Targeting the Renin-Angiotensin-Aldosterone System in Fibrosis

Abstract

Fibrosis is characterized by excessive deposition of extracellular matrix components such as collagen in tissues or organs. Fibrosis can develop in the heart, kidneys, liver, skin or any other body organ in response to injury or maladaptive reparative processes, reducing overall function and leading eventually to organ failure. A variety of cellular and molecular signaling mechanisms are involved in the pathogenesis of fibrosis. The renin-angiotensin-aldosterone system (RAAS) interacts with the potent Transforming Growth Factor β (TGFβ) pro-fibrotic pathway to mediate fibrosis in many cell and tissue types. RAAS consists of both classical and alternative pathways, which act to potentiate or antagonize fibrotic signaling mechanisms, respectively. This review provides an overview of recent literature describing the roles of RAAS in the pathogenesis of fibrosis, particularly in the liver, heart, kidney and skin, and with a focus on RAAS interactions with TGFβ signaling. Targeting RAAS to combat fibrosis represents a promising therapeutic approach, particularly given the lack of strategies for treating fibrosis as its own entity, thus animal and clinical studies to examine the impact of natural and synthetic substances to alter RAAS signaling as a means to treat fibrosis are reviewed as well.

Fibrosis as a Response to Injury

All mammalian organs respond to tissue injury, whether from trauma or various diseases, by initiating a cascade of cellular and molecular events that can eventually lead to tissue fibrosis. The tissue response to injury usually involves an integrated sequence of cellular events that include tissue inflammation, proliferation and regeneration [1]. The regenerative stage aims to restore the tensile strength of the organ, and scar formation occurs during this stage [2–4]. Tissue injury stimulates different effector cells such as inflammatory cells, epithelial cells, endothelial cells and fibrogenic cells to activate and deactivate subsets of intracellular signaling pathways that are common in different body organs, for example the transforming growth factor beta (TGFβ) pathway.

Early in the proliferative phase, fibroblasts migrate to the site of the wound and acquire bundles of microfilaments composed of α- and β-cytoplasmic actin to form stress fibers, at which time they are variously described as activated fibroblasts or proto-myofibroblasts based on their ability to generate small tractional forces [5]. Eventually, α-smooth muscle actin (αSMA) becomes expressed and incorporated into these microfilaments, marking the conversion of protomyofibroblasts to myofibroblasts [1, 6]. Myofibroblasts synthesize the elements of the extracellular matrix (ECM) and restore the tensile strength of the organ [7, 8]. ECM proteins that contribute to fibrosis are similar in different tissues, and include collagens (primarily types I and III), fibronectin, basement membrane proteins such as laminin and other less common proteins and glycoproteins, however the relative amount of these proteins may vary by tissue [9]. As myofibroblasts express the contractile protein αSMA, they are responsible for wound contraction and maturation of the granulation tissue [10, 11]. Once the wound closes, myofibroblasts may undergo regulated cell death, apoptosis, to resolve the wound healing process [12]. If myofibroblasts persist after wound healing is otherwise complete, they may contribute to tissue deformation by contracture, as well as by maintaining elevated production of ECM constituents including type I collagen [13].Contractures in the skin manifest as hypertrophic scars [14]. The hallmark of scleroderma, also referred to as ―stiff skin disease,‖ is progressive fibrosis and contracture of multiple organs. Myofibroblast persistence can result in devastating outcomes when contractures and fibrosis affect vital organs such as the heart, lung, liver and kidney [15].

The molecular signaling pathways of fibrosis comprise a complex and sophisticated network that involves myriad inflammatory mediators such as cytokines, chemokines, and growth factors that recruit and activate a wide range of inflammatory cells that in turn lead to activation of stromal fibrogenic effector cells including fibroblasts [2].These signaling pathways include growth factors that alter fibrogenic cell behavior, including TGFβ, connective tissue growth factor (CTGF, now referred to as CCN2), and platelet-derived growth factor (PDGF), which generally exert pro-fibrosis effects [16]. However, mechanical force is also an important regulator of cell activation in fibrosis, thus the ECM and its interaction with cells via adhesion molecules such as integrins and cadherins also provide alternative signaling mechanisms [17]. Vasoactive peptides including endothelin-1 but in particular angiotensin II (Ang II) appear capable of tuning these responses, in part by impacting TGFβ signaling [18]. TGFβ is synthesized and secreted from inflammatory and fibrogenic cells, yielding both paracrine and autocrine modes of action [19, 20]. The TGFβ signaling pathway interacts with other cellsignaling mechanisms, including in particular the renin-angiotensin-aldosterone system (RAAS), which has been extensively shown to contribute to the development of fibrosis in multiple organs [20, 21]. This review will discuss the role of RAAS in the development and progression of fibrosis in various organs, with a focus on RAAS crosstalk with TGFβ, and highlights recent efforts to target RAAS to ameliorate fibrosis.

The Renin-Angiotensin-Aldosterone System

RAAS is best recognized for its essential role in the physiological regulation of blood volume, blood pressure, and sodium homeostasis [22]. Given this critical role, it is not surprising that altered RAAS signaling is also involved in the pathogenesis of several significant clinical conditions such as hypertension, heart failure, extracellular matrix remodeling and fibrosis [23]. Beside the systemic RAAS, detailed studies have revealed that local, tissue-level RAAS is active in many body organs [24]. Ang II, a major hormone produced in the RAAS, acts through specific receptors to exert a wide range of biological actions affecting almost every tissue in the body including the kidneys, cardiovascular system, liver, brain, skin and immune system [25, 26]. The components of RAAS can be considered to arise from two individual pathways: the classical, and the alternative pathway (Figure 1).

