Jagged/Notch proteins promote endothelial‐mesenchymal transition‐mediated pulmonary arterial hypertension via upregulation of the expression of GATAs

Abstract This study tested the hypothesis that Jagged2/Notches promoted the endothelial‐mesenchymal transition (endMT)‐mediated pulmonary arterial hypertension (PAH) (i.e. induction by monocrotaline [MCT]/63 mg/kg/subcutaneous injection) through increasing the expression of GATA‐binding factors which were inhibited by propylthiouracil (PTU) (i.e. 0.1% in water for daily drinking since Day 5 after PAH induction) in rodent. As compared with the control (i.e. HUVECs), the protein expressions of GATAs (3/4/6) and endMT markers (Snail/Zeb1/N‐cadherin/vimentin/fibronectin/α‐SMA/p‐Smad2) were significantly reduced, whereas the endothelial‐phenotype markers (CD31/E‐cadherin) were significantly increased in silenced JAG2 gene or in silenced GATA3 gene of HUVECs (all p < 0.001). As compared with the control, the protein expressions of intercellular signallings (GATAs [3/4/6], Jagged1/2, notch1/2 and Snail/Zeb1/N‐cadherin/vimentin/fibronectin/α‐SMA/p‐Smad2) were significantly upregulated in TGF‐ß/monocrotaline‐treated HUVECs that were significantly reversed by PTU treatment (all p < 0.001). By Day 42, the results of animal study demonstrated that the right‐ventricular systolic‐blood‐pressure (RVSBP), RV weight (RVW) and lung injury/fibrotic scores were significantly increased in MCT group than sham‐control (SC) that were reversed in MCT + PTU groups, whereas arterial oxygen saturation (%) and vasorelaxation/nitric oxide production of PA exhibited an opposite pattern of RVW among the groups (all p < 0.0001). The protein expressions of hypertrophic (ß‐MHC)/pressure‐overload (BNP)/oxidative‐stress (NOX‐1/NOX‐2) biomarkers in RV and the protein expressions of intercellular signalling (GATAs3/4/6, Jagged1/2, notch1/2) and endMT markers (Snail/Zeb1/N‐cadherin/vimentin/fibronectin/TGF‐ß/α‐SMA/p‐Smad2) in lung parenchyma displayed an identical pattern of RVW among the groups (all p < 0.0001). Jagged‐Notch‐GATAs signalling, endMT markers and RVSBP that were increased in PAH were suppressed by PTU.


