Chronic RAS (renin–angiotensin system) activation by both AngII (angiotensin II) and aldosterone leads to hypertension and perpetuates a cascade of pro-hypertrophic, pro-inflammatory, pro-thrombotic and atherogenic effects associated with cardiovascular damage. In 2000, a new pathway consisting of ACE2 (angiotensin-converting enzyme2), Ang-(1–9) [angiotensin-(1–9)], Ang-(1–7) [angiotensin-(1–7)] and the Mas receptor was discovered. Activation of this novel pathway stimulates vasodilation, anti-hypertrophy and anti-hyperplasia. For some time, studies have focused mainly on ACE2, Ang-(1–7) and the Mas receptor, and their biological properties that counterbalance the ACE/AngII/AT1R (angiotensin type 1 receptor) axis. No previous information about Ang-(1–9) suggested that this peptide had biological properties. However, recent data suggest that Ang-(1–9) protects the heart and blood vessels (and possibly the kidney) from adverse cardiovascular remodelling in patients with hypertension and/or heart failure. These beneficial effects are not modified by the Mas receptor antagonist A779 [an Ang-(1–7) receptor blocker], but they are abolished by the AT2R (angiotensin type 2 receptor) antagonist PD123319. Current information suggests that the beneficial effects of Ang-(1–9) are mediated via the AT2R. In the present review, we summarize the biological effects of the novel vasoactive peptide Ang-(1–9), providing new evidence of its cardiovascular-protective activity. We also discuss the potential mechanism by which this peptide prevents and ameliorates the cardiovascular damage induced by RAS activation.
RAS activation has a relevant role in hypertension and in the development of secondary target organ damage. Currently, pharmacological blockade of the RAS is a major tool in preventing and treating target organ damage. However, these drugs are not always effective in reducing target organ damage and, as a consequence, both residual CV and renal damage persist. Available evidence from experimental models indicates that Ang-(1–9) plasma levels decrease when hypertension or heart failure are established, and that Ang-(1–9) is a natural anti-hypertensive molecule. These observations may reflect a role for this peptide in long-term BP regulation, suggesting that Ang-(1–9) could be relevant for treating hypertension. Ang-(1–9) infusion also protects against or reverses cardiac and vascular remodelling in experimental hypertension, as well as in heart failure. Considering that Ang-(1–9) levels increase in plasma under ACE inhibition and AT1R-blocking conditions [8,9], it is plausible that the beneficial clinical effect of ACEIs or ARBs may also imply increased circulating Ang-(1–9). On the basis of the hypothesis that AT2R mediates Ang-(1–9) activity, activation of AT2R by Ang-(1–9) may represent a counter-regulatory effect on classical RAS activation. Interesting questions yet to be addressed include whether the anti-hypertensive action of Ang-(1–9) involves renal effects and if Ang-(1–9) protects against the development of chronic renal damage.
Moreover, in patients with hypertension, heart failure or kidney organ damage currently on RAS blocker therapy, Ang-(1–9) might have an additive effect when the AT1R is blocked and the AT2R remains free. In this sense, it is possible to speculate that the remodelling-protective mechanisms induced by Ang-(1–9) act mainly by AT1R antagonism, rather than by lowering AngII levels through ACE inhibition. The effects of even higher Ang-(1–9) levels on these clinical situations and the appropriate strategies to increase them should be investigated further.
The classical RAS (renin–angiotensin system) pathway is a major regulator of BP (blood pressure) and CV (cardiovascular) function [1,2]. In addition, the RAS has a role in vascular response to injury and inflammation . Chronic RAS activation by AngII (angiotensin II) or aldosterone leads to hypertension and perpetuates a cascade of pro-hypertrophic, pro-inflammatory, pro-thrombotic and atherogenic effects that trigger end-organ damage [1,2].
AngII, the major RAS effector molecule, binds mainly to the AT1R (angiotensin type 1 receptor), a G-protein-coupled receptor, which results in vasoconstriction, sodium reabsorption, proliferation and inflammation . However, AngII also binds to the AT2R (angiotensin type 2 receptor), an inhibitory G-protein-coupled receptor, which leads to vasodilation, diuresis, natriuresis and anti-inflammation .
