Getting to the heart of myofibroblast differentiation: implications for scleraxis in ECM remodeling and therapeutic targeting

Myofibroblasts are major contributors to cardiac remodeling after injury. These cells derive partly from the differentiation of resident cardiac fibroblasts, express a more tensile phenotype, and exhibit very high levels of extracellular matrix (ECM) synthesis. The ECM is a dynamic, acellular scaffold of structural proteins (such as collagen, fibronectin, integrins, and cadherins) maintained by fibrocytes (quiescent, immotile, low ECM-modifying) and fibroblasts (proliferative, motile, high ECM-modifying) within the interstitial space. The essential ECM regulates cell migration, cardiomyocyte alignment, and maintains proper chamber morphology and mechanical function (9). In the infarcted heart, ECM produced by fibroblasts and differentiated myofibroblasts replaces necrotic cardiomyocytes. Cell death and progressive cardiomyocyte loss weaken the structural integrity of the ventricular wall, making it prone to hemorrhagic rupture and sudden cardiac death. As cardiomyocytes possess little regenerative capacity in the adult heart, increased ECM deposition in the form of a fibrotic scar is essential to maintain ventricular wall integrity and function after myocardial infarction. Although fibrotic remodeling plays an important role in adaptive repair, excess ECM synthesis and matrix type can become maladaptive. In the peri-infarct zone, cardiomyocytes struggle to survive and adapt to the increased hemodynamic load brought on by the loss of contractile cells. Progressive mechanical stretch due to cell death, protracted inflammation, and hemodynamic stress stimulate extensive interstitial fibrosis in the peri-infarct region (Fig. 1). Loss of elasticity and misalignment of myocardial contractile units (i.e., myocardial slippage) from diffuse interstitial fibrosis decreases cardiac compliance and contributes to systolic and/or diastolic dysfunction (8). Pathological fibrotic remodeling further impairs proper cardiac function by obstructing normal electrical conduction pathways, contributing to arrhythmias (9). Thus, maladaptive ECM deposition in the later stages of ventricular remodeling is a significant contributor to heart failure or sudden cardiac death and is an important target for post-infarct therapeutics. Unfortunately, the therapeutic options for pathological fibrosis in the heart are very limited. Given the prominent role of myofibroblasts in the fibrotic pathophysiology of cardiac injury, understanding myofibroblast differentiation and the mechanisms regulating myofibroblast-mediated fibrosis are essential to developing successful antifibrotic therapies.

Fig. 1.

Mechanisms of myofibroblast differentiation post-myocardial infarction. Scleraxis (Scx) is a key transcriptional regulator of transforming growth factor (TGF)-β₁-mediated myofibroblast transdifferentiation. AGTRI, angiotensin II receptor 1; α-SMA, α-smooth muscle actin; ECM, extracellular matrix; EMT, epithelial-to-mesenchymal transition; LAP, latency-associated peptide; MAPK, mitogen-activated protein kinase; MMP, matrix metalloprotease; Snai1, zinc finger protein SNAI1; TAK1, TGF-β-activated kinase 1; TGF-βRI/RII, TGF-β receptor I/receptor II; Twist1, twist-related protein 1.

A critical step in understanding the pathophysiology of cardiac fibrosis lies in the transition state between fibroblasts and myofibroblasts. This has long been the subject of debate, largely due to the use of transient, nonspecific fibroblast markers. As fibroblasts exhibit significant genetic and phenotypic plasticity, the challenge of identifying fibroblast and myofibroblast origins has been difficult. However, lineage tracing using transcription factor-21 (Tcf21) reporter tags has revealed that resident fibroblasts develop from Tcf21-positive epicardial cells during development (1). This process of fibroblast conversion is termed epithelial-to-mesenchymal transition (EMT). EMT occurs when adherent epithelial cells reduce their expression of cell adhesion proteins (e.g., adherins and tight junctions), losing basal-apical polarity and increasing their motility to sculpt the ECM in vivo. Although epicardial cells are progenitors for both cardiac fibroblasts and coronary vascular smooth muscle cells (or pericytes), only Tcf21-positive epicardial cells commit to the cardiac fibroblast lineage. Therefore, Tcf21 expression is a highly specific marker for resident cardiac fibroblasts and persists from development through to adulthood and cardiac injury (1). Identification of myofibroblast origins has been equally challenging; myofibroblasts also exhibit significant genetic variability depending on their stage of differentiation and the ratio of resident fibroblasts as the sole source of myofibroblasts remains to be determined (i.e., relative to EMT or medullary sources). Periostin, a nonstructural protein secreted into the ECM, has been used as a positive marker for myofibroblast differentiation by Kanisicak et al. (7). In mouse hearts containing both periostin and Tcf21 lineage-tracing reporters, Tcf21-positive cells give rise to nearly all periostin-positive cells, suggesting that fibroblasts may be the largest contributor of differentiated myofibroblasts in the infarcted heart (7). However, the ability of other cell types to differentiate into myofibroblasts should not be discounted when developing antifibrotic strategies.

Initiation of myofibroblast differentiation after myocardial infarction canonically via resident fibroblasts or by EMT is largely attributed to transforming growth factor (TGF)-β₁ and the accompanying inflammatory cytokine signaling mechanisms (Fig. 1) (2). TGF-β₁ is released in response to mechanical strain, catecholaminergic signaling, and inflammation. As cardiomyocytes die, resident fibroblasts are subjected to an elevated and redistributed hemodynamic load. This mechanically stretches fibroblasts and their extracellular network, leading to the release of latent TGF-β₁ from the ECM as well as activation of mechanosensitive angiotensin receptors. Proteolytic cleavage of latency-associated peptide from TGF-β₁ subsequently enables TGF-β₁ to activate TGF-β₁ tyrosine kinase receptor-II (Fig. 1). Activation of TGF-β₁ receptor-II results in the recruitment of TGF-β₁ receptor-I, which forms a heterodimeric receptor complex regulating key proteins in fibroblast proliferation, migration and differentiation. TGF-β₁ decreases epithelial cell-like properties while increasing myofibroblast-like characteristics; for example, TGF-β₁ decreases expression of cellular adhesion proteins (e.g., E-cadherin and tight junctions) and signals the increased synthesis of ECM proteins [e.g., collagen, fibronectin, and matrix metalloproteases (MMPs)] and α-smooth muscle actin (α-SMA). TGF-β₁ levels are further elevated after myocardial infarction through increased catecholamine release. Cardiac injury results in the elevated sympathoadrenal system to release catecholamines into circulation causing an increase in heart rate to compensate for decreased cardiac output (5). Although initially adaptive, chronic catecholamine release can result in sympathetic overdrive and persistent release of renin and angiotensin II by the kidneys, perpetuating mechanotransduction-mediated activation of TGF-β₁ and angiotensin II receptors and stimulating excessive fibrotic remodeling. Inflammation also plays a central role in TGF-β₁-mediated fibrosis. The inflammatory response is first initiated with cardiomyocyte and endothelial cell necrosis as dead cells spill their inflammatory cytokines and chemokines, such as TNF-α, IL-1β, IL-6β, endothelin-1, and TGF-β₁, into the interstitial space. The release of inflammatory cytokines results in the recruitment and infiltration of monocytes and macrophages, which, in turn, release more TGF-β₁. Thus, TGF-β₁ is released through a variety of mechanisms to regulate myofibroblast differentiation.

Current interventional strategies targeting pathological fibrotic remodeling are limited (2). Considering the role of fibrosis in systolic and diastolic dysfunction, this is a critical gap in our ability to effectively prevent patients from developing heart failure. Clinically, therapeutic strategies remain focused on inhibition of the renin-angiotensin-aldosterone system (RAAS). However, only a modest reduction in cardiac fibrosis is observed in patients with heart failure prescribed RAAS inhibitors; thus, the development of novel therapeutic targets is warranted. Therapeutics such as pirfenidone and nintedanib are clinically approved for antifibrotic use in patients with idiopathic pulmonary fibrosis; however, their use has yet to be examined in models of cardiac fibrosis. Another therapeutic strategy has been the targeting of relaxin. The pharmacological interest in relaxin stems from its ability to reverse collagen deposition through the upregulation of MMPs. Relaxin has also been shown to reduce inflammation, increase vasodilation, and stimulate angiogenesis in animal models. However, in a clinical trial completed in 2017, relaxin failed to meet its primary end points of time to confirmed cardiovascular death and worsening of heart failure, potentially due to its half-life of 30 min. Given the significance of TGF-β₁ signaling to fibrotic remodeling and myofibroblast differentiation, inhibition of TGF-β₁ has been explored as a novel therapeutic strategy to reduce fibrosis (2). In experimental rat models of myocardial infarction, inhibition of TGF-β₁ by small-molecule inhibitors administered 1 wk post infarct for 4 wk significantly reduces systolic dysfunction and collagen synthesis (10). Similar results have been shown with anti-TGF-β₁ antibodies and TGF-β₁ receptor decoys. Yet, it is important to keep in mind the beneficial roles that TGF-β₁ plays in infarct repair when considering it as a target for inhibition. In addition to its roles in stimulating myofibroblast differentiation and fibrosis, TGF-β₁ exerts checkpoint regulation (important for tumor suppression) and modulates normal immune regulation (which is critical to recruit monocytes for early infarct repair). Because pathological fibrotic remodeling develops over a prolonged period of time, the timing of administration of antifibrotics is essential. Thus, consideration of both the maladaptive and beneficial signaling, as well as drug specificity, is critical for therapeutics targeting TGF-β₁ signaling. The therapeutic targeting of scleraxis, a downstream mediator of TGF-β₁, may overcome these challenges associated with direct TGF-β₁ inhibition.

The report published in a recent issue of the American Journal of Physiology-Heart and Circulatory Physiology by Al-Hattab et al. (2a), “Scleraxis regulates Twist1 and Snai1 expression in the epithelial-to-mesenchymal transition,” represents the authors’ latest work in characterizing scleraxis as an essential regulator of myofibroblast differentiation and fibrosis (Fig. 1). Here, the authors showed that scleraxis, a basic-helix-loop-helix transcription factor, directly transactivates Twist1 and Snai1 in cardiac fibroblasts and epithelial cells. Synergistically, Twist1 and Snai1 reduce cell adhesion and increase structural ECM protein expression, respectively. This is one of a series of articles by this group aimed at collectively improving our understanding of scleraxis as it relates to myofibroblast transdifferentiation. Previous studies from this group have shown that scleraxis is also a critical and direct inducer of myofibroblast markers such as α-SMA (3) and collagens (4). Previously, TGF-β₁ was thought to be the principal regulator of Twist1 and Snai1. However, here, Al-Hattab et al. clearly show that TGF-β₁-mediated myofibroblast differentiation and fibrosis are dependent on direct scleraxis upregulation of Twist1 and Snai1.

The present findings by Al-Hattab et al. have important implications for novel antifibrotic therapies; however, critical translational barriers remain to be overcome. In particular, our limited understanding of the effects of scleraxis inhibition in vivo and our understanding of myofibroblast temporality represent major challenges. Scleraxis expression is not limited to cardiac cells and is highly expressed in tissues such as tendons, ligaments, and cartilage. As inhibition of scleraxis has yet to be explored in in vivo models of myocardial infarction, aging, or female subjects, it remains unclear how to best advance the target clinically. Cardiac-specific delivery of scleraxis inhibitors could overcome this risk in part; however, this itself is a challenge. First, the potential effects of scleraxis inhibition on nonfibroblast cardiac cells remain unclear. Current approaches to targeting scleraxis include the use of adenoviral vectors or are aimed at developing novel small-molecule inhibitors. As scleraxis is intracellularly expressed, novel small-molecule inhibitors may first have to target cardiac fibroblasts before being internalized, a potentially difficult task considering the void of unique extracellular fibroblast and myofibroblast markers. This difficulty applies to adenoviral methods of delivery as well, since researchers would be hard pressed to find clinicians willing to administer viral vectors to their patients after myocardial infarction. Although this may seem like an insurmountable barrier to the therapeutic translation of scleraxis inhibition, one way to address this quickly would be through the trial of existing libraries of clinically approved drugs. Searching these libraries for drugs that suppress scleraxis expression could provide a solid basis with which to model novel scleraxis-specific inhibitors. The second critical barrier relates to the temporality of fibroblast and myofibroblast phenotypes within the infarcted heart. Fibroblasts do not suddenly convert to myofibroblasts; instead, they exist on a phenotypic spectrum, activating different genes with different intensities and at different stages of infarct remodeling. In a recent study by Fu et al. (6), the authors began to map both the temporality and spatial localization of fibroblast and myofibroblast phenotypes and gene expression. As previously mentioned, cardiac fibrosis after myocardial infarction is a double-edged sword. Fibrosis is both essential and detrimental to cardiac function and highly dependent on the remodeling stage after injury. Understanding when myofibroblasts play a critical role in ventricular remodeling versus when they contribute to further pathogenesis will be crucial to understanding how scleraxis should be modulated for therapeutic benefit. A deeper understanding of scleraxis activity in vivo, temporally, regionally, and in a cell-specific context during cardiac myofibroblast progression is needed.

The myofibroblast is given short shrift when it comes to postmyocardial infarction interventions. As we aim to advance antifibrotic pharmacology, we must commit to further understanding the regulation of myