SIX1 mediates pulmonary arterial hypertension endothelial dysfunction through the IL-6/STAT3 axis

Abstract

Background

Pulmonary arterial hypertension (PAH) is a life-threatening and cardiovascular disease characterized by elevated pulmonary artery pressure leading to right ventricular hypertrophy. Its features include endothelial dysfunction and vascular remodeling. Inflammation is a key factor in PAH progression. This study focuses on the role and mechanism of sine oculis homeobox 1 (SIX1) in PAH through transcriptomics analysis, as well as validation in vivo and in vitro models, aiming to identify novel therapeutic strategies.

Methods

To analyze the transcriptomic data from mouse lung tissues (control vs. PAH), we employed an integrated strategy that combined Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment with weighted gene co-expression network analysis (WGCNA). This approach identified SIX1 as a key factor in PAH. We predicted potential binding sites between SIX1 and STAT3 using the Joint Academic Site for Protein And Regulatory (JASPAR) database. Through an integrated approach leveraging both in vivo and in vitro PAH models, we uncovered a pivotal role for SIX1 in disease pathogenesis, including hemodynamics, histopathology, protein expression (Western blot) and gene expression [reverse transcription quantitative polymerase chain reaction (RT-qPCR)], immunofluorescence (IF), and cellular phenotypes (knockdown, viability, migration, proliferation, apoptosis).

Results

Through linear regression analysis in the transcriptome sequencing data, we found that SIX1 expression positively correlated with clinical and pathological indices of PAH. Subsequently, KEGG pathway analysis revealed that STAT3 was enriched in multiple pathways. Using JASPAR, multiple binding sites between SIX1 and STAT3 were predicted. Consistent with our transcriptomic results, SIX1 was significantly induced in vivo, with corresponding increases observed not only in messenger RNA (mRNA) but also in protein, and IF staining confirmed its predominant localization in endothelial cells. In vitro, the inflammatory cytokine interleukin-6 (IL-6), a key driver of PAH, significantly induced SIX1 expression and activated STAT3 in human umbilical vein endothelial cells (HUVECs). Notably, SIX1 knockdown effectively reversed the IL-6-driven proliferation and migration of endothelial cells. This phenotypic rescue was accompanied by a significant inhibition of STAT3 phosphorylation and a disruption of SIX1-STAT3 protein colocalization.

Conclusions

We conclude SIX1 aggravates endothelial dysfunction via the IL-6/STAT3 axis, thereby identifying SIX1 as a therapeutic target for PAH.

Highlight box.

Key findings

• Sine oculis homeobox 1 (SIX1) is upregulated in pulmonary arterial hypertension (PAH) and correlates with disease severity.

• SIX1 mediates interleukin-6 (IL-6)-induced STAT3 activation and colocalization in endothelial cells.

• Knockdown of SIX1 reverses IL-6-driven endothelial proliferation, migration, and mesenchymal transition.

What is known and what is new?

• The IL-6/STAT3 axis is a critical driver of inflammation and vascular remodeling in PAH. SIX1 is a developmental transcription factor implicated in cancer and fibrosis.

• This study is the first to identify SIX1 as an upstream mediator linking IL-6 to STAT3 activation specifically in PAH endothelial cells, and to demonstrate that SIX1 knockdown ameliorates endothelial dysfunction.

What is the implication, and what should change now?

• SIX1 represents a novel therapeutic target for PAH. Future studies should explore endothelial-specific SIX1 inhibition as a strategy to attenuate vascular remodeling and right heart failure.

Introduction

Pulmonary arterial hypertension (PAH) a progressive disease with a poor long-term prognosis, characterized by persistent vasoconstriction and remodeling of the distal pulmonary arterioles. In 2021, the estimated global prevalence of PAH was 192,000 cases, with approximately 22,000 deaths attributed to the disease (1). According to the most recent clinical guidelines on PAH, a mean pulmonary artery pressure (mPAP) exceeding 20 mmHg at rest, as measured by right heart catheterization, is diagnostic of PAH (2). Although therapies targeting vasoactive pathways have improved clinical outcomes, PAH remains a progressive disease with high mortality, indicating that its molecular pathogenesis is not yet fully elucidated. Common mechanisms include small pulmonary vascular remodeling, inflammation and immune dysregulation. Among these, inflammatory mechanisms play a critical role in the initiation and progression of PAH.

Interleukin-6 (IL-6) is a circulating pro-inflammatory cytokine involved in the acute-phase response, hematopoiesis, and immune regulation. Its sustained dysregulation is associated with several chronic inflammatory diseases (3). In idiopathic PAH, high serum IL-6 concentration portends an unfavorable prognosis (4). Activation of the JAK-STAT3 pathway is a key downstream effect of IL-6 signaling, promoting Human Pulmonary Artery Smooth Muscle Cells (PASMCs) proliferation, migration, and resistance to apoptosis. It also exacerbates inflammatory responses, increases endothelial permeability and disrupts endothelial phenotype, and drives fibroblast-to-myofibroblast differentiation, thereby amplifying vascular remodeling (5-7). Evidence suggests that IL-6 receptor antagonists can improve hemodynamics, pulmonary and cardiac function, tissue remodeling, and inflammation in PAH (8). Following the administration of monocrotaline to rats, which recapitulates key features of PAH, increased pulmonary IL-6 downregulates BMPR2 and activates STAT3, stimulating PASMC proliferation (9). These findings highlight that the IL-6/STAT3 axis as a critical regulator orchestrating multiple facets of PAH pathology.

Sine oculis homeobox 1 (SIX1) is a key transcription factor during embryogenesis that participates in the morphogenesis of the lung and vascular smooth muscle. Upregulation of SIX1 has been reported in pulmonary fibrosis and in the reactivation of the cardiac fetal gene program. Cross-model evidence indicates that SIX1 activates STAT3 signaling, a process that requires EYA1, thereby promoting thyroid cancer cell proliferation and invasion (10), suggesting its potential role in regulating inflammatory and proliferative networks. SIX1 drives ZEB1 expression, enabling ZEB1 to function as a direct transcriptional activator of the IL-6 gene and thereby promote IL-6 production and release. The resulting IL-6 activates autocrine STAT3 signaling and promotes STAT3 phosphorylation, while secreted IL-6 can act in a paracrine fashion to trigger STAT3 activation and acquisition of stem-like traits in neighbouring cells with low SIX1 expression (11). In liver malignancy models, macrophage SIX1 induces IL-6 upregulation in hepatocellular carcinoma cells and enhances STAT3 activity, leading to increased matrix metalloproteinase-9 (MMP-9) levels and greater invasiveness. However, the role of SIX1 in PAH and whether it drives pulmonary vascular remodelling via the STAT3/IL-6 axis remains to be directly demonstrated. We present this article in accordance with the ARRIVE and MDAR reporting checklists (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0101/rc).

