Increased Methyl-CpG-Binding Domain Protein 2 Promotes Cigarette Smoke-Induced Pulmonary Hypertension

Pulmonary hypertension (PH) is a chronic vascular proliferative disorder. While cigarette smoke (CS) plays a vital part in PH related to chronic obstructive pulmonary disease (COPD). Methyl-CpG-Binding Domain Protein 2 (MBD2) has been linked to multiple proliferative diseases. However, the specific mechanisms of MBD2 in CS-induced PH remain to be elucidated. Herein, the differential expression of MBD2 was tested between the controls and the PH patients’ pulmonary arteries, CS-exposed rat models’ pulmonary arteries, and primary human pulmonary artery smooth muscle cells (HPASMCs) following cigarette smoke extract (CSE) stimulation. As a result, PH patients and CS-induced rats and HPASMCs showed an increase in MBD2 protein expression compared with the controls. Then, MBD2 silencing was used to investigate the function of MBD2 on CSE-induced HPASMCs’ proliferation, migration, and cell cycle progression. As a consequence, CSE could induce HPASMCs’ increased proliferation and migration, and cell cycle transition, which were suppressed by MBD2 interference. Furthermore, RNA-seq, ChIP-qPCR, and MassARRAY were conducted to find out the downstream mechanisms of MBD2 for CS-induced pulmonary vascular remodeling. Subsequently, RNA-seq revealed MBD2 might affect the transcription of BMP2 gene, which furtherly altered the expression of BMP2 protein. ChIP-qPCR demonstrated MBD2 could bind BMP2’s promotor. MassARRAY indicated that MBD2 itself could not directly affect DNA methylation. In sum, our results indicate that increased MBD2 expression promotes CS-induced pulmonary vascular remodeling. The fundamental mechanisms may be that MBD2 can bind BMP2’s promoter and downregulate its expression. Thus, MBD2 may promote the occurrence of the CS-induced PH.


INTRODUCTION
Pulmonary hypertension (PH) is a complex vascular disease characterized as a rise in mean pulmonary arterial pressure (mPAP) of ≥ 25 mmHg at rest, which is measured by right heart catheterization (RHC) (1,2). It is a multifactorial pathophysiological disorder caused by increased pulmonary artery pressure, resulting in chronic elevation of pulmonary arterial resistance, right ventricular (RV) hypertrophy, RV failure, and ultimately death (3)(4)(5).
Data show that the incidence of PH in chronic obstructive pulmonary disease (COPD) is high (6). PH is often present as a result of hypoxia associated with COPD. However, hypoxia can't account for all PH related to COPD. An increasing amount of evidence suggests that cigarette smoke (CS) may directly impact on the pulmonary vasculature system (7)(8)(9)(10). The pathogenesis of PH includes pulmonary vascular remodeling and vasoconstriction, while pulmonary vascular remodeling is the fundamental mechanism in chronic PH. CS is closely associated with pulmonary vascular remodeling (11). Multiple biological processes, such as cell proliferation, migration, and cell cycle progression of pulmonary artery smooth muscle cells (PASMCs), are associated with pulmonary vascular remodeling in PH. Among these, PASMCs proliferation is the key component of the pulmonary vascular remodeling that occurs in PH (12). Cigarette smoke extract (CSE) can directly lead to PASMCs proliferation (13), which results in CS-induced pulmonary vascular remodeling. Hence, focusing on the pathogenesis of CS-induced PASMCs proliferation is one of the objectives in the development of treatment strategies for PH related to COPD. However, the specific molecular mechanisms need to be further investigated.
DNA methylation is the most common manifestation of epigenetics (13). Compelling evidence has been linking epigenetics to PH (12,14,15). Smoking or environmental CS exposure significantly modifies gene methylation, which is associated with diseases like COPD (16,17). One of the regulatory elements of DNA methylation is the methyl-CpGbinding domain protein family (the MBD protein family). The MBD2 protein is part of the MBD protein family. Among all the members, MBD2 has exhibited the highest binding affinity with methylated DNA (18,19). As a reader in DNA methylation, MBD2 can mediate transcriptional repression or activation through directly combining methylated DNA or gathering proteins to form a suppressive compound (18,20,21). Furthermore, MBD2 is associated with many proliferative diseases (18,22,23).
As a proliferative illness, PH is closely related to many different tumor-related genes. However, limited research has been conducted on the molecular mechanisms involved in CSinduced PH. As an element with multiple functions, MBD2 may be involved in the pathogenesis of CS-induced PH. Nonetheless, the critical role of MBD2 in CS-induced PH has not been clarified yet. Thus, in this study, the role of MBD2 in CSinduced pulmonary vascular remodeling was investigated.