Figure 1. The Renin-Angiotensin-Aldosterone System.

RAAS consists of two parallel axes, representing the classical and alternative signaling pathways. The classical RAAS includes angiotensinogen that is converted into angiotensin I by the action of renin. Angiotensin I is subsequently converted into angiotensin II (AngII) by the action of Angiotensin Converting Enzyme (ACE). The classical RAAS induces its pro-fibrotic actions by activating the angiotensin II type 1 receptor (AT1R). The alternative RAAS axis comprises angiotensin-(1–9), angiotensin-(1–7) and alamandine (Amdn), generated by decarboxylation (DeCxn) of angiotensin-(1–7) or step-wise conversion of angiotensin II to angiotensin A by decarboxylation followed by processing by ACE2. The alternative RAAS pathway exerts anti-fibrotic actions through the angiotensin II type 2 receptor (AT2R), Mas receptor (MasR) and Mas-related GPCR, member D (MrgD). Angiotensin I can also be converted to angiotensin-(1–7) by the action of neutral endopeptidase (NEP) or prolylendopeptidase (PEP).

The Classical RAAS pathway

Activation of the classical RAAS pathway begins with the synthesis of the aspartyl protease renin. Renin is produced primarily from juxtaglomerular cells of the renal afferent arteriole [27]. Low body fluid volume, high sodium, decreased renal perfusion and sympathetic activity act as stimuli for renal renin release [28]. In the circulation, renin converts liver-derived angiotensinogen into angiotensin I (Ang I). Ang I in turn is converted into the biologically active octapeptide Ang II by the action of angiotensin converting enzyme (ACE) [29].

It is noteworthy that the renin precursor prorenin can also be released by the kidneys and has a receptor that is expressed in the kidneys as well as other tissues. Binding of prorenin to its receptor amplifies the classical pathway by increasing the efficiency of angiotensinogen cleavage by renin [30]. However, receptor-bound prorenin may induce distinct signaling pathways. Prorenin activates the extracellular signal-regulated kinase (ERK) 1/2 and p38 pathways, and increases the production of TGFβ and the matrix proteins plasminogen activator inhibitor (PAI1) and fibronectin [31–33].

Ang II is a versatile molecule that acts in intracrine/autocrine/paracrine fashions in almost all body tissues, performing its biological functions via interaction with two G protein-coupled receptors (GPCR), the Ang type 1 (AT1R) and Ang type 2 (AT2R) receptors [34]. Most biological functions of Ang II have been attributed to its binding to AT1R, which leads to the activation of phospholipases C, A2 and D, and inhibition of adenylate cyclase [35, 36]. AT1R binding of Ang II also activates tyrosine kinase phosphorylation and activation of phospholipase C-γ, leading to the activation of several downstream signals such as Janus kinases and mitogenactivated protein kinases (MAPKs), and eventually changes in gene transcription resulting in other cellular effects. Ang II activation of AT1R results in the opening of Ca²⁺ channels and the influx of extracellular Ca²⁺ into the cell [37, 38]. Ang II also stimulates the Smad signaling pathway similar to the action of TGFβ, activating the receptor-regulated R-Smads Smad2 and Smad3 even in the presence of antagonists of TGFβ, leading to downstream pro-fibrotic effects [39, 40] (Figure 2).

Figure 2. Schematic representation of the interactions of TGFβ and RAAS signaling.

Ligand activation of TGFβ receptors leads to Smad2/Smad3 phosphorylation, leading to recruitment of Smad4 and translocation to the nucleus to activate transcription of fibrotic genes such as collagen, TGFβ, PDGF, CCN2 and periostin. TGFβ also activates non-canonical pathways via MAPKs such as ERK, JNK and p38.Binding of AngII to its type 1 receptor (AT1R) activates MAPKs and Smads, leading to synthesis and release of TGFβ and creating a positive feedback loop to amplify fibrotic TGFβ signaling. Aldosterone activation of mineralocorticoid receptor (MR) induces transcriptional activation of pro-fibrotic proteins. Inhibition of AT1R and MR attenuates receptor activation and inhibits pro-fibrotic signaling. Activation of AT2R by Compound 21 or Ang II, or of MasR by Ang-(1–7), inhibits TGFβ fibrotic signaling and counteracts the effects of the classical RAAS pathway. Pirfenidone may both inhibit AT1R and activate MasR.

The Alternative RAAS Pathway

The main components of the alternative RAAS pathway are ACE2, the heptapeptide Ang-(17), and the Ang-(1–7) Mas receptor (MasR/MAS1) (Figure 2) [41]. Interestingly, the effects of the alternative pathway typically antagonize those of the classical pathway. ACE2 is a close homologue of ACE with 42% similarity in the catalytic domain [42]. ACE2 cleaves one amino acid at the carboxy-terminal end of Ang I to produce Ang-(1–9), and cleaves Ang II to produce Ang-(1–7) [43]. Ang-(1–7) can also be generated by the ACE-mediated cleavage of Ang-(1–9) [44], or by cleavage of Ang I by neutral endopeptidase or prolylendopeptidase [45, 46]. Ang-(17) and Ang-(1–9) mediate their patho/physiological functions by binding to MasR and AT2R, respectively [47]. MasR is an oncogenic GPCR with a ubiquitous expression pattern; it is abundant in the brain and gonads, but under physiological conditions is expressed at low levels in the lungs, heart, blood vessels, kidney and skeletal tissues [48]. Recently, a new member of the RAAS family named Ala1-Ang-(1–7), or alamandine, has been identified as a decarboxylation product of Ang-(1–7) [49]. Alamandine is also produced by Ang II decarboxylation to form angiotensin A which is converted to alamandine by the action of ACE2 [49, 50]. This peptide exerts anti-hypertensive and anti-fibrotic actions through its binding to the Mas-related GPCR, member D (MrgD) [49, 51].

The Ang II/AT2R axis produces actions that counteract those stimulated by Ang II/AT1R. In this regard, Ang II/AT1R usually activates phosphorylation and protein synthesis, while Ang II/AT2R dephosphorylates signaling molecules and inhibits protein synthesis. However, both axes produce actions that are dose- and location-dependent. ACE/ACE2 activity is essential for both RAAS axes, and regulates local shifts in the balance of the two pathways [52].

ACE2/Ang-(1–7)/MasR may regulate several important functions whose perturbation may cause various diseases. For example, ACE2 knockout in mice resulted in impaired cardiac contraction and increased plasma Ang II levels [52]. ACE2 mRNA and protein were reduced in rat hearts with hypertension [52]. Renal fibrosis was attenuated through ACE2 upregulation in long-term renal denervation in spontaneously hypertensive rats [53]. This RAAS axis is mediated by diverse signaling pathways including nitric oxide (NO), stimulation of cyclic AMPdependent protein kinase, and reduced MAP kinase activities [54–56]. Moreover, activation of this pathway leads to anti-inflammatory, anti-fibrotic, and anti-thrombotic activities, and inhibition of cell growth [56].

RAAS Physiological Functions

As noted above, RAAS components are expressed by almost all tissues of the body and are involved in myriad homeostatic functions. In the cardiovascular system, Ang II increases blood pressure through direct and indirect actions that target vascular tone and sodium/water balance [57]. Ang II activates the sympathetic nervous system leading to increased heart rate and contractility and induces general vasoconstriction, resulting in an increase in total peripheral resistance. Ang II also increases the pro-thrombotic potential of the blood by stimulating plasminogen activator inhibitor protein PAI-1 and PAI-2 [58]. In the renal system, Ang II promotes sodium reabsorption in the proximal tubule and induces the secretion of aldosterone from the adrenal cortex, which in turn regulates salt homeostasis by stimulating sodium reabsorption and potassium secretion in the kidney distal tubules [59]. Furthermore, RAAS stimulates the thirst reflex and prevents fluid urinary loss through its action on the hypothalamus to stimulate the antidiuretic hormone (ADH/vasopressin) release from the posterior pituitary gland [60]. Ang II stimulates the release of adrenocorticotropic hormone (ACTH) from the anterior pituitary to regulate cortisol secretion from the adrenal gland [61]. Ang II has been shown to influence prostaglandin release, amplifying the vasoconstriction effect of Ang II [62]. In the adipose tissues, Ang II increases lipogenesis and the risk of adipose inflammation, and is linked to glucose homeostasis and insulin release [63].

Importantly, Ang II and aldosterone influence the synthesis of extracellular matrix proteins such as collagen I, fibronectin and plasminogen activator inhibitor proteins PAI-1 and PAI-2. These actions suggest crucial roles for RAAS in tissue fibrosis [64, 65].

TGFβ Signaling

Ang II has been widely associated with fibrosis due to its ability to activate TGFβ signaling pathways in addition to its stimulatory effect on TGFβ synthesis and secretion [66]. Overactivation of TGFβ is a key feature of fibrosis in almost all tissues and is responsible for fibroblast to myofibroblast activation and ECM protein synthesis. TGFβ can inhibit the activity of matrix metalloproteinases (MMPs) by stimulating synthesis of tissue inhibitors of metalloproteinases (TIMPs) [67].The impact of TGFβ on MMP/TIMP balance favors ECM production, and thus amplifies the fibrotic signals, although the specific effect of TGFβ on the expression of individual MMPs and TIMPs can vary, and in fact TGFβ can up-regulate some MMPs to promote ECM remodeling [68, 69]. Additionally, TGFβ enhances the synthesis of the fibrotic mediators CCN2 and PDGF [70, 71].

TGFβ mediates its biological function by binding to its receptor, TGFβ type II receptor (TβRII). Subsequently, TβRII complexes with and trans-phosphorylates the TGFβ type I receptor (TβRI) [72]. In turn, TβRI activates intracellular Smad signaling pathways by phosphorylating Smad2 and Smad3 (Figure 2), which bind to a common mediator Smad4 and translocate to the nucleus where this complex acts as transcription factor for genes encoding ECM proteins such as collagen 1α1, αSMA, periostin, and fibronectin. Smad3 can activate Smad7 as well, which acts as a negative feedback mechanism to inhibit TGFβ-mediated downstream effects [73].

