The role of the mineralocorticoid receptor in immune cells in cardiovascular disease

Chronic low‐grade inflammation and immune cell activation are important mechanisms in the pathophysiology of cardiovascular disease (CVD). Therefore, targeted immunosuppression is a promising novel therapy to reduce cardiovascular risk. In this review, we identify the mineralocorticoid receptor (MR) on immune cells as a potential target to modulate inflammation. The MR is present in almost all cells of the cardiovascular system, including immune cells. Activation of the MRs in innate and adaptive immune cells induces inflammation which can contribute to CVD, by inducing endothelial dysfunction and hypertension. Moreover, it accelerates atherosclerotic plaque formation and destabilization and impairs tissue regeneration after ischaemic events. Identifying the molecular targets for these non‐renal actions of the MR provides promising novel cardiovascular drug targets for mineralocorticoid receptor antagonists (MRAs), which are currently mainly applied in hypertension and heart failure.


| INTRODUCTION
Cardiovascular disease (CVD) is the leading cause of loss of disabilityadjusted life years and deaths worldwide (Joseph et al., 2017). It accounts for approximately one third of all deaths globally, with CVDrelated morbidity having increased by over 10% in the last decade.
Importantly, the majority of cardiovascular events is not prevented by current therapeutic regimens, despite a growing number of pharmacological agents targeting classical cardiovascular risk factors, such as hyperlipidaemia and hypertension (Joseph et al., 2017). Not surprisingly, the impetus to further delineate the pathophysiological mechanisms driving CVD to identify novel targets for personalized therapies is high.
In the last two decades, research has changed our vision on the pathophysiology of CVD and highlighted its complex pathophysiology.
Perhaps the most novel change was that chronic low-grade inflammation was acknowledged as an important additional risk factor for CVD (Swirski & Nahrendorf, 2013). Inflammatory cells seem to affect, overall, the atherosclerotic burden and tissue regeneration. Antiinflammatory drugs hold promising value for treatment of CVD.
Canakinumab (a monoclonal antibody against IL-1β) and colchicine significantly reduced morbidity and mortality in high CV risk patients.
Other anti-inflammatory drugs failed to prevent or reduce the CVD burden (Ridker et al., 1997). General loss of immune function may underlie increased susceptibility to infection and needs further investigation, as well as identification of targets for CVD-specific modulation of inflammation.
One promising target is the mineralocorticoid receptor (MR) that primarily binds the mineralocorticoid aldosterone, but has an additional high affinity for the glucocorticoid cortisol and androgens. General MR activation is undisputedly associated with CVD. First, the autonomous adrenal overproduction of aldosterone (primary aldosteronism, PA) results in a two to three-fold increased risk of cardiovascular events (Monticone et al., 2018). Second, treatment of patients in heart failure with a MR antagonist (MRA) reduced CVD morbidity and mortality (Pitt et al., 1999(Pitt et al., , 2003. Moreover, in patients with established CVD but without PA, higher plasma aldosterone levels were associated with an increased risk of myocardial infarction, stroke, and cardiovascular mortality (Ivanes et al., 2012). Although the MR is mainly known by its classical effect-to promote renal sodium reabsorption in epithelial cells in the distal nephron after binding of aldosterone and to regulate blood pressure and salt homeostasis (Funder & Zennaro, 2017)-these effects appear in part to be independent of blood-pressure levels. Ongoing research showed the MR to be present on virtually all cells of the cardiovascular system, including endothelial cells, vascular smooth muscle cells (VSMCs), and cells of the adaptive as well as innate immune system (van den Berg et al., 2014). Activation of these extra-renal MRs seems to have additional detrimental effects on cardiovascular health (Bene et al., 2014;Faulkner & Belin de Chantemele, 2019;van den Berg et al., 2014;van der Heijden, Deinum, et al., 2018).
This review aims to summarize the effects of the MR on innate and adaptive immune cells on the process of atherosclerosis and CVD, in order to highlight promising novel drug targets to prevent CVD.

