Smad-dependent pathways in the infarcted and failing heart

Abstract

In infarcted and failing hearts, TGF-β superfamily members play an important role in regulation of inflammatory, reparative, fibrogenic and hypertrophic responses through activation of Smad-dependent and Smad-independent cascades. This review manuscript discusses the mechanisms of regulation and role of Smad pathways in myocardial infarction and in heart failure. Cardiomyocyte-specific Smad1 activation exerts protective anti-apoptotic actions following ischemia/reperfusion. In contrast, the role of the Smad1/5/8 cascade in reparative, immune and vascular cells infiltrating the infarcted heart is unknown. Smad3, but not Smad2 is implicated in repair of the infarcted heart, by activating reparative myofibroblasts and by promoting anti-inflammatory transition in macrophages. However, prolonged activation of Smad3 may promote adverse remodeling and fibrosis. The inhibitory Smad, Smad7 restrains TGF-β-induced fibroblast activation, but also exerts TGF-independent actions through inhibition of receptor tyrosine kinase signaling. Cell-specific approaches targeting Smad pathways may hold therapeutic promise in myocardial infarction and in heart failure.

1. Introduction

The members of the TGF-β superfamily are highly conserved across species and serve as central regulators of inflammatory, reparative and fibrotic responses in many different organs and pathologic conditions [1],[2]. In humans, the superfamily is comprised of more than 30 structurally related proteins, which can be subclassified into several subfamilies, including the TGF-βs (TGF-β1,-β2 and -β3), the bone morphogenetic proteins (BMP2/10 and BMP12/15)[3], the Growth differentiation factors (GDFs), the inhibins, activins, Nodal, Lefty and anti-Mullerian hormone proteins. TGF-β superfamily members signal by binding to specific combinations of type I and type II TGF-β receptors, thus transducing downstream signaling cascades, involving intracellular effectors, the receptor-activated Smads (R-Smads) [4], or Smad-independent cascades [5]. R-Smad signaling is negatively regulated through induction of the inhibitory Smads (I-Smads), Smad6 and Smad7, which suppress TGF-β superfamily signaling. The relative significance of Smad-mediated and non-Smad signaling is dependent on the specific cell type stimulated by TGF-β superfamily members, and on context-dependent factors.

Cardiac injury is associated with induction of a wide range of TGF-β superfamily members, including TGF-βs [6], BMPs [7] and GDFs [8], resulting in downstream activation of Smad-dependent [9] and non-Smad signaling pathways [10]. A growing body of evidence suggests that activation of Smad signaling cascades plays an important role in regulation of myocardial remodeling, inflammation, and fibrosis. The current review manuscript discusses our knowledge on the effects of Smad-dependent signaling in repair, remodeling and fibrosis in infarcted and failing hearts, with an emphasis on recently published studies examining the cell-specific actions of specific R-Smads, and the role of I-Smads in myocardial pathologic conditions.

2. The biology of Smad signaling cascades

TGF-β superfamily members signal by binding to dual specificity cell surface receptors, called TGF-β receptors (TβR). Humans have seven type I TβRs (ALK1-7) and 5 type II TβRs (TβRII, ActRII, ActRIIB, AMHRII and BMPRII). Binding to a member of the TGF-β superfamily results in formation of a heterotetramer, a complex comprised on 2 type I and 2 type II receptors [11]. Depending on the cell type and context, individual TGF-β superfamily members bind to their respective type II TβRs that then phosphorylate and partner with their cognate type I TβR. The activated type I receptor then phosphorylates the carboxyterminal end of a member of the R-Smad family (Smad1-3, Smad5 and Smad8), thus activating the canonical arm of TGF-β signaling. TGF-βs, activins and Nodal typically activate Smad2 and Smad3, whereas BMPs act predominantly through phosphorylation of Smad1, Smad5 and Smad8. After phosphorylation by the type I receptor, R-Smads dissociate from the receptor and form trimeric complexes with the common Smad, Smad4. Subsequently, the R-Smad/Smad4 complex translocates to the nucleus through importin-mediated actions, and binds to Smad-binding elements in the promoter regions of target genes, regulating their transcription. Nuclear translocation of the Smad complex is stimulated by TGF-β ligands, but can also be regulated by TGF-β-independent cascades [12]. For example, phosphorylation of the linker region of R-Smads through activation of Erk MAPK [13], or protein kinase G [14] pathways inhibits nuclear shuttling of the Smad complex.

