The renin-angiotensin system: going beyond the classical paradigms

Abstract

Thirty years ago, a novel axis of the renin-angiotensin system (RAS) was unveiled by the discovery of angiotensin-(1−7) [ANG-(1−7)] generation in vivo. Later, angiotensin-converting enzyme 2 (ACE2) was shown to be the main mediator of this reaction, and Mas was found to be the receptor for the heptapeptide. The functional analysis of this novel axis of the RAS that followed its discovery revealed numerous protective actions in particular for cardiovascular diseases. In parallel, similar protective actions were also described for one of the two receptors of ANG II, the ANG II type 2 receptor (AT₂R), in contrast to the other, the ANG II type 1 receptor (AT₁R), which mediates deleterious actions of this peptide, e.g., in the setting of cardiovascular disease. Very recently, another branch of the RAS was discovered, based on angiotensin peptides in which the amino-terminal aspartate was replaced by alanine, the alatensins. Ala-ANG-(1−7) or alamandine was shown to interact with Mas-related G protein-coupled receptor D, and the first functional data indicated that this peptide also exerts protective effects in the cardiovascular system. This review summarizes the presentations given at the International Union of Physiological Sciences Congress in Rio de Janeiro, Brazil, in 2017, during the symposium entitled “The Renin-Angiotensin System: Going Beyond the Classical Paradigms,” in which the signaling and physiological actions of ANG-(1−7), ACE2, AT₂R, and alatensins were reported (with a focus on noncentral nervous system-related tissues) and the therapeutic opportunities based on these findings were discussed.

INTRODUCTION

The renin-angiotensin system (RAS) is one of the most potent cardiovascular regulators and an important target for therapeutic drugs. From 120 yr ago, when renin was discovered (148), until 30 yr ago, only one peptide was thought to be active in this system: angiotensin II (ANG II). It was shown to be produced by successive digestion of angiotensinogen by the enzymes renin and angiotensin-converting enzyme (ACE), with ANG I as an intermediate, and to interact with a receptor, now called the ANG II type 1 receptor (AT₁R). Physiologically, AT₁R activation by ANG II elicits vasoconstriction, water intake, and Na⁺ retention. Pathophysiologically, activation of ACE/ANG II/AT₁R signaling is associated with oxidative stress, hypertrophy, fibrosis, and inflammation. This view of the RAS began to change when, in 1988, Santos et al. (125) discovered the formation of ANG-(1−7) from ANG I by an ACE-independent pathway, and Schiavone et al. (131) showed the vasopressin-releasing activity of this heptapeptide in neural tissue explants. One year later, the first in vivo actions of ANG-(1−7) were described (16). At about the same time, a second receptor for ANG II [the ANG II type 2 receptor (AT₂R)] was found (21, 159). In the following 30 yr until now, the physiological functions of these two alternative pathways of the RAS have been extensively studied, and it was shown that ANG-(1−7) via its receptor, Mas (127), as well as ANG II via AT₂R counteract the classical RAS resulting in vasodilation, anti-inflammation, antifibrosis, and antiapoptosis, conferring beneficial effects in the settings of cardiovascular diseases (19, 128, 144). A central role in regulating the relative activities of the RAS arms in a tissue is played by the local levels of the enzyme ACE2, which transforms ANG II into ANG-(1−7) (31, 149). It was only a logical consequence of these findings that compounds were developed to activate the two protective arms of the RAS, which are undergoing clinical evaluation at the moment (5). However, the full complement of angiotensin peptides is still not completed, because Lautner et al. (74) and Jankowsky et al. (60) recently discovered a novel class, which we suggest to call “alatensins,” with an alanine at the amino-terminal position instead of the aspartate in the classical angiotensin peptides. Although Ala¹-ANG II interacts with the same receptors as ANG II, AT₁R, and AT₂R, Ala¹-ANG-(1−7) (alamandine) has its own receptor, Mas-related G protein-coupled receptor D (MrgD) (74).

This review will summarize novel findings on the physiological and pathophysiological actions of the new beneficial RAS components, which were presented at the symposium on “The Renin-Angiotensin System: Going Beyond the Classical Paradigms” at the International Union of Physiological Sciences 2017 Rhythms of Life Congress in Rio de Janeiro, Brazil.

ANG-(1−7) AS A CRITICAL MEDIATOR OF THE ACE2-RELATED PROTECTIVE AXIS OF THE RAS

ACE2 is a monocarboxypeptidase that cleaves away phenylalanine from ANG II, converting it to ANG-(1−7), and leucine from ANG I, converting it to ANG-(1−9) (Fig. 1A) (104, 109). Together, ACE2, ANG-(1−9), and ANG-(1−7) act as endogenous negative regulators of the RAS. ANG-(1−7) is a biologically active heptapeptide exerting cardioprotective effects in the settings of cardiovascular diseases by antagonizing maladaptive signaling attributed to ANG II (1, 4, 88, 97). ANG-(1−7) primarily acts via the endogenous receptor Mas because the Mas antagonist A779 blocks the majority of ANG-(1−7) effects (1, 124, 127, 145, 171). More importantly, ANG-(1−7) effects are not observed in Mas-deficient animals (53, 127, 128, 164). MrgD can also mediate some ANG-(1−7) actions (127, 147). The AT₂R has been linked to cardioprotective actions of ANG-(1−9) (36, 37, 94, 95). The cardioprotective effects of ACE2 are a combination of 1) the degradation of ANG I to ANG-(1−9), thereby limiting the availability of the substrate for ACE, and 2) the degradation of ANG II to cardioprotective ANG-(1−7). Therefore, loss of ACE2 activity results in loss of protection from maladaptive signaling of the ANG II/AT₁R axis, promoting progression of cardiovascular diseases, whereas increased ACE2 activity leads to activation of ACE2/ANG-(1−7) and ACE2/ANG-(1−9) axes mediating cardiovascular protection (Fig. 1B).

Fig. 1.

Enzymatic cascade involving the renin-angiotensin system, key receptor systems, and biological effects. A: renin-angiotensin system cascade showing the angiotensin peptide metabolic pathway. ANG I is cleaved by angiotensin-converting enzyme (ACE) to ANG II, which is metabolized by ACE2 to ANG-(1−7). ANG II binds to ANG II type 1 receptors (AT₁Rs) and ANG II type 2 receptors (AT₂Rs), whereas ANG-(1−7) binds to Mas receptors (MasRs) and opposes ANG II/AT₁R actions. B: decreased ACE2 shifts the balance in the renin-angiotensin system toward the ANG II/AT₁R axis, resulting in the progression of cardiovascular disease. Increased ACE2 shifts the balance to the ANG-(1−7)/Mas axis, leading to protection from cardiovascular disease. ADH, antidiuretic hormone; APA, aminopeptidase A; PCP, prolyl carboxypeptidase (also known as angiotensinase C). [Reproduced with permission from Ref. 109.]

Biochemistry and Regulation of ACE2

ACE2 is a type I transmembrane protein consisting of an extracellular amino-terminal domain containing the catalytic site and an intracellular carboxy-terminal tail (31, 149). ACE2 functions as a carboxymonopeptidase with the Arg²⁷³ residue being critical for substrate binding and enzymatic activity (49). As a carboxymonopeptidase, ACE2 degrades ANG II to generate the heptapeptide ANG-(1−7) and ANG I to generate the nonapeptide ANG-(1−9) by hydrolyzing the carboxy-terminal amino acid (31, 121, 149). ANG-(1−9) can be further converted to ANG-(1−7) by ACE. In the same manner, ACE2 can also cleave and partially inactivate pyr-apelin-13 and apelin-17 peptides (155). In addition to being an enzyme, ACE2 is a transmembrane protein that also acts as an amino acid transporter and is targeted by severe acute respiratory syndrome coronavirus (109). ACE2 is widely expressed in the cardiovascular system, kidneys, lungs, and brain to protect these organs from excessive ANG II signaling (32, 42, 54, 71, 99, 161). In the heart, both ACE2 and Mas are expressed in cardiomyocytes, cardiofibroblasts, and the coronary vasculature (58, 103, 108, 123, 124, 126). ACE2 activity is regulated by a disintegrin and metalloproteinase domain-containing protein 17 (ADAM-17; also known as TNF-α-converting enzyme) and apelin. ADAM-17 acts as a negative regulator of ACE2 tissue activity by proteolytically cleaving ACE2, resulting in the shedding of ACE2 into the interstitium, leading to decreased ACE2 activity in the tissue and elevated circulating ACE2 activity (103, 109). Deletion of this negative regulator (ADAM-17 or TNF-α-converting enzyme) protects the heart and vasculature in conditions of myocardial infarction, hypertension, pressure overload, and aortic aneurysm (34, 35, 43, 135, 136). Unlike ADAM-17, apelin regulates ACE2 expression. In mice, knockout of apelin is associated with lower ACE2 expression levels and exacerbated ANG II-induced cardiac hypertrophy and dysfunction (172). Treatment of apelin knockouts with pyr1-apelin-13 normalized ACE2 expression and counteracted ANG II-induced cardiac hypertrophy and dysfunction (172). Interestingly, ACE2 can also degrade and inactivate apelin peptides, creating a self-regulating negative feedback loop (155).

