RAS inhibition in resident fibroblast biology

Abstract

Angiotensin II (Ang II) is a primary mediator of profibrotic signaling in the heart and more specifically, the cardiac fibroblast. Ang II-mediated cardiomyocyte hypertrophy in combination with cardiac fibroblast proliferation, activation, and extracellular matrix production compromise cardiac function and increase mortality in humans. Profibrotic actions of Ang II are mediated by increasing production of fibrogenic mediators (e.g. transforming growth factor beta, scleraxis, osteopontin, and periostin), recruitment of immune cells, and via increased reactive oxygen species generation. As a result, drugs that inhibit Ang II production or action, collectively referred to as renin angiotensin system (RAS) inhibitors, are first line therapeutics for heart failure. Moreover, transient RAS inhibition has been found to persistently alter hypertensive cardiac fibroblast responses to injury providing a useful tool to identify novel therapeutic targets. This review summarizes the profibrotic actions of Ang II and the known impact of RAS inhibition on cardiac fibroblast phenotype and cardiac remodeling.

1. Introduction

Cardiac pathology as a product of acute or chronic injury typically manifests in cardiac fibrosis mediated by cardiac fibroblasts. The classical renin angiotensin system (RAS) positively regulates the progression of fibrosis by signaling via angiotensin receptors in the heart. Angiotensin II (Ang II) is a vasoactive peptide with systemic and local effects on the heart via angiotensin type 1 (AT1) and type 2 (AT2) receptors. Angiotensinogen is cleaved by the enzyme renin, to form angiotensin I, which in the presence of angiotensin converting enzyme (ACE) or chymase, is further cleaved to produce angiotensin II. ACE inhibitors and AT1 receptor antagonists are known to both reduce blood pressure and minimize cardiac fibrosis following injury. Cardiac tissues have been shown to uptake extracellular angiotensinogen and renin^1,2, and increase expression of both following injury^3,4. The presence of ACE has been well-established in the normal⁵ and remodeling heart^6,7 allowing for the generation of Ang II. Thus, Ang II elicits cardiac effects by signaling from the circulation⁸ as well as by local production of Ang II⁹. In the heart, Ang II induces cardiomyocyte hypertrophy and cardiac fibroblast activation which together potentiate cardiac fibrosis and dysfunction. Angiotensin-(1–7) (Ang (1–7)) is an alternative product of Ang I by the action of neprilysin or of Ang II by ACE2. Ang (1–7) antagonizes the effects of Ang II via actions at the Mas receptor¹⁰ and as a systemic vasodilator¹¹.

The RAS plays a primary role in the long-term control of arterial pressure. Dysregulation of the RAS systemically can result in hypertension. Excess local activation of the RAS can result in tissue remodeling, including fibrosis in the heart¹² and other tissues¹³. All the necessary components for the generation of Ang II can be found in the left ventricle (LV)^14,15. Moreover, cardiac fibroblasts also express genes for renin, ACE, angiotensinogen, as well as the AT1 and AT2 receptors^4,16–21. Angiotensin II also stimulates aldosterone release from the adrenal cortex, which promotes fibrosis in the heart via the mineralocorticoid receptor. As such, aldosterone antagonists have also shown to be effective in attenuating cardiac fibrosis¹³. The cardioprotective effects of RAS inhibition are therefore linked to both cardiomyocyte and cardiac fibroblast specific signaling. Fibroblast activation is directly reduced while diminished cardiomyocyte hypertrophy removes additional stimuli. ACE inhibitors and angiotensin receptor blockers (ARBs) are standard of care to slow the development of heart failure induced by chronic hypertension and acute MI and have been associated with reduction in cardiac fibrosis and mortality. More recent combination therapies are proving to be more effective than these monotherapies. Thus, a better understanding of downstream targets in the RAS will help elucidate therapeutic targets that may be more proficient in preventing heart failure.

The pathological impact of Ang II is primarily mediated by AT1 receptor agonism, while AT2 receptor activation can attenuate the fibrogenic signaling²². The relative contribution of AT1 vs. AT2 receptors in mediating fibrosis has not been fully elucidated. AT2 receptor knockout mice have been shown to have lower cardiac collagen content and reduced collagen 1a1 gene expression²³, and in response to Ang II infusion, AT2 receptor knockout mice exhibited attenuated collagen deposition²⁴. In contrast, AT1aR deletion alone²⁵ or in combination with overexpression of AT2R, resulted in a greater reduction in fibrosis²⁶. Given that each of these studies were performed in global knockout mice, the relative contribution of blood pressure changes or cardiomyocyte-fibroblast crosstalk is not known. The development of the TCF21-cre and POSTN-cre lines for fibroblast or activated fibroblast-specific knockouts would be important for future studies to identify the importance of fibroblast-specific RAS signaling in the development of fibrosis. Crosstalk between fibroblasts and cardiomyocytes includes pro-inflammatory, pro-death, and activating signals. Whether the injury is slow and progressive or sudden, fibroblasts respond to apoptosis and necrosis of cardiomyocytes which in turn triggers fibroblast activation. Both fibroblasts and cardiomyocytes contribute to the inflammatory milieu, communicate with immune cells, and exacerbate fibrotic remodeling. The details of cardiac cell crosstalk during cardiac remodeling has been reviewed in depth elsewhere^27,28.

