipidaemia. TZDs have been tested for PH in animals and showed beneficial effects in preventing PH (Hansmann et al. 2007; Hart, 2008). A clinical trial was started for PH patients with insulin resistance, but it has been terminated, perhaps because of difficulty recruiting enough patients (ClinicalTrials.gov Identifier: NCT00825266).

Glucagon‐like peptide‐1 (GLP‐1) receptor agonist is a new class of anti‐diabetic drug that stimulates secretion of insulin from the pancreatic β cells. Roan et al. (2017) demonstrated that a synthetic GLP‐1, exendin‐4, attenuates RVSP in monocrotaline‐induced PH rats via the reduction of PA media thickness and vasoconstriction in PAs.

Statins are the most studied drug for PH. They were developed for the purpose of lowering lipids, and there are at least seven different statins on the market in USA. Statins have been used not only for the management of plasma lipid levels, but also as a vasoprotective drug by restoring endothelial functions (Maron et al. 2000; Margaritis et al. 2014), so it is natural to test whether statins affect PH. There are several good review articles regarding statin treatment of PH patients (Katsiki et al. 2011; Wang et al. 2017). However, the results are not conclusive. There are some clinical trials showing a beneficial effect of statins on PH patients, whereas there are some trials which demonstrate no effect of statins on the progression and survival of PH patients. The key point is that statin treatment has NOT shown any adverse effect on PH patients so far, which means that further trials with careful intervention may lead to a definitive conclusion regarding their effectiveness.

Conclusion

We can be sure that pulmonary hypertension does not lead to metabolic syndrome. However, metabolic syndromes have been found to alter pulmonary vascular functions, and may potentially serve to instigate pulmonary hypertension, exacerbate its symptoms, or worsen the survival rate of patients with pulmonary hypertension. Therefore, special care is necessary during the choice of treatment for pulmonary hypertension in patients with metabolic syndrome. Some treatments could be beneficial for both pulmonary hypertension and metabolic syndrome, but some could be maladaptive to pulmonary hypertension. ‘Precision medicine’ is becoming an increasingly important concept during the selection of medication, but there is insufficient information available to classify the many possible outcomes of certain medications. We have just started to acknowledge the effect of metabolic syndrome on pulmonary vascular functions during the past decade, and further experimental and clinical information is required.

Additional information

Competing interests

The authors declare that there are no competing interests.

Author contributions

C.W., M.W. and A.T. drafted and reviewed the manuscript. A.M. conceived the project, drafted, reviewed, and edited the manuscript. All authors approved the final version of the manuscript.

Funding

This work was supported by a grant from the National Heart, Lung, and Heart Institute of the National Institutes of Health (HL142214 to A. Makino).

Biographies

Conor Willson is a fourth year undergraduate student working under the supervision of Dr. Makino at the University of Arizona. His research work focuses on endothelial dysfunction in pulmonary hypertension.

Ayako Makino is an Associate Professor of Physiology and Medicine at the University of Arizona. Her research interests center on the modulation of vascular function in pathological states, as well as the pathogenic mechanisms involved in the development and progression of pulmonary and coronary vascular diseases.