Experimental pulmonary arterial hypertension in mice with a pathogenic SOX17 variant

Abstract

The SRY-box transcription factor 17 gene (SOX17) has been identified as a causative gene for pulmonary arterial hypertension (PAH), but its underlying mechanisms remain unclear. We conducted experiments using mice carrying the Sox17 variant confirmed in patients with PAH. Mice with the heterozygous Sox17 c.397 C > G (p.Pro133Ala) variant (Sox17^+/Pro133Ala mice), identified in severe PAH patients resistant to combination therapy with pulmonary vasodilators, were created and raised in a hypoxic environment. The Fulton index, right ventricular systolic pressure, and medial wall thickness were increased in Sox17^+/Pro133Ala mice compared with those in wild-type controls, suggesting the presence of PAH in Sox17^+/Pro133Ala mice. RNA sequencing of murine lungs demonstrated that Cyp1b1 expression was elevated in the Sox17^+/Pro133Ala mice. CYP1B1 encodes cytochrome P450 1B1, an enzyme that plays a key role in the metabolism of various endogenous and exogenous compounds, particularly in the oxidative metabolism of steroid hormones such as estrogens. CH-223,191, an aryl hydrocarbon receptor antagonist, suppressed Cyp1b1 expression, alleviating the PAH phenotype. Furthermore, the expressions of Bmpr2, Col4a1, and Col4a2, which were downregulated in Sox17^+/Pro133Ala mice, were restored. These findings suggest that multiple pathways, including Cyp1b1, Bmpr2, and Col4a1/Col4a2, are interconnected and altered by a single base substitution in Sox17, and that PAH patients with SOX17 variants may develop PAH through these pathways. These results demonstrate that Sox17^{+/p.Pro133Ala} mice exhibited right heart overload and pulmonary vascular remodeling when subjected to hypoxic stress, confirming that this mouse model mimics the human PAH phenotype.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-46893-0.

Introduction

Pulmonary arterial hypertension (PAH), an intractable disease that causes right heart failure because of the thickening of the pulmonary artery wall, can be fatal. Genetic variants in bone morphogenetic protein receptor type 2 (BMPR2) were identified as the cause of familial PAH¹, and various causative genes, including activin A receptor like type 1 and endoglin, have been reported². In 2018, Grӓf et al. reported the results of a large-scale observational study in Europe, and SRY-box transcription factor 17 gene (SOX17) was identified as a new causative gene³. Supporting this, there have been reports of families in which PAH was caused by the loss-of-function of SOX17⁴. Furthermore, patients with SOX17 variants develop PAH at an early age⁵, and pathogenic variants in SOX17 have been identified in patients with PAH concomitant with congenital heart disease⁶. However, the mechanism of PAH development caused by SOX17 variants remains largely unknown.

Several studies have been conducted using Sox17 knockout mice to investigate the role of SOX17. Park et al. reported that HGF/c-Met signaling was associated with the development of PAH in Sox17 knockout mice⁷, while Sangam et al. demonstrated that 16α-OHE induced downregulation of SOX17 contributed to the development of PAH⁸. In humans, however, single-base substitution variants in SOX17 accumulate in the high mobility group box (HMG box) region identified in patients with PAH⁹. Thus, it remains unclear whether the mechanisms identified in knockout mouse models mimicking the effects of genetic alterations observed in patients with PAH.

In this study, we introduced the variant observed in patients with severe PAH into mice to investigate the resulting phenotypes and what mechanisms influence the onset of PAH.

Methods

Human subjects

This genetic study regarding patients with PAH was approved by the ethical review board of Keio University Hospital (#20140203), and research involving human research participants was performed in accordance with the Declaration of Helsinki. The genetic testing was performed with informed written consent from all patients, followed by the genetic counseling if patients requested it. The variants in our patients were identified by whole exome sequencing as previously reported¹⁰. In this study, we retrospectively assessed the data from 29/Sep/ 2014 to 30/Sep/2021, and data was accessed on 1/Dec/2023. We targeted a patient with severe PAH who carries the heterozygous SOX17 c.397 C > G (p.Pro133Ala) variant. Informed written consent was obtained, and the lung specimens obtained from autopsy were analyzed. Details of patient characteristics were described previously¹¹.