In addition to the canonical Smad signaling pathway, TGFβ stimulates several non-canonical pathways by activating multiple kinases, including the MAPKs ERK, p38, c-Jun N-terminal kinase (JNK), phosphatidylinositol-3-kinase (PI3K), Akt/PKB, and ROCK [74, 75]. Activated kinases in turn regulate the Smad signaling pathways positively or negatively, depending on cell type [76]. Non-canonical TGF pathways play several roles in fibrosis of different organs [77]. Additionally, TGFβ induces the nuclear translocation of serum response factor/myocardinrelated transcription factors (SRF/MRTF). A number of SRF/MRTF target genes such as αSMA and CCN2 are involved in tissue fibrosis [78].

RAAS in Liver Fibrosis

Hepatic fibrosis is a multifactorial process that involves hepatic stellate cells (HSC) and Kupffer cells (KC), growth factors, chemokines and cytokines that eventually results in disruption of homeostatic functions and organ failure. Similar to fibrosis in other tissues, hepatic fibrosis proceeds by HSC and KC activation and proliferation, resulting in the generation of myofibroblasts or the production of TGFβ, respectively [8, 79]. These actions lead to the synthesis of increased amounts of ECM components, ECM remodeling and the contraction of scar tissues [79]. Ample evidence demonstrates the involvement of RAAS over-production at different stages of liver fibrosis [80]. Ang II leads to increases in intrahepatic resistance and ECM protein production via its vasoconstriction effect, enhances oxidative stress, and acts as a local and systemic cytokine, thus enhancing pro-inflammatory mediators [81, 82].

Renin and aldosterone levels have been shown to be elevated in patients with liver cirrhosis [83]. In healthy human liver, quiescent HSCs do not express RAAS components, but activation of these cells in the cirrhotic liver or in culture models of liver damage leads to substantial expression of renin, ACE, and Ang II [84]. This increase could be a physiological reflex to systemic vasodilatation that takes place in patients with cirrhosis, or due to a local activation of the RAAS [85]. Hepatic fibrosis results, in part, from Ang II-mediated conversion of HSC to myofibroblasts and subsequent contraction through TGFβ stimulation[86]. Moreover, Ang II induces the production of vascular endothelial growth factor (VEGF), which is one of the major angiogenic factors upregulated during fibrosis [87]. VEGF stimulates HSC activation and exacerbates hepatic fibrosis [87].

Ang II acts on AT1R to activate HSCs via several signaling molecules, including MAPK pathways, Janus kinase 2 (JAK2), the phosphoinositide/Ca²⁺ pathway, and the generation of reactive oxygen species via phosphorylation of the p47phox subunit of Nox [88, 89]. In quiescent and activated HSCs, Ang II-produced TGFβ mRNA and protein expression is dependent on ERK1/2- and Nox-dependent pathways, but independent of protein kinase C [90]. AT1R stimulation in wild type rats and mice resulted in activation of HSCs and production of liver fibrosis by the activation of intracellular JAK2, which in turn activated RhoA/Rho-kinase [89]. Liver fibrosis was prevented by deleting the AT1R gene in vivo, and by pharmacological inhibition of JAK2 [89]. Ang II enhances periostin expression in HSCs, and periostin in turn upregulates both TGFβ and collagen 1α1expression to drive fibrosis [91]. Furthermore, Ang II induced the expression of TIMP-1 via AT1R and phosphokinase C [91].

Conversely, the alternative RAAS pathway counterbalances the effects of the classical RAAS pathway through its components ACE2, Ang-(1–7), and MasR [41]. This ‘alternate axis’ produces anti-fibrotic effects, thus protecting tissues from the potentially harmful effects of classical RAAS activation [41]. In animal models of cirrhosis, ACE2 and Ang-(1–7) expression levels were upregulated [92]. Ang-(1–7) in turn reduced the vascular tone in the mesenteric bed of cirrhotic animals [92]. Moreover, the MasR agonist AVE 0991 reduced expression of the profibrotic proteins collagen 1α1, αSMA, TGFβ, and ACE [93]. Furthermore, loss of ACE2 exacerbated fibrosis in BALB/c mice subjected to bile duct ligation, while recombinant ACE2 inhibited key features of liver fibrosis in chronic liver injury models, suggesting a protective role of the alternative RAAS pathway in liver fibrosis [94].

RAAS in Cardiac and Renal Fibrosis

Cardiac fibrosis is a significant contributor to mortality and morbidity worldwide [95, 96]. It is characterized by the excessive production and deposition of ECM in the cardiac wall or septum, and is associated with most cardiac pathologic conditions such as acute myocardial infarction, pressure overload with hypertension and aortic stenosis, volume overload due to valvular diseases, hypertrophic cardiomyopathy, post-viral dilated cardiomyopathy, and aging [97]. Cardiac fibrosis is variable with different pathological insults; however, the cellular mechanisms involved in fibrosis are similar. The RAAS is a major pathway that can contribute to cardiac fibrosis, with TGFβ being a key downstream signaling molecule of Ang II/AT1R that acts in an autocrine or paracrine manner to induce fibrosis in the heart, although the full mechanisms involved are not fully understood [98]. The Ang II/AT1R/p38 MAPK pathway is activated in cardiac and renal fibrosis, leading to an imbalance in the ratio of ACE/ACE2 which, if restored, could provide a novel approach to treat fibrotic diseases [99].

