Right Ventricle and Epigenetics: A Systematic Review

There is an increasing recognition of the crucial role of the right ventricle (RV) in determining the functional status and prognosis in multiple conditions. In the past decade, the epigenetic regulation (DNA methylation, histone modification, and non-coding RNAs) of gene expression has been raised as a critical determinant of RV development, RV physiological function, and RV pathological dysfunction. We thus aimed to perform an up-to-date review of the literature, gathering knowledge on the epigenetic modifications associated with RV function/dysfunction. Therefore, we conducted a systematic review of studies assessing the contribution of epigenetic modifications to RV development and/or the progression of RV dysfunction regardless of the causal pathology. English literature published on PubMed, between the inception of the study and 1 January 2023, was evaluated. Two authors independently evaluated whether studies met eligibility criteria before study results were extracted. Amongst the 817 studies screened, 109 studies were included in this review, including 69 that used human samples (e.g., RV myocardium, blood). While 37 proposed an epigenetic-based therapeutic intervention to improve RV function, none involved a clinical trial and 70 are descriptive. Surprisingly, we observed a substantial discrepancy between studies investigating the expression (up or down) and/or the contribution of the same epigenetic modifications on RV function or development. This exhaustive review of the literature summarizes the relevant epigenetic studies focusing on RV in human or preclinical setting.


Method 1.Search Strategy and Study Selection
All the published data were searched and collected between the inception of the study and 1 January 2023, according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) recommendations for systematic reviews [1].The electronic search was conducted on PubMed and designed to capture maximum sensitivity for the right ventricle (RV) and epigenetic modification (DNA methylation, histone modification, long non-coding RNA, microRNA, and chromatin modification).A comprehensive description of the search strategy is available in the supplementary materials (Figure S1, Table S1).
The following inclusion criteria were used to obtain more specific search results: original research articles accepted or published with availability in electronic databases by 1 January 2023, and articles only in English.Review articles and conference abstracts were excluded.The articles that did not directly focus on epigenetic modifications and the RV in healthy, developmental, and pathological conditions were excluded.

Introduction
The right ventricle (RV) is a thin-walled crescent structure of the heart, connecting the systemic venous return to the pulmonary circulation.In the physiologic condition, the RV works in a low-pressure system that receives venous blood from the atrium through the tricuspid valves and expulses it to the pulmonary artery through the outflow tract.The RV shape and pressure stress make it significantly different to the cone-shaped left ventricle (LV), which is adapted for a high-pressure system.Moreover, RV and LV have distinct embryologic origins.The RV (and outflow track) emerges from the second heart field whilst the LV, the muscular interventricular septum, and the atria rise from the first heart field [2].Consequently, the RV has a more distinct transcriptomic signature than the LV or atria [3,4].This specificity precludes any extrapolation of the knowledge from the LV to the RV.
In terms of physiopathology, RV dysfunction is associated with ischemic, non-ischemic (e.g., arrhythmogenic cardiomyopathy), congenital (e.g., tetralogy of Fallot (TOF), pulmonary stenosis), and arrhythmogenic cardiomyopathy (ACM).Moreover, RV dysfunction and failure are often the culminating point of chronic lung (e.g., chronic obstructive pulmonary disease (COPD) and pulmonary hypertension (PH), amongst others) and left heart diseases.Whatever the underlying condition, impaired RV function is almost systematically associated with patients' decreased quality of life and increased morbidity and mortality [5][6][7].However, until recently, the RV received far less attention than the LV and most molecular knowledge on RV dysfunction was extrapolated from the LV.
Functionally, RV dysfunction and failure are associated with drastic changes involving, among others, cardiomyocyte hypertrophy, metabolic reprogramming, increased fibrosis and inflammation, and impaired angiogenesis [8].How and what orchestrates those pathological modifications remains nebulous; but epigenetic regulation of gene expression has emerged as a critical determinant of embryogenesis, cardiovascular diseases, LV hypertrophy, and left heart failure [9].Epigenetics is defined as a "stable heritable phenotype resulting from changes in a chromosome without alterations in the DNA sequence".It results in the modulation of gene expression in response to the environment, such as pathologic or physiologic stressors, through: (1) DNA methylation, (2) post-translational modifications of histones, and (3) non-coding RNAs [10,11].
Although less investigated, recent studies pinpointed the potential contribution of epigenetics to the physiologic and pathologic regulation of RV function.The role of epigenetics in RV development and the progression of RV dysfunction, its capacity to predict RV dysfunction, and the potential of epigenetic-based therapy remain largely elusive.We thus performed a systematic exhaustive review of the literature and to summarize the relevant epigenetic studies focusing on the RV in humans and animal models.

Results
Our literature search retrieved 817 studies that were assessed for eligibility.We excluded 708 studies and included 109 separate publications reporting epigenetic modifications.The reasons for excluding studies appear in Figure S1.The included studies explored epigenetic modifications related to non-coding microRNAs (n = 65, Table 1), long non-coding RNA (lncRNA, n = 12, Table 2), circular RNA and small nucleolar RNA (n = 3, Table 3), histones modification (n = 15, Table 4), and DNA methylation (n = 17, Table 5).