Protein structure analysis

Molecular dynamic simulation of the three-dimensional structure of the HMG box of SOX17, with and without the Pro133Ala mutation was performed. Newton’s second law, F (force) = m (mass) × a (acceleration), was solved for all atoms when constructing the HMG box of SOX17 along with the water molecules surrounding them. CHARMM-36 was used as a force field. The initial structure of the wild-type form was obtained from solution structure by nuclear magnetic resonance (protein data bank: 2yul)¹². The initial structure of the mutant form was obtained by inducing an amino-acid substitution in the wild-type structure using the mutate residue plugin of Visual Molecular Dynamics (VMD) version 1.9.3 as described elsewhere¹³. The water molecules in the wild-type structure were removed and re-solvated under the same conditions as the structure with the mutations. The water molecules were modeled as transferable intermolecular potential water molecules (TIP3P)¹⁴. The position and velocity vector of each atom and water molecule were calculated for each 2.0 femto-second (1 × 10^{− 15} sec). Particle mesh Ewald (PME) summation¹⁵ with a cut-off length of 12 Å for direct interactions was used for predicting long-range electrostatic interactions.

All results were visualized with VMD version 1.9.3. The root mean square deviation (RMSD) of all atoms, excluding water molecules, was calculated for each 10 pico-seconds. The calculation was assumed to be stabilized when the RMSD stopped changing with an extension of calculation time. The multi-dimensional calculations of Newton’s second law were conducted with NAnoscale Molecular Dynamics software¹⁶.

Mice and materials

All animal studies were performed in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health Guidelines. All experiments were performed in accordance with relevant guidelines and regulations. This study was designed in accordance with the ARRIVE guidelines. The experimental procedures were approved by the Boards for Studies in Experimental Animals of Keio University (#D2017-013) and Nippon Veterinary and Life Science University (#2019K-11). Mouse experiments were performed using male C57BL/6 mice purchased from Charles River Laboratories (Wilmington, MA). Mice carrying the heterozygous variant (Sox17^{+/p.Pro133Ala} mice) were created with clustered regularly interspaced short palindromic repeats/CRISPR-associated protein 9 technology. Target sequence of CRISPR/Cas9 was designed using CHOPCHOP website¹⁷, and guide RNA was determined as CCACCCCAACTACAAGTACCGGC. Single-stranded oligodeoxynucleotide was described in supplemental figure S1A. Sox17^{+/p.Pro133Ala} mice were generated using electroporation technique done by Charles River Laboratories (Supplemental Figure S1A). As the genotyping, genomic DNA was extracted from a tail using Easy-DNA gDNA Purification Kit (Thermo Fisher Scientific, Waltham, MA), and PCR was performed with AmpliTaq Gold 360 Master Mix (Thermo Fisher Scientific, Waltham, MA) and the primers of F: AATAGGACCGGGCTTCTCAC, and R: CTGCATAGTCCGAGACTGGAG. Initial denaturation at 95 °C for 10 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 60 °C for 30 s, extension at 72 °C for 1 min, and final extension at 72 °C for 2 min. PCR products were analyzed by electrophoresis on a 2% agarose gel containing ethidium bromide. Bands were extracted with x-tracta™ Gel Extractor (Promega Corporation Fitchburg, WI), and sequence analysis was performed to identify the genotype. Mice aged 9–12 weeks were maintained in a normoxic or a hypoxic environment (10% oxygen) for 4 weeks (Supplemental Figure S1B).Aryl hydrocarbon receptor (AhR) antagonist (CH-223191, Sigma-Aldrich, St. Louis, MO) was administrated, in which CH-223,191 (10 mg/kg) diluted in dimethyl sulfoxide (DMSO) and corn oil, intraperitoneally to mice twice a week in addition to hypoxic environment (Supplemental Figure S1C). The dose and administration schedule of CH223191 were determined based on previously published studies¹⁸. Each group contained 7 or 8 mice. Mice were euthanized at humane endpoints if they exhibited decreased activity, more than 20% lower body weight compared with controls, weight loss exceeding 25% within 7 days, cachexia, tachypnea, or labored breathing. All procedures were performed under isoflurane anesthesia, and euthanasia was carried out by cervical dislocation, with every effort made to minimize distress.