The deleterious effects of AngII have stimulated the search for a natural counter-regulatory axis of the classical RAS. By 2000, a new pathway consisting of ACE2 (angiotensin-converting enzyme 2), Ang-(1–9) [angiotensin-(1–9)], Ang-(1–7) [angiotensin-(1–7)], and its Mas receptor was discovered. Activation of this new non-canonical RAS pathway induces vasodilation and prevents CV hypertrophy and hyperplasia. For some time, studies focused mainly on ACE2, Ang-(1–7), the Mas receptor, and their biological properties that counterbalance the ACE/AngII/AT1R axis .
Before 2006, the biological activity of Ang-(1–9) was unknown [6,7]. However, the experimental evidence accumulated since then suggests that Ang-(1–9) may decrease BP and protect the heart, blood vessels and possibly the kidney from adverse CV remodelling stimulated by hypertension or heart failure [8–10]. Thus Ang-(1–9) emerges as a novel vasoactive peptide that prevents and/or reduces hypertension and pathological CV remodelling and dysfunction [9,10].
In this regard, ACE2 activation seems to counterbalance the damage due to RAS activation by increasing Ang-(1–9) [9,10]. In the present review, the origin, degradation, potential mechanism of action and tissue effects of Ang-(1–9), as well as its clinical and therapeutic applications in CV diseases, are discussed.
METABOLISM OF ANG-(1–9)
Ang-(1–9) is present in healthy volunteers and in rat plasma in the range of 2 to 6 fmol/ml [9,11]. Higher concentrations of Ang-(1–9) than AngII are found in the kidney . Circulating levels of Ang-(1–9) are increased by some pathological conditions [i.e. early after MI (myocardial infarction)]  and in control animals treated with ACEIs (ACE inhibitors) or ARBs (AT1R blockers) [6,8,9]. Thus there is an alternate pathway of AngI (angiotensin I) metabolism by ACE2, and this pathway may be amplified in the presence of ACEIs and ARBs. Both circulating and tissue levels of Ang-(1–9) depend on its catabolism and anabolism.
Ang-(1–9) is a peptide with the sequence Asp-Arg-Val-Tyr-Ile-His-Pro-Phe-His, generated from AngI by several carboxypeptidase-type enzymes (Figure 1), including CxA (carboxypeptidase A), CpA (cathepsin A) [13,14], and ACE2 . The last one is probably least relevant, since the generation rate of Ang-(1–9) from AngII is relatively slow compared with the generation rate of Ang-(1–7) from AngII . In addition, the ACE2 inhibitor has no effect on Ang-(1–9) formation, whereas the CxA inhibitor benzylsuccinate abolishes the formation of Ang-(1–9) and increases AngI levels in cardiac membranes .
Simplified view of the pathways for Ang-(1–9) formation, receptors and conversion of Ang-(1–9)
In human heart tissue, the main products of AngI degradation are Ang-(1–9) and AngII . However, in platelets, the main metabolite of AngI is Ang-(1–9), not AngII, which possesses inhibitory activity against ACE [18,19].
Chen et al.  examined the metabolism of AngI, Ang-(1–9) and Ang-(1–7) in stable transfected CHO (Chinese-hamster ovary) cells expressing human ACE and human B2R [BK (bradykinin) B2 receptor] coupled to GFP. They found that Ang-(1–9) was hydrolysed 18 times slower than AngI and 30% slower than Ang-(1–7) , suggesting that Ang-(1–9) may have a longer half-life in plasma compared with other RAS peptides.
Ang-(1–9) can be cleaved to Ang-(1–7) by ACE or via the action of alternative enzymes including POP (prolyl endopeptidase), NEP (neutral endopeptidase) and TOP (thimet-oligopeptidase) (Figure 1) [21,22]. Recently, by using a computational system biology approach applied to peptidomic data and in vitro experiments, Schwacke et al.  observed that Ang-(1–9) is converted into Ang-(2–9) by APA (aminopeptidase A) in glomerular podocytes (Figure 1) . This step might be important as an alternative pathway for Ang-(1–9) degradation, in addition to Ang-(1–9) degradation to Ang-(1–7) by ACE. Therefore, although the relative contribution of the carboxypeptidases has not been explored directly, increased steady-state levels of Ang-(1–9) could be due to decreased catabolism by ACE or other peptidases and/or by increased production of Ang-(1–9) by ACE2 as a result of increased AngI substrate availability.
MECHANISMS OF ACTION OF ANG-(1–9)
The information available to date indicates that the biological actions of Ang-(1–9) may be mediated by both direct and/or indirect mechanisms.
Direct effects of Ang-(1–9)
The principal studies regarding potential receptors for Ang-(1–9) relied mainly on a pharmacological approach [9,10,24,25]. The first evidence that Ang-(1–9) can bind and activate the AT2R was provided by Flores-Munoz et al. , using radioligand-binding assays in the H9c2 cell line and adult rabbit cardiomyocytes expressing AT1R or AT2R. Although the data from this study showed ~100-fold lower affinity of the AT2R for Ang-(1–9) than for AngII, Ang-(1–9) was able to compete with radiolabelled AngII with pKi values of 6.6±0.3 and 6.3±0.1 at the AT1R and AT2R respectively . Interestingly, the presence of Ang-(1–9) in the culture medium prevented AngII- or vasopressin-induced cardiomyocyte hypertrophy. This action was blocked by PD123319, implicating the AT2R . Furthermore, in SHRSPs [stroke-prone SHRs (spontaneously hypertensive rats] , with already-established experimental hypertension due to AngII, and in the rat Goldblatt model (2-kidney, 1-clip) , chronic Ang-(1–9) administration significantly reduced hypertensive CV damage. Co-administration of Ang-(1–9) with the Mas receptor blocker A779 did not affect the ability of Ang-(1–9) to reverse this damage. However, co-administration of Ang-(1–9) with the AT2R blocker PD123319 thwarted the beneficial effects of Ang-(1–9). These findings support the hypothesis that Ang-(1–9) could act directly on the AT2R.
Although AT1R mediates most of the recognized detrimental effects of AngII, AT2R stimulation partly counteracts the effects mediated by the AT1R . The AT2R is expressed in adult tissues in lower amounts than the AT1R . The effects and cell signalling mechanisms of AT2R activation are less well-characterized than those of the AT1R [27,28]. Current knowledge suggests that AT2R stimulation mediates vasodilatory, pro-apoptotic and anti-inflammatory effects, and inhibits cell growth [29,30]. Hence the AT2R can modulate CV remodelling, as well as progression of atherosclerosis, and the binding of Ang-(1–9) to the AT2R may counterbalance the AT1R-mediated effects on BP and target organ damage in hypertension (and possibly other CV diseases).
Recently, to test whether Ang-(1–9) has a direct effect on the AT2R, Ocaranza et al.  used aortic rings with intact endothelium pre-contracted with adrenaline (epinephrine; 1 mM). Ang-(1–9) caused a dose-dependent vasodilation with a maximal effect of 33±3% relaxation compared with control and an EC50 value of 22 pM, a concentration similar to the Ang-(1–9) levels found in the plasma of hypertensive rats infused with Ang-(1–9). The vasodilation induced by Ang-(1–9) was prevented by the AT2R pharmacological blocker and was resistant to the Mas receptor and AT1R blocker, which is consistent with the hypothesis of Ang-(1–9) action via the AT2R. Ang-(1–9)-induced vasodilation was also prevented by endothelium ablation and by eNOS (endothelial NO synthase) blockade with L-NAME (NG-nitro-L-arginine methyl ester) , a result that is consistent with the AT2R-dependent mechanism, considering that the stimulation of AT2R activates the NO/cGMP-dependent pathway through BKs and/or increased eNOS activity . The available studies examining the role of the AT2R in mediating the effects of Ang-(1–9) have relied mainly on the use of the AT2R antagonist PD123319. However, this inhibitor could have ‘a great deal of non-specificity’ . It is well known that the AT2R is the predominant angiotensin receptor subtype in the fetal and early post-natal periods, and that the absence of the AT2R negatively affects cardiac and renal maturation and growth [33–35], limiting the use of AT2R-KO (knockout) mice. Thus studies with tissue-specific and inducible AT2R KO in mice will help to clarify the role of the AT2R in the beneficial actions of Ang-(1–9).
