Differences in the expression of DNA methyltransferases and demethylases in leukocytes and the severity of pulmonary arterial hypertension between ethnic groups

Abstract The loss of ten‐eleven translocation (TET2) methylcytosine dioxygenase expression contributes to the pathobiology of pulmonary arterial hypertension (PAH). However, whether the expression and activity of other TETs and DNA methyltransferases (DNMTs) are altered in PAH remains enigmatic. Therefore, our objective was to determine the expression of DNMT (1, 3a, and 3b) and TET (1, 2, and 3) and their total activity. We assessed the expression of DNMT and TET enzymes in the leukocytes and their activity in extracellular vesicles (EVs). Expression of DNMT (1, 3a, and 3b), TET (2 and 3) in leukocytes, and total activity in EVs, from PAH patients was higher than in healthy controls. Additionally, we noticed there were difference in expression of these epigenetic enzyme based on ethnicity and found higher DNMT1 and lower TET2/TET3 expression in Caucasian than Hispanic/African American (combine) patients. Since loss‐of‐function mutation(s) and down‐regulation of TET enzymes are associated with hematological malignancies and cytokine production, we determined the expression of genes that encode cytokines in samples of Caucasian and Hispanic/African American patients. Expression of IL6, CSF2, and CCL5 genes were higher in the leukocytes of Caucasian than Hispanic/African American patients, and CSF2 and CCL5 negatively correlated with the decreased expression of TET3. Interestingly, the expression of gene encoding CD34, a marker of myeloid and lymphoid precursor cells, and CD163, a monocyte/macrophage protein, was higher in the leukocytes of Caucasian than Hispanic/African American patients. Furthermore, Hispanic/African American patients having higher TET2/TET3 expression had higher pulmonary capillary wedge pressure. In conclusion, our results revealed higher DNMT1 and lower TET2/TET3 in Caucasian than Hispanic/African American patients together potentially augmented genes encoding inflammation causing cytokines, and CD34+‐derived immunogenic cells, and the severity of PAH.


| INTRODUCTION
Evidence that alteration of DNA and histone methylation or acetylation (epigenetic modifications) are associated with the development of all types of pulmonary hypertension is accumulating (Napoli et al., 2019). It has been proposed, for example, that altered methylation and acetylation marks on histones produce a persistent inflammatory phenotype in fibroblasts and an apoptosisresistant and hyperproliferative phenotype of endothelial and smooth muscle cells from hypertensive patients and animal models (Hu et al., 2019). DNA methylation is regulated by DNA methyltransferases (DNMTs; 1, 3a, and 3b) and TET methylcytosine dioxygenases (TETs; 1, 2, and 3), DNA demethylases (Lachat et al., 2020).
In recent studies, we and others observed that differential DNA methylation and a reduction in TET2 mRNA expression may be key to the pathogenesis of pulmonary hypertension in mice and humans Potus et al., 2020). In those studies, our laboratory found that DNA methylationdependent transcription of pulmonary hypertension-related genes and Tet2 expression is suppressed in the lungs of hypoxic mice . Moreover, we observed that hypoxia-induced metabolic reprogramming contributes to the reduction in Tet2 expression, and that inhibiting glucose-6phosphate dehydrogenase activity increases pulmonary Tet2 expression in hypoxic mice and reduces pulmonary hypertension . Others have found that inherited and acquired TET2 abnormalities occur in only 0.39% of pulmonary arterial hypertension (PAH) cases, but lower expression of TET2 in mononuclear cells was detected, with elevated proinflammatory cytokines, in >86% of inflammation-associated PAH cases (Potus et al., 2020). Consistent with those findings, Tet2 deletion leads to increases in right ventricular (RV) pressure in 7-month-old, but not in 2-month-old, mice (Potus et al., 2020). Altogether these findings highlight an association between reduced TET2 and the pathogenesis of pulmonary hypertension in humans and mice.
While loss-of-function mutations in TET2 are associated with diverse blood cell malignancies in humans (Solary et al., 2014), loss of a single TET enzyme is not sufficient to efficiently promote malignancy and promote synthesis of pro-inflammatory cytokines . Although reduction of TET2 has been implicated as a contributing factor in the development of PAH, information regarding the expression of DNMT and other TET enzymes is differentially regulated in patients with PAH remains unclear. More so, the activity of these epigenetic writers and eraser enzymes, which are metabolically regulated, remains undetermined in PAH cases. Therefore, we assessed DNMT (1, 3a, and 3b) and TET (1, 2, and 3) expression in leukocytes as well as their activity in extracellular vesicles (EVs), submicron intact vesicles (Blair et al., 2016), isolated from plasma collected from scleroderma (scl) and idiopathic (i) PAH patients. Furthermore, whether differential expression of DNMTs or TETs influences maladaptive cytokine gene expression and severity of PAH is unknown. Therefore, we analyzed the expression of DNMTs and TETs enzyme in Caucasian, Hispanic, and African American patients with PAH. Our findings suggest that DNMT (1, 3a, and 3b), TET2, and TET3 are increased in humans with PAH. What is more, the results suggest that higher DNMT1 and lower TET2 and TET3 expression is directly correlated with increased cytokines (IL6 and CCL5) and severity of PAH in Caucasian and Hispanic/African American (combined) patients.

| Drugs and reagents
All chemicals and reagents were purchased from Sigma, Thermo-Fisher, R&D Systems, MyBioSource, and VWR.

| Collection of blood from control individuals and PAH patients
All protocols were approved by the New York Medical College and Westchester Medical Center Institutional Review Board (Protocol #11775). Patient demographics are presented in Table 1. Blood samples were collected into heparinized tubes from PAH patients and control individuals by venipuncture and stored at 4 °C for 24 h. White blood cells (WBCs/leukocytes) were isolated as described previously . Briefly, the cells were collected by centrifugation at 300 g for 5 min at 4 °C. The supernatant was aspirated, and the cell pellet was resuspended in 10 ml of red cell lysis buffer (0.15 M NH 4 Cl; 0.01 M KHCO 3 ; 0.1 M Na 2 EDTA, pH 7.2-7.4) and incubated on ice for 15 min. The cells were then collected by centrifugation (1200 g), washed twice with PBS and immediately suspended in Qiazol (700 μl; Qiagen), after which total RNA was isolated and used for real-time PCR as described below. We used leukocytes for gene expression analysis, as a surrogate for lung tissue or pulmonary artery, because they are increased in blood of PAH patients and a

| Quantitative real-time PCR
Real time PCR was used to analyze mRNA expression. Briefly, total RNA was extracted from lungs and human leukocytes using a Qiagen miRNeasy kit (Cat # 217004). The quality and concentration of the input RNA were measured on the Synergy HT Take3 Microplate Reader (BioTek) and cDNA was prepared using SuperScript™ IV VILO™ Master Mix with ezDNase™ Enzyme (Cat # 11766500, Invitrogen) for mRNA. Quantitative PCR was performed in duplicate using TaqMan™ Fast Advanced Master Mix (Cat # 44-445-57) for mRNA using a Mx3000p Real-Time PCR System (Stratagene). The primers for the qPCR were purchased from Thermo Fisher Scientific/ TaqMan. mRNA levels were normalized to internal control TUBA1A, and relative mRNA expression was reported.

| Isolation and identification of EV
Because we used the leukocytes to extract RNA for PCR, we could measure activity of epigenetic enzymes in the immune cells. Therefore, we isolated EVs from plasma collected from PAH patients and healthy controls as previously described  (Blair et al., 2016). Briefly, we generated platelet-rich plasma by centrifugation of samples at 1500 × g for 20 min at room temperature. The platelet-rich plasma was then used to generate platelet-free plasma (PFP) by centrifugation at 13,000 g for 10 min at room temperature. PFP was diluted 1:40 with PBS and ultracentrifuged for 60 min at 100,000 g at 4° in an Optima XE-100 Ultracentrifuge (Beckman Coulter). The supernatant plasma was then removed, and the recovered EV pellet was dissociated in PBS. We quantitated EVs using Zeta potential analysis and identified by transmission electron microscopy (TEM).

| Measurement of DNMT and TET activity
DNMT and TET activity was measured using kits from Epigentek. DNMT and TET activity were measured in EVs isolated from plasma collected from PAH patients and healthy controls.