Methods

Animal model

We used 8–10-week-old male C57BL/6J mice purchased from Zhejiang Vital River Laboratory Animal Technology Co., Ltd. (Zhejiang, China). Use of males only was pre-specified to reduce hormonal variability in the SuHx model. Control mice were housed under specific pathogen-free (SPF) conditions in a barrier facility with ambient oxygen at 21% (22±2 ℃, 50–60% humidity, 12:12 h light-dark cycle). Animals had ad libitum access to food and water. Mice were randomly assigned to Control or SuHx groups, with group sizes specified in each figure legend. Investigators performing hemodynamics and histomorphometry were blinded to group allocation. In the SuHx group (SU5416 + hypoxia), mice received subcutaneous SU5416 (S2845, Selleck) at 20 mg/kg starting on day 1, once weekly for three weeks. After the first injection, animals were placed in a normobaric hypoxia chamber (RK-Ox-450L, RAYKEN) set to 10% O₂, 22 °C, and 50–60% relative humidity for four consecutive weeks. Throughout this period, hypoxia exposure was sustained for over 23 hours daily, with brief interruptions only for cage maintenance (e.g., feeding, bedding change) and the weekly drug administration. All other housing conditions were identical between the two groups (12). Mice were included if they were clinically healthy at baseline and completed the full SuHx or control protocol. No animals or data points were excluded from the analyses. Experiments were performed under a project license (No. 197521) granted by the Animal Care and Use Committee of Soochow University, in compliance with relevant national or institutional guidelines for the care and use of animals. A protocol was prepared before the study without registration.

Reagents and cell culture

A working stock of recombinant human IL-6 (C610007, Sangon Biotech, Shanghai, China) was prepared in sterile phosphate-buffered saline (PBS). For leukemia inhibitory factor (LIF) stimulation, cells were treated with 30 ng/mL recombinant human LIF (Catalog No. 300-05, PeproTech, Suzhou, China) for 24 h. Vehicle control cells received an equivalent volume of PBS. For IL-6 neutralization, recombinant human IL-6 (C610007, Sangon Biotech) at 100 ng/mL was pre-incubated with or without 10 µg/mL of neutralizing antibody against human IL-6 (MAB206, R&D Systems, Shanghai, China) for 1 h at 37 ℃ to allow immune complex formation. The mixture was then added to the cells and incubated for 24 h. Human umbilical vein endothelial cells (HUVECs), obtained from the American Type Culture Collection (ATCC), were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), L-glutamine and antibiotics at 37 ℃ with 5% CO₂ and high humidity.

Hemodynamic assessment

Throughout the study, a standard laboratory diet and water were provided for the animals. At the end of modeling, right ventricular systolic pressure (RVSP) was assessed under inhalational isoflurane anesthesia (2–3%). RVSP was measured using right ventricular catheterization (Millar Instruments, SPF1030, Shanghai, China) through the external jugular vein. RVSP was recorded with LabChart 8 software (ADInstruments, Australia), and the value for each mouse was documented. Following pressure recording, heart and lung tissues were collected after the mice were euthanized. The degree of right ventricular hypertrophy was assessed by calculating the ratio of the right ventricular (RV) weight to the combined weight of the left ventricle (LV) and septum (S) [RV/(LV + S)], known as the Fulton index [right ventricular hypertrophy index (RVHI)] (13). For molecular analyses, the heart and the lung tissues were perfused with PBS, then rapidly snap-frozen in liquid nitrogen for subsequent Western blotting and quantitative RT-qPCR. For morphological examination, tissues were fixed by immersion in 4% neutral-buffered formalin for 72 hours prior to analysis. Sample size was chosen based on our previous experience with the SuHx model and published studies using similar endpoints.

Hematoxylin-eosin (H&E) staining

Fixed tissues were processed into paraffin blocks, from which 5-µm-thick sections were cut and mounted on slides. H&E staining was performed to evaluate the thickness of the pulmonary vasculature. Images were acquired using a PDone panoramic whole-slide scanner (Bainuo), and morphometry was performed with OlyVIA software. For each small pulmonary artery, the external and internal diameters as well as the external and luminal areas were measured. Medial wall thickness was calculated as: thickness = (outer diameter – inner diameter)/outer diameter (14). Based on external diameter, vessels were stratified into two groups: <50 µm and 50–100 µm. For each mouse, five randomly selected fields from lung sections were analyzed, and the mean vessel wall thickness per mouse was calculated and used for statistical analysis.

Masson’s trichrome staining

Fixed tissues were processed into paraffin blocks, from which 5-µm-thick sections were cut and mounted on slides. Sections were stained with Masson’s trichrome to assess perivascular collagen deposition (15). Images were captured using a PDone panoramic whole-slide scanner (Bainuo). Slides were reviewed in OlyVIA software, and perivascular collagen deposition in small pulmonary arterioles was quantified using Image J. The percentage of collagen deposition was calculated as: (collagen fiber area/total vessel area) × 100%. For each mouse, five randomly selected fields from lung sections were analyzed, and the mean perivascular collagen deposition per mouse was calculated and used for statistical analysis.

Immunofluorescence (IF)

Fixed mouse lung tissue was sectioned. HUVECs were grown on glass coverslips and fixed with ice-cold methanol. After deparaffinization, using a microwave oven for low-temperature thermal processing in sodium citrate buffer (pH 6.0) for 30 mins, quenched endogenous peroxidase by treating the samples with 3% H₂O₂, permeabilization with 0.3% Triton X-100, and blocking with 3% goat serum for 30 mins. Following blocking, select an appropriate concentration of primary antibody and incubated with the sections overnight at 4 °C. The secondary horseradish peroxidase (HRP)-conjugated antibodies were then added, and tyramide signal amplification (TSA) was performed according to the manufacturer’s protocol. For multiplex staining, a cyclic strategy was used: brief re-retrieval in pH 6.0 sodium citrate buffer between cycles, followed by repetition of the secondary/Tyramide Signal Amplification (TSA) steps. Nuclei were labeled using 4',6-diamidino-2-phenylindole (DAPI) (C1002, Beyotime, Shanghai, China). Images were acquired using a PDone panoramic whole-slide scanner and reviewed in OlyVIA; quantification and colocalization (Pearson’s r, overlap R) were performed in Image J. Colocalization analysis was performed using Image J software. For each acquired image, the multichannel stack was split into individual fluorescence channels. The nuclear channel (DAPI, blue) was discarded, while the red and green channels corresponding to the markers of interest were retained and converted to 8-bit grayscale images. Pseudo-colors were then applied to the grayscale images to facilitate visualization. Colocalization analysis was conducted using the “Colocalization Finder” plugin. Two colocalization metrics were calculated: Pearson’s correlation coefficient (Pearson’s R), which measures the strength of linear correlation between pixel intensities in the two channels, with values closer to 1 indicating a stronger positive correlation. Manders’ overlap coefficient (Overlap R), which represents the degree of colocalization independent of signal intensity, ranging from 0 to 1, where values >0.6 are generally considered to indicate significant colocalization.