Cardiac fibroblast activation and transdifferentiation into extracellular matrix (ECM)-producing cells, often referred to as myofibroblasts, is the cornerstone to fibrotic remodeling of the heart²⁹. The structural ECM includes collagen, fibronectin, proteoglycans, glycoproteins and integrins. Activated fibroblasts also secrete modulatory matricellular proteins that transduce inflammatory, growth, and fibrogenic signals. The roles of key matricellular proteins periostin, osteopontin, thrombospondins, tenascin-C, and CCN1-CCN6 are reviewed here and elsewhere³⁰. Markers of cardiac fibroblast activation continue to be refined and remain difficult to determine given low fibroblast specificity and high fibroblast heterogeneity. Thus, the activated fibroblast is currently best determined by a collection of markers which include ɑ-smooth muscle actin (ɑ-SMA), periostin, fibronectin, and collagens I and III among others³¹. Some effector pathways downstream of AT1 receptor are mitogen-activated protein kinases (MAPK) including extracellular signal-regulated 1/2 (ERK1/2), p38 MAPK, and AKT, which modulate proliferation. Additionally, Ang II induces reactive oxygen species (ROS) production and upregulation of ECM components mainly via transforming growth factor beta (TGF-β) signaling. This review focuses on cardiac-specific RAS signaling and direct effects of Ang II and RAS inhibition on cardiac fibroblast responses.

2. Ang II and cardiac fibroblast activation

Ang II infusion in rodent models produces a phenotype of hypertension, cardiac hypertrophy and cardiac fibrosis^32–37. In addition, production of local Ang II has been shown to increase in response to myocardial infarction and play a major role in the wound healing response^38,39. Thus, chronic elevations of Ang II and acute increase in local Ang II following an ischemic event each result in fibroblast activation and expansion. Studies that investigate the direct impact of Ang II on cardiac fibroblasts reveal increased measures of activation including proliferation and fibrogenic gene and protein expression. Ang II-induced fibroblast activation and associated ECM production is mediated by fibrogenic signaling molecules, inflammation, and oxidative stress (Figure 1) as outlined in this section. Additionally, epigenetic modulation of gene expression is influenced by the RAS and therefore increases the complexity of Ang II actions on cardiac fibroblast activation. Current and future investigations utilizing single cell RNA-sequencing are working to elucidate novel targets for therapy that range outside of general RAS inhibition.

Figure 1. RAS-mediated mechanisms of cardiac fibroblast activation.

Angiotensin II (Ang II) binds angiotensin type 1 receptors (AT1R) on the fibroblast to initiate three key processes that result in fibroblast activation: fibrogenic gene upregulation, immune cell recruitment, and oxidative stress. Crosstalk between components of each pathway supports redundant feed forward mechanisms that together with cell death resistance perpetuate activated fibroblasts with cardiac injury. Transforming growth factor-beta (TGF-β), scleraxis (SCX), signal transducer and activator of transcription 3 (STAT3), extracellular signal-regulated kinase 1/2 (ERK1/2), cellular communication network factor 2 (CCN2), osteopontin (OPN), periostin (POSTN), alpha-smooth muscle actin (α-SMA), toll-like receptor 2 (TLR2), nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), interleukin-6 (IL-6), monocyte chemoattractant protein-1 (MCP-1), NADPH oxidase (NOX4), reactive oxygen species (ROS), B-cell lymphoma 2 (Bcl-2). Created with BioRender.com.

2.1. Pro-Fibrogenic actions of Ang II-mediated signaling molecules

Ang II has been widely shown to increase TGF-β signaling in the cardiac fibroblast via AT1 receptor stimulation. The pro-fibrotic effects of Ang II pay be mediated in large part by TGF-β signaling. Whole animal Ang II infusion and in vitro stimulation of cardiac fibroblasts with Ang II induces increases in TGF-β mRNA and protein^40–42. Activated fibroblasts produce TGF-β, which then can act in an autocrine capacity to bind TGF-β receptors on the cell surface⁴³ and signal through ERK1/2 and Smad proteins^41,44. TGF-β-mediated ERK1/2 activation plays a large role in cardiac fibroblast proliferation for expansion of ECM-producing cells in the remodeling heart. Binding of TGF-β also leads to phosphorylation of Smad2 and Smad3, which transfer to the nucleus to increase transcription of both pro-inflammatory and ECM proteins via a process that involves several accessory and modulatory proteins.

TGF-β/Smad3 signaling has been linked to several signaling partners that work to maintain the fibrogenic capacity of fibroblasts. Some more recent investigations have highlighted a role of signal transducer and activator of transcription 3 (STAT3) in maintaining Smad signaling that is initiated in a TGF-β independent manner. JAK/STAT signaling occurs downstream of AT1 receptor⁴⁵ and toll-like receptor 4 (TLR4)⁴⁶ in the heart. Ang II infusion upregulates STAT3 activity and is linked with cardiac fibrosis via thrombospondin 1⁴⁷. In the heart, loss of STAT3 prevents Ang II-induced upregulation of fibrogenic TGF-β and collagen⁴⁶ demonstrating the critical role of TGF-β in potentiating fibrotic remodeling. Upstream inhibition of STAT3 activity by global TLR4 knockout⁴⁶ or direct pharmacological inhibition⁴⁸ protects mice from fibrosis induced by Ang II. The STAT/Smad3 association has also been shown to be maintained by the transmembrane ligand, EphrinB2. This ligand is elevated co-incident with fibrosis after 4 weeks of Ang II infusion in mice and in vitro treatment with EphrinB2 stimulates cardiac fibroblast activation⁴⁹. Further, STAT3-mediated protection of mitochondria prolongs fibroblast activation⁵⁰. Other agents that inhibit the actions of TGF-β also suppress proliferation and fibrotic signals in models of Ang II infusion in vivo⁵¹ and in vitro in cardiac fibroblasts^52–56.