Generation and Metabolism of ANG-(1−7)

Recombinant human ACE2 (rhACE) produces both ANG-(1−7) and ANG-(1−9) (from ANG II and ANG I, respectively), whereas recombinant murine ACE2 generates predominantly ANG-(1−7) from ANG II (6, 169). Studies using rhACE2 and ACE2 purified from sheep tissues showed that ANG II is the preferred substrate for ACE2 (48, 107, 134, 149, 153, 162, 173). In rats, changes in ACE2 correlated with plasma ANG-(1−9) levels (93), whereas rhACE2 efficiently converted infused ANG II into ANG-(1−7) (80). In mice, rhACE2 counteracts detrimental ANG II signaling, and these therapeutic effects are mediated via ANG-(1−7) (102, 107, 173). Also, in human studies, rhACE2 clearly lowered plasma ANG II levels and increased plasma ANG-(1−7) levels (6, 109). In the heart, ANG-(1−7) is primarily generated by ACE2 (41, 173), whereas plasma ANG-(1−7) is primarily produced by neprilysin from ANG I because inhibition of neprilysin reduced ANG-(1−7) levels and increased ANG II levels (111). The degradation of ANG-(1−7) is governed predominantly by the amino-terminal catalytic domain of ACE, which degrades ANG-(1−7) to ANG-(1−5) (28). Inhibition of ACE increases plasma levels of ANG-(1−7) in both rodents and humans (3, 6, 20).

Role of ACE2 and ANG-(1−7) in Heart Failure

Heart failure (HF) is driven by activation of multiple neurohumoral and signaling pathways resulting in pathological hypertrophy and maladaptive ventricular remodeling. Activation of the RAS and increased ANG II levels play a pivotal role in adverse myocardial remodeling and disease progression (6, 85, 87) contributing to systolic and diastolic dysfunction in patients with HF (13, 57). Use of ACE inhibitors does not necessarily result in lower ANG II formation because ANG II levels can remain elevated in optimally treated patients with HF. In fact, ~50% of patients using current ACE inhibition therapies exhibit elevated levels of ANG II, which is probably the result of increased ANG I levels and activity of mast cell chymase (6, 62, 78, 113, 158). This ACE-independent ANG II production highlights a need for a better understanding of the contribution of various parts of the RAS to HF progression.

Our current understanding of the role of various parts of the RAS in the development of HF is shown in Fig. 1B. In models of HF with reduced ejection fraction, loss of ACE2 worsened the HF phenotype (67, 102). Under the conditions of elevated ANG II levels, loss of protective ACE2 activity worsens cardiac dysfunction, hypertrophy, and fibrosis, leading to greater diastolic dysfunction (2, 173). Supplying rhACE2 decreased plasma and myocardial ANG II levels and increased plasma ANG-(1−7) levels, resulting in attenuated pathological remodeling and corrected diastolic dysfunction (173). Even partial loss of ACE2 (in heterozygous female mice) is sufficient to enhance the susceptibility to heart disease, which is a clinically relevant observation because ACE2 is downregulated in human hearts with dilated cardiomyopathy (156). Increase in soluble ACE2, which reflects loss of tissue-bound ACE2, is associated with severity of HF and serves as an independent predictor of major adverse cardiac events (6, 33, 119). Because shedding of ACE2 is mediated via ADAM-17 (59, 103, 163), it suggests a possible involvement of ADAM-17 in the development of HF. Involvement of ADAM-17 in HF development has been shown primarily in relation to the TNF-α signaling pathway in humans (129, 130) and in mouse models of heart disease (34, 66). However, in light of the findings that ANG II induces ADAM-17-dependent cleavage of ACE2, it becomes clear that ANG II, ADAM-17, and ACE2 create a potentially dangerous positive feedback loop conducive to the development of HF (103). Conversely, loss of ANG-(1−7) actions exacerbated heart disease in an ACE2-deficient state (102), whereas inhibition of ANG-(1−7)/Mas signaling prevented rhACE2-mediated cardioprotection (107), confirming the importance of ANG-(1−7) contribution to the cardioprotective effects of ACE2. Supplying ANG-(1−7) directly was also cardioprotective in preclinical models of heart disease (81, 88, 118, 145). ANG-(1−7) suppressed cardiomyocyte growth and myocardial infarction-induced ventricular hypertrophy and decreased myocardial levels of proinflammatory cytokines, leading to a reduction in myocardial inflammation (118, 145). ANG-(1−7) activation of Mas results in the activation of phosphatidylinositol 3-kinase (PI3K)-Akt-endothelial nitric oxide (NO) synthase (eNOS), upregulation of MAPK phosphatase (84), inhibition of PKC-p38 MAPK and reactive oxygen species production (50, 102, 173), and suppression of collagen expression (reducing myocardial fibrosis) (29, 44, 124), thereby antagonizing pathological effects of ANG II (88, 109, 127). In addition to ANG-(1−7), ANG-(1−9), the product of ACE2 degradation of ANG I, is antihypertrophic, antifibrotic, and antihypertensive in models of hypertension and myocardial infarction (36, 37, 94, 95).