Non-Smad signaling pathways (such as p38, Erk, and JNK MAPK, PI-3K and Rho kinase cascades) can also be activated by TGF-β superfamily members. TGF-β/R-Smad signaling is tightly regulated through several negative feedback mechanisms that can operate not only at the receptor level, but also upstream or downstream of the receptors. The I-Smads, Smad6 and Smad7 act as prominent negative regulators of TGF-β cascades through several distinct mechanisms. First, I-Smads restrain TGF-β-induced Smad signaling by competitively inhibiting R-Smad activation through binding to type I TβRs [15]. Second, interactions of I-Smads with Smad4, prevent formation of R-Smad-Smad4 complexes [16]. Third, I-Smad-mediated recruitment of the ubiquitin ligases Smurf 1 and 2 mediate type I receptor ubiquitination and degradation [17], thus suppressing TGF-β signaling.

3. Cell-specific actions of Smad cascades in the infarcted myocardium

Myocardial infarction increases the levels of bioactive TGF-β in the myocardium through activation of latent stores of TGF-β, via release of TGF-β stored in platelet granules, and through de novo synthesis and secretion of TGF-β isoforms by ischemic cardiomyocytes, macrophages, vascular cells and fibroblasts [18],[6],[19]. Unfortunately, experimental evidence on the mechanisms of TGF-β activation in the infarcted heart is limited, and the relative significance of de novo synthesis of TGF-β isoforms vs. activation of latent stores is not known. Thus, the current paradigm is primarily informed by in vitro experiments, or extrapolation of findings from other organs. Cell surface αV integrins, specialized matrix proteins (such as thrombospondin-1 [20] and ED-A fibronectin[21]) and proteases (including, cathepsins [22], serine proteases [23],[24], matrix metalloproteinases (MMPs) [25], and members of the ADAMTS (a disintegrin and metalloproteinase with thrombospondin motifs) family [26],[27]) are capable of activating TGF-β in vitro, and may be implicated in TGF-β activation in the infarcted myocardium.

In addition to the 3 TGF-β isoforms, several other members of the TGF-β superfamily, including BMPs [7] and GDFs [28] are also upregulated in the infarcted myocardium. Binding of TGF-β superfamily members to their receptors results in activation of both Smad1/5 and Smad2/3 cascades in border zone cardiomyocytes and in immune cells, fibroblasts and vascular cells infiltrating the infarct. Studies using cell-specific loss-of-function models in mice have provided key insights into the role of specific Smad-dependent pathways in repair and remodeling of the infarcted heart (Table 1).

Table 1:

Cell-specific effects of Smad proteins in the infarcted myocardium

Smad member	Pattern of activation in the infarcted heart	Cell-specific actions	Proposed Mechanism	Ref
Smad1	• Early Smad1 activation in the infarcted myocardium. • Localization in cardiomyocytes was noted in a model of ischemia/reperfusion. • Activated Smad1 was found in endothelial cells in the infarct zone.	• Cardiomyocyte-Smad1 protects from ischemia/reperfusion injury, attenuating cardiomyocyte death. • The role of Smad1 signaling in other cell types has not been studied	Cardiomyocyte apoptosis inhibition by decreasing Caspase 3 and enhancing anti-apoptotic Bcl-xL expression.	[29],[9], [79]
Smad2	• Smad2 activation was noted in the border zone of reperfused infarcts, localized in both cardiomyocytes and infiltrating cells • Smad2 activation in myofibroblasts peaked during the proliferative phase of infarct healing (7 days after coronary occlusion).	• Myofibroblast Smad2 has a modest and transient effect in decreasing systolic function after infarction but does not play a major role in repair and fibrosis of the infarcted heart. • The role of Smad2 signaling in cardiomyocytes and infiltrating cells in the infarct has not been studied.	• In vitro, cardiac fibroblast Smad2 stimulates synthesis of fibronectin, versican, periostin, and collagen V. These actions were not associated with an in vivo phenotype in fibroblast-specific Smad2 KO mice. • In contrast to Smad3, Smad2 does not play a role in TGF-β-induced expression of integrins and suppression of RhoA expression.	[35],[9], [80]
Smad3	Increased Smad3 activation has been reported following infarction, localized in fibroblasts, immune cells and cardiomyocytes.	• Myofibroblast Smad3 protects the infarcted heart from cardiac rupture and adverse remodeling by promoting formation of an organized scar composed of well-aligned myofibroblast arrays and collagen fibers. • Cardiomyocyte Smad3 contributes to systolic dysfunction. • Myeloid cell–specific Smad3 contributes to cardiac repair.	• In fibroblasts, Smad3 mediates α-SMA expression and induction of extracellular matrix proteins, CCN2 and TIMPs. Moreover, Smad3 mediates the anti-proliferative effects of TGF-β on cardiac fibroblasts. • Smad3 activates fibroblasts through stimulation of α5 integrin-induced NOX2 synthesis and induction of RhoA. • Cardiomyocyte Smad3 accentuates NOX2 expression and nitrosative stress, promoting apoptosis and MMP2-dependent degradation of the extracellular matrix. • Phagocytosis stimulates Smad3 signaling in macrophages, which in turn contributes to activation of a phagocytic program, inducing Mfge8 expression. • Macrophage Smad3 promotes macrophage transition to an anti-inflammatory phenotype.	[30],[9], [32],[35]
Smad7	• Smad7 is upregulated in border zone cardiomyocytes, in fibroblasts and in immune cells infiltrating the infarcted myocardium. • Expression of Smad7 is higher in α-SMA-expressing myofibroblasts, when compared to α-SMA-/PDGFRa+ fibroblasts.	Smad7 induction in infarct myofibroblasts protects the infarcted heart from heart failure related mortality, dysfunction and fibrosis.	• Smad7 induction in myofibroblasts acts as an endogenous TGFβ1 induced negative feedback mechanism that inhibits post-infarction fibrosis by restraining Smad-dependent and- independent TGFβ1 responses. • Smad7 attenuates R-Smad and non-Smad pathway activity, without affecting TGF-β receptor activation. The anti-fibrotic actions of Smad7 also involve TGF-independent effects on Erbb2 signaling.	[61]

The role of the Smad1/5/8 cascade in myocardial injury, repair and remodeling after myocardial infarction is poorly understood. In a model of coronary ischemia/reperfusion, early Smad1 activation was noted in ischemic cardiomyocytes. Cardiomyocyte-specific Smad1 signaling has been suggested to exert protective actions, attenuating cardiomyocyte apoptosis in the ischemic and reperfused myocardium [29]. Although Smad1 activation in the infarcted heart is not limited to cardiomyocytes, the role of Smad1 signaling in modulating phenotype of immune cells, fibroblasts and vascular cells remains unknown.