Human Tissue
Human pulmonary arteries were obtained from patients undergoing surgery for pulmonary lump at Tongji Hospital, Tongji Medical College of Huazhong University of Science and Technology, Wuhan, China. Participants were assigned into the PH group and the control group according to their medical history, pulmonary function, and echocardiography. The PH group's clinical diagnosis was mild to moderate COPD without hypoxemia. All enrolled PH patients had smoking history. All the included participants underwent echocardiography, and the peak tricuspid regurgitation velocity (TRV) was greater than 3.0 m/s in PH patients (1,24). The exclusion criteria were comprised of left ventricular systolic or diastolic dysfunction, coronary heart disease, other chronic lung diseases, diabetes mellitus, infections, autoimmune diseases, severe hepatic, endocrine, or renal diseases, and medical treatment, including immunosuppressive agents or corticosteroids (25)(26)(27)(28). Prior to enrollment, written informed consent was signed, and this study was approved by the Ethics Committee of Tongji Hospital. The baseline characteristics of the PH patients and controls are shown in Table 1.

Western Blot Analysis
Total proteins were obtained from HPASMCs and human pulmonary arteries. The concentration of proteins was assessed

Animal Models
Adult male Sprague-Dawley (SD) rats were obtained from the Experimental Animal Center of Tongji Hospital (Wuhan, China). Animal experiments were performed in line with the Animal Care and Use guidelines of the Chinese Council on Animal Care. Rats were randomly divided into two groups, the air group and the smoking group. The smoking groups were exposed to CS generated by ten Hong Jin Long cigarettes as previously noted (13). Rats' lung sections (CS models and controls) came from Li, Q and Wu, J et al., and the phenotype data (Supplementary Figure 1) was acquired as previously described (13).

Preparation of Cigarette Smoke Extract (CSE)
Cigarette smoke extract (CSE) was freshly prepared by bubbling the smoke coming from two research cigarettes (Code 3R4F, University of Kentucky, USA), at a frequency of 1 cigarette/ 5 min, into a conical tube of 50 ml which contained 20 ml of culture medium, aided by vacuum extraction. A 0.22 mm filter was used to filter the extract, which was considered to be CSE of 100% concentration (13,30,31).

CCK8 Assay, Cell Cycle Assay, and EDU Assay
To evaluate the role of MBD2 in the proliferation of HPASMCs, HPASMCs were treated for 24h with MBD2si and then incubated for 24h with 2% CSE. A Cell Counting Kit-8 (Dojindo, Japan) and a cell cycle detection kit (KeyGEN BioTECH, China) were used to assess cell proliferation in accordance with instructions of the manufacturer. An ELx800 Universal Microplate Reader (Bio-Tek Instruments, Inc., Winooski, VT) and a flow cytometer (BD Bioscience, USA) were used to conduct CCK8 and cell cycle analyses, respectively. Furthermore, HPASMCs proliferation was also measured by 5-ethynyl-2′-deoxyuridine (EdU) staining (RiboBio, China). Apollo 567 was used to label EdU, and the cells were examined through a fluorescence microscope.

Transwell Assay
To perform migration assays, membranes with 8-mm pores (Corning Costar, Cambridge, MA) in 24-well Transwell ® plates were used. Prior to being digested and counted, the HPASMCs were first transfected with MBD2si for 24h. Approximately 10,000 cells were transferred to the upper chamber to adhere. Next, the medium in the upper chamber was substituted with 200 ml fresh medium containing 1% FBS, while 700 mL of medium with 20% FBS with or without 2% CSE was supplemented to the lower compartment. After 24h, the cells found in the membranes of the chambers at the bottom were fixed, stained with crystal violet, and then imaged.