ACE2 and ANG-(1−7) also play important roles in vascular disease and hypertension (109). For example, rhACE2 pretreatment partially reversed ANG II-mediated hypertension because of decreased plasma ANG II and increased plasma ANG-(1−7) levels (80, 162), whereas loss of ACE2 exacerbates ANG II-mediated hypertension (108). Cyclodextrin-encapsulated ANG-(1−7) and Mas agonists (AVE0091 and CGEN856S) have demonstrated antihypertensive effects in preclinical models (146). These antihypertensive effects of ACE2/ANG-(1−7) suggest new possible therapeutic options against hypertensive heart disease.

Role of ACE2/ANG-(1−7) in Obesity-Associated Cardiomyopathy

Diabetes and obesity are major causes of cardiovascular morbidity and mortality worldwide and result in microvascular and macrovascular complications, including hypertension (174). Diabetic heart disease, linked to systolic and diastolic dysfunctions, and HF (39, 47, 70, 174) are associated with activation of the RAS (11, 12, 56, 90). Obesity, the major metabolic precursor to type 2 diabetes, is an independent risk factor for the development of HF with preserved ejection fraction (HFpEF) (68, 69, 98, 166). Emerging preclinical and clinical data strongly support a key pathogenic role for ACE2 and ANG-(1−7) in the progression of cardiovascular diseases (109, 132, 142). In a model of type 2 diabetes (db/db), ANG-(1−7) rescued diastolic dysfunction and reduced cardiac hypertrophy, fibrosis, lipotoxicity, and adipose inflammation (89, 100). Conversely, in the settings of high-fat diet-induced obesity, loss of ACE2 worsened heart disease because of increased epicardial adipose tissue inflammation, myocardial lipotoxicity, and worsened cardiac insulin resistance (105). ANG-(1−7) prevented these changes and rescued HFpEF in ACE2 knockout mice (105). These findings, coupled with the protective effects of ACE2 and ANG-(1−7) in the vasculature and adipose tissue, support inflammation and microvascular dysfunction as key mediators of HFpEF (110, 116). These results suggest that in patients with HFpEF, epicardial adipose tissue inflammation may be related to cardiac dysfunction and adverse remodeling (105, 106). In patients with obesity, administration of ANG-(1−7) improved insulin-stimulated endothelium-dependent vasodilation and blunted endothelin-1-dependent vasoconstrictor tone (132). Therefore, enhancing the ACE2 and ANG-(1−7) pathways represents a potential therapy for HFpEF, a condition with an adverse prognosis that lacks effective therapy.

ALATENSINS

In 2008, Jankowski and coworkers (60) described a novel octapeptide derived from ANG II, which they called ANG A. This peptide can be formed from ANG II by decarboxylation of the Asp¹ residue to Ala¹. The affinity of Ala¹-ANG II for AT₁Rs and AT₂Rs, as determined by displacement of ¹²⁵I-Sar¹-Ile⁸-ANG II, was similar to ANG II. However, the pressor effect induced by ANG A was smaller than ANG II, suggesting the involvement of other mechanisms on this effect. This observation led to the hypothesis that reduced pressor effect was because of the formation of an ANG-(1−7)-like peptide [Ala¹-ANG-(1−7)] from ANG A. In pursuing this hypothesis, Santos’ group identified and characterized the heptapeptide Ala¹-ANG-(1−7), which they called alamandine (74). Differing from ANG A, which stimulates the same receptors as its precursor, alamandine did not act through Mas. Actually, alamandine stimulates MrgD (74). This receptor was initially believed to be an IB4-nociceptive neuron-specific receptor (30, 76). However, there is growing evidence that MrgD is present in other tissues/cells, including cardiomyocytes and endothelial cells (51, 61, 96). It is not clear yet if the lower pressor effect of ANG A compared with ANG II is because of the formation of alamandine.

Although only two Ala¹-ANG-related peptides have been reported so far (i.e., ANG A and alamandine), we believe that this peptide family is larger than anticipated [probably containing Ala¹-ANG I, Ala¹-ANG-(1−9), Ala¹-ANG-(1−5), etc.]. To avoid confusion related to angiotensin peptides, we are introducing the term “alatensins” to refer to the Ala¹-ANG family. We are currently studying the physiological presence of these peptides using a LC-MS/MS platform with promising results that shall be published elsewhere soon.

