This article is mentioned in:

Introduction

Pulmonary arterial hypertension (PAH) is one of the most severe types of chronic cardiopulmonary disorder with an incidence of 15–50/million, high mortality and poor prognosis, and the 5-year survival rate is 38%. The PAH is caused by a multitude of diseases and diverse pathogenic mechanisms (such as left heart failure, Interstitial lung disease with hypoxemia) and characterized by progressive pulmonary artery remodeling, resulting in elevated mean pulmonary arterial pressure (mPAP) >25 mmHg at rest and pulmonary vascular resistance, culminating in right heart failure and mortality (1). At present, the pathogenesis of PAH has not been completely elucidated. However, PAH pulmonary vascular remodeling is histopathologically similar to malignant tumors, including cell proliferation and glucose metabolism pathway abnormalities (2). These findings have led to PAH being denoted as the ‘cancer of the cardiovascular world’. Like cancer, PAH is marked by autonomous growth signaling, resistance to antiproliferative cues, evasion of programmed cell death and persistent neovascularization (3). Patients with PAH exhibit inactivated or deleted oncogenes, including protein tyrosine phosphatase, large tumor suppressor kinase 1 gene, forkhead box O1 and phosphoprotein p53, alongside an increased expression of telomerase reverse transcriptase in pulmonary arterial smooth muscle cells (PASMCs) (4).

The pathological characteristics of both cancer and PAH can be regarded as a manifestation of glucometabolic reprogramming resulting from cell proliferative disorder (5). Previous studies have demonstrated that dysregulated activation of hypoxia-inducible factor-1α (HIF-1α) influences the activity of a number of aerobic glycolytic enzymes, including glucose transporter 1 (Glut1), hexokinase 2 (HK2), pyruvate dehydrogenase kinase 1(PDK1), and lactate dehydrogenase (LDH), which redirects ATP generation from mitochondria to the cytosolic space, a phenomenon that can transpire under the circumstance of sufficient oxygen availability and is commonly referred to as aerobic glycolysis or the Warburg effect (6,7). These changes potentiate uptake of glucose in PASMCs, suppress the activity of the tricarboxylic acid (TCA) cycle and increase rapid cell proliferation by the upregulation of the pentose phosphate pathway in synthesis of purine nucleotides (8). Dichloroacetate (DCA), a PDK1 inhibitor, inhibits pulmonary artery remodeling and decreases pulmonary vascular resistance by promoting oxidative phosphorylation of glucose and facilitating the TCA metabolic pathway (9). Typically, mitochondrial oxidative phosphorylation is the primary source of energy in differentiated cells, whereas a number of tumor cells predominantly rely on aerobic glycolysis, commonly referred to as the Warburg effect (10). Despite the low efficiency of ATP production by aerobic glycolysis, it provides raw materials for tumor cell proliferation and promotes tumor cell metastasis (11). In summary, inhibition of the Warburg effect by targeting regulation of enzymes associated with to glucose metabolism could impede pulmonary vascular remodeling and decrease pulmonary circulation resistance.

Oroxylin A (OA), a flavonoid derivative of Scutellaria baicalensis, possesses therapeutic properties against numerous types of malignant tumors (12). OA has been demonstrated as a therapeutic candidate for breast cancer by decreasing HK2 expression, leading to a substantial decrease in proliferation of MDA-MB-231 and MCF-7 breast cancer cell lines (13). Moreover, OA inhibits HIF-1α and glycolysis via the downregulation of aerobic glycolytic enzymes, including HK2, LDH and PDK1. It also decreases the levels of complex III in the electron transport chain. These mechanisms contribute to decreased lactate production in hypoxic HepG2 hepatocellular carcinoma cells (14,15). Overall, OA as an anticancer drug achieves therapeutic effects by inhibiting cell proliferation and glycolysis, which suggests potential use of OA in PAH.

The present study aimed to assess therapeutic potential of OA in a monocrotaline (MCT)-induced PAH rat model to evaluate underlying mechanisms of the beneficial effects of OA in PAH, particularly through mitigating Warburg effect.

