Reduced Circulating HIF‐1α, HIF‐2α, HIF‐3α, and VHL Levels in Patients With Pulmonary Arterial Hypertension

ABSTRACT

Pulmonary arterial hypertension (PAH) involves complex vascular remodeling. The hypoxia‐inducible factor (HIF) pathway is a central oxygen‐sensing mechanism, but its circulating profile in PAH is poorly defined. This case‐control study included 71 patients with PAH and 93 age‐ and sex‐matched healthy controls. Serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL were measured by ELISA. Serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL were significantly lower in PAH patients compared with controls (all p < 0.05). HIF‐1α demonstrated the highest diagnostic accuracy (AUC: 0.891; 95% CI: 0.836–0.947, p < 0.001). HIF‐2α levels were significantly lower in patients in the intermediate–high risk category compared to the low–intermediate risk category (p = 0.0449). Circulating HIF pathway proteins are downregulated in PAH. These markers, particularly HIF‐1α and HIF‐2α, show potential for diagnostic screening and risk stratification, suggesting a systemic exhaustion of oxygen‐sensing mechanisms.

Abbreviations

6 MWD

6‐min walk distance

AUC

area under the curve

CI

confidence interval

ELISA

enzyme‐linked immunosorbent assay

HIF

hypoxia‐inducible factor

mPAP

mean pulmonary arterial pressure

NT‐proBNP

N‐terminal pro‐B‐type natriuretic peptide

PAH

pulmonary arterial hypertension

PVR

pulmonary vascular resistance

ROC

receiver operating characteristic

TAPSE

tricuspid annular plane systolic excursion

VHL

von Hippel‐Lindau

Pulmonary hypertension (PH) is a progressive and potentially fatal disorder defined by pathologically elevated pressure within the pulmonary arterial circulation, and is associated with substantial morbidity and mortality despite contemporary advances in therapy. Hemodynamically, PH is diagnosed when the resting mean pulmonary artery pressure is ≥ 20 mmHg, as measured by right heart catheterization [1]. Although the clinical and hemodynamic framework of PH is well established, the molecular mechanisms that drive disease progression and determine inter‐individual variability in clinical severity remain incompletely understood.

PH is a heterogeneous clinical syndrome rather than a single disease entity, and is classified into five major groups based on shared hemodynamic characteristics and underlying pathophysiology: pulmonary arterial hypertension (PAH), PH due to left heart disease, PH associated with lung diseases and/or hypoxia, chronic thromboembolic PH (CTEPH), and PH with multifactorial or unclear mechanisms [2]. Among these, PAH is a distinct and particularly severe form, characterized by pre‐capillary hemodynamics with elevated pulmonary vascular resistance (PVR) and normal left‐sided filling pressures.

PAH arises from a complex interplay between endothelial dysfunction, aberrant smooth muscle cell proliferation, inflammation, metabolic reprogramming, extracellular matrix remodeling, and in situ thrombosis within the pulmonary arteriolar bed [3]. These processes culminate in progressive narrowing of the pulmonary vascular system, increased right ventricular afterload, and eventual right ventricular failure. Notably, many of these pathological features mirror cellular responses classically associated with hypoxic stress, even in the absence of sustained systemic hypoxemia, suggesting a fundamental disturbance in oxygen‐sensing and adaptive signaling pathways.

Clinically, PAH presents with nonspecific symptoms such as exertional dyspnea, fatigue, syncope, and angina, frequently leading to delayed diagnosis and advanced disease at presentation. Therefore, risk stratification is central to patient management and prognosis [4]. Although current algorithms integrate functional capacity, circulating biomarkers, imaging, and invasive hemodynamics, they remain imperfect surrogates of underlying disease biology, highlighting the need for molecular markers that directly reflect pathogenic signaling pathways.

Therapeutic strategies for PAH have largely focused on modulating vasoregulatory pathways, including endothelin, nitric oxide–soluble guanylyl cyclase–cyclic guanosine monophosphate, and prostacyclin signaling [5, 6, 7]. Although these approaches have improved clinical outcomes, they do not fully reverse pulmonary vascular remodeling or prevent disease progression. This therapeutic ceiling highlights the importance of identifying upstream regulatory pathways that integrate hypoxia, inflammation, metabolism, and vascular remodeling.