TGF-β mediated Smad signaling was shown to be modulated by microRNAs (miRNA). Ang II-mediated pressure overload in mice was shown to be linked to repression of miRNA-221/222 resulting in unrestrained TGF-β signaling⁵⁷. Protective actions of Klotho include suppression of pro-fibrotic miRNA-132 downstream of Ang II-induced Smad activation⁵⁸. Four weeks of Ang II suppresses miRNA-1297-mediated inhibition of ULK1 with potential reduction in autophagy, which may contribute to cardiac fibroblast activation⁵⁹. Cardiac fibroblast proliferation mediated by cyclin D2 and cyclin-dependent kinase 6 (CDK6) is blunted by miRNA-1, which was shown to be downregulated in cardiac fibroblasts from Ang II-infused rats⁶⁰. On the other hand, miRNA-96 promotes Ang II-induced proliferation and ECM production in murine cardiac fibroblasts⁶¹.

TGF-β stimulates expression of the profibrotic transcription factor scleraxis via both Smad-dependent and Smad-independent mechanisms^62,63. In turn, scleraxis is critically required for various aspects of TGF-β/Smad signaling in the transdifferentiation of cardiac fibroblasts, including the TGF-β-dependent up-regulation of collagen Iɑ2 and ɑ-SMA, as well as myofibroblast contraction^63,64. Further, the production of fibronectin⁶⁵ and MMP2⁶⁶ is prevented in cardiac fibroblasts isolated from scleraxis null mouse hearts. Scleraxis null cardiac fibroblasts fail to undergo transdifferentiation to myofibroblasts in response to TGF-β or mechanical stretch^64,67. We and others have reported that Ang II increases scleraxis expression in cardiac fibroblasts^68,69. Our preliminary data indicates that Ang II-mediated up-regulation of ColIα1, ColIα2 and ColIIIα1 is attenuated by scleraxis knockdown, implicating scleraxis as an effector of Ang II fibrotic signaling (M. Czubryt, personal observation).

Periostin and osteopontin are two downstream products of Ang II-mediated fibroblast activation. Periostin is increased after MI and following Ang II infusion^41,47,70,71. Genetic lineage tracing of cardiac fibroblasts has confirmed that activated fibroblasts express periostin and arise from resident cardiac fibroblasts of the Tcf21 lineage and that these cells modulate adaptive healing and fibrosis⁷². Periostin depletion can cause cardiac rupture with MI⁷³, but limiting the persistence of periostin-expressing activated fibroblasts during 4 weeks of Ang II infusion or 3 weeks of MI attenuates fibrosis and improves cardiac function⁷⁰. Both Tcf21 and periostin promoter-driven therapeutic interventions have been used during Ang II infusion^74,75 to attenuate collagen deposition – suggesting a causal role for periostin in Ang II-induced fibrosis.

Osteopontin is associated with fibrogenic signaling but is also involved in the pro-inflammatory status of activated fibroblasts. The Ang II-induced increase in osteopontin, IL-6, and monocyte chemoattractant protein 1 (MCP-1) gene expression in mouse cardiac fibroblasts is prevented in cells from osteopontin knockout mice⁷⁶. These reductions in gene expression correlate with an attenuation in cardiac fibrosis linking the fibrogenic effects of osteopontin induction with pro-inflammatory signaling. In another study, osteopontin knockout mice exhibit similar increases in ECM gene expression, but reduced adhesion and proliferation which attenuates interstitial cardiac fibrosis in response to Ang II compared to wildtype animals⁷⁷.

2.2. Role of inflammation and immune cell infiltration in Ang II-dependent fibrosis

In addition to their role in promoting ECM production, activated fibroblasts initiate and perpetuate inflammation, which in turn exacerbates fibrosis in hypertensive hearts⁷⁸. Cardiac fibrosis that occurs with Ang II infusion is dependent on bone marrow-derived inflammatory cells. Macrophages are recruited to the heart mainly via MCP-1. Macrophage infiltration into the heart peaks after 3 days of Ang II infusion in rodents and prevention of this influx limits fibrosis⁷⁹. Direct effects of macrophages on fibroblast phenotype have been demonstrated using co-culture experiments. Specifically, macrophages stimulate fibroblasts to produce IL-6 which induces TGF-β-mediated Smad activation in the cardiac fibroblast⁸⁰ as well as increased expression of collagen and ɑ-SMA⁷⁹. Suppression of IL-6 in vitro also attenuates Ang II-induced TGF-β expression and downstream production of collagen and ɑ-SMA by fibroblasts⁸¹. Furthermore, using the in vivo model of Ang II infusion, TGF-β-mediated ECM and inflammatory signals are attenuated in MCP-1³⁴ and IL-6⁸⁰ knockout mice. However, hypertension and cardiac hypertrophy are not impacted by loss of MCP-1, differentiating the systemic hemodynamic response and hypertrophy from the fibrotic response³⁴. Loss of tumor necrosis factor alpha (TNFɑ) receptor results in a reduction of cardiac fibrosis after 1 week of Ang II infusion in mice⁸². The relationship between Ang II and TNFɑ signaling is tightly linked to ROS production and reviewed here⁸³. IL-6 and TNFɑ are increased by nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB)-mediated transcription downstream of toll-like receptor (TLR) stimulation. TLR2 knockout mice exhibit a marked attenuation in macrophage infiltration and the associated fibrosis upon Ang II infusion⁸⁴. Given that Ang II mainly upregulated TLR2 expression in innate immune cells, this finding supports a main role for macrophages and neutrophils in the fibrotic response to Ang II.

