The Impact of Sex Chromosomes in the Sexual Dimorphism of Pulmonary Arterial Hypertension

Abstract

Pulmonary arterial hypertension (PAH) is a sex-biased disease with a poorly understood female prevalence. Emerging research suggests that nonhormonal factors, such as the XX or XY sex chromosome complement and sex bias in gene expression, may also lead to sex-based differences in PAH incidence, penetrance, and progression. Typically, one of females' two X chromosomes is epigenetically silenced to offer a gender-balanced gene expression. Recent data demonstrate that the long noncoding RNA X-inactive specific transcript, essential for X chromosome inactivation and dosage compensation of X-linked gene expression, shows elevated levels in female PAH lung specimens compared with controls. This molecular event leads to incomplete inactivation of the females' second X chromosome, abnormal expression of X-linked gene(s) involved in PAH pathophysiology, and a pulmonary artery endothelial cell (PAEC) proliferative phenotype. Moreover, the pathogenic proliferative p38 mitogen-activated protein kinase/ETS transcription factor ELK1 (Elk1)/cFos signaling is mechanistically linked to the sexually dimorphic proliferative response of PAECs in PAH. Apprehending the complicated relationship between long noncoding RNA X-inactive specific transcript and X-linked genes and how this relationship integrates into a sexually dimorphic proliferation of PAECs and PAH sex paradox remain challenging. We highlight herein new findings related to how the sex chromosome complement and sex-differentiated epigenetic mechanisms to control gene expression are decisive players in the sexual dimorphism of PAH. Pharmacologic interventions in the light of the newly elucidated mechanisms are discussed.

Pulmonary arterial hypertension (PAH) is a sex-biased disease characterized by pulmonary artery remodeling and hallmark plexiform lesions, leading to right heart failure and untimely death.1, 2, 3 Histopathologic findings at the level of small pulmonary arteries, the vascular bed predominantly affected by the disease, include medial hypertrophy, intimal hyperplasia, fibrosis, and development of hallmark plexiform lesions.^4,5 The interplay between the hyperproliferative endothelial and pulmonary artery smooth muscle cells, the accumulation in the perivascular spaces of fibroblasts, myofibroblasts, and pericytes, and increased infiltration of inflammatory cells (B and T lymphocytes, mast cells, dendritic cells, and macrophages) contributes to the pathophysiological features of PAH.^6,7 In human and experimental PAH, progressive obliteration of precapillary arteries, followed by their loss, leads to pulmonary vascular rarefaction (dead-tree picture), a transformation happening on a background enriched in pro-angiogenic factors, on the endothelial layer, and overexpressed Notch3 receptor on the smooth muscle cell layer.^8,9 Functional metabolic changes, such as the glycolytic shift, altered lipid metabolism, and insulin resistance, favor endothelial cells' propensity toward increased proliferative capacity, mesenchymal transformation, and decreased sensitivity to apoptosis.10, 11, 12, 13, 14 These dysfunctional pulmonary endothelial cells also exhibit a proinflammatory phenotype characterized by high levels of the integrin ligands intracellular adhesion molecule-1, vascular cell adhesion molecule-1, E-selectin, and P-selectin, as well as high levels of expression and secretion of different chemokines, cytokines, and growth factors.^15,16 These endothelial cells are responsible for the impairment of endothelial-dependent vasodilatation in favor of vasoconstriction and reduced anticoagulant properties of the luminal surface of the endothelium.¹⁷ Up-regulation of hypoxia-inducible factor in vascular cells leads to the production of bone marrow–mobilizing factors that recruit pro-angiogenic progenitor cells to the pulmonary circulation, where they contribute to angiogenic remodeling of the vessel wall.¹⁸

Notch3, a member of the Notch receptor family, is expressed only in arterial smooth muscle cells of the human vasculature; and its signaling controls their proliferation.¹⁹ Recent studies of lung tissue of PAH patients and rodent models of PAH demonstrated elevated expression of Notch3 that correlate with the severity of the disease.⁸ Moreover, mice deficient in the C-C chemokine ligand type 2 exposed to chronic hypobaric hypoxia develop spontaneous PAH, via increased Notch3 signaling.²⁰ Furthermore, Notch3 knockout mice are resistant to hypoxia-induced PAH, whereas blocking Notch signaling prevents pulmonary hypertensive vascular pathology in vivo. More sex-specific analyses are needed to provide additional information on Notch3 possible involvement in PAH sex paradox.

