Whole Exome Sequencing of Patients With Heritable and Idiopathic Pulmonary Arterial Hypertension in Central Taiwan

Background Genetic variants could be identified in subjects with idiopathic and heritable pulmonary arterial hypertension (PAH). The 6th World Symposium on Pulmonary Hypertension (WSPH) provided a list of genes with evidence of association with PAH. However, reports using whole exome sequencing (WES) from southeastern Asian PAH cohorts were scarce. Methods Subjects with idiopathic and heritable PAH (N = 45) from two medical centers in central Taiwan were screened for PAH related gene variants. The genomic DNA was prepared from peripheral blood lymphocytes. We performed WES for all patients enrolled in this study. All identified gene variants were validated by polymerase-chain reaction and Sanger sequencing. The clinical and hemodynamic data were compared between bone morphogenetic protein receptor type-2 (BMPR2) gene variants carriers vs. non-carriers. Results Eight patients (8/45 = 17.8%) was identified carrying BMPR2 gene variants and 8 patients (8/45 = 17.8%) had other WSPH-listed PAH-related gene variants (1 with ACVRL1, 1 with ENG, 1 with SMAD9, 1 with SMAD1, 1 with ATP13A3 and 3 with AQP1). In addition, a total of 14 non-WSPH-listed PAH-related genetic variant sites (ABCC8, NOTCH1, NOTCH2, NOTCH3, JAG1, BMP10, GGCX, FBLN2, ABCA3 and PTGIS) were found in this PAH cohort. Subjects carrying BMPR2 gene variant (N = 8) were younger at diagnosis of PAH (30 ± 11 vs 49 ± 13 years, p = 0.001) than the non-carrier group (N = 37). BMPR2 variant carriers had a trend toward having higher mean pulmonary arterial pressure (PAP) (61 ± 19 vs. 51 ± 13 mmHg, p = 0.076) than the non-carriers upon initial diagnosis. Pulmonary vascular resistance, right atrial pressure, cardiac output, as well as functional class were similar between BMPR2 variant carriers and non-carriers at initial diagnosis. Conclusions We identified 17.8% of patients with BMPR2 gene variants and 17.8% subjects with other 6th WSPH-listed PAH-related gene variants in a Taiwanese idiopathic and heritable PAH cohort. PAH patients carrying BMPR2 variants presented at a younger age with a trend toward having higher mean PAP at initial diagnosis.


INTRODUCTION
It is now well established that around 70-80% of heritable pulmonary arterial hypertension (PAH) and 10-20% of idiopathic PAH cases are related with genetic variants in bone morphogenetic protein receptor type-2 (BMPR2) (1,2). BMPR2 deficiency leads to an abnormal over-activation of the transforming growth factor β (TGF-β) signaling pathway, resulting in over-proliferation of vascular smooth muscle cells at pulmonary arteriole (3).
Genes not directly impacting on the TGF-β/BMP pathway, including aquaporin 1 (AQP1), which encodes the plasma membrane water channel aquaporin 1, and ATPase family homolog up-regulated in senescence cells 1 (ATP13A3), an ATPases involved in ion channel transport, were reported to be associated with PAH (2,6,8). A distinct form of PAH, pulmonary veno-occlusive disease (PVOD) or pulmonary capillary hemangiomatosis (PCH), was shown recently to be caused by homozygous mutations in eukaryotic translation initiation factor 2 alpha kinase 4 (EIF2AK4), a kinase in the integrated stress response (6,(8)(9)(10). The 6 th World Symposium on Pulmonary Hypertension (WSPH) provided a list of pulmonary arterial hypertension genes with higher or lower level of evidence (2).
In addition to the genetic variants listed by WSPH (2), there are several new candidate genes with related reports for their associations with PAH. ABCC8 (ATP-binding cassette, subfamily C, member 8) encodes SUR1 (sulfonylurea receptor 1), a regulatory subunit of the ATP-sensitive potassium channel (11). ABCC8 loss-of function mutations in both pediatric and adult-onset PAH had been reported (11,12). NOTCH signaling is involved in vascular morphogenesis and function (13). A mechanistic link between Notch3 signaling and PAH is supported by the observation that NOTCH3 KO mice are protected from hypoxia-induced PAH and that PAH can be treated by a g-secretase inhibitor that inhibits the NOTCH pathway (14). Chida et al. (15) identified two novel missense mutations (G840E and T900P) of the NOTCH3 gene in two cases of childhood idiopathic PAH.
PAH patients with BMPR2 genetic variants are trended to have earlier onset and worse clinical scenarios (1,18,19). Some PAH Asian cohorts had a higher rate for carrying PAH-related genetic variants than the Caucasian cohorts (18)(19)(20). However, reports using whole exome sequencing (WES) for identifying genetic variants from Southeasten Asian idiopathic and heritable PAH cohorts are still rare. The aim of the current research is using WES to identify PAH-related genetic variants and investigate their hemodynamic and clinical profiles in a Taiwanese idiopathic and heritable PAH cohort.