| Isolation of leukocytes and flow cytometry
Blood samples were collected from PAH patients. After red blood cell lysis using lysing buffer (BD Biosciences), 10 6 cells suspended in 90 µl of buffer were treated with 10 µl of FcR blocking reagent (Miltenyi Biotec) for 10 min at 4 °C and stained with 10 µl of fluorescent antibodies for 15 min at 4 °C. We used fluorescein (FITC)-conjugated anti-CD34 antibody purchased from Miltenyi Biotec. Flow cytometry was performed as described before . Negative control (without) primary antibody treated cells were used each time as validation of antibodies.

| Statistical analysis
Statistical analysis was performed using GraphPad Prism 9 software. Values are presented as mean ± standard error (SE). Statistical comparisons of samples were made for two groups with Mann-Whitney test. Values of p < 0.05 were considered significant.

| Treatment and hemodynamic changes in scleroderma-associated and idiopathic PAH cases
We collected blood from eight scl-PAH and five i-PAH patients and seven healthy controls. We had four Caucasians and four Hispanics in the scl-PAH group, four Caucasians and one African American in the i-PAH group, and six Caucasians and one Indian American in the control group. Their ages ranged from 56 to 78 years in the PAH group and from 24 to 64 years in the control group (Table  1). Right atrial pressure (RAP), pulmonary arterial pressure (PAP), pulmonary capillary wedge pressure (PCWP), transpulmonary gradient (TPG), pulmonary vascular resistance (PVR), and cardiac output and index did not differ between the scl-PAH and i-PAH groups (Table 2).

| DNMT and TET expression and activity in leukocytes and EV from PAH patients
To determine whether the expression and activity of DNMTs were modified in PAH, we assessed expression of three major DNMTs (1, 3a, and 3b) in leukocytes and total DNMT activity in EVs from PAH patients and controls. Real-time PCR showed that DNMT3a and DNMT3b, but not DNMT1, expression was significantly higher in the PAH than the control group (p < 0.05; Figure 1a-c). Similarly, we measured the expression of TET enzymes. While expression of TET1 did not change, TET2 and TET3 increased in the PAH than the control group (p < 0.05; Figure 1d-f). We did not find a correlation between age and expression of DNMTs and TETs (data not shown). To determine whether there was a corresponding increase in DNMT and TET activity in PAH cases, we isolated EVs from PAH patients and controls. After using zeta potential analysis and electron microscopy to confirm the diameter and appearance of the isolated EV ( Figure 2a,b), we determined their associated enzyme activity. Total activity of DNMT and TET was significantly higher in EVs from the PAH patients ( Figure 2c,d).

Caucasian vs Hispanic/African American (combine) PAH patients
Since lower expression of TET2 in mononuclear cells was detected in >86% of inflammation-associated PAH cases (Potus et al., 2020), our findings suggest higher TET2 expression in PAH patients with scleroderma was somewhat perplexing and surprising to us. Therefore, to determine the potential cause for the differences in the results of the present and previous studies, we analyzed/compared the expression of DNMTs and TETs between different ethnic groups. Furthermore, along those lines, we noticed the difference in their expression based on the ethnicity and hence separated the patients based on ethnicity and in Caucasian and non-Caucasian groups to determine differences in the expression and activity of epigenetic enzymes and cytokines.
In the non-Caucasian group we combined Hispanics and African American patients. Interestingly, we found higher DNMT1 and lower TET2 and TET3 expression in leukocytes of Caucasians as compared with Hispanic/African American (combined) patients (Figure 3a-c), and this was not because of differences in the number of leukocytes between Caucasian and Hispanic/African American patients (Table 3).