| BACKG ROU N D
Histopathological features of pulmonary vasculatures in idiopathic pulmonary arterial hypertension (PAH) have been clearly identified as complicated morphological changes, including vasoconstriction of resistance artery, in situ thrombosis in small arteries and pulmonary vascular remodelling by proliferation of smooth muscle cells along with intimal fibrosis, medial hypertrophy and adventitial thickening. [1][2][3][4][5][6] This, ultimately, causes not only an overproduction of vasoconstrictors such as endothelin-1 and tissue proliferative factors but also a concomitant reduction in vasodilators and antiproliferative factors, such as prostacyclin and nitric oxide. 7 Currently, there still lacking the effective treatment for PAH, raising that the scientists, in addition to developing a new strategic management for this unmet need, should clearly delineate the underlying mechanism of PAH for a precise treatment.
Activation of JAG2 (i.e. gene expression) in breast tumour cells induced epithelial mesenchymal transition (EMT) and promoted cell survival and migration has been well recognized by previous study. 8 Interestingly, a study has previously demonstrated that depleted JAG2 gene intensely inhibited pancreatic cancer cell migration, invasion and metastasis. 9 Additionally, Jagged2 (i.e. protein expression) was also identified to be remarkably increased in bone marrow stromal cells under hypoxia and promoted the self-renewal of cancer stem-like cells by activating Notch signalling. 10 Presently, Notch activation is identified to be almost always dependent upon Jagged2 ligand engagement. 11 Moreover, a previous study has demonstrated that expression of Notch ligand of Jagged2 was closely correlated with different grades of metastatic and recurrent bladder carcinoma. 12 This finding 12 suggested that Jagged2 plays an important signalling role in tumour metastasis in bladder cancer through initiation of EMT process.
Of the GATA family, [13][14][15][16] the GATA3 is the most critical regulator of mammary epithelial cells. 17 Additionally, transcription factors of the GATA family are essential regulators of the specification, differentiation and proliferation of numerous tissues and a broad variety of cancers in humans, including leukaemia, breast cancer, gastrointestinal cancers and others. [13][14][15][16] The endMT is a process in which endothelial cells (ECs) lose polarity and cell-to-cell contacts, and undergo a dramatic remodelling of the cytoskeleton, called cellular trans-differentiation. [18][19][20] The endMT is a complex biological process in which ECs adopt a mesenchymal phenotype displaying typical mesenchymal cell morphology and functions, including the acquisition of cellular motility and contractile properties. [18][19][20][21][22] Studies have further established that the ECs undergoing endMT lose the expression of epithelial cell-specific proteins and initiate the expression of mesenchymal cell-specific genes as well as the production of their encoded proteins. 23 Intriguingly, the process of EMT/endMT is regulated by several transcriptional suppressor families, including the zinc-finger proteins (i.e. Zeb 1 and Zeb 2), Snail1 and Snail2. [24][25][26] Zeb1 and Zeb2 expressions are regulated by the microRNA-200 (i.e. called family miR-200). [27][28][29] Furthermore, a previous study has shown upregulation of miR-200 locks the tumour cells into an epithelial state and abrogates their metastatic capacity. 30 Moreover, one study has exhibited that Jagged2 was found to promote metastasis of mouse lung adenocarcinoma cells by increasing the expression of GATA-binding factors, which suppressed the expression of miR-200 family that targeted at the transcriptional repressors that, in turn, drove EMT and thereby induced EMT, suggesting that GATA3 and miR-200 are mutually inhibitory and have opposing effects on EMT and metastasis. 8 Intriguingly, studies have further demonstrated that PAH was associated with GATA2/6 deficiency. [31][32][33] On the hand, studies have shown that GATA4 is upregulated in setting of PAH. 34 It is well known the pathogenesis PAH shares quite similarities with carcinogenesis. Additionally, some studies have previously demonstrated that (EndMT) played a crucial role in PAH through upregulating the vascular wall remodelling. 21,35,36 Accordingly, it is rationale to hypothesize that endMT plays a principal role in initiation and progression of PAH. However, the characteristics of endothelial cells which have undergone endMT in PAH have not yet been fully investigated. Some previous studies have shown that activation of Jagged-Notch signalling enhance pulmonary vascular smooth muscle cell proliferation. 37,38 However, there have no data to address (%) and vasorelaxation/nitric oxide production of PA exhibited an opposite pattern of RVW among the groups (all p < 0.0001). The protein expressions of hypertrophic (ß-MHC)/pressure-overload (BNP)/oxidative-stress (NOX-1/NOX-2) biomarkers in RV and the protein expressions of intercellular signalling (GATAs3/4/6, Jagged1/2, notch1/2) and endMT markers (Snail/Zeb1/N-cadherin/vimentin/fibronectin/TGF-ß/α-SMA/p-Smad2) in lung parenchyma displayed an identical pattern of RVW among the groups (all p < 0.0001). Jagged-Notch-GATAs signalling, endMT markers and RVSBP that were increased in PAH were suppressed by PTU.

K E Y W O R D S
endothelial-mesenchymal transition biomarkers, intercellular signalling, propylthiouracil, pulmonary artery hypertension, vascular wall proliferation the role of Jagged2 or GATA3 on setting of PAH. Based on the aforementioned mentioned issues, [18][19][20] we proposed that a complex signalling of Jaggeds-Notches-GATAS-family miR-200 might participate in the initiation and propagation of PAH though regulating the endMT process in PA.
The PTU is well-known for treatment of hyperthyroidism. In addition to its standard clinical use for patients with hyperthyroidism, because of its thyroid-suppressing function, PTU has also been identified to own a strong antioxidant ability, 39 augment NO production, inhibit vascular smooth muscle proliferation and migration, as well as collagen production. 40,41 Accordingly, previous studies, 42,43 including ours, 44 have revealed that PTU has a strong anti-atherosclerotic effect. Furthermore, another of our previous study 45 has shown that PTU therapy attenuated monocrotaline (MCT)-induced PAH through augmenting NO production and suppressing smooth muscle cell proliferation. However, the underlying mechanism of PTU therapy on ameliorating PAH is still not yet fully clarified by our previous study. 45 2 | MATERIAL S AND ME THODS

| Transwell migration assay
Cells were first trypsinized, and 5 × 10 4 cells were then added to the Boyden chambers (8 μm pore size; Millipore) with 0.5% FBScontaining medium and assay media that contained 10% FBS was added to the culture plates. After 24 h incubation, the non-motile cells at the top of the filter were removed and the motile cells at the bottom of the filter were fixed with methanol and stained with one-tenth dilution of Giemsa (Sigma Corporation). The number of migrated cells in each chamber was carefully counted in five randomly chosen fields under the microscope for three independent experiments.