The hypoxia‐inducible factor (HIF) pathway is a central molecular hub that links these processes. HIFs are transcription factors that coordinate cellular adaptation to hypoxia by regulating genes involved in angiogenesis, erythropoiesis, glucose metabolism, proliferation, and survival [8]. The HIF family comprises three oxygen‐sensitive α‐subunits: HIF‐1α, HIF‐2α, and HIF‐3α, which exhibit distinct but overlapping biological functions. Under normoxic conditions, HIF activity is tightly controlled by the von Hippel–Lindau (VHL) protein, which mediates oxygen‐dependent ubiquitination and degradation of HIF α subunits.

Importantly, accumulating evidence indicates that HIF signaling may be activated independently of ambient oxygen tension through inflammatory mediators, mechanical stress, oxidative imbalance, and genetic alterations [9, 10, 11]. Dysregulation of this pathway has been implicated in pulmonary vascular remodeling, altered vascular tone, metabolic reprogramming, and maladaptive right ventricular responses. Germline mutations affecting components of the HIF–VHL axis cause human disorders such as erythrocytosis, and population‐based genetic variants modulating HIF activity influence physiological adaptation to hypoxia [12]. Collectively, these observations suggest that PAH is a compelling human model of maladaptive HIF signaling in the cardiopulmonary system.

Although HIF proteins primarily function as intracellular transcriptional regulators, emerging evidence suggests that components of the HIF pathway can be detected in the systemic circulation [13, 14]. Therefore, circulating HIF‐related proteins may represent an integrated, non‐invasive readout of tissue hypoxia, cellular stress, and dysregulated oxygen sensing across the affected organs. However, whether such circulating signals meaningfully reflect disease severity or hemodynamic burden in PAH patients remains largely unexplored.

We hypothesized that dysregulation of the HIF/VHL axis is systemically reflected in patients with PAH and that circulating levels of key HIF pathway components reflect clinically relevant aspects of disease severity. Accordingly, this study aimed to comprehensively characterize the serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL in patients with PAH and delineate their relationships with functional status, echocardiographic parameters, N‐terminal pro‐B‐type natriuretic peptide (NT‐proBNP) levels, 6‐min walk distance (6 MWD), and invasive hemodynamics. By investigating the systemic signature of HIF signaling, we aimed to provide new insights into the pathophysiological relevance of oxygen‐sensing pathways in PAH and their potential role in disease stratification.

1. Methods

1.1. Study Design and Population

This study was designed as a single‐center, cross‐sectional study comparing patients diagnosed with PAH with healthy controls. This study was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all participants. Ethical approval was granted by the Local Ethics Committee.

A total of 71 patients with PAH and 93 age‐ and sex‐matched healthy controls were included. PAH diagnosis was confirmed by right heart catheterization according to current clinical guidelines (mean pulmonary artery pressure ≥ 20 mmHg, pulmonary arterial wedge pressure ≤ 15 mmHg, PVR > 2 Wood units [WU]). Patients with PH secondary to left heart disease, primary lung disease, or CTEPH, as well as those with active infections or malignancies, were excluded. To minimize potential confounding related to chronic lung disease, all participants underwent pulmonary evaluation, and Group 3 pulmonary hypertension associated with lung diseases and/or hypoxia was excluded following consultation with the Department of Chest Diseases. All patients with PAH were clinically stable at the time of the evaluation. The control group consisted of volunteers without known cardiopulmonary, inflammatory, or malignant diseases; normal physical examination findings; and no history of chronic medication use. All controls had normal resting oxygen saturation levels at the time of enrollment.

1.2. Clinical, Hemodynamic, and Functional Assessment

Clinical and hemodynamic data, including WHO functional class [15], NT‐proBNP levels, right heart catheterization findings, echocardiographic parameters, and 6 MWD, were recorded.

Right‐heart catheterization was performed using standard techniques. The measurements included right atrial pressure, pulmonary artery systolic, diastolic, and mean pressures, pulmonary capillary wedge pressure, and systemic arterial pressure. Cardiac output was calculated using the indirect Fick method based on estimated oxygen consumption and simultaneous arterial and mixed venous oxygen saturation measurements. PVR, systemic vascular resistance, transpulmonary gradient, cardiac index, cardiac power output, pulmonary artery pulsatility index, and mixed venous oxygen saturation were calculated using the standard formulas [16].