2.3. Oxidative stress promotes fibrosis downstream of AT1 receptor activation

Oxidative stress is a major inducer of hypertension-related fibrosis and is linked to TGF-β signaling⁸⁵. Ang II induces ROS generation, which stimulates fibroblast activation. NADPH oxidase (NOX) 1, 2, and 4 are enzymes that produce superoxide and hydrogen peroxide in times of cell stress and are expressed in both cardiomyocytes and cardiac fibroblasts⁸⁶ which exacerbates cardiac fibrosis⁸⁷. Of all NOX isoforms, cardiac fibroblasts primarily express NOX4⁸⁸. Ang II promotes increased NOX4 expression and cardiac overexpression of NOX4 worsens Ang II-induced cardiac hypertrophy and fibrosis⁸⁷. The impact of NOX4 modulation likely depends on the duration of Ang II infusion. The absence of NOX4 does not alter cardiac fibrosis after 2 weeks of Ang II infusion⁸⁹, but suppression of NOX4 protein and activity does attenuate fibrosis when assessed after 4 weeks of Ang II infusion in rodents⁸⁶. Knockdown or inhibition of hydrogen peroxide-producing NOX4 attenuates proliferation, migration, and collagen secretion induced by Ang II in adult mouse cardiac fibroblasts⁹⁰. Further, physical interaction of NOX4 with the AT1 receptor is enhanced by Ang II⁹⁰. NOX4-dependent cardiac fibroblast migration and collagen production was shown to be mediated by Ang II-induced downregulation of regulator reversion-inducing-cysteine-rich protein with Kazal motifs (RECK) and subsequent unrestrained MMP expression⁹¹.

AT1 receptor/NOX mediated ROS production and fibrogenic gene expression (collagen I, collagen III, ɑ-SMA, TGF-β) is inhibited by a cardioprotective gasotransmitter, hydrogen sulfide (H₂S) in isolated neonatal cardiac fibroblasts⁹². In addition, H₂S has been shown to reduce fibrosis in hypertensive rats in vivo⁹², as well as to blunt Ang II-induced upregulation of NOX4⁹³. To combat the elevated ROS, antioxidants are initially increased in cardiac tissue after 24 hours of Ang II infusion, but this was not evident after 2 weeks - suggesting a reduced capacity to neutralize free radicals with prolonged RAS activation⁹⁴. Longer term Ang II infusion with antioxidant co-treatment attenuates fibroblast activation and fibrosis⁸⁵. Thus, agents that neutralize ROS reduce Ang II-induced fibroblast proliferation, migration, and activation. Various ROS scavengers and antioxidants including N-acetyl cysteine⁹⁵, tempol⁹⁰, propofol/alpha-tocopherol⁹⁶, and combined antioxidant therapy⁹⁷ prevent Ang II-induced collagen production and ɑ-SMA expression in cardiac fibroblasts, TGF-β activity, and fibroblast proliferation. Ang (1–7) also attenuates Ang II-induced ROS and fibrosis⁹⁸.

Ultimately, ROS production downstream of AT1 receptor binding is mediated by NOX4-dependent ERK1/2 signaling. Thus, modulation of Ang II, ROS, or other endogenous regulators of the ERK1/2 pathway have an impact on fibrogenic signaling of cardiac fibroblasts. Targets of interest for ERK1/2 range from simple reduction in direct ROS scavenging to more interconnected alterations in ECM components and regulators. Importantly, there is a feed forward mechanism of Ang II and ROS production, which sustains expression and activity of enzymes like NOX4. In addition to the non-canonical responses to ROS, fibroblasts are known to be resistant to apoptosis due to constitutive expression of anti-apoptotic Bcl-2⁹⁹. Recently, Ang II activation of ERK1/2/RSK1 downstream of AT1 receptor stimulation led to inhibition of caspase-3 cleavage in rat and human cardiac myofibroblasts, adding to the anti-apoptotic properties of activated fibroblasts¹⁰⁰. This combination of ROS-induced activation and resistance to cell death is key to the persistence of ECM-producing fibroblasts in the hypertensive and injured heart.

2.4. Epigenetic modulation by the RAS

In additional to direct modulation of signaling cascades, Ang II has been shown to induce epigenetic alterations of DNA and histones by methylation and acetylation to influence expression of pro-fibrotic genes^101,102. Moreover, fibroblast transdifferentiation into myofibroblasts may also be regulated epigentically¹⁰³. Histone acetylation promotes relaxation of DNA and increased transcription of genes. Thus, histone deacetylases (HDAC) condense DNA resulting in decreased gene transcription. Ang II¹⁰⁴ and hypertension¹⁰⁵ increase HDAC activity, while HDAC inhibitors attenuate hypertrophy, fibrosis and cardiac dysfunction in Ang II-infused mice^106,107. Inhibition of Ang II-induced HDAC activity attenuates fibrogenic signaling via MMP9/RECK and inhibits the production of ECM proteins fibronectin and collagen in murine cardiac fibroblasts¹⁰⁴.