| Animal model of PAH and animal grouping
It is well recognized that MCT treatment is one of the most popular method for creating a validated animal model of PAH. 45,46 The procedure and protocol were based on our previous reports. 45,46 In detail, by Day 0, pathogen-free, adult male Sprague-Dawley (SD) rats, weighing 325-350 g (Charles River Technology, BioLASCO Co., Ltd., n = 24) were given one subcutaneous injection of monocrotaline (MCT) (63 mg/kg; Sigma). On Day 5, these MCT-treated animals were assigned to two experimental groups: group 2 (MCT alone, n = 8) and group 3 (MCT + 0.1% PTU [Sigma] in drinking water daily, n = 8). Another group 1 (i.e. sham-operated control [SC], n = 8) that receiving only subcutaneous injection of 3 ml normal saline served as SC. PTU therapy was implemented immediately after the assignment. The dosage of MCT utilized here was based on our previous reports 45,46 with minimal modification of MCT dosage.

| Rationale of the PTU dosage
The dosage and the time point of the PTU administration were according to our previous description. 45 Briefly, the 0.1% PTU in water 100 ml was equivalent to 0.27 mg per 100 ml of drinking water which was the average daily amount of water consumption for each rat in our study. It was well recognized that a rat usually drank 15-18 cc water per day. Thus, it was easily to calculate the dosage of PTU for one rat per day.

| Hemodynamic measurement
The procedure and protocol have been described in our previous studies. 45,46 In detail, on Day 42 after MCT administration, the rats were anaesthetised by inhalational 2.0% isoflurane, placed supine on a warming pad at 37°C for midline laparotomies. After being shaved on the chest, animals in each group were endotracheally intubated with positive-pressure ventilation (180 ml/min) with room air using a small animal ventilator (SAR-830/A, CWE Inc.). The heart was exposed by left thoracotomy. A sterile 20-gauge, soft plasticcoated needle was inserted into the right ventricle (RV) and femoral artery of each rat to measure the right ventricular systolic pressure (RVSP) (an indirect indicator of pulmonary arterial systolic blood pressure) and arterial pressure, respectively. The pressure signals were first transmitted to pressure transducers (UFI, model 1050) and were then exported to a bridge amplifier (ML866 PowerLab 4/30 Data Acquisition Systems. ADInstruments Pty Ltd.) where the signals were amplified and digitized. The data were recorded and later analysed with the Labchart software (ADInstruments).
After hemodynamic measurements, the rats were euthanized in each group with the hearts and lungs to be harvested for individual study. For each animal, the right ventricular weight, whole heart weight, septal + left ventricular (LV) weight and body weight were recorded. The left lung was fixed in 4% formaldehyde and then embedded in paraffin blocks. The right lung was cut into pieces, frozen in liquid nitrogen and then stored at −80°C until future use.

| Measurement of pulmonary arterial contractility
At the end of study, pulmonary artery was isolated, cleaned and cut into slices of 2 mm in length for evaluating the contractile and relaxant response as we previously reported. 47 In detail, pulmonary arterial rings were carefully mounted on an isometric force transducer

