ECs play a wide range of different functions within the microvasculature. They regulate the vascular tone by virtue of the dynamic crosstalk with VSMCs, PVAT and renin-angiotensin-aldosterone (RAAS) system membrane receptors and neuronal terminations. ECs also transduce mechanical stimuli from the local bloodstream flow and can act as nutrient sensors and mediators of insulin signaling. Moreover, they have been reported to be promoters and perpetrators of the microvascular inflammatory response, impaired angiogenesis, the modulation of salt sensitivity and the response to sex hormones. The centrality of the vascular endothelium in all these functions ^[1][31] makes it an essential actor in the healthy-to-disease transition and has placed it at the center of vascular research over the last forty years ^[2][32]. The endothelium is characterized by an array of different epigenetic signals that fine-tune the transcriptome, thus securing cellular identity and function. For example, specific DNA and histone methylation patterns define an active vs. quiescent endothelium. These two endothelial phenotypes change during life but might be simultaneously present within the same organ or tissue and even within the same microvessel. This can be observed by quantifying the different expression of the von Willebrand Factor (vWF) among neighboring endothelial cells from the same vascular bed. Whilst, during life, in some tissues, vWF expression is either low (e.g., liver sinusoidal) or constant (e.g., aorta), indicating a more quiescent endothelium, in others (e.g., heart, skeletal muscle, brain), the expression is higher and very variable, thus suggesting endothelial activation. This dynamic epigenetic mosaicism represents the molecular basis of adaptive endothelial homeostasis. It is regulated by a DNA methylation of CpG islets of the vWF promoter, which occurs randomly and dynamically during the lifetime, allowing for a faster response time and adaptation to biological stimuli ^[3][33].

Many of these functions are dependent on the capacity of the endothelium to generate nitric oxide (NO)—a crucial mediator of ECs function and homeostasis—by endothelial nitric oxide synthase (eNOS). Reduced NO availability is the hallmark of endothelial dysfunction ^[4][8]. The eNOS promoter exhibits a unique histone code characterized by the enrichment of active epigenetic signatures, including H3K9Ac, H4K12Ac, H3K4me2, H3K4me3, H3K27me3 and the H2A.Z histone variants in ECs ^[5][34]. In AH, this code is altered and leads to an impairment in eNOS expression and activity ^[6][35]. Endothelial-dependent relaxation impairment in AH, as well as in other cardiometabolic diseases ^[7][16], is also mediated by epigenetic pathways involving the adaptor p66Shc ^[8][36] and JunD ^[9][10][37,38]. In hypertensive rats, resveratrol, a non-selective silent information regulator 1 (SIRT1) activator ^[11][25], is associated with the restored methylation of H3K27me3 on the eNOS promoter ^[12][39] rescuing endothelial function, as also observed in human models of cardiometabolic disease ^[13][24].

The endothelium also modulates the response of the vessels to neurohormonal stimuli. Changes in the epigenetic landscape play an essential role in this respect. The enrichment of H3K4me3 and decreased H3K9me2 levels were found at the angiotensin-converting enzyme (ACE)-1 promoter, which is associated with ACE-1 upregulation ^[14][40]. In vitro and in vivo experiments have shown that AH leads to the hypermethylation of ACE-1 promoter/reporter constructs of the somatic ACE (sACE) promoter, downregulating its expression and reducing its transcriptional activity ^[15][30]. The angiotensin II receptor type 1a gene (ATGR1A) promoter in ECs from the aorta and mesenteric arteries of SHR rats shows progressive hypomethylation with age compared to wild-type animals ^[16][41]. Human endothelial cells isolated from the placenta of pre-eclamptic women show higher miR-155 levels ^[17][42] as a compensatory mechanism for reducing the levels of angiotensin II receptor type 1 (AGTR1) mRNA. This miRNA targets the polymorphic sequence in the 3′UTR of AGTR1 mRNA ^[18][43].