| THE MR AND ITS EXPRESSION IN THE CARDIOVASCULAR SYSTEM
The MR is a cytosolic receptor, which after ligand binding shuttles into the nucleus to bind to a DNA sequence known as the hormone response element (HRE) (Figure 1). Upon binding, the MR forms homodimers, or combines with the glucocorticoid receptor (GR) to form heterodimers-thus initiating different transcriptional pathways (Ong & Young, 2017;Savory et al., 2001). Next to a ligand-binding and DNAbinding domain, it contains an amino terminal domain, which interacts with cofactors to alter specificity of gene activation. MR activation can also result in rapid effects which suggest non-genomic routes. Moreover, mineralocorticoids may also function as a ligand of cell membraneassociated receptors, such as the G protein-coupled oestrogen receptor (GPER). These complex MR signalling cascades and mineralocorticoid-receptor interaction have been extensively reviewed elsewhere (Ong & Young, 2017;Ruhs et al., 2017), In the cardiovascular system, the MR is expressed by various cell types. In immune cells, the MR is expressed in monocytes and macrophages (innate immune cells), and most studies on MR-induced phenotypic modification of immune cells focus on this myeloid MR. However, the MR is also expressed and functional in dendritic cells (Bene et al., 2014), which link the innate and adaptive immune system, as well as in T-and B-cells (Armanini et al., 1988) (adaptive immune cells). MR expression in lysozyme M (LysM) positive cells of innate immune cells has been shown by PCR and Western blotting (Rickard et al., 2009).
Moreover, MR expression was found by PCR in myeloid as well as Tand B-cells. In addition, MR protein expression was shown in myeloid cells (Montes-Cobos et al., 2017). Also, MR expression in myeloid cells was found by PCR and Western blotting (Usher et al., 2010). The same was seen for the MR in dendritic cells (Herrada et al., 2010).
Expression of the MR in lymphocytes has been shown by radioreceptor assay (Armanini et al., 1988) others showed expression of the MR in T-cells by FACS analysis, PCR, and Western blotting . Therefore, there are convincing data that the MR is expressed in cells of the innate and adaptive immune system, although, to our knowledge, the presence of MRs in neutrophils has not been studied.
Apart from the kidney and immune cells, the MR is also expressed in several other cells of cardiovascular tissues. Experimental studies using cell type-specific gene targeting of the MR gene in mice have revealed the importance of this extra renal aldosterone signalling in cardiomyocytes, endothelial cells, and vascular smooth cells. As this is not the focus of the present manuscript, the reader is referred to recent reviews (Biwer et al., 2019;Lother et al., 2015;van den Berg et al., 2014). Knockout of the MR gene in fibroblasts was without effect in cardiac hypertrophy (Lother et al., 2011).
Recent data suggest that some of the sex differences found in CVD are related to the endothelial MR. An interaction of the MR with oestrogen receptors could be the underlying mechanisms, as reviewed previously, (2019). However, translation of these findings to clinical practice is still lacking. In the recent trial on the effect of finerenone on chronic kidney disease outcomes in Type 2 diabetes, 1/3 of participants were female. However, no significant gender difference was found for reduction of primary composite endpoint by finerenone (Bakris et al., 2020). More work is necessary to establish the role of MRs in gender differences in patients with CVD.

| EVOLUTIONARY CLUES THAT THE MR IS IMPORTANT FOR MORE THAN BLOOD PRESSURE ALONE
Generally, aldosterone is seen as the main physiological MR ligand.
Nonetheless, evolutionary clues suggest that the MR originated to serve different ligands: Cartilaginous and bony fish exhibit the MR but no mineralocorticoids. In these fish, cortisol is likely to be the main ligand of MRs, important in stress responses. Aldosterone was first discovered in lungfish, who developed millions of years later (Funder, 2017).
In mammals, the MR still has a similarly high affinity for the glucocorticoids (cortisol in humans and corticosterone in rodents) ( Figure 1) (Arriza et al., 1987). In the physiological state, the plasma concentration of cortisol is much higher than that of aldosterone. Therefore, mechanisms exist to exert specificity at the MR. The most important one is co-expression of the enzyme hydroxysteroid 11-β dehydrogenase 2 (11β-HSD2) that converts cortisol into cortisone which has negligible affinity for the MR. In the vasculature, 11β-HSD2 is expressed by vascular smooth muscle cells and endothelial cells (van den Berg et al., 2014). However, monocytes and macrophages do not express 11β-HSD2, while in lymphocytes, its presence has not F I G U R E 1 The effects of myeloid and lymphoid MR on the immune cell phenotype. Top: When MR signalling is intact, aldosterone and/or cortisol bind to the MR in the cytosol, after which the MR-ligand complex shuttles into the nucleus to bind to a hormone-response element (HRE). In monocytes/macrophages, this induces transcription of various pro-inflammatory cytokines and chemokines. In addition, the inflammasome is activated. ROS production increases, as does the production of certain matrix metalloproteinases (MMPs). Macrophages polarize towards a M1 or inflammatory phenotype. In Tcells, aldosterone and/or cortisol acting through the mineralocorticoid receptor interact with the transcription factors, NFAT1 and AP-1. This promotes production of IFN-γ. Bottom: Data from myeloid specific MR-KO models and MRA treatment show skewing towards an M2 or "antiinflammatory" phenotype with induction of specific M2-markers,such as IL-10, Arg1, and PPAR-γ and up-regulation of cholesterol efflux transporters been studied. Interestingly, even in cells that do not express 11β-HSD2, the MR is more sensitive to aldosterone than to cortisol. Multiple post-binding mechanisms orchestrate this ligand-specific functional outcome of MR-ligand binding, which results in significantly stronger MR transcriptional activation at lower concentrations when bound to aldosterone than cortisol (Arriza et al., 1987). These mechanisms, which are complex and incompletely understood, have been extensively reviewed elsewhere (Fuller et al., 2017). For cells of the immune system, it is yet to be investigated which of these mechanisms are of importance. However, regardless of the ligand, MR signalling is available for drug-targeting.
Moreover, it is broadly acknowledged that immune sensors are also detecting and rectifying deviations from cell homeostasis to maintain cell physiology. Blood pressure control and host defence are essential for evolutionary survival of mammals, and one might speculate that it is therefore not surprising that evolution incorporated immune cells as active participants in the regulation of blood pressure and cardiovascular homeostasis. Infection can cause hypotension via fluid loss during fever, tachypnoea, and diarrhoea and vascular hyperpermeability. Thus, the risk of hypotension related to inflammation might have favoured selection of mechanisms that link immune and MR activation to blood pressure increases for short-term survival benefits. Such an evolutionary force may explain why important antimicrobial effectors like monocytes/macrophages and lymphocytes could have direct hypertensive effects by promoting vasoconstriction or sodium retention (Wenzel et al., 2016).

| IMMUNE CELLS IN CARDIOVASCULAR DISEASE
Conceptual and technological innovations have importantly increased our understanding of CVD as a chronic inflammatory disorder (Swirski & Nahrendorf, 2013). Atherosclerosis is the pathophysiological process that underlies most CVD, and is driven by leukocytes, in particular monocytes and macrophages which are the most abundant cell types in the plaque, together with vascular smooth muscle cells (VSCM). Single-cell RNA sequencing in a murine low-density lipoprotein receptor deficient (LDLR À/À ) mouse model of atherosclerosis recently identified 30% of plaque leukocytes in early, and 50% in more mature plaques to be macrophages, while CD8 + T-cells accounted for another 20% (Cochain et al., 2018). Preventing monocyte influx into the arterial wall prevents the formation of atherosclerotic plaques (Moore et al., 2013) and progression of established lesions in preclinical atherosclerosis models (Inoue et al., 2002). Communication of these innate immune cells with cells of the adaptive immune system and local nonimmune cells (i.e., endothelial cells and vascular smooth muscle cells) is necessary to initiate and maintain a local pro-inflammatory environment that promotes expansion and destabilization of atherosclerotic lesions, while in the steady state, the interaction between these cell types is pivotal to maintain a healthy local vascular environment (Moore et al., 2018). This illustrates the remarkable plasticity of this system and its cells, where the balance can shift from homeostasis to disease in the wrong circumstances.
When pro-atherosclerotic factors, including disturbed shear stress and hypertension, activate endothelial cells, circulating monocytes are the first to adhere to leukocyte adhesion molecules that are expressed on these activated endothelial cells, to migrate into the intima. Their differentiation into macrophages that engulf oxidized lipid particles, such as oxidized LDL (oxLDL) via scavenger receptors, mainly CD36 and the scavenger receptor-A (SR-A, CD204) to form foam cells, further accelerates atherosclerotic plaque formation (Moore et al., 2013). Importantly, although present in lower numbers and studied to a lesser extent in atherosclerosis models than macrophages, for virtually every immune cell, a role in atherosclerosis has been suggested, either athero-promoting or athero-protective (Swirski & Nahrendorf, 2013).
In brief, dendritic cells reside in the aorta and affect atherogenesis, probably through interaction with T-cells. T-cells can exert diverse effects on the atherosclerotic process, being either athero-promoting (T helper 1 and T helper 17 subsets) or athero-protective (T helper 2 and regulatory T-cell subsets) (Swirski & Nahrendorf, 2013).
Neutrophils, amongst others, produce a range of chemotactic agents that attract monocytes, produce myeloperoxidase that oxidizes lipoproteins, and produce neutrophil extracellular traps (Silvestre-Roig et al., 2020).
Immune cells are important not only in the atherosclerotic process itself but also in determining one of the most important risk factors for its development, hypertension. Here, it is mainly the adaptive immune system that plays an important role. The first conclusive evidence for a role of the adaptive immune system in the pathogenesis of arterial hypertension was provided by Guzik et al., by showing that the increase in blood pressure caused by angiotensin II infusion was significantly blunted in mice lacking T-and B-cells (Guzik et al., 2007).
This finding was confirmed in several other laboratories and genetic models (Ji et al., 2014;Madhur et al., 2020;Mattson et al., 2013). Surprisingly, we did not observe resistance to angiotensin II in B6.
Rag1 À/À mice (that do not have mature T-and B-cells) (Seniuk et al., 2020). The Sandberg laboratory also reported that Jackson B6. Rag1 À/À mice lost their resistance to angiotensin II-induced hypertension (Ji et al., 2017). There is no doubt that lymphocytes play a role in hypertension. The negative data illustrate that there are important, but still unidentified confounders (Madhur et al., 2020;Rios et al., 2020).
Last, tissue regeneration after a CVD event is importantly regulated by immune cells (Swirski & Nahrendorf, 2013). Most studied is the role of leukocytes after myocardial infarction. After the onset of ischaemia, endothelial cells up-regulate adhesion molecules facilitating neutrophil influx in the injured heart, which accelerates local inflammation. This influx of inflammatory neutrophils quickly overwhelms the population of macrophages (Pinto et al., 2012) and dendritic cells (Choi et al., 2011) that reside in the healthy heart. Inflammatory monocytes are the other subclass of immune cells that infiltrate the injured heart in the early stages after myocardial infarction, further releasing pro-inflammatory mediators and inducing proteolysis that causes tissue destabilization. After several days, they are replaced by anti-inflammatory monocytes that support angiogenesis and extracellular matrix synthesis (Swirski & Nahrendorf, 2013). Monocytes also infiltrate the non-injured myocardium where they can exert either protective effects through induction of angiogenesis or destructive effects through mediators that induce cardiac dilatation and fibrosis (Swirski & Nahrendorf, 2013).
Although not a part of the cardiovascular system, the recent discovery of the increasing levels of MR expression during (white) adipose cell differentiation (Fu et al., 2005) is also particularly relevant to CVD. MR activation has been suggested to potentiate white adipose tissue inflammation, oxidative stress, fibrosis, and insulin resistance, although part of these effects are attributed to effects of MR-induced activation of myeloid cells embedded in adipose tissue, with proinflammatory macrophages promoting a local inflammatory environment and decreasing insulin sensitivity (Wada et al., 2017). Knockout of the MR in adipocytes in mice fed a high-fat diet decreased the expression of genes involved in adipogenesis (Ferguson et al., 2020).

| ANTI-INFLAMMATORY THERAPY REDUCES CARDIOVASCULAR DISEASE IN THOSE AT RISK
Although preclinical data on the importance of inflammation in CVD are plentiful, only recently, the first landmark trial showed that systemic anti-inflammatory treatment reduced cardiovascular morbidity and mortality in those at high risk. Canakinumab, an antibody directed against the macrophage-derived driver of inflammation, IL-1β, reduced cardiovascular events, by an additional 15%, when added to standard therapy in patients with a recent myocardial infarction, at the expense of increased fatal infections (Ridker et al., 2017). A "broad-spectrum" anti-inflammatory approach using low dose colchicine in those with coronary disease in different stages also gave promising results (Tardif et al., 2019). However, low-dose methotrexate was unable to reduce atherosclerotic cardiovascular events in a similar population (Ridker et al., 2019), illustrating that the mechanisms linking chronic inflammation to atherogenesis and adverse outcomes are only partly understood. Unravelling these is essential in the development of new drug therapies.

| IMMUNOMODULATION THROUGH THE MR OF INNATE IMMUNE CELLS AFFECTS CARDIOVASCULAR DISEASE
The studies discussed below describe the current insights in the inflammatory effects of the myeloid MR, as well as the MR of dendritic cells. The immunomodulatory effect of the monocyte/ macrophage MR is also summarized in Figure 1. Moreover, clinical and experimental data are summarized in Table 1.
Upon activation by various stimuli, monocytes differentiate into macrophages with a diverse spectrum of phenotypes. For practical reasons, these macrophage phenotypes are often divided into "classically activated" inflammatory M1-macrophages and "alternatively activated" or "anti-inflammatory" M2-macrophages. Importantly, these phenotypes are oversimplifying biology and should be considered as two extremes of an extensive spectrum (Nahrendorf & Swirski, 2016) but, in general, M1 macrophages are also considered pro-atherogenic, while pro-fibrotic characteristics might be present in both (Nahrendorf & Swirski, 2016).
Various murine models combine in vivo, ex vivo, and in vitro techniques to characterize the potential of MR signalling to modify the monocyte and macrophage phenotype. These models use either aldosterone infusion, DOCA (deoxycorticosterone acetate, a mineralocorticoid)/salt infusion, or sodium restriction (which results in an endogenous activation of the renin-angiotensin-aldosterone system [RAAS]) to induce MR (over)stimulation, or myeloid-specific MR knockout or MRAs to reduce MR signalling in vivo. Many of these models suffer from the lack of a hypertensive control group, and therefore, in theory, some of the observed changes might be secondary to hypertension and release of mediators after vascular stress rather than to MR stimulation per se. In knockout models, the cre/loxrecombination system allows the selective deletion of the MR from inflammatory cells, and the LysM promoter is widely used to drive cre expression in monocytes/macrophages. Of note, the use of the LysM promoter does not allow strict macrophage-specific gene inactivation, but in parallel targets dendritic cells, neutrophils, and monocytes.
Human data investigating the immunomodulatory effects of the MR are scarce and consist of in vitro studies (co)incubating monocytes and monocyte-derived macrophages with aldosterone or MRAs. Additionally, some studies report on ex vivo of monocytes obtained from patients with primary aldosteronism. Some studies of the effects of hyperaldosteronism on circulating markers of inflammation, such as hsCRP and circulating IL-6, largely show a heightened systemic inflammatory state as reflected in these biomarkers (Remde et al., 2016;Staermose et al., 2009;Tzamou et al., 2013) (although some disagree; Somloova et al., 2016;van der Heijden, Groh, et al., 2020). However, these studies are unable to answer the question which cell type and mechanisms are responsible for the induction van der HEIJDEN ET AL. Interestingly, circulating IL-1β is increased in mice after aldosterone infusion (Bruder-Nascimento et al., 2016). This is particularly relevant since the CANTOS trial showed improvement of cardiovascular morbidity and mortality in high risk patients treated with IL-1β targeted therapy, as mentioned previously (Ridker et al., 2017). IL-1β is a pro-inflammatory cytokine derived from macrophages, which is high in the inflammatory signalling cascade. Its production is dependent of the NLRP3 inflammasome, which activates caspase-1 to cleave pro-interleukin-1β (and pro-IL-18). Aldosterone/salt infusion in rats, and DOCA/salt infusion in mice, increased the expression of inflammasome components in the kidneys (Doi et al., 2014;Krishnan et al., 2016). Mice treated with a NLRP3 receptor inhibitor, or that are deficient in an adapter protein of NLRP3, are protected against the hypertension and renal inflammation seen after DOCA/salt infusion (Krishnan et al., 2016). Surprisingly, Anakinra (an IL-1 receptor antagonist) lowered blood pressure in mice treated with DOCA/salt, but it did not alter the renal immune cell infiltration. In mice, aldosterone increased caspase-1 activation, and NLRP3 and Il-1β gene expression in bone-marrow derived macrophages. This activation was prevented by NLRP3 knockout (Ling et al., 2017). In vivo, caspase-1 deficiency prevented vascular dysfunction, aldosterone induced vascular cell adhesion protein 1 (VCAM-1) and ICAM-1 expression and macrophage adherence to the arterial wall and vascular remodelling after aldosterone infusion. To further distinguish whether NLRP3 inflammasome activation of the vasculature or immune cells were responsible for these effects, they transplanted the bone marrow from NLRP3 À/À mice into wild-type mice. Aldosterone-induced changes in vascular reactivity were partly blocked in these mice, and some protection from aldosterone-induced vascular inflammation was seen. Overall, these findings suggest that NLRP3 inflammasome in myeloid cells contributes to vascular damage induced by aldosterone  (Keidar et al., 2004;Sonder et al., 2006;Sun et al., 2016). This, together with findings in other models reviewed elsewhere (Funder, 2013), suggests that the MRA can act as inverse agonist.

T A B L E 1 Effects of the MR on inflammation and immune cell phenotype
Inverse agonists have the ability to not only block a receptor for binding with a true agonist, but to exert a (pharmacological) response opposite to that of the agonist after binding. This has important consequences for their use. In the context of inflammation, this would imply that MRA treatment could exert anti-inflammatory effects, independent of aldosterone excess, and therefore be of interest to a varied patient population.
We recently showed that aldosterone is able to induce a long lasting pro-inflammatory effects in human monocyte-derived-macrophages, a phenomenon known as trained immunity, which was partly abolished by co-incubation with the MRA spironolactone. Trained immunity initially described the ability of human monocytes to build a long-term immunological memory upon stimulation with microbial products (Netea et al., 2016), driven by broad epigenetic and immunometabolic changes (van der Heijden, Noz, et al., 2018). The concept has been extended to include pro-atherogenic, sterile compounds such as oxLDL (Bekkering et al., 2014), lipoprotein (a) (van der Valk et al., 2016), and catecholamines (van der Heijden, Groh, et al., 2020). In vivo, trained immunity was confirmed in a murine model, using a Western-type diet (Christ et al., 2018), and it has been implicated in the low-grade, long-term inflammation that is characteristic of atherosclerosis (Christ et al., 2016;Leentjens et al., 2018).
Monocytes isolated from patients with PA showed heightened responsiveness ex vivo in some studies. We could not confirm these findings (van der Heijden, Smeets, et al., 2020), but instead showed that monocytes obtained from PA patients that were differentiated into macrophages ex vivo for a week expressed more TNFA and IL6

| THE MYELOID MR ACCELERATES ATHEROSCLEROSIS
Although data on the relationship between MR activation in myeloid cells and atherosclerotic burden in humans are lacking, surrogate markers for atherosclerosis, such as the carotid artery intima-media thickness (IMT), are increased in PA patients (Holaj et al., 2007;Strauch et al., 2006). These changes reverse after adrenalectomy, although no effect was seen of spironolactone treatment (Strauch et al., 2008). We recently showed that patients with PA have an increased signal from the large arteries using PET-CT with [ 18 F] deoxyglucose, compared to controls with essential hypertension, suggestive of macrophage infiltration of the arterial wall (van der Heijden, Smeets, et al., 2020).
The total number of plaque macrophages importantly contributes to plaque size and vulnerability. It reflects the balance between infiltration by circulating monocytes and local proliferation of resident macrophages and macrophage egress, apoptosis, and efferocytosis (the clearance of apoptotic macrophages by phagocytic leukocytes) (Swirski & Nahrendorf, 2013). In murine models, aldosterone increased (McGraw et al., 2013), and myeloid-specific MR-KO decreased, the macrophage numbers in the atherosclerotic plaques (Shen, Morgan, et al., 2016). Moreover, in inflammatory and NOdeficiency models of kidney and heart fibrosis, MRA treatment (Martin-Fernandez et al., 2016) and MR-KO (Usher et al., 2010) reduced macrophage infiltration, although this finding was not confirmed by all groups (Bienvenu et al., 2012). Macrophage infiltration in femoral arteries after wire-injury was significantly less in myeloidspecific MR-KO than in wild-type mice (Sun et al., 2016). Although it is known that aldosterone induces the expression of adhesion molecules on vascular endothelial cells, transwell experiments showed that the migratory ability of peritoneal macrophages from MR-KO mice was markedly decreased, indicating that intrinsic changes through the macrophage MR further contribute to migratory capacity. Also, the MR was shown to influence the proliferative potential of macrophages (Sun et al., 2016). Myeloid-specific MR deficiency was also shown to significantly decrease the number of apoptotic cells in atherosclerotic lesions and increase efferocytosis (Shen, Morgan, et al., 2016).
Foam cell formation-the ingestion of (partly modified) lipoproteins by plaque macrophages-also importantly adds to the expansion of the atherosclerotic plaque and formation of an instable necrotic core, prone to rupture.
After oxidation, the oxidized LDL (oxLDL) is mainly ingested in macrophages through the scavenger receptors CD36 and scavenger receptor class A (SR-A), while the ATP-binding cassette transporters A1 (ABCA1), G1 (ABCG1), and scavenger receptor BI (SR-BI) mediate cholesterol efflux. Up-regulating RAAS activity with a low sodium diet augmented cholesterol accumulation in circulating and peritoneal mice macrophages, an effect which was inhibited by eplerenone. Others also report a decreased oxLDL content of peritoneal mouse macrophages after eplerenone treatment (Keidar et al., 2003). In myeloidspecific MR-KO mice receiving a high-fat diet, foam cell formation in vitro and in vivo was decreased (Mattson et al., 2013). In this model, gene expression of cholesterol efflux receptors was up-regulated in myeloid specific MR-KO macrophages while expression of genes involved in oxLDL uptake was not significantly altered, indicating that the MR mainly influences cholesterol efflux pathways (Figure 1). This was also reported in a different myeloid-specific knockout model (Usher et al., 2010).
Several intrinsic factors determine the stability of the atherosclerotic plaque. Next to apoptosis of foam cells that leads to the formation of a necrotic core, plaque composition with a high number of inflammatory cells contributes to instability. Plaque instability is further caused by local production of matrix metalloproteinases (MMPs) that lead to degradation of the protective fibrous cap overlying a thrombogenic core (Kojima et al., 2014). MMP-1, MMP-2, MMP-9, and MMP-12, are all produced by macrophages (next to other cell types), and play a role in the pathogenesis of atherosclerosis and atherothrombosis (Ketelhuth & Back, 2011;Wagsater et al., 2011).
The role of the MR in determining MMP production has been little studied. MMP-2 protein levels were higher in purified kidneys from rats treated with aldosterone (Wenzel et al., 2016), but many MMP producing cell types could have contributed to this finding, since in several non-immune cells, the MR was suggested to induce metalloproteinase expression (Rude et al., 2005;Sun et al., 2016). In a murine model of cardiac tissue remodelling, intact MR signalling in macrophages was required to increase MMP-12 production above baseline levels ( Figure 1) (Shen, Chen, et al., 2016a). The aldosteroneto-renin ratio and circulating MMP-2 concentrations were correlated in patients with PA in vivo, but no control group with normal aldosterone values was included (Lim et al., 2016) and adrenalectomy did not reduce MMP-2 levels in PA patients (Hung et al., 2015). We did not The role of the myeloid MR has been studied in different models of hypertension. Knockout of the MR in monocytes and macrophages uniformly showed a decrease in vascular and cardiac fibrosis (Rickard van der HEIJDEN van der Heijden, Smeets, et al., 2020) (Figure 2), independent of the model of hypertension used. Cardiac hypertrophy and fibrosis induced by aortic constriction were also significantly attenuated in myeloid MR-deficient mice. Macrophage infiltration in the heart was inhibited, and the expression of inflammatory genes decreased in the deficient mice. In addition, aortic fibrosis and inflammation were also attenuated . However, the effects on blood pressure are conflicting. MR knockout mice showed less increase in blood pressure and cardiac fibrosis in compared to wildtype mice after 8 weeks of DOCA salt (Rickard et al., 2009;Shen, Morgan, et al., 2016). In contrast, no effect on blood pressure was found in MR knockout mice in a model of hypertension induced by administration of the NOS inhibitor, N G -nitro-L-arginine methyl ester (L-NAME) and salt. LysM cre MRflox mice paradoxically even displayed a higher systolic blood pressure than wild-type mice (Usher et al., 2010) in a model of hypertension mediated by administration of L-NAME, in combination with angiotensin II infusion and high salt diet, although vascular and cardiac fibrosis were decreased. In summary, these studies clearly demonstrate a role for the MR in macrophages in hypertensive cardiovascular remodelling. In the DOCA salt model of hypertension, the monocyte/macrophage MR also regulates blood pressure, while in models with NOS inhibition, the data are less clear.
Intriguingly, macrophages have been proposed as a factor in the control of salt sensitivity by modulating the lymphatic response in the skin (Machnik et al., 2009). The interstitium of the skin has emerged as important organ involved in maintaining body sodium balance. Skin monocytes and macrophages regulate non-osmotic storage of salt in the skin by up-regulating NFAT5 in T-cells and secreting vascular endothelial growth factor-C (VEGF-C). The latter increases lymph capillary density which acts as a fluid buffer and attenuates the blood pressure response to high salt. Failure of this macrophage-driven escape results in skin electrolyte overload and hypertension (Wenzel et al., 2021). No data are available on the role of the MR in skin myeloid cells and non-osmotic salt storage in the skin, but it is tempting to speculate that the blood pressure reducing effect of myeloid cell MR-KO, which occurs specifically in DOCA salt models, might occur through this pathway. In contrast, NO deficiency promotes hypertension independently of macrophage function and MR deficiency in these cells has therefore no effect on blood pressure. Importantly, the F I G U R E 2 Effect of MR signalling in monocytes/macrophages on cardiovascular outcome. MR signalling in monocytes/macrophages and lymphocytes affects blood pressure, hypertension-induced end organ damage and atherosclerosis as well as cardiac tissue remodelling after ischaemia. In models of hypertension, inactivation of the MR reduced blood pressure and cardiac fibrosis. In atherosclerosis, macrophage infiltration into the vascular wall is increased, both through intrinsic changes in the macrophage itself as well as through MR mediated effects on endothelial cells. The combination of increased foam cell formation, reduced efferocytosis and increased production of MMPs results in an unstable plaque phenotype. In the (post)ischaemic heart, intact myeloid MR signalling importantly contributes to cardiac fibrosis development of cardiac fibrosis in the models of NO inhibition and the protection by monocyte/macrophage MR deficiency are independent of blood pressure changes. Next to CD8 + T-cells, also Th17 might be phenotypically dependent on MR signals. Th17 differentiation in a rat DOCA model was suggested to be MR-dependent (Amador et al., 2014). Moreover, in this model, MRA inhibited DOCA-induced suppression of the Foxp3 transcription factor that controls regulatory T-cell (Treg) differentiation. This leads to the hypothesis that the MR could control the Th17/Treg balance, in the presence of mineralocorticoid abundance.
In addition, data from an earlier study by Herrada et al. (Herrada et al., 2010), suggest that both the enhancement of CD8 + T-cell activation and promotion of Th17-polarization by the MR are critically dependent on modulation of dendritic cell responses by aldosterone.