| Assessment for the arterial basal nitric oxide (NO) release
Vascular basal nitric oxide release was calculated as the percentage of difference between PE-induced vasocontractile response in the absence and presence of L-NAME according to our previous study. 47 2.9 | Histological assessment of lung injury and crowded score of lung parenchyma and vascular wall remodelling of pulmonary artery To elucidate the alveolar sac distribution in lung parenchyma, left lung specimens from all animals were fixed in 10% buffered formalin before embedding in paraffin and the tissue was sectioned at 5 μm for light microscopic analysis. The H&E staining was performed to determine the number of alveolar sacs according to our previous study 48 in a blind fashion. Three lung sections from each rat were analysed, and three randomly selected high-power fields (HPFs) (200×) were examined in each section. The mean number per HPF for each animal was then determined by summation of all numbers divided by 9. In addition, alveolar wall thickness and the presence or absence of haemorrhage were determined under light microscopy. The extent of crowded area, which is defined as region of thickened septa in lung parenchyma associated with partial or complete collapse of alveoli on H&E-stained sections, was performed in a blind fashion. The scoring system adopted was as follows: 0 = no detectable crowded area; 1 = <15% of crowded area; 2 = 16%-25% of crowded area; 3 = 26%-50% of crowded area; 4 = 51%-75% of crowded area; 5 = 76%-100% of crowded area/per high-power field (200×). Additionally, this scoring system was also adapted for identification of the lung inflammatory cell infiltration score (i.e. to replace the 'detectable crowded area' by 'inflammatory cell infiltration area').
The procedure and protocol for assessing the muscularization (i.e. an index of vascular wall remodelling) of pulmonary arterioles have described in detail in our previous report. 45,49 Briefly, three measurements were examined for the analysis of thickness of pulmonary arterioles. Muscularization of the pulmonary arterial medial layer at the level of middle lung level was defined as a mean thickness of vessel wall >50% of the lumen diameter in a vessel of diameter >30 μm. Measurement of arteriolar diameter and wall thickness was achieved using the Image-J software (NIH).
In this study, histopathological findings were examined by Olympus BX51 microscope (Shinjuku). The lens used in this study were the Olympus UPlanSApo 40× Plan Apo objective lens (40×/0.90) for 400× magnification and the Olympus UPlanSApo

| Histological quantification of lung fibrosis
The procedure and protocol have been described in our previous reports. [44][45][46][47][48] In detail, Masson's trichrome staining was used for identification of the lung fibrotic area. Three serial sections of lung in each animal were prepared at 4 μm thickness by Cryostat (Leica CM3050S). The integrated areas (μm 2 ) of fibrosis on each section were calculated using the Image Tool 3 (IT3) image analysis software (University of Texas, Health Science Center, San Antonio, UTHSCSA; Image Tool for Windows, Version 3.0). Three randomly selected high-power fields (HPFs) (100×) were analysed in each section. After determining the number of pixels in each fibrotic area per HPF, the numbers of pixels obtained from three HPFs were calculated. The procedure was repeated in two other sections for each animal. The mean pixel number per HPF for each animal was then determined by summing up all pixel numbers and divided by 9. The mean integrated area (μm 2 ) of fibrosis in parenchyma per HPF was obtained using a conversion factor of 19.24 (since 1 μm 2 corresponds to 19.24 pixels).

| Statistical analysis
Quantitative data were expressed as mean ± standard error of the mean (SEM). Statistical analysis was adequately performed by anova followed by Bonferroni multiple comparison post hoc test. Statistical analysis was performed using SAS statistical software for Windows version 8.2 (SAS Institute). A probability value <0.05 was considered statistically significant.

| TGF-ß1 upregulated endothelialmesenchymal transition (endMT) process
First, to prove that the TGF-ß1 participated in endMT processing, we categorized the HUVECs into HUVECs only (i.e. the control group) and HUVECs +10 ng/ml of TGF-ß1 for two time points (i.e.

| Protein expressions of endMT biomarkers undergoing the TGF-ß1 and MCT treatments
Second, for the purpose of MCT would be utilized for PAH induction in rats, we compared whether the MCT treatment was comparable to TGF-ß1 on upregulating the endMT biomarkers ( Figure S2).

| Jagged2 regulated the protein expression of GATAs
To elucidate whether the Jagged2 regulated the protein expression of GATAs, the gene JAG2 was silenced in HUVECs and the culturing cells were categorized into HUVECs only, HUVECs + scramble and siRNA-JAG2 (Figure 1 and Figure S3). The result showed that the protein expressions of GATA3, GATA4 and GATA6 were significantly lower in siRNA-JAG2 group than in the former two groups but they did not differ between the former two groups, suggesting the expressions of GATAs were regulated by jagged2 (i.e. served as the upstream signalling mediated the downstream signalling of GATAs).