RNA Sequencing (RNA-Seq)
RNAs were extracted by using the TRIzol reagent according to the manufacturer's protocol. The Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) was used to assess RNA integrity. Then the TruSeq Stranded mRNA LT Sample Prep Kit (Illumina, San Diego, CA, USA) was applied to construct the libraries based on the manufacturer's instructions. The transcriptome sequencing and subsequent analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China). The libraries were sequenced on an Illumina HiSeq X Ten platform, and 150 bp paired-end reads and raw reads for each sample were generated. Raw reads in fastq format were first processed with Trimmomatic; subsequently, the low-quality reads were eliminated to retrieve the clean reads for each sample for further analyses. HISAT2 was used to map the clean reads to the human genome (GRCh38). Cufflinks were used to calculate the FPKM of each gene, and HTSeq-count was used to obtain the read counts of each gene. The DESeq (2012) R package was used to perform differential expression. A P-value < 0.05, foldchange > 2, or foldchange < 0.5 were set as the thresholds for significantly differential expression. Chromatin Immunoprecipitation Quantitative PCR Assays (ChIP-qPCR)

MassARRAY
ChIP assay was conducted with technical support by Genecreate (Wuhan, China). The cells were crosslinked with formaldehyde and then quenched with glycine. Then, the cells were centrifugated, washed, and harvested. The cell pellets were lysed in lysis buffer and sonicated. After centrifugation, the supernatant was diluted in dilution buffer, and part of the samples were taken as the input. Then, the immunoprecipitation was done by incubating overnight with MBD2 antibody (Abcam) or the control rabbit IgG antibody. Then, wash buffer was applied. Then, the samples were eluted and the crosslinking was reversed. Finally, the DNA was purified. The immunoprecipitated chromatin was quantified by RT-qPCR. The procedures of RT-qPCR were conducted as previously described.
The primer sequences used for ChIP-qPCR are shown in Table 3.

Statistical Analysis
All data were presented as mean ± standard deviation (SD). The results were analyzed with GraphPad Prism 8 (GraphPad Software Inc, San Diego, CA, USA). Statistically significant differences were determined using the one-way analysis of variance (ANOVA) or Student's t-test. The RNA-seq and the MassARRAY assay were assessed by paired test, while other assays were assessed by un-paired test. P < 0.05 was regarded as statistically significant.

MBD2 Protein Expression Was Increased in the Pulmonary Arteries of PH Patients and CS-Exposed Rats
The MBD2 protein expression was assessed by Western Blot. The expression of MBD2 in human pulmonary arteries of the  PH groups was higher than that in the normal controls ( Figures 1A, B). To further verify our findings, we conducted an immunofluorescence assay with tissue slices, and the MBD2 of the PH groups was confirmed to be higher than that in the controls in the pulmonary arteries of humans and CS-exposed rats ( Figures 1C-E).

MBD2 Protein Expression Was Elevated in HPASMCs After CSE Stimulation
To examine the expression of the MBD2 protein by CSE stimulation in HPASMCs, we experimented with different concentrations (0%, 0.5%, 1%, 2%) and durations (0h, 12h, 24h, 48h) of CSE stimulation. Interestingly, we found that the expression of MBD2 was increased in the CSE stimulation groups compared with the control groups (Figures 2A-D, Supplementary Figure 2). Simultaneously, we carried out the immunofluorescence experiment to further verify the previous results. The outcomes indicated that the MBD2 protein was mainly expressed in the cell nucleus, and stimulation with CSE also elevated its expression ( Figures 2E, F).

MBD2si Suppressed CSE-Induced HPASMCs Proliferation and Migration
As PH is a proliferative disease, we mainly focused on the proliferative ability of HPASMCs. We used MBD2si to discover the specific functions of the MBD2 protein in PH. Firstly, we designed three different sequences of siRNA and verified the transfection effect of the MBD2si (Figures 3A, B), and we found that the MBD2si1 was the most effective. As a result, we chose MBD2si1 for the subsequent experiments. Next, we performed a CCK8 assay to assess the effect of MBD2si on CSE-induced HPASMCs proliferation. As the HPASMCs proliferated due to the CSE, MBD2si prevented CSE-induced HPASMCs proliferation ( Figure 3C). To further verify the effect of MBD2si on HPASMCs proliferation, we performed the EDU assay and found that MBD2si inhibits the multiplication of HPASMCs (Figures 3D, E). Simultaneously, Ki67 immunofluorescence was performed, and results similar to the EDU assay were discovered ( Figures 3F, G). Then, we tested the PCNA protein in the HPASMCs and found that MBD2si could suppress the upregulation of PCNA due to CSE stimulation ( Figures 3H, I). Finally, we did a transwell assay and found that MBD2si could also inhibit CSE-induced migration of the HPASMCs (Figures 3J, K).

MBD2si Inhibited CSE-Induced HPASMCs Cell Cycle Transition
CSE can result in the cell cycle transition, including a decreased G0/G1 phase populations, along with increased S and G2/M phase populations. In our findings, MBD2si could suppress CSE-induced cell cycle alteration, thus raising CSE-induced G0/G1 phase populations, reducing CSEinduced S phase and G2/M phase populations ( Figures 4A, B). According to our previous study (32)(33)(34), CSE can increase the cyclin D1 protein expression in HPASMCs. Cyclin D1 may play an effective role in the CSE-induced cell cycle transition change. Therefore, we tested the cell cycle protein, cyclin D1. The results indicated that the cyclin D1 expression in HPASMCs was elevated through CSE stimulation, while MBD2si prevented the increase in cyclin D1 protein and CCND1 mRNA ( Figures 4C-E).

Expression of the BMP2 Was Elevated After Treatment With MBD2si
To evaluate why MBD2si can affect the CSE-induced proliferation, migration, and cell cycle of HPASMCs, we conducted the RNA-seq assay to determine the downstream gene of MBD2. Surprisingly, we found that MBD2 was linked to the BMP2 gene ( Figure 5A, B). Therefore, to further verify the results of the RNA-seq in the HPASMCs, we used RT-qPCR and WB assay to confirm. The results showed that CSE could lead to BMP2 downregulation, while MBD2si could increase the expression of BMP2 ( Figure 5C-E, Supplementary Figure 3), which indicated that MBD2 could change the transcription of BMP2 gene and its protein expression.

MBD2 Could Bound to BMP2's Promoter Region and Regulated Its Expression
As a reader for interpreting DNA methylation encoded information, MBD2 tended to function by directly binding methylated CpG DNA and forming an inhibitory complex or crosstalk with the histone-modifying and chromatin-remodeling enzymes to achieve a follow-up transcription program (35). To investigate the relationship between MBD2 and BMP2, we designed the overexpression adenovirus vector of MBD2 (MBD2ad, Figures 6A, B), and used MBD2si as previously described. We assumed MBD2 could directly bind to the promoter area of the BMP2 gene in HPASMCs. Therefore, we performed a ChIP-qPCR assay to verify our hypothesis. We designed seven primers according to the BMP2's promoter CpG island region. The results confirmed our hypothesis by showing that MBD2 could bind to the BMP2's promoter region ( Figure 6C). Futhermore, we assumed that MBD2 could influence BMP2's methylation status. We conducted a MassARRAY assay to confirm our hypothesis. Before the MassARRAY assay, the BMP2 and the MBD2 mRNA of the samples were evaluated by RT-qPCR ( Figure 6D), which indicated that MBD2 could directly influence BMP2's transcription in its specific regulation mode. After the assay, we discovered that the methylation of the BMP2 gene was not changed by manipulating MBD2 (Figures 6E, F), indicating that MBD2 exerted its transcription repression not by regulating gene's methylation status but by bind its promoter and gathering proteins to form a suppressive compound.