Cardiac assessment

Mice were anesthetized with isoflurane, then their bodyweights (BW) were measured, and electrocardiograms were monitored. To measure the right ventricular systolic pressure (RVSP), a micro-tip catheter (Millar Instruments, Inc., Houston, TX, USA) was inserted into the right jugular vein under anesthesia with isoflurane inhalation. Hemodynamic data were analyzed with Lab Chart 8 (ADInstruments Ltd., Oxford, UK). In addition, to assess the degree of right heart load, hearts were harvested and separated into the right ventricle (RV) and left ventricle plus septum (LV + S), and the weights of the RV and LV + S were measured to calculate the Fulton index (RV/LV + S).

Vascular remodeling analysis

The lungs of mice were fixed with 10% formalin, embedded in paraffin, sectioned at 4-µm thickness, and stained with Elastica van Gieson (Sigma-Aldrich, Inc., Saint Louis, MO). We measured the medial wall thickening in arteries with diameters ranging from 40 to 80 μm. Four arteries per section were measured. Images were captured by light microscopy (BZ-9000; Keyence, Osaka, Japan).

Gene expression analysis

RNA was extracted from whole mouse lung tissue using TRlzol LS Reagent (#10296028, Thermo Fisher Scientific, Waltham, MA) and subjected to RNA sequencing using Illumina NovaSeq 6000 (Illumina Inc., SAN DIEGO, CA) with the read-depth of approximately 25 million per sample. Read counting using fastq files, Differentially Expressed Genes (DEGs) analysis using iDEP2.01, and Gene Ontology (GO) analysis using Metascape were performed¹⁹. Fold-changes over 1.5 and a p-value less than 0.05 were considered as statistically significant. In addition, to investigate which gene expressions were affected by the AhR antagonist, genes were narrowed down based on the following criteria: (i) genes whose expression levels were altered in Sox17^{+/p.Pro133Ala} mice compared with wild-type mice; (ii) genes with changed expressions associated with the Sox17 heterozygous variant (Sox17^{+/p.Pro133Ala}), which were reversed by the AhR antagonist; and (iii) genes whose expressions were not altered by the AhR antagonist in wild-type mice. The concentration of extracted RNA was measured using Nanodrop (Thermo Fisher Scientific), and complementary DNA was synthesized from 1 µg of RNA using SuperScriptIII First-Strand Synthesis SuperMix (#18080400 Thermo Fisher Scientific). Quantitative polymerase chain reaction (qPCR) was performed using a KAPA SYBR Fast qPCR Master Mix (2X) Kit (KAPA BIOSYSTEMS, Wilmington, MA). Polymerase Chain Reaction was performed using the ViiA7 system (Applied Biosystems, Waltham, MA). The expression levels were compared by measuring fold changes using the ΔΔCT method. The primers used were as follows: F: TGCACCACCAACTGCTTAG and R: GGATGCAGGGATGATGTTC for glyceraldehyde-3-phosphate dehydrogenase (Gapdh)²⁰ as an endogenous control gene; F: GCCACTATTACGGACATCTTCGG and R: ACAACCTGGTCCAACTCAGCCT for cytochrome P450 family 1 subfamily B member 1 (Cyp1b1) (SKU MP203248, ORIGENE, MD, USA); F: AAGCTGCTGGAGCTGATTGG and R: AACTGGACGCTCATCCAAGG for Bmpr2²¹; F: CTGGCACAAAAGGGACGAG and R: ACGTGGCCGAGAATTTCACC for collagen, type IV, alpha 1 (Col4a1)²²; and F: CCCGGATCTGTACAAGGGTG and R: TGATGCCTTCCTCGCCTTTT for collagen, type IV, alpha 2 (Col4a2)²². Publicly available ChIP-seq datasets were interrogated using ChIP-Atlas²³ to evaluate potential SOX17 binding near the target gene loci.