For each mouse, five randomly selected fields from lung sections were analyzed, and the mean colocalization coefficient across these fields was calculated and used for subsequent statistical analysis. Primary antibodies: SIX1 (10709-1-AP, Proteintech, Wuhan, China), STAT3 (#9132, CST, Shanghai, China), and CD31 (ab222783, Abcam, Shanghai, China) (16).

Reverse transcription quantitative polymerase chain reaction (RT-qPCR)

Total RNA was isolated from tissues using Trizol reagent, and from cultured cells using the MicroElute Total RNA Kit (R6831, Omega, Shanghai, China). Complementary DNA (cDNA) was synthesized from RNA using an applicable reverse transcription kit (E047-01B, Novoprotein, Suzhou, China). Gene expression was assessed by RT-qPCR using the SYBR Green Master Mix (B21202, Bimake, Shanghai, China) on a Roche LightCycler 480 II system. Expression levels were normalized to β-actin. Primer information is provided in the Appendix 1.

Western blotting

Protein lysates were prepared from harvested cell pellets or mouse tissues. Cell pellets were lysed in lysis buffer (18 mM Tris-HCl, 1.5% Triton X-100, 120 mM NaCl, pH 7.0, 10% glycerol, 0.5 mM Na₃VO₄, 25 mM β-glycerophosphate) containing 0.1 mM PMSF and protease/phosphatase inhibitor cocktails (P1006, P1082, Beyotime), and then incubated on ice for 30 min. Mouse tissues were homogenized and resuspended in the same buffer, and kept on ice. Samples were vortexed for 10 s and incubated on ice for 30 mins, then clarified by centrifugation at 12,000 rpm for 15 mins. Supernatants were subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting. Primary antibodies included SIX1 (10709-1-AP, Proteintech), β-actin (PSH-03-63, HUABIO), phospho-STAT3 (9145, CST), VE-cadherin (27956-1-AP, Proteintech), VCAM-1 (SA05-04, HUABIO), α-SMA (14968, CST) and STAT3 (9132, CST). Band intensities were quantified using Image J.

Small interfering RNA (siRNA) transfection

HUVECs were seeded the day before transfection. HUVECs were introduced with SIX1-targeting siRNA using a transfection reagent (siSIX1; 90 nM) or a GC content-matched stealth negative control duplex (siNC; 60 nM) using the siRNA transfection kit (A10001, GenePharma, Suzhou, China), as described by the supplier. The concentrations of siSIX1 (90 nM) and siNC (60 nM) were based on the manufacturer’s optimized transfection protocols for different siRNA sequences. Throughout the transfection experiments, no overt morphological changes or reduction in cell viability was observed under light microscopy in any of the transfection conditions, indicating that the transfection procedure itself did not induce cytotoxicity. The siRNA sequences are provided in the Appendix 1.

Cell viability and proliferation assays

HUVECs were seeded in 96-well plates 10 hours prior to experimentation, and subjected to standard treatment protocols. Cell viability was determined from cellular adenosine triphosphate (ATP) using the CellTiter-Glo Luminescent (G7570, Promega, Beijing, China) (17). Luminescence was measured on a SpectraMaxi 3x multimode microplate reader (Molecular Devices). Cell proliferation was additionally assessed with the manufacturer’s protocol provided in the 5-ethynyl-2'-deoxyuridine (EdU) cell proliferation kit manual (product number: C0071S, Beyotime) (18). Briefly, cells (2×10⁴ per well) were plated in 24-well plates, exposed to the indicated stimuli for 24 hours, then incubated with 2× EdU working solution for 90 mins prior to fixation and staining.

Wound-healing (scratch) assay

For wound-healing assays, cells were passaged in 6-well plates until the HUVECs reached 100% confluence. A linear scratch was created in the HUVECs monolayer with a 200 µL pipette tip. Subsequently, the culture medium was removed by aspiration, and the HUVECs were washed with sterile PBS. HUVECs were then treated with IL-6 at the indicated concentrations for 24 hours. Wound closure was imaged every 12 hours using an inverted microscope.

Transferase-mediated deoxyuridine triphosphate-nick end labelling (TUNEL) assay

Apoptosis was detected using a one-step TUNEL apoptosis assay kit (Beyotime, C1088) according to the manufacturer’s instructions. Briefly, HUVECs were seeded on coverslips in 24-well plates and subjected to the indicated treatments. Cells were then fixed with 4% paraformaldehyde for 30 min at room temperature, followed by permeabilization with 0.1% Triton X-100 in PBS for 5 min on ice. After washing with PBS, cells were incubated with TUNEL reaction mixture for 50 min at 37 ℃ in a humidified chamber in the dark; TUNEL-positive cells exhibited green fluorescence. Images were captured directly under a fluorescence microscope connected to a computer. For each condition, five randomly selected fields were analyzed, and the percentage of TUNEL-positive cells was calculated as (number of TUNEL⁺ cells/total number of cells) × 100%. All experiments were performed in triplicate with at least three independent biological replicates.

Data sources and preprocessing

We obtained transcriptome [RNA sequencing (RNA-seq)] data from lung tissue samples collected from PAH model mice. Raw sequencing reads were quality-controlled and adapter-trimmed using Cutadapt, aligned to the mouse reference genome using STAR, and summarized into gene-level count matrices using HTSeq-count. Gene-level count data were imported into R and organized for downstream analyses. Lowly expressed genes were filtered based on counts per million (CPM), retaining genes expressed above the predefined threshold in at least 25% of samples. Differential expression analysis was then performed in DESeq2. Library size and composition biases were corrected using DESeq2 size-factor normalization, and normalized expression values were transformed to log₂CPM for visualization and exploratory analyses. All analyses were conducted in R 4.3.1.

Differential expression and Gene Set Variation Analysis (GSVA)

Differential expression between groups was assessed using DESeq2 with a group-based design matrix. Results were extracted using the specified contrast, and genes with FDR <0.05 were considered significantly differentially expressed. For visualization and descriptive summaries, normalized counts were further represented on the log₂CPM scale.

Weighted gene co-expression network analysis (WGCNA)

A weighted co-expression network was constructed from the normalized expression matrix using the WGCNA package in R (R 4.3.1; WGCNA 1.72.1). The soft-thresholding power (β) was set to 8 based on the scale-free topology criterion. The adjacency matrix was computed and transformed into a topological overlap matrix (TOM); 1 − TOM was used as the dissimilarity metric for hierarchical clustering to enhance the robustness of module detection. Modules were identified with the dynamicTreeCut algorithm (hybrid cutting method) with a minimum module size of 20 (minModuleSize =20). Module eigengenes (MEs) were calculated, and modules with highly correlated MEs were merged according to eigengene correlation distance. Pearson correlations between MEs and clinical traits (e.g., RVSP, collagen deposition) were computed to obtain correlation coefficients and corresponding P values.

Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis

Genes from WGCNA modules were subjected to KEGG pathway enrichment using the black module gene set as the background. Gene identifiers were first mapped to Entrez Gene IDs with org.Mm.eg.db, and results were converted back to gene symbols for presentation. Over-representation analysis (ORA) was conducted with clusterProfiler (≥4.0), using a full gene list ordered by the relevant statistic as input. Significant pathways were visualized with setReadable and the barplot functions.

Quantification and statistical analysis

Statistical analyses were conducted with GraphPad Prism software (version 10.1.2.; GraphPad Software, La Jolla, CA, USA). All data are presented as the mean ± standard deviation (SD). All quantitative data were first tested for normality using the Shapiro-Wilk test and for homogeneity of variance using the F-test. The results confirmed that the data were normally distributed (P>0.05) and met the assumption of equal variances (P>0.05), justifying the use of parametric tests. Accordingly, comparisons between two groups were performed using unpaired two-tailed Student’s t-test, and comparisons among multiple groups were performed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test for multiple comparisons. A P-value below 0.05 was considered statistically significant (*, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001; ns, not significant). Unless otherwise stated, all experiments included at least three independent biological replicates.

Results

WGCNA identifies a SIX1-associated module linked to PAH-related phenotypes

Initial transcriptomic analyses showed clear separation between control and PAH samples and broad pathway activation in PAH, particularly in immune-, adhesion-, and membrane-associated processes (Figure S1). We performed WGCNA to further identify coordinated gene networks associated with disease. Sample-level hierarchical clustering with trait annotation demonstrated a clear separation between control and PAH samples and showed consistent between-group differences in RVSP, pulmonary artery thickness, collagen fiber deposition, and RV/(LV + S), indicating concordance between transcriptomic variation and disease phenotypes (Figure 1A). WGCNA identified multiple co-expression modules associated with PAH-related traits, among which the black module exhibited the strongest positive correlations with RVSP, pulmonary artery thickness, collagen area, and RV/(LV + S) (Figure 1B). Within this module, SIX1 ranked among the highly connected genes by module membership, supporting its close association with a disease-relevant co-expression network (Figure 1C). Moreover, SIX1 expression was positively correlated with RVSP and collagen fiber deposition, further linking SIX1 to disease severity in this dataset (Figure 1D). Consistent with this association, KEGG enrichment analysis stratified by SIX1 expression highlighted inflammatory and immune-related pathways, including chemokine signaling, cytokine-related signaling, phagosome, and antigen-processing pathways (Figure S1). Furthermore, integrative Gviz/JASPAR motif-track visualization identified putative SIX1 and Stat3 motif distributions in nearby genomic regions, suggesting possible parallel regulatory involvement at selected loci; however, this observation remains predictive and requires experimental validation (Figure S1).

Figure 1.

WGCNA identifies a SIX1-associated co-expression module linked to PAH-related phenotypes. (A) Hierarchical clustering dendrogram of control and PAH lung samples with corresponding trait heatmap. Phenotypic variables shown include RVSP, PAT, CFD, and right ventricular hypertrophy index [RV/(LV + S)]. (B) Heatmap of module-trait relationships derived from WGCNA. Each cell contains the Pearson correlation coefficient and corresponding P value. The black module exhibited the strongest positive correlations with PAH-related traits. (C) Module membership of representative genes in the black module, with SIX1 highlighted. (D) Scatter plots showing positive correlations of SIX1 expression with RVSP and CFD. CFD, collagen fiber deposition; LV, left ventricle; PAH, pulmonary arterial hypertension; PAT, pulmonary artery thickness; RV, right ventricular; RVSP, right ventricular systolic pressure; S, septum; WGCNA, weighted gene co-expression network analysis.

Characteristics of the SuHx-induced mouse model of PAH and elevated IL-6 expression

We comprehensively characterized the SuHx-induced pulmonary hypertension mouse model and found that it successfully recapitulated the classic pathological features of PAH. Specifically, mice in the SuHx group developed a significant increase in RVSP (Figure 2A,2B; ****, P<0.0001) along with a higher right ventricular hypertrophy index [RV/(LV + S), Fulton index; Figure 2C; ****, P<0.0001], indicating right ventricular dysfunction resulting from elevated pulmonary pressure. Histological examination of lung tissues by H&E staining, including low-magnification images to show overall lung architecture, revealed marked medial wall thickening and luminal narrowing in small pulmonary arteries (diameter <50 and 50–100 µm; Figure 2D). Quantitative analysis, with each data point representing the mean value from an individual mouse, confirmed a significant increase in vascular wall thickness in the SuHx group compared to the Control group (Figure 2E,2F; *, P<0.05; **, P<0.01). Wall thickness was calculated as (outer diameter – inner diameter)/outer diameter. Furthermore, Masson’s trichrome staining detected significantly increased collagen deposition around the vessels in SuHx-treated mice (Figure 2G,2H, **, P<0.01), as quantified by collagen volume fraction [(collagen fiber area/total vessel area) × 100%]. Each data point represents the mean value from an individual mouse. Finally, to assess the inflammatory response, we measured IL-6 mRNA levels and found them to be significantly upregulated in the lung tissues of the SuHx group (Figure 2I; **, P<0.01).

Figure 2.

SuHx group induces pulmonary arterial hypertension and increased mRNA expression of IL-6. (A,B) RVSP was significantly higher in the pulmonary hypertension group (SuHx) compared to the control group (Control) (n=6; ****, P<0.0001). (C) The right ventricular hypertrophy index [RV/(LV + S), Fulton index] was also significantly increased in the SuHx group relative to the Control group (n=6; ****, P<0.0001). (D) H&E staining revealed marked vessel wall thickening and lumen narrowing in small pulmonary arteries with diameters <50 µm and 50–100 µm in the SuHx group. Black arrows indicate vascular structures. Scale bars: 20 µm, 100 µm. (E,F) Quantitative analysis of vascular thickness in all mice was performed using Image J software. The intimal thickness was calculated using the formula: (outer diameter – inner diameter)/outer diameter. The bar graph shows pulmonary artery thickness (µm) on the y-axis, with the Control and SuHx groups on the x-axis; each dot represents the mean vascular thickness of an individual mouse. Pulmonary artery thickness was significantly greater in the SuHx group than in the Control group (n=6; *, P<0.05; **, P<0.01). (G,H) Collagen deposition in pulmonary vessels was assessed by Masson staining. Black arrows indicate vessels, and blue areas represent collagen fibers. The percentage of collagen deposition was calculated as: (collagen fiber area/total vessel area) × 100%. Quantitative analysis of perivascular collagen deposition in all mice was performed using Image J software. Each dot represents the mean percentage of collagen deposition in an individual mouse. The results showed that the percentage of perivascular collagen deposition was significantly higher in the SuHx group than in the Control group (n=6; **, P<0.01). Scale bars: 20 µm, 100 µm. (I) RT-qPCR analysis of lung tissue showed that IL-6 mRNA expression levels were significantly higher in the SuHx group than in the Control group (n=6; **, P<0.01). Statistical analysis: all quantitative data were obtained from 6 independent mice. Data were analyzed using GraphPad Prism 10.1.2 software. Data are presented as mean ± standard deviation. Normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the F-test. The results showed that the data conformed to a normal distribution (P>0.05) and exhibited homogeneity of variance (P>0.05). Comparisons between two groups were performed using an independent samples t-test (two-tailed), and P<0.05 was considered statistically significant. H&E, hematoxylin-eosin; IL-6, interleukin-6; LV, left ventricle; mRNA, messenger RNA; RT-qPCR, reverse transcription quantitative polymerase chain reaction; RV, right ventricular; RVSP, right ventricular systolic pressure; S, septum.