Histone methylation is associated with both repression and activation of gene transcription, depending on the specific location and number of methyl groups on the various histones. However, Ang II¹⁰⁸ has been shown to increase levels of histone demethylases in the heart in association with fibrogenic gene activation. Further, the expression of the ACE gene is increased in spontaneously hypertensive rats (SHR) and the ACE promoters are mostly unmethylated¹⁰⁹. It is known that pressure overload and in vitro stimulation of cardiac fibroblasts with Ang II induces activation of the pro-fibrotic Wnt/β-catenin pathway¹¹⁰. Increased expression of the demethylase JMJD3 was recently linked with Ang II-mediated β-catenin and ECM production in cardiac fibroblasts¹¹¹. Demethylation of histones was also recently shown to be controlled by metabolic shifts that occur during fibroblast transdifferentiation,¹¹² further implicating histone methylation in cardiac fibroblast activation. Methylation of DNA within the promoter region of genes by DNA methyltransferases (DNMT) leads to inhibition of specific fibrogenic genes. For example, TGF-β induced ɑ-SMA occurs via inhibition of DNMT1 in cardiac fibroblasts and is correlated with reduced DNMT1 expression in the infarcted heart¹¹³. DNA hypomethylation is associated with hypertension¹¹⁴ and heart failure¹¹⁵. The epigenetic regulation of cardiac fibrosis has been reviewed in depth elsewhere^116,117.

2.5. Single cell RNA-sequencing: Identifying the impact of Ang II on cardiac fibroblasts and novel therapeutic targets

Technological advances in single cell transcriptomics have identified multiple cardiac fibroblast populations and fibroblast-specific changes in gene expression in the remodeling heart. Cardiac fibroblasts treated in vitro with Ang II exhibit upregulation of 126 genes compared to control with IL-1β and MMP3 being identified as potential therapeutic targets¹¹⁸. In a study of cardiac cells from control vs. Ang II infused mice, single cell RNA sequencing analysis identified nine subpopulations of cardiac fibroblasts. Two of these subpopulations were primarily composed of fibroblasts from Ang II-infused hearts¹¹⁹. These two pathological subpopulations were defined by the expression of either Cilp or Thbs4 and were characterized by increased cell adhesion, wound healing, and ECM production compared to other fibroblast subpopulations. Further, the abundance of Thbs4 expressing fibroblasts was almost doubled in females compared to males among other gene changes that may underlie altered ECM quality and remodeling between the sexes. Another study showed that genetic deletion of Fstl1, an Ang II-responsive gene¹²⁰, in cardiac fibroblasts increased cardiac rupture and mortality¹²¹. Proliferative and fibrogenic gene expression in periostin-traced cardiac fibroblasts after combination Ang II/phenylephrine infusion are reversed after a 2-week washout of the treatment demonstrating the capacity to downregulate these processes after injury resolution⁷². Several single cell assessments of cardiac cells, including fibroblasts, have been performed following MI. A comprehensive study evaluated a time course of MI in mice from diverse genetic backgrounds and found that mice prone to hypertension (mouse strain 129) exhibited an increased predisposition to cardiac rupture in the proliferative phase of wound healing¹²². The number of fibroblast subpopulations remained consistent, but the proportion of cells within the myofibroblast subpopulations was increased in mouse strain 129. Ultimately, the altered ratio of subpopulation composition resulted in increased activation and decreased fibrinolysis gene expression, which increased the rate of cardiac rupture¹²². As Ang II signaling is implicated in the pathology of MI, markers of activated fibroblasts in normotensive hearts may reveal novel targets for RAS inhibition as well. Gladka et al. identified Ckap4 as a novel marker of activated fibroblasts following ischemia/reperfusion injury, a marker which was modulated by TGF-β in vitro¹²³. Farbehi et al. identified three myofibroblast subsets that emerge from activated fibroblasts 7 days post-MI¹²⁴. Two myofibroblast populations expressed genes associated with fibrogenic pathways with markers of Scx and Thbs4¹²⁴. The genetic makeup and relative proportion of cardiac fibroblast subpopulations holds valuable information for exacerbated cardiac fibrosis in hypertension and cardiac injury.

3. RAS inhibition alters cardiac fibroblast phenotype and limits fibrosis

As described above, Ang II stimulates fibrogenic and pro-inflammatory cascades in the cardiac fibroblast that fuel fibrotic remodeling. Thus, the inhibition of the production or actions of Ang II are known to attenuate adverse cardiac remodeling. Though the impact of RAS inhibition encompasses hemodynamic alterations and other tissue-specific benefits¹³, this section focuses on the known effects of RAS inhibition on cardiac fibrosis.

3.1. RAS inhibition and fibrosis in hypertensive heart disease

ACE inhibitors^125–133, AT1 receptor antagonists^134–137, and aldosterone antagonists¹³⁴ have been shown to attenuate or reverse cardiac fibrosis in experimental models. Given the direct effects of Ang II on cardiac fibroblasts, it is perhaps not surprising that drugs that inhibit the production or action of this peptide would be effective in reducing fibrosis. Associated with RAS inhibition is an increase in collagenase activity, decreased collagen production and inhibition of TGF-β signaling^138–140. In cardiac fibroblasts, activation of ERK1/2, p38 MAPK, and JAK/STAT pathways by Ang II are reduced by ACE inhibition¹⁴¹. Downstream of similar signaling pathways, periostin⁷¹ is upregulated in cardiac fibroblasts stimulated with Ang II, which is also shown to increase CCN2 (previously known as CTGF)¹⁴², both of which are attenuated by AT1 receptor blockers (ARBs). In addition to suppressing Ang II production, ACE inhibitors also inhibit hydrolysis of N-acetyl-seryl-aspartyl-lysyl-proline (Ac-SDKP)¹²⁹ thus increasing plasma Ac-SDKP levels which reduces inflammatory cell infiltration and fibrotic remodeling¹⁴³. Ac-SDKP has been shown to inhibit TGFβ-induced fibroblast activation, fibroblast proliferation, and collagen production - independently of Ang II^144,145. Since fibroblasts have been shown to express the requisite components for Ang II production as well as the angiotensin receptors^18,146, targeting this local RAS has been a proposed reason for the consistent efficacy of these drugs to attenuate or reverse cardiac fibrosis. Recently, FDA-approved angiotensin receptor-neprilysin inhibitors (ARNI), a combination of ARB/neprilysin inhibitors, not only antagonize AT1 receptors but also prevent the breakdown of vasodilatory natriuretic peptides¹⁴⁷. Treatment with ARNIs during pressure overload produces superior cardioprotection than that achieved with either agent alone and is linked with PKG-dependent inhibition of RhoA activation in cardiac fibroblasts¹⁴⁸.