PAH, a cause of pulmonary hypertension (PH), is classified as group I PH by the World Health Organization.²¹ On the basis of the disease's underlying cause, the complex, multifactorial pathophysiology, and therapeutic options, PH was categorized into five different diagnostic groups, each one with multiple subgroups, termed the World Health Organization groups. They include the following: Group 1: PAH, defined as PH with specific hemodynamic criteria (mean pulmonary artery pressure ≥25 mmHg, pulmonary artery occlusion pressure ≤15 mm Hg, and pulmonary vascular resistance ≥3 Wood units); it can be idiopathic or heritable and associated with connective tissue disease, congenital left-to-right shunt, hemoglobinopathies, HIV disease, schistosomiasis, and liver disease.²² Group 2: PH due to left heart disease. Group 3: PH due to lung diseases and/or hypoxia. Group 4: PH due to pulmonary obstructions. Group 5: PH with unclear or multifactorial etiologies.²³ In 2018, the 6^th World Symposium on PAH, the European Society of Cardiology, and European Respiratory Society revisited and updated the hemodynamic definition and classification of PH.²³ Briefly, the revised hemodynamic assessment suggests a new pressure level to define an abnormal elevation in the mean pulmonary artery pressure (>20 mmHg) and the need for pulmonary vascular resistance ≥3 Wood units to define the presence of precapillary PH. The updated classification maintain the five PH diagnostic groups; however, several separate entities are added in group I (PAH) that includes three new subgroups: drug and toxin-induced PAH, PAH patients who are long-term responders to calcium channel blockers and PAH with overt features of venous/capillary involvement, such as heritable pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis, resulting from eukaryotic translation initiation factor 2 α kinase 4 or other gene mutations, and PAH associated with occupational exposure, in particular organic solvents. Moreover, group 5 is simplified by removing splenectomy and thyroid disorder and classification of several conditions (lymphangioleiomyomatosis-associated PH, with other parenchymal lung diseases). Despite revisions, it became clear that more studies are needed to better understand various conditions associated with PH and disease development.

Population-based studies show a female prevalence in PAH of around 2 to 4 over men for all races and ethnicities and across all ages that have been studied to date.²⁴ Although sex is the leading risk factor for the disease, other factors, such as infections, autoimmune diseases, inflammation, obesity, sleep apnea, and genetic predisposition, may be associated with PAH.²⁴ Emerging research strongly suggests that nonhormonal factors such as sex chromosomes and sex bias in gene expression may also lead to sex-based differences in PAH penetrance and progression. Mammals have two distinct heteromorphic chromosomes that govern sex determination: the Y chromosome that contains few genes, approximately 70, present only in males; and the X chromosome containing between 900 and 1500 genes, present in two copies in female and one copy in male.^25,26 This distribution generates a gene dosage imbalance between X chromosome–linked and autosomal genes and between the sexes.²⁷ For the maintenance of the two sexes, this imbalance is relieved via two dosage compensation mechanisms: by increasing expression levels of dosage-sensitive X-linked genes (X up-regulation), in both sexes, and by random silencing of one X chromosome (X chromosome inactivation) in females.²⁸ X up-regulation is achieved by increased expression levels of genes on the single active X chromosome to balance expression with the autosomes, which are present in two copies. In addition, X chromosome inactivation is achieved by the long noncoding RNA (lncRNA) X-inactive specific transcript (Xist).²⁶ Studies of X chromosome's effects on cardiovascular diseases, metabolism, and inflammation showed that two X chromosomes confer a more significant disease load than one single X chromosome, and proposed as a possible explanation the enhanced expression of genes that escape X chromosome inactivation.29, 30, 31 Historically, the study of sex effects on PAH was hampered by confounding determinants of sex, the chromosomes, and the gender, the sex hormones, which affect tissue composition, metabolic activity, and response to active agents, combined with the lack of animal models displaying the same sexual dimorphism as humans.

This review highlights new findings related to how the sex chromosome complement and sex-differentiated epigenetic mechanisms to control gene expression are decisive players in the sexual dimorphism of PAH.

Genetic and Epigenetic Mechanisms in PAH

The complex pathogenesis of PAH is not well understood. As a general agreement, genetic, environmental, and local factors are modifiers of vascular structure-function involved in the initiation, propagation, therapeutic response, and ultimate outcomes of the disease. A plethora of genomic mechanisms are increasingly recognized to participate in the pathogenesis of PAH. This article aims to outline new developments in the genomics of sexual dimorphism of PAH; for other facets of the disease, recent excellent reviews are available.32, 33, 34

The traits of familial PAH are inherited in an autosomal dominant manner, display incomplete penetrance, affect the female primarily, and have more severe outcomes in the male. Heritable PAH accounts for 6% to 10% of PAH cases,³⁵ with the bone morphogenetic protein receptor 2 (BMPR2) gene encoding a receptor in the transforming growth factor-β superfamily, being the most mutated. More than 380 different BMPR2 mutations are found in about 70% of heritable PAH and 10% to 40% of idiopathic PAH cases.³⁶ In BMPR2 mutation carriers, the penetrance is estimated to be 14% for males and 42% for females.³⁷ Even if the female sex is the most critical factor influencing the penetration of the BMPR2 mutations in human PAH, there is no clearly defined molecular driver of this trend.³⁸ BMPR2 mutations influence phenotype more obviously in male PAH patients.³⁹ Limited studies of the impact of genotype on phenotype and whether this influence is associated with sex reveal that the overall survival difference between mutation carriers and noncarriers was more evident in male patients, which is reflected by a higher mortality risk of male mutation carriers than that of male noncarriers after adjustment for age at diagnosis. In females, this trend does not reach statistical significance. The authors hypothesize that pathogenesis of female PAH patients is more complicated, and the influence of BMPR2 mutations may be modified by other unknown factors (ie, critical molecules in the BMPR2 signaling pathways, level of BMPR2 expression, and sex hormones), making disparities in the prognosis between female mutation carriers and noncarriers less evident.

