Endoplasmic Reticulum Stress‐Induced Endothelial Cell Pyroptosis Contributes to Pulmonary Vascular Remodeling and Pulmonary Arterial Hypertension via IRE1α/Caspase‐3/GSDME Pathway

Abstract

Background

Endothelial cell (EC) injury is regarded as the initiating trigger of pulmonary arterial hypertension (PAH). Excessive endoplasmic reticulum (ER) stress could cause early damage to ECs with subsequent cell death. Pyroptosis leads to EC damage and accelerates PAH progression. However, whether and how ER stress plays a role in regulating EC pyroptosis, especially in PAH progression, remains unclear.

Methods

The activation level of ER stress and endothelial pyroptosis were assessed in the lungs of a PAH model. Pharmacological inhibitors, small‐interfering RNA, and specific inhibitors were used to explore the role and the mechanism of ER stress in regulating EC pyroptosis in PAH in vivo and in vitro, respectively.

Results

ER stress and endothelial pyroptosis were activated in the early stage of monocrotaline‐induced PAH rats. Inhibition of ER stress suppressed the activation of the endothelial GSDME (gasdermin E) in PAH rats. Prolonged and severe ER stress increased the level of the GSDME‐NT (N‐terminal of gasdermin E) and LDH (lactic dehydrogenase) release in ECs. Silencing GSDME or caspase‐3 reversed the effect of ER stress‐induced EC pyroptosis. Mechanistically, the IRE1α (inositol‐requiring kinase 1α) kinase activity mediated the activation of ER stress‐triggered caspase‐3/GSDME. Inhibition of the IRE1α kinase activity by KIRA6 (IRE1α kinase inhibitor) treatment alleviated the development of PAH by inhibiting caspase‐3/GSDME‐mediated endothelial pyroptosis and subsequent endothelial integrity disruption.

Conclusions

These results demonstrated the critical role of prolonged and unresolved ER stress‐induced IRE1α activation in modulating EC pyroptosis, leading to early endothelial cell injury and the acceleration of PAH progression.

Nonstandard Abbreviations and Acronyms

4‐PBA

4‐phenylbutyric acid

ATF4

activating transcription factor 4

ATF6α

activating transcription factor 6α

BMPR2

bone morphogenetic protein type II receptor

CHOP

C/EBP homologous protein

EC

endothelial cell

eIF2α

α‐subunit of eukaryotic translational initiation factor 2

ER

endoplasmic reticulum

ERN1

endoplasmic reticulum‐nuclei‐1

GRP78

glucose regulatory protein 78

GSDMD

gasdermin D

GSDMD‐NT

gasdermin D N‐terminal

GSDME

gasdermin E

GSDME‐NT

gasdermin E N‐terminal

HSP60

heat shock protein 60

IRE1α

inositol‐requiring kinase 1α

JNK

JUN N‐terminal kinase

NLRP3

nucleotide‐binding oligomerization domain, leucine‐rich repeat and pyrin domain‐containing protein 3

p‐eIF2α^S51

phospho‐α‐subunit of eukaryotic translational initiation factor 2

p‐IRE1α^S724

phosphorylation of IRE1α at S724

PAEC

pulmonary artery endothelial cell

PAH

pulmonary arterial hypertension

PASMC

pulmonary arterial smooth muscle cell

PERK

protein kinase R‐like endoplasmic reticulum kinase

RVSP

right ventricular systolic pressure

TRAF2

tumor necrosis factor receptor‐associated factor 2

UPR

unfolded protein response

XBP1s

spliced X‐box binding protein 1

Research Perspective.

What Is New?

• Prolonged and excessive endoplasmic reticulum stress triggers GSDME (gasdermin E)‐mediated pyroptosis in endothelial cells, thereby accelerating the progression of pulmonary arterial hypertension.

• The kinase activity of IRE1α (inositol‐requiring enzyme 1α) induces caspase‐3/GSDME‐dependent pyroptosis in the pulmonary endothelium, thereby promoting pulmonary arterial hypertension.

What Question Should Be Addressed Next?

• Further studies using targeted approaches are required to delineate the specific regulation of caspase‐3/GSDME by IRE1α kinase activity in pulmonary arterial hypertension and to validate its clinical relevance, which encompasses determining the level of phosphorylation of IRE1α at S724 in patients with pulmonary arterial hypertension and exploring the feasibility of therapeutic intervention.

Pulmonary arterial hypertension (PAH) is a devastating vascular disease, characterized by progressive elevated pulmonary vascular resistance and pulmonary arterial pressure. ¹ Despite advanced therapies, there is no cure for PAH, with a 3‐year mortality of 21%. ² One of the reasons is that the current pharmacological therapies mainly focus on the imbalance between vasoconstriction and vasodilation in the late stage of vascular remodeling instead of the initiating event. ³ , ⁴ It is currently believed that endothelial cell (EC) injury serves as a key trigger in the pathogenesis of PAH. ³ , ⁵ , ⁶ , ⁷ Studies have shown that the early stage of PAH development involves EC injury and apoptosis, whereas apoptosis‐resistant ECs emerge later as the disease progresses. ⁸ , ⁹ , ¹⁰ In a transgenic mouse model of Fas‐induced EC apoptosis, PAH and pulmonary arteriopathy were observed, providing direct evidence that lung EC damage acts as a trigger for PAH. ¹¹ Genetically, the loss of BMPR2 (bone morphogenetic protein type II receptor) signaling in the endothelium has been implicated as an initiating factor in PAH. Furthermore, selective enhancement of endothelial BMPR2 signaling with its ligand, BMP receptor ligand 9, decreased the sensitivity of the endothelium to apoptotic stimuli in PAH models and thus ameliorated disease phenotypes. ¹² Thus, it is urgently necessary to elucidate the mechanism underlying EC damage during PAH progression. Evidence has shown that unresolved endoplasmic reticulum (ER) stress could cause early damage to ECs, ultimately leading to cell death. ¹³ , ¹⁴ , ¹⁵ Microvascular endothelial cells in the presence of glucosamine initiates an amplified unfolded protein response (UPR). The amplified and sustained UPR resulted in a proinflammatory, prothrombotic, and proapoptotic phenotype that recapitulates the major endothelial dysfunctions associated in vivo with cardiovascular diseases, which can be hampered by the inhibition of ER stress. ¹⁶ Prolonged endoplasmic reticulum stress resulted in CHOP (C/EBP homologous protein)‐dependent cardiomyocyte apoptosis, leading to hypertrophic and failing heart after aortic constriction. Although, depletion of CHOP attenuated cardiac hypertrophy and dysfunction. ¹⁷ Recent studies suggested that ER stress promoted pyroptosis in various cell types such as hepatocytes ¹⁸ and trophoblasts. ¹⁹ However, it remains unclear whether ER stress plays a role in endothelial pyroptosis, and how varying degrees and durations of ER stress influence this process, especially in the progression of PAH.