DISCUSSION
Our results indicate the role of increased MBD2 expression in the development of CS-induced pulmonary vascular remodeling. Herein, MBD2 were increased in the pulmonary arteries of the PH patients and CS-exposed rats. The in vivo findings showed that MBD2 increased along with the concentration and stimulation time of CSE. Then, it was found that the absence of MBD2 in HPASMCs could reduce the proliferation and migration caused by CSE. Furthermore, a deficiency in MBD2 could decrease the upregulation of cyclin D1 induced by CSE, and reverse the cell cycle transition after CSE stimulation. To investigate the possible mechanisms of those effects, we performed an RNA-seq assay and found that MBD2 could influence the transcription of BMP2 gene, which furtherly changed the expression of BMP2 protein. In addition, we verified with a ChIP-qPCR assay that MBD2 could bind directly to the promoter of BMP2. All in all, our research shows that MBD2 may act as a potential therapeutic target for CS-induced PH.
Of the epigenetic changes, DNA methylation is a very notable category and can profoundly affect cell function. Epigenetics is associated with CS as well as PH (11,36). DNA methylation is featured by covalently adding a methyl group on the fifth carbon of cytosine, which forms 5-methylcytosine. This structure enables binding of the methyl binding domain proteins, followed by the recruitment of the histone-modifying and chromatin-remodeling enzymes, and leading to a chromatinrepressing structure and inhibited gene expression (37). The family of the methyl-CpG-binding domain proteins functions as a methylated reader by binding directly to the methylated CpG DNA (38). MBD2 belongs to the family of the methyl-CpGbinding domain proteins and has been connected to multiple diseases (18,19,39,40), such as cancers and inflammations. As a proliferative disease comparable to tumors, PH may also be influenced by multiple tumor-associated genes, including MBD2. Other biomolecules interacted with MBD2 are broadly reported in PH relevant research. For instance, MBD2 is believed to repress the MYC gene transcription, thus suppressing glycolysis (41,42), and MBD2 is also reported to interplay with non-coding RNA related to PH (23). Whereas in other cases, MBD2 is also associated with many proliferative diseases. For example, MBD2 is associated with cancer-related methylations in prostate cancer (43), and it is also interrelated to chronic myeloid leukemia (44). Furthermore, it is also stated that MBD2 can correlate with cancer angiogenesis (45). In most cases above, the expression of MBD2 elevated, while in some other instances, MBD2 decreased. Sometimes, MBD2 acted as a pathogenic factor, while in other cases, MBD2 served as a protective factor. MBD2 is conventionally considered a transcriptional repressor that combines inhibitory complexes (46), while in some other research, MBD2 has been regarded as a transcriptional activator that activates gene expression (47). However, it activates some regulative genes to inhibit the downstream functional genes' transcription. Various studies have shown that methylation is multifunctional (38). Therefore, we consider its transcriptional activation may be the other manifestation of its transcriptional inhibition function. Our previous research showed that DNA methylation plays a vital part in CS-induced PH (13). Therefore, we hypothesized that MBD2 might also play a role in CS-induced PH. We conducted assays to detect the expression of MBD2 in the pulmonary arteries of the PH patients and CS-exposed rats. We also performed assays to investigate MBD2 in HPASMCs following CSE stimulation. The results indicated that MBD2 was elevated in the situations mentioned above. Our results are similar to most studies associated with MBD2 in diseases. Its elevated expression is similar to that in other proliferative diseases, suggesting that it may function as a pathogenic factor in CS-induced PH. According to the basic functions of MBD2, elevated MBD2 protein may operate by binding methylated DNA and inhibiting the transcription of the downstream functional genes. To further verify the specific effects of MBD2 in the HPASMCs, functional assays were performed.
In healthy adults, PASMCs are in their differentiated contractile state and exhibit low synthetic and proliferative activity (48). However, like other vascular smooth muscle cells (VSMCs), PASMCs are not terminally differentiated and can undergo a phenotypic transformation and change their state to a de-differentiated phenotype with a high rate of proliferation, migration, and protein synthesis in response to the environment (48,49). CS stimulation is such an environment. Among various cellular processes, PASMCs proliferation plays a central role in the pulmonary vascular remodeling in PH (50,51). We performed cell proliferation related assays to detect the proliferative process of the HPASMCs. We conducted CCK8 assay and EDU assay, and found that MBD2 interfering could attenuate the increased proliferation of the HPASMCs caused by CSE stimulation. The CSE stimulation can increase proliferative markers, such as PCNA and Ki67, and our results indicated that MBD2 interference could diminish those