Statistical analysis

All values are expressed as the mean ± standard deviation and analyzed with the Student’s t-test. A p-value less than 0.05 was considered to indicate a significant difference. Statistics were performed using SPSS (IBM SPSS Statistics for Windows, Version 29.0. Armonk, NY: IBM Corp), and figures were prepared using GraphPad Prism (Prism 8 for Windows 64-bit version 8.4.3; GraphPad Software, MA).

Results

Clinical features of a PAH patient with the SOX17 variant

The heterozygous SOX17 c.397 C > G (p.Pro133Ala) variant was located in the HMG box, which is known as a “hot-spot” area in patients with severe PAH^8,10. We report the case of a patient with severe PAH who had this variant. The patient was diagnosed with PAH at the age of 33 years and she died approximately 1.5 years after the diagnosis. Despite a combination of three types of pulmonary vasodilators, sufficient reductions in the mean pulmonary artery pressure and pulmonary vascular resistance were not achieved, and this patient died because of right heart failure (Fig. 1A).

Fig. 1.

Clinical course of a patient with SOX17 p.Pro133Ala variant and its structural changes. (A) The clinical time-course of the patient with severe pulmonary arterial hypertension who has the heterozygous SOX17 c.397 C > G (p.Pro133Ala) variant. The patient died shortly after right heart catheterization at X + 14 months. The blue line shows mean pulmonary arterial pressure (mPAP), the orange line shows pulmonary vascular resistance (PVR), and the black line shows serum levels of B-type natriuretic peptide (BNP). The medications are shown above the graph. (B) The 3D structure of HMG box of SOX17 for wild type (red) and Pro133Ala mutation (yellow). A change of p.Pro133Ala position is demonstrated as the black structure. The double-strand DNA structure is indicated in light blue. PDE5, phosphodiesterase type 5; p.o., per os.

Protein structural analysis

The effect of the heterozygous SOX17 c.397 C > G variant on the SOX17 protein structure was simulated (Fig. 1B). The SOX17 protein with c.397 C > G variant had different distance to DNA compared to the wild-type SOX17 protein, suggesting that this variant has a pathogenic effect.

PAH induced in mice with the Sox17 p.Pro133Ala variant

Sox17^{p.Pro133Ala/p.Pro133Ala} mice had embryonic lethality. Sox17^{+/p.Pro133Ala} mice and wild-type mice were raised under 10% hypoxic conditions (Supplemental Figure S1B and S1C). Sox17^{+/p.Pro133Ala} mice developed elevated RV/BW (1.089 ± 0.162 mg/g vs. 0.777 ± 0.076 mg/g, p < 0.001), elevated Fulton index (0.267 ± 0.032 vs. 0.221 ± 0.028, p = 0.016), and elevated RVSP (34.9 ± 2.1 mmHg vs. 32.5 ± 1.9 mmHg, p = 0.045), compared with control mice. Furthermore, in Sox17^{+/p.Pro133Ala} mice, Elastica van Gieson staining of lung tissues confirmed a significant thickening of the medial wall in pulmonary arteries compared with control mice (0.239 ± 0.053 vs. 0.186 ± 0.045, p < 0.001). These results demonstrate that Sox17^{+/p.Pro133Ala} mice exhibited right heart overload and pulmonary vascular remodeling when subjected to hypoxic stress, confirming that this mouse model mimics the human PAH phenotype.

Gene expressions

RNA-sequencing of whole lungs of mice confirmed that 451 genes were upregulated and that 651 genes were downregulated in Sox17^{+/p.Pro133Ala} mice compared with wild-type mice (Fig. 2A and B). GO analysis revealed that the expressions of genes related to xenobiotic metabolism by cytochrome P450 (Cyp450) showed marked changes (Fig. 2C). The significant increase in the expression of Cyp1b1 in Sox17^{+/p.Pro133Ala} mice was confirmed by qPCR.

Fig. 2.