Micro-RNA
Non-coding RNA with a length of <22 nucleotides are classified as micro-RNA (miRNA).miRNAs are, by far, the most studied epigenetic modifications in RV dysfunction (Table 1).The canonical miRNA biogenesis includes three sequential steps [12].The miR-NAs are transcribed from DNA sequences into primary miRNAs, processed into precursor miRNAs, and then into mature miRNAs.Functionally, mature miRNAs act as posttranscriptional regulators of gene expression.In most cases, miRNAs interact with target messenger RNAs (mRNAs) to induce mRNA degradation and/or translational repression.Notably, miRNAs can activate translation or regulate transcription in a non-canonical pathway [13].miRNAs can also be secreted into extracellular fluids and transported to target cells via vesicles, such as exosomes, or by binding to proteins.Interestingly, a phylogenic study conducted in healthy rats, dogs, and monkeys showed that the miRNA profile is relatively well conserved in the RV, compared to the apex, septum, LV, and papillary muscle, but is poorly conserved across the species [14].Not surprisingly, miRNAs have been repeatedly associated with diverse conditions leading to RV dysfunction.

Arrhythmogenic Cardiomyopathy
ACM is a genetic disease [79][80][81][82][83] characterized by a progressive fibro-fatty substitution of the ventricular myocardium, which is particularly pronounced in the RV, associated with life-threatening ventricular arrhythmias and, ultimately, heart failure [84,85].Traditionally, ACM was known as arrhythmogenic right ventricular cardiomyopathy (ARVC), stemming from the initial belief that the condition was confined exclusively to the RV chamber of the heart.ACM diagnosis is challenging and is often achieved after disease onset or postmortem [86].The quantification of circulating miRNA represents an attractive approach to uncover novel biomarkers.For example, two independent studies reported increased miR-185-5p levels in the blood of ACM patients compared to control patients (the sample consisted of healthy controls, patients with idiopathic ventricular tachycardia, and patients with borderline or possible ACM) [19,21].Interestingly, circulating miRNA allows for a granular classification of ACM and an increased miR-494 expression successfully discriminates ACM patients with recurrent ventricular arrhythmia post radiofrequency catheter ablation from patients without recurrent ventricular arrhythmia [21].Similarly, Sommariva and colleagues reported that decreased levels of circulating miR-320a successfully discriminate ACM from idiopathic ventricular tachycardia and healthy patients [15].Although miR-185-5p, miR-494, and miR-320a might represent a potent biomarker for ACM, whether their expression is also affected in cardiac and RV tissue remains uncertain.Comparing both cardiac tissue and blood, Bueno Marinas et al. reported that six miRNAs (miR-122-5p, miR-133a-3p, miR-133b, miR-142-3p, miR-182-5p, and miR-183-5p) are differentially expressed in the blood and RV of ACM patients compared to healthy controls [17].Interestingly, miR-122-5p (as well as miR-1-3p, miR-21-5p, miR-206, and miR-3679-5p) was also differentially expressed in pericardial fluid from ACM patients, compared to patients with post-infarct ventricular tachycardia [20].Increased miR-21-5p and decreased miR-135b expression observed in ACM in the RV correlate with myocardial fibrosis and adipose tissue deposition [17].Nonetheless, whether targeting miR-21-5p and miR-135b improve ACM and have a therapeutic effect remains unexplored.Mazurek and colleagues demonstrated that forced miR-130a expression results in ACM phenotypes (leading to RV hypertrophy, arrhythmogenic dysfunction associated with RV fibrosis, lipid accumulation, and cardiomyocyte death) in mice [16].Functionally, they demonstrated that miR-130a mediated the translational repression of Desmocollin2, an important protein for cell adhesion that is dysregulated in human ACM (OMIM 610472).Interestingly, treatments or therapeutic interventions can also affect miRNA expression profiles.Thus, compared to non-paced children, blood from chronically paced children exhibit an increased and decreased expression of 296 and 192 miRNA, respectively [22].In conclusion, the study of miRNA is mainly descriptive and restrained to the biomarkers field in ACM.