Coordinated activation of SIX1 and STAT3 in pulmonary endothelial cells of SuHx mice (Figure 3)

Figure 3.

Co-activation of SIX1 and STAT3 in lung endothelial cells of SuHx group. (A,B) Western blot and RT-qPCR analyses of lung tissue samples showed that SIX1 expression levels were significantly higher in the SuHx group than in the Control group (n=5–6, **, P<0.01; ****, P<0.0001). (C,D) Western blot analysis revealed that the protein level of p-STAT3 was markedly higher in the SuHx group compared to the Control group (n=6, **, P<0.01). (E) IF staining was performed to detect the colocalization of SIX1 (green) with CD31⁺ endothelial cells (red) in the Control and SuHx groups. Nuclei were stained with DAPI (blue), and white arrows indicate colocalized cells. Scale bar: 10 μm. (F) Quantitative analysis of colocalized cells in all mice was performed using Image J software. The bar graphs display the Pearson’s correlation coefficient (Pearson’s R) and the overlap coefficient (Overlap R) for SIX1 and CD31, each dot represents the mean colocalization rate of colocalized cells in an individual mouse. Pearson’s R reflects the linear correlation of fluorescence intensity distributions between the two channels, with values closer to 1 indicating a stronger correlation. Overlap R represents the actual proportion of overlapping signals between SIX1 and CD31, with values closer to 1 indicating a higher degree of colocalization. The colocalization rate of SIX1 with CD31 in pulmonary vessels was significantly higher in the SuHx group than in the Control group (n=6; **, P<0.01; ***, P<0.001). (G) IF staining was performed to detect the colocalization of STAT3 (green) with CD31⁺ endothelial cells (red) in the Control and SuHx groups. Nuclei were stained with DAPI (blue), and white arrows indicate colocalized cells. Scale bar: 10 μm. (H) Quantitative analysis of colocalized cells in (G) was performed using Image J software, following the same method as described in (F). The results showed that the colocalization rate of STAT3 with CD31 in pulmonary vessels was significantly higher in the SuHx group than in the Control group (n=6; **, P<0.01; ***, P<0.001). Statistical analysis: all quantitative data were obtained from 5–6 independent mice (specific sample sizes are detailed in the respective figure legends). Data were analyzed using GraphPad Prism 10.1.2 software. Data are presented as mean ± standard deviation. Normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the F-test. The results showed that the data conformed to a normal distribution (P>0.05) and exhibited homogeneity of variance (P>0.05). Comparisons between two groups were performed using an independent samples t-test (two-tailed), and P<0.05 was considered statistically significant. DAPI, 4',6-diamidino-2-phenylindole; IF, immunofluorescence; p-STAT3, phosphorylated STAT3; RT-qPCR, reverse transcription quantitative polymerase chain reaction.

Western blot and RT-qPCR analyses revealed that SIX1 expression was significantly upregulated in the SuHx group at both the protein and mRNA levels compared to the Control group (Figure 3A,3B; **, P<0.01; ****, P<0.0001). To determine the cellular localization of SIX1, IF staining was performed on lung sections. Colocalization analysis showed that SIX1 was predominantly localized in CD31⁺ endothelial cells, with significantly higher Pearson’s R and Overlap R coefficients in the SuHx group compared to controls (Figure 3E,3F; **, P<0.01; ***, P<0.001). Each data point represents the mean value from an individual mouse (n=6). These results indicate that pulmonary vascular endothelial cells are the primary site of SIX1 upregulation. Given the established role of IL-6/STAT3 signaling in PAH, we next examined STAT3 activation in the same model. Western blot analysis revealed that phosphorylated STAT3 (p-STAT3) protein levels were significantly elevated in the SuHx group (Figure 3C,3D, **, P<0.01). IF staining demonstrated that STAT3 was also predominantly localized in CD31⁺ endothelial cells. Quantitative colocalization analysis confirmed significantly higher Pearson’s R and Overlap R coefficients in the SuHx group compared to controls (Figure 3G,3H, **, P<0.01; ***, P<0.001), with each dot representing an individual mouse (n=6). To further confirm STAT3 activation in endothelial cells, we performed IF staining for p-STAT3. As shown in Figure S2, p-STAT3 was significantly colocalized with CD31⁺ endothelial cells in the SuHx group compared to controls (**, P<0.01; ***, P<0.001), providing direct evidence of STAT3 phosphorylation within the pulmonary endothelium. To verify the cell-type specificity of SIX1 and STAT3 expression, we performed co-staining with markers for smooth muscle cells (α-SMA), lymphocytes (CD3), and myeloid cells (CD11B). As shown in Figure S3, no significant colocalization was observed between SIX1 or STAT3 and any of these cell types in either the Control or SuHx groups (ns), confirming that the upregulation and activation of SIX1 and STAT3 in SuHx lungs are specifically localized to endothelial cells. Collectively, these data demonstrate that SuHx challenge induces the expression and activation of both SIX1 and STAT3 specifically within pulmonary microvascular endothelial cells. The endothelial-specific upregulation of SIX1, along with the activation and phosphorylation of STAT3 in the same cell population, suggests a potential functional interaction that may contribute to the vascular remodeling and endothelial dysfunction characteristic of this PAH model. The negative results from cell-type specificity experiments (Figure S3) further support that these effects are endothelial-specific.