Ang-(1–9) may induce NO release and/or enhance B2R activity. The incubation of CHO cells with Ang-(1–9) potentiated the release of arachidonic acid by [Hyp3Tyr(Me)8]-BK, elevated [Ca2+]i (intracellular [Ca2+]) and resensitized the B2R desensitized by BKs . Moreover, Jackman et al.  showed that Ang-(1–9) was significantly more active than Ang-(1–7) in CHO cells and in human pulmonary endothelial cells, enhancing the effect of an ACE-resistant BK analogue on the B2R, probably by combining with the active site of ACE, inducing a conformational change in a heterodimer complex formed by ACE and the B2R. In addition, Ang-(1–9) increased the release of arachidonic acid and NO by kinins . More recent findings show that Ang-(1–9) stimulates the release of NO in the rat aorta , as discussed below.
In transfected CHO cells expressing human ACE and the human B2R, Ang-(1–9) resensitized the desensitized GFP–B2R, an effect that was independent of ACE inhibition . This was reflected by release of arachidonic acid through a mechanism involving the induction of conformational changes in the ACE–B2R complex and certainly not through blocking potential inactivation of the ACE-resistant BKs . According to these authors, Ang-(1–9) could act as an endogenous allosteric modifier of the ACE–B2R complex . On the other hand, activation of the AT2R counteracts some of the effect of AngII, possibly via the B2R . If Ang-(1–9) resensitizes the B2R, Ang-(1–9) could trigger AT2R signalling, leading to PLA2 (phospholipase A2) activation and arachidonic acid release without affecting basal BK levels. By this mechanism, Ang-(1–9) may also enhance the therapeutic usefulness of ACEIs.
Some studies suggest that Ang-(1–9) may be an endogenous ACEI. Donoghue et al.  proposed that Ang-(1–9) is a competitive inhibitor of ACE because it is an ACE substrate. Under conditions of ACE inhibition, such as after long-term ACEI administration in rats, Ang-(1–9) levels increase significantly in plasma and the kidney [6,7]. Consistent with this hypothesis, we observed that the infusion of Ang-(1–9) prevented MI- and hypertension-induced increases in AngII plasma levels and LV (left ventricular) ACE activity [8–10]. Therefore it is possible that, when an ACEI blocks the hydrolysis of AngI to AngII, other enzymes such as ACE2 may still release AngI metabolites such as Ang-(1–9), enhancing the efficacy of the ACEI. To corroborate this hypothesis, we evaluated ACE activity and AngII levels in MI rats treated with vehicle or Ang-(1–9) at picomolar levels. Our results showed that chronic administration of Ang-(1–9) to MI rats significantly decreased plasma and LV ACE activity and plasma AngII levels .
Recently, neurolysin (EC184.108.40.206), a neurotensin-degrading neutral metalloendopeptidase, has been identified as a non-AT1R/non-AT2R angiotensin-binding site, thereby widening its putative function to the RAS . Neurolysin exists primarily as a cytosolic protein; however, depending on the cell type, it can also be secreted, bound to the plasma membrane or targeted to mitochondria [39–42]. Competitive binding assays have shown that neurolysin binds Ang-(1–9) >50% at 10 μM (Figure 1), and exhibits 59±3 and 75±2% inhibition of [125I]-Sar1,Ile8-AngII binding and IC50 values of 7.0 and 3.3 μM for brain and testis respectively . However, neurolysin also metabolizes AngII and AngI [44–46]. The products of AngII metabolism are the inactive (1–4) and (5–8) fragments of AngII [45,46], whereas the product of AngI metabolism is Ang-(1–7) [47,48]. By degrading AngII as well as forming Ang-(1–7), neurolysin can antagonize the well-known pathophysiological effects of AngII mediated by the AT1R . Thus neurolysin may play an important role in regulating the RAS.
Neurolysin may be a key enzyme responsible for metabolic processing of angiotensins in certain pathophysiological conditions, with perhaps even more importance than other peptidases. This notion is supported by recent observations indicating decreased density of this angiotensin-binding site in brain CV centres of SHRs . Thus the binding of Ang-(1–9) to neurolysin (Figure 2) may serve to counter-regulate the action of AngII binding to the AT1R by activation of intracellular signalling pathways by membrane-bound neurolysin in response to Ang-(1–9), similar to other peptidases . Future studies focusing on metabolic processing of Ang-(1–9) or molecular mechanisms activated by neurolysin will provide further insights into the role of this peptidase in relation to the function of the RAS.