Pulmonary Hypertension
Pulmonary hypertension (PH) is hemodynamically defined by an increased mean pulmonary arterial pressure above 20 mmHg, which imposes a substantial hemodynamic stress load on the RV [87].As a consequence of this inescapable afterload rise, patients exhibit a progressive decline in cardiac function, which culminates in RV failure and premature death.The current classification system divides PH into five groups, including pulmonary arterial hypertension (PAH or group 1 PH), PH due to left heart disease (PH-LHD or group 2 PH), PH due to lung disease and/or hypoxia (or group 3 PH), chronic thromboembolic PH (CTEPH or group 4 PH), and PH with unclear and/or multifactorial mechanisms.Regardless of the group, RV function is one of the most important predictors of both morbidity and mortality in PH.In the past 5 years, tremendous efforts have been undertaken to understand the molecular mechanisms underlying RV failure in PAH.In this context, epigenetic modifications, including miRNA, have been widely studied.
Beyond the large-scale characterization of the miRNA landscape, miR-21 is the most studied miRNA in PH-RV dysfunction.In group 3PH patients, circulating miR-21 correlated with RV dysfunction (decreased tricuspid annular plane systolic excursion and tissue Doppler RV S wave) [51].Moreover, an artificial increase in miR-21 induced cardiomyocyte hypertrophy and the expression of a cardiac stress marker (namely, brain natriuretic peptide) [60].miR-21 was also increased in the RV, but not the LV, of a group 2 PH canine model [48].Interestingly, Chang and colleagues reported dynamic changes in miR-21 expression in chronic heart disease-associated PAH patients (CHD-PAH) [38].Compared to CHD-PAH without heart failure, CHD-PAH patients with heart failure had decreased circulating mir-21 levels, which are associated with increased heart failure hospitalization.In a CHD-PAH rat model, RV miR-21 expression was increased in early response to aorto-venous fistula, and then decreased.Similarly, cardiomyocyte exposed to shear stress exhibited an increased and decreased miR-21 expression at 3 h and 6 h post-treatment, respectively.Functionally, decreased miR-21 expression is associated with cardiomyocyte apoptosis, which can be reversed by forced miR-21 expression.This observation was confirmed by Liu and colleagues, who reported that decreased miR-21 expression in a decompensated RV from a preclinical model of PH was associated with increased cardiomyocyte apoptosis [39].However, as observed in human plasma, whether the miR-21 expression is increased or decreased in the RV from preclinical models of PH or pressure overload remains conflictual.For example, the miR-21 expression was reported to be increased in RVs from female monocrotaline (MCT) rats [40] and male Sugen + hypoxia rats [31], but decreased in PAB female rats [32] and in decompensated (male Sugen + hypoxia rats and PAB + low copper diet) vs. compensated RVs (hypoxia and PAB) [57].Moreover, miR-21 expression was decreased in the basal part of the RV and unchanged in the apex of PAB animals, suggesting a topologic regulation of miR-21 in this model [60].Combined with the dynamic regulation of miR-21 [38], this topologic regulation of the miRNA might explain some discrepancies between the studies.miR-223 is another example of the complex regulation of miRNA in PAH-associated RV dysfunction.miR-223 was increased in MCT rat RVs [40] and decreased in hypoxia-PH and PAB mice [26,58].Although an antagomir-223 treatment failed to decrease miR-223 expression and improve RV function in MCT rats [40], the genetic depletion of miR-223 decreased the RV function [26].Moreover, artificial miR-223 over-expression improved RV function and decreased RV fibrosis, which was likely mediated by the lower expression of the insulin-like growth factor-I receptor (IGF-IR, a miR-223 target).Similarly, Levchenko and colleagues reported increased miR-197 and miR-146b levels in RVs from human PAH patients and Sugen + hypoxia rats, when compared to healthy donors and control animals [27].In vitro, the authors demonstrated that miR-197 and miR-146b regulated metabolism, and that forced miR-197 and miR-146b expressions decreased the expression of factors that drive fatty acid oxidation (e.g., CPT1B, and FABP4).Interestingly, Chouvarine and colleagues observed increased mir-197 and decreased miR-146b expression in the RVs of mice exposed to hypoxia [58].In vitro, they reported that hypoxia decreased miR-146b and induced TRAF6, IL-6, and CCL2 (MCP-1) pro-inflammatory signalling.This controversial observation, finding a putative beneficial and deleterious role of miR-146b on RV function, highlights the complexity of miRNA regulation and contribution upon various stimuli.
Functionally, miRNA provided insight into the mechanisms regulating RV hypertrophy in PH.Connolly and colleagues reported that miR-424(322) expression was decreased in MCT rat RVs and likely contributed to the increased expression of the pro-hypertrophic growth factor insulin-like growth factor-1 [28].However, Baptista and colleagues showed that increased levels of circulating miR-424(322), observed in PH patients, correlated with higher symptom severity and hemodynamics (e.g., miR-424(322) inversely correlated with cardiac output) [29].In a subgroup of CHD-PAH, miR-424(322) was an independent marker of survival.Using a translational approach, the authors demonstrated that increased levels of circulating miR-424(322) were associated with RV hypertrophy in MCT rats.miR-214 expression was increased in RVs harvested from MCT rats, Sugen + hypoxia rats, and mice models of PH [30,33].Moreover, in Sugen + hypoxia mice, the genetic depletion of miR-214 led to increased RV hypertrophy, likely mediated by PTEN overexpression [30].PTEN is an important regulator of cardiac hypertrophy, which is also regulated by miR-495 in failing MCT rats' RVs [34].Decreased miR-495 expression was associated with increased PTEN and decreased cardiac stress markers (e.g., NPPA) in MCT RVs and cultured cardiomyocytes.Increased miR-335-5p was associated with a decreased expression of calumenin (miR-335-5p target) and pathologic RV hypertrophy in preclinical models of PH [35].The inhibition of miR-335-5p expression decreases RV fibrosis, cardiomyocyte hypertrophy, and apoptosis, and improves RV function.
miRNAs are differentially expressed across cell types and tissues.For example, my-omiRs are a class of miRNAs that are mainly expressed in muscles.Kmecova and colleagues studied the expression of four myomiRs (miR-1, miR-133a, miR-208, and miR-499) which are known to be enriched in cardiac tissue [33].They reported a global decreased expression of those four cardio-myomiRs in RVs from MCT rats, compared to control animals [33].Decreased miR-1 expression in MCT RVs was associated with increased TGF-β signalling, which might contribute to cardiac hypertrophy [36].Intriguingly, miR-1 expression was unchanged in RVs from female PAB rats [32] but increased in hypertrophic RVs isolated from rats exposed to hypoxia [37].Moreover, under hypoxic conditions, miR-1 silencing decreased RV hypertrophy, RV fibrosis, and fibroblast collagen production [37].Mir-133a expression was also decreased in decompensated RVs (Sugen + hypoxia or PAB + low copper diet), when compared to animals with compensated RVs (hypoxia or PAB surgery) [57].The expression of miR-208, and specifically miR-208a, was also decreased in RVs from Sugen + hypoxia PH rats [31].Paulin and colleagues demonstrated that miR-208a expression in the RV negatively correlated with RV hypertrophy in MCT rats [53].Moreover, they reported that decreased miR-208a activated the complex mediator of the transcription 13/nuclear receptor corepressor 1 axis, which in turn, promoted Mef2 inhibition, likely contributing to cardiomyocyte hypertrophy and decreased angiogenesis.
AngiomiRs are a category of miRNA involved in the regulation of angiogenesis [88].Among them, miR-126 is one of the most studied and important angiomiRs.Interestingly, a decreased mir-126 expression in PAH decompensated RVs correlates with capillary rarefaction and decreased RV function in both human patients and MCT rat models of PH [56].This effect was likely mediated by the increased expression of SPRED-1, a downstream inhibitor of the VEGF/VEGFR2 angiogenic pathway.Conversely, artificial miR-126 overexpression decreased SPRED-1 expression, improved RV endothelial cells' angiogenic capacity, and increased RV capillary density and function in MCT rats.Interestingly, miR-126 was also decreased in RVs from Sugen + hypoxia rats' models of PH [57] and in failing RVs from a dog model of group 2PH [48].Beyond PAH, decreased miR-126 RV levels were associated with post-transplantation complications, such as cardiac allograft vasculopathy [23].MiR-17 is another angiomiR that was decreased in the blood of PH patients (compared to healthy controls) [55], which correlated with RV function [59].However, miR-17 was increased in the RVs of female rats exposed to MCT [40].
FibromiR refers to a miRNA involved in fibrotic processes.Yi Tang and colleagues demonstrated that decreased miR-325 expression in MCT rats' RVs negatively correlated with RV fibrosis [52].Moreover, forced miR-325 expression in rat fibroblasts and MCT animals decreased collagen deposition and relieved RV fibrosis.Decreased miR-200b expression observed in MCT rats' RVs was associated with the increased expression of the pro-fibrotic protein kinase c alpha [54].Functionally, miR-200b overexpression decreased protein kinase c alpha expression in both the RVs and the lungs of MCT rats.Powers and colleagues reported the impaired expression of 26 miRNAs, including increased miR-21 and miR-221, in RV samples harvested from a dog model of group 2PH [48].Hormonal (aldosterone) and mechanical (stretch) stress increased miR-21 and miR-221 expression in canine RV fibroblasts (but not in LV fibroblasts).Similarly, the inhibition of miR-21 and miR-221 decreased RV (but not LV) fibroblast proliferation.

Congenital Heart Diseases
Congenital heart diseases (CHDs) are the most common type of birth defect and the leading cause of inherited perinatal and infant mortality.CHD refers to structural anomalies of the heart and blood vessels that arise during cardiac development and represents a broad spectrum of malformations, including septal and valve defects, and lesions affecting the outflow tract and ventricle.CHDs include tetralogy of Fallot (TOF), pulmonary atresia, and pulmonary valve stenosis, among other diseases.Qian Yang and colleagues reported 20 differentially regulated miRNAs in RV outflow tract (RVOT) samples isolated from CHD patients, compared to controls [24].Among them, increased miR-29b-3p was associated with reduced cardiomyocyte proliferation, cardiac malformation, and lethality in zebrafish.
-Tetralogy of Fallot TOF is a complex congenital heart defect consisting of pulmonary stenosis, ventricular septal defect, overriding aorta, and RV hypertrophy.TOF is the most common cyanotic congenital heart defect and accounts for about 10% of all congenital heart malformations.Although the contribution of miRNA in TOF etiology remains poorly understood, several studies characterized the miRNA landscape associated with TOF development in RV myocardium, RV outflow tract RVOT, or plasma samples from human patients.
Grunert and colleagues reported that 172 miRNA (111 upregulated and 61 downregulated) were differentially regulated in TOF RVs [68].They identified miR-1 and miR-133 as potential key miRNAs in disease etiology, notably by regulating KCNJ2, FBN2, SLC38A3, and TNNI1 expression.Interestingly, miR-1/miR133 clusters might contribute to the regulation of the 41 differentially expressed miRNA between male and female TOF patients, and thus contribute to a potential miRNA sexual dimorphism in TOF [78].Comparing TOF to control RVs, Zhang et al. observed 47 differentially expressed miRNA, including upregulation of three members of the miR-29 family (known to be involved in LV failure) [76,89].Bittel and colleagues identified 61 differentially regulated miRNA (32 upregulated and 29 downregulated) in myocardial RV samples harvested from TOF and control patients [75].They showed that increased miR-421 expression decreases SOX4 expression, a key regulator of the Notch pathway involved in the cardiac outflow tract and TOF development [70].A study, investigating the expression of miRNA in RV samples from patients with TOF, RVOT obstruction, combined pulmonary insufficiency (PI), and stenosis (PS), with and without end-stage RV failure, reported increased miR-21 expression in tissue samples from PI/PS patients, compared to TOF and RVOTO patients.Circulating miR-21 expression correlated with RV function and exhibited a dynamic regulation, as it increased in the plasma of PI/PS patients without RV failure and decreased in patients with RV failure.However, Dilorenzo and colleagues did not report any association between circulating miR-21 and RV function assessed by MRI [77].Thus, the potential biomarker value of circulating miR-21 remains to be further investigated.
The quantification of circulating miRNA is an attractive approach to identifying novel and potent biomarkers.Lai and colleagues reported decreased expressions of miR-766 and miR-99b in the plasma of TOF patients [74].Abu-Halima and colleagues identified 49, 58, and 77 circulating miRNAs that were differentially expressed in, firstly, TOF patients vs. controls, secondly, TOF patients with right heart failure vs. controls, and, thirdly, TOF patients without symptomatic right heart failure vs. controls [73].Similarly, Weldy and colleagues assessed the expression of circulating RNA in TOF patients with normal RV, mild to moderate RV dilatation, and severe RV dilatation [72].The authors identified (and validated) increased miRNA 28-3p, 433-3p, and 371b-3p to be associated with increased RV size and decreased RV systolic function.