Pyroptosis, a form of programmed and inflammatory cell death, is mediated by gasdermin family proteins. The pyroptotic cell undergoes membrane blebbing before the plasma membrane ruptures and then releases intracellular contents including LDH (lactic dehydrogenase). ²⁰ This rapid mode of cell death, typically occurs in the early stages of a disease, enabling the body to respond swiftly to external stress. ²¹ , ²² This process is closely associated with endothelial injury, with both being pivotal early events contributing to the pathogenesis of vascular disease. The pyroptosis of vascular endothelial cells has been reported to play a significant role in the occurrence and progression of various cardiovascular diseases. For instance, in lipopolysaccharide‐induced blood–brain barrier disruption, brain ECs, as the most prominent and probably the first‐line responders, undergo pyroptosis. Activated GSDMD (gasdermin D) directly targets and permeabilizes the plasma membrane of brain ECs, finally leading to blood–brain barrier leakage. ²³ Furthermore, endothelial pyroptosis can be induced by the high mobility group box 1/advanced glycation end‐products signaling pathway, which is well recognized for its role in the early acute phase of Kawasaki disease. This pathway promotes the release of IL‐1β (interleukin‐1β) and IL‐18 (interleukin‐18), thereby exacerbating vascular inflammation. ²⁴ Correspondingly, in the early stages of diabetic mice, the expression of pyroptosis‐related molecules (eg, NLRP3 [nucleotide‐binding oligomerization domain, leucine‐rich repeat and pyrin domain‐containing protein 3], IL‐18, IL‐1β) is upregulated. Inhibition of NLRP3 has been shown to suppress endothelial pyroptosis and vascular inflammation, offering protection against early diabetic retinopathy. ²⁵ The role of pyroptosis in the pathogenesis of pulmonary hypertension, particularly within endothelial cells, remains poorly understood. Recent studies have demonstrated that hypoxia‐induced GSDMD pyroptosis in pulmonary artery smooth muscle cells contributes to the development of PAH. Endothelial cell injury is recognized as an early event in PAH. Our previous studies have shown that caspase‐3/GSDME (gasdermin E)‐mediated pyroptosis in pulmonary artery endothelial cells (PAECs) compromises endothelial integrity, thereby promoting PAH. ²⁶ Furthermore, in vivo interventions targeting GSDME using lentivirus have shown potential in alleviating PAH. ²⁷ However, the upstream regulatory mechanisms governing endothelial pyroptosis in PAH and the time course of endothelial pyroptosis during the progression of PAH remain to be fully elucidated. In this study, we demonstrated that endothelial pyroptosis occurred in the early stages of pulmonary hypertension. Sustained and unresolved ER stress initiated endothelial cell pyroptosis and contributed to PAH pathogenesis.

The latest study suggested that IRE1α (inositol‐requiring enzyme 1α) exacerbated herpes simplex virus type 2 infection‐induced neuron pyroptosis. ²⁸ IRE1α is considered the most conserved branch among the 3 pathways of the UPR, ²⁹ serving as a crucial link between ER stress and cell fate. The endoribonuclease activity of IRE1α typically promotes cell survival, whereas the kinase activity leads to cell death. Results show that pharmacological inhibition of IRE1α/XBP1s (spliced X‐box binding protein 1) pathway attenuated the development of PAH by inhibiting pulmonary arterial smooth muscle cells (PASMCs) proliferation. ³⁰ However, whether and how IRE1α mediates endothelial cell pyroptosis in PAH and what the role is of IRE1α kinase activity in PAH remain unknown. In the present study, we demonstrated that ER stress promoted PAECs pyroptosis in the PAH model. Suppression of ER stress alleviated EC pyroptosis and PAH progression. Mechanistically, the IRE1α/caspase‐3/GSDME pathway mediated ER stress‐induced PAEC pyroptosis. Inhibiting IRE1α kinase activity suppressed the caspase‐3/GSDME activation, thus ameliorating endothelial cell injury and barrier dysfunction, consequently attenuating pulmonary vascular remodeling in PAH.

METHODS

Data Availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

For expanded methods, see Data S1.

Inclusion and Ethics

Preclinical PAH experiments in rats and mice were performed according to the guidelines of the National Institutes of Health and approved by the Animal Care and Use Committee at the Shanghai Jiao Tong University School of Medicine.

Experimental Animals

Male Sprague–Dawley rats weighing from 200 to 250 g received a single intraperitoneal injection of monocrotaline (60 mg/kg) to establish a PAH rat model. For 4‐phenylbutyric acid (4‐PBA) or KIRA6 treatment, rats that received monocrotaline were treated with 4‐PBA (500 mg/kg) through oral gavage daily ³¹ or the IRE1α kinase activity inhibitor KIRA6 (7 mg/kg) via intraperitoneal injection daily. ³² Animals were anesthetized and euthanized 21 days after monocrotaline injections for hemodynamic assessment and tissue harvesting. Eight‐week‐old male C57BL/6 mice were administered 20 mg/kg Sugen5416 by subcutaneous injection once weekly for 3 weeks ²⁶ and were sealed in a hypoxic chamber containing 10% O₂ for 3 weeks to develop Sugen5416 combined with the chronic hypoxia mouse model.

Statistical Analysis

All results were presented as mean±SEM of independent experiments. Each replicate represents an independent experiment. Comparisons between the 2 groups were analyzed by an unpaired, 2‐tailed Student t test. Results were analyzed by 1‐way ANOVA among ≥3 groups. Permutation tests were applied to assess significance in small sample sizes (n<6). For comparisons involving ≥3 groups, pairwise tests with Bonferroni correction were conducted if the initial permutation test reached significance. Statistical analysis was performed with the GraphPad Prism software version 10.0, and P<0.05 was considered to be statistically significant.

RESULTS

ER stress‐mediated pulmonary vascular endothelial cell pyroptosis promoted the pulmonary vascular remodeling and PAH, which was reversed by an ER stress inhibitor.