Proposed signalling pathways for Ang-(1–9) and therapeutic targets
Indirect effects of Ang-(1–9)
Recently, Cha et al.  showed that Ang-(1–9) increased ANP (atrial natriuretic peptide) secretion and its plasma concentration without varying atrial contractility. At a concentration of 3 μM, Ang-(1–9) increased Ang-(1–9)-induced ANP secretion from 5 to 60% during the low-stretch state of the atrium. Secretion was increased more in the high-stretch than the low-stretch atrial state, and adding 1 μM Ang-(1–9) increased Ang-(1–9)-induced ANP secretion. This stimulatory effect of Ang-(1–9) on ANP secretion was attenuated by pre-treatment with an AT2R antagonist, but not by AT1R or Mas receptor antagonists. In addition, pre-treatment with inhibitors of PI3K (phosphoinositide 3-kinase), Akt, eNOS and sGC (soluble guanylate cyclase) blocked Ang-(1–9)-induced ANP secretion. The acute infusion of Ang-(1–9) in rats increased plasma ANP levels without altering arterial BP . This effect was attenuated by pre-treatment with an AT2R antagonist, but not by a Mas receptor antagonist or losartan . These results suggest that Ang-(1–9) may induce cardiac ANP release by activating the AT2R/PI3K/Akt/eNOS/sGC signalling pathway (Figure 2). The relevance of the increased levels of ANP and NO may be related to the beneficial effects of Ang-(1–9) on the CV system. Further in vitro and in vivo studies are needed to determine and characterize the downstream effectors of Ang-(1–9).
To date, all of the available information suggests that the main beneficial CV effects of Ang-(1–9) are direct rather than via transformation into Ang-(1–7). However, in vitro evidence shows that incubating Ang-(1–9) with ACE generates Ang-(1–7) , and that Ang-(1–7) negatively regulates CV damage [53,54]. Therefore it remains unclear in vivo whether Ang-(1–9) is only active after conversion into Ang-(1–7).
Ocaranza et al.  used the Mas receptor blocker A779 to investigate whether Ang-(1–7) could mediate the anti-hypertrophic effects of Ang-(1–9) in an MI model in rats. Although A779 increased circulating Ang-(1–7) levels by 2.7-fold, this compound did not modify the Ang-(1–9)-dependent suppression of cardiomyocyte hypertrophy induced by MI . In vitro experiments with cultured cardiomyocytes treated with noradrenaline (norepinephrine; 10 μM) or with IGF-1 (insulin-like growth factor-1; 10 nM) showed that Ang-(1–9) also prevented hypertrophy and that this effect was not modified by co-incubation with A779 . In agreement with these findings, Flores-Munoz et al.  observed in both the rat H9c2 cell line and adult rabbit cardiomyocytes that Ang-(1–9) prevented AngII- and vasopressor-triggered hypertrophy. However, this effect was not inhibited by the ACEI captopril nor by A779, supporting previous evidence that Ang-(1–9) acts independently of Ang-(1–7).
Moreover, in three experimental hypertension models, chronic administration of Ang-(1–9) significantly reduced hypertensive CV damage [10,25]. Co-administration of Ang-(1–9) with A779 did not affect the ability of Ang-(1–9) to reverse this damage, supporting the hypothesis that Ang-(1–9) may act independently of Ang-(1–7).
FUNCTIONAL ASPECTS OF ANG-(1–9) IN THE CV AND RENAL SYSTEMS
Several studies in the literature have shown that Ang-(1–9) has beneficial effects on structural and functional CV remodelling.
The first observations on the role of the ACE2/Ang-(1–9) axis showed that circulating and LV enzymatic activities of ACE2 were down-regulated in the long-term phase of LV dysfunction due to experimental MI in rats. At 8 weeks post-MI, AngII levels remained higher, but circulating Ang-(1–9) levels were lower, than in control rats. The ACEI enalapril prevented cardiac hypertrophy and dysfunction, as well as the changes in LV ACE2 levels . Plasma Ang-(1–9) levels were significantly increased when MI rats or sham-operated rats were treated with enalapril for 8 weeks. However, circulating Ang-(1–7) levels were not modified by enalapril at that time . On the basis of these findings, we proposed that Ang-(1–9) rather than Ang-(1–7) may act as a counter-regulator of AngII in experimental heart failure .