The Smad3 cascade, on the other hand, plays a central role in regulating phenotype and function of cardiomyocytes, immune cells and fibroblasts in the infarcted heart. In a model of reperfused myocardial infarction, cardiomyocyte-specific Smad3 signaling was found to promote dysfunction and adverse remodeling through actions involving nitrosative stress and matrix metalloproteinase (MMP)2 activation [30]. In a model of non-reperfused infarction, cardiomyocyte-specific TGF-β signaling was involved in the pathogenesis of left ventricular rupture [31], through effects attributed to suppression of cardioprotective genes, such as interleukin (IL)-33, GDF-15 and thrombospondin (TSP)-4. Whether these actions involve Smad-mediated mechanisms is not known. TGF-β/Smad3 signaling is also centrally involved in regulation of immune cell phenotype. Studies using mice with myeloid cell-specific Smad3 loss showed that Smad3 signaling in macrophages mediates a phagocytotic program and promotes anti-inflammatory transition, which is important in repair of the infarcted heart [18]. Activation of Smad3 in a-smooth muscle actin (α-SMA)-expressing myofibroblasts infiltrating the infarct plays a critical role in repair of the infarcted heart. Myofibroblast-specific Smad3 signaling protects the infarcted heart from cardiac rupture and adverse remodeling, mediating formation of organized well-aligned myofibroblast arrays. These protective effects of fibroblast Smad3 have been attributed to stimulation of an α5 integrin/reactive oxygen axis, required for acquisition of a reparative phenotype [30]. Smad3 signaling in fibroblasts has also been implicated in TGF-β-mediated induction of extracellular matrix genes, fibrogenic mediators (such as CCN2), and anti-proteases (such as TIMPI) [9],[32]. In vitro, Smad3 plays a central role in upregulation of α-SMA and was found to mediate the anti-proliferative effects of TGF-β1 on cardiac fibroblasts [32]. Thus, fibroblast Smad3 signaling mediates reparative functions of fibroblasts and promotes matrix synthesis, but also restrains proliferative expansion of the myofibroblast pool in vitro and in vivo [32]. Although timely Smad3 activation is central for repair of the infarcted heart, prolonged stimulation of TGF-β-induced Smad signaling in fibroblasts may accentuate fibrotic remodeling after myocardial infarction [33],[34], contributing to the pathogenesis of diastolic dysfunction. In contrast to the critical actions of fibroblast Smad3 in repair of the infarcted heart, Smad2 activation in myofibroblasts does not significantly contribute to the reparative process [35].

The cellular effects of Smad cascades in the infarcted heart may not be limited to actions on cardiomyocytes, fibroblasts and macrophages. Smad3 has been implicated in activation of regulatory T cells (Tregs) [36], a subpopulation of lymphocytes with an important role in negative regulation of post-infarction inflammation and repair [37],[38]. Moreover, in injury sites, TGF-β activation of R-Smads has been suggested to regulate the inflammatory and angiogenic properties of endothelial cells [39]. In vitro, the effects of TGF-β on endothelial cells were found to be either angiogenic or angiostatic, depending on the balance between Smad1 vs Smad2/3 activation [40]. However, the significance of these mechanisms in cardiac repair is unknown, as studies investigating lymphocyte or endothelial cell-specific actions of Smad cascades in the infarcted heart have not been performed.

4. Smad signaling cascades in conditions associated with chronic heart failure

Chronic heart failure in patients exhibits remarkable pathophysiologic heterogeneity. TGF-β signaling cascades have been suggested to play an important role in the pathogenesis of cardiac remodeling and dysfunction in several heart failure-associated pathophysiologic conditions. Patients with end-stage Heart Failure with reduced Ejection Fraction (HFrEF) exhibit persistent myocardial activation of Smad2 and Smad3 cascades, associated with increased levels of TGF-β1 [41]. Evidence showing activation of Smad-dependent signaling in Heart Failure with preserved Ejection Fraction (HFpEF) is lacking; however, the prominent fibrotic remodeling found in a subset of HFpEF patients may reflect, at least in part, activation of TGF-β/Smad3-mediated cascades.

Left ventricular pressure overload underlies heart failure in patients with a wide range of conditions, including hypertension and aortic stenosis. Activation of mechanosensitive pathways (such as the YAP/TAZ cascade [42]) and neurohumoral mediators (such as norepinephrine and angiotensin II) induce and activate TGF-βs and BMPs in the pressure-overloaded myocardium, resulting in activation of Smad2/3 and Smad1/5 signaling cascades [43],[44]. Considering that Smad cascades are activated in all cells involved in cardiac remodeling, cell-specific targeting approaches are needed to understand the role of specific Smads in regulation of the cellular responses following pressure overload. Although cardiomyocyte-specific TGF-β signaling has been implicated in dysfunction and adverse remodeling of the pressure-overloaded heart [45], whether these actions involve Smad-dependent signaling remains unknown. Two independent studies have investigated the role of the TGF-β/Smad2/3 cascade in regulating fibroblast phenotype in the pressure-overloaded heart. Fibroblast-specific activation of TGF-β/Smad2/3 signaling was involved in late development of interstitial fibrosis and dysfunction following left ventricular pressure overload [46]. In contrast, in another study, myofibroblast-specific Smad3 activation had protective actions, preventing early development of systolic dysfunction, by attenuating collagenase activity. Thus, TGF-b-induced Smad3 activation in fibroblasts may exert important matrix-preserving actions that inhibit generation of injurious and pro-inflammatory matrix fragments [47]. Taken together these seemingly conflicting studies may represent two different aspects of fibroblast-specific Smad3 signaling during the course of the cardiomyopathy caused by pressure overload. Early Smad3 activation in fibroblasts may serve to protect cardiomyocytes, by preserving the surrounding extracellular matrix, which plays an important role in homeostatic function. However, prolonged activation of TGF-β/Smad3 signaling in response to a persistent pressure load may result in excessive deposition of matrix proteins, thus contributing to dysfunction and the progression of heart failure.

In addition to its effects on the myocardium, Smad3 may also play a role in fibrotic remodeling of the valves. In vitro studies using cultures of aortic valve leaflets demonstrated that TGF-β-mediated Smad3 activation promotes conversion of valve interstitial cells into matrix-secreting myofibroblasts, thus promoting valve fibrosis [48]. Moreover, inhibitory effects of TGF-β1 on osteoblastic differentiation of valve interstitial cells were attributed to inhibition of the Smad1/5/8 pathway [48].

5. Inhibitory signals that restrain Smad-mediated signaling: the role of the I-Smads and c-Ski

Effective cardiac repair is dependent on timely suppression of TGF-β/Smad responses. This is particularly important for fibroblast-mediated actions, as unrestrained TGF-β/Smad2/3 signaling in fibroblasts would be expected to stimulate progression of fibrosis and development of heart failure. During the maturation phase of infarct healing, activated matrix-synthetic myofibroblasts convert into matrifibrocytes [49], a differentiated fibroblast phenotype that does not express significant amounts of α-SMA and structural collagens, but serves to maintain the scar and expresses tendon and cartilage genes. Considering the central role of TGF-β/Smad2/3 cascades in myofibroblast conversion, it is plausible to hypothesize that inhibitory signals suppressing Smad-dependent signaling may be implicated in transition of myofibroblasts to matrifibrocytes.

Although the I-Smads, Smad6 and Smad7 have been demonstrated to restrain responses triggered by TGF-β superfamily members, their role in myocardial diseases remains poorly understood. In vitro evidence suggests that I-Smads can function as negative regulators of TGF-β superfamily pathways at several different levels; however, the in vivo significance of these actions is unclear.

The role of Smad6 in myocardial infarction and heart failure has not been investigated. Smad6 is expressed in the myocardium at the early stages of development [50] and inhibits BMP signaling, thus regulating cardiomyocyte differentiation [51]. Global loss of Smad6 in mice is associated with embryonic lethality, due to valve hyperplasia and outflow tract defects. Some Smad6 KO survive into adulthood, but exhibit aortic ossification and elevated blood pressure [52]; the cellular basis for these perturbations remains poorly defined. In order to address the role of Smad6 in adult cardiac pathology, there is a need for development of conditional cell-specific genetic models of Smad6 loss.

An emerging body of evidence suggests an important role for cell-specific Smad7 actions in repair and remodeling of the injured myocardium. Early studies have used a mouse model with germline deletion of the coding region of exon 1 to explore the role of Smad7 in cardiac remodeling (Smad7^Δex1 mice). These mice exhibit only partial loss of Smad7 functions, due to the preservation of the MH2 domain, which is responsible for Smad7 interactions with the TGF-β signaling pathway [53]. The Smad7^Δex1 mice are viable but have a smaller size due to alterations in skeletal muscle growth [54,55]. Due to the preservation of MH2 domain actions, experimental studies using the Smad7^Δex1 model are of limited value in understanding the potential role of Smad7 in cardiac remodeling [56,57]. On the other hand, mice with global germline loss of Smad7 exhibit embryonic lethality [58], whereas mutant mice lacking the MH2 domain (Smad7^ΔMH2) die within a few days after birth, with the survivors being of smaller size [59,60]. Thus, considering the involvement of Smad7 in embryonic development, cell-specific conditional approaches are needed to understand the role of Smad7 in adult myocardial diseases.