All participants underwent comprehensive transthoracic echocardiography in the left lateral decubitus position using an Epiq 7 ultrasound system (Philips Medical Systems, Andover, MA, USA) equipped with a 5–1 MHz transducer. The examinations were performed by two experienced cardiologists who were blinded to the clinical and laboratory data. All measurements were performed in accordance with the American Society of Echocardiography [17].

Standard left‐ and right‐sided cardiac dimensions and functional parameters were assessed, including indices of right ventricular size, systolic function, pulmonary artery pressure, interventricular dependence, and right atrial remodeling. Congenital shunts and pericardial effusions were systematically evaluated in all patients.

A 6‐min walk test was performed on all patients, and walking distance, heart rate, and oxygen saturation were recorded before and after the test.

1.3. Blood Sampling, Biomarker, and Statistical Analysis

Venous blood samples were collected from the antecubital vein and centrifuged within 30 min (1500 × g, 15 min), and serum aliquots were stored at − 80°C until analysis. Serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL were quantified using commercial enzyme‐linked immunosorbent assay (ELISA) kits (Bioassay Technology Laboratory, Shanghai, China) according to the manufacturer's instructions. The minimum detectable concentration (sensitivity) was 0.01 ng/mL for HIF‐1α (Cat. No: E0422Hu), 0.044 ng/mL for HIF‐2α (Cat. No: E7164Hu), 3.32 ng/L for HIF‐3α (Cat. No: E4763Hu), and 7.03 ng/L for VHL (Cat. No: E6401Hu). For all kits, the intra‐ and inter‐assay coefficients of variation were < 8% and < 10%, respectively. High‐sensitivity cardiac troponin T (hs‐TnT) and NT‐proBNP levels were measured using electrochemiluminescence immunoassay.

Blood sampling, echocardiography, and the 6‐min walk test were performed within a maximum interval of 24–48 h to ensure clinical and hemodynamic stability. Risk assessment was performed using the 4‐strata model based on the WHO functional class, 6 MWD, and NT‐proBNP levels, as recommended by the 2022 ESC/ERS guidelines [2].

Statistical analyses were performed using SPSS v26.0. The normality of the distribution was assessed using the Shapiro–Wilk test. Normally distributed variables are expressed as mean ± SD, while skewed data are reported as median (IQR). Group comparisons were performed using the independent samples t‐test or Mann–Whitney U‐test, where appropriate. Categorical variables were analyzed using the chi‐squared test. Correlations between serum protein levels and clinical parameters were evaluated using Spearman's rank correlation test. Statistical significance was set at p < 0.05.

2. Results

2.1. Demographic and Clinical Data

The demographic characteristics, WHO functional classes, 6 MWD test results, and etiologies of the PAH cohort are summarized in Table 1. The PAH cohort was predominantly composed of idiopathic and congenital heart disease–associated PAH, with the majority of patients classified as WHO functional classes I–II.

Table 1.

Demographic characteristics and clinical evaluation results of the PAH group.

Height (cm)	161 ± 6.99
Weight (kg)	74.6 ± 17.5
Body surface area (m2)	1.81 ± 0.24
Diabetes mellitus status	n = 13 (% 18)
Type 1 PAH class	n = 71 (Total)
Idiopathic	n = 41 (57.7%)
Drug and toxin related	n = 1 (1.4%)
Connective tissue disease related	n = 7 (9.86%)
Congenital heart disease related	n = 22 (31%)
WHO functional capacity class	n = 71 (Total)
1	n = 26 (36.6%)
2	n = 19 (26.8%)
3	n = 18 (25.4%)
4	n = 8 (11.2%)
6‐min walking test – Distance (m)	330 (150–435)
Pre‐test heart rate (/min)	83.1 ± 18
Pre‐test oxygen saturation (%)	93 (88.5–95)
Post‐test heart rate (/min)	99.2 ± 22.7
Post‐test oxygen saturation (%)	92.5 (84–96)

Abbreviations: PAH, pulmonary arterial hypertension; WHO, World Health Organization.