The development of high-throughput sequencing approaches led to novel causal genes and additional pathways involved in PAH susceptibilities, such as pathogenic or likely pathogenic genetic variants in potassium channels [potassium two pore domain channel subfamily K 3, ATP binding cassette subfamily C member 8, and transcription factors (T-box transcription factor 4 and SRY-box transcription factor 17)].40, 41, 42, 43, 44, 45 Recently, Hodgson et al⁴⁶ reported the association of the heterozygous mutations in the gene encoding the growth differentiation factor 2 and in two ligands for the BMPR2, the BMP type 9 and type 10, resulting in BMP9 loss of function and PAH.⁴⁶ Patients with PAH who carry these mutations exhibit reduced plasma levels of BMP9 and reduced BMP activity. Interestingly, although overall BMP9 and BMP10 levels did not differ between patients with PAH and control subjects, BMP10 levels were lower in PAH females.

Besides genetic mutations, the changes in gene expression that occur without alteration of the DNA sequence may contribute to pulmonary artery endothelial cells (PAECs) and pulmonary artery smooth muscle cell proliferation and vascular remodeling in PAH. DNA methylation, histone modification, and noncoding RNAs, able to turn the genes off and on, have been associated with PAH pathophysiology. DNA methylation analysis identified genes involved in lipid transport pathway, actin cytoskeletal rearrangements, cell migration, and proliferation, which could be relevant to PAH pathophysiology.^47,48 Although limited studies report accurate quantitative methylation measurements and their correlation with gender, abnormal DNA methylation patterns with a tendency toward higher methylation in males are observed in many diseases, such as tumors and imprinting disorders.^49,50 Studies have shown that histone deacetylases (HDACs) play crucial roles in controlling left ventricular cardiac remodeling in response to stress.⁵¹ Cardioprotection, resulting from the absence of HDACs 5 or 9 in female mice, can be attributed, at least in part, to enhanced neo-angiogenesis in the infarcted region via up-regulation of the estrogen receptor target gene vascular endothelial growth factor-α.⁵² It has also been reported that HDAC Sirtuin 1 (SIRT1), a potent negative regulator of pulmonary artery smooth muscle cell proliferation, exerts sexually dimorphic effects in atherosclerotic cardiovascular disease.^53,54 In PAH, SIRT1 deficiency can be a factor favoring vascular remodeling.⁵⁵ Bromodomain proteins have been identified as critical epigenetic drivers for cardiovascular disease, including PAH.^56,57 During mammalian development, one of the X chromosomes must be silenced in females to dosage compensate the X-encoded gene expression levels between males (XY) and females (XX). The lncRNA Xist gene, a set of 15,000 to 20,000 nucleotide sequences localized in the X chromosome inactivation center of chromosome Xq13.2, is essential for X chromosome transcriptional silencing; this is a tightly regulated process that involves epigenetic mechanisms, including histone deacetylation and methylation of CpG islands.58, 59, 60 The noncoding RNAs, including miRNAs and lncRNAs, are less studied mechanisms that play a critical role in gene expression by degrading their target mRNA and/or inhibiting their translation.⁶¹

Given the low number of cases identified/studied and lack of large-scale genetics/genomics studies, the available genetic data are limited, and the influence of most genetic variants on phenotype and whether this influence is associated with sex are still not known. The need for these genotype-phenotype correlations in causative genes and PAH phenotypes is critical. They may have clinical implications for lung vascular function and can help identify molecular determinants and mechanistic pathways for novel therapeutic approaches.

Sexual Dimorphism in PAH

There is a general agreement that PAH is a sexually dimorphic disease, more common in women.⁶² Data reported in several registries indicate a female predominance in PAH of around 2 to 4 over men for all races and ethnicities and across all ages that have been studied to date, except for HIV-associated and portopulmonary hypertension, which occur more often in men.⁶³ Yet, the severity and prognosis of the disease are worse in male than female subjects; these discrepancies are known as the estrogen paradox or estrogen puzzle of PAH.^3,64,65 Additional inconsistencies are related to the cardioprotective effects of estrogen in the right ventricle, and disease-promoting effects in the pulmonary vasculature. It is also unclear why estrogens are protective in various animal models of PAH, whereas PAH registries consistently demonstrate a female susceptibility to the disease. Another PAH enigma is why estrogens are protective in several animal models of PAH, but disease promoting in others. The worse disease severity for the male gender was rationalized to result from the strong protective effect of estrogens; however, effects of multiple sex hormones, receptors, metabolites, a better right ventricle functional adaptation in the female subjects, sex chromosomes, genetics, and epigenetics should be considered for explaining and understanding the estrogen paradox or estrogen puzzle of PAH.^3,65

Recent studies by Oliva et al,⁶⁶ designed to identify how sex and genetics interact to influence complex traits and disease, reported sex differences in gene expression level in all tissues, with most of the genes being relevant to disease and clinical phenotypes. Although most sex-biased genes were autosomal, the most robust sex bias was observed for X chromosome genes.⁶⁶ Sex differences were also identified in tissue cell type composition and cis expression quantitative trait loci, with sex-differentiated effects. Gene-trait associations driven by genetic regulation of gene expression in single sex have been identified, as well. The high tissue specificity of sex-biased gene expression and the enrichment of transcription factor binding sites link specific transcription factors in mediating sex-biased gene.⁶⁶ Signaling events via two of X-linked transcription factors, Elk1 and androgen receptor, have already been implicated in PAH pathophysiology in a sex-specific manner.67, 68, 69 Studies to identify gene sets enriched for genes highly expressed in females or males across multiple tissues identified a top-scoring cluster of genes corresponding to targets of the polycomb repressive complex 2 and trimethylation of histone H3 at Lys27 predominantly driven by female-biased genes. Significantly, this complex induces gene silencing and is involved in chromosome X inactivation.⁶⁶ The binding of the polycomb repressive complex 2 subunit Ezh2 to Xist RNA is critical for Xist-mediated chromosome silencing.⁷⁰