The Systemic Right Ventricle
The RV, which normally supports low-pressure pulmonary circulation, has to go through various adaptive mechanisms to sustain the systemic load when placed in the systemic position.Compared to normal condition, the systemic RV (SRV) represents a distinctly different model in terms of its anatomic spectrum, short-and long-term adaptation, and clinical phenotype.The common clinical scenarios where an SRV is encountered are as follows: the complete transposition of the great arteries (TGA) with previous atrial switch repair; double discordance or congenitally corrected TGA, operated or native; double-inlet RV; and hypoplastic left heart syndrome, which occurs in patients with aortic atresia or hypoplasia who require reconstruction of the aorta from the pulmonary artery (Norwood protocol).
Decreased miR-486 expression was also observed in RV samples from patients with hypoplastic left heart syndrome and in a surgical model of the disease induced by aortopulmonary vascular grafts [25].In vitro, forced miR-486 expression decreased FOXO1 and SMAD signalling, and increased STAT1 expression and cardiomyocyte contractility.Interestingly, Lai and colleagues reported/validated the increased expressions of 11 miRNA, including miR-486, in blood from patients with TGA + SRV.Moreover, they observed that miR-486 and miR-18 negatively correlated with systemic ventricular contractility.An independent study identified 106 differentially expressed miRNA in plasma samples from TGA + SRV [67].Four miRNAs, namely, miR-150-5p, miR-1255b-5p, miR-423-3p, and miR-183-3p, were associated with the clinical worsening of heart failure, with miR-183-3p being the only independent predictor of worsening heart failure in a multivariate model.Nonetheless, Tutarel and colleagues reported that miR-423-5p levels were unchanged in TGA + SRV patients, compared to healthy controls, and failed as a biomarker of heart failure [65].

Dilated Cardiomyopathy
Dilated cardiomyopathy (DCM) is characterized by the enlargement and dilation of one or both of the ventricles, along with impaired contractility, defined as an LV ejection fraction (LVEF) of less than 50% [90].While often restrained to the LV, patients with DCM may also develop diastolic dysfunction and impaired RV function.In DCM patients, impaired RV function is associated with increased morbidity and lower survival [91].
Limited studies focused on the RV in DCM.Nonetheless, the impaired expression of miR-21 and miR-133 was described in most of them.Although Parikh et al. reported decreased miR-21 expression in DCM RVs (compared to controls) [41], Wang and colleagues observed an increase in miR-21 [43], and Rubis showed no differences in miR-21 levels [44].Similarly, depending on the study, miR-133a expression was reported to be decreased [43,44] or unaffected [41] in RVs from DCM patients, compared to controls.Moreover, miR-133a expression was increased in DCM patients with RV dysfunction, compared to patients without [45].The observed discrepancies in miR expression levels could potentially be attributed to variations in the selection and characterization of control groups, for which there is often limited information reported (e.g., healthy controls vs. non-DCM patients).In plasma, miR-21 levels correlated with RV morphology.Circulating Mir-133a was associated with RV morphology [42] and RV dysfunction [45], and was an independent predictor of mortality in DCM [44].

Right Ventricle Dysfunction Associated with Neonatal Hypoxia and Irradiation
Rats exposed to early hypoxia (2-20 days post-natal) exhibited increased inflammation and RV hypertrophy, which was associated with decreased miR-206 expression [47].Interestingly, decreased miR-206 expression was also observed in the blood of children suffering from TOF [73] and patients with ARVC [20], suggesting that, beyond contributing to RV hypertrophy, miR-206 might be involved in several RV-related pathological processes.Radiation of the chest during cancer therapy is deleterious to the heart, mostly due to oxidative stress and inflammation-related injury.Indeed, RV and LV dysfunction and remodeling are common complications observed in human patients exposed to radiotherapy [92].Similar to other types of RV dysfunction, RV from rats exposed to a single sub-lethal dose of irradiation exhibited an increase in miR-21 expression which was reversed by aspirin, atorvastatin, and sildenafil treatment [46], suggesting that this miRNA might be a critical determinant in RV function and homeostasis.