In addition, adoptive transfer of regulatory T-cells ameliorated vascular and renal effects of aldosterone infusion in mice (Kasal et al., 2012). DOCA treatment in vivo also decreased the abundance of regulatory T-cells in a MR-dependent manner (Amador et al., 2014). However, the cell type in which MR activation mediates these effects remains unclear.

| FUTURE PERSPECTIVES
Two decades ago, two hallmark trials showed that treatment with MRA substantially reduced morbidity and mortality in patients with heart failure. The observed effect was larger than would have been expected from the modest decrease in blood pressure or the diuretic effect (Pitt et al., 1999;Pitt et al., 2003). This underscores the presence of direct adverse cardiovascular effects of the MR, independent of blood pressure regulation, as discussed here. Targeting of the MR in a cell-specific manner could further improve the benefit to risk (side effect) ratio of intervention with MR signalling.
In humans, replication of the evidence from experimental animals that the MR in inflammatory cells causes CVD is a major challenge. A better and deeper understanding of the role of the MR in these cells van der HEIJDEN ET AL.
is important and clinically relevant for the identification of new therapeutic interventions. Clinical studies which could verify current experimental findings are necessary, but also difficult to undertake.
Inflammatory cells serve many different functions in host defence.
Accordingly, their non-selective inhibition, as a strategy to treat CVD disease, may result in severe and unwanted immunosuppressive effects. A better understanding of the exact cellular, temporal and spatial contributions of the MR in the immune system to CVD is needed. Only after this step has been taken, can the novel therapeutic approaches targeting the MR in immune cells to combat CVD be addressed in drug development programs. This will require an approach that integrates epidemiological data, data sets about genetic variations in patients and the functional probing of cytokines and immune cells, as well as their crosstalk (Wenzel et al., 2021).
Non-steroid MR antagonists, such as finerenone, have stronger anti-inflammatory effects and less effects on potassium levels, than the steroid MR antagonists. It is primarily thought that this is mediated by a differential cofactor recruitment and tissue distribution (Agarwal et al., 2021). Whether the stronger anti-inflammatory effects of the non-steroid MR antagonists are mediated by preferential inhibition of the MR in inflammatory cells is not known but needs to be further explored.
In the future, we expect nanotechnology to provide important therapeutic advances,by specifically targeting myeloid cells in CVD (Duivenvoorden et al., 2019). Next to targeted delivery of drugs to distinct cell populations, the ability to label nanoparticles in order to follow their migration to and accumulation in, tissues of interest is an important additional advantage in research. Although the use of nanotechnology is still seldom used in the field of cardiovascular diseases, over 20 nanotechnological approaches for the delivery of therapeutic modalities have been FDA-approved in various areas of medicine (Duivenvoorden et al., 2019), mainly in the field of oncology and infectious diseases. In the cardiovascular field, research into the application of nanotechnology includes trials investigating the targeting of PCSK9 with small interfering RNA (siRNA), for cholesterol lowering and angiotensinogen targeting for blood pressure reduction (Fitzgerald et al., 2014). Alternatively, nanoparticles could be used to deliver immunomodulating drugs (such as MRAs) to plaque macrophages, as has already been explored in a HDL nanoparticle model for the delivery of simvastatin (Duivenvoorden et al., 2014).
Again, when designing myeloid-cell targeting nanoparticles, the most important challenge is to ensure development of technologies with enough precision as not to affect the host defence system, in general.

| CONCLUSION
Thirty-five years after the cloning of the MR, there is clear evidence that it plays an important role in inflammatory CVD. Virtually, all cells in the cardiovascular system express the MR. Especially for myeloid cells, and to a lesser extent lymphoid cells, strong phenotypic changes are induced by MR signalling. Under the influence of increased MR signalling, monocytes and macrophage polarize towards an M1-like phenotype, with increased pro-inflammatory cytokine and chemokine production. Lymphocytic MR signalling promotes IFN-γ production and hypertension. Although preclinical data are plentiful, the lack of human data makes it difficult to translate these findings to clinical practice. As MR antagonists are readily available and the number of patients in which they are administered is growing, answering many of the questions that remain on the role of the MR on immune cells in hypertension, atherosclerosis and tissue remodelling seems within reach. The growing field of nanotechnology will help to target inflammation in a cell-and tissue-specific manner, and nanotechnological targeting of the myeloid and lymphoid MR is an interesting novel and topic for future investigations.

| Nomenclature of targets and ligands
Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, and are permanently archived in the Concise Guide to PHARMACOLOGY 2021/22 ( Alexander, Christopoulos, et al., 2021;Alexander, Cidlowski, et al., 2021;.

CONFLICT OF INTEREST
The authors declare no conflicts of interest.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article because no new data were created or analysed in this study.