Hispanic/African American (combine) PAH patients
Altered DNA methylation and reduced expression of TET enzymes have been implicated to elevate inflammatory cytokines and associated with hematological disorders or malignancies Solary et al., 2014). Furthermore, increased cytokine levels have been concurrently observed with the loss-of-function mutation in TET2 and reduction of TET2 expression in PAH patients (Potus et al., 2020). Therefore, we determined expression of IL6, CCL5, CCL2, and CSF2 genes in leukocytes of Caucasian and Hispanic/African American patients. We found Caucasian as compared with Hispanic/African American patients expressed more IL6 and CCL5 (Figure 4a,b). Next, to determine if elevation of these cytokines are dependent on altered DNMT1 and/or TET2/TET3 expression, we performed Pearson correlation analysis. Interestingly, while neither IL6 nor CCL5 depend on DNMT1 (Figure 4c,d) and TET2 (Figure 4e,f), expression of both the cytokine genes and especially CCL5 gene negatively correlated (p < 0.05) with TET3 expression (Figure 4g,h). Similarly, we found CSF2 expression was higher in Caucasian versus Hispanic/ African American patients and CSF2 expression negatively correlated (p < 0.05) with TET3 expression (Figure 4i-l). CCL2 expression did not differ between Caucasian and Hispanic/African American patients.

| CD163 gene levels positively correlates with the DNMT1 expression in Caucasian versus Hispanic/African American (combine) PAH patients
Next, we determined expression of genes encoding immunogenic cell markers and found higher expression of CD163 gene in leukocyte samples of Caucasian as  compared with Hispanic/American African patients ( Figure 5a). Furthermore, expression of CD163 depended on DNMT1 (Figure 5b) but not on TET2/TET3 ( Figure  5c,d). Additionally, although the expression of ITGAM, a gene that encodes integrin αM chain expressed on monocytes and macrophages, trended to be higher in Caucasian (0.10 ± 0.03) than Hispanic/African American (0.06 ± 0.01) patients, we did not find significant differences between the groups and correlation with the expression of DNMT1.

| CD34 + cells and CD34 gene levels in Caucasian versus Hispanic/African American (combine) PAH patients
CD34 + cells are precursors of immunogenic myeloid and lymphoid cells. These cells are increased in blood and lungs of PAH patients. Therefore, we determined whether these cells and the gene encoding CD34 protein are elevated in Caucasian compared to Hispanic/African American patients with PAH. As we expected, CD34 + cell numbers and the expression of the gene encoding CD34 were higher in Caucasian than Hispanic/African American patients ( Figure 5e,f) Furthermore, we found expression of CD34 and CD163 positively correlated (Figure 5g).

| Correlation between epigenetic enzymes and PCWP in Caucasian versus Hispanic/African American (combine) PAH patients
To determine whether there is any difference in the severity of PAH between Caucasian and Hispanic/African American patients, we compared the hemodynamic results. Our results revealed lower (p = 0.004) PCWP ( Figure 6a) and a trend toward higher PVR (Caucasian: 9.74 ± 1.49 vs. Hispanic/African American: 6.40 ± 0.89; in mmHg; p = 0.075; Table 4) in Caucasian than Hispanic/ African American patients. Next, to determine whether the expression of DNMT1 and/or TET2/TET3 has any bearing on the severity of PAH in Caucasian and Hispanic/ African American patients, we performed Pearson correlation analysis. We found PCWP positively correlated (p < 0.05) with the expression of TET2 and TET3 ( Figure  6b,c) and not with DNMT1 (data not shown).