After cardiomyocyte injury, local RAAS components are dramatically elevated in the heart [100]. Ang II regulates the expression of integrins – transmembrane receptors that play important roles in the physiology and pathophysiology of the heart, including fibrosis [101]. Acting on AT1R, Ang II upregulates αv, β1, β3, β5,and α8β1 integrins at the mRNA and protein levels [101–103]. Ang II enhances the binding of cardiac fibroblasts to a variety of ECM proteins, including collagen I, fibronectin, vitronectin, and laminin, and this binding activates focal adhesion kinase (FAK) [104]. Additionally, β3 integrin plays an important role in Ang IIinduced attachment of these ECM proteins [104]. Conversely, β1 integrin regulates angiotensinogen expression in cardiac fibroblasts by activating and inhibiting the intracellular kinases Rac1 and RhoA, respectively [105]. Ang II induces c-Fos, Early Growth Response Protein 1 and MAPK activities [106]. Furthermore, Ang II up-regulates the expression of profibrotic TGFβ, the ECM proteins laminin and fibronectin, collagen I and II, as well as PAI-1 through AT1R [101]. Further, Ang II upregulates Tenascin-C which is a matricellular protein that exacerbates the fibrotic process by activating macrophages and fibroblasts to generate more pro-fibrotic cytokines in the myocardium [107]. Angiotensin-mediated activation of Tenascin-C occurs via activation of the integrin αvβ3 and nuclear translocation of phosphorylated NF-κB, and eventually increased production of collagen I by cardiac fibroblasts [107].

Osteopontin [108] is an adhesion molecule that is important for renal and cardiac fibrosis [108]. Ang II augments its expression in cardiac fibroblasts, and osteopontin depletion has been shown to prevent Ang II-mediated induction of cardiac fibroblast collagen gel contraction by an integrin-dependent mechanism [109–111]. Osteopontin promotes fibrosis by enhancing macrophage activation and fibroblast proliferation stimulated by Ang II [112].

Ang II also upregulates CCN2 in renal and cardiac tissues. Ang II-induced CCN2 expression via AT1R occurs in a protein kinase C (PKC)-dependent manner [113]. Ang II stimulates TGFβ and CCN2 in cultured fibroblasts by activating p38 MAPK. AT1R inhibition using losartan and a p38 MAPK inhibitor attenuates CCN2 expression [114].

Aldosterone, via the mineralocorticoid receptor (MR), induces cardiac and renal fibrosis that can be prevented by the MR antagonists spironolactone and eplerenone [115–117] (Figure 2). Aldosterone recruits and activates macrophages, leading to increased expression of inflammatory cytokines such as intercellular adhesion molecule 1, and activates fibroblast conversion to myofibroblasts [118]. Aldosterone potentiates cardiac and renal fibrosis signaling by directly stimulating pro-fibrotic proteins TGFβ, CCN2, endothelin 1, placental growth factor, PAI-1, galectin-3, and osteopontin [119, 120]. Generation of reactive oxygen species (ROS) is another mechanism by which Ang II and aldosterone contribute to fibrosis [120, 121]. Ang II and aldosterone both stimulate several enzymes required for ROS production such as nicotinamide adenine dinucleotide phosphate oxidase through AT1R- and MR-dependent mechanisms, respectively [122]. In cardiac and renal fibrosis, the effects of aldosterone are in part due to the inhibition of NO and cGMP pathways and the subsequent activation of NF-κB signaling, which in turn may induce the expression of pro-fibrotic proteins [123, 124].

In addition to the transcriptional activation of MR, aldosterone can induce rapid non-genomic effects that are insensitive to transcriptional and translational inhibitors [125]. These nongenomic effects may include the activation of various intracellular signaling pathways such as ERK1/2, PI3K, diacylglycerol, PKC, and JNK. These affects might be due to aldosterone activation of the newly characterized G protein-coupled receptor 30 [125–127]. The interactions between Ang II and aldosterone are evident in a variety of experimental settings and diseases. For example, Ang II directly induces the nuclear translocation of MR, which may partially account for the gene expression induced by Ang II stimulation. Moreover, Ang II-induced cardiac fibrosis has been shown to depend on the ability of Ang II to induce local aldosterone synthesis in heart tissues by activating AT1R, steroidogenic acute regulatory protein and aldosterone synthase protein expression [128].

While the classical RAAS pathway described above produces pro-fibrotic responses, the alternative RAAS components act in the heart and kidney to counteract these pathological actions. The key enzyme of alternative RAAS activation, ACE2, is expressed in the heart, kidney and other tissues [52, 92, 94, 129]. Several studies have reported that ACE2/Ang-(1–7)/MasR pathway activation has anti-fibrotic effects in the heart and kidney [130]. Genetic deletion of ACE2 in mice results in severe cardiac dysfunction and up-regulation of hypoxia-induced genes with a concomitant increase of Ang II expression [52]. Moreover, loss of ACE2 in the left anterior descending coronary artery ligation model of myocardial infarction leads to increased infarct size that is associated with oxidative stress, increased MMP expression and greater inflammation [131]. These changes were associated with increased classical RAAS pathway activity that mediated the up-regulation of nicotinamide adenine dinucleotide phosphate oxidase and enhanced ROS formation [131]. Similarly, ACE2 deletion or inhibition leads to Ang IIdependent oxidative renal damage in aged mice [131]. Furthermore, these effects could be due to Ang II-dependent activation of the MAPK pathway as phosphorylation of ERK1/2 and JNK1/2 was increased in the infarct area of ACE2 knockout mice and AT1R blockade prevented these adverse effects [131].Over-expression of ACE2 using lentiviral delivery in rats attenuates the deleterious cardiac effects and fibrosis induced by Ang II [132, 133]. Moreover, recombinant ACE2 treatment in Col4a3 knockout mice results in reduction in ECM protein accumulation, inhibition of TGFβ signaling and less fibrosis in the kidney [129]. More recently, Wang et al. showed that ischemic post-conditioning could serve as an adjunct to treat fibrosis after MI. Postconditioning was associated with reduced ACE and AT1R and increased ACE2 and AT2R expression in addition to a coincident reduction in the TGFβ/Smad fibrotic pathway [134]. Similar to the cardiac role of ACE2, loss of this enzyme resulted in a four-fold increase in the ratio of local renal Ang II/Ang-(1–7) that was associated with increased expression of fibrotic proteins, including αSMA and collagen I [135]. These changes were due to activation of Ang II and the ERK1/2, TGFβ/Smad2/3, and NF-κB signaling pathways [135]. Moreover, ubiquitin degradation of inhibitory Smad7, mediated by the E3 ligase Smurf2, was reported in the same study, and could act to amplify TGFβ/Smad2/3-mediated renal fibrosis [135].