We first confirmed the activation of ER stress and pyroptosis in our PAH model. Following 21 days of monocrotaline injection, the protein levels of both GRP78 (glucose regulatory protein 78), a stress‐inducible ER chaperone and a marker of ER stress activation, ³³ and components of the ER stress induced‐UPR pathways were upregulated in the lung (Figure 1A). Pyroptosis executors were induced in the PAH model as well (Figure 1B). To clarify whether pyroptosis occurred in the endothelium of the PAH model, immunofluorescence analysis was performed. The results revealed that endothelial cell marker CD31 (platelet endothelial cell adhesion molecule‐1) was colocalized with GSDME, which was significantly upregulated following monocrotaline treatment (Figure 1C and 1D). These results indicated that pyroptosis occurred in the endothelium of pulmonary arteries in the PAH model. In addition, GSDMD was primarily overlapped with α‐SMA (α‑smooth muscle actin) (Figure 1E and 1F). To determine whether ER stress mediates endothelial cell pyroptosis in the PAH model, we used 4‐PBA, a recognized ER stress inhibitor, to treat PAH rats. ³¹ A dosage of 500 mg/kg per day of 4‐PBA was administered by oral gavage to PAH rats. Rats treated with monocrotaline showed increased right ventricular (RV) systolic pressure (RVSP) and RV hypertrophy (RV/(LV+S), right ventricle/(left ventricle+septum)), which were attenuated by 4‐PBA treatment (Figure 2A and 2B). The 4‐PBA administration also effectively reduced pulmonary vascular remodeling, as evidenced by reduced medial wall thickness in both small (<50 μm) and large (>50 μm) vessels in PAH rats (Figure 2C and 2D). These results were consistent with previous studies. ³¹ Notably, compared with the monocrotaline group, 4‐PBA treatment lowered the cleavage of GSDMD, GSDME, and caspase‐3 (Figure 2E). Furthermore, treatment with 4‐PBA led to a decrease in the GSDME protein level in the pulmonary vascular endothelium (Figure 2F). These data suggested that pulmonary endothelial pyroptosis was attenuated by the ER stress inhibitor.

Figure 1. Activation of ER stress and pyroptosis in a PAH model.

A, Representative immunoblots and the quantified data of ER stress‐related molecules, including GRP78, p‐IRE1α^S724, IRE1α, p‐JNK, JNK, XBP1s, p‐eIF2α^S51, eIF2α, ATF4, and ATF6α‐N in lung tissue of MCT‐PAH rats at day 21 post‐MCT injection (n=5–8). B, Representative immunoblots and the quantified data of GSDME‐NT and GSDMD‐NT in lung tissue of MCT‐PAH rats (n=5–8). C and D, GSDME, CD31, and α‐SMA immunofluorescence staining in control and MCT‐PAH rats. Nuclei were stained with DAPI. Representative fluorescent‐stained sections are shown (n=6). E and F, GSDMD, CD31, and α‐SMA immunofluorescence staining in control and MCT‐PAH rats. Nuclei were stained with DAPI. Representative fluorescent‐stained sections are shown (n=6). Statistical analysis was performed using an unpaired t test, unpaired t test with Welch's correction, or permutation test. Scale bar=50 μm. α‐SMA indicates α‑smooth muscle actin; ATF4, activating transcription factor 4; ATF6α‐N, cleaved nuclear variants of activating transcription factor 6α; CD31, platelet endothelial cell adhesion molecule‐1; Con, control; eIF2α, α‐subunit of eukaryotic translational initiation factor 2; DAPI, 4',6‐diamidino‐2‐phenylindole; ER, endoplasmic reticulum; GRP78, glucose regulatory protein 78; GSDMD, gasdermin D; GSDMD‐FL, gasdermin D full length; GSDMD‐NT, gasdermin D N‐terminal; GSDME, gasdermin E; GSDME‐FL, gasdermin E full length; GSDME‐NT, gasdermin E N‐terminal; IRE1α, inositol‐requiring kinase 1α; JNK, JUN N‐terminal kinase; MCT, monocrotaline; PAH, pulmonary arterial hypertension; p‐eIF2α^S51, phospho‐α‐subunit of eukaryotic translational initiation factor 2; p‐IRE1α^S724, phosphorylation of IRE1α at S724; p‐JNK, phospho‐JUN N‐terminal kinase; and XBP1s, spliced X‐box binding protein 1.

Figure 2. Inhibition of ER stress attenuated the activation of GSDMD and GSDME in a PAH model.

A through F, MCT‐PAH rats treated with or without 4‐PBA (500 mg/kg per day) (n=6). RVSP (A) and the ratio of RV/(LV+S) (B) in vehicle‐ or 4‐PBA‐treated MCT‐PAH rats. C, Representative images of hematoxylin and eosin staining of lung tissue in vehicle‐ or 4‐PBA‐treated MCT‐PAH rats. D, Ratio of vascular media thickness to CSA (media/CSA) in MCT‐PAH rats treated with vehicle or 4‐PBA. E, Representative immunoblots and quantified data of GSDME‐NT, GSDMD‐NT, and cleaved caspase‐3 in lung tissue of MCT‐PAH rats treated with or without 4‐PBA. F, GSDME and CD31 immunofluorescence staining in Con, MCT, and MCT+4‐PBA rats. Nuclei were stained with DAPI. Representative fluorescent‐stained sections are shown. Statistical analysis was performed using 1‐way ANOVA followed by Tukey test or Brown‐Forsythe and Welch ANOVA tests. Scale bar=50 μm. 4‐PBA indicates 4‐phenylbutyric acid; CD31, platelet endothelial cell adhesion molecule‐1; Con, control; CSA, cross‐sectional area; DAPI, 4',6‐diamidino‐2‐phenylindole; ER, endoplasmic reticulum; GSDMD, gasdermin D; GSDMD‐FL, gasdermin D full length; GSDMD‐NT, gasdermin D N‐terminal; GSDME, gasdermin E; GSDME‐FL, gasdermin E full length; GSDME‐NT, gasdermin E N‐terminal; MCT, monocrotaline; PAH, pulmonary arterial hypertension; RV/(LV+S), right ventricle/(left ventricle+septum); and RVSP, right ventricular systolic pressure.

GSDME, Rather Than GSDMD, Was Selectively Activated in ER Stress‐Induced PAECs Pyroptosis

To test whether ER stress could induce PAECs pyroptosis in vitro, we used tunicamycin, a potential pharmacologic ER stress inducer as previously described, ³⁴ , ³⁵ to treat PAECs at different concentrations (0, 1, 5 μg/mL). The results showed that tunicamycin treatment at concentrations of both 1 and 5 μg/mL for 24 hours elevated the protein level of GRP78 (Figure 3D). Treatment with tunicamycin at 5 μg/mL induced PAECs pyroptosis featuring cytoplasmic swelling and membrane rupture, whereas no significant changes were observed in PAECs treated with a lower level of tunicamycin at 1 μg/mL. Meanwhile, an LDH release assay revealed increased LDH release in 5 μg/mL tunicamycin‐treated PAECs rather than 1 μg/mL (Figure 3A). We also examined the time‐course effect of ER stress on PAEC pyroptosis. The results showed that short‐term (12 hours) tunicamycin stimulation upregulated the GRP78 protein level (Figure 3E) but did not show the pyroptotic cells and LDH release changes (Figure 3B). The morphological features of pyroptosis and elevated LDH release were exhibited in PAECs treated with tunicamycin for 24 hours (Figure 3B). To identify the pyroptotic executors in tunicamycin‐treated PAECs, we measured the mRNA levels of gasdermins in PAECs with tunicamycin treatment under different concentrations. Results showed that treatment with tunicamycin at 1 μg/mL did not affect the expression of gasdermin mRNAs, whereas treatment with tunicamycin at 5 μg/mL increased the mRNA expression of GSDME (Figure 3C, Table S1). GSDME and GSDMD are 2 main pyroptotic executors that to date were the only gasdermins implicated in mediating endothelial cell pyroptosis. ²⁶ , ³⁶ , ³⁷ We also detected the protein level of GSDMD and GSDME in PAECs treated with tunicamycin. Consistent with the reverse transcriptase–quantitative polymerase chain reaction results, Western blotting showed that neither GSDME nor GSDMD protein levels were changed in 1 μg/mL tunicamycin‐treated PAECs. Nevertheless, treating PAECs with tunicamycin at 5 μg/mL, the GSDME‐NT (GSDME N‐terminal) was activated, whereas the GSDMD‐NT (GSDMD N‐terminal) was not induced (Figure 3D). The duration of the tunicamycin treatment test showed that GSDME‐NT was activated in PAECs treated with tunicamycin for 24 hours (Figure 3E).

Figure 3. ER stress selectively induced GSDME activation and pyroptosis in PAECs.

A, Photomicrographs and LDH release by PAECs treated with different concentrations of Tm (0, 1, 5 μg/mL) for 24 hours. The arrows indicate pyroptotic cells (n=6). B, Photomicrographs and LDH release by PAECs treated with 5 μg/mL Tm for 6, 12, and 24 hours. The arrows indicate pyroptotic cells (n=6–8). C and D, PAECs were treated with different concentrations of Tm (0, 1, 5 μg/mL) for 24 hours. C, The mRNA expression of GSDMs family genes (n=5). D, Representative immunoblots and quantified data of GSDME‐NT and GRP78 in PAECs (n=5–6). E, Representative immunoblots and quantified data of GSDME‐NT and GRP78 in PAECs treated with 5 μg/mL Tm for 6, 12, 24 hours (n=4–6). Statistical analysis was performed using 1‐way ANOVA followed by Tukey test, Brown‐Forsythe, Welch ANOVA tests, and unpaired t test or permutation test. Scale bar = 20 μm. ER indicates endoplasmic reticulum; GRP78, glucose regulatory protein 78; GSDM, gasdermin; GSDME, gasdermin E; GSDME‐NT, gasdermin E N‐terminal; GSDMA, gasdermin A; GSDMB, gasdermin B; GSDMC, gasdermin C; GSDMD, gasdermin D; GSDMD‐FL, gasdermin D full length; GSDMD‐NT, gasdermin D N‐terminal; GSDME, gasdermin E; GSDME‐FL, gasdermin E full length; GSDME‐NT, gasdermin E Nterminal; LDH, lactic dehydrogenase; PAEC, pulmonary artery endothelial cell; PJVK, pejvakin; and Tm, tunicamycin.

Caspase‐3/GSDME Mediated ER Stress‐Induced Pyroptosis in PAECs

To further determine which candidate mediated PAECs pyroptotic cell death, we knocked down GSDME or GSDMD by transfecting PAECs with siRNA, followed by stimulation with tunicamycin (5 μg/mL) for 24 hours. The effect of ER stress‐induced GSDME activation and LDH release was abolished when GSDME was knocked down (Figure 4A and 4B; Figure S1A and S1B; Table S2). In addition, LDH release was similarly decreased in PAECs with GSDMD knockdown (Figure 4D, Figure S1C, Table S2). These findings motivated us to explore whether there is mutual regulation between GSDMD and GSDME. We found that the knockdown of GSDMD reduced GSDME‐NT activation, whereas the knockdown of GSDME did not affect GSDMD‐NT activation (Figure 4A and 4C).

Figure 4. ER stress‐induced pyroptotic cell death in PAECs through caspase‐3/GSDME.

A and B, Representative immunoblots and quantified data of GSDME and GSDMD (A) and LDH release (B) by PAECs pretreated with si‐GSDME transfection for 24 hours followed by 5 μg/mL Tm treatment for another 24 hours (n=4). C and D, Representative immunoblots and quantified data of GSDME and GSDMD (C) and LDH release (D) by PAECs pretreated with si‐GSDMD transfection for 24 hours followed by 5 μg/mL Tm treatment for another 24 hours (n=4–6). E, Representative immunoblots and quantified data of GSDME‐NT and cleaved caspase‐3 in PAECs pretreated with different concentrations of the pan‐caspase inhibitor Z‐VAD‐FMK (0, 10, 50 μM) for 0.5 hours followed by 5 μg/mL Tm treatment for another 24 hours (n=5–6). F, Representative immunoblots and quantified data of GSDME‐NT and cleaved caspase‐3 and LDH release by PAECs pretreated with si‐CASP3 transfection for 24 hours followed by 5 μg/mL Tm treatment for another 24 hours (n=5–7). Statistical analysis was performed using 1‐way ANOVA followed by Tukey test, Brown‐Forsythe, and Welch ANOVA tests or permutation test. CASP3 indicates caspase‐3; DMSO, dimethyl sulfoxide; ER, endoplasmic reticulum; GSDMD, gasdermin D; GSDMD‐FL, gasdermin D full length; GSDMD‐NT, gasdermin D N‐terminal; GSDME, gasdermin E; GSDME‐FL, gasdermin E full length; GSDME‐NT, gasdermin E N‐terminal; LDH, lactic dehydrogenase; PAEC, pulmonary artery endothelial cell; Si‐CASP3, siRNA targeting CASP3; Si‐GSDMD, siRNA targeting GSDMD; Si‐GSDME, siRNA targeting GSDME; Si‐NC, non‐targeting control siRNA; Tm, tunicamycin; and Z‐VAD‐FMK, carbobenzoxy‐valyl‐alanyl‐aspartyl‐[O‐methyl]‐fluoromethylketone.

Caspases are known to be critical for the activation of gasdermins that mediate pyroptotic cell death. ³⁸ To confirm whether tunicamycin‐induced PAEC pyroptosis was dependent on caspase activation, we pretreated PAECs with pan‐caspase inhibitor Z‐VAD‐FMK (carbobenzoxy‐valyl‐alanyl‐aspartyl‐[O‐methyl]‐fluoromethylketone) for 30 minutes, and tunicamycin (5 μg/mL for 24 hours) was then used to activate ER stress in PAECs. Western blotting results showed that both 10 and 50 μM of Z‐VAD‐FMK partially inhibited tunicamycin‐induced PAEC pyroptosis (Figure 4E). Evidence has been provided that GSDME cleavage was triggered by caspase‐3 activation. ³⁹ Results showed that the cleaved caspase‐3 protein was upregulated by tunicamycin in PAECs, and was suppressed by Z‐VAD‐FMK treatment (Figure 4E). To clarify whether caspase‐3 was involved in the tunicamycin‐induced PAEC pyroptosis, we used siRNA to knock down CASP3 (encoding caspase‐3 protein) in PAECs. Compared with the si‐NC (non‐targeting control siRNA) group, the si‐CASP3 (siRNA targeting CASP3) group inhibited GSDME cleavage and decreased LDH release (Figure 4F, Figure S2, Table S2). The above results suggest that silencing CASP3 partially eliminates the effect of tunicamycin on GSDME‐dependent pyroptosis in PAECs.

ER Stress‐Induced IRE1α Kinase Branch Activation Led to Caspase‐3/GSDME Activation

ER stress triggers the UPR, including IRE1α, PERK (protein kinase R‐like endoplasmic reticulum kinase), and ATF6α (activating transcription factor 6α) pathway. IRE1α exerts fundamental effects on cell fate under various stress conditions. ⁴⁰ The kinase activity of IRE1α is particularly significant, because it activates pathways related to inflammation and cell death when ER stress becomes unabated. ⁴⁰ To further explore the mechanisms mediating ER stress‐induced PAEC pyroptosis, we used specific inhibitors to intervene each UPR pathway. ³³ , ⁴¹ By treatment with IRE1α kinase inhibitor KIRA6, tunicamycin‐induced caspase‐3 and GSDME activation were partly abolished. Moreover, the tunicamycin‐induced LDH release from PAECs showed a corresponding reduction upon KIRA6 treatment (Figure 5A). To further characterize pyroptosis‐like cell death mediated by ER stress, PAECs were costained with Hoechst 33342 and PI (propidium iodide). Consistent with the increased LDH release and GSDME‐NT protein level, tunicamycin treatment significantly increased the percentage of PI‐positive cells, which was rescued by KIRA6 treatment (Figure S3). Inhibition of the IRE1α endoribonuclease by STF‐083010 significantly downregulated its downstream XBP1s protein level but did not alter the level of GSDME‐NT or LDH release (Figure 5B), which was in line with the role of XBP1s as a prosurvival transcription factor. The PERK inhibitor GSK2656157 suppressed its downstream ATF4 (activating transcription factor 4) expression and eIF2α (α‐subunit of eukaryotic translational initiation factor 2) activation, as indicated by eIF2α phosphorylated at the Ser51 site, but increased the activation of GSDME‐NT and LDH release (Figure 5C). The ATF6α inhibitor Ceapin‐A7 reduced GRP78 protein level, whereas it did not affect the cleavage of GSDME or LDH release from PAECs (Figure 5D). These results indicated that among the 3 downstream sensor pathways mediating ER stress, only the IRE1α kinase pathway specifically mediates ER stress‐induced pyroptosis in PAECs, not the PERK or ATF6α pathways. To further demonstrate the impact of IRE1α activation in ER stress‐induced PAEC pyroptosis and GSDME cleavage, we silenced ERN1 (endoplasmic reticulum‐nuclei‐1, encoding the IRE1α protein) in PAECs by transfecting PAECs with siRNA. Consistent with the findings from KIRA6 intervention, the knockdown of ERN1 significantly suppressed the activation of GSDME (Figure S4A, Table S2) and the release of LDH induced by tunicamycin (Figure S4B).

Figure 5. Inhibition of IRE1α kinase activity attenuated GSDME‐dependent pyroptosis in PAECs.

A, Representative immunoblots and quantified data of p‐IRE1α^S724, IRE1α, GSDME‐NT, and cleaved caspase‐3, and LDH release in PAECs pretreated with 1 μM KIRA6 (IRE1α kinase inhibitor) for 30 minutes followed by 5 μg/mL Tm treatment for another 24 hours (n=6–7). B, Representative immunoblots and quantified data of GSDME‐NT and LDH release in PAECs pretreated with 50 μM STF‐083010 for 30 minutes followed by 5 μg/mL Tm treatment for another 24 hours (n=5–6). C, Representative immunoblots and quantified data of p‐eIF2α^S51, eIF2α, GSDME‐NT, and LDH release in PAECs pretreated with 0.5 μM GSK2656157 for 30 minutes followed by 5 μg/mL Tm treatment for another 24 hours (n=6). D, Representative immunoblots and quantified data of GSDME‐NT and LDH release in PAECs pretreated with 5 μM Ceapin‐A7 for 30 minutes followed by 5 μg/mL Tm treatment for another 24 hours (n=6–9). Statistical analysis was performed using 1‐way ANOVA followed by Tukey test, Brown‐Forsythe, and Welch ANOVA tests or permutation test. ATF4 indicates activating transcription factor 4; eIF2α, α‐subunit of eukaryotic translational initiation factor 2; GRP78, glucose regulatory protein 78; GSDME, gasdermin E; GSDME‐FL, gasdermin E full length; GSDME‐NT, gasdermin E N‐terminal; IRE1α, inositol‐requiring kinase 1α; LDH, lactic dehydrogenase; PAEC, pulmonary artery endothelial cell; p‐eIF2α^S51, phospho‐α‐subunit of eukaryotic translational initiation factor 2; p‐IRE1α^S724, phosphorylation of IRE1α at S724; Tm, tunicamycin; and XBP1s, spliced X‐box binding protein 1.

Inhibition of the IRE1α Kinase Pathway Prevented the Activation of the Caspase‐3/GSDME Pathway in Endothelium and Attenuated Pulmonary Vascular Remodeling