Materials and methods

Animals and ethics approval

A total of 48 adult male Sprague-Dawley rats [age, 3 months; weight, 250–280 g; specific-pathogen-free (SPF)-grade; certificate no. SCXK 2019–0004] was purchased from Hunan Slake Jingda Experimental Animal Co., Ltd.). The rats were housed in an SPF-grade animal facility of Zunyi Medical University [Zunyi, China; certificate no. SYXK (QIAN) 2021–0003] and had ad libitum access to food and water, 12/12- h alternation of day and night at 20–24°C and 50–60% humidity. Standard feed and water were provided for 1 week as acclimatization. The general condition of rats was observed every day during the experiment The study was approved by the Experimental Animal Ethics Committee of Zunyi Medical University (approval no. ZMU11-2203-273).

Grouping and drug administration

Sprague-Dawley rats were allocated randomly into five groups as follows: Control (n=6); PAH (n=12); OA 40 and 80 mg/kg/day (both n=10) and the 100 mg/kg/day DCA (n=10). The rats were administered 55 mg/kg MCT (InvivoChem LLC) or an equal volume of normal saline as a control via intraperitoneal injection (16). The rats in OA40, OA80 and DCA groups were treated with 40 or 80 mg/kg OA (Jiangsu Yongjian Pharmaceutical Technology Co., Ltd.) (17,18) or DCA (100 mg/kg; InvivoChem LLC) by intragastric administration for 2 weeks, respectively. The control and PAH groups were administered an equivalent volume of normal saline via the same route. Observe the rats daily, record their weight.

Measurement of mPAP by right heart catheterization

Rats were anesthetized by intraperitoneal injection of pentobarbital sodium solution (50 mg/kg). The measurement of mPAP was performed using the central venous catheter technique, as previously reported (19). A central venous catheter (Secalon, 16 G/1.6×400.0 mm, Viggo products) was positioned into the right subclavian vein, extending through the superior vena cava, right atrium, right ventricle and into the pulmonary artery. The catheter was linked to a PowerLab physiological recording system (ADInstruments Pty Ltd.) and pressure transducer for the real-time display of pulmonary artery pressure. The recorded data were analyzed to determine the mPAP. After the pressure measurement, rats were anesthetized by intraperitoneal injection of 1% sodium pentobarbital (130 mg/kg). After anesthesia, the rats were sacrificed by cervical dislocation, and the heart and lung tissues of the rats were collected. The humane endpoints were as follows: Inability to eat or drink without anesthesia or sedation or stand for up to 24 h; poor condition including hypothermia with a body temperature <37°C in the absence of anesthesia/sedation and central nervous system depression, tremor, paralysis or pain that does not respond to analgesics. A total of two rats in the PAH and one each in the OA40 and OA80 group were euthanized in compliance with the humane endpoints.

Histopathological assessment

A total of five rats from each group were randomly chosen. The lower right lobe of the lung was excised and fixed in 10% neutral formaldehyde at room temperature for 24 h, followed by dehydration through a series of graded alcohol concentrations. The tissues were embedded in paraffin and sectioned at 4 µm thickness, followed by staining steps at room temperature. Hematoxylin and eosin (HE) staining was used to examine pulmonary artery remodeling. The tissue sections were immersed in hematoxylin staining solution for 7 min and rinsed with running water for 15 sec. Next, the sections were soaked in 1% hydrochloric ethanol, differentiated for 3 sec, and rinsed with running water for 15 sec. Finally, the tissue sections were immersed in eosin staining solution, stained for 3 min, and rinsed with running water for 15 sec. Masson's trichrome staining was applied to assess pulmonary fibrosis. The sections were immersed in Bouin solution (60°C, 30 min) and rinsed with running water for 15 sec. Next, the sections were soaked in azure blue, stained for 3 min, and rinsed with running water for a few seconds. After that, the tissue sections were immersed in Mayer hematoxylin, stained for 3 min, and rinsed with running water for 15 sec. The sections were immersed in 1% hydrochloric ethanol, differentiated for 3 sec, and rinsed with running water for 15 sec. The tissue sections were further immersed in Ponceau magenta solution, stained for 10 min, and rinsed with running water for 15 sec. The tissue sections were then immersed in phosphomolybdic acid solution for 10 min. Next, the tissue sections were directly immersed in aniline blue solution and stained for 8 min. The final treatment was carried out with a weak acid solution for 2 min. All tissue sections were subjected to gradient dehydration using alcohol at the end of staining and were sealed and preserved by dropping neutral gum. Digital photographs were captured by the light microscope (Olympus Corporation; cat. no. BX43; magnification, ×400). The cross-sectional dimensions and wall area (WA) of all small arteries were quantified using Image-Pro Plus software version 50.100 (Media Cybernetics, Inc.). Pulmonary vessel WA was calculated as a percentage of the vessel cross-sectional area as follows: WA=(vessel WA/vessel cross-sectional area) ×100. Fibrosis was measured using Image-Pro Plus 6.0 (Media Cybernetics, Inc.) and the percentage of pulmonary fibrosis (area ratio) was calculated as follows: Area ratio=(blue collagen fibrosis staining area/total test area) ×100.