In order to examine the role of Smad7 in regulating fibroblast phenotype following myocardial infarction, we have recently developed myofibroblast-specific Smad7 KO mice [61]. We found that Smad7 expression is highly induced in α-SMA+ myofibroblasts (but not in α-SMA-negative/PDGFRa+ fibroblasts) through a Smad3-mediated pathway. Myofibroblast-specific Smad7 expression protects the infarcted myocardium from heart failure and attenuates adverse remodeling and fibrosis, by restraining collagen accumulation and myofibroblast conversion. Although Smad7 does not play a role in myofibroblast to matrifibrocyte transition, expression of Smad7 attenuates matrix gene synthesis in matrifibrocytes. The anti-fibrotic effects of Smad7 are only in part related to its suppressive actions on Smad-dependent and non Smad-mediated cascades downstream of TGF-β receptor activation. Unbiased transcriptomic and proteomic experiments identified a new mechanism responsible for the effects of Smad7 in activated myofibroblasts, involving a direct TGF-β-independent inhibition of the receptor tyrosine kinase Erbb2, a crucial fibrogenic mediator in several disease models [62-64]. Our findings showed that I-Smads do not serve only as negative regulators of the TGF-β system, but also exert regulatory actions on receptor tyrosine kinase signaling pathways [61].

In addition to the I-Smads, several additional mechanisms have been implicated in negative regulation of TGF-β/Smad signaling cascades. c-Ski is a nuclear protein that can serve as a TGF-β repressor through several different mechanisms. First, c-Ski can shuttle to the cytoplasm, where it can bind to TβRI [65]. The TβRI/c-Ski interaction may interfere with the R-Smad/Smad4 complex, thus perturbing its nuclear translocation and inhibiting subsequent activation of Smad-dependent transcription. Second, nuclear c-Ski may inhibit transcription of TGF-β-induced signals by stabilizing inactive Smad complexes on DNA [66]. Third, c-Ski can serve as a co-repressor [67] that interferes with interactions between R-Smads and transcriptional co-activators (such as p300) [68]. In the myocardium, c-Ski expression was found to be markedly increased in the infarcted myocardium during the maturation phase of infarct healing [69]. In vitro, c-Ski inhibits α-SMA expression in myofibroblasts, and might be involved in conversion of myofibroblasts into differentiated matrifibrocytes. In addition, part of the ani-fibrotic actions of c-Ski have been attributed to TGF-independent effects, involving interactions with the YAP/TAZ signaling cascade [70].

6. Sex-specific effects of Smad cascades in cardiac remodeling

In the mouse model of non-reperfused myocardial infarction, female animals have better outcome, exhibiting a much lower incidence of cardiac rupture than males [71]. This observation may reflect sex-specific differences in the cellular responses involved in cardiac repair and remodeling [72]. Comparative analysis of data from studies on the role of Smad cascades in the infarcted myocardium have suggested, at least in some cases, sex-specific effects. Although both male and female myofibroblast-specific Smad7 KO mice have accentuated adverse remodeling and fibrosis following myocardial infarction, only male animals show increased mortality [61]. High mortality in mice lacking Smad7 in myofibroblasts is independent of cardiac rupture, and may be due to a higher incidence of heart failure, or arrhythmic events. Considering the comparable accentuation of fibrotic responses observed in male and female myofibroblast-specific Smad7 KOs, the basis for the increased impact of Smad7 loss on male mouse infarct mortality is unclear. Male sex and androgen administration have been suggested to accentuated TGF-β-mediated R-Smad responses in vascular smooth muscle cells [73], in fibroblasts, and in H9c2 cardiomyocyte-like cells [74]. On the other hand, estrogens were found to inhibit TGF-β/R-Smad signaling by promoting Smad2 and Smad3 degradation [75]. However, the significance of these mechanisms in myocardial infarction has not been examined.