3.2. Antihypertensive agents and cardiac fibroblast apoptosis

Hypertension is associated with fibroblast hyperplasia in the hypertrophied LV.^149,150 ACE inhibitors, AT1 receptor antagonists, and calcium channel blockers (CCBs), when administered to adult male SHR, each induce a burst of apoptosis within the first 2 weeks of treatment that results in a 30% loss of LV fibroblasts^149–154. In addition, AT1 receptor blockade has been shown to induce cardiac fibroblast apoptosis associated with reduced collagen deposition in a model of nitric oxide synthase (NOS) inhibitor-induced cardiac injury¹⁵⁵. Similar treatment does not induce apoptosis of fibroblasts in the right ventricle, nor in the LV of normotensive WKY rats,¹⁵⁴ suggesting that the excess fibroblasts that are present in the hypertrophied LV may be particularly susceptible to RAS inhibitor and CCB-induced cell death. Moreover, these drugs also protect against cardiomyocyte^156–160 and endothelial cell^159,161,162 death, suggesting LV fibroblast-specific effects.

The mechanisms by which RAS inhibitors and CCBs each induce cardiac fibroblast apoptosis are unknown, however the extrinsic (i.e. death receptor) pathway is frequently implicated in other cell types^163–166. For example, increased levels of TNFα-related apoptosis inducing ligand (TRAIL) accompanied ACE inhibitor-induced apoptosis in intimal cells.¹⁶⁴ Fas associated death domain (FADD) and tumor necrosis factor receptor type 1-associated death domain (TRADD) were both upregulated in response to ACE inhibition in apoptotic erythroid precursor cells.¹⁶⁵ In addition, caspase-8 was upregulated during aortic vascular smooth muscle cell (VSMC) apoptosis in response to CCB treatment¹⁶⁷. The molecular differences that render these subsets of cells uniquely susceptible to apoptosis by these antihypertensive drugs with distinct mechanisms of action have yet to be identified.

The stimulus that leads RAS inhibitors and CCBs to induce cell death also remains elusive. Fibroblast apoptosis appears to be independent of blood pressure lowering effects of these drugs, as beta blockers and hydralazine were unable to produce cell death despite effective antihypertensive response¹⁵⁴. It may be that the fibroblasts of the hypertrophied LV are particularly dependent upon Ang II and calcium, and a loss of these survival factors triggers cell death. Given that fibroblast apoptosis precedes regression of LV hypertrophy, it is unlikely that this cell loss is merely secondary to a reduction in mass¹⁵³. Interestingly, the spatiotemporal association of fibroblast apoptosis and regression of cardiomyocyte hypertrophy suggests that the fibroblasts may play a role in supporting the increased cardiomyocyte size¹⁵³. This is similar to more recent studies in which a 50% loss of cardiac fibroblasts following germline deletion of scleraxis was accompanied by a ~30% decrease in cardiomyocyte size⁶⁴.

During angiotensin receptor blockade, selective antagonism of the AT1 receptor leaves the AT2 receptor unopposed. While AT2 receptor stimulation has been shown to have growth inhibitory and pro-apoptotic effects, combined AT1 and AT2 blockade was unable to prevent valsartan-induced apoptosis of interstitial left ventricular fibroblasts¹⁵². During ACE inhibition, there is a prolongation of the half-life for bradykinin. Activation of bradykinin B1 and B2 receptors has also been proposed as a potential mechanism for ACE inhibitor-induced fibroblast apoptosis. Although this has not been tested in the LV, evidence from aortic VSMC suggests that while bradykinin B2 receptor activity did not block ACE inhibitor-induced apoptosis, there may be a role for the bradykinin B1 receptor¹⁶⁸. In addition, the ACE inhibitor enalaprilat has been shown to prevent Ang II-induced proliferation as well as ROS/p38 activation in cardiac fibroblasts, in vitro¹⁶⁹. Future studies aimed at identifying the mechanisms by which these antihypertensive agents induce apoptosis of LV fibroblasts may reveal a novel target for antifibrotic therapies.

