Introduction

Pulmonary hypertension (PH) is a severe and incurable disease that seriously affects the quality of life and life expectancy of patients [1]. Although targeted drugs have been developed in recent years, they still have limitations with regard to efficacy and safety [2, 3]. Therefore, further exploration of novel medications for treating PH to improve treatment effectiveness and alleviate patient burden is urgently needed.

The primary pathological alterations in PH involve the impairment and destruction of pulmonary artery endothelial cells (PAECs), abnormal growth of pulmonary artery smooth muscle cells (PASMCs), and the accumulation of extravascular collagen [4, 5]. These pathological processes are also accompanied by the oxidative stress response of pulmonary artery cells [6]. Oxidative stress is primarily characterized by an abundance of reactive oxygen species (ROS) within cells. ROS are also critical signaling molecules for vascular cell proliferation [7, 8]. Vitry et al. showed that PH-PASMCs could exert hyperproliferative and antiapoptotic effects by hijacking persistent oxidative stress [9]. Hennigs et al. reported a PPARγ- and P53-mediated mechanism for vascular protection in PH-PAECs in response to DNA damage and oxidative stress [10]. Yeligar et al. reported that depletion of PPARγ in human PASMCs increased mitochondria-derived ROS production and disrupted mitochondrial bioenergetics [11]. These results suggest that PPARγ-driven inhibition of oxidative stress is critical for vascular remodeling in PH. Consequently, finding an effective drug targeting these pathological hallmarks is of interest for the management of PH.

Canagliflozin (CANA) is an effective hypoglycaemic drug that is approved in the USA for type II diabetes treatment. Nevertheless, in recent years, an increasing number of studies have revealed its favorable protective effects on the cardiovascular system. For example, numerous studies have indicated that CANA might impact vasorelaxation in arterioles from human visceral adipose tissue [12], hinder the proliferation of vascular smooth muscle cells in blood vessels and decrease the progression of atherosclerosis [13, 14]. Furthermore, a study revealed that CANA can substantially reduce the occurrence of major cardiovascular incidents in individuals with diabetes and chronic kidney disease [15] while also notably diminishing the cardiovascular hazard for patients suffering from heart failure [16]. The protective effect of CANA on the cardiovascular system may be attributed to its ability to inhibit cell proliferation [17], fibrosis [18], and oxidative stress [19]. However, the therapeutic effect and underlying mechanisms of CANA on PH remain unclear, and a better understanding of these effects could provide us with novel insights into PH treatment.

Here, we combined network pharmacology, transcriptomics, and biological methods to evaluate the promising protective effect of CANA against PH. The pharmacological protective effect of CANA on PH was investigated in three common PH rodent models and a hypoxia-induced PH cell model in this study. Network pharmacology and transcriptomics methods revealed that PPARγ was the target and signaling molecule for the anti-PH effect of CANA, which suppressed oxidative stress and excessive proliferation of PASMCs. Subsequent validation experiments were conducted in vivo and in vitro to assess the anti-PH effects of CANA through PPARγ. Moreover, we showed that CANA targeted PPARγ and inhibited PPARγ S225 phosphorylation, thus exerting its antiproliferative and antioxidative effects on stress. Notably, this is the first time that CANA has been investigated and validated in the field of PH via multiple bioinformatics and biological methods. Figure 1 displays a schematic illustration of the investigation.

Fig. 1: Flow chart of the study.
figure 1

The animation graph condensed the study method and proposed mechanism demonstrated in this work. Here, we integrate network pharmacology, transcriptomics, molecular docking, and experimental verification to uncover the role and mechanism of CANA on PH. 5 animal models are used as shown in the image. We report that CANA inhibits PPARγ Ser225 phosphorylation, promoting nuclear translocation of PPARγ, and enhancing oxidative stress-related gene translation, thus inhibiting ros accumulation, proliferation, and migration of rPASMCs.

Full size image

Materials and methods

Animal models

This study involved the creation of various animal models, including a mouse model of hypoxia-induced (Hx) PH, mouse and rat models of severe PH induced by Sugen 5416-hypoxia (Su/Hx), a model for rescuing established PH and a rat model of PH induced by monocrotaline (MCT) [4, 5, 20]. All mice and rats were purchased from Weitong Lihua Limited Company (Bei**g, China). Before the experiments, all animals were allowed to acclimate to the pathogen-free animal habitat for a minimum of 1 week. Additionally, an ample supply of food (consisting of 24% protein, 15% fat, and 61% carbohydrates, sourced from Jiangsu ** rat brain: effect of metal mixture and troglitazone in astrocytes. Cell Death Dis. 2014;5:e1033." href="/article/10.1038/s41401-024-01286-9#ref-CR56" id="ref-link-section-d82699633e2139">56]. PPARγ activation also suppressed rPASMC proliferation and promoted eNOS expression, increasing the release of NO from endothelial cells [57, 58]. The results of this study indicate that CANA significantly increases the expression of PPARγ both in vivo and in vitro. Based on the evidence above, we presumed that PPARγ plays a key role in the anti-PH effects of CANA. The other 4 core genes (eNOS, iNOS, ACE, MMP9) were not initially considered for the following reasons: (1) our transcriptome data indicated significant enrichment of the PPAR signaling pathway after CANA treatment, highlighting the importance of PPARγ in the anti-PH effect of CANA; (2) PPARγ showed a strong association with oxidative stress, which was also significantly enriched in the transcriptome and network pharmacology analysis; and (3) our previous studies emphasized the crucial role of PPARγ in PH pathology in both rPASMCs and endothelial cells, while eNOS, iNOS, and ACE were primarily associated with endothelial cells [45, 46, 59], and MMP9 was related to fibroblasts [60, 61]. Since PAMSCs are key cells involved in vascular remodeling, our study focused specifically on the role of rPASMCs in the pathogenesis of PH. However, further research is needed to determine whether CANA affects lung endothelial cells through PPARγ.