7. Therapeutic implications and conclusions

A growing body of evidence suggests that activation of Smad pathways regulates cellular responses with a critical role in repair, remodeling and dysfunction of the infarcted and failing heart [76]. However, the cell-specific and context-dependent actions of Smad signaling cascades pose major challenges in designing therapeutic interventions to improve outcome in patients with myocardial infarction, or heart failure. For example, Smad3 plays an important protective role in the early stages of myocardial infarction by promoting activation of reparative myofibroblasts and by mediating anti-inflammatory transition of macrophages, but is also implicated in chronic cardiomyocyte dysfunction [30],[18]. In older subjects, perturbed activation of Smad3 signaling in fibroblasts may be responsible for impaired repair and formation of a defective scar [77]. On the other hand, persistent activation of Smad3 in fibroblasts may contribute to progressive fibrotic remodeling, promoting heart failure. Thus, design of effective therapeutics targeting Smad3 requires cell-specific approaches, and consideration of the temporal sequence of events involved in repair and remodeling of the infarcted heart [78].

Moreover, our knowledge on the mechanisms of regulation of TGF-β superfamily signaling cascades remains limited. The relative role of specific TGF-β isoforms, BMPs and GDFs in repair and remodeling of the infarcted heart is poorly understood. The role of the Smad1/5/8 cascade remains enigmatic. The identify and function of the inhibitory signals that restrain Smad-dependent signaling is not known. In addition to dissection of the fundamental mechanisms involved in regulation of Smad cascades, clinical studies are needed, in order to identify subsets of patients with heart failure or myocardial infarction who may exhibit perturbed or overactive Smad responses in order to tailor therapeutic interventions.

Figure 1: Effects of Smad cascades on infarct fibroblast function regulate repair and remodeling of the infarcted heart.

A: TGF-β-mediated Smad-dependent cascades in cardiac fibroblasts. (1) Latent TGFβ is stored in the myocardium bound to the latency-associated peptide (LAP) and a member of the Latent TGF-β Binding Protein (LTBP) family, forming the large latent complex. Cell surface integrins, specialized matrix proteins (such as thrombospondin (TSP)-1) and proteases co-operate to release mature TGFβ from this complex, mediating TGF-β activation. (2) Active TGF-β binds to TGFβ receptor (TβR) complexes on the cell surface, stimulating both canonical Smad 2/3 signaling and non-canonical pathways, such as Erk and Akt. (3) Receptor-activated Smads (R-Smads, Smad2/3) bind to the common Smad4, forming an R-Smad-Smad4 complex that translocates to the nucleus and binds to Smad binding elements (SBE) in the promoter region of TGFβ target genes, regulating their transcription. (4) Smad3 activation in infarct myofibroblasts also induces expression of the inhibitory Smad, Smad7 which inhibits TGF-β mediated Smad2/3, Erk and Akt signaling, without affecting TβR activation. (5) In addition, Smad7 has TGF-independent actions, mediated through inhibition of the receptor tyrosine kinase Erbb2. B: The role of Smad cascades in fibroblast-mediated repair and remodeling of the infarcted heart. (6) Smad3 activation plays an essential role in cardiac fibroblast function by promoting formation of well-aligned arrays of activated α-smooth muscle actin (α-SMA)+ myofibroblasts and by stimulating extracellular matrix (ECM) protein synthesis. Effects of Smad3 on cardiac fibroblasts may involve activation of an integrin-NOX2 axis and increased RhoA GTPase expression. The activating effects of Smad3 on infarct fibroblasts are associated with suppression of their proliferative capacity. Smad3-induced myofibroblast activation mediates the formation of an organized collagen-based scar, protecting the infarcted heart from rupture. (7) On the other hand, the inhibitory Smad7, restrains synthesis of structural and matricellular matrix-associated genes, protecting the infarcted heart from adverse remodeling, and reducing heart-failure related mortality. The anti-fibrotic effects of Smad7 involve not only suppression of TGF-β cascades, but also inhibition of Erbb2 signaling.