Additionally, angiotensin II recruits SET1, a histone H3K4 tri-methyltransferase, to the promoter of endothelin-1 ^[19][44] by activating protein 1 (AP1) to methylate H3K4. This increases the expression of endothelin-1 (ET-1), leading to uncontrolled AH and subsequent organ damage, namely, cardiac hypertrophy. miR-125a-5p and miR-125b-5p were shown to downregulate ET-1 expression in ECs, in line with their reduced level in hypertensive rats ^[20][45]. In normal conditions, aldosterone regulates sodium reabsorption due to the inactivation of cortisol to cortisone by 11β-hydroxysteroid dehydrogenase (HSD11B2). In AH, the ECs HSD11B2 gene promoter is hypermethylated ^[21][17], while its H3K36me3 levels are reduced ^[22][46]. This leads to a tetrahydrocortisol/tetrahydrocortisone ratio, which is also a hallmark of AH ^[23][47].

The inflammatory response is also filtered by the ECs’ capacity to modulate or activate it firsthand. Inflammatory stimuli such as lipopolysaccharide (LPS) treatment promote the demethylation of H3K27me3 on the promoter of inflammatory genes by activating the demethylase JMDJ3 ^[24][48]. The inflammatory response provoked by S-adenosylhomocysteine by NF-κB also collaterally downregulates the methyltransferase EZH, thus reducing H3K27 trimethylation ^[25][49].

In ECs, a high-salt diet promotes the hypomethylation of the Solute Carrier Family 2 Member 2 (SLC2A2) gene, with subsequent higher levels of membrane transporter Na⁺-K⁺-2Cl-cotransporter 1 (NKCC1) ^[26][50] and the downregulation of histone lysine-specific demethylase 1 (LSD1), leading to an increased methylation of histone H3K4 or H3K9 ^[17][42]. Additionally, the hypertensive signature leads to impaired angiogenesis by upregulating miR-505 in ECs ^[27][51].

VSMCs function is tightly dependent on ECs. However, preserving VSMCs homeostasis is crucial, as their level of contraction/relaxation determines the vasomotor tone, with subsequent changes in the lumen diameter and arterial resistance. In the microcirculation, VSMCs contract in response to blood pressure variations within the physiological range. Their myogenic activity has both a phasic and static component. While the first is involved in the acute response to stimuli, the latter maintains the flow in homeostatic conditions. In AH, both components are impaired ^[28][52]. Another major VSMCs feature lies in their high plasticity. In healthy conditions, they display a fully mature contractile phenotype characterized by low proliferation and a hyperexpression of contractile proteins. In response to noxious stimuli, they can de-differentiate to a synthetic phenotype, characterized by the production of the extracellular matrix, proliferation, migration and angiogenesis. When sustained, it leads to inward hypertrophic remodeling ^[29][53], a hallmark of AH ^[30][54].

High blood pressure induces epigenetic changes that foster the de-differentiation and the consequent proliferative activity of VSMCs. A recent trans-ancestry genome-wide association study (GWAS) showed that different blood pressure phenotypes display opposite methylation levels in four genes involved in vascular tone and VSMC plasticity: insulin-like growth factor binding protein 3, potassium two-pore domain channel subfamily K member 3, phosphodiesterase 3A and PR domain-containing protein 6 ^[31][55]. AH was also associated with the downregulation of ten-eleven translocation-2 (TET2) ^[32][56]. TET2 is responsible for cytosine 5-hydroxymethylation on the promoters of myocardin, serum response factor (SRF) ^[33][57] and myosin heavy chain 11 ^[32][56], which are all involved in the contractile VSMCs phenotype. SRF and myocardin are also involved in the tight control of proliferation by upregulating the miR-143/145 cluster ^[34][58]. Mice lacking both miR-143 and miR-145 have reduced blood pressure but are more prone to vascular injury ^[35][59]. Although the reduced blood pressure might reflect the laxity of the vessel wall, reduced levels of miR-143 are also associated with contractile dysfunction in skeletal muscle cells from older mice ^[36][60]: the microcirculation structure and function might thus be directly affected by the VSMCs down-regulation of the miR-143/145 axis in age-related diseases such as AH. However, further studies are needed to elucidate miR-143/145-specific contribution in this context. Higher angiotensin II levels in AH decreased miR-365, increasing VSMC proliferation ^[37][61]. Similarly, miR-34b levels are reduced in spontaneously hypertensive rats, leading to a higher VSMCs proliferation rate ^[38][62]. On the other hand, the upregulation of miR-181b-5p prevents angiotensin II-induced VSMCs proliferation ^[39][63], while the upregulation of miR-155-5p inhibits VSMCs proliferation via suppressing angiotensin-converting enzyme expression ^[40][64]. However, persistently high levels of angiotensin II will eventually lead to the downregulation of miR-181b-5p and miR-155-5p, which are critical regulators of the VSMCs phenotype. VSMCs are also involved in hypertension-related microvascular inflammation, which, in turn, impairs VSMCs functionality. In hypertensive rats, the histone acetylases EP300-binding protein and CREB-binding protein increase the H3K9ac in the NLRP3 promoter. Its activation promotes the hyperactivation of the NF-κB pathway, ultimately resulting in VSMC remodeling, transformation and proliferation ^[41][65].