Sex is correlated with the tissue cellular composition.^66,71 The study of 53 healthy human tissues has established that >6500 protein-coding genes are differentially expressed in men and women and generate a molecular basis to explain why some diseases are more prevalent in one sex or another.⁷¹ This data-driven computational biology approach yielded a plethora of discoveries. The study pointed out that the genes expressed only in the left ventricle of the female heart might play a role in sex-dependent patterns of cardiovascular diseases. It is also worth mentioning that of the 6500 protein-coding genes differentially expressed in men and women, >1000 belong to the female extra X chromosome. Furthermore, a significant association between sex-specific gene transcription and reduced selection efficiency and accumulation of deleterious mutations, which might affect the prevalence of different traits and diseases, has been reported. Altogether, the observations strongly suggest that the disease may alter cellular abundances in a sex-differentiated manner or in sex-specific pathology.

Sex Chromosomes in PAH

The study of the effects of sex on PAH was hampered by two confounding determinants of sex: the chromosome complement and the sex; the sex hormones that affect the tissue composition, metabolic activity, and response to active agents, combined with the lack of animal models displaying the same sexual dimorphism as the humans, are considered additional hampering factors. The X chromosome carries 900 to 1500 genes, of which 867 are known protein-coding genes and the rest code for different RNA types.^25,26 Pathogenic variants of many of these protein-coding genes that induce complete loss of function may be lethal to fetuses of both sexes. Yet, several less severe pathogenic variants or pathogenic variants occurring in less-essential protein genes cause at least 533 X-linked diseases that affect males more severely than females.²⁵ Rather than influencing sexual development, most of these genes play a role in nonreproductive human tissues, including brain, bone, blood, ears, heart, lung, liver, kidney, retina, skin, and teeth. As a general rule, most women do not manifest X-linked disorders as they are not homozygous for the pathogenic variant and because their variant cells (those expressing the deleterious allele) receive sufficient gene product to perform the essential metabolic function from the cells that transcribe the normal allele. For an in-depth analysis of the female X chromosome protective effect, refer to several comprehensive reviews.^25,72,73

To delineate the role of sex chromosomes from that of the gonadal hormones in disease, the four core genotypes (FCG) mouse model was generated and used to produce significant insights into the effects driven by the chromosomes for different illnesses, PAH included.⁷⁴ Data generated using this FCG mouse, XX mice with either ovaries or testes and XY mice with either ovaries or testes,⁷⁵ have revealed that X chromosome dosage impinges in adiposity, hyperlipidemia,⁷⁶ inflammation, metabolic regulation,⁷⁷ and disease susceptibility to cardiovascular diseases irrespective of male or female gonads.⁷⁸ Although informative for chromosomal participation in different cardiovascular diseases, the use of FCG mice has shown that the magnitude of angiotensin II–induced hypertension is more significant in gonadectomized XX than the XY mice, regardless of their born sex, male or female.⁷⁹ Females with intact ovaries are protected from hypertension because of vasodilatory mechanisms mediated by endothelium-derived hyperpolarizing factors and the angiotensin II receptor. Although the endothelium-derived hyperpolarizing factor only requires estradiol, angiotensin II receptor-mediated vasodilation requires both estradiol and the XX sex chromosome complement.⁸⁰ Thus, ovarian hormones and sex chromosomes appear to have complex effects on the susceptibility to hypertension, similar to that in other cardiovascular diseases such as ischemia/reperfusion injury.⁷² Therefore, dissecting the impact of sex chromosomes from the sex hormones is not straightforward, and the need to have animal models of PAH that recapitulate the human disease remains.

Using the FCG murine model, Umar et al⁷⁴ have found that XY mice, irrespective of the gonadal sex, developed less severe PAH compared with XX mice. Several chromosome Y genes expressed in the lung (Ddx3y, Eif2s3y, Kdm5d, and Uty) with potential impact on cell proliferation, apoptosis, and epigenetic regulation support a protective role of the Y chromosome in PAH.⁸¹ Therefore, it was hypothesized that one or more of these genes might be responsible for the protective effect of the Y chromosome. However, in subsequent studies, each of the four candidate Y chromosome genes was independently knocked down via intratracheal instillation of siRNA in male mice exposed to hypoxia. Interestingly, except the knock down of the ubiquitously transcribed tetratricopeptide repeat-containing, Y-linked (UTY), the studies failed to validate a protective role of the Y-linked genes in PAH.^82,83 The same group also reported that UTY is protective in PAH by reduction of proinflammatory cytokines and protection against PAEC death.⁸³ A protective role of the Y chromosome in PAH is also supported by recent chromatin immunoprecipitation studies using male fibroblast cell lines; the endogenous Y chromosome testis-determining gene Sry, a member of the Sry-like box family of transcription factors, expressed exclusively in males, bound to and positively regulates the BMPR2 promoter. As a reduced expression of the BMPR2 is detrimental in PAH, the positive regulation of BMPR2 by Sry may contribute to the protective role of the Y chromosome in PAH.⁸¹

Although useful for directly testing the contributions of sex chromosome complement versus gonadal sex, the FCG model does not control for the X chromosome copy number or the presence of a Y chromosome. To address this limitation and to distinguish whether a phenotypic effect is caused by the number of X chromosomes or the presence/absence of the Y chromosome, studies using the XY∗ model, discovered by Eva M. Eicher, have been proposed for assessing the X or Y candidate genes and sex chromosome effects.^78,79

Previous work has recognized most, but not all, of the attributes and limitations of available animal models of PAH.84, 85, 86 Although these animal models are informative for molecular mechanisms and cellular participants in the PAH pathogenesis, essential characteristics of human PAH, such as its sexually dimorphic incidence, development, and therapeutic response, are not sufficiently addressed. As already known, females are more often affected, and looking to the mechanisms behind female susceptibility to PAH is of paramount importance. X chromosome participation in determining human PAH sexual dimorphism is still underappreciated. Possible mechanisms behind a female's susceptibility to X chromosome–linked diseases may come from having more than one copy of a pathologic variant, a relevant second variant, a variant in an essential gene that does not permit males to survive, an enhanced susceptibility to chance to skew, which favors the mutant allele, the effects of X chromosome aberrations (ie, translocations), and monozygotic twinning on inducing unfavorable skewing.⁸⁷ Therefore, the affected endothelial cells in PAH display more severe DNA damage, chromosomal abnormalities, and microsatellite instability, factors that underline their unstable genome.88, 89, 90, 91 How and whether all these modified endothelial cell features participate in sexual dimorphism of the disease is an area of active research and debate.

Mechanisms of Sexual Dimorphism Related to the X Chromosome: Potential Roles of X-Linked Genes

Recently, using an intersectin-1s (ITSN) protein fragment with endothelial cell proliferative potential, it has been shown that the cultured human PAECs expressing this fragment are more proliferative, similar to the human PAECs of PAH patients.^69,92 This proliferative ITSN fragment, EH_ITSN, is the NH₂-terminal fragment of ITSN (a prominent protein of the lung tissue⁹³), result of granzyme B cleavage under inflammatory conditions associated with PAH; ITSN is a substrate for granzyme B, with a cleavage site at IDQD²⁷¹GK, well conserved throughout the evolution.^92,94 The studies have also shown that the human PAECs are sex dimorphic in the proliferative potential, with female cells being more proliferative than the male ones. Moreover, in vivo studies demonstrated that prolonged expression of the EH_ITSN in the intersectin-1s heterozygous knockout (K0^ITSN+/–) murine lung triggers a severe PAH phenotype, including plexiform arteriopathy; similar to human, the PAECs of female EH_ITSN-transduced K0^ITSN+/– mice have a higher proliferation rate compared with the PAECs of EH_ITSN-transduced K0^ITSN+/– male mice.⁶⁸ The increased responsiveness of female endothelial cells to the EH_ITSN expression is due to up-regulation above basal levels of the lncRNA Xist, essential for X chromosome inactivation and dosage compensation of X-linked gene expression.^26,68 This EH_ITSN-transduced K0^ITSN+/– mouse model of PAH recapitulates most of the sex-specific differences of human disease. Like human female endothelial cells and human female PAH lung specimens, the female EH_ITSN-K0^ITSN+/– mice show up-regulation of the expression and activity of lncRNA Xist compared with baseline healthy state. Down-regulation of Xist gene expression via siRNA or its inhibition via a particular inhibitor of the EH_ITSN abolishes the increase of cell proliferation, suggesting that up-regulation of lncRNA Xist expression and transcriptional activity may explain the sexual dimorphism in the proliferative potential of female PAECs and the increased women's susceptibility for PAH.⁶⁸ The expression of EH_ITSN in the lung tissue of several PAH animal models and human patients supports this idea.^92,95

The Xist locus (X chromosome: 73820650-73852753 in humans, and X chromosome: 103460372-103483233 in mouse) plays a critical role in X chromosome inactivation to ensure equivalent levels of X-linked genes between male and female.⁹⁶ X chromosome abnormalities have been reported in PAH.⁸⁹ lncRNA Xist expression was related to fibroblast and smooth muscle cell proliferation.^97,98 Improper Xist expression has been linked to abnormal cell proliferation and poor survival in different human cancers.^92,99,100 Altered Xist expression and mislocalization have been related to female-specific overexpression of X-linked genes in autoimmune diseases that predominantly affect women.¹⁰¹ However, how the improper Xist levels may cause sex-specific alterations in the expression of protein-encoding genes carried on human and mouse X chromosome and documented to be involved in PAH pathogenesis⁶⁸ is still unclear. X-linked genes may also encode for miRNAs and lncRNAs, altogether involved in altered metabolism, apoptosis, proliferation and vascular remodeling, vasoconstriction, and inflammation, frequent pathologic manifestations characteristic to PAH.^68,69,102 The mechanisms underlying the effects of lncRNA Xist up-regulation by EH_ITSN expression might involve multiple functions of the lncRNA Xist regulatory network, like its direct interaction with gene promoters, the involvement of cis-regulatory elements for different genes, control of transcription factor expression, chromatin modifications toward an active or repressed conformation, higher-order chromatin structure, and miRNA activation or repression via sponging (Figure 1). Several studies proposed that Xist may act as a competing endogenous RNA by depleting miRNAs. By doing so, specific RNA targets cannot be degraded, leading to dysregulation of downstream genes.¹⁰³ Recent studies from several laboratories demonstrate that many PAH inductors converge on the p38 mitogen-activated protein kinase pathways.^68,92,104, 105, 106, 107 Thus, besides charting a general molecular mechanism underlining PAECs' enhanced proliferation in PAH (Figure 1), the recent studies add new molecular players centered on p38-Elk1 activation as an alternative to the suggested mechanisms centered on embryonic vasculogenesis.¹⁴ lncRNA Xist up-regulation augments the expression of Elk1, an X-linked gene implicated in PAH pathobiology. The increased expression and activity of Elk1 contribute to the severity of lung phenotype in the female EH_ITSN-K0^ITSN+/– mouse model of plexiform arteriopathy. Recent peptide-based kinase activity studies revealed Elk1 as a substrate of the cdk2 (the regulatory kinase of Ccna1), overactivated in PAH,⁶⁷ suggesting a positive feedback loop in which cdk2 phosphorylates Elk1 to regulate Ccna1 expression and cell cycle progression.⁶⁸ Ccna1 is required for entry into the S and M phases of the cell cycle,¹⁰⁸ whereas Ccnd2 promotes progression through G₁.¹⁰⁹ Besides, the Ccna1 has a putative binding site for Elk1, shown to contribute to cell cycle progression in a tissue- and cell-specific manner.^110,111 The Ccna1 promoter (the region spanning –1040/–980 in human Ccna1 and –980/–920 in mouse Ccna1) contains the Elk1 consensus motif.^110,112,113 Also, the Ccnd2 is a target of Elk1.¹¹⁴ Xist up-regulation may also account for the sex bias in PAH by modulation of its downstream target, Kruppel-like factor 2 (Klf2).⁶⁸ Klf2, a central transcriptional switch point between the quiescent and activated endothelium,¹¹⁵ shows sex-specific repression.⁶⁸ Although expression of Klf2 is protective in PAH,^116,117 its protective role has never been related to sex differences and female prevalence in PAH. Up-regulation of the lncRNA Xist leads to differential sex-specific modulation of X-linked Elk1 transcription factor^67,118 and of a cell cycle regulatory protein, cyclin A1 (ccna1). These molecular events are more prominent in female EH_ITSN-transduced K0^ITSN+/– murine model of PAH and human female lung specimens compared with males, strongly suggesting that the increase in the lncRNA Xist expression in the female PAECs accounts, at least in part, for the sex/ratio imbalance in PAH. Furthermore, the lncRNA Xist, by regulating the expression of the X chromosome–linked gene Elk1, is a determining factor behind the sexually dysmorphic response of endothelial cells to the EH_ITSN fragment. Thus, Xist participates directly and differentially in establishing the proliferative profile of PAECs, strongly suggesting a sex-specific mechanism for endothelial cell augmented proliferation in PAH. Moreover, a cross talk via P-Elk1/Ccna1/d2 signaling between the general and sex-specific mechanisms cannot be ruled out, and thus, p38 mitogen-activated protein kinase signaling may be connected to the sexually dimorphic response of endothelial cells in PAH.

Figure 1.

Proposed NH₂-terminal fragment of intersectin-1s (EH_ITSN)/X-inactive specific transcript (Xist)–mediated molecular mechanism of the pulmonary artery endothelial cells’ (PAECs') sexual dimorphism in the proliferative response. The EH_ITSN protein fragment translocates to the nucleus to activate p38 mitogen-activated protein kinase (MAPK). In turn, the phosphorylated p38 kinase (P-p38) not only phosphorylates but also interacts with the transcription factor Elk1, a molecular event required for the recruitment of activated Elk1 [phosphorylated Elk1 (P-Elk1)] to the c-Fos promoter. Thus, p38 MAPK activation mediates the proliferative potential of EH_ITSN, via downstream activation of the Elk-1 transcription factor and of the immediate early response gene c-Fos. This p38/Elk1/c-Fos signaling is a general molecular pathway behind PAEC proliferation and vascular remodeling in pulmonary arterial hypertension (PAH; blue design). The nuclear translocation of the EH_ITSN protein fragment also leads to a sex-specific increase in the expression/activity of the long noncoding RNA (lncRNA) Xist; the downstream molecular events triggered by the EH_ITSN to increase the expression/activity of Xist (triple dashed lines) are under investigation. Moreover, down-regulation of Klf2, a Xist target and a negative regulator of vascular endothelial growth factor receptor 2 (VEGFR2) signaling module, leads to additional PAEC proliferation, via a sex-specific mechanism (purple design), involving the cyclin A1 (Ccna1) and cyclin D2 (Ccnd2), two cell cycle regulatory proteins. A cross talk via P-Elk1/Ccna1/Ccnd2 signaling between the general and sex-specific mechanisms (dashed-dotted line) cannot be ruled out. By up-regulating the genes of two transcription factors, Xist participates directly and differentially in establishing the proliferative profile of PAECs. The finding of lncRNA Xist direct modulation of Klf2, one of the pillars of endothelial phenotypic stability, brings a new dimension of how the sexual dimorphism displayed by the endothelial cells is determined. The plethora of new molecular pathways, still to be discovered, centered on lncRNA Xist (gray inset), should be studied to have a complex picture of the sexual dimorphism determination and to unravel more precise therapies for PAH.