2.2. Hemodynamic and Echocardiographic Findings

The hemodynamic parameters obtained using right heart catheterization are shown in Table 2. They confirmed pre‐capillary pulmonary hypertension with markedly elevated mean pulmonary arterial pressure (mPAP), PVR, and transpulmonary gradient, along with preserved pulmonary capillary wedge pressure.

Table 2.

Hemodynamic parameters of the PAH group obtained by right heart catheterization.

	Value (mean ± SD or median [25th–75th percentile])	Minimum‐maximum
Pulmonary capillary wedge pressure (mmHg)	11.8 ± 3.82	3–22
Systolic pulmonary artery pressure (mmHg)	64 (50–88)	32–153
Mean pulmonary artery pressure (mmHg)	40 (30.8–54)	20–90
Diastolic pulmonary artery pressure (mmHg)	22.5 (18–34)	5–65
Systolic aortic pressure (mmHg)	140 ± 27.1	95–200
Mean aortic pressure (mmHg)	102 ± 17.9	68–153
Diastolic aortic pressure (mmHg)	75.8 ± 13.7	40–118
Right atrium pressure (mmHg)	10 (7.5–14)	3–27
Aortic oxygen saturation (%)	88.8 ± 5.26	76–96
Pulmonary artery oxygen saturation (%)	62.3 ± 13.5	16–89
IVC oxygen saturation (%)	65.6 ± 11.6	29–86
SVC oxygen saturation (%)	61 (58–67.5)	22–88
Mixed venous oxygen saturation (%)	64 (58–68.3)	14.5–85.3
Cardiac output (L/min)	4.63 ± 1.24	2.2–8
Cardiac index (L/min/m2)	2.42 (2.15–2.66)	1.38–4.47
Transpulmonary gradient (mmHg)	26 (18.5–40)	13–82
Pulmonary vascular resistance (Wood Unit)	6.08 (3.85–9.1)	2.19–20.5
Systemic vascular resistance (dyne*sn*cm−5)	1507 ± 503	560–2880
Pulmonary artery pulsatility index	4 (3–6.25)	1.33–23.3
Cardiac power output (mmHg*L/min)	0.96 ± 0.32	0.39–1.57

Abbreviations: IVC, inferior vena cava; SVC, superior vena cava.

The key echocardiographic parameters reflecting right ventricular size are summarized in Table 3, and a comprehensive list of all the measured echocardiographic indices is provided in the Supporting Information (Table S1). Echocardiographic assessment demonstrated right ventricular enlargement and impaired systolic function in the PAH cohort, as evidenced by increased right ventricular dimensions, reduced tricuspid annular plane systolic excursion (TAPSE) and FAC values, and elevated estimated pulmonary artery systolic pressure.

Table 3.

Echocardiographic findings of the PAH group.

	Value (mean ± SD or median [25th–75th percentile])	Minimum‐maximum
Right atrium area (cm2)	22.7 ± 7.82	8.5–42.2
Right atrium volume (mL)	82.2 ± 43.7	17–211
Right ventricular basal diameter (mm)	46.5 ± 8.15	28–72
Right ventricular fractional area change (%)	37.9 ± 10.7	10–62
Tricuspid annular plane systolic displacement (mm)	21.1 ± 6.06	10–37
Tricuspid diastolic tissue Doppler S' velocity (cm/s)	12.4 ± 3.33	6.5–21
Inferior vena cava diameter (cm)	1.87 ± 0.57	0.7–3
Pulmonary artery systolic pressure (mmHg)	64.2 ± 29.6	21–170
Tricuspid regurgitation rate (m/s)	3.6 ± 0.86	2.12–6.12

2.3. Biomarker Levels in Study Groups

The serum NT‐proBNP and hs‐cTnT levels in the study population are shown in Table 4. In our study, hs‐cTnT levels in patients with PAH were found to be significantly higher than those in the control group (10.8 (6.81–21.3) vs. 3.0 (3.0–3.64), p < 0.0001), which is consistent with the current literature [18].

Table 4.

High‐sensitivity cardiac troponin T, N‐terminal pro‐B‐type natriuretic peptide, serum HIF‐1α, HIF‐2α, HIF‐3α, and VHL levels of the control group and PAH patients.