A VSMCs-altered contractile response is also a hallmark of AH. Whilst, in normal conditions, TET2 modulates the expression of SRF, allowing for the VSMCs between synthetic and contractile states, the hypertension-related downmodulation of TET2 ^[32][56] leads to a reduced response to vascular injury, and when VSMCs switch to the contractile state, this leads to the hyperacetylation of H3 and H4 on the smooth muscle cell 22 (SM22) ^[42][66] and myocardin genes by SRF. This induces increased vascular stiffness, a hallmark of AH ^[43][67]. VSMCs also overexpress miR-431-5p, worsening vascular stiffening ^[44][68]. In murine models, miR-153 is elevated and targets potassium voltage-gated channels, increasing the myogenic basal tone ^[45][69]. On the other hand, miR-328 is downregulated in VSMCs, and the loss of inhibition on the L-type voltage-gated calcium channel activity promotes AH. Peculiarly, aging upregulates miR-328 in the same murine model, pointing out its hypertension-specificity ^[46][70]. A high-salt diet leads to the hypomethylation of the SLC2A2 gene, increasing NKCC1 also in VSMCs ^[26][50]. Angiotensin also upregulates NKCC1, increasing H3Ac and decreasing H3K27me3 ^[47][71]. Moreover, the adaptor p66Shc restricts the activation of transient receptor potential cation channels, downregulating Ca2+ influx ^[48][72] and reducing spontaneous Ca2+ oscillations ^[49][73] on VSMCs.

Finally, several mechanisms implicated in VSMCs homeostasis are progressively impaired in AH. The overexpression of specific sirtuins, a histone deacetylase family, such as SIRT1 and SIRT6, inhibits VSMCs proliferation and extracellular matrix synthesis in hypertensive rats ^[50][74] and protects from vascular senescence by reducing telomere H3K9 acetylation ^[51][75], respectively.

Perivascular adipose tissue (PVAT) is crucial in maintaining microcirculatory homeostasis. First, it modulates the inflammatory response thanks to its paracrine and juxtracrine capacities. PVAT-secreted molecules (i.e., adipokines) such as tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), monocyte chemoattractant protein-1, leptin as well as angiotensinogen ^[52][76] directly affect the microvascular function. They modulate the inflammatory pathways ^[19][44], the metabolic response to external stimuli such as increased blood pressure or nutrient overload and also the microvascular contractility ^[52][76]. Moreover, in healthy subjects, PVAT has a protective anti-contractile property and a protective brown phenotype whose origin is still debated ^[53][77]. These features are lost in AH ^[54][78]. However, there is a lack of a complete understanding of the involvement of PVAT in the cellular crosstalk in the microvascular unit in conditions of health and disease, and evidence is sometimes difficult to interpret ^[55][79]. The molecular mechanisms of its epigenetic signatures under hypertensive damage are almost uncharted territory. In rats, a high-salt diet induces hypomethylation around two CCAAT/enhancer binding protein (CEBP) binding sites and a transcription start site, ultimately leading to upregulated angiotensinogen gene expression in PVAT ^[56][80]. Additionally, in the visceral adipose tissue of obese hypertensive patients, the β3-adrenergic receptor was found to be hypermethylated ^[57][81].