Patient Population
We enrolled 45 idiopathic and heritable PAH patients from the Cardiology divisions of Taichung Veterans General Hospital (Taichung, Taiwan) and China Medical University Hospital (Taichung, Taiwan) for PAH genetic analysis between Mar. and Sep. 2021 (Figure 1). The diagnosis of PAH was based on the measurements of a mean PAP>20 mmHg at rest and a pulmonary arterial wedge pressure ≤15 mmHg by right heart catheterization (21). Secondary pulmonary hypertension was excluded by history, blood tests, chest computed tomography scan, nuclear ventilation-perfusion scan and echocardiography (21). We retrospectively collected information including age of initial PAH diagnosis, functional class at diagnosis, and medications and clinical worsening events after initial PAH diagnosis. This study was reviewed and approved by the Institutional Review Board of Taichung Veterans General Hospital (Approval Number, SE21052A) and China Medical University Hospital (Approval Number, CMUH109-REC3-198) (Taichung, Taiwan) and all patients gave their written consent to participate.

Hemodynamic Measurements
All diagnoses were confirmed by right heart catheterization to measure hemodynamic values for all patients at their initial diagnosis (21). The mean PAP, pulmonary arterial wedge pressure, cardiac output, and PVR were recorded by standard techniques (21). The date of 1st right heart catheterization was deemed as the time of initial PAH diagnosis.

Diagnosis Criterion for Hereditary Hemorrhagic Telangiectasia
The diagnosis is based on the Curacao criteria. These criteria include (1) recurrent and spontaneous epistaxis; (2) visceral localization (gastrointestinal telangiectasia, pulmonary, hepatic, cerebral or spinal arteriovenous malformations) (3) an affected first-degree family member; and (4) the presence of mucocutaneous telangiectases. When an individual shows three or more criteria, they are considered to have HHT. When they meet two criteria the diagnosis is possible and with one or none criteria, HHT is considered unlikely (7,22).

DNA Extraction, Whole Exome Sequencing and Data Analysis
Total genomic DNA was isolated from peripheral blood using NucleoSpin R Blood Kit (Macherey-Nagel, Duren, Germany) according to manufacturer protocols. We performed the WES for all patients enrolled in this study. Whole exome libraries were processed using KAPA HyperExome Plus and TopBP1 (26,38)]. Those variants with read depth <30x were filtered out. The coverage of the specific gene was analyzed using WES Binary Alignment/Map (BAM) files. In this study, the average coverages of all candidate genes have >100×. All variants on those genes were compared with widely used normal subject control databases (gnomAD database, ExAC, 1000 Genomes Project and Taiwan Biobank) to filter those variants allelic frequencies more than 0.1% in East Asian group (EAS). The Polyphen2 and SIFT in silico software were applied to predict the impact of variants on specific genes. For structure variations on BMPR2 gene, we applied ExomeDepth from a free R package (version 1.1.15) to detect CNVs on the BMPR2 gene of all PAH patients (39). The suspected CNVs on BMPR2 from ExomeDepth analysis software were subjected to validation through multiplex ligationdependent probe amplification technology using SALSA MLPA Probemix P093 HHT/HPAH kit (MRC-Holland, Netherlands). The evaluation of clinical impacts followed the 2015 ACMG guidelines (The American College of Medical Genetics and Genomics) (40).

Polymerase-Chain Reaction and Sanger Sequencing
All filtered variants identified from candidate genes were validated by polymerase-chain reaction (PCR) and Sanger sequencing. The specific PCR primers were designed manually to cover entire exons of genes. The PCR was conducted using PCRBIO Ultra Polymerase (PCR Biosystems, London, UK) and SensoQuest Labcycler (SensoQuest, Göttingen, Germany). The purified PCR products were directly sequenced with an ABI PRISM terminator cycle sequencing kit v3.1 on the ABI 3730 DNA sequencer (Applied Biosystems, CA, USA).