Physiological Effects of Alamandine

Alamandine has been reported to reverse hyperhomocysteinemia-induced vascular dysfunction. Activation of PKA appears to be involved in this effect (117). In adipose tissue, an alamandine-induced reduction of leptin has been described through a mechanism involving Src/p38 MAPK (150). As previously described for ANG-(1−7), an inclusion compound of alamandine/2-hydroxypropyl (HP)-β-cyclodextrin reduced blood pressure in nonanesthetized spontaneously hypertensive rats. Likewise, as previously observed for ANG-(1−7) (83), oral administration of alamandine/HP-β-cyclodextrin attenuated the cardiac profibrotic effect of isoproterenol in Sprague-Dawley rats (50). More recently, a cardioprotective effect of alamandine has been described in an experimental model of sepsis (79). In keeping with these effects, it has been reported that knockdown of MrgD in mice leads to marked dilated cardiomyopathy (96). Centrally, the action of alamandine resembles those described for ANG‐(1−7) in the rostral ventrolateral medulla, caudal ventrolateral medulla, and hypothalamus (38, 74, 137, 140). These observations suggest that, as described for ANG II and ANG‐(1−7), alamandine may act as a neuronal excitatory molecule in the brain. However, striking differences between these two peptides are becoming progressively evident (61, 82). For instance, in the insular cortex, alamandine but not ANG‐(1−7) promotes excitatory cardiovascular effects, including increases in blood pressure and heart rate associated with increased renal sympathetic activity (82). Moreover, ANG-(1−7) [and ANG II/AT₂Rs (112, 168)] promotes NO release mainly by activating the PI3K/Akt/eNOS pathway (124), whereas alamandine leads to NO release by a mechanism not involving this pathway (Fig. 2). AMP-activated protein kinase (AMPK) appears to be the main target for alamandine-induced NO formation (61).

Fig. 2.

Signaling pathways for nitric oxide (NO) formation. ANG-(1−7)/Mas and ANG II/ANG II type 2 receptors (AT₂Rs) induce NO formation via phosphatidylinositol 3-kinase (PI3K)/Akt signaling, whereas alamandine/Mas-related G protein-coupled receptor D (MrgD) leads to NO formation through AMP-activated protein kinase (AMPK) activation.

It was well demonstrated by Jankowski and colleagues that ANG A binds and produces biological effects through interactions with AT₁Rs (60). This finding was confirmed by other authors (23, 167). However, the relationship of ANG A and AT₁Rs apparently is not as straightforward as it appears. The vasoconstrictor effect of ANG A is smaller than that of ANG II, despite the fact that the affinity of both peptides for AT₁Rs is remarkably similar (60). This could be because of the formation of alamandine, as we have previously hypothesized (74). However, in isolated cardiomyocytes, ANG A was essentially ineffective in promoting electrically stimulated Ca²⁺ influx (23). In the same preparation, ANG II was active. Accordingly, a 10-fold difference between ANG II and ANG A for intracellular Ca²⁺ release in vascular smooth muscle cells was observed by Jankowski et al. (60). These observations suggest that ANG A could be a biased agonist of AT₁Rs. This possibility has not been explored yet.

Novel Findings on Alamandine and AT₁R Interactions

Canta et al. (17) compared the effects of alamandine and ANG-(1−7) in nonanesthetized normotensive Sprague-Dawley rats. In contrast to ANG-(1−7), which has no effect on blood pressure, alamandine produced a dose-response U-shaped decrease in blood pressure. The maximal decrease in blood pressure was achieved by intravenous administration of 20 pg/animal. To test whether the U-shaped dose-response was because of stimulation of AT₁Rs, the experiments were repeated in losartan-treated rats. In this condition, a magnification of the hypotensive response was observed (Fig. 3). However, the U-shaped form was still evident. These results suggest that alamandine is more potent than ANG-(1−7) as a vasodilator and that AT₁Rs are involved in alamandine effects. A similar conclusion was drawn by Soltani Hekmat et al. (141) in pentobarbital-anesthetized rats. However, a more general conclusion about the modulatory role of AT₁Rs on the cardiovascular effects of alamandine cannot be drawn in other rat strains (Wistar and spontaneously hypertensive stroke-prone rats), because losartan abolished rather than increased alamandine vasorelaxant action (75). In addition, in the caudal ventrolateral medulla, blockade of AT₁Rs did not alter the hypotensive effect of alamandine (140).

Fig. 3.

Effect of alamandine on blood pressure in nonanesthetized 12-wk-old Sprague-Dawley male rats. Administration of drugs and blood pressure measurement were performed with a cannula inserted in the femoral vein and artery, respectively. Alamandine was administered intravenously in bolus in increasing amounts (0.02, 0.2, 1, 5, 20, and 80 ng). A subsequent alamandine amount was administrated only when mean arterial pressure (MAP) returned to the basal level. Losartan (5 mg/kg) was administrated intravenously in bolus 30 min before intravenous administration of alamandine. A U-shaped effect of alamandine was observed. *P < 0.05 vs. saline; #P < 0.05 vs. alamandine. Statistical significance was obtained by two-way ANOVA followed by a Bonferroni test. Each bar represents the mean ± SE.

NOVEL DOWNSTREAM PLAYERS OF ANG-(1−7)/MAS AND ALAMANDINE/MRGD SIGNALING NETWORKS IDENTIFIED BY PHOSPHOPROTEOMICS

Cell signaling is a communication process governing cellular actions and is mainly driven by reversible phosphorylation of downstream effectors. In the past decade, phosphoproteomics has emerged as a powerful approach to study phosphorylation dynamics. Figure 4 shows a general phosphoproteomic workflow to study cell signaling dynamics. This technology has been used to study signaling of ANG-(1−7)/Mas in human endothelial cells (152) and alamandine/MrgD signaling in Chinese hamster ovary MrgD and cancer cells (138).