	Control group (n = 93)	PAH group (n = 71)	p value
NT‐proBNP (pg/mL)	—	553 (171–1974)
hs‐Tn (ng/L)	3 (3–3.64)	10.8 (6.81–21.3)	< 0.0001
HIF‐1α (ng/mL)	4.06 (2.89–7.84)	1.39 (0.9–1.9)	< 0.0001
HIF‐2α (ng/mL)	7.95 (4.87–18.2)	5.73 (5.09–6.5)	0.002
HIF‐3α (ng/mL)	264 (124–825)	109 (73.2–141)	< 0.0001
VHL (ng/mL)	654 (352–1555)	506 (361–716)	0.0295

Abbreviations: HIF, hypoxia‐inducible factor; Hs‐Tn, high‐sensitivity cardiac troponin; NT‐proBNP, N‐terminal pro‐B‐type natriuretic peptide; PAH, pulmonary arterial hypertension; VHL, von Hippel‐Lindau.

Serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL were significantly lower in the PAH group than in the control group (all p < 0.05) (Table 4, Figure 1).

Figure 1.

Serum HIF‐1α (A), HIF‐2α (B), HIF‐3α (C), and VHL (D) levels in the control group and PAH patients. HIF, hypoxia inducible factor; PAH, pulmonary arterial hypertension; VHL, von Hippel‐Lindau.

2.4. HIF‐2α and Disease Severity

Among the measured proteins, only HIF‐2α showed a significant stepwise decrease as the disease severity (risk score) increased. Based on the 4‐strata PAH risk assessment model, patients classified as low–intermediate risk (1–2 points) had significantly higher HIF‐2α levels than those in the intermediate–high risk category (3–4 points) (6.0 [5.42–6.82] vs. 5.42 [4.86–6.27] ng/mL, p = 0.0449) (Figure 2).

Figure 2.

HIF‐2α results of patients according to the 4‐strata PAH risk assessment. HIF, hypoxia inducible factor; PAH, pulmonary arterial hypertension.

2.5. Diagnostic Performance Analysis

Receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic potential of these markers (Figure 3; Tables 5 and S2). HIF‐1α demonstrated the highest diagnostic accuracy in distinguishing PAH patients from healthy controls, with an area under the curve (AUC) of 0.891 (95% confidence interval [CI]: 0.836–0.947, p < 0.001). HIF‐3α also showed strong diagnostic performance (AUC: 0.755, p < 0.001), followed by HIF‐2α (AUC: 0.640), and VHL (AUC: 0.599).

Figure 3.

ROC curves of serum HIF pathway proteins for the identification of pulmonary arterial hypertension.

Table 5.

Diagnostic performance of HIF pathway proteins for identifying PAH.

Biomarker	Cut‐off (ng/mL)	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	AUC (95% CI)	Youden index
HIF‐1α (ng/mL)	2.14	93.55	81.69	87.00	90.62	0.891	0.752
HIF‐2α (ng/mL)	6.87	58.06	81.69	80.60	59.79	0.640	0.324
HIF‐3α (ng/mL)	178.4	64.52	84.51	84.51	64.52	0.755	0.490
VHL (ng/mL)	772	45.16	80.28	75.00	52.78	0.599	0.254

Abbreviations: AUC, area under the curve; HIF, hypoxia‐inducible factor; NPV, negative predictive value; PAH, pulmonary arterial hypertension; PPV, positive predictive value; VHL, von Hippel‐Lindau.

Subgroup analysis based on PAH etiology, including idiopathic, congenital heart disease‐associated, and connective tissue disease‐associated forms, revealed no significant differences in serum HIF‐1α, HIF‐2α, HIF‐3α, and VHL levels (all p > 0.05; Table S3). This suggests that the observed downregulation of the HIF pathway is consistent across PAH etiologies.

No significant linear correlations were observed between serum HIF‐1α, HIF‐2α, HIF‐3α, or VHL levels and individual hemodynamic parameters, including mPAP, PVR, and cardiac index, or markers of right ventricular function, such as the TAPSE/PASP ratio and NT‐proBNP (all p > 0.05). The complete correlation matrix is presented in Table S4.

3. Discussion

PAH is a rare but serious disease characterized by progressive narrowing and remodeling of the pulmonary vascular bed, which can result in right ventricular failure and mortality [1]. Numerous mechanisms, including endothelial dysfunction, smooth muscle hyperplasia, inflammation, thrombosis, and genetic factors, are involved in its pathogenesis; these processes have been shown to be closely linked to oxygen sensing and response systems at the cellular level [7]. HIFs and VHL, which regulate these proteins, are among the key regulators of cellular responses to oxygen levels and play important roles in the pathophysiology of PAH [19, 20].