Statistical Analysis
Continuous variables are expressed as mean ± SD and categorical data as percentages. The study subjects were grouped as BMPR2 genetic variant carriers and non-carriers. Non-normally distributed continuous variables were compared with the Mann-Whitney U test, while normally distributed variables were compared with Student's t test. Categorical variables were compared by the Chi-square test with Yate's correction or Fisher's exact test. A two-tailed p-value < 0.05 was considered statistically significant. The SPSS 12.1 statistical software package (SPSS, Inc., Chicago, IL, USA) was used for all calculations.

Details of WSPH-Listed PAH-Related Gene (Other Than BMPR2) Variants
The 6th WSPH provided a list of pulmonary arterial hypertension genes with higher or lower level of evidence (2). In this study, 8 patients (8/45 = 17.8%) had WSPH-listed PAH-related gene (other than BMPR2) variants (1 with ACVRL1, 1 with ENG, 1 with SMAD9, 1 with SMAD1, 1 with ATP13A3 and 3 with AQP1) ( Table 1 and Figure 1). All the mis-sense variant effects were checked by software PolyPhen2 and SIFT and were listed in

Details of Non-WSPH-Listed PAH-Related Gene Variants
Among the study cohort, we totally identified 14 genetic variants sites, which are not listed in WSPH PAH-related genes, from 12 patients (Figure 1 and Table 1). Novel genetic variants related with Notch pathway, such as NOTCH3, NOTCH2, NOTCH1, and JAG1 were identified in this cohort, too ( Table 1). One patient carried an ABCC8 genetic variant, which is deemed likely pathogenic ( Table 1).

Family History of PAH in the Study Cohort
Among the 45 study subjects, 3 of them had family aggregate of heritable PAH ( Table 2). Patient (A256) with BMPR2