Fig. 4.

Phosphoproteomic approach to study signaling pathways. A: activation or inhibition of a given receptor leads to phosphorylation and dephosphorylation of target proteins from the plasma membrane toward the nucleus in a time-dependent manner. B: general phosphoproteomic workflow for temporal cell signaling studies. To study phosphorylation and dephosphorylation dynamics, experimental groups are divided based on treatment duration (e.g., from seconds to days). Untreated cells are used as controls and are considered time = 0 min (i). Cells are lysed to extract proteins and phosphoproteins and then digested using specific proteases (e.g., trypsin) (ii). Generated peptides are labeled with nonradioactive isotopes for multiplexing analysis (iii). Phosphopeptides are enriched to reduce sample dynamic range (iv). Samples are analyzed using liquid chromatography coupled to mass spectrometry (LC-MS), and bioinformatics tools are used to identify and quantify phosphoproteins, including their phosphorylation sites, to ultimately build dynamic signaling networks affected by the given treatment (v).

ANG-(1−7)/Mas Signaling

Because ANG-(1−7) was identified as the endogenous ligand of Mas in 2003 (127), Western blot-based studies have contributed to build a solid knowledge of ANG-(1−7)/Mas signaling. For example, using this method, it was reported that ANG-(1−7)-induced NO production is dependent on the activation of the PI3K/Akt/eNOS pathway (124) and that ANG-(1−7) activates Src homology region 2 domain-containing phosphatase (SHP)-2 to counterregulate ANG II/AT₁R signaling (123).

Phosphoproteomics has been used to build a comprehensive time-resolved signaling network of ANG-(1−7) in human endothelial cells (152). ANG-(1−7) stimulation led to differential regulation of 121 phosphosites from 79 proteins, including dephosphorylation of Ser²⁵⁶ on the transcription factor FOXO1, leading to its activation. FOXO1 is an important regulator of tumor suppression and cell metabolism (46, 72). FOXO1 activation by ANG-(1−7) seems to be an important molecular event on the antitumoral effect attributed to this heptapeptide. One could not anticipate that ANG-(1−7) would induce FOXO1 activation before this phosphoproteomic study because Akt1, a crucial player of ANG-(1−7) signaling (124), induces FOXO1 phosphorylation and inhibition (15). Thus, this is a fine example of how phosphoproteomics can help one building an unbiased and comprehensive signaling network. Nevertheless, it is clear that other upstream players yet to be identified are responsible for FOXO1 dephosphorylation upon Mas activation.

ANG-(1−7)/Mas and Alamandine/MrgD Induce NO Formation Via Different Signaling Pathways

NO formation is a shared outcome of ANG-(1−7) and alamandine treatment. This event is associated with many beneficial effects observed for these peptides (45, 74). However, as mentioned above in alatensins, ANG-(1−7) and alamandine do not trigger a common pathway to activate NOS (Fig. 2). Although ANG-(1−7)/Mas induces NO production via the PI3K/Akt/eNOS pathway on endothelial cells and cardiomyocytes (29, 124), alamandine increases NO levels in a LKB1/AMPK-dependent manner in cardiomyocytes. Using ventricular cardiomyocytes isolated from hearts of 10- to 12-wk-old male C57BL/6 mice, de Jesus and collaborators (61) showed that alamandine induced phosphorylation of AMPK-α (Thr¹⁷²) and its upstream effector LKB1 (Ser⁴²⁸). Although ANG-(1−7) induced phosphorylation of Akt at its activation site (Ser⁴⁷³), the authors did not observe its phosphorylation using alamandine. Using cardiomyocytes from Mas-deficient mice, the authors ruled out the contribution of Mas in LKB1 and AMPK-α phosphorylation induced by alamandine (61).

Although alamandine and ANG-(1−7) have 86% sequence identity (six of seven amino acid residues are the same), the replacement of the negatively charged residue (Asp) to the neutral one (Ala) in position 1 changes the physicochemical feature of alamandine, influencing its receptor affinity as alamandine activates MrgD but not Mas (74). As reviewed here, by activating different receptors, alamandine and ANG-(1−7) induce the activation of different signaling pathways, adding new therapeutic clues and insights for the understanding of the RAS.

AT₂Rs: FROM ENIGMA TO THERAPEUTICS

In the late 1980s, the development of new angiotensin receptor ligands by some pharmaceutical companies resulted in the realization that some of these ligands, such as Dupont compound DUP753 (later losartan) or Ciba-Geigy compound CGP42112A, were able to distinguish between two different angiotensin receptor subtypes because of different affinity (21, 159). It soon became the consensus to name the receptor to which DUP753 bound with high affinity and CGP42112A with low affinity the ANG II type 1 receptor (AT₁R) and the receptor to which DUP753 bound with low affinity and CGP42112A with high affinity the ANG II type 2 receptor (AT₂R) (25).