Untargeted metabolomics assessment

A total of five rats were randomly selected from each group for untargeted metabolomics assessment. A sample of 25 mg rat lung tissue was combined with 500 µl extraction solution, consisting of a 2:2:1 ratio of acetonitrile, methanol and water, supplemented with isotopically-labelled internal standard mixture (Shanghai Zhenzhun Biotechnology Co., LTD, Merck Serono). Then, lung tissue samples were ground for 4 min at a frequency of 35 Hz and subjected to ultrasonic processing at a frequency of 40 kHz in ice-water (ultrasound 5 sec/interval 5 sec, 30 times, total time 5 min). The grinding and ultrasonic treatment were repeated three times. After incubation at −40°C for 1 h, samples were centrifuged at 13,800 × g at 4°C for 15 min. The supernatant was transferred into sample vials for analysis. Equal volumes of supernatant from all samples were combined to create a quality control (QC) sample for instrument testing.

LC-MS/MS analyses were performed using an UHPLC system (Vanquish; Thermo Fisher Scientific, Inc.) with a UPLC BEH Amide column (2.1×100 mm, 1.7 µm) coupled to Orbitrap Exploris 120 mass spectrometer (Orbitrap MS, Thermo). Chromatographic conditions: Waters ACQUITY UPLC BEH Amide (2.1×100 mm, 1.7 µm) column, mobile phase A:25 mmol/l ammonium acetate and 25 mmol/l ammonia water/1 l ultrapure water, mobile phase B: acetonitrile. Gradient elution (0–0.5 min, 95% B; 0.5–7 min, 95–65% B; 7–8 min, 65–40% B B; 8 to 9 min, 40% B; 9–9.1 min, 40–95% B; 9.1–12 min, 95% B), and column temperature 30°C, sample room temperature of 4°C, the flow rate of 0.5 ml/min, 2 µl sample quantity.

Orbitrap Exploris 120 mass spectrometer (room temperature, nebulizer pressure, 87 psi, flow rate 0.3 l/min), controlled by the acquisition software (Xcalibur 4.4; Thermo Fisher Scientific, Inc.), was used due to its ability of acquiring tandem mass spectrometry (MS) spectra in information-dependent acquisition mode. Under this operational setting, the software consistently evaluates the complete full scan MS spectrum. The electrospray ionization source parameters were as follows: Sheath gas flow rate, 50 arbitrary units (Arb); auxiliary gas flow rate, 15 Arb; capillary temperature, 320°C; full MS resolution, 60,000; MS/MS resolution, 15,000; collision energy in Normalized Collision Energy mode, 10–60 units and spray voltage, 3.8 or −3.4 kV for positive and negative ionization polarity, respectively.

The raw data was converted into mzXML format using the ProteoWizard 3.0 (http://proteowizard.sourceforge.net/) software suite and subjected to a custom processing pipeline (20) developed using R version 4.3.3 (R Core Team, R-project.org/) built upon the XCMS framework (21). This proprietary program used peak detection, extraction, alignment and integration. Subsequently, an internally constructed MS2 database was used for metabolite annotation purposes, with the threshold for annotation established at 0.3. Principal component analysis (PCA) and Orthogonal Projections to Latent Structures-Discriminant Analysis (OPLS-DA) were performed using SIMCA software (V16.0.2, Sartorius Stedim Data Analytics AB) Logarithmic (LOG) and centralized (CTR) format, then automatic modeling analysis. In order to verify the quality of the model, we used 7-fold cross validation to test. Then R2Y (interpretability of the model to categorical variables Y) and Q2 (predictability of the model) obtained after cross-validation were used to evaluate the effectiveness of the model. Finally, through permutation test, the permutation order of categorical variable Y is changed at random to obtain different random Q values, and the validity of the model is further tested. By mapping the differential metabolites to authoritative metabolite databases such as KEGG and PubChem (22) (kegg.jp/, we obtained matching information for the differential metabolites and then searched and analyzed metabolic pathways for the corresponding species Rattus norvegicus (rat). Regarding the hierarchical clustering analysis of differential metabolites, the Euclidean distance matrix of quantitative values of differential metabolites is calculated, and the differential metabolites are clustered using the complete linkage method.

Western blotting

A total of five rats were randomly selected from each group for western blotting to determine protein expression levels. Frozen lung tissue (−80°C) samples were homogenized and immersed in RIPA lysis buffer supplemented with 1 nM PMSF (both Beijing Solarbio Science & Technology Co., Ltd.). The samples were homogenized and centrifuged at 12,000 × g at 4°C for 10 min. The supernatant was harvested, total protein content was quantified using a BCA Protein Assay kit (Beijing Solarbio Science & Technology Co., Ltd.). According to the molecular weight of the protein, different concentrations of SDS-PAGE (8–10%) were used, and the protein loading volume was 30 µg. The separated proteins were transferred onto polyvinylidene fluoride membranes (0.45 µm; MilliporeSigma) and blocked with 5% skimmed milk solution (Beijing Solarbio Science & Technology Co., Ltd.) in 1X TBST buffer for 2 h at room temperature. The membranes were incubated overnight at 4°C with primary antibodies against Glut1 (1:1,000; Abcam; ab115730), HK2 (1:1,000; Abcam; ab209847), pyruvate kinase (PK) (1:1,000; Cell Signaling Technology; #3186), PDK1 (1:1,000; Abcam; ab110025), LDH (1:1,000; Proteintech; 19987-1-AP), isocitrate dehydrogenase 2 (IDH2; 1:1,000; Proteintech; 15932-1-AP), β-tubulin (1:1,000; Proteintech; 66031-1-Ig) and β-actin (1:10,000; Proteintech; 66009-1-lg). Subsequently, membranes were incubated with a horseradish peroxidase-conjugated secondary antibody (1:5,000; Proteintech; RGAR001) for 1 h at room temperature under gentle agitation. Protein bands were visualized using enhanced chemiluminescence substrate (Melunbio; MA0186) and protein expression was quantified using Quantity One software (version 1.4.6; Bio-Rad Laboratories, Inc.).

Statistical analysis

All data were evaluated using SPSS Statistics Professional software (version 29.0; IBM Corp.) and presented as mean ± SEM (n=5). The data was evaluated for normal distribution using Kolmogorov-Smirnov test. For data that follow a normal distribution with homogeneous variance, statistical analysis was performed using one-way ANOVA with Tukey's post hoc test. For data that did not conform to a normal distribution or had uneven variance, Welch ANOVA with Tamhane T2 test was used. P<0.05 was considered to indicate a statistically significant difference.

Results

OA reduces mPAP and inhibits pulmonary vascular remodeling and vascular fibrosis in PAH model rats

To assess the therapeutic effect of OA on PAH, OA was administered to MCT-induced PAH rats at doses of 40 and 80 mg/kg/d for 14 days (Fig. 1A). MCT-induced PAH model in rats is characterized by late-stage complications including heart failure, hepatic congestion and ascites, which culminate in a notable mortality rate (23). The mortality of these rats was associated with the toxic properties of MCT. MCT administration reduced the growth rate, while OA (40 mg/kg) increased the growth rate (Fig. 1C).

Figure 1. OA alleviated the development of MCT-induced PAH in rats. (A) Timeline of OA treatment. (B) mPAP quantitative analysis. (C) Growth rate of rats. (D) Representative hematoxylin and eosin staining of pulmonary artery (scale bar, 50 µm). (E) Quantitative analysis of vascular remodeling. (F) Representative pulmonary artery after Masson staining (scale bar, 50 µm). (G) Quantitative analysis of vascular fibrosis. *P<0.05 vs. control. #P<0.05 vs. PAH, OA, oroxylin A; PAH, pulmonary arterial hypertension; mPAP, pulmonary arterial pressure; MCT, monocrotaline; DCA, dichloroacetate; w, week; WA, wall area.

In the PAH model group, the mPAP value was 43.6 mmHg, which was significantly increased compared with the control group at 15.5 mmHg. Furthermore, the OA40 and OA80 groups had a significantly lower mPAP compared with the PAH group, with a reduction of 33.4 and 39.2%, respectively, demonstrating a dose-dependent response. Likewise, the DCA group also had a significant decrease of 19.9% in mPAP value compared with the PAH group (Fig. 1B). These results confirmed establishment of the PAH model and the efficacy of OA in decreasing mPAP.

Pathological changes in pulmonary arteries in the PAH group included notable narrowing of vessel lumens and intima-media thickening along with marked arterial remodeling (Fig. 1D). WA of the PAH group was significantly increased by 34.6% compared with the control. Moreover, OA40, OA80 and DCA groups had a significantly decreased WA compared with the PAH group (Fig. 1E). Masson's staining demonstrated pulmonary vascular fibrosis surrounded by a large number of collagen fibrosis in the PAH group (Fig. 1F). Area ratio was significantly higher in the PAH compared with Control group. OA at different doses significantly improved pulmonary arteriolar fibrosis and OA40 mg/kg had an improvement comparable to DCA (Fig. 1F and G).

OA alters the metabolite profile of lung tissue in PAH based on untargeted metabolomics analysis

PCA and OPLS-DA scoring plots of the metabolic profiles of lung tissues were calculated (Fig 2); PCA score plot shows that the PAH group is initially separated from the Control group, indicating that the PAH model induced by MCT was successfully established (Fig. 2A, B). The initial separation results of the lung tissue metabolome spectra in the OA40 and 80 groups and the DCA group are not obvious (Fig. 2C, E, G). Therefore, OPLS-DA model was established for greater differentiation between treatment groups. Screening of effective differential metabolites was performed. The predictive capacity of the OPLS-DA model was satisfactory, effectively distinguishing between the groups (Fig. 2B, D, F and H). These findings indicated that OA exerts an impact on the metabolic levels of PAH rats.

Figure 2. Metabolic analysis of OA-treated PAH lung samples by ultra performance liquid chromatography-mass spectrometry. Score plots of samples between control and PAH groups using (A) PCA and (B) OPLS-DA, PAH and OA80 using (C) PCA and (D) OPLS-DA models, PAH and OA40 using (E) PCA and (F) OPLS-DA models and PAH and DCA groups using (G) PCA and (H) OPLS-DA models. OA, oroxylin A; PAH, pulmonary arterial hypertension; PCA, principal component analysis; OPLS-DA, orthogonal partial least squares discriminant analysis; DCA, dichloroacetate.

Metabolic pathway analysis and association with pathway metabolites

To assess the metabolic pathways involved in PAH, metabolites exhibiting differences between the PAH and control groups were analyzed using R. This analysis focused on pathways in which these differentially expressed metabolites were involved, identifying enrichment in the ‘glycolysis or gluconeogenesis’ and ‘citrate cycle (TCA cycle)’ pathways (Fig. 3A). To assess the association between metabolites and the Warburg effect in PAH and OA-treated groups, a Euclidean distance matrix was created based on relative levels of metabolites within the pathway and data were normalized using Z-scores. The heatmap indicated that the majority of metabolites varied across the five groups (Fig. 3B). Further evaluation of these metabolites aimed to elucidate their association with PAH in the context of Warburg effect. Notably, eight types of metabolites demonstrated notable variation between the PAH and control groups (Fig. 3B). Specifically, levels of α-D-glucose, D-glucose 6-phosphate, phosphoenolpyruvic acid, pyruvic acid, cis-aconitic acid, isocitric acid and succinic acid were significantly lower in the PAH compared with the control group, whereas L-lactic acid levels were higher in the PAH compared with the control group. However, after administration of OA and DCA, it promoted the generation of metabolites in the TCA cycle and inhibited the generation of L-lactic acid. Preliminary evidence suggests that Warburg effect occurred in MCT-PAH model rats, and administration of OA can effectively inhibit the occurrence of Warburg effect and promote normal glucose metabolism in rats (Fig. 4).

Figure 3. Analysis of metabolic pathways and metabolites of PAH affected by OA. (A) Metabolic pathway analysis between the Control group and the PAH group. (B) Hierarchical clustering of differential metabolites involved in glucose metabolic pathways for each group. OA, oroxylin A; PAH, pulmonary arterial hypertension; DCA, Dichloroacetate.

Figure 4. Changes in metabolites in the Warburg effect of PAH treated with OA. (A) α-D-glucose. (B) D-glucose 6-phosphate. (C) Phosphoenolpyruvic, (D) pyruvic, (E) cis-aconitic, (F) isocitric, (G) succinic and (H) L-lactic acid. *P<0.05 vs. control. #P<0.05 vs. PAH. OA, oroxylin A; PAH, pulmonary arterial hypertension; DCA, dichloroacetate.

OA decreases the function of Warburg effect in PAH model rats

To assess the potential association between the protective action of OA against MCT-induced PAH and the Warburg effect, the protein expression of Glut1, HK2, PK, PDK1, LDH and IDH2 was evaluated. These proteins exhibited significant differences between the control and PAH groups (Fig. 5). Increased protein expression of Glut1 and HK2 in the PAH group was indicative of enhanced glucose uptake. Conversely, downregulation of PK and IDH2, accompanied by the upregulation of LDH and PDK1, indicates that the normal glucose metabolism pathway is inhibited, and more pyruvate reacts to produce lactate under the action of LDH, enhancing the Warburg effect. Compared with the PAH group, OA (40 mg/kg/d) or DCA treatment can inhibit the Warburg effect by altering enzymes related to glucose metabolism, promoting normal energy metabolism conversion of glucose. These findings indicate that the protective effect of OA in the MCT induced PAH model is related to its ability to regulate the Warburg effect.

Figure 5. Expression of glycolysis-associated proteins in lung tissue from PAH model rats treated with OA. Representative western blotting and quantitative analysis of (A) Glut1, (B) HK2, (C) PK, (D) LDH, (E) PDK1 and (F) IDH2. *P<0.05 vs. control. #P<0.05 vs. PAH. OA, Oroxylin A; PAH, pulmonary arterial hypertension; DCA, dichloroacetate; Glut1, glucose transporter 1; HK2, Hexokinase 2; PK. Pyruvate kinase; LDH, Lactate dehydrogenase; PDK1, Pyruvate dehydrogenase kinase 1; IDH2, Isocitrate dehydrogenase 2.

Discussion

To the best of our knowledge, the present study is the first to assess the protective properties of OA against PAH and its underlying mechanisms. OA exerted a protective influence on mPAP and attenuated pulmonary vascular remodeling in the MCT-induced PAH rat model. The mechanism of action of OA may be due to the blockade of Warburg effect, suggesting that OA has potential application prospects in treatment of PAH.

OA is a naturally derived flavonoid with potential therapeutic applications in inhibiting abnormal glycolysis, angiogenesis, invasion, metastasis, and anti-tumor effects (24,25). In PAH, PASMC, as an effector cell of pulmonary vasoconstriction, exhibits many characteristics of cancer cells under the influence of factors such as inflammatory factors, growth factors, and vasoactive substances. Its proliferation and synthesis abilities are enhanced, and apoptosis is hindered, accompanied by upregulation of oncogene expression such as p53 and c-Myc or expression of cancer markers. Over proliferating PASMC is the main cellular component of pulmonary artery remodeling (26). Based on the inhibitory effect of OA on cancer cell proliferation, it was hypothesized that OA may have protective effects in PAH (27). To assess this, a MCT-induced PAH rat model was used to evaluate the effects of OA. Previous studies have reported that in rats, subcutaneous injection of 60 mg/kg MCT for 5 weeks results in a mortality rate of up to 35%, with the right ventricular systolic pressure reaching 80 mmHg, leading to severe pulmonary hypertension (16,28). In the study of pulmonary hypertension model, the dose of MCT usually 50~60 mg/kg. Considering the efficacy and toxicity of MCT, we select 55 mg/kg dose in the model. In the present 55 mg/kg MCT-induced PAH rat model, pulmonary artery remodeling was accompanied by right ventricular hypertrophy. It was hypothesized that pulmonary artery remodeling was the primary cause of progressive mPAP increase. OA effectively reduced the MCT-induced mPAP increase, pulmonary artery wall thickening and the progression of pulmonary fibrosis. These changes suggested that OA can serve a protective role in PAH.

To the best of our knowledge there are no studies using OA in the treatment of PAH. The OA doses of 40 and 80 mg/kg referred to the dose of OA in numerous types of cancer (29,30). However, further research is needed to determine whether 40 mg/kg is close to the minimum effective dose. Similarly, the role of OA in reducing mPAP through concentration gradient remains to be further studied. But we still found for the first time that OA has a therapeutic effect on PAH.