DISCUSSIONS
Idiopathic and heritable PAH patients with genetic variants are usually younger with worse hemodynamic or clinical profiles in Caucasians and Northeastern Asian cohorts (1,18). However, reports from southeastern Asian cohorts are scarce. In this study, we identified 17.8% of patients with BMPR2 gene variants and 17.8% subjects with WSPHlisted PAH-related gene (other than BMPR2) variants in a Taiwanese PAH cohort. Idiopathic and heritable PAH patients carrying BMPR2 variant presented at a younger age with a trend toward having higher mean PAP at initial diagnosis. BMPR2 genetic variants will lead to receptor protein deficiency and result in abnormal overactivation of the TGF-β1 signaling pathway (3). The TGF-β pathway over-activation leads to proliferation of vascular smooth muscle cells and increase of PVR in pulmonary circulation in primary PAH (3,46). Asian cohort from Korea reported that the prevalence of BMPR2 variants in Korean primary PAH patients was 22% (19). A Japanese group reported BMPR2 variants in all 4 familial (4/4 = 100%) and 12 (12/30 = 40%) of the sporadic PAH cases, which was higher than the proportions reported in Caucasian patients (20). Another Japanese group demonstrated that with state-of-the-art therapy, the long-term survival of patients with PAH was high, regardless of the BMPR2 mutation status (47). Evans et al. conducted a meta-analysis of 1,550 patients with idiopathic, heritable, and anorexigen-associated PAH from eight cohorts and showed that 29% had BMPR2 gene mutations (1). In our study from central Taiwan, we identified BMPR2 variants in 17.8 % (8/45) of primary PAH patients, which is somewhat lower than previously reported prevalence from other cohorts. However, we identified three novel BMPR2 gene variants (c.1512 deletion, c.479 duplicate, c.1165 G>A) in the current study (Table 1), which were not reported in literature before.
The 6th WSPH listed BMPR2, EIF2AK4, TBX4, ATP13A3, GDF2, SOX17, AQP1, ACVRL1, SMAD9, ENG, KCNK3 and CAV1 mutations as higher level of evidence and SMAD4, SMAD1, KLF2, BMPR1B and KCNA5 with lower level of evidence of association with primary PAH (2). Mutations of ACVRL1, ENG, SMAD1, SMAD4, and SMAD9 are closely related with alterations of the TGF-β/BMPR2 signal pathways in pathogenesis of primary PAH (2,(4)(5)(6). In this study, we used WES and identified 17.8% subjects with 6th WSPH listed PAH-related gene variants (other than BMPR2) in a Taiwanese idiopathic and heritable PAH cohort. The protein product of EIF2AK4 belongs to a family of kinases that regulates angiogenesis. Specifically, these kinases phosphorylate the α subunit of eukaryotic translation initiation factor 2, a protein that downregulates protein synthesis in response to varied cellular stresses (9). In this study, none of our study subjects had homozygous EIF2AK4 variants and nor did anyone present with evidence of PVOD or PCH (9,10,48). Three of study patients had EIF2AK4 heterozygous variants (Supplementary Table 1) (48). The mutations of ACVRL1, a TGF-β/BMPR2 signal pathway member, have been found to be associated with PAH. In addition, ACVRL1 mutation has been found in patients with HHT combined with PAH (7). One study subject (A249) carried ACVRL1 variant but did not show clinical evidence of HHT. Previous report from Zhang et al. (49) reported primary PAH patients with ACVRL1 mutation have rapid disease progression, high overall mortality rate and no response to the acute pulmonary vasodilation test.
In terms of non-WSPH-listed PAH related genes, we identified 14 novel sites in 12 patients ( Table 1). ABCC8 encodes SUR1, a regulatory subunit of the ATP-sensitive potassium channel (11). ABCC8 loss-of function mutations in both pediatric and adult-onset PAH had been reported (11,12). We identified a genetic variant of c.206dup, which is deemed likely pathogenic. Notch receptors, several ligands, and components of the downstream signaling machinery are expressed in the vessel and have been reported to be related with PAH (13,14,50). We identified several novel genetic variant sites in NOTCH1, NOTCH 2, NOTCH 3 and JAG1 in this PAH study cohort ( Table 1). However, we did not conduct further in vitro studies to prove their genetic impacts on the downstream cell signals.
This study had several limitations. First, the PAH patients were recruited from two medical centers in central Taiwan. Thus, the patient numbers were limited. Second, the PAH patients were enrolled for genetic analysis in a specific time frame between Mar. and Sep. 2021. Their clinical and hemodynamic data were then retrospectively traced back to their initial diagnosis date. Those who had the worst clinical scenario with rapid disease progression to death had no chance to be retrospectively enrolled in this genetic study. Third, we identified several novel variant sites for BMPR2 as well as other WSPH-listed or non-listed PAH-related genes. We only used software PolyPhen2 and SIFT to exam all the mis-sense mutations but did not have further in vivo studies to confirm that the genetic variants did have functional changes to the downstream proteins.
In conclusion, we identified 17.8% of patients with BMPR2 gene variants and 17.8% subjects with other WSPH-listed PAH-related gene variants in a Taiwanese idiopathic and heritable PAH cohort. In addition, we also identified 14 none WSPH-listed PAH-related genetic variants sites from 12 patient. PAH patients carrying BMPR2 variants presented at a younger age with a trend toward having higher mean PAP at initial diagnosis.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories (NCBI BioProject Accession: PRJNA824211).

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Taichung Veterans General Hospital, China Medical University Hospital (Taichung, Taiwan). The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
K-WL wrote the manuscript and analyzed/interpreted the data. S-KC performed the laboratory work. K-YW, K-WL, W-WL, W-JT, and Y-WC recruited the patients and critically reviewed the manuscript for important intellectual content. K-YW designed the study. All authors read and approved the final manuscript.

FUNDING
The whole exome sequencing, polymerase-chain reaction and Sanger sequencing of the genomic DNA were sponsored by Excelsior Biopharma Inc. and conducted at its Genetic Laboratory. The author S-KC is employed by Excelsior Biopharma Inc. The Excelsior Biopharma Inc. was not involved in the study design, collection, analysis, interpretation of data, and the writing of this article or the decision to submit it for publication.

SUPPLEMENTARY MATERIAL
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fcvm. 2022.911649/full#supplementary-material The quantification plot of copy number variation of A450 on BMPR2, ENG, and ACVRL1 genes using the SALSA MLPA Probemix P093 HHT/HPAH kit (MRC-Holland, Netherlands). BMPR2 5: the sample analytical number of A450. WT: the sample derived from normal subjects.