CANA, dapagliflozin, and empagliflozin are classified as inhibitors of sodium-glucose cotransporter 2 (SGLT2), where SGLT2 accounts for the majority of renal glucose reabsorption in the kidneys [62]. Various studies have shown that CANA has a weaker inhibitory effect on SGLT2 than dapagliflozin and empagliflozin, suggesting that CANA might exhibit more specific and potent inhibition of other proteins [63, 64]. Similar to scholarly research indicating the absence of SGLT2 in any blood vessel other than the kidney, our investigation using the THPA database revealed a low level of SGLT2 protein in rPASMCs. Hence, we did not examine the effect of CANA on SGLT2 in our models. Our findings confirmed that CANA can directly bind to PPARγ, thereby hindering its phosphorylation. This discovery introduces a novel perspective on CANA. Notably, in addition to SGLT2 inhibitors regulating blood sugar levels, SGLT2 inhibitors such as CANA are being increasingly used due to their supplementary advantages for cardiovascular health [65, 66].

In PH associated with heart failure with preserved ejection fraction (HFpEF), Taijyu et al. [67] reported that empagliflozin decreased mitochondrial ROS and increased pulmonary arterial remodeling while exercising. ZSF-1 obese rats were used as a rodent model for HFpEF and PH, either with or without Sugen treatment. These models exhibit numerous characteristics similar to those of type II diabetes in humans [68]. According to the authors, in this particular situation, empagliflozin ameliorated metabolic syndrome, thereby inhibiting the initial development of mitochondrial ROS [67]. We utilized hypoxia-related pulmonary hypertension (PH) models from the third category, which lacked a distinct metabolic profile (such as diabetes) similar to that of obese ZSF-1 rats. Hence, these metabolic features were not evaluated in this study. Similar to the role of empagliflozin in HFpEF, CANA also ameliorated oxidative stress and pulmonary artery remodeling in hypoxia-related PH in our study. However, we highlight the role of PPARγ in the anti-PH effects of CANA. Indeed, PPARγ and the PPAR coactivator PGC1-α are thought to be closely related to mitochondrial oxidative stress. In addition, there is no direct evidence that CANA modulates oxidative stress via SLGT2 inhibition.

In short, different SGLT2 inhibitors might have different effects on different clinical types of PH. Hess et al. [69] found that empagliflozin attenuated MCT-induced PH, but its molecular mechanism of action was nonspecific. Wu et al. [70] showed that dapagliflozin has no protective effect on PH models. Recently, Tang et al. [71] showed that CANA had a protective effect on hypoxia-induced PH, which is consistent with our conclusions. However, our exploration of the role and mechanism of CANA in PH is undoubtedly multiangled, comprehensive, and in depth.

PPARγ activation has been found to prevent PH progression. According to our previous investigation, rosiglitazone, a conventional activator of PPARγ, can effectively alleviate pulmonary vascular remodeling and right ventricular remodeling in a mouse model of hypoxia-induced PH [28]. Legchenko et al. reported that the PPARγ agonist pioglitazone reversed PH and prevented right heart failure [52]. Pioglitazone was also found to prevent the development of MCT-induced PH in rats [72]. However, several clinical trials with these PPARγ agonists, such as rosiglitazone, have demonstrated that they may increase heart failure. This finding can be explained by their side effects, with a focus on mechanisms such as edema and sodium retention. The clinical use of full-spectrum PPARγ agonists, such as rosiglitazone, is limited due to the cardiovascular risks associated with their long-term use. In the treatment of diabetes, one strategy is the development of selective PPARγ modulators, SPPARMs, to preserve the expression level of PPARγ in vivo and regulate the transcription of therapeutic-related target genes while avoiding the regulation of target genes linked to adverse reactions [73]. Our data provide some support for the use of CANA as a selective PPARγ modulator, which holds promise for the treatment of PH. However, more experiments are needed to support this hypothesis.

Protein functions can be regulated by post-translational modifications, most commonly phosphorylation [74]. Depending on the cell type and stimulation, the site of phosphorylation of PPARγ and its biological effects are completely different [75, 76]. The phosphorylation sites that have been identified in PPARγ include Ser112 and Ser273. PPARγ Ser112 phosphorylation typically leads to reduced PPARγ activity [77], while PPARγ phosphorylation at Ser273 occurs mainly in adipose tissue [78]. Phosphorylation of PPARγ Ser225 (S225) has not been reported. Our results suggested that PPARγ S225 expression was altered in the pathogenesis of PH, which inhibited PPARγ transfer into the nucleus and thus reduced PPARγ transcriptional activity. However, CANA directly interacts with PPARγ, inhibiting PPARγ S225 phosphorylation and promoting PPARγ activity, resulting in excellent antioxidant stress effects. However, a larger in vivo dataset would result in further insight into this issue.

The findings of this research suggest that CANA has potential for use as a treatment for PH. This drug improved the progression of PH by reducing damage caused by oxidative stress and inhibiting the proliferation of smooth muscle cells in the pulmonary artery. This effect is partially achieved by activating PPARγ. Mechanistically, CANA inhibits PPARγ S225 phosphorylation and promotes PPARγ function. Taken together, the results of this study may help to elucidate the molecular process underlying the emergence and progression of PH while also revealing a novel avenue for the practical utilization of CANA.