Regarding the biological properties of Ang-(1–9), our studies have shown that Ang-(1–9) regulates cardiomyocyte hypertrophy both in vivo and in vitro . In rats with MI randomized to receive either vehicle or drugs for 8 weeks, both enalapril and candesartan prevented LVH (LV hypertrophy) and also increased Ang-(1–9) plasma levels several-fold . In those experiments, Ang-(1–9) levels correlated inversely with different LVH markers, with or without adjustment for reduced BP . This effect was specific, since Ang-(1–7), AngII nor BK levels correlated with LVH. Chronic administration of Ang-(1–9) prevented cardiomyocyte hypertrophy , and A779 did not modify the Ang-(1–9)-dependent suppression of cardiomyocyte hypertrophy induced by MI . Ang-(1–9) also prevented the hypertrophy of cultured cardiomyocytes incubated with noradrenaline or IGF-1. Consistent with these results, the anti-hypertrophic effect of Ang-(1–9) was not modified by co-incubation with A779 .
Recently, Ocaranza et al.  have shown that chronic Ang-(1–9) administration reversed experimental hypertensive CV damage. In rats infused with AngII for 2 weeks or in rats with Goldblatt unilateral renal artery clipping (2-kidney, 1-clip) for 4 weeks, Ang-(1–9) infusion for 2 weeks normalized or ameliorated high BP . Ang-(1–9) plasma levels at the end of the experimental periods (4 weeks in AngII-infused rats or 6 weeks in Goldblatt rats) were approximately 40% lower compared with the respective control values in both models, whereas infusion of Ang-(1–9) increased circulating Ang-(1–9) levels 2-fold compared with control rats. In addition, we observed that Ang-(1–9) infusion blunted the increase in LVH, the increase in cardiomyocyte area and perimeter, cardiac fibrosis, and the changes in LV systolic function in both hypertensive models. All of these data show that increasing plasma Ang-(1–9) levels reverses cardiac damage in a setting of high AngII and already-established hypertension. Consistent with the hypothesis that the AT2R is the main receptor mediating the effects of Ang-(1–9), PD123319 abolished the beneficial effects of Ang-(1–9) in both AngII-infused and Goldblatt rats, but co-administration of A779 with Ang-(1–9) did not decrease the effectiveness of Ang-(1–9) .
In agreement with these findings, Flores-Muñoz et al.  observed that Ang-(1–9) infusion reduced cardiac fibrosis and collagen I expression in the SHRSP . This anti-fibrotic action of Ang-(1–9) was blocked by PD123319 co-infusion, which is consistent with the role of AT2R stimulation in mediating the effects of this peptide . Ang-(1–9) also inhibited fibroblast proliferation in vitro in a PD123319-sensitive manner . These results show that AT2R activation by Ang-(1–9) has an important cardiac anti-fibrotic action that may indicate a direct effect on cardiac fibroblasts.
Collectively, these data support the idea that chronic AT2R stimulation by Ang-(1–9) is potentially a good approach for reducing CV disease. The mechanism of cardiac hypertrophy/fibrosis reversion by Ang-(1–9) in MI rats  and in hypertensive rats [10,25] may reflect both direct AT2R stimulation and/or the beneficial effect of afterload reduction by reducing BP. In the SHR model, cardiac AT2R overexpression attenuated cardiac hypertrophy despite elevated BP .
Regarding the role of Ang-(1–9) in vascular damage, Kramkowski et al.  showed that Ang-(1–9) enhanced thrombosis in electrically injured arteries of rats, an effect that was abolished by the AT1R antagonist losartan. Ang-(1–9) also enhanced platelet aggregation ex vivo and in vitro. Recently, the same group showed in a rat experimental model that Ang-(1–9) increased thrombosis development, decreased plasma concentration of tissue plasminogen activator and increased the level of its inhibitor [PAI-1 (plasminogen-activator inhibitor-1)] . The action of Ang-(1–9) was reversed by the selective AT1R antagonist losartan, but not by the Ang-(1–7) antagonist A779. Ang-(1–9) did not bind to the AT1R. All of these effects seem to be mediated by the AT1R and by the conversion of Ang-(1–9) into AngII. However, these studies have several points that should be clarified. First, the pro-thrombotic effect of Ang-(1–9) was much weaker than the pro-thrombotic action of AngII [58,59]. Secondly, Ang-(1–9) caused only a modest increase in platelet aggregation in vitro . Thus the potential pro-thrombotic action of Ang-(1–9) needs further evaluation.