Despite these new experimental tools for distinguishing between AT₁R and AT₂R, it took more than 5 more years until the scientific community began to really understand that AT₁Rs and AT₂Rs generally mediate opposing actions. There were several reasons as to why this process took so long. For example, 1) CGP42112A was initially regarded to be an antagonist, which led to wrong interpretation of experimental data (159), 2) for some time, several researchers regarded the AT₂R as binding site without function (139), and 3) as discussed in detail below, the G protein coupling of the AT₂R is quite unusual (170), which makes studies on its signaling and function difficult.

Signaling of the AT₂R and Cross-Talk With Mas

The mid-1990s brought a real breakthrough in the understanding of the AT₂R. Four groups independently discovered that activation of phosphatases seemed to be a major signaling mechanism of the AT₂R, whereas the AT₁R signals mainly through kinase-driven signaling cascades. Although Bottari et al. (9) found in 1992 that, in general, AT₂R activation leads to tyrosine dephosphorylation, in subsequent years, the groups of Sumners, Dzau, and Nahmias identified protein phosphatase 2 (PP2A) (63), MAPK phosphatase 1 (165), and SHP-1 (92) as specific AT₂R-stimulated phosphatases. At the same time, it was excluded that the AT₂R couples to “conventional” G proteins such as G_q or G_s proteins (10). However, coupling to G_i proteins was demonstrated and shown to be involved in the modulation of ion channel currents and activation of PP2A (55, 63). The lack of conventional G protein coupling of the AT₂R was elegantly explained by a Nature publication from 2017, which reported the crystalline structure of the AT₂R (170). Although this study confirmed that the AT₂R displays all characteristics of a 7-transmembrane, class A G protein-coupled receptor, the authors unexpectedly discovered that upon activation of the AT₂R, intracellular helix 8 changes its orientation in a way that it interacts with intracellular helixes III, V, and VI, thereby sterically blocking binding of conventional G proteins and β-arrestins.

Although these newest findings explain the lack of conventional G protein coupling of the AT₂R, the actual signaling mechanisms of the receptor are still only incompletely understood. This holds true, in particular, for the initiation of signaling upon receptor activation. What is known, however, is that the third intracellular loop and COOH-terminal end of the AT₂R seem crucial for AT₂R signaling (114, 115). In fact, some initial signaling molecules, such as SHP-1, PP2A, and AT₂R-interacting protein, interact directly with the AT₂R upon receptor activation: SHP-1 and PP2A (probably under involvement of G_i) with the third intracellular loop (64, 133) and AT₂R-interacting protein with the COOH-terminal end (8). Certain kinases may be involved, too, in this early initiation of signaling, such as tyrosine kinase c-Src (133).

From a functional perspective, the lack of conventional G protein signaling and the activation of G_i and phosphatases, which again interfere with kinase-driven signaling in an inhibitory way, make sense and are in accordance with known AT₂R actions, which oppose actions of cytokines, growth factors, and classical G protein-couple receptors such as the AT₁R (22, 24, 25, 122, 151, 157).

Signaling mechanisms of the AT₂R and Mas have many similarities like, e.g., involvement of SHP-1/SHP-2 and signaling through PI3K/Akt/eNOS (Fig. 2). Moreover, the AT₂R and Mas form heterodimers, at least in certain tissues, which may explain the shared signaling pathways and also the phenomenon that often effects of ANG-(1−7) can be inhibited by an AT₂R antagonist and effects of an AT₂R agonist by A779 (77, 101, 154). Interestingly, when the AT₂R and Mas dimerize, they seem to depend on each other functionally, because knockout of one of the receptors leads to loss of function of the other receptor in the respective cell or tissue (77).

Physiological and Pathophysiological Actions of the AT₂R

In the physiological situation, the AT₂R is usually expressed at low levels and in most tissues appears to be dormant (25). Exceptions seem to be a role in the central regulation of blood pressure (143), a weak vasodilation (160), a natriuretic effect (52), and an impact on cell differentiation, e.g., in neurons (86), uterus (26), or fetal tissue (18).

In the pathophysiological situation, the AT₂R mediates a variety of tissue protective actions, which again very much resemble the actions of ANG-(1−7) through Mas and which comprise, for example, anti-inflammation, immune modulation, antifibrosis, inhibition of sympathetic outflow, antiapoptosis, and neuroregeneration (91). In the context of HF, several of these actions work together in a well-orchestrated way. For example, in rats with HF caused by myocardial infarction, AT₂R stimulation acts in an anti-inflammatory manner by reducing cytokine synthesis and in an antifibrotic manny by inhibition of transforming growth factor-β generation, thus ameliorating peri-infarct remodeling, which again results in improved cardiac function (65, 73). An antifibrotic effect was also seen in the right ventricles of rats with pulmonary hypertension (14). Right ventricular fibrosis leading to HF is in fact a major pathomechanism determining mortality in patients with pulmonary hypertension. In another model of ischemia-induced HF (coronary ligation model), central administration of the AT₂R agonist C21 for 7 days by intracerebroventricular infusion significantly reduced sympathetic outflow and improved baroreceptor sensitivity, both mechanisms with a proven beneficial effect on HF (40).