Although it might seem difficult to imagine the bloodstream flow as a homeostatic modulator, the pressure exerted on the microvascular wall, particularly its frictional component, i.e., the shear stress, impacts the functionality of the microvascular unit. Indeed, endothelial cells are professional flow sensors, and the mechanical signal exerted on them is transduced to all the actors involved in microvascular homeostasis. Indeed, physiologic shear stress exerts a protective function on the vascular wall in healthy conditions by regulating the expression of protective genes through mechanotransduction. On the other hand, in hypertensive subjects, its alterations lead to the hyperexpression or downregulation of specific elements, often by epigenetic means ^[58][82]. Different types of flows (e.g., laminar vs. disturbed) induce different methylation patterns in endothelial cells ^[59][83]. The adaptor protein p66Shc was shown to be implicated in endothelial damage by promoting a pro-oxidative environment after the cyclic stretch to the vascular wall ^[60][84].

Disturbed flow, as observed in the mouse after the partial ligation of the carotid artery, led to upregulation of the DNMT1. In HUVECs, oscillatory shear stress similarly led to DNMT1 upregulation, and the inhibition of DNMT1 prevents the monocyte adhesion that is a consequence of oscillatory shear stress ^[61][85]. Disturbed flow also leads to the hyperactivation of DNMT3A, leading to the hypermethylation of the Krupper-like factor 4 promoter, an anti-inflammatory and anti-thrombotic protein also involved in the modulation of eNOS signaling ^[62][86]. Histone deacetylase 3 in endothelial cells is also enhanced by a disturbed flow ^[63][87]. Similarly, it promotes the expression and the accumulation of histone deacetylases 1, 3, 5 and 7 ^[64][88], while it does not affect the expression of SIRT1, which is increased during normal pulsatile flow ^[65][89]. Finally, a disturbed flow regulates the expression of specific long noncoding RNA such as STEEL ^[66][90] and LEENE, which are both enhancers of eNOS function that are increased during pulsatile flow but decreased during oscillatory flow ^[67][91].

Though this rentryview will not extensively discuss their role, neural terminations and inflammatory cells are involved in the microcirculatory crosstalk. Indeed, the theory of the neurogenic component of AH strongly supports a direct epigenetic crosstalk between the central and the autonomic nervous systems and inflammatory elements ^[68][92]. However, their epigenetic cues have yet to be investigated accurately. On the other hand, there is evidence about the epigenetic networks involved in the transcriptional regulation of inflammation in different vascular cells in AH. Euchromatic histone-lysine n-methyltransferase 2, a methyltransferase specifically overexpressed in leukocytes in healthy conditions, is hypomethylated in hypertensive patients, as reported by a longitudinal GWAS ^[69][93]. Peripheral blood mononuclear cells from patients with AH have a specific miRNA profile, showing lower levels of miR-133, miR-143 and miR-145 and higher levels of miR-1 and miR-21 ^[70][94]. In macrophages, the deacetylation of NLRP3 by the overexpression of SIRT2 NAD+-dependent deacetylase represses inflammasome activation, preventing chronic inflammation ^[71][95] and suggesting its protective role in the context of cardiometabolic diseases such as AH. Finally, telocytes deserve to be mentioned. As an interstitial cell type found in various organs of the human body, including the microvasculature, they have a specific role in the context of the cardiovascular system, regulating the cardiac and vascular development, structure and tone ^[72][96]. Although their function in the microcirculatory system has yet to be elucidated, they have been recently shown to be involved in crosstalk with ECs ^[73][97] mediated by mi-21-5p signaling to improve the angiogenetic response after tissue injury ^[74][98]. Additionally, they are spatially related to VSMCs and contribute to regulating the vascular tone in different body districts, from the peripheral to the cardiac and cerebral microcirculation ^[74][75][98,99]. When their physiologic function is lost, they are responsible for derangement in the vascular wall homeostasis, contributing to an increased susceptibility to vasospasm, which is another determinant in the pathogenesis of AH ^[76][100]. However, direct evidence of their role and, in particular, their epigenetic signature in the context of microcirculation in AH needs to be adequately investigated by further studies.