The observations concerning the beneficial effects of Ang-(1–9) on vascular remodelling have emerged from studies of pharmacological inhibition of the RhoA/Rho-kinase pathway with fasudil . Inhibition of this pathway reduced the up-regulation of genes that promote vascular remodelling [TGF-β1 (transforming growth factor-β1), PAI-1, and MCP-1 (monocyte chemoattractant protein-1)] in DOCA (deoxycorticosterone acetate)-salt rats. Fasudil-treated animals also had increased ACE2 activity and higher plasma Ang-(1–9) levels. Interestingly, the increase in ACE2/ACE and Ang-(1–9) levels were present only during fasudil treatment, both in sham and DOCA-salt hypertensive rats . This novel effect of RhoA/Rho-kinase inhibition might contribute to the benefits of fasudil in hypertension, atherosclerosis, and CV and renal pathological remodelling. Thus vascular remodelling may be more dependent on the tissue ACE2/Ang-(1–9) axis than on Ang-(1–7) levels in normotensive, as well as in hypertensive, rats, and possibly in clinical hypertension. Supporting this hypothesis, chronic administration of Ang-(1–9) in hypertensive rats prevented and reduced vascular remodelling . In already-established experimental hypertension, chronic administration of Ang-(1–9) for 2 weeks significantly decreased thoracic aortic thickening, collagen and TGF-β1 protein content, and protected the endothelium of hypertensive rats . These effects were not modified by the Mas receptor antagonist, but were abolished in the presence of PD123319 .
These protective effects of Ang-(1–9) in the CV system suggest that this vasoactive peptide targets CV cells, i.e. cardiomyocytes, fibroblasts, endothelial cells and vascular smooth muscle cells, inducing beneficial effects that could be mediated by the AT2R and the AT2R-dependent cell signalling cascade.
Several lines of evidence indicate that Ang-(1–9) has anti-hypertensive properties. Studies in DOCA-salt hypertensive rats and in normotensive control animals showed that treatment with the Rho kinase inhibitor fasudil reduced high BP, and increased ACE2 activity and Ang-(1–9) plasma levels . The changes in Ang-(1–9) levels in the DOCA-salt rats were not associated with altered Ang-(1–7) levels . Moreover, Ang-(1–9) infusion reduced experimental hypertension in AngII-infused and Goldblatt rats . Co-administration of Ang-(1–9) with PD123319 inhibited the anti-hypertensive effects of Ang-(1–9), indicating that AT2R activity is necessary for the anti-hypertensive action of Ang-(1–9) in vivo . A recent study in SHRSPs did not find an anti-hypertensive effect of Ang-(1–9) infusion . This was possibly due to the Ang-(1–9) dose used, which was six times lower than that used by Ocaranza et al.  (100 ng/kg per min compared with 600 ng/kg per min).
One contributing factor in reducing BP in hypertensive animals may be a decrease in total peripheral resistance resulting from higher Ang-(1–9) plasma levels. At least two mechanisms may explain Ang-(1–9)-induced vasodilation in hypertension: first, prevention/reversion of endothelial dysfunction; and secondly, activation of the endothelial AT2R signalling cascade by Ang-(1–9). In support of the first mechanism, our ex vivo functional studies in resistance arteries from AngII-treated rats showed that co-infusion with Ang-(1–9) preserved endothelium-dependent relaxation induced by acetylcholine . Ang-(1–9) also increased aortic eNOS mRNA levels, an effect that was associated with higher nitrate plasma levels. These effects of Ang-(1–9) were blocked by PD123319, showing that Ang-(1–9) increased NO bioavailability, implicating the AT2R . In agreement with our results, eNOS blockade significantly increased the contractile response to phenylephrine in aortic rings of SHRSPs that received chronic Ang-(1–9) infusion, whereas eNOS blockade did not alter contractile response to phenylephrine of aortic rings from vehicle-infused SHRSPs . Supporting a direct action of Ang-(1–9) on the endothelial AT2R, in experiments with aortic rings pre-contracted with a maximal dose of adrenaline, the vasodilator action of Ang-(1–9) was prevented by the presence of PD123319 . Moreover, mechanical ablation of the endothelium or eNOS blockade with L-NAME abolished the vasodilator action of Ang-(1–9). AT2R-mediated relaxation is a well-known effect in isolated vessels . Vascular AT2R activation by AngII releases BK and NO .