This study aimed to evaluate the serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL proteins in patients with PAH to reveal the relationship between these biomolecules and the disease and to examine their relationship with clinical parameters. These findings indicate that all HIF isoforms and VHL protein levels are significantly lower in patients with PAH than in healthy individuals, and that HIF‐2α levels, in particular, may be associated with disease risk and severity.

In our study, HIF‐1α levels in patients (1.39 (0.9–1.9)) were significantly lower than those in the control group (4.06 (2.89–7.84)) (p < 0.0001) (Table 4, Figure 1A). However, conflicting findings regarding HIF‐1α levels in PAH patients have been reported. Increased HIF‐1α accumulation in the pulmonary vascular walls of patients with PAH has been reported to be a result of chronic hypoxia, inflammation, and other pro‐PH stimuli. Some studies have shown increased HIF‐1α levels in endothelial and smooth muscle cells of the lungs of patients with PAH [21, 22]. In contrast, it has been reported that HIF‐1α protein expression is significantly reduced in pulmonary artery smooth muscle cells (PASMC) isolated from idiopathic PAH compared to controls [8]. Barnes et al. reported decreased HIF‐1α protein levels and increased vasoconstrictor tone in primary PASMC cultures obtained from PAH patients compared with healthy controls [23]. This finding suggests that HIF‐1α suppression in smooth muscle cells may enhance pulmonary arterial vasoconstriction and contribute to PAH pathogenesis. Conversely, it is also known that systemic activation of HIF‐1α can promote vascular remodeling in animal models of hypoxic PH; in fact, mice with partial HIF‐1α deficiency display delayed development of pulmonary hypertension and reduced vascular wall thickening under chronic hypoxia [24]. These data suggest that HIF‐1α signaling in PAH may have different effects depending on the cell type and disease stage, and that both excessive HIF‐1α activity and insufficient HIF‐1α levels may play a role in its pathogenesis. Our ROC analysis revealed that HIF‐1α possesses excellent diagnostic accuracy (AUC: 0.891), suggesting its clinical utility as a non‐invasive screening tool in the diagnostic work‐up of PAH (Figure 3).

In our study, HIF‐2α levels in patients (5.73 (5.09–6.5)) were significantly lower than those in the control group (7.95 (4.87–18.2)) (p < 0.002) (Table 4, Figure 1B). Increasing evidence suggests that HIF‐2α plays a significant role in the pathophysiology of PAH. Various studies have shown that HIF‐2α expression is particularly elevated in pulmonary artery endothelial cells in the lungs of PAH patients [22]. HIF‐2α has been reported to be a transcription factor that controls the expression of genes that regulate long‐term cellular and systemic adaptation processes such as erythropoiesis and vascular remodeling [25]. Excessive stabilization of endothelial HIF‐2α under hypoxic conditions results in significant pulmonary vascular remodeling, as observed in experimental models [26, 27]. These findings support the hypothesis that the HIF‐2α signaling pathway is a central driving factor in the pathogenesis of PAH.

According to the 4‐strata PAH risk assessment model [2], patients with a score of 1–2 were classified as low–intermediate risk, whereas those with a score of 3–4 were classified as intermediate‐high‐risk. A significant difference was observed in HIF‐2α levels between the low‐risk (6.00 (5.42–6.82)) and high‐risk group (5.42 (4.86–6.27)) (p = 0.0449) (Figure 2). The observation that HIF‐2α levels were significantly lower in high‐risk patients than in low‐ and intermediate‐risk patients suggests its potential utility as a non‐invasive biomarker for risk stratification and monitoring disease severity in PAH. Furthermore, our ROC analysis supported this finding, demonstrating that HIF‐2α can discriminate between disease states, although its prognostic power should be validated in larger longitudinal cohorts.