| GATAs regulated expressions of the EMT biomarkers
To verify whether the GATAs regulated the expressions of endMT, the silencing of the GATA3 gene in HUVECs was performed ( Figure 2). The result showed that the protein expressions of Snail, Zeb1, N-cadherin, vimentin, fibronectin, α-SMA and phosphorylated (p)-Smad2, six indices of endMT biomarkers, were significantly reduced in siRNA-GATA3 group than in HUVECs only and HUVECs + scramble, whereas the protein expressions of CD31 and E-cadherin, two indicators of the epithelial phenotype biomarkers, exhibited an opposite pattern of Snail among the groups. However, they were similar between HUVECs only and HUVECs + scramble groups. Our findings implicated that the role of the GATAs was essential for augmenting the expressions of EMT biomarkers.

| HUVECs migratory assay, miR-200 expression and cell viability in HUVECs
To verify whether the GATAs participated in regulating the propagation of endMT processing, resulting in enhancement of migratory capacity, the HUVECs (2.0 × 10 5 /well were seeded) were categorized into HUVECs only, HUVECs + scramble and silenced GATA3 gene in HUVECs (i.e. siRNA-GATA3) migratory assay ( Figure 3). All the cell cultures were carried for 72 h prior to the cell were collected for individual study. As we expected, the migratory ability was significantly suppressed in siRNA-GATA3 group than in the former two groups, but it did not differ between the former two groups. This result proved that in situation of intact of cell viability, losing migratory ability of HUVECs implicated a reflection of endEMT processing. Accordingly, our finding might imply that GATAs participated in upregulating the endothelial to mesenchymal processing (i.e.

| Pulmonary artery relaxation and nitric oxide (NO) release by Day 42 after MCT treatment
The baseline NO released from endothelial cells of PA was significantly reduced in MCT group than in SC that was significantly reversed in MCT-PTU group ( Figure 5).
Additionally, to elucidate the impact of PTU on regulating vessel relaxation and contraction, PA was cut into pieces and mounted on the DMT (Danish Myo Technology). As expected, vasoconstriction induced by phenylephrine was significantly increased in MCT group than in SC and was significantly reversed in MCT-PTU group.
On the contrary, vasorelaxation caused by acetylcholine exhibited an opposite pattern of vasoconstriction among the three groups.
Nevertheless, the vasorelaxation caused by sodium nitroprusside did not differ among the groups. Our results supported that PTU therapy could upregulate the PA relaxation and NO production. The ratio of RV weight and WWLW to tibial length was significantly increased in MCT group than in SC and MCT-PTU groups, and significantly increased in MCT-PTU group than in SC group. On the other contrary, the ratio of septum + LV weight to tibial length showed no significant difference among the three groups.
The RVSBP was significantly higher in MCT group than in the other two groups and significantly higher in MCT-PTU group than in the SC group. On the contrary, the FASBP was significantly higher in SC than in MCT and MCT-PTU groups, but it showed no difference between the latter two groups. Nevertheless, the LVSBP did not differ among the three groups. Our finding discovered that PTU therapy not only reduced PAH but also reduced cardiac hypertrophy in setting of PAH.

| The lung injury score, lung consolidation area, inflammatory cell infiltration and arterial muscularization Day 42 after MCT treatment
To assess the degree of lung injury score and lung consolidation, we utilized the microscopic tool for identification of lung histopathological feature ( Figure 7). As we expected, the number of alveolar sacs was significantly reduced in MCT-PTU group and more significantly reduced in MCT only group as compared with SC group, whereas the lung parenchymal crowding demonstrated an opposite pattern compared to that of the number of alveolar sacs among the three groups, suggesting that the lung injury score (i.e. combined reduction of Other statistical analyses were performed by one-way anova, followed by Bonferroni multiple comparison post hoc test. Symbols (*, †, ‡) indicate significance (at 0.05 level), this means that * versus †, p < 0.05; * versus ‡, p < 0.05; † versus ‡, p < 0.05 and * versus † versus ‡, p < 0.01. HUVECs, human umbilical vein endothelial cells; MCT, monocrotaline; PTU, propylthiouracil; siRNA-GATA3, silenced the GATA3 gene. alveolar sacs and increased crowding lung parenchyma) was notably reversed in PAH animals after receiving the PTU therapy.
Additionally, the lung consolidation area and inflammatory cell infiltration score exhibited an identical pattern of lung injury score among the three groups. Our finding proved that PTU therapy safeguarded the lung architectural integrity in setting of PAH.
Furthermore, histopathological analysis using H&E staining demonstrated that the percentage of thickened arterioles and the number of muscularized arterioles displayed an identical manner compared to that of the crowded score among the three groups.