RAAS in Skin Fibrosis

RAAS components are present in normal skin, suggesting that local RAAS may be a homeostatic regulator of the skin [136]. However, the expression level of RAAS components is not abundant. AT1R is the predominant Ang II receptor present in unwounded skin, with distribution in epidermis keratinocytes and blood vessels, and low expression in dermis fibroblasts [137]. Angiotensinogen, renin, MasR, ACE and ACE2 are also present in normal skin, and colocalized with AT1R [137, 138].

Upon injury, RAAS is a key regulator of all steps of wound healing in the skin, including hemostasis, recruitment and activation of keratinocytes and fibroblasts, wound contraction, and ECM remodeling, and it does so by stimulating synthesis and release of various pro-fibrotic proteins [139]. Injury of the skin induces time-dependent expression changes in various RAAS components [139]. Inflammatory cells and resident cell byproducts such as cytokines and chemokines act locally to increase or decrease RAAS players in the skin. Inhuman skin, both AT1R and AT2R are up-regulated 24 hours after cutaneous injury, with higher expression of AT2R in the latter stages of the healing process [140, 141]. The roles of AT1R and AT2R in skin healing have been demonstrated in loss-of-function knockout animals. AT1R deletion in mice delayed skin healing, while deleting AT2R halted healing at early stages but accelerated the healing process at later stages [142]. AT2R knockout mice showed weaker healed skin compared to wild type animals [142]. Ang-(1–7) promotes wound healing via activation of MasR, leading to stem cell proliferation and eventually re-epithelialization [143]. ACE expression is upregulated in the skin after injury, and inhibition of ACE is correlated with a reduction in TGFβ and tissue necrosis factor α suggesting that ACE is involved in skin healing [144]. Furthermore, ACE deletion or inhibition with ramipril reduced scar formation and expression of fibrotic markers [145]. ACE appeared to induce these effects via TGFβ-induced-Smad2/3 phosphorylation and a TGFβ-activated kinase 1-dependent mechanism [145]. These observations suggest that the temporal balance in the expression of local AT1R and AT2R regulates proper skin wound healing.

Several debilitating diseases such as diabetes, aging and chronic burns induce disturbances in the local skin RAAS that lead to dysregulated wound healing [146]. Ang II induces collagen synthesis in skin fibroblasts via AT1R and augments insulin-like growth factor 1 (IGF-1)induced collagen synthesis. Conversely, activation of AT2R by Ang II inhibits collagen production and IGF-1-induced collagen synthesis [146]. Moreover, Ang II opposes MMPs activity by regulating the level of TIMP-1 [147]. AT1R activation enhances TIMP-1 expression and thus inhibits collagen degradation, whereas AT2R activation inhibits TIMP-1 expression to promote collagen degradation [146]. The inhibitory effect of AT2R could be partially due to activation of protein tyrosine phosphatase SHP-1 [146].

Targeting RAAS to Treat and Prevent Fibrosis

Fibrotic disorders and resultant organ failure account for an increasing and significant socioeconomic burden on patients worldwide, yet with few exceptions, there are no specific and potent medications available for the treatment of fibrosis [148]. The molecular mechanisms responsible for fibrosis exhibit both overlap and distinction across body organs, however in almost all tissues, activation of the classical RAAS pathway leads to pro-fibrotic effects through AT1R while activation of the alternative pathway Ang-(1–7)/ACE2/AT2R or MasR leads to inhibition of fibrosis [149]. There is a growing body of evidence that fibrosis in various body organs, at a minimum, can be attenuated, and may even be reversible to some degree, which would have significant positive effects on tissue/organ function and recovery from disease (Table 1). In this regard, targeting RAAS represents a novel and promising approach.

Table 1.