Long Non-Coding RNA
Non-coding RNA with a length of >200 nucleotides are classified as lncRNA.During the past decade, numerous studies have highlighted the importance of lncRNA in orchestrating heart development, cardiovascular cell signalling, and cardiac function [93].Functionally, lncRNAs exert a plethora of functions, which make them key regulators of biological functions, but complicate their study.Indeed, depending on their localization and their specific interactions with DNA, RNA, and proteins, lncRNAs can modulate chromatin function, regulate the assembly and function of nuclear bodies, alter the stability and translation of cytoplasmic mRNAs, and interfere with signalling pathways.Tissue-specific and condition-specific expression patterns suggest that lncRNAs are potential biomarkers and provide a rationale to target them clinically.Studies investigating the role of lncRNAs in RV physiologic and pathologic development are summarized in Table 2.
lncRNAs are critical players in RV development.Touma and colleagues investigated the neonatal LV and RV and identified 21,916 known and 2033 novel lncRNAs.One hundred and ninety-six exhibited significant dynamic regulation along the maturation process, and eleven were specifically upregulated in the RV [94].Keeping in mind that lncRNAs are poorly conserved across the species, the authors identified five lncRNAs with conserved orthologs in humans, including UCP2-lncRNA (NONMMUT062940), n420212 (NONM-MUT041263), Fus-lncRNA (NONMMUT063779), Ppp1r1b-lncRNA (NONMMUT011874), and H19 (NONMMUT064276).Heart And Neural Crest Derivatives Expressed 2 (Hand2) is a critical gene for the embryonic development of the heart and the differentiation of the outflow tract myocardium, RV, and right atria [95].Hand2 is expressed throughout the primary heart tube during early embryonic stages, with dominant expression in the RV and outflow tract as the heart tube loops.Consequently, Hand2-null (Hand2 −/− ) mice showed severe RV hypoplasia and growth delay from E9.5, with death by E10.5.Ritter and colleagues demonstrated that the lncRNA locus Handsdown (Hdn), active in the early heart cells, regulated Hand2 expression and was essential for early mouse development [96].Indeed, the transcriptional activity of the Hdn locus, independent of its RNA, suppressed its neighboring gene Hand2.Consequently, Hdn-heterozygous adult mice exhibited increased Hand2 expression and RV wall hyperplasia.Interestingly, Andersson and colleagues demonstrated that another lncRNA, namely, upperhand (Uph), is required to maintain Hand2 transcription [97].They reported that Uph-heterozygous adult mice exhibited decreased Hand2 expression and RV wall hypoplasia.Those data suggest a tight epigenetic regulation of Hand2 in the embryogenic development of the heart and RV.
The RV myocardium of an acute rat model of PH right heart failure exhibited an impaired expression of 169 lncRNA [98].Consistent with the potential role of lncRNA in PAH RV dysfunction, Omura and colleagues reported that increased expression of the lncRNA H19 correlated with RV fibrosis and cardiomyocyte hypertrophy in PAH patients and two preclinical models of RV failure [99].Notably, H19 is known to serve as the primary precursor of miR-675, which is also increased in decompensated RV.Functionally, they reported that targeting H19 (siRNA) improved RV function and decreased RV fibrosis and hypertrophy in two preclinical models of RV failure.In vitro, silencing H19 decreased cardiomyocyte hypertrophy and the expression of cardiac stress genes, while H19 overexpression had the opposite effect.Moreover, the authors reported that increased circulating H19 in the blood of PAH patients with decompensated RVs was an independent predictor of survival in PAH.This study demonstrated the therapeutic potential and the biomarker value of lncRNA in PAH-RV failure.
Focusing on lncRNA, several studies dissected RNA-sequencing data to describe the changes in the non-coding RNA landscape associated with RV dysfunction across various pathological conditions.RV samples from ischemic and non-ischemic RV failure expressed over 10,831 lncRNA [100].Di-Salvo and colleagues demonstrated that RV myocardium harvested from human patients with ischemic and non-ischemic RV failure exhibited impaired expressions of 105 lncRNAs [101].Similarly, RV myocardium from patients with cardiomyopathy and tricuspid regurgitation exhibited impaired expressions of 163 lncR-NAs and 201 miRs [102].Another descriptive study demonstrated that hypertrophic RV (from TOF/pulmonary stenosis patients) showed impaired expressions of twenty-five lncRNA and eight miRs [103].Wang and colleagues observed that increased expression of the lncRNA HA117 in the RVOT of TOF patients was associated with the severity of the disease and might lead to adverse short-term outcomes in TOF patients [104].Beyond disease development and progression, Yaping Shan and colleagues reported that the lncRNA profile was affected by surgical procedures [105].They observed that the implantation of expanded polytetrafluoroethylene's (a biomaterial commonly used in cardiovascular surgery) into RVOT was associated with the impaired expression of 246 lncRNAs, likely contributing to RV inflammation and fibrosis in animals.Although descriptive, those studies suggest a potential role for lncRNA in RV dysfunction.

Other Non-Coding RNA
Beyond miRNA or lncRNA, the impaired expressions of other classes of non-coding RNA, namely circular RNAs (circRNAs) and small nucleolar RNAs (SnoRNAs), have been associated with RV dysfunction (Table 3).CircRNAs are single-stranded, covalently closed RNA molecules that are ubiquitous across species.CircRNAs exert biological functions by acting as transcriptional regulators, miRNA sponges, and protein templates.Circulating levels of CircRNAs have been proposed as potential biomarkers of RV function.For example, increased circRNA_0068481 and decreased circulating expression of its "sponged miRNA" (miR-646 and miR-570) predicted RV hypertrophy in PH patients [35].SnoRNAs are non-coding RNAs, found in the nucleolus, that primarily guide nucleotide modifications of rRNA (ribosomal RNA).Hypertrophic RV myocardium from TOF patients exhibited impaired expressions of 135 snoRNA [75] and 3 circRNA, namely, ENSG00000171517:LPAR3, ENSG00000155657:TTN, and ENSG00000127914:AKAP9:chr7 [103].