| Fibrotic area and antioxidant biomarker in lung parenchyma by Day 42 after MCT treatment
To clarify whether PTU treatment could attenuate the MCT-induced lung fibrosis and upregulation of antioxidant, histopathologic level of lung parenchyma was investigated by using the microscopic finding ( Figure 8). The result of Masson's trichrome stain demonstrated that the fibrotic area of the lung parenchyma was significantly higher in MCT group than that of the SC group, and that was significantly

| Immunofluorescent stain for identification of specific inflammatory cells in the lung parenchyma by Day 42 after MCT treatment
By using the immunofluorescent (IF) microscopic finding, we further clarified the inflammatory cell distribution in the lung parenchyma ( Figure 9). The result showed that cellular expressions of CD11b and CD68, two indicators of inflammation, were significantly increased in MCT group than in SC and MCT-PTU groups and significantly increased in MCT-PTU group than in the SC group, suggesting that PTU therapy could attenuate inflammation in lung parenchyma of PAH. Our finding provided the evidence that PTU therapy suppressed PAH-induced inflammatory reaction.

| Protein expressions of promoters and transcriptional regulators of endMT, and gene expression of miR-200 in lung parenchyma by Day 42 after MCT treatment
To verify whether the promoters and transcriptional regulators of endMT were upregulated in the lung of PAH rodents, Western blot analysis was utilized in the present study ( Figure 11). The result demonstrated that the protein expressions of Jagged1/Jagged2 and

| Protein expressions of endMT biomarkers in lung parenchyma by Day 42 after MCT treatment
To clarify whether the PTU therapy effectively attenuated the expression of endMT biomarkers in PHA rat, Western blot analysis of the lung tissue was performed again (Figure 12). The result demonstrated that the protein expressions of Snail, Zeb1, fibronectin, Vimentin, TGF-ß, p-Smad2 and p-Smad3, seven indices of endMT biomarkers, were significantly increased in MCT group than in the SC group, and were significantly reversed in MCT-PTU group, suggesting PTU therapy may be potential for the PAH patients.

| DISCUSS ION
This study which investigated the role of Jaggeds-Notches-GATATs-endMT signalling pathway on the initiation and propagation of MCT-induced PAH in rodent and the therapeutic impact of The distinctively histopathological feature of pulmonary artery in setting of PAH has been well recognized as endlessly increasing in the intimal, medial and adventitial layers, resulting in excessive pulmonary vascular remodelling and sustained pulmonary vasoconstriction, as a consequence of PAH and right-side heart failure. 50 It is well investigated that the signal transduction in Notch pathway is elicited by binding of Notch receptors (i.e. Notch1 to 4) with ligands (i.e. Jagged1, Jagged2, Delta-like, Delta-like3 and Delta-like4). 51,52 Additionally, Notch signalling has been identified to not only participate in proliferation, invasion and metastasis of cancer cells/cancer stem-like cells 10,11 but also be particularly important in vascular remodelling by promoting proliferation of endothelial cells and smooth muscle cell recruitment in setting of PAH. 50,53,54 One important finding in the present study was that the protein expressions of Notch1/Notch2 and Jagged1/Jagged2 were significantly upregulated in HUVECs undergoing the TGF-ß or MCT treatment. Additionally, these parameters were also identified to be markedly increased in lung parenchyma of MCT-induced PAH rats.
Furthermore, the protein expressions of the endMT markers were substantially increased in HUVECs-treated by TGF-ß or MCT and lung tissue of PAH rats. On the contrary, when looked at the in vitro study, we also identified that silenced JAG2 gene in HUVECs significantly suppressed the protein expressions of the endMT markers. shown that a critical GATA-6 deficiency played a critical role in the development and disease progression in PAH. 55 Another previous study has also revealed that the levels of GATA-6 mRNA and protein were markedly decreased in those of pneumonectomy and MCTtreated PAH rats. 56 In contrast to these studies, 55,56 Suzuki et al previously demonstrated that GATA-4 activation crucially regulated pulmonary arterial smooth muscle cell proliferation, suggesting targeting at the GATA-4 activation could be a therapeutic potential for preventing the PAH. 28 In this way, our finding was consistent with the finding of the previous study. 28 We remain uncertain regarding why the protein expressions of GATAs were remarkably increased not only in the in vitro study (i.e. HUVECs treated by TGF-ß or by MCT) but also in the lung parenchyma of PAH rats. Perhaps, one main reason could be due to an intrinsic response of the HUVECs and the cells in lung parenchyma to the MCT stimulation or hypoxia/ elevated RVSBP, that is an index of PA systolic blood pressure.
The miR-200 family are universally accepted to target at the transcriptional repressors, resulting in inhibiting the EMT and cancer cell metastasis. 10,25 Surprisingly, while an association between tumour metastasis along with upregulation of EMT and downregulation of miR-200 has been extensively investigated 8,28,57 link between miR-200 and PAH has not yet been reported. An essential finding in the present study was that the miR-200a-3p and miR-200c-3P were markedly suppressed by MCT treatment (i.e. in vitro study) and in lung parenchyma of PAH rat. Accordingly, our findings, in addition to extending of the previous studies, 8,28,57 highlight that this finding could possess novelty, and therefore, would provide a more complete jigsaw puzzle on the understanding of PAH mechanism.
PAH always causes the RV pressure overload, RV hypertrophy and finally, right-side heart failure. [1][2][3]5 A principal finding in the present study was that the parameters of RV weight, lung weight and the RVSBP were significantly increased in MCT group than in SC control, whereas the SaO 2 exhibited an opposite pattern between these two groups. Additionally, protein expressions of pressure overload, hypertrophic and oxidative biomarkers remarkably increased in MCT animals than in the SC animals. Our findings were consistent with the findings of those previous studies, [1][2][3]5 suggesting that we had successfully created a reliable animal model of PAH for the study.
Interestingly, our previous study has proved that PTU therapy significantly attenuated MCT-induced PAH, RVSBP, RV hypertrophy and lung injury in rodent. 45 However, the limitation of our study 45 was that the underlying mechanisms of how the PTU therapy could ameliorate the above-mentioned parameters and the PAH were not clearly verified. Our present study also demonstrated that the PTU therapy effectively improved the MCT-induced PAH and protected the right ventricle and lung organ against the PAH-induced injury. In this way, our finding corroborated with the finding of our previous study. 45 Of distinctively important finding in the present study was that the underlying mechanism of PTU therapy attenuated the MCT-induced PAH was clearly clarified due to inhibiting the endMT processing up-regulators (i.e. Jagged1/Jagged2 and Notch1/ Notch2) and endMT transcriptional regulators (i.e. GATAs [3,4,6]

| Study limitation
This study has limitations. First, the dosage of PTU utilized in the present study was merely based on our previous report. 45 Accordingly, this study did not verify what was the optimal dose for offering the best impact on inhibiting the PAH. Second, although previous studies have demonstrated that the protective effect of PTU on inhibiting the atherosclerosis, smooth muscle cell proliferation and PAH was independent of its hypothyroid effect [39][40][41][42][43][44][45] and no PTU-related hypothyroidism side effect was observed by previous studies, 41,45 the present study did not measure the circulating level of TSH, T3 or free T4 in the rodent. However, we did not find any PTU-related hypothyroidism side effect in the animals. Third, this study just utilized the HUVECs rather than to utilize the rat circulatory-derived endothelial cells or primary endothelial cells from PAH patient to conduct the endEMT, suggesting that its justifiability might not be enough. Finally, although extensive works were done in the present study, the precise mechanism of PAH might be more complicated that was out of the scope of this study. Based on the results of our study, we schematically illustrated the underlying mechanism of how the PTU therapy suppressed the development of MCT-induced PAH in rodent in Figure 13. and Guangzhou Science and technology planning project, China (202103000010).

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors declare no conflict of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets of present study can be available from the corresponding author upon request.

R E FE R E N C E S F I G U R E 1 3
Schematically illustrating the underlying mechanism of monocrotaline-induced endothelial damage mimicking the clinical setting of pulmonary arterial hypertension (PAH) and the therapeutic impact of propylthiouracil on against the initiation and propagation of PAH. EC, endothelial cell; endMT, endothelialmesenchymal transition; PABP, pulmonary arterial blood pressure; RV, right ventricular; SM, smooth muscle. Black arrows indicated medium layer thickness after monocrotaline (MCT) treatment.