The protective effects of AT₂R stimulation have been observed and associated with improved outcome in multiple other disease models. including cardiovascular disease, diabetic end-organ damage, autoimmune disease, neurological disease, etc. For more details on the protective actions of the AT₂R in a broad spectrum of diseases, the reader is referred to recent review articles (22, 24, 25, 27, 122, 151, 157).

Targeting the AT₂R in Drug Development

As a result of the better understanding of AT₂R actions and the realization that the receptor promotes tissue protection, repair, and regeneration in the context of several different pathologies, drug development projects have been initiated for the development of AT₂R agonists. In the meantime, some of these projects have reached the clinical phases of development. The nonpeptide AT₂R agonist compound 21 (C21), a proprietary molecule of Vicore Pharma (https://vicorepharma.com/), has successfully undergone phase I clinical testing and will soon be forwarded into a phase IIa clinical study in patients with idiopathic pulmonary fibrosis. The cyclic peptide MOR107 (Morphosys, https://www.morphosys.com/; previously LP-2 by Lanthio Pharma, https://www.lanthiopharma.com/) is currently being tested in a phase I clinical trial, whereas the clinical testing of MP-157 (Mitsubishi Tanabe, https://www.mt-pharma.co.jp/) has been discontinued. The AT₂R antagonist EMA401 (Novartis, https://www.novartisclinicaltrials.com/TrialConnectWeb/home.nov; previously Spinifex) has been successfully tested in a phase II trial for the treatment of neuropathic pain (120). More phase II studies with EMA401 are currently being initiated.

CONCLUSIONS AND FUTURE DIRECTIONS

The numerous beneficial actions of the novel RAS arms summarized in this review warrant clinical verification and therapeutic exploitation.

ACE2 has emerged as the dominant mechanism for negative regulation of the RAS by metabolizing ANG II into the beneficial peptide ANG-(1−7). This heptapeptide has emerged as a major protective peptide in the cardiovascular system. Clinical and experimental findings have demonstrated that ACE2 and ANG-(1−7) comprise the dominant mechanism for protective regulation of the RAS in many types of HF. ANG-(1−7) generated by ACE2-dependent conversion of ANG II is a crucial mediator of the cardioprotective effects, which makes ANG-(1−7) a promising therapy for HF.

After a long period of preclinical research to understand AT₂R signaling and function better, this knowledge is currently being translated into developments for a potential, future clinical use of drugs targeting the AT₂R. Most of these drugs in development are AT₂R agonists, with current primary indications being fibrotic diseases and diabetic nephropathy. The AT₂R antagonist EMA401 is being developed for the treatment of neuropathic pain. Phase II clinical studies are currently being initiated and will provide information about the therapeutic potential of drugs targeting the AT₂R.

The first compounds activating the Mas and AT₂ axis are available, and clinical trials have been started. For alatensins, the physiological and pathophysiological importance has still to be confirmed before therapeutic approaches can be initiated. However, it is quite likely that we will soon have novel therapies for cardiovascular and other diseases based on beneficial RAS peptides. Indeed, a recent study has described the beneficial effects of an oral formulation of hydroxypropyl-β-cyclodextrin/ANG-(1−7) in volunteers submitted to isometric overload muscle damage (7). This study, which is the first to test the effects of an oral formulation of ANG-(1−7) in humans, opens new possibilities for testing the actions of ANG-(1−7) and therapeutic effects in patients.

Finally, this review highlights the potential of phosphoproteomics to investigate ANG-(1−7) and alamandine signaling in an unbiased way, allowing the identification of unanticipated downstream effectors of these signaling networks.

GRANTS

This work was supported by the Canadian Institutes of Health Research and Heart and Stroke Foundation (to G. Oudit) and the Brazilian Funding Agencies CNPq (Grant 421021/2016-0, to T. Verano-Braga), FAPEMIG (Grant APQ-03139-16, to R. Santos, and Grant APQ-03242-16, to T. Verano-Braga), and CAPES (to R. Santos and T. Verano-Braga).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.A.S.S., G.Y.O., T.V.-B., and G.C. prepared figures; R.A.S.S., G.Y.O., T.V.-B., G.C., U.M.S., and M.B. drafted manuscript; R.A.S.S., G.Y.O., T.V.-B., G.C., U.M.S., and M.B. edited and revised manuscript; R.A.S.S., G.Y.O., T.V.-B., G.C., U.M.S., and M.B. approved final version of manuscript.