Histone Modifications
Histones directly interact with DNA and are the principal component of chromatin.Post-transcriptional modification of histones (including histone acetylation, methylation, phosphorylation, ubiquitination, etc.) is one of the main epigenetic mechanisms regulating chromatin compaction and subsequent gene expression.Histone modifications are tightly regulated processes.For example, histone acetylation status is regulated by two groups of enzymes exerting opposite effects; the histone acetyltransferases (HATs) and the histone deacetylases (HDACs).Similarly, histone methylation is regulated by enzymes which add (histone methyl transferase, HMT) or remove (histone demethylase) methyl groups from the histone.Impaired regulation of histone modifications results in aberrant gene expression and pathologic conditions, including LV failure [106].Therefore, the development of pharmacological inhibitors of enzyme-regulating histone modifications (e.g., HDAC, HAT, and HMT) opens the doors to novel epigenetic-based therapeutic approaches.Studies focusing on the role of histone modification in RV physiologic and pathologic development are summarized in Table 4.
Histone modifications are also critical for regulating embryonic gene expression and developing the fetal heart, including the formation of the RV.Indeed, homozygous depletion of the HMT SET-MYND-domain 1 (Smyd1) in cardiomyocytes and the outflow tract resulted in a truncated RV and outflow tract and embryonic lethality at E9.5 in mice [107].Functionally, Smyd1 depletion impaired the expansion and proliferation of the second heart field, which is involved in the differentiation of the RV and outflow tract.Similarly, homozygous depletion of the Hdac m-BOP gene resulted in decreased hand2 expression, resulting in RV hypoplasia and ventricular maturation defects [108].Genetically engineered mice lacking Hdac2 exhibited abnormalities of the right chamber, with an almost complete loss of the RV lumen [109].Beyond the expression, the cellular localization of HDAC is critical for RV development.For example, mice lacking apelin receptor APJ had greater endocardial Hdac4 and Hdac5 nuclear localization and reduced expression of MEF2 and its transcriptional target Krüppel-like factor 2, resulting in poorly looped hearts and an aberrant RV [110].In adult mice, enhanced recruitment of the histone acetyltransferase p300 led to increased acetylation of H3K9/14 and H4, as well as demethylation of H3K4.This epigenetic modification was associated with heightened expression of cardiac stress genes, including atrial and brain natriuretic peptides, predominantly in the left ventricle (LV) as compared to the right ventricle (RV).Interestingly, Notch depletion in mouse cardiomyocytes isolated from the LV resulted in increased H3kme3 marks on the Hrt2 promoter, while Notch depletion had no effect on mouse cardiomyocytes isolated from the RV [111].Taken together, those observations suggest that post-natal histone modification contributed to distinct gene expression patterns in the healthy LV and RV (likely reflecting the distinct afterload and wall stress applied to both cavities) [112].
HDACs are clustered into five classes, namely, HDAC class I (includes HDAC 1, 2, 3, and 8), HDAC class IIA (HDACs 4, 5, 7, and 9), HDAC class IIB (HDACs 6 and 10), HDAC class III (HDACs called sirtuins), and HDAC class IV (HDAC 11).RVs isolated from the PH rat models (SUHX, and MCT) and pressure overload (PAB) model exhibited increased activity of HDAC classes I, IIa, and IIb, while the RVs from animals exposed to hypoxia alone showed only increased activity of HDAC class IIb [113,114].However, the expression of Hdac 1, 2, and 3 (Class-I HDAC) was increased in RVs from rats exposed to hypoxia [115], and Hdac 1 was increased in hypoxic calves' RV tissues [116].Consequently, the HDAC class I, II, and IV inhibitor, vorinostat, improved RV function in hypoxic calves [116].The HDAC class I inhibitor sodium valproate decreased RV hypertrophy and fibrosis in the MCT and PAB rat models [117], and GCD0103, which selectively inhibited Hdac 1, 2, and 3 (HDAC class I) decreased the expression of cardiac stress genes, inhibited the proapoptotic caspase activity, and reduced the proinflammatory protein expression in the RV of hypoxic rats [115].Conversely, in PAB rats, the HDAC class I and II inhibitor trichostatin-A failed to prevent RV hypertrophy, increased RV fibrosis, decreased RV capillary density, and increased cardiomyocyte apoptosis [118].
Increased expression of the Hdac GCN5 was observed in RV samples from ACM patients and associated with lipid accumulation and reactive oxygen species (ROS) production [119].Impaired post-translational histone modification mechanisms were associated with congenital heart diseases affecting the RV.For example, RV samples from children with single ventricle heart disease exhibited increased HDAC Class I, IIa, and IIb catalytic activity and protein expression [120].RVOT from TOF children showed decreased mRNA and protein expression of the histone chaperone HIRA (which facilitates the activity of HDACs) [121].

DNA Methylation
DNA methylation is a biological process by which methyl groups are added to cytosine or adenine nucleotides.It is a dynamic process tightly regulated by the expression/activity of DNA methylation "writers", which add methyl groups to the DNA (e.g., DNA methyl transferases 1, (DNMT1), DNMT3a, and DNMT3b), and the DNA methylation erasers, which remove methyl groups (e.g., the ten-eleven translocation (TET1, TET2, TET3)).Functionally, DNA methylation regulates gene expression by recruiting proteins involved in gene repression or by inhibiting the binding of transcription factor(s) to DNA.Although an impaired DNA methylation landscape has been associated with (left) heart failure, the contribution of DNA methylation to RV dysfunction remains poorly investigated [122] (Table 5).
Interestingly, the LV and the RV exhibit distinct DNA methylation profiles.RVs from human patients with DCM exhibited 1828 differentially methylated loci (compared to the LV), likely contributing to the regulation of cardiac development genes [123].This observation suggests a tissue specificity in DNA methylation patterns.Consistently, blood leucocytes and right atrial tissues displayed distinct DNA methylation profiles in patients that underwent coronary artery bypass surgery [124].In TOF RVOT or RV myocardium, an impaired DNA methylation landscape affected ETS1, RXR, and SP1 binding sites and contributed to decreased DLK1, NOTCH4, and TBX20 gene expression [125][126][127][128][129][130].Notably, hypomethylating drugs influenced the interaction between RXR, ETS1, and their respective binding sites in vitro [126,130].Additionally, the impaired DNA methylation of NKX2-5, EGFR, EVC2, TBX5, CFC1B, RXRa, VANGL2, TBX20, NR2F2, NOTCH4, and DLK1 correlated or was associated with the impaired expression of their respective transcripts [126][127][128][129][130][131][132][133][134].Interestingly, the DNA methylation of TBX20 (affected in TOF patients) was unaffected in the RVs and blood of DCM patients, suggesting a potential specificity of DNA methylation changes for the pathologic condition affecting the RV [135,136].The DNA methylation of LINE-1 was decreased in RV samples from male TOF patients (compared to male controls) but not in females [125].This observation supports the potential effect of sex on the DNA methylation landscape in TOF.
In PAH, DNA methylation contributes to pathological processes involved in RV dysfunction.Indeed, decompensated RV from PAH patients exhibited an increased expression of DNMT3a and DNMT3b, associated with increased DNA methylation of the angiomiR-126 gene, likely contributing to its downregulation and the angiogenic defect observed in patients' RVs [56].In vitro, the hypomethylating drug named hydralazine increased miR-126 and the angiogenic potential of endothelial cells isolated from PAH decompensated RVs.Moreover, RV fibroblasts harvested from a preclinical model of PH with RV failure exhibited an increased expression of DNMT1, compared to fibroblasts isolated from control (non-PH) animals [137].In vitro, the hypomethylating drug decitabine decreased DNMT1 expression and global DNA methylation, resulting in an improved fibroblast mitochondrial metabolism and a potentially decreased PH RV fibroblast pro-proliferative and pro-fibrotic phenotype.This observation echoes another study, which demonstrated that increased RV fibrosis was associated with DNMT1 and DNMT3a overexpression in hypoxic mice [138].In vitro, hypoxia increased DNA methylation, fibroblast proliferation, and fibrosis.

Conclusions
In comparison to other cardiovascular conditions, such as left heart failure and atherosclerosis, research on the role of epigenetic modifications in RV development, dysfunction, and failure is relatively sparse.To address this gap, we performed a thorough systematic review of the literature, focusing on studies that investigated the relationship between various epigenetic mechanisms (including DNA methylation, histone modification, and non-coding RNA) and RV pathophysiology.Our search on PubMed, up to 1 January 2023, yielded 109 original peer-reviewed articles.
Our review revealed that the bulk of the literature is descriptive in nature.A scant number of studies ventured into proposing epigenetic-based therapeutic strategies or biomarker development.This scarcity of interventional research could reflect the current limitations in pharmaceutical interventions that target epigenetic modifications, particularly those affecting DNA methylation and non-coding RNAs.The primary agents available, pan DNMT or HDAC inhibitors, induce widespread and non-selective effects, which are less than ideal.
However, the advent of sophisticated biomolecular techniques, such as the second generation of CRISPR/Cas9 systems, now permits precise DNA methylation editing at the single-nucleotide resolution [139] and targeted histone acetylation modifications at the single-gene level [140,141].Despite their potential, these technologies are predominantly confined to preclinical studies.Additionally, devising RV-specific therapies that do not impact other organs (e.g., LV, lungs, and kidneys) poses a significant technical hurdle, thus impeding the progression to clinical trials for RV dysfunctions.
A further complication in the field is the inconsistency of study results, particularly concerning miRNA expression patterns in RV dysfunction.Conflicting data on whether RV dysfunction correlates with an upregulation or downregulation of specific miRNAs, such as miR-21, miR-223, or miR-146b, underscores the dynamic and intricate nature of epigenetic regulation throughout disease progression.It also highlights the necessity for standardized protocols in terms of control groups, epigenetic assessment technologies, and sample processing.
Moreover, the epigenetic underpinnings of various physiological processes affecting the RV, notably during the aging process, are poorly understood, and, consequently, the literature is very limited.For instance, similarly to the LV, aging markedly impairs RV function-this is manifested through reduced systolic and diastolic functions, increased fibrosis, cardiomyocyte death, and stiffness [142][143][144]-yet the molecular drivers of these age-related changes in the RV remain unclear, with no dedicated studies investigating their epigenetic foundations.
Interestingly, the RV exhibits a greater susceptibility to damage from intense physical exercise than the LV.Highly trained athletes, for example, display well-documented morphological and functional changes in the RV, such as decreased ejection fraction and increased wall stress, particularly after prolonged endurance activities [145][146][147][148].These changes have been correlated with an elevated risk of atrial fibrillation.The causal factors behind RV dysfunction post-exercise are thought to be related to the ventricle's unique physiological characteristics.Nonetheless, the precise molecular and potential epigenetic contributions to these changes are yet to be fully delineated.
In summary, studies have only just scratched the surface of the role of epigenetics in RV development and dysfunction, and despite promising results, additional investigations are needed to pinpoint the contribution of epigenetics in physiological and pathological RV development and the development of epigenetic-based therapy.

Table 2 .
Long non coding RNA and right ventricle.

Table 3 .
Circular RNA, Small nucleolar RNA and right ventricle.

Table 4 .
Histone modification and right ventricle.

Table 5 .
DNA methylation and right ventricle.