Dissecting the complicated relationship between lncRNA Xist and X chromosome–linked genes and between the lncRNA Xist and the expression of somatic genes and how all these gene control levels integrate into a sexually dimorphic proliferation of PAECs remain challenging. Thus, more work is needed to: i) decipher the steps from the molecular pathways by which lncRNA Xist imposes a dimorphic female phenotype, ii) develop tools using Xist as a diagnostic and prognostic marker for PAH, and iii) try to use this as new therapeutic targets for precision PAH treatment.

New Molecular Mechanisms: New Pharmacologic Opportunities

Despite advances in our understanding of PAH pathogenesis, there is no curative treatment for treating PAH. Identifying novel, specific therapeutic targets and more effective treatments against the progression of PAH is necessary. As recent studies suggest that epigenetic modifications [ie, DNA methylation, histone post-translational modifications (acetylation and methylation), miRNAs, and other noncoding RNAs] may play a critical role in the pathogenesis of PAH, the recent identification of new pharmacologic drugs targeting these epigenetic dysregulations has opened new therapeutic avenues for PAH.¹¹⁹ The great potential and advantage for epigenetic therapies reside in the fact that, unlike genetic abnormalities, epigenetic changes are reversible, allowing recovery of function for affected genes with normal DNA sequences. Thus, given the development of new pharmacologic drugs for treating several cancers, some of them approved by the US Food and Drug Administration, new opportunities to exploit the current therapeutic strategies used in cancer to treat PAH, which displays many common features with cancer (ie, proliferation, apoptosis-resistance, and dysregulation of tumor suppressor genes^119,120), should be considered. Of interest are the bromo-domain containing protein 4 inhibitors, currently tested in clinical trials for cancer. Molecular and pharmacologic inhibition of bromo-domain containing protein 4 reversed established PAH in the Sugen/hypoxia rat model.⁵⁷ Of particular interest, the isoform-selective HDAC inhibition seems to be better tolerated than nonselective pan-HDAC inhibition and inhibition of the HDAC6 that displays deacetylation activity toward nonhistone proteins.^121,122 It has been reported that tubastatin and ACY-775 improves pulmonary hypertension in the Sugen/hypoxia and monocrotaline rat model, as demonstrated by decreased right ventricular systolic pressure, mean pulmonary artery pressure, cardiac output, total pulmonary vascular resistance, right ventricle hypertrophy, and vascular remodeling, and restoration of the proliferation/apoptosis balance.^61,123

p38 kinase is a central molecular mediator of vascular remodeling in both human disease and experimental models, and a converging signaling pathway in PAH.⁶⁸ Pharmacologic inhibition of p38, in both chronic hypoxic and monocrotaline rodent models of pulmonary hypertension, prevents and reverses the pulmonary hypertensive phenotype.¹⁰⁴ The newly developed p38 inhibitors, with greater selectivity and inhaled version, may provide the opportunity, still to be validated by recently completed clinical trials, to enhance p38 inhibition in the lung, increasing its tolerability, safety, and efficacy, while reducing unwanted systemic effects of p38 inhibition.¹²⁴ Finally, exploring the therapeutic potential of targeting the lncRNA Xist to inhibit female PAEC proliferation and ameliorate vascular remodeling could be another exciting new opportunity for PAH treatment. Recent studies demonstrate that a small EH_ITSN inhibitory peptide or siRNA against Xist decreases female PAEC proliferation via inhibition of the p38-Elk1-cFos pathway.⁶⁹ In addition to its original role in X chromosome dosage compensation, lncRNA Xist also participates in the development of multiple types of tumors, including brain tumor, leukemia, lung cancer, breast cancer, and liver cancer. lncRNA Xist also contributes to other human diseases, such as pulmonary fibrosis, inflammation, neuropathic pain, cardiomyocyte hypertrophy, and osteoarthritis.⁶⁰ Further studies to better understand the mechanisms of lncRNA Xist action and how its regulatory network participates in various signaling pathways related to PAH may provide important insight into how lncRNA Xist can be used as a therapeutic opportunity in female-biased disorders.

Gender Bias in Treating PAH: The X Factor

PAH presents a gender bias regarding not only its mechanisms, prevalence, and prognosis, but also the response to treatment. The past 20 years have seen major changes in the treatment of PAH. The use of combination therapy for simultaneous targeting of more than one of the signaling pathways implicated in disease progression or to address the multiple PAH comorbidities improve outcomes.¹²⁵ The early diagnosis and treatment of comorbidities became an important tool together with the treatment of PAH itself. Understanding the magnitude and significance of comorbidities in PAH is of paramount importance for clinicians. PAH comorbidities, based on their prevalence in patients with idiopathic PAH in the Registry to Evaluate Early and Long-Term PAH Disease Management, include systemic hypertension, obesity, obstructive sleep apnea (OSA), clinical depression, obstructive airway disease, such as chronic obstructive pulmonary disease (COPD), thyroid disease, diabetes, and ischemic cardiovascular events.¹²⁶ Most of PAH comorbidities are sex biased. Type 2 diabetes and ischemic heart disease are more prevalent in men, but hypertension and depression are more prevalent in women with OSA.¹²⁷ Human sex chromosome as well as mouse models with engineered alterations in sex chromosome complement support an important role for sex chromosomes in obesity and metabolism. Females are more likely to develop obesity.²⁹ Syndromic obesity arises from discrete genetic defects or chromosomal abnormalities at several genes and can be autosomal or X-linked. FCG mouse model has revealed an effect of X chromosome dosage on adiposity and hyperlipidemia, with no influence from the presence of the Y chromosome.¹²⁸