To examine the contribution of IRE1α kinase activity in PAH, we first detected the expression and localization of p‐IRE1α^S724 (the phosphorylation of IRE1α at S724). The protein level of p‐IRE1α^S724 was upregulated at 3 days after monocrotaline injection in lung tissue of the PAH rat model and then gradually decreased (Figure 6A). Consistently, immunofluorescence analysis showed that the expression of p‐IRE1α^S724 in lung endothelial cells increased 3 days after monocrotaline injection but declined by 21 days (Figure 6B). We also detected the protein levels of ER stress markers and pyroptosis effectors at days 3, 7, and 14 post‐monocrotaline injection. The results indicated that the protein levels of GRP78 and GSDME‐NT were elevated in PAH rat lung tissue as early as day 3 post‐monocrotaline injection, with a gradual upregulation observed over time. The level of GSDMD‐NT protein exhibited a tendency to increase after 3 and 7 days of monocrotaline injection with no statistical significance, reaching statistical significance by day 14 post‐monocrotaline injection (Figure S5A). Moreover, immunofluorescence staining showed that GSDME, colocalized with CD31, was upregulated 3 days after monocrotaline injection (Figure S5B). These results indicated that ER stress, especially IRE1α pathway and endothelial pyroptosis, were activated at the early stage of PAH. To further determine the clinical potential of IRE1α‐targeted therapy in the development of PAH, we used the inhibitor of IRE1α kinase activity KIRA6 in PAH rats. We first examined the effective target engagement of KIRA6, and the results indicated that, compared with the monocrotaline group, KIRA6 significantly reduced the protein levels of p‐IRE1α^S724, cleaved caspase‐3, and GSDME‐NT at day 3 following monocrotalineinjection (Figure S6). Furthermore, the KIRA6 treatment group exhibited slight PAH, characterized by decreased RVSP, and RV hypertrophy (Figure 6C). Meanwhile, pulmonary vascular remodeling was alleviated, as shown by a reduction in medial wall thickness in both small (<50 μm) and large (>50 μm) vessels in the KIRA6 treatment group (Figure 6D). Notably, Western blotting confirmed the efficacy of KIRA6, as evidenced by the reduction in p‐IRE1α^S724. KIRA6 administration counteracted the activation of caspase‐3/GSDME in PAH rats (Figure 6E). Moreover, immunofluorescence analysis revealed a significant increase in GSDME protein level in the pulmonary vascular endothelium in lung tissue sections from PAH rats, which was partly prevented by KIRA6 treatment (Figure 6F). These results suggested that increased p‐IRE1α^S724 drives endothelial pyroptosis via caspase‐3/GSDME in PAH, which could be effectively suppressed by inhibiting IRE1α kinase activity. We also tested the p‐IRE1α^S724 and GSDME‐NT protein level in the Sugen5416 combined with 10% chronic hypoxia‐induced PAH mouse model (SuHx). Both were higher in lung tissue from SuHx mice compared with the normoxia group (Figure S7). To further demonstrate whether inhibition of IRE1α kinase activity could be a promising therapeutic target for PAH, we evaluated the efficacy of KIRA6 treatment in the SuHx model. First, by using 2 preventive administration schemes for KIRA6, we tested the efficacy of KIRA6 in preventing PAH development in mice. SuHx mice treated with KIRA6 starting 1 week before Sugen5416 injection exhibited improved hemodynamics as shown by a decrease in RVSP and RV/(LV+S) (Figure S8A and S8B). In addition, pulmonary vascular remodeling in SuHx mice treated with KIRA6 was significantly improved, as indicated by a reduction in medial wall thickness (Figure S8C and S8D). Similarly, we also observed the alleviated PAH phenotype (Figure S8E through S8H) in the SuHx mice treated with KIRA6 on the day of Sugen5416 injection in comparison with the controls. Second, we also established PAH therapeutic model to illustrate the therapeutic effect of KIRA6 on PAH. KIRA6 was administered for 2 weeks once the SuHx model was established, which was confirmed by a significant increase in RVSP (Figure S9A). Compared with vehicle treatment, KIRA6 administration reduced RVSP, RV/(LV+S), and pulmonary vascular remodeling (Figure S9B through S9E). These results indicate that inhibition of IRE1α kinase activity could not only prevent PAH but also alleviate established PAH.

Figure 6. IRE1α inhibition prevented the activation of caspase‐3/GSDME in the lung tissue of MCT‐PAH rats.

A, Representative immunoblots and quantified data of p‐IRE1α^S724 and IRE1α protein levels in lung tissue of MCT‐PAH rats at indicated days after intraperitoneal injection of MCT (n=6). B, p‐IRE1α^S724 and CD31 immunofluorescence staining in control and MCT‐PAH rats at days 3 and 21 post‐MCT injection. Nuclei were stained with DAPI. Representative fluorescent‐stained sections are shown (n=6). C through F, MCT‐PAH rats treated with or without the IRE1α kinase inhibitor KIRA6 (7 mg/kg per day) (n=6). RVSP and the ratio of RV/(LV+S) in vehicle‐ or KIRA6‐treated MCT‐PAH rats (C). Representative images of hematoxylin and eosin staining of lung sections and ratio of vascular media thickness to CSA (media/CSA) in vehicle‐ or KIRA6‐treated MCT‐PAH rats (D). Representative immunoblots of p‐IRE1α^S724, IRE1α, GSDME‐NT, and cleaved caspase‐3 protein levels in lung tissue of MCT‐PAH rats treated with or without KIRA6 (E). GSDME and CD31 (platelet endothelial cell adhesion molecule‐1) immunofluorescence staining in Con, MCT, and MCT+KIRA6 rats (F). Nuclei were stained with DAPI. Representative fluorescent‐stained sections are shown. Statistical analysis was performed using 1‐way ANOVA followed by Tukey test or Brown‐Forsythe and Welch ANOVA tests. Scale bar=50 μm. CD31 indicates platelet endothelial cell adhesion molecule‐1; Con, control; CSA, cross‐sectional area; D, day; DAPI, 4',6‐diamidino‐2‐phenylindole; GSDME, gasdermin E; GSDME‐FL, gasdermin E full length; GSDME‐NT, gasdermin E N‐terminal; IRE1α, inositol‐requiring kinase 1α; MCT, monocrotaline; PAH, pulmonary arterial hypertension; p‐IRE1α^S724, phosphorylation of IRE1α at S724; RV/(LV+S), right ventricle/(left ventricle+septum); and RSVP, right ventricular systolic pressure.

DISCUSSION

In this study, we confirmed that ER stress induced PAECs pyroptosis in a PAH model. This was dependent on the caspase‐3/GSDME axis. Among the 3 branches of unfolded protein response in response to ER stress, the activation of caspase‐3/GSDME was regulated by IRE1α, which induced pyroptosis in PAECs. KIRA6, the inhibitor of IRE1α kinase activity, attenuated pulmonary vascular remodeling and PAH symptoms associated with lower‐level activation of the caspase‐3/GSDME pathway in a monocrotaline‐induced PAH rat model.

Endothelial dysfunction serves as the primary instigating factor in PAH, triggering a cascade of events that drive disease progression. The injured endothelium secretes a range of growth factors, cytokines, and chemokines that fundamentally alter smooth muscle cell behavior, promoting survival, proliferation, and a synthetic phenotypic switch. These paracrine signals foster a hyperproliferative and apoptosis‐resistant PASMC population, ultimately culminating in the characteristic structural remodeling of the pulmonary vascular wall. ³ , ⁴² Therefore, vascular remodeling caused by abnormal proliferation of PASMCs occurs at a relatively late stage of PAH. Early intervention at the level of endothelial injury could therefore mitigate downstream PASMC hyperproliferation by limiting exposure to endothelial‐derived signaling. Current research on ER stress in PAH has largely focused on PASMCs, ³⁰ , ⁴³ , ⁴⁴ , ⁴⁵ linking it to vascular remodeling at later stages of PAH. However, the role of endothelial ER stress, particularly in early PAH, remains largely unexplored. Our findings provide preliminary evidence that ER stress contributes to endothelial injury as an early event in PAH pathogenesis.

Our previous work has established that endothelial pyroptosis played a significant role in the pathogenesis of PAH. Inhibition of caspase‐4/11 alleviated PAH by reducing endothelial pyroptosis. ²⁶ Building upon our prior research, we further demonstrated that endothelial cell pyroptosis occurs during the early stages of PAH. In addition, the present study identified the specific upstream driver that initiates this pyroptosis. We discovered that sustained ER stress triggered endothelial pyroptosis during the early stages of PAH.

A recent study has demonstrated that homocysteine accelerates atherosclerosis progression by enhancing macrophages pyroptosis through the promotion of endoplasmic reticulum stress. ⁴⁶ This finding indicates that endoplasmic reticulum stress‐mediated pyroptosis contributes to the pathogenesis of cardiovascular diseases. Our study first identified that ER stress‐regulated endothelial pyroptosis is involved in the progression of PAH. Pharmacological inhibition of ER stress using 4‐PBA resulted in reduced GSDME activation in the endothelium. We also noted that another pyroptotic executor, GSDMD, was primarily activated in the PASMCs of lung tissue in PAH rats, and the cleavage of GSDMD was decreased in response to 4‐PBA administration. Further investigation is required to elucidate the role of ER stress in regulating PASMCs pyroptosis and the function of GSDMD in PAH.

Distinct stimuli activate different gasdermin subtypes in induction of pyroptosis in different cell types. In vascular endothelial cells, low shear stress triggers GSDMD‐mediated endothelial cell pyroptosis via the 10‐11 translocation 2/succinate dehydrogenase B/reactive oxygen species pathway. ⁴⁷ Caspase‐1/GSDMD activation mediates homocysteine‐induced pyroptosis in endothelial cells. ⁴⁸ In our study, we observed the cleavage of GSDME instead of GSDMD in ECs under a relatively high concentration or a long duration of ER stress stimulation. Moreover, knocking down GSDME or CASP3s prevented ER stress‐induced PAECs pyroptosis. We revealed that the caspase‐3/GSDME pathway was selectively activated in PAECs pyroptosis in response to ER stress. Unexpectedly, the knockdown of GSDMD also inhibited ER stress‐triggered pyroptosis in PAECs. Some research suggested a regulation between GSDMD and GSDME‐mediated pyroptosis. It has been demonstrated that GSDMD‐deficient cells are still susceptible to caspase‐1‐mediated macrophage death. This involves caspase‐1‐induced activation of the BH3 (bcl‐2 homology 3)‐interacting domain death agonist‐caspase‐9‐caspase‐3 axis, which can be followed by GSDME‐dependent secondary pyroptosis. ⁴⁹ In addition, the knockdown of GSDMD led to a reduction of GSDME‐NT via the inactivation of caspase‐3 in human aplastic thyroid carcinoma cell lines. ⁵⁰ In our research, we demonstrated that the inhibition of ER stress‐triggered pyroptosis upon GSDMD knockdown was associated with a decrease in GSDME cleavage, whereas knocking down GSDME did not affect the GSDMD protein level. These findings highly suggest the presence of a regulatory mechanism whereby GSDMD modulates GSDME activation during ER stress‐induced pyroptosis. However, in this study, we did not examine the mechanism by which GSDMD regulates GSDME in the context of ER stress, which will be further elucidated in a future study.

Our previous work reported that caspase‐4/11 contributes to endothelial pyroptosis both directly via GSDMD cleavage and indirectly through caspase‐3‐mediated GSDME cleavage. ²⁶ In the present study, we observed that ER stress (tunicamycin treatment) triggers caspase‐4 activation in PAECs, as evidenced by increased protein level of cleaved caspase‐4. Knockdown of CASP4 (encoding caspase‐4 protein) significantly suppressed the ER stress‐induced increases in both cleaved caspase‐3 and GSDME‐NT (Figure S10A, Table S2). These results suggest that caspase‐4 activation upon ER stress contributes to the subsequent cleavage of caspase‐3 and GSDME. To further elucidate the absence of GSDMD cleavage during ER stress‐induced pyroptosis, we performed a coimmunoprecipitation assay as previously described. ⁵¹ Our results showed that, in the condition of ER stress activation, caspase‐4 could not interact with GSDMD, which may explain why GSDMD remains uncleaved in this context (Figure S10B).

Although our previous work demonstrated that pro‐PAH factors such as TNF‐α (tumor necrosis factor‐alpha) trigger a caspase‐4/11‐dependent pathway that cleaves both GSDMD and GSDME, our current data suggest that the persistent ER stress may preferentially engage the caspase‐4/caspase‐3/GSDME axis, with GSDMD cleavage appearing less prominent. Several factors could potentially dictate this substrate preference. Existing literature has established that calcium overload can trigger the activation of caspase‐4 within the Apaf‐1 complex, a process that subsequently directs signaling toward the caspase‐3/GSDME axis rather than GSDMD. ⁵² , ⁵³ Notably, calcium dyshomeostasis is also present during ER stress. It is therefore possible that the intracellular calcium dyshomeostasis associated with ER stress might contribute to a preference for caspase‐3/GSDME activation, whereas GSDMD cleavage does not appear to be the primary event in this specific context. In addition, our prior research suggested that O‐GlcNAc (O‐linked N‐acetylglucosamine) modification of GSDMD could hinder its binding to caspase‐4. ⁵⁴ Given that ER stress is typically associated with increased protein O‐GlcNAcylation, it is possible that in this context, GSDMD is less accessible to activated caspase‐4, which could potentially preclude the cleavage and activation of GSDMD.

To illustrate the mechanism that triggers ER stress‐induced endothelial cell pyroptosis, we tested the signal transduction by 3 ER‐resident transmembrane proteins that orchestrate the UPR, including IRE1α, ATF6α, and PERK. We reported that inhibiting IRE1α kinase activity effectively suppressed the activation of the caspase‐3/GSDME pathway and attenuated pyroptosis in PAECs. Nevertheless, neither interfering with the IRE1α endoribonuclease nor the ATF6α activity affects GSDME cleavage and PAECs pyroptosis. IRE1α is the cell fate executor among UPR. Dissimilar effects on cell survival could be observed in response to different IRE1α‐induced pathways. Under mild ER stress, the endoribonuclease activity of IRE1α typically promotes cell homeostasis by activating XBP1s. ⁵⁵ However, if ER stress becomes irremediable, IRE1α becomes hyperactivated. Activation of kinase activity induces phosphorylation of downstream pathways to promote inflammation and cell demise. ⁵⁶ , ⁵⁷ Recent studies reported that the kinase activity of the IRE1α‐induced pathway exacerbated herpes simplex virus‐2 infection‐induced neuron pyroptosis. ²⁸ Our investigation revealed that inhibiting of the kinase activity of IRE1α mitigated severe ER stress‐triggered caspase‐3/GSDME activation and subsequent pyroptosis in PAECs. Surprisingly, GSK2656157, an inhibitor of PERK kinase activity, unexpectedly facilitated GSDME cleavage and exacerbated PAEC pyroptosis. A previous study reported that EIF2AK3 ^KO (EIF2AK3 encoding the PERK protein, EIF2AK3 knockout) melanoma cells underwent ER mass enlargement, expanded misfolded protein accumulation, and had augmented reactive oxygen species levels after treatment with thapsigargin. ⁵⁸ Another research study found that the loss of PERK signaling impedes mitochondrial respiration, indicating impaired mitochondrial function. ⁵⁹ Given that ER stress, reactive oxygen species, and mitochondrial dysfunction have been well established as inducers of pyroptosis, ⁶⁰ , ⁶¹ , ⁶² it is plausible that loss of PERK may induce pyroptosis via some of these pathways. Notably, our study demonstrated that inhibition of PERK signaling upregulated GSDME cleavage and amplified PAEC pyroptosis. The specific mechanisms underlying these observations need further investigation in our future studies.

We observed that the changes in total IRE1α and phosphorylated IRE1α exhibited parallel trends in the lung tissues of monocrotaline‐induced PAH rats. The dynamics of total IRE1α protein may be governed by a positive‐feedback mechanism responsive to changes in p‐IRE1α^S724 levels. This concept is supported by the existing literature, which demonstrates that IRE1α transcription can be autoregulated through a kinase/JNK (JUN N‐terminal kinase)‐dependent feed‐forward loop. ⁶³ Thus, we speculate that the observed trend toward increased total IRE1α protein expression may be attributed to positive feedback by the p‐IRE1α^S724 protein itself, resulting in a constant ratio.

IRE1α is activated when the endoplasmic reticulum and cellular homeostasis are disrupted, thereby participating in physiological regulation and disease development by determining cell fate. ⁴⁰ , ⁶⁴ , ⁶⁵ Early endothelial cell dysfunction is thought to be an initiating event in the development of PAH. In our study, we detected IRE1α activation at the early stage of PAH, with specific localization and activation in endothelial cells. These results indicate that the early and sustained IRE1α activation may contribute to endothelial cell injury. Additionally, our data further demonstrated GSDME activation in the pulmonary endothelium as early as 3 days after PAH induction. The temporal concurrence of endothelial injury, IRE1α activation, and GSDME activation suggests a mechanistic link in which early ER stress‐induced IRE1α activation may trigger pyroptosis through GSDME, thereby amplifying endothelial dysfunction and contributing to PAH progression.

To test this hypothesis, we used KIRA6, an IRE1α kinase activity inhibitor, for both in vitro and in vivo interventions. The results showed that tunicamycin‐induced caspase‐3 and GSDME activation were significantly abolished by KIRA6. Consistently, the elevation of LDH release triggered by tunicamycin declined by KIRA6. As for in vivo intervention, results showed that KIRA6 treatment reduced RVSP, RV/(LV+S), as well as pulmonary vascular remodeling. Moreover, activation of caspase‐3 and GSDME in monocrotaline rats were alleviated after KIRA6 administration. We further interfered with the expression of IRE1α by using siRNA targeting ERN1 in vitro. The results showed a considerable decrease in the activation of GSDME and the release of LDH induced by tunicamycin when silencing ERN1, confirming the pivotal role of IRE1α in ER stress‐induced PAEC pyroptosis. Because the endothelial injury is recognized as the initiating factor of PAH, our results provide preliminary evidence that early IRE1α activation could participate in endothelial pyroptosis and potentially influence the early stages of PAH pathogenesis.

Our assessment of IRE1α kinase activity in PAH relied solely on pharmacological interventions, which have inherent limitations in specificity. Therefore, future research is necessary to confirm the precise role of IRE1α kinase activity in PAH, potentially using more specific methodologies such as the generation of IRE1α kinase active site mutant mice. Moreover, we did not clarify the detailed mechanism of IRE1α in the regulation of caspase‐3. Studies in other cell types have provided some clues to the mechanisms. First, IRE1α interacts with TRAF2 (tumor necrosis factor receptor‐associated factor 2) to activate downstream MAPK (mitogen‐activated protein kinase) signaling pathways, ⁶⁶ including JNK and p38 (p38 mitogen‐activated protein kinase), in several cell types such as rat pheochromocytoma cells ⁶⁵ and rat cardiomyocytes. ⁶⁷ Second, IRE1α interacts with TRAF2, which in turn interacts with procaspase‐12 to promote its activation, ⁶⁸ in several cell types such as human non‐small cell lung cancer cells ⁶⁹ and human glioma cells. ⁷⁰ Third, receptor‐interacting serine/threonine‐protein kinase 1 interacts with IRE1α to mediate caspase‐8‐dependent activation of caspase‐3 in mouse embryonic fibroblasts. ⁷¹ Whether any of these established mechanisms or a novel pathway is responsible for IRE1α‐mediated caspase‐3 activation in our context requires further investigation.

KIRA6 has been reported to exert IRE1α‐independent effects on targeting HSP60 (heat shock protein 60) ⁷² and p38, ⁷³ respectively. Given that there are nearly no studies addressing the role of HSP60 in pulmonary hypertension, it remains unclear whether KIRA6 may exert its effects in preventing PAH through HSP60. However, because p38 is a common pathway involved in the pathogenesis of pulmonary hypertension, ⁷⁴ , ⁷⁵ , ⁷⁶ we cannot rule out the possibility that KIRA6 may partially inhibit pulmonary hypertension through the p38 pathway.

In summary, our findings established a new link between ER stress and endothelial cell pyroptosis during PAH. Inhibiting ER stress attenuated PAECs pyroptosis via the IRE1α/caspase‐3/GSDME pathway. Pharmacological interfering with the IRE1α kinase activity downregulated the caspase‐3/GSDME signaling in pulmonary vascular endothelium and subsequent endothelial integrity damage, thus decelerating the progression of PAH. Our findings provided new evidence that manipulation of IRE1α activity induced by ER stress could be a novel therapeutic approach to prevent PAECs pyroptosis and associated injury involved in PAH.

Sources of Funding

This work was partly supported by grants from the Ministry of Science and Technology of China (2021YFA0804803, 2022YFA1104204), the National Natural Science Foundation of China (82495172, 82421001), and the Shanghai Magnolia Talent Plan Pujiang Project (24PJA078).

Disclosures

None.

Supporting information

Data S1: Supplemental Methods

Tables S1–S2

Figures S1–S10

Reference 77

Unedited Gels