| DISCUSSION
PAH is a heterogeneaous and complex angio-proliferative disease. PAH also exhibits cancer-like physiognomies, and hence some studies have characterized PAH as cancerlike disease (Culley & Chan, 2018;Leopold & Maron, 2016). One of the hallmarks of cancer is an alteration of DNA and histone methylation resulting in increased expression of oncogenes and/or decreased expression of tumor suppressor genes (Ehrlich, 2019;Samudio-Ruiz & Hudson, 2012). Increased methylation frequently silences DNA repair genes and miRs involved in cancer biology (Lakshminarasimhan & Liang, 2016). As in cancer,  Note: Mean ± SD. differential DNA and histone methylation is associated with different sub-types of PAH in experimental models and humans (Archer et al., 2010;Joshi et al., 2020;Ke et al., 2018;Napoli et al., 2019;Potus et al., 2015Potus et al., , 2020. Because methyltransferases and demethylases regulate the level of DNA and histone methylation and can thus alter gene transcription (Lachat et al., 2020), we investigated expression and activity of DNMT and TET enzymes in leukocytes and EVs from PAH patients. While DNMT1, DNMT3a, and DNMT3b expression increased in leukocytes from PAH patients as compared to healthy controls, our data showing higher DNMT1 expression and PCWP in Caucasian than Hispanic/ African American PAH patients suggests Caucasian patients with higher DNMT1 appear to have more severe PAH. Therefore, this implies augmented expression and activity of DNMTs, at least partly, contributed to the development and progression of PAH. In that regard, a recent study suggests expression of DNMT3b increases and other DNMTs decrease in lungs of congestive heart disease-associated PAH, and knockdown of Dnmt3b in rats augments monocrotaline and hypoxia induced pulmonary hypertension (Yan et al., 2020). Another recent study shows an association between up-regulation of DNMT1 and down-regulation of BMPR2 expression in lungs of PAH patients (Bisserier et al., 2021). Similarly, increased DNA methylation in the lungs of hypoxic mice reduces expression of genes encoding proteins, including BMPR1B and Krebs cycle enzymes, implicated in the pathogenesis of PAH . While increased DNA hypermethylation leads to suppression of genes (Sod2 and Nos3) encoding vasodilatory proteins and antioxidants in lungs of hypoxic mice, pulmonary artery SMCs of fawn-hooded rats, and endothelial cells of fetal lamb with pulmonary hypertension (Archer et al., 2010;Joshi et al., 2020;Ke et al., 2018). Furthermore, DNA hypermethylation and downregulation of miR-126 contributes to right heart failure in PAH (Potus et al., 2015), and play various roles in the regulation of genes during the development of atherosclerosis (Dong F I G U R E 4 Expression of cytokines in leukocytes from Caucasian than Hispanic/African American patients with sclerodermaassociated and idiopathic-PAH patients. Expression of genes encoding cytokines was compared between the different ethnic groups with scl/i-PAH. (a, b) IL-6 and CCL5 expression was higher in leukocytes from Caucasian as compared with Hispanic/African American patients. Pearson's correlation between (c, d) DNMT1, (e, f) TET2 and (g, h) TET3, and cytokines in leukocytes from Caucasian than Hispanic/African American scl/i-PAH patients is shown. (i) CSF2 expression was higher in leukocytes from Caucasian as compared with Hispanic/African American patients. Pearson's correlation between (j) DNMT1, (k) TET2 and (l) TET3, and CSF2 expression in leukocytes from Caucasian than Hispanic/African American scl/i-PAH patients is shown. A significant negative correlation between decreased TET3 and CSF2 was observed. N = 6 in the Caucasian group and N = 5 in the Hispanic (N = 4)/African American (N = 1) group. Comparisons were made using Student's t-test with Welch's correction. *p < 0.05 and **p < 0.005 . Because DNMT1 is the key maintenance methyltransferase, and DNMT3a and DNMT3b enzymes methylate unmethylated and hemimethylated CpG islands, we suggest increased DNMT3a and DNMT3b expression and activity potentially leads to de novo hypermethylation of genes in PAH patients. Furthermore, our results suggest higher DNMT1 and lower TET2 and TET3 expression in Caucasian patients caused hypermethylation and downregulation of more protective genes genome-wide, including those mentioned earlier, increasing severity of PAH. In addition to increased expression of DNMT (1, 3a, and 3b), our results revealed PAH patients as compared with healthy controls had higher total DNMT activity, which is driven by metabolites of one-carbon metabolism, the polyamine pathway, and the Krebs cycle (Puleston et al., 2017;Solary et al., 2014;Stover, 2009). Each of those pathways is reprogrammed in various cell types residing in the pulmonary artery and lungs of PAH patients and animal models (D'Alessandro et al., 2018). Onecarbon metabolism and polyamine pathway metabolites (s-adenosylmethionine and spermine, respectively) are methyl-group donors and regulators of DNA methylation (Amelio et al., 2014;Puleston et al., 2017) and are elevated in lungs/fibroblasts/SMCs of PAH patients . Along with higher expression and activity of DNMT, we found increased expression of TET2 and TET3 and total TET activity in PAH patients