Thus these data indicate that Ang-(1–9)-induced vasodilatation is endothelium-dependent and may imply endothelial NO production due to AT2R activation. However, it is important to note that, in addition to NO, the release of arachidonic acid, another vasodilator, may be implicated in the vasodilator action of Ang-(1–9) .
Although it is tempting to speculate that the anti-hypertensive action of Ang-(1–9) may result from vascular AT2R activation, vasodilatation and reduced total peripheral resistance, the influence of AT2R on BP regulation in vivo is still controversial. Studies using AT2R-KO mice show elevated SBP (systolic BP) and enhanced sensitivity to the vasopressor effects of AngII . On the other hand, overexpression of the AT2R in the vasculature did not alter SBP, but markedly impaired AngII-induced vasopressor activity . The injection of lentiviral particles of the AT2R into the left ventricle of 5-day-old SHRs did not affect BP or cardiac hypertrophy compared with wild-type rats . Conversely, subsequent studies in rats with AT2R adenoviral infection found a significant decrease in BP . Selective AT2R stimulation with the AT2R agonist CGP42112 in Sprague–Dawley rats reduced BP , and similar results were observed after treating SHRs with compound 21, another AT2R agonist . Reduction in SBP using an AT2R agonist was potentiated by concomitant AT1R blockade [66,67].
AT2R is present in the renal vasculature and in proximal tubule epithelial cells . Several studies have demonstrated that AT2R activation enhances urinary sodium excretion [68–70]. Thus the anti-hypertensive action of Ang-(1–9) may also involve increased natriuresis due to renal AT2R activation and enhanced NO/cGMP cascade in the kidney.
The recent experimental evidence shows that Ang-(1–9) is a natural anti-hypertensive molecule that causes vasodilatation and protects the heart and vessels from adverse remodelling in hypertension and heart failure. These beneficial effects of Ang-(1–9) are not modified by the Mas receptor antagonist A779, but are abolished in the presence of PD123319, suggesting that the beneficial effects of Ang-(1–9) are mediated by the AT2R. However, further studies are necessary to confirm that Ang-(1–9) binds to the AT2R and leads to activation and AT2R-dependent cell signalling in cardiomyocytes, endothelial cells vascular smooth muscle cells and renal cells. Moreover, further studies in genetically modified mice are necessary to evaluate the relevance of AT2R to the beneficial effects of Ang-(1–9) in hypertension and heart failure. The available experimental evidence indicates that circulating Ang-(1–9) may have important clinical implications for long-term therapy and suggests that new approaches aimed at increasing Ang-(1–9) levels may be useful for prevention and treatment of hypertension and as well as CV organ damage.
- Ang-(1–7) etc.
- AngI etc.
angiotensin I etc. ANP, atrial natriuretic peptide
angiotensin type 1 receptor
angiotensin type 2 receptor
bradykinin B2 receptor
endothelial NO synthase
insulin-like growth factor-1
NG-nitro-L-arginine methyl ester
plasminogen activator inhibitor-1
soluble guanylate cyclase
spontaneously hypertensive rat
transforming growth factor-β1
Our own work was supported by the Fondo Nacional de Desarrollo Científico y Tecnológico (FONDECYT) [grant numbers 1100874 (to M.P.O., M.C., L.M. and J.E.J.), 1130550 (to L.M.), 1085208 (to J.E.J., M.P.O. and S.L.)]; the Fondo de Fomento al Desarrollo Cientifico y Tecnologico (FONDEF) [grant number D11I1122 (to M.P.O., L.M., M.C., S.L. and J.E.J.)]; a Comisión Nacional de Investigación Científica y Tecnológica (CONICYT) [grant number Anillo ACT-1111 (to S.L. and M.C.)]; the Fondo de Investigación Avanzada en Areas Prioritarias (FONDAP) [grant number 15130011 (to S.L. and M.C.)]; and the Millennium Institute of Immunology and Immunotherapy [grant number P09/016-F (to L.M)].