Studies of RAAS blockade and fibrosis inhibition

Drug	Study Type	Target Organ	Mechanism of Action	Outcome	References
Candesartan	Randomized, open-label controlled study	Liver	ARB	Reduction in αSMA, TGFβ1, Col-1, TIMP-1 and MMP2	[150]
Olmesartan, candesartan, losartan	Meta-analysis of 4 randomized controlled trials	Liver	ARB	Reduction in liver fibrosis score and area	[152]
A-779	Animal	Liver	Mas receptor antagonist	Aggravation of liver fibrosis and elevation of TGFβ	[93]
Aliskiren	In vitro	Heart, kidney	Renin inhibitor	Reduction in fibroblast proliferation, αSMA, TGFβ and collagen	[156, 159]
Candesartan	Clinical trial	Heart	ARB	Reduction of type I procollagen-Npeptide and type III procollagenN-peptide	[160]
Spironolactone	Animal	Heart, liver	Aldosterone antagonist	Reduction of TGFβ and reactive oxygen species	[151, 162, 167]
Lisinopril	Randomized clinical trial	Heart	ACE inhibitor	Reduction of myocardial fibrosis and collagen volume	[163]
Enalapril	Animal	Heart, kidney	ACE inhibitor	Reduction in fibrosis and long-term protection against organ damage	[164, 165]
Pirfenidone	Animal	Heart	Classical/alternative RAAS balance	Reduction in fibrosis	[99]
Compound 21 and A8011	In vitro	Heart, skin	AT2R agonist	Reduction in fibrosis, collagen, TIMP1, MMP1–9 and TGFβ	[171-173, 175]
Ramipril	Animal and in vitro	Skin	ACE inhibitor	Decreased collagen, TGFβ1 expression and attenuated TGFβ-activated kinase-1 phosphorylation	[145]
Topical losartan and valsartan	Pilot placebocontrolled single blinded study	Skin	ARB	Reduction in scar score	[174, 175]

Targeting RAAS in Liver Fibrosis

The efficacy of RAAS blockade in preventing and treating liver fibrosis has been tested in several in vitro and in vivo investigations. AT1R blockers (ARBs), ACE inhibitors, and selective aldosterone blockers have been reported to be effective in suppressing hepatic fibrosis [150, 151]. The ARB candesartan, administered for 6 months in patients with compensated alcoholic liver disease, ameliorated liver fibrosis and significantly reduced fibrotic areas, scores and related markers such as αSMA and hydroxyproline levels [150]. A recent meta-analysis of randomized controlled trials in patients with liver fibrosis revealed that RAAS inhibition significantly reduced measures of liver fibrosis, and that the therapeutic agents were welltolerated [152]. Ang-(1–7) reduces hepatic fibrosis by inhibiting activation of hepatic stellate cells [153, 154]. These data suggest that RAAS antagonism represents a promising therapeutic approach for liver fibrosis, however additional translational and controlled randomized clinical trials are needed for adoption as front-line therapy.

Targeting RAAS in Cardiac and Renal Fibrosis

Inhibiting renin could represent a successful approach to combat cardiac and renal fibrosis as it is the rate limiting step in the synthesis of Ang II [155]. Aliskiren is an oral renin inhibitor approved by the United States Food and Drug Administration for the treatment of hypertension [155]. It has been shown that aliskiren attenuates collagen deposition, oxidative stress, TGFβ synthesis and the infiltration of inflammatory cells in gingival tissues of both normal and diabetic mice [156–158]. Aliskiren may exert its anti-fibrotic effects independently of Ang II via activation of the prorenin receptor [159].

Blocking ACE/AT1R/MR has also been shown to improve cardiac and renal fibrosis. Candesartan reduces markers of cardiac fibrosis such as type I procollagen-N-peptide and type III procollagen-N-peptide in patients with atrial fibrillation [160]. Whether blood pressure reduction alone is sufficient to attenuate cardiac fibrosis has not been fully determined. However, it is noteworthy that the beneficial effect of RAAS inhibition on fibrosis can be independent of blood pressure regulation [151]. Queisser et al. have reported that blocking the MR receptor with spironolactone ameliorated hepatic fibrosis in hypertensive rats [151]. At sub-therapeutic doses, spironolactone prevented heart fibrosis as well [161]. The beneficial effects of RAAS inhibition could be due to reduction in reactive oxygen species in cardiac tissues [162]. In one prospective clinical trial, the ACE inhibitor lisinopril significantly reduced collagen volume fraction in participating patients, suggesting its potential use as an anti-fibrotic agent [163]. Reports have indicated that even transient ACE inhibition can protect against fibrotic stimuli [164]. Two weeks of enalapril treatment of adult spontaneously hypertensive rats produced persistent physiological changes in cardiac fibroblasts that lasted beyond the treatment course [164]. These changes were able to protect the heart from future noxious stimuli such as nitric oxide synthase inhibition [164]. Similarly, this transient treatment approach also produced long-term protection against renal interstitial fibrosis in response to nitric oxide synthase inhibition [165]. The AT1R antagonist telmisartan inhibited cardiac fibrosis in TGFβ over-expressing mice by modulating the MMP/TIMP ratio [166]. Aldosterone antagonists have also been reported to inhibit cardiac fibrosis. Treatment with spironolactone attenuates collagen deposition, inflammation and increases antioxidants in hyperthyroid rats [167].