In our study, HIF‐3α levels in patients (109 (73.2–141]) were significantly lower (p < 0.0001) than those in the control group (264 (124–825)) (Table 4, Figure 1C). Although HIF‐3α is a less‐characterized member of the HIF family, data in the literature suggest that it may play a regulatory role in the pathophysiology of PAH. One study showed that the HIF‐3α protein is expressed particularly in pulmonary fibroblasts in the lung, but HIF‐3α levels in pulmonary fibroblasts obtained from patients with PAH were significantly lower than those in the controls [8]. Some HIF‐3α isoforms may limit angiogenesis and fibrosis by suppressing the transcription of HIF‐1α target genes [28]. Therefore, a decrease in HIF‐3α levels in PAH may facilitate pathological angiogenesis or remodeling. In our study, the suppression of all HIF isoforms (HIF‐1α, HIF‐2α, and HIF‐3α) in PAH patients suggests that the HIF pathway is generally impaired and that this imbalance may play a role in the pathophysiology of the disease.

In our study, VHL levels were significantly lower in the patients (506 (361–716)) than in the control group (654 (352–1555)) (p = 0.0295) (Table 4, Figure 1D). In PAH pathogenesis, the VHL protein is responsible for maintaining the balance of HIF proteins as a central element of the cellular oxygen‐sensitization mechanism. Under normoxic conditions, VHL suppresses HIF‐mediated gene expression. Under hypoxia, VHL binding to HIF is inhibited, leading to HIF‐α accumulation [29]. Both genetic and acquired impairments in the functioning of the VHL‐HIF axis have been reported in patients with PAH. Loss‐of‐function mutations in VHL lead to uncontrolled HIF‐1α and HIF‐2α activity, resulting in the excessive production of potent vasoconstrictors and mitogenic mediators, such as endothelin‐1 and vascular endothelial growth factor, thereby elevating PVR and pressure [30, 31]. The simultaneous decrease in both VHL and HIF proteins suggests global downregulation of the oxygen‐sensing machinery in chronic PAH rather than a simple enzyme‐substrate relationship.

The low HIF‐α levels observed in patients with PAH in our study suggest that the VHL‐prolyl hydroxylase domain (PHD) system may be overactive. Increased PHD activity may cause cells to rapidly degrade HIF proteins despite a hypoxic environment, impairing the adaptive HIF response. This may contribute to the sustained contraction of smooth muscle cells deprived of the vasodilator and anti‐proliferative effects of HIF‐1α and accelerate vascular remodeling [23]. However, this remains speculative because peripheral serum concentrations do not directly reflect tissue‐specific regulatory mechanisms. Alternative explanations include reduced cellular release, increased proteolytic degradation in the plasma, or assay‐related differences in the detection of intracellular and extracellular forms.

The relationship between VHL and HIF proteins in PAH requires a tightly balanced regulatory system. Excessive HIF signaling (e.g., due to VHL loss‐of‐function) can induce pulmonary vascular remodeling and hypertension [12, 22], whereas insufficient HIF signaling (e.g., due to heightened PHD activity) may worsen the disease [23].

Although hypoxia‐driven mechanisms are central to PAH pathogenesis, the phenotype of our patient group appeared to be characterized more by endothelial dysfunction than by hypoxia‐induced pulmonary hypertension. HIF‐2α is selectively expressed in vascular endothelial cells and plays a critical role in regulating endothelial integrity and intravascular signaling pathways. As reported in literature, changes in HIF‐2α expression are associated with endothelium‐mediated vascular remodeling and proliferative processes [32]. In our study, the significant difference in HIF‐2α levels between high‐risk and low‐risk PAH patients supports the closer relationship between HIF‐2α and disease severity and endothelial damage.

In patients with chronic pulmonary hypertension, progressive impairment of endothelial function reduces the release of vasodilatory mediators. The decreased HIF protein levels observed in our study may reflect long‐standing endothelial fibrosis and the loss of physiological hypoxia‐responsive signaling within the pulmonary vascular bed. It is also plausible that circulating HIF levels vary over the course of the disease, being higher at the time of initial hypoxic stress but diminishing in later stages due to cellular desensitization. Given that the interval between symptom onset and definitive diagnosis of PAH frequently exceeds 2 years, this temporal decline may partly explain the lower HIF concentrations detected in our cohort. A longitudinal assessment of HIF levels in patients diagnosed at earlier stages of the disease could help clarify this dynamic pattern and provide further insight into the progression of hypoxia‐related pathways in PAH.