SIX1 mediates IL-6-induced STAT3 activation and interaction in HUVECs

We next investigated whether IL-6 directly regulates SIX1 and STAT3 in HUVECs. Using HUVECs, we found that IL-6 induced SIX1 expression and activated STAT3 signaling, which was significantly attenuated by SIX1 knockdown. CellTit er-Glo assay showed that IL-6 stimulated HUVEC proliferation in a concentration-dependent manner, with the most significant effect observed at 100 ng/mL (Figure 4A, ****, P<0.0001). Proliferation rates were normalized to the NC group and expressed as “Cell proliferation (% of NC)”. RT-qPCR and Western blot analyses confirmed that IL-6 (100 ng/mL) significantly upregulated SIX1 expression at both the mRNA and protein levels compared to the NC group (Figure 4B,4C, **, P<0.01, *, P<0.05). To establish the role of SIX1 in IL-6-induced signaling, we transfected HUVECs with three independent siRNAs targeting SIX1. Among these, siSIX1-3 exhibited the highest knockdown efficiency at both protein and mRNA levels (Figure 4D,4E, *, P<0.05; ****, P<0.0001). Western blot analysis showed that IL-6 stimulation significantly increased SIX1 expression and STAT3 phosphorylation (p-STAT3) compared to the NC group (Figure 4F,4G, **, P<0.01). IF analysis revealed that SIX1 and STAT3 colocalized in HUVECs. IL-6 stimulation enhanced this colocalization, as quantified by Pearson’s R and Overlap R coefficients (Figure 4H,4I, **, P<0.01; ***, P<0.001). Conversely, SIX1 knockdown significantly reduced the colocalization signal (IL-6 + siSIX1 vs. IL-6 + siNC; **, P<0.01; ***, P<0.001). Consistent with the Western blot results, anti-IL-6 + IL-6 treatment also significantly diminished the colocalization (**, P<0.01; ***, P<0.001).

Figure 4.

SIX1 mediates IL-6-induced STAT3 activation and interaction in HUVECs. (A) CellTiter-Glo assay was used to assess the proliferation rate of HUVECs stimulated with different concentrations of IL-6 (0–200 ng/mL) for 24 h. The results showed that, compared to the NC group, IL-6 at 100 ng/mL exhibited the most significant pro-proliferative effect (****, P<0.0001). (B) RT-qPCR results indicated that, compared to the NC group and other IL-6 concentrations, the mRNA expression level of SIX1 was highest in the IL-6 (100 ng/mL) group (**, P<0.01). (C) Western blot analysis of SIX1 protein expression in HUVECs treated with NC or IL-6 (100 ng/mL, 24 h). Compared with the NC group, IL-6 stimulation significantly upregulated SIX1 protein expression (*, P<0.05). Subsequent experiments were performed using IL-6 at 100 ng/mL. (D,E) The knockdown efficiency of SIX1 was assessed by Western blot and RT-qPCR. Compared with IL-6 combined with siRNA negative control (IL-6 + siNC), the IL-6 + SIX1 siSIX1-3 (IL-6 + siSIX1-3) group exhibited the highest knockdown efficiency (*, P<0.05; ****, P<0.0001). (F,G) Western blot analysis of SIX1, total STAT3, p-STAT3, and β-actin protein expression in HUVECs under different treatment conditions: NC, vehicle control, IL-6 (100 ng/mL, 24 h), IL-6 combined with siNC (IL-6 + siNC), IL-6 combined with SIX1 siRNA (IL-6 + siSIX1), LIF stimulation (30 ng/mL, 24 h) as an alternative activator of the JAK-STAT pathway, and IL-6 neutralizing antibody pretreatment followed by IL-6 stimulation (anti-IL-6 + IL-6). Quantitative results showed that compared to the NC group, IL-6 significantly increased the expression of SIX1 and p-STAT3 (**, P<0.01); compared to the IL-6 + siNC group, the IL-6 + siSIX1 group exhibited decreased p-STAT3 and SIX1 protein expression (*, P<0.05; ***, P<0.001); compared to the IL-6 group, the anti-IL-6 + IL-6 group also showed significantly reduced SIX1 and p-STAT3 expression (**, P<0.01). (H) IF analysis of SIX1 (green) and total STAT3 (red) colocalization in HUVECs under different treatment conditions. Nuclei were stained with DAPI (blue). Merged images show colocalization signals in white. Scale bar: 10 μm. (I) Quantitative analysis of colocalization signals using Image J software. Pearson’s R and Overlap R were used to evaluate the degree of colocalization, following the same method as in (F). The results showed that compared to the NC group, the IL-6 group exhibited significantly increased colocalization of SIX1 and STAT3 (**, P<0.01; ***, P<0.001); compared to the IL-6 + siNC group, the IL-6 + siSIX1 group showed significantly reduced colocalization (**, P<0.01; ***, P<0.001); compared to the IL-6 group, the anti-IL-6 + IL-6 group also exhibited significantly reduced colocalization (**, P<0.01; ***, P<0.001). Statistical analysis: all quantitative data were obtained from three independent experiments, each performed in triplicate. Each data point in the figures represents the mean of one independent experiment (n=3). Data analysis was performed using GraphPad Prism 10.1.2 software. All data are presented as mean ± standard deviation. Normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the F-test. The results confirmed that the data followed a normal distribution (P>0.05) and exhibited homogeneous variances (P>0.05). Comparisons between two groups were conducted using an independent samples t-test (two-tailed). Multiple group comparisons were performed using one-way ANOVA, followed by Tukey’s multiple comparison test for pairwise comparisons when overall significance was detected. A P value <0.05 was considered statistically significant. ANOVA, analysis of variance; DAPI, 4',6-diamidino-2-phenylindole; HUVEC, human umbilical vein endothelial cell; IF, immunofluorescence; IL-6, interleukin-6; LIF, leukemia inhibitory factor; mRNA, messenger RNA; NC, negative control; p-STAT3, phosphorylated STAT3; RT-qPCR, reverse transcription quantitative polymerase chain reaction; siNC, negative control siRNA; siRNA, small interfering RNA; siSIX1-3, the third of three independent SIX1-targeting siRNAs.

SIX1 is a critical mediator of IL-6-induced endothelial dysfunction

We next investigated the functional role of SIX1 in IL-6-induced endothelial dysfunction. Using HUVECs, we assessed proliferation, migration, EndoMT-related marker expression, and apoptosis under different treatment conditions (Figure 5). CellTiter-Glo assay showed that IL-6 stimulation significantly increased HUVEC proliferation compared to the NC group (Figure 5C; ****, P<0.0001), with proliferation rates normalized to NC and expressed as “Cell Proliferation (% of NC)”. This effect was significantly attenuated by SIX1 knockdown (IL-6 + siSIX1 vs. IL-6 + siNC; ****, P<0.0001). Consistently, EdU assay revealed that the percentage of EdU-positive proliferating cells was significantly higher in the IL-6 group than in the NC group (Figure 5A,5D; ***, P<0.001), and this increase was suppressed by SIX1 knockdown (**, P<0.01). Wound healing assay demonstrated that IL-6 stimulation significantly enhanced HUVEC migration after 24 h compared to the NC group (Figure 5B,5E; **, P<0.01). This pro-migratory effect was markedly reduced by SIX1 knockdown (IL-6 + siSIX1 vs. IL-6 + siNC; **, P<0.01). To further evaluate the role of SIX1 in PAH-relevant endothelial dysfunction, we examined EndoMT-related markers and adhesion molecule expression by Western blot. As shown in Figure S4A,S4B, IL-6 stimulation significantly upregulated the expression of SIX1, VCAM-1 (adhesion molecule), and α-SMA (mesenchymal marker) compared to the NC group (*, P<0.05; **, P<0.01). In contrast, the endothelial marker VE-cadherin showed no significant change (ns). Knockdown of SIX1 significantly reversed the IL-6-induced upregulation of VCAM-1 and α-SMA (*, P<0.05; ***, P<0.001), while VE-cadherin remained unchanged (ns). These results suggest that SIX1 mediates IL-6-induced endothelial-to-mesenchymal transition and adhesion molecule expression. To determine whether apoptosis contributes to these effects, TUNEL staining was performed. Quantitative analysis showed no significant differences in apoptotic cell numbers among any of the treatment groups (Figure S4C,S4D; ns), indicating that IL-6 stimulation and SIX1 knockdown do not affect HUVEC apoptosis under the current experimental conditions. This negative result suggests that the observed effects of IL-6 and SIX1 on proliferation, migration, and EndoMT marker expression are not confounded by changes in cell survival. Taken together, these data demonstrate that SIX1 mediates IL-6-induced endothelial proliferation, migration, adhesion molecule expression, and mesenchymal transition in HUVECs. No significant changes in apoptosis were observed under these experimental conditions. These findings suggest that SIX1 may contribute to IL-6-induced endothelial dysfunction and could represent a potential therapeutic target for vascular remodeling in PAH.

Figure 5.

SIX1 promotes IL-6-induced excessive proliferation and migration of HUVECs. (A) EdU assay was performed to assess the proliferation of HUVECs. Proliferating cells were labeled with EdU (green), and nuclei were stained with DAPI (blue). Scale bar: 100 μm. Quantitative analysis of (A) shown in (D) revealed that compared to the NC group, IL-6 stimulation significantly promoted cell proliferation (***, P<0.001), whereas knockdown of SIX1 (IL-6 + siSIX1) significantly suppressed the IL-6-induced proliferative effect (compared to the IL-6 + siNC group, **, P<0.01). (B) Wound healing assay was performed to evaluate the migration ability of HUVECs. Yellow lines indicate the wound edges. Scale bar: 200 μm. Quantitative analysis of (B) shown in (E) demonstrated that compared to the NC group, IL-6 stimulation significantly promoted cell migration (**, P<0.01), whereas knockdown of SIX1 significantly suppressed the IL-6-induced migratory effect (compared to the IL-6 + siNC group; **, P<0.01). (C) CellTiter-Glo assay was used to measure cell proliferation rate. The results showed that compared to the NC group, HUVECs in the IL-6 group exhibited a higher proliferation rate (****, P<0.0001), whereas knockdown of SIX1 significantly attenuated this effect (compared to the IL-6 + siNC group, ****, P<0.0001), further confirming that SIX1 knockdown inhibits HUVEC proliferation. Statistical analysis: all quantitative data were obtained from three independent experiments, each performed in triplicate. Each data point in the figures represents the mean of one independent experiment (n=3). Data analysis was performed using GraphPad Prism 10.1.2 software. All data are presented as mean ± standard deviation. Normality was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the F-test. The results confirmed that the data followed a normal distribution (P>0.05) and exhibited homogeneous variances (P>0.05). Multiple group comparisons were performed using one-way ANOVA, followed by Tukey’s multiple comparison test for pairwise comparisons when overall significance was detected. A P value <0.05 was considered statistically significant. ANOVA, analysis of variance; DAPI, 4',6-diamidino-2-phenylindole; EdU, 5-ethynyl-2'-deoxyuridine; HUVEC, human umbilical vein endothelial cell; IL-6, interleukin-6; NC, negative control; siNC, negative control siRNA; siSIX1, the most efficient SIX1‑targeting siRNA, corresponding to siSIX1‑3.

Discussion

As an inflammation-driven vasculopathy, PAH is characterized by tight coupling of the IL-6/STAT3 axis with vascular remodeling and aberrant endothelial activation. In this context, we applied a multidimensional, transcriptomics-anchored integrative strategy that pointed to global activation of immune and adhesion programs and a putative cooperative SIX1-STAT3 module. In the SuHx model, we verified IL-6 upregulation as well as worsened hemodynamics and fibrotic remodeling, and observed concomitant increases and colocalization of SIX1 and p-STAT3 within CD31⁺ endothelium. Cell-type specificity experiments confirmed that this upregulation was specifically localized to endothelial cells, with no significant colocalization observed in smooth muscle cells, lymphocytes, or myeloid cells (Figure S3). In HUVECs, IL-6 induced SIX1 expression and drove STAT3 activation, whereas SIX1 knockdown partially reversed IL-6-enhanced proliferation and migration, as well as IL-6-induced upregulation of adhesion molecules and mesenchymal markers (Figure S4). Collectively, these data support a functional “IL-6/SIX1/STAT3” axis and suggest its tractability as a therapeutic target.

We first performed multidimensional transcriptomic profiling in the SuHx-induced PAH model and delineated system-wide activation of immune response and cell-adhesion pathways in diseased lungs. WGCNA further pinpointed a SIX1-centered module tightly associated with right ventricular functional readouts and co-enriched with STAT3-indicated inflammatory/adhesion signals (Figure 1, Figure S1). The positive correlations between SIX1 expression and both RVSP and the extent of vascular remodeling (Figure S1) reinforce this connection, suggesting that SIX1 and STAT3 may constitute a pathogenic transcriptional regulatory unit. This transcriptomic landscape aligns with the prevailing consensus that the inflammatory–immune axis is central to PAH pathogenesis (1,8). Multiple studies place the IL-6/STAT3 axis at the core of aberrant activation of pulmonary vascular cells (5,7,19). In parallel, work in oncology and developmental biology on SIX1 and its co-activator complexes (e.g., SIX1-EYA) shows that SIX1 can cooperate with other transcription factors to amplify downstream gene expression, providing an a priori rationale for our testable hypothesis of the “SIX1-STAT3 co-regulation” (10).