than control individuals. These findings are somewhat perplexing because lower expression of TET2 in mononuclear cells was detected in >86% of inflammation-associated PAH cases (Potus et al., 2020) and smooth muscle cells isolated from i-PAH patients . Although reasons for these differences in our results and previous studies are unclear, different source of genetic material (leukocytes vs. mononuclear cells/smooth muscle cells), methodologies (real-time PCR vs. microarray) employed to measure gene expression, disease state, and race of the patients, presumably contributed to the observed differences. TET2 expression levels were higher in Hispanic/African American than Caucasian patients who had more severe PAH. Therefore, we propose TET2 and TET3 expression and activity are increased to potentially antagonize actions of elevated DNMTs or vice versa in PAH patients.
Next, because loss-of-function mutation in TET2 and reduction of TET2 expression is associated with cytokine production in PAH cases (Potus et al., 2020), we determined the expression of genes that encode cytokines in samples of Caucasian and Hispanic/African American patients. We found higher CSF2 and CCL5 levels significantly correlated with decreased TET3, but not with TET2 and DNMT1, in Caucasian and Hispanic/African American patients. Increased IL6 did not correlate with DNMT1 and TE2/TET3, indicating IL-6 expression may not be directly regulated by altered DNA methylation status in Caucasian and Hispanic/African American patients. Interestingly, augmented CD163 gene correlated with the expression of DNMT1 but not with the expression of either TETs. Although DNMT1 is the key maintenance methyltransferase, down regulation miR-124 by increased DNMT1 facilitates M1 alveolar macrophage polarization in acute lung injury (Wang et al., 2021). Furthermore, we observed significantly higher CD34 + cells and CD34 gene expression in Caucasian than in Hispanic/African American patients. Since the expression of CD34 and CD163 were positively correlated, this potentially implies increased CD34 + cells convert to CD163 gene expressing monocytes, and decreased TET3 led to the upregulation of CSF2 and CCL5 in Caucasian and Hispanic/African American patients. Indeed, the reduction of TET2 and TET3 expression is responsible for hematologic disorders/malignancies and for increasing cytokine production Solary et al., 2014). Inflammation caused by elevated cytokines, in perivascular region of the pulmonary artery, contributes to the remodeling of the pulmonary artery and increase pulmonary vascular resistance (El Kasmi et al., 2014). This is reflected by a positive correlation between PCWP and TET2/TET3 expression. Therefore, it is not unreasonable to suggest that lower TET2/TET3 expression contributed to the severity of PAH in Caucasian than in non-Caucasian (Hispanic/ African American) patients. Although our findings clearly suggest differences in the expression and activity of epigenetic writers and erasers are potentially associated with the severity of PAH, these results should be cautiously interpreted as they could be confounded by the treatment for PAH. However, this is likely to be a minor point because there was no difference in expression of all genes, we examined, in Hispanic patients that were on treatment and one patient on no treatment. Furthermore, a small sample size in Caucasian and non-Caucasian (Hispanic/African American) precludes us from affirmatively suggesting that differences in the epigenetics between the groups is a cause of the severity of PAH, the statistically significant differences between the groups certainly hints DNA methyltransferases and demethylases must be important to eliciting the severity of PAH in Caucasian group and this must be further studied. Nevertheless, DNMT and TET activity in circulating EVs could be further developed as diagnostic markers for determining the severity of PAH in different ethnic groups. EVs are cargos that transport second messengers, genetic information, and proteins (Blair et al., 2016;Sayner et al., 2019). They are emerging diagnostic tools. Therefore, further multicenter studies in more PAH patients with diverse ethnic backgrounds are needed to identify if circulating EVs are released from the lungs or leukocytes, and to correlate the activity of epigenetic enzymes in EVs with the severity of the disease for diagnostic purposes and determining the effectiveness of the therapy in PAH patients.
In summary, our findings support the hypothesis that increased DNMT-catalyzed DNA methylation potentially contributes to the development and progression of PAH pathology in humans. Moreover, we have identified a potential association between expression and activity of increased DNMT1 and decreased TET2/TET3 and the excited inflammatory cytokine production and severity of PAH in Caucasian versus Hispanic/African American patients. Although some analyses suggest minorities may have a poor outcome in PAH, the role of ethnicity/race in PAH remains a controversial topic (Medrek & Sahay, 2018). Based on our findings, additional multicenter studies with more patients in different ethnic groups are warranted to confirm these findings and to develop new ethnicity/race-based therapies for PAH.