Considering that our patients had relatively long follow‐up durations after diagnosis, and assuming an equally long pre‐diagnostic disease period, our hypothesis is plausible. Moreover, the markedly elevated mean PVR values in our cohort [6.08 WU (3.85–9.1)] likely reflect advanced fibrosis in the peripheral pulmonary arterioles, which may indicate reduced HIF production within the pulmonary vasculature. These interpretations are based solely on circulating levels obtained from peripheral blood. Samples derived directly from the pulmonary circulation or molecular analyses from lung biopsies or cultured pulmonary cells may provide more accurate mechanistic information than peripheral serum measurements. The observed reduction in circulating HIF isoforms and VHL may reflect an altered systemic turnover or disease stage‐dependent changes in tissue expression. However, causal inferences regarding PHD/VHL overactivity cannot be made using peripheral serum measurements alone.

To the best of our knowledge, this is the first human study to simultaneously and comparatively evaluate the serum levels of HIF‐1α, HIF‐2α, HIF‐3α, and VHL proteins in PAH patients. Only a few studies have measured serum HIF‐1α levels in patients with PAH associated with connective tissue diseases [13]. Furthermore, to our knowledge, no human study has evaluated HIF‐2α, HIF‐3α, or VHL levels in patient serum. Therefore, our study fills an important gap and represents a novel step toward defining the systemic roles of these key molecular regulators in PAH pathophysiology.

3.1. Limitations

Owing to its single‐center observational design, the external validity of our findings may be limited. Serum biomarker concentrations were measured using ELISA, a method that is inherently sensitive to pre‐analytical and analytical variability. The lack of long‐term clinical outcomes and confirmatory tissue‐level analyses in this study limited our ability to assess the true prognostic role of these biomarkers.

Another important limitation of the present study is the absence of comparator groups with other pulmonary diseases. Although patients with Group 3 pulmonary hypertension were excluded through pulmonary consultation, circulating HIF pathway proteins may also be altered in chronic lung diseases associated with hypoxia and inflammation. Therefore, future studies including patients with chronic obstructive pulmonary disease, interstitial lung disease, and hypoxia‐related pulmonary hypertension may provide further insight into the disease specificity and biomarker potential of HIF‐ and VHL‐related molecules in PAH.

In conclusion, circulating HIF‐1α, HIF‐2α, HIF‐3α, and VHL protein levels were significantly reduced in patients with PAH compared with healthy controls. These findings suggest a potential association between dysregulation of the HIF pathway and the pathophysiology of PAH. However, further studies involving larger cohorts and other pulmonary disease groups are required to better define the disease specificity, diagnostic performance, and clinical utility of these molecules as potential biomarkers in PAH.

Author Contributions

Hasan Kan: conceptualization, methodology, investigation, data curation, writing – original draft. Yakup Alsancak: supervision, conceptualization, methodology, project administration, writing – review and editing, final approval. Meltem Uyaner Kan: laboratory analysis (ELISA), biochemical data interpretation, data collection, literature review. Emine Göktaş: laboratory analysis (ELISA), manuscript review. Mustafa Çelik: critical review of the manuscript. Ahmet Seyfeddin Gürbüz: critical review of the manuscript. Mehmet Akif Düzenli: critical revision of the manuscript. During the preparation of this manuscript, the authors used ChatGPT (OpenAI) solely for linguistic and stylistic improvements. The tool was not used for data analysis, interpretation, or the generation of scientific conclusions. Napkin.ai was used to generate the graphical abstract. All content was reviewed, revised, and approved by the authors, who take full responsibility for the final manuscript.

Funding

The authors have nothing to report.

Ethics Statement

This study was conducted in accordance with the Declaration of Helsinki and was approved by the Necmettin Erbakan University Meram Faculty of Medicine Ethics Committee (Approval No 2024/5036, Application ID: 19860). Written informed consent was obtained from all participants.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File

Kan H., Alsancak Y, Uyaner Kan M., et al., “Reduced Circulating HIF‐1α, HIF‐2α, HIF‐3α, and VHL Levels in Patients With Pulmonary Arterial Hypertension,” Pulmonary Circulation 16 (2026): e70326. 10.1002/pul2.70326.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request. Hasan Kan and Yakup Alsancak had full access to all of the data in the study and took responsibility for the integrity of the data and the accuracy of the data analysis.