Building on our bioinformatic predictions, the SuHx model not only reproduced elevated pulmonary arterial pressure, right ventricular hypertrophy, and perivascular collagen deposition (Figure 2A-2H), but also showed upregulated pulmonary IL-6 expression (Figure 2I). Histological assessment included low-magnification images to visualize overall lung architecture, and quantitative analyses were performed with each data point representing the mean value from an individual mouse. Consistent with prior work, these findings support IL-6 as a key inflammatory driver in PAH (19). IL-6 receptor antagonism has been reported to improve hemodynamics and fibrosis in experimental models (8), and IL-6 promotes T helper type 17 (Th17) differentiation and EndoMT, thereby exacerbating vascular remodeling (4,20). Notably, we observed synchronous activation and spatial colocalization of SIX1 and STAT3 within the pulmonary vascular endothelium (Figure 3), with further confirmation that p-STAT3 was also localized to CD31⁺ endothelial cells (Figure S2). Cell-type specificity experiments confirmed that this activation was endothelial-specific, as no significant colocalization was detected with smooth muscle, lymphoid, or myeloid cell markers (Figure S3). This extends the extensively studied STAT3 signaling paradigm from PASMCs to endothelial cells (21,22), underscoring cell type-specific differences in PAH pathobiology.

To delineate the functional linkage between SIX1 and STAT3, we demonstrated in HUVECs that IL-6 induces SIX1 expression in a concentration-dependent manner (Figure 4B,4C) and further promotes STAT3 phosphorylation together with pronounced nuclear colocalization of the two proteins (Figure 4F-4I). The specificity of IL-6-mediated STAT3 activation was verified using LIF as a positive control and IL-6 neutralizing antibody, which abolished IL-6-induced SIX1 and p-STAT3 upregulation (Figure 4F-4I). Critically, SIX1 knockdown (Figure 4D,4E) markedly attenuated activation of the IL-6/STAT3 axis and diminished their colocalization (Figure 4F-4I), consistent with prior evidence that pharmacologic JAK2/STAT3 inhibition mitigates vascular remodeling in PAH (23,24). These data position SIX1 as an upstream modulator that contributes to STAT3 activation following IL-6 stimulation. This finding also echoes observations in thyroid cancer in which SIX1, in cooperation with EYA1, augments STAT3 activation (10); however, our study is the first to propose and validate an IL-6/SIX1/STAT3 pathological signaling axis within the endothelial context of PAH. It should be noted that previous studies in cancer cells have reported a SIX1→IL-6 axis, whereas our findings reveal an IL-6→SIX1 direction in PAH endothelial cells, suggesting that the relationship between IL-6 and SIX1 may be context-dependent, varying by cell type and disease setting.

At the functional level, loss of SIX1 effectively reversed the IL-6-induced hyperproliferation and migration of endothelial cells (Figure 5), indicating that the SIX1/STAT3 axis is a central hub through which IL-6 triggers endothelial dysfunction. This finding accords with the canonical view that STAT3 promotes proliferation and suppresses apoptosis (22-25) and is consistent with the pro-survival and pro-proliferative roles of STAT3 reported in PAH-related cells and tissues (23-26). Notably, our study further identifies SIX1 as a critical node within this pathway. To further assess PAH-relevant endothelial outcomes, we examined EndoMT markers, adhesion molecule expression, and apoptosis. IL-6 stimulation significantly upregulated the adhesion molecule VCAM-1 and the mesenchymal marker α-SMA, while the endothelial marker VE-cadherin remained unchanged—suggesting a partial EndoMT phenotype. SIX1 knockdown significantly reversed these IL-6-induced changes (Figure S4A,S4B). TUNEL staining revealed no significant apoptosis in any treatment group (Figure S4C,S4D), indicating that the observed effects of IL-6 and SIX1 on proliferation, migration, and marker expression are not confounded by changes in cell survival. Moreover, prior work in pulmonary fibrosis has shown that SIX1 regulates collagen production and fibrogenic progression, offering a mechanistic clue to the perivascular collagen deposition observed in our model (Figure 2G,2H) and suggesting that SIX1 may contribute to PAH through multiple routes (27,28).

To our knowledge, this study is the first to establish SIX1 as a pivotal mediator that links the inflammatory cytokine IL-6 to STAT3 signaling in PAH, and to demonstrate its importance across multiple levels—from transcriptional regulation and signal activation to cellular phenotypic outputs. These findings not only broaden our understanding of the pleiotropy of IL-6 signaling, but also identify SIX1 as a potential therapeutic target for addressing endothelial dysfunction in PAH. Several limitations warrant consideration. First, while our in vitro mechanistic studies were conducted in HUVECs—a well-characterized model for initial endothelial investigation—these cells may not fully recapitulate the properties of human pulmonary artery endothelial cells (PAECs). Therefore, future validation in primary PAECs or human pulmonary microvascular endothelial cell lines, as well as endothelial cell-specific SIX1 knockout models, is necessary to confirm the pulmonary-specific relevance of our findings. Second, the precise molecular mechanism by which SIX1 regulates STAT3 remains to be elucidated; co-immunoprecipitation, chromatin immunoprecipitation (ChIP)-qPCR, or luciferase reporter assays are needed to determine whether SIX1 directly interacts with STAT3 or transcriptionally regulates components of the JAK/STAT pathway. Third, key functional experiments were performed using a single siRNA sequence; future studies using multiple independent siRNAs or CRISPR/Cas9-mediated knockout are warranted to further confirm the specificity of SIX1 knockdown effects. HUVECs differ from pulmonary microvascular endothelial cells; future validation in primary pulmonary endothelial cells or human pulmonary microvascular endothelial cell lines is necessary. Fourth, some Western blot analyses had reduced sample sizes (n=5–6) due to technical exclusions (protein degradation or loading inconsistencies), and although the observed statistical differences support the robustness of the main conclusions, future studies with larger and more consistent sample sizes are warranted. Fifth, the lack of validation in human PAH samples is a limitation; analysis of SIX1 expression in patient lung tissues or serum would strengthen the clinical relevance of our findings. Next, we will generate endothelial-specific SIX1 knockout mice to test therapeutic potential in the SuHx model; apply interaction proteomics to systematically identify SIX1-binding partners and elucidate its regulation of STAT3; and analyze PAH patient serum and lung specimens to assess correlations between SIX1 expression and clinicopathologic parameters, thereby advancing SIX1 toward clinical translation.

Conclusions

This study demonstrates that SIX1 is upregulated in PAH and mediates endothelial dysfunction via the IL‑6/STAT3 axis. Knockdown of SIX1 reverses IL‑6‑induced proliferation, migration, and partial endothelial‑mesenchymal transition. These findings identify SIX1 as a novel therapeutic target for PAH.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Experiments were performed under a project license (No. 197521) granted by the Animal Care and Use Committee of Soochow University, in compliance with relevant national or institutional guidelines for the care and use of animals. No human studies are presented in the manuscript. No potentially identifiable images or data are presented in this study.

Data Sharing Statement

Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2026-1-0101/dss

jtd-18-04-392-dss.pdf ^{(98.6KB, pdf)}

DOI: 10.21037/jtd-2026-1-0101