Pirfenidone is an anti-fibrotic agent that could ameliorate fibrosis in various tissues [99]. The mechanism of action of pirfenidone is not fully understood, however recent studies have shown that this small molecule converges at the RAAS pathway in cardiac fibrosis. Pirfenidone seems to balance the ACE/ACE2 ratio to favor the anti-fibrotic RAAS axis [99]. Moreover, pirfenidone inhibited AT1R/p38 MAPK, corrected the ACE/ACE2 ratio and shifted the RAAS balance that eventually ameliorated MI-induced cardiac fibrosis [99]. The ARB losartan similarly inhibited AT1R/p38 MAPK and increased the expression of Liver X Receptor α (LXRα) which exerts anti-fibrotic effects as well [99, 168]. In support of these data, deletion of the small GTPase Rho A in activated fibroblasts using a tamoxifen-inducible periostin-Cre promoter (RhoA^fl/fl::PSTNCre/+) attenuated cardiac fibrosis and was associated with reduced TGFβ and non-canonical p38-MAPK signaling [169].

A recent study suggests a combination therapy of anti-oxidants and serelaxin, a recombinant form of the human ovarian hormone relaxin-2, may be effective to combat cardiac fibrosis without affecting the progression of related cardiac diseases. This combination therapy in a normotensive mouse model completely abrogated cardiac fibrosis to levels comparable to the fibrosis measured in control mice [170].

Targeting AT2R is a tempting approach to treat fibrosis. However, at present there are few selective AT2R agonists. Compound 21 (C21) is a highly selective AT2R agonist that has been widely used in preclinical research for this purpose. Treatment of myocardial infarction in rats with C21 significantly reduced scar size, and this was associated with reduced phosphorylation of the p44/42 and p38 MAPKs [171]. Additionally, C21 treatment of induced myocardial infarction for 6 weeks resulted in a substantial reduction in collagen content in peri-infarct myocardium, elevation of TIMP-1 expression and reduction in MMPs 1 and 9 [172]. Recently, Wang et al. were able to produce selective AT2R ligands through single β-amino acid substitutions into various angiotensin fragments, producing a β-substituted angiotensin peptide MU37 and comparing its efficiency as an anti-fibrotic agent to C21. Both reagents produced a dose-dependent reduction in TGFβ-induced collagen and αSMA expression, and significantly reduced high salt-induced cardiac fibrosis [173].

Targeting RAAS in Skin Fibrosis

As noted above, skin expresses local RAAS components that may undergo dysregulation during wound healing and in skin pathologies, resulting in the formation of hypertrophic scars and keloids. Blocking ACE/AT1R and/or activating ACE2/AT2R may thus be beneficial in these conditions. Deletion of ACE in mice or treatment with the ACE inhibitor ramipril reduced scar formation after full thickness skin wounding [145]. AT1R inhibition with losartan significantly reduced scar formation, but took more time than ramipril. ACE inhibition resulted in decreased fibroblast proliferation, and decreased collagen and TGFβ₁ expression [145]. A pilot singleblinded placebo-controlled clinical trial was designed to test the effect of topical losartan on patients with hypertrophic scars and keloids [174]. Results of this study showed that scar scores were significantly reduced in the losartan group compared to placebo, suggesting that AT1Rselective antagonism using losartan 5% can alleviate the keloid and hypertrophic scar [174].

In mouse and porcine models of chronic diabetic wounds, and in aging diabetic pigs, application of 1% losartan gel resulted in faster wound healing and enhanced the quality of healing skin compared to other tested medications and controls [175]. The roles of AT1R and AT2R in skin healing are phase-dependent, and it is the late application of AT1R blocker that enhanced the wound closure time and the quality of healing. Notably, AT2R deletion abolished the effects of losartan on wound healing through AT1R, which may explain why ACE inhibition failed to produce beneficial effects similar to losartan. That is, ACE inhibitors reduce Ang II production resulting in a lack of stimulation of both AT1R and AT2R while losartan selectively blocks AT1R allowing unopposed binding of AT2R.

In a xenograft model of Dupuytren’s Disease, C21 significantly reduced fibrotic markers such as CCN2, TGFβ, Smad3,Smad4and αSMA both in vitro and in vivo [176]. C21 may thus represent a novel therapeutic for this condition, which can be challenging to treat, and suggests that RAAS modulation in general may be a useful strategy in this setting.

Conclusion

Organ fibrosis is associated with increased mortality and morbidity worldwide. A large body of literature has demonstrated that, in the pathogenesis of fibrosis in different systems of the body, including the liver, heart, kidney, skin and others, TGFβ signaling represents a central mechanism. The RAAS pathway interacts with TGFβ at multiple levels to enhance or retard tissue fibrosis. The classical RAAS pathway, including renin, ACE, Ang II, AT1R and aldosterone, acts to exacerbate fibrosis. Conversely, ACE2, Ang-(1–7), AT2R and MasR, constituting the alternative RAAS pathway, act to ameliorate and prevent fibrosis.

Although targeting the RAAS pathways has been shown to be a promising approach to combat fibrosis, more investigation is needed to precisely understand the RAAS molecular and cellular mechanisms at different stages of the disease in various body organs, particularly since the timing of any intervention in RAAS signaling may be critical to success. Moreover, additional clinical studies are needed to confirm RAAS-based treatment effectiveness in patients at various stages of disease progression, and to establish the long-term safety of this approach in treating different forms of fibrosis, since such strategies are likely to involve extended periods of treatment.

Highlights.

• The classical renin-angiotensin-aldosterone system (RAAS) has been implicated as an inducer of fibrosis in multiple tissue and organ types.

• Recent evidence demonstrates that the alternative RAAS counteracts the classical RAAS to attenuate fibrosis.

• Augmentation of the alternative RAAS, or altering the balance of the classical and alternative RAAS, holds promise for the therapeutic treatment of fibrosis.