Gene expression is regulated by the Sox17^{+/p.Pro133Ala} variant. Changes in gene expression in Sox17^{+/p.Pro133Ala} mice (hetero) compared with wild-type mice were analyzed using RNA-sequencing and displayed as a volcano plot (A) and heat map (B). (C) Changes in gene expression in both groups were assessed using gene ontology analysis.

Effects of an AhR antagonist on PAH with the Sox17 c.397 C > G variant

Next, we conducted experiments using CH-223,191, an AhR receptor antagonist, to inhibit the elevated Cyp450 in the lungs of Sox17^{+/p.Pro133Ala} mice. CH-223,191 (10 mg/kg) was administered intraperitoneally twice a week while the mice were kept in a hypoxic environment, designed to evaluate preventive effects (Supplementary Figure S1C). The effect of the AhR antagonist was confirmed by qPCR, which showed a decrease in Cyp1b1 expression (Fig. 3A). Although no significant changes were observed in wild-type mice, the RV/BW, Fulton index, and RVSP were significantly reduced by the AhR antagonist (Fig. 3B-D). A preventive effect on the remodeling of pulmonary arteries demonstrated by the medial wall thickness was observed in Sox17^{+/p.Pro133Ala} mice, but not in wild-type mice, after treatment with the AhR antagonist (Fig. 3E and F). Macroscopic examination at sacrifice revealed no evident adverse effects, including cutaneous lesions, musculoskeletal abnormalities, or gross hepatic abnormalities in mice with CH223191.

Fig. 3.

Therapeutic intervention of the AhR signaling pathway in Sox17^{+/p.Pro133Ala} mice. (A) Relative gene expression of cytochrome p450 1b1 (Cyp1b1). Right ventricle (RV)/body weight (BW) (B), Fulton index (C), right ventricular systolic pressure (RVSP) (D) were measured in wild-type and Sox17^{+/p.Pro133Ala} mice (hetero) with or without CH-223,191, respectively. (E) Images of the pulmonary arteries of mice observed by Elastica van Gieson staining. Scale bars, 50 μm. (F) Medial wall thickness (MWT) was also measured. ns = not significant. *p < 0.05.

Gene expression changes caused by the AhR antagonist

RNA-sequencing was performed on the lungs of mice treated with CH-223,191, and the gene expression were compared among four groups; wild type, wild type treated with CH-223,191, Sox17^{+/p.Pro133Ala}, and Sox17^{+/p.Pro133Ala} treated with CH-223,191. The effect of the AhR antagonist on the lungs of wild-type mice was limited to changes in the expressions of 91 genes, whereas in the Sox17^{+/p.Pro133Ala} mice, it affected 1712 genes (Fig. 4A). Next, we narrowed down these genes to those that were closely related to the onset of PAH because of the Sox17 heterozygous variant (Sox17^{+/p.Pro133Ala}) using the criteria described in the Methods section. We found 674 genes fit these criteria (Fig. 4B). Among them, genes including Bmpr2, Col4a1, and Col4a2, which were previously reported to vary in human PAH patients^22,24,25, were included, and changes in their expressions by the AhR antagonist were confirmed by qPCR (Fig. 4C-E). These results confirmed that Cyp1b1 fluctuated because of the Sox17 heterozygous variant (Sox17^{+/p.Pro133Ala}), and that the Bmpr2 and Col4a1/Col4a2 pathways, which are related to the pathology of PAH, were also changed.

Fig. 4.

Gene expressions associated with Sox17 variant and AhR signals. (A) The number of genes whose expression was changed by genotype and administration of CH-223,191 in RNA-sequencing using lung tissues. (B) The number of genes narrowed down by the criteria described in the method section. Upper panel: genes whose expression is decreased in Sox17^{+/p.Pro133Ala} mice (hetero) compared with wild-type mice, lower panel: genes whose expression is increased in Sox17^{+/p.Pro133Ala} mice compared with wild-type mice. Relative gene expressions of Bmpr2 (C), Col4a1 (D), and Col4a2 (E) measured by qPCR. ns = not significant. *p < 0.05. AhR, aryl hydrocarbon receptor; Bmpr2, bone morphogenetic protein receptor type 2; Col4a, collagen, type IV, alpha; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Discussion

In this study, mice carrying the heterozygous Sox17 c.397 C > G (p.Pro133Ala) variant, a single nucleotide variant identified in human PAH patients, developed PAH in a hypoxic environment. PAH was ameliorated by the administration of CH-223,191, suggesting that AhR signaling might be involved in the development of SOX17-associated PAH. Fluctuations in Bmpr2, Col4a1/Col4a2, and Cyp1b1 suggest that they interact with each other, based on the single base substitution of Sox17. The summary of this research is demonstrated in Fig. 5.

Fig. 5.

Graphical abstract. AhR, aryl hydrocarbon receptor; Bmpr2, bone morphogenetic protein receptor type 2; Col4a, collagen, type IV, alpha; Cyp1b1, cytochrome P450 family 1 subfamily B member 1; SOX17, SRY-box transcription factor 17. Created in https://BioRender.com.

The SOX17 c.397 C > G (Pro133Ala) variant is located within the HMG-box domain, a region critical for DNA binding and bending. Pathogenic variants within this domain (e.g. c.388 C > T, c.392 A > G, c.394 C > G, c.397 C > T, c.398 C > T) have previously been identified in patients with PAH^3,6, highlighting the functional importance of this region. In silico analysis by Gräf S, et al. demonstrated that HMG-box variants such as His132Asp and Pro133Ser were predicted to disrupt interactions between SOX17 and transcriptional co-factors, including Oct4 and beta-catenin³, suggesting that alterations in this domain may affect not only DNA binding but also protein–protein interactions critical for transcriptional regulation. Moreover, several prediction tools indicated a deleterious effect of the p.Pro133Ala variant (REVEL score 0.93; AlphaMissense 0.995; SIFT predicted deleterious), supporting the possibility of structural alteration. Although our molecular modeling suggested altered spatial relationships between the HMG-box and DNA, the direct functional impact of this variant on transcriptional regulation remains to be experimentally validated.

Cyp1b1 was previously reported to be increased in the pulmonary arteries of PAH model mice and human PAH patients²⁶. An increase in the RVSP was suppressed in Cyp1b1 knockout mice, and the use of a Cyp1b1 inhibitor rescued the right heart load and attenuated RVSP and vascular remodeling in PAH models^26,27. Furthermore, the upregulation of Cyp1b1 was confirmed in human idiopathic or heritable PAH patients^26,28. Thus, the onset of PAH is closely associated with increased expression of Cyp1b1. In this study, hypoxic loading in Sox17^{+/p.Pro133Ala} mice significantly increased the expression of Cyp1b1 compared with wild-type mice, and the inhibition of Cyp1b1 with CH-223,191, an AhR inhibitor, improved the PAH phenotype. It was reported that PAH patients with high AhR activity have a poor prognosis^29,30, and the increased expression of Cyp1b1, a downstream of AhR, observed in this study supports the idea that PAH develops in association with enhanced AhR-CYP1B1 signaling due to SOX17 variants. Our RNA-sequencing data did not establish whether the altered expression of Cyp1b1 represents direct transcriptional regulation by Sox17. Publicly available ChIP-Atlas dataset²³ found the SOX17 occupancy near the CYP1B1 locus in human datasets. Further study to elucidate the detailed mechanism by which the SOX17 variant mediates increased Cyp1b1 expression is warranted.

Regarding the association between SOX17 and BMPR2, prior studies have demonstrated that BMP9, an essential ligand for BMPR2 signaling, increases the expression of SOX17^31,32. Conversely, silencing of SOX17 has been reported to decrease BMPR2 expression³³. In our RNA-sequencing dataset, Bmp9 expression showed only a modest, non-significant reduction in Sox17 mutant mice; therefore, its involvement in our model remains speculative. Although these findings suggest potential interaction between these pathways, the precise regulatory relationship remains to be fully elucidated. Furthermore, the expressions of BMPR2, COL4A1, and COL4A2 were decreased in PAH patients, Col4a1 and Col4a2 were variably expressed in association with Bmpr2²³, and the decreased expressions of Col4a1 and Col4a2 impaired cell migration and adhesion²⁵. Moreover, loss-of-function variants of BMPR2 are associated with the development of PAH^32,33, and these results are consistent with our study data demonstrating low expressions of Bmpr2, Col4a1, and Col4a2 in hypoxia-fed Sox17^{+/p.Pro133Ala} mice. These findings suggest that multiple signal pathways that regulate Cyp1b1, Bmpr2, and Col4a1/Col4a2 are interconnected and affected by the single base substitution in Sox17 (Sox17^{+/p.Pro133Ala}), causing PAH in patients with SOX17 variants. Thus, the therapeutic intervention in these multiple signal pathways should be considered for PAH patients with SOX17 variants.

Several studies have reported the pathogenesis of PAH mediated by Sox17. One study reported that pulmonary endothelial cell-specific Sox17 knockdown induced PAH and E2F transcription factor 1 signaling leading to endothelial dysfunction³², whereas another study using endothelial-specific Sox17 knockout mice demonstrated that hepatocyte growth factor/c-Met signaling was involved in the pathogenesis of PAH⁷. In our study, these pathways were not significant in the lungs of Sox17^{+/p.Pro133Ala} mice, suggesting that the signal cascade resulting from whole Sox17 knockout may differ from that resulting from a single nucleotide change in Sox17, which is present in human PAH.

This study had several limitations. First, We investigated the effect of Sox17^{+/p.Pro133Ala} but did not examine other Sox17 variants, and thus it is unclear whether similar results would be obtained for other Sox17 variants. Second, the AhR antagonist may act on Cyp1b1 as well as other Cyp450 pathways, but this has not been verified. Research into the molecular biological relationship between Sox17 and other factors such as Bmpr2 should be performed. Third, although RNA-sequencing analysis identified multiple differentially expressed genes, our investigation was centered on CYP1B1. Thus, the contribution of other altered factors to the development of PAH cannot be ruled out. Fourth, while Cyp1b1 emerged as a prominent differentially expressed gene, whether Sox17 directly regulates transcription of Cyp1b1 requires further verification. Fifth, although molecular dynamics simulations suggested altered SOX17–DNA interactions, direct verification of DNA-binding affinity or transcriptional activity should be considered in the future. Finally, in the mice model generated in this study, we observed significant pulmonary vascular remodeling and right ventricular hypertrophy; however, the overall phenotype was relatively mild. Differences between this model and the clinical manifestations observed in patients with SOX17 variants remain an important subject for future investigation.

In conclusion, our findings indicate that introduction of a single base substitution in Sox17, corresponding to a variant reported in human PAH patients, is associated with the development of a mild pulmonary hypertension phenotype in mice. Furthermore, Sox17^{+/p.Pro133Ala} affected various pathways, including Cyp1b1, Bmpr2, Col4a1, and Col4a2, which are linked to each other in the development of PAH caused by SOX17 variants.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Yoshiki Shinya: Data curation, Formal analysis, Investigation, Validation, Writing – original draft. Takahiro Hiraide: Conceptualization, Data curation, Formal analysis, Funding acquisition, Methodology, Supervision, Validation, Writing – original draft. Mizuki Momoi: Data curation, Formal analysis, Investigation, Validation. Shinichi Goto: Data curation, Investigation, Methodology, Software, Visualization. Yoshinori Katsumata: Formal analysis, Project administration, Resources, Visualization, Writing – review & editing. Takumi Inami: Data curation, Formal analysis, Validation. Yoji Hakamata: Conceptualization, Investigation, Methodology, Project administration, Resources, Supervision. Masaharu Kataoka: Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing – review & editing. Masaki Ieda: Project administration, Resources, Supervision, Writing – review & editing.

Funding

This study was partially funded by Miyata foundation bounty for Pediatric cardiovascular research.

Data availability

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

Declarations

Competing interests

The authors declare no competing interests.