Women have a 1.5- to 2-fold increased lifetime prevalence of depression compared with men.¹²⁹ Sex differences in psychiatric disorders are the result of a complex interplay between genetic, hormonal, immunologic, and psychosocial factors. However, growing evidence suggests that additional sexually dimorphic biological factors, including X chromosome and Xist, the master regulator of X chromosome inactivation, are associated with observed sex differences in depression and anxiety.129, 130, 131

The female lung is more susceptible to obstructive airway diseases (COPD; ie, asthma, chronic bronchitis, emphysema, cystic fibrosis, and bronchiolitis) than the male lung, and women develop symptoms of the disease at a younger age with less tobacco exposure than men.^132,133 Women with COPD are often misdiagnosed and disproportionally experience comorbid conditions, including anxiety and depression. Most recently, whole-genome sequencing analysis of 19,996 participants from the National Heart, Lung, and Blood Institute Trans-Omics for Precision Medicine Program revealed a strong association with colocalization and replication of X-chromosome variants in the RGN (regucalcin; alias senescence marker protein 30) gene in association with lung function measures enriched for COPD.¹³⁴ Anatomic factors and the concept of dysanapsis (a mismatch between the size of the airway tree and lungs; the luminal area of the larger conducting airways is smaller in women than in men when matched for lung size) may explain why women are at increased risk of airway diseases.¹³⁵ Several cohort studies of patients in the Genetic Epidemiology of COPD study support the existence of a sex-related genetic component in COPD onset, including a higher risk of early-onset severe COPD in female smokers.^136,137 The X chromosome may be involved because COPD in a mother increases the risk of COPD onset in a daughter who smokes.¹³⁶ Women are twice as likely as men to have asthma; it has been reported that this gender difference may be caused by the effects of sex hormones on lung cells and the presence of a susceptibility gene for asthma on chromosome X.^138,139

OSA, another PAH comorbidity, shows marked sexual dimorphism in disease prevalence and progression; earlier studies indicated that men were as much as nine times as likely to have OSA as women. Today, the gap between men and women is nowhere so large. Limited genome-wide associated tests identified an association with apnea-hypopnea index in men in a region that includes strong biological candidate genes, but not in women.^140,141 The different OSA prevalence between males and females has generated interest in sex-related aspects of OSA pathophysiology. Anatomic and physiological features of the upper airways, the modulating effects of sex hormones on control of breathing, and sex-dependent features of fat distribution in obesity may account for OSA and other sleep-disordered breathing's sexual dimorphism.¹⁴⁰ Although the prevalence and severity of OSA may be lower in women compared with men, the consequences of the disease are similar, if not worse. Women with OSA may have greater risk for hypertension and endothelial dysfunction, be more likely to develop comorbid conditions such as anxiety and depression, and have increased mortality. Animal models have been developed, allowing to assess sex-related differences in sleep structure and acquisition of phenotypes during early development. Interesting findings from the FCG mouse models suggest that sex chromosome complement contributes to the establishment of sex differences in sleep.¹⁴² Following a period of sleep deprivation, females with the XY compliment acquire more sleep during their mid-active phase and have higher non–rapid eye movement sleep delta power (quantitative measure of sleep intensity; ie, analogous to slow wave activity in human) than XX females, suggesting that the processes mediating recovery from sleep loss are partially dependent on sex chromosomes.¹⁴³

Women have about 10 times higher risk for thyroid disease than men. Accumulated evidence indicates that the degree of X chromosome inactivation is an important contributor to the females' predisposition in developing autoimmune thyroid diseases.¹⁴⁴ It is well established that the incidence and progression of heart disease is markedly higher in men than in age-matched women before menopause. The female sex hormones were thought to account for the lower incidence of coronary artery disease in women.¹⁴⁵ Interestingly, studies of the FCG mice demonstrated that the number of X chromosomes also plays an important role in myocardial susceptibility to ischemia/reperfusion injury. XX mice have higher vulnerability to myocardial ischemia/reperfusion injury compared with XY mice, which is due to the number of X chromosomes rather than the absence of the Y chromosome.¹⁴⁵

Over the past few years, it became clear that sex affects cell physiology, metabolism, and many other important biological functions. Sex affects the symptoms and manifestations of disease. Sex affects responses to treatment. This knowledge is critical for transition into an era of precision/personalized medicine, beyond oncology. The prevention, management, and therapeutic treatment of many diseases, PAH included, should reflect the most obvious and important risk factor for the patient: sex. These findings indicate the importance of considering sex when developing and testing epigenetic therapeutics, specifically those that target misregulation of X-linked genes. Despite significant challenges in developing sex-specific interventions to specifically target abnormal cells with minimal or no damage to normal cells, the epigenetic therapeutics in the treatment of disease are promising.

Conclusions and Future Perspectives

This review highlighted some of the latest findings related to how sex chromosomes and sex-differentiated epigenetic mechanisms controlling gene expression are decisive players in the sexual dimorphism of PAH. Investigations of lncRNA Xist may lead to a better understanding of its functions and regulatory networks, and a better understanding of how the lncRNA Xist contributes to disease in humans. The new findings may have clinical implications for lung vascular function and can help identify mechanistic pathways and molecular determinants suitable for pharmacologic intervention. Reversing abnormal expression of the X-linked genes in affected females may potentially be a new strategy for future treatment of X-linked disorders.