effects. Simultaneously, we tested migration and cell cycle progression functions. We found that MBD2si could decrease CSE-induced increased migration and prevent HPASMCs from cell cycle transition caused by CSE, which is consistent with the reported conclusion that MBD2 interference can attenuate migration and decelerate cycle transition (18,52). According to our previous research (32-34, 53, 54), cyclin D1 increases in CS-induced PH, which plays a vital role. Furthermore, studies have shown that MBD2 protein knockdown can decelerate cell cycle progression (18,52), and MBD2si can reduce the expression of cyclin D1 protein (18). Therefore, we concentrated further on cyclin D1 protein. In our study, CSE prominently increases the cyclin D1 protein and CCND1 mRNA in HPASMCs, while MBD2si can decrease those changes markedly. The ability of MBD2si to diminish cyclin D1 mechanistically explains its function to CSEinduced cell cycle transition. In conclusion, according to our findings, MBD2 interference can prominently attenuate CSEinduced HPASMCs' proliferation, migration, and cell cycle transition. The interruption of those functions by MBD2si implies that it can attenuate CS-induced PH and serve as a therapeutic target for it. However, the underlying mechanisms remain to be further investigated.
As a member of the methyl-CpG-binding domain proteins, MBD2 can combine methylated DNA and possess the highest affinity (18,19). Thus, we hypothesized that MBD2si could attenuate CS-induced PH by reducing MBD2's combination to the genes that exert antiproliferative influences in HPASMCs. We performed RNA-seq to search for the downstream genes of MBD2. We compared the CSE and the MBD2si with CSE group and acquired various differential genes. After looking through multiple articles, we eventually focused on the BMP2 gene. BMP2 is part of the transforming growth factor-beta (TGF-beta) superfamily. It is extensively reported to be involved in PH as its receptor BMPR2 (1,15,(55)(56)(57)(58). The research has revealed that BMP2 can be originated from different cell types such as HPASMCs and endothelial cells. It can be endogenously expressed in HPASMCs, which can be synthesized and secreted to the intercellular space and can activate BMP2 receptors on adjacent HPASMCs (59). It is also reported to correlate with methylation and CS (60)(61)(62). After treatment of 5-Aza-2'-deoxycytidine (5Aza), BMP2 mRNA was reported to increase in the 5Aza treating group (61). An in vivo study by Andersen et al. suggested that BMP2 has a protective role in PH (63). It has also been reported that BMP2 has an antiproliferative role in PASMCs of secondary PH patients and normal subjects (59,64). In our study, we mainly focused on the secondary (CS-induced) PH, and most literature tends to report the antiproliferative role of BMP2 in PASMCs (57)(58)(59). There is literature stating that BMP6 is down-regulated in CS-induced COPD (65). To further verify all the data and hypotheses, we performed the afterward experiments. RT-qPCR and WB were conducted to investigate whether BMP2 was actually influenced by the CSE and MBD2 to verify the results of the RNA-seq. Although there was no literature in which MBD2 is associated with BMP2, it has been stated already that MBD2 could control the BMP4 response and DNA methylation may have a role BMP4 signaling (66), while BMPR2 also seemed to be related with hypermethylation in heritable PH (15), which is all similar to our findings. Therefore, the hypothesis that MBD2 can directly bind to the BMP2 promoter seemed worthy of investigation. Hence, a ChIP-qPCR assay was performed to determine whether MBD2 could bind to BMP2 promoter, and the conclusion was that MBD2 indeed could bind to the BMP2 promoter in multiple CpG sequence sites. For verification of whether MBD2 could affect DNA methylation, MassARRAY assay was performed. The findings suggested that MBD2 didn't directly change the methylation of BMP2 promoter. In sum, MBD2 can influence CS-induced PH by binding the BMP2 gene and downregulating its expression.
There are some limitations to our study. Although we indicate that the increased MBD2 promotes cigarette smoke-induced pulmonary vascular remodeling in in vitro assays, the smooth muscle cell-specific MBD2 deficient mice still should be designed to further prove the assumption. However, the MBD2 conditional knockout mice need more time to generate.

ETHICS STATEMENT
The investigations of human participants were reviewed and approved by the Human Research Ethics Committees of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology. The participants gave their written informed consent to join the study. The animal experiments were reviewed and approved by the Animal Care and Use Committees of Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology.

AUTHOR CONTRIBUTIONS
JGX and JX designed the research and revised the manuscript. JW performed most of the experiments, contributed to the conception, and drafted the manuscript. QH, QL, YG, and YZ took part in cell culture experiments and analyzed and discussed the data. TW, JC, ZZ, YL, and JZ helped collect the human tissue samples and analyzed data. All authors contributed to the article and approved the submitted version.