Metabolomics has been used in examination of metabolic perturbations associated with PAH (31). The discovery of metabolic changes in PAH may provide new avenues for its treatment. Some literature suggests that several main metabolic pathways are related to PAH, including glucose and fatty acid oxidation, glutamine breakdown, arginine metabolism, one carbon metabolism, TCA, electron transfer chain, calcium homeostasis, and glycine metabolism (32,33). However, further research is required to assess how OA modulates these metabolic changes in PAH. The present study used DCA as a positive control and a metabolomics approach to study the impact of OA on metabolic profiling in PAH model rats. Untargeted metabolomics strategy was used to identify and screen differentially expressed metabolites. It was demonstrated that most of these metabolites belonged to the ‘citrate cycle (TCA cycle)’ and ‘glycolysis or gluconeogenesis’ pathways. Glucose metabolism provides energy for cell proliferation. The phenomenon of increased glycolysis promoting lactate production in abnormal glucose metabolism is known as the Warburg effect, which can also be observed in patients with cancer patients. Its characteristic is that under normal oxygen supply conditions, the main form of energy generation in tumor cells ranges from mitochondrial oxidative phosphorylation to lower efficiency aerobic glycolysis (34,35).

In PAH, the Warburg effect of glucose metabolism transformation can promote the proliferation of PASMC, and HIF-1 α activation of glycolytic genes is considered key to metabolic adaptation to hypoxia, by increasing the conversion of glucose to pyruvate and subsequently to lactate (36,37). DCA is a potent metabolic drug for the treatment of PAH that acts by inhibiting PDK1. DCA promotes glycolysis by inhibiting phosphorylation of pyruvate dehydrogenase, which converts pyruvate to acetyl-CoA into the TCA cycle (38). The present study used metabolomics methods to accurately reveal whether glycolytic metabolic pathways contribute to the protective effect of OA on PAH. The data shows that the relative content of pyruvate and some TCA cycle products in PAH rats decreases, while the relative content of lactate increases. The OA and DCA treatment groups can change this phenomenon. Western blot was used to detect expression level of glucose metabolism related proteins. The results showed significant changes in the expression levels of glucose transporter Glut1, enzymes in the glycolytic pathway (HK2, PK, LDH), and enzymes in TCA (PDK1, IDH2). In the PAH model, expression of Glut1, HK2, PDK1, and LDH was upregulated, indicating enhanced glycolysis in animal models; Meanwhile, downregulation of PK and IDH2 expression indicates inhibition of normal glucose metabolism pathways, with more pyruvate reacting with LDH to generate lactate, enhancing the Warburg effect. OA administration can significantly alter this change, indicating that abnormal glucose metabolism in PAH is related to the Warburg effect. Inhibiting the Warburg effect is key to reversing pulmonary artery remodeling.

In summary, OA may decrease PAH by modulating the Warburg effect. However, the specific mechanism by which OA ameliorates this is not known. Therefore, future studies should assess therapeutic targets for OA to treat PAH by targeting the Warburg effect.

In conclusion, OA has a protective effect on PAH and can disrupt endogenous metabolic disorders in PAH rats by regulating the Warburg effect. The results of this study contribute to understanding the mechanism by which OA reduces PAH and provide a new approach for further developing drugs for treating PAH.

Acknowledgements

Not applicable.

Funding

The present study was supported by National Natural Science Foundation of China (grant nos. 82460780, 82260780 and U1812403), Guizhou Provincial Science and Technology Plan Project [Qiankehe Foundation-ZK (2023) General 500] and Zunyi Medical University Postgraduate Research Fund Projects (grant no. ZYK196).

Availability of data and materials

The data generated in the present study are included in the figures and/or tables of this article. The data generated in the present study may be found in the CNGB Sequence Archive (CNSA) of China National GeneBank DataBase (CNGBdb) under the accession number (CNP0006089) using the following URL: https://db.cngb.org/.

Authors' contribution

YW, YF and YZ performed experiments and wrote the article. TC and JL analyzed and interpreted data. SX and LL conceived and designed the study and revised the manuscript. All authors have read and approved the final manuscript. YW and LL confirm the authenticity of all the raw data.

Ethics approval and consent to participate

The present study was conducted in accordance with the Ethical Guidelines for the Welfare of Laboratory Animals of the Zunyi Medical University (approval no. ZMU21-2203-622).

Patient consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

References