3.3. Long-term consequences of transient RAS inhibition on cardiac fibroblasts

Inhibitors of the RAS produce a reduction of arterial pressure and LV hypertrophy that persists even after stopping treatment in adult male SHR^14,170–174. Moreover, RAS inhibitors have been shown to be effective in producing long-term resistance to aging-induced cardiac fibrosis - an effect that persisted up to 15 months after stopping treatment^{134–137,175}. In addition, a 2-week treatment of adult male SHR with the ACE inhibitor enalapril conferred protection against a future insult induced by NOS inhibition^{14,170,175,176}. Specifically, rats that were previously treated with an ACE inhibitor showed a reduction in fibrosis, inflammation, and cellular proliferation in response to NOS inhibition^14,176. This protection does not appear to be secondary to a reduction in cardiomyocyte loss. In fact, for a given area of infarction there were fewer macrophages infiltrating the hearts of previously-treated rats.¹⁴ This protection against NOS inhibitor-induced pathological remodeling was found to be related to a phenotypic difference in the cardiac fibroblasts following transient ACE inhibition. Cardiac fibroblasts isolated from SHR transiently treated with an ACE inhibitor had a lower proliferation rate, collagen I gene expression, and secretion of chemokines when compared to fibroblasts isolated from naive SHR that were treated with a NOS inhibitor¹⁷⁷. Taken together, inhibition of the RAS produces persistent effects in the cardiac fibroblasts leading to a less fibrogenic phenotype. Whether this is a result of a loss of a subset of fibrogenic fibroblasts during treatment, or epigenetic changes in the fibroblasts during treatment, is unknown.

3.4. RAS inhibition and protection from myocardial infarction

With MI, the ischemic area of the heart undergoes massive tissue necrosis and the fibrotic remodeling that occurs is necessary for the repair of the heart. Regions remote to the injury also accumulate interstitial fibrosis that contributes to myocardial stiffness and ultimately heart failure. Activation of the RAS occurs with MI^178–181 and therefore inhibitors of the RAS have been shown to slow cardiac remodeling and improve cardiac function. Despite no change in infarct size, ACE inhibition improved survival and decreased LV hypertrophy and remote fibrotic remodeling in normotensive^182–187 and hypertensive hearts¹⁸⁸. The protection in SHRs was associated with a reduced capacity to induce ventricular arrhythmias¹⁸⁸. Interestingly, treatment of SHRs with either ramipril or losartan reduces LV fibrosis, but not collagen mRNA suggesting that the protection occurs post-translationally¹⁸⁹. Reduction of oxidative stress¹⁸⁶ and improved antioxidant activity¹⁹⁰ have been associated with ACE inhibitor-mediated protection. A longer-term study found that 25 weeks after MI, collagen and CCN2 protein were elevated, but only collagen was reduced by captopril implying that CCN2 is an important regulator of late cardiac remodeling independent of RAS inhibition¹⁸⁴. The combination of an ACE inhibitor with resveratrol¹⁹⁰, bradykinin¹⁹¹, soluble guanylyl cyclase activator¹⁹², or mineralocorticoid receptor antagonist¹⁹³ further suppresses collagen deposition.

Like ACE inhibitors, ARBs do not reduce the size of the infarct and therefore also alter remote cardiac remodeling and hypertrophy to improve outcomes¹⁹⁴. Mice with MI that were treated with valsartan exhibited reduced hypertrophy, fibroblast proliferation and fibrosis¹⁹⁵. Olmesartan administration to mice with MI prevented cardiac rupture due to attenuated periostin expression¹⁹⁶. The ARB, telmisartan, is also a peroxisome proliferator-activated receptor-gamma (PPAR-γ) activator, which further reduces hypertrophy, fibrosis, and macrophage infiltration to improve hemodynamic function compared to losartan. Moreover, isolated cardiac fibroblasts pretreated with telmisartan produce less MMP2, MMP9, CCN2, and osteopontin, but more TIMP1 then losartan treated fibroblasts in hypoxic conditions¹⁹⁷. Elevated CCN2 in failing hearts after MI is attenuated with losartan¹⁴².

Most studies find equivalent protection afforded by either ACE inhibitors or ARBs, suggesting that the reduction in Ang II is the primary mediator of the beneficial effects of these interventions. However, one study found that post-MI rats had greater reductions in fibroblast proliferation and LV collagen content with enalapril compared to losartan with similar reductions in hypertrophy¹⁹⁸, while another reported treatment post-MI with valsartan reduced collagen, immune cell infiltration, and TGF-β in the LV to a greater extent than fosinopril¹⁹⁹. Both studies evaluated 2 weeks post-MI in normotensive rats that received therapeutic intervention 24 hours after surgery, therefore the differences in findings may be due to drug or dose, rather than class-specific findings. Similar to findings in pressure overload, ARNIs produce greater reductions in fibrotic remodeling compared to an ARB²⁰⁰ or ACE inhibitor²⁰¹ alone after MI likely due to added reduction in cardiac hypertrophy²⁰⁰.

Cardioprotection can also be achieved by activation of the alternative RAS pathway. Activation of ACE2, which converts Ang II to Ang (1–7), was recently shown to attenuate pathological remodeling in the normotensive rat heart when measured 4 weeks post-MI²⁰². The mechanism of action for Ang (1–7) may be linked to the proinflammatory signaling that occurs acutely following the insult. Treatment of rats with Ang (1–7) promoted resolution of inflammation and reduced mitochondrial dysfunction related to downregulation of C-X-C chemokine receptor 4 (CXCR4) during MI²⁰³. The mechanisms of protection discussed here underlie the benefits of RAS inhibition in humans as well. ACE inhibitors^204–208, and ARBs^209,210 reduce post-MI mortality and secondary events. Collectively, there is consensus on improved cardiac remodeling with RAS inhibition post-MI. Given that the size of the infarct is often unaffected, limiting the fibrotic remodeling of the viable myocardium is of great importance to prevent the development of cardiac dysfunction and heart failure. Clinical use of RAS inhibitors is common, but alternative strategies and novel targets are necessary for development of anti-fibrotic therapies. The known actions of Ang II in the cardiac fibroblast can be used to identify these targets and potentially determine more direct methods of intervention for heart failure.

4. Conclusions and Future Directions: Cardiac fibroblasts as therapeutic targets

Mounting evidence supports that limiting the extent and duration of cardiac fibroblast activation is necessary for cardioprotection after injury and slowed progression toward heart failure. It is important to allow for the appropriate initiation of wound healing for cardiac repair and elasticity for contraction without excessive fibrotic remodeling. These collective data highlight multiple players that contribute to fibroblast activity and ECM production downstream of Ang II, including various fibrogenic signaling pathways, inflammation, and oxidative stress. Thus, protection is afforded by RAS inhibitors as summarized in Figure 2, but the timing of treatment dictates the extent of that protection. During pressure overload and cardiac injury, cardioprotection is achieved with ACE inhibitors, ARBs, and ARNIs when treatment occurs prior to or during the insult. Given the lack of protection after the establishment of fibrotic remodeling, novel targets and approaches are needed. Future directions for identification of mechanisms of cardiac fibroblast activation are summarized in Figure 3.

Figure 2. RAS inhibition reduces cardiac fibrosis and improves outcomes after chronic or acute injury.

Hypertension and cardiac injury (e.g. myocardial infarction) induce changes in the heart that result in cardiac fibrosis, impaired function, and increased mortality. These insults impact several cell types resulting in cardiomyocyte hypertrophy, fibroblast proliferation and activation, and immune cell recruitment to the heart. Together, the adaptive responses of cardiac cells cause heart wall thickening and extracellular matrix production that reduce cardiac contractility and output that are attenuated by RAS inhibition. Created with BioRender.com.

Figure 3. Current knowledge gaps and future areas of study.

Important foci for future investigation are highlighted.

Research in the area of cardiac RAS signaling and modulation of adverse cardiac remodeling has produced the most widely used therapeutics for patients with heart failure secondary to hypertension and MI. The growing knowledge base in fibrogenic proteins, epigenetics, and cardiac fibroblast subpopulations are elucidating novel targets for intervention that may increase the anti-fibrotic efficacy of existing approaches. Furthermore, transient ACE inhibition has been shown to produce persistently altered cardiac fibroblasts with reduced proliferative, pro-inflammatory, and fibrogenic profiles^{14,170,176,177}. Understanding the ways in which cardiac fibroblasts are reprogrammed by transient ACE inhibition, and therefore targeting Ang II-mediated pathways, to manifest suppressed responses to injury may hold new insights for limiting fibrotic remodeling. As single cell investigations of cardiac fibroblasts become more numerous, genetic candidates for excessive cardiac fibroblast activation are being described and providing more avenues for future work. Thus far, TGF-β has been described as the main fibrogenic mediator downstream of AT1 receptor stimulation. However, it is possible that some fibrogenic signals may work independently of TGF-β and warrants investigation. Additionally, the regulation of AT1 receptor expression and the incidence of Ang II-induced cell priming, or feed-forward mechanisms are impact Ang II-mediated signaling. It is also possible that different cardiac cell types have different Ang II receptor expression bias to regulate cell-specific responses. The roles for multiple cell types including cardiac fibroblasts, cardiomyocytes, and immune cells during fibrotic remodeling, cell-specific responses to ROS and the timing and extent of oxidative stress are important areas of further investigation of cell-cell communication. Lastly, most of the reviewed studies were performed in males. The known differences in cardiac remodeling due to sex were recently reviewed²¹¹ and the role of estrogen in various fibrogenic signaling cascades was discussed. However, sex-specific differences in fibrotic remodeling remains an important and understudied area of investigation. Research efforts in these areas will increase our mechanistic understanding of how to prevent, limit and reverse fibrosis.

Highlights.

• Angiotensin induces fibroblast activation, proliferation, and inflammation

• Renin angiotensin system inhibitors attenuate fibrotic remodeling

• Targeting angiotensin-mediated pathways may produce lasting effects in fibroblasts

Abbreviations:

RAS

renin angiotensin system

ACE

angiotensin converting enzyme

Ang II

angiotensin II

AT1

angiotensin receptor type 1

AT2

angiotensin receptor type 2

Ang (1–7)

angiotensin-(1–7)

ARB

angiotensin receptor blocker

MI

myocardial infarction

ECM

extracellular matrix

ɑ-SMA

ɑ-smooth muscle actin

MAPK

mitogen-activated protein kinases

ERK1/2

extracellular signal-regulated kinase 1/2

ROS

reactive oxygen species

TGF-β

transforming growth factor beta

STAT3

signal transducer and activator of transcription 3

TLR4

toll-like receptor 4

miRNA

microRNA

CDK6

cyclin-dependent kinase 6

MCP-1

monocyte chemoattractant protein 1

AP-1

activator protein-1

TNFɑ

tumor necrosis factor alpha

NFκB

nuclear factor kappa-light-chain-enhancer of activated B cells

NOX

NADPH oxidase

RECK

reversion-inducing-cysteine-rich protein with Kazal motifs

DNMT

DNA methyltransferase

HDAC

histone deacetylases

Ac-SDKP

N-acetyl-seryl-aspartyl-lysyl-proline

ARNI

angiotensin receptor-neprilysin inhibitors

PPAR-γ

peroxisome proliferator-activated receptor-gamma

CCB

calcium channel blocker

SHR

spontaneously hypertensive rat

NOS

nitric oxide synthase

LV

left ventricle

VSMC

vascular smooth muscle cell

TRAIL

TNFα-related apoptosis inducing ligand

FADD

Fas associated death domain

TRADD

tumor necrosis factor receptor type 1-associated death domain

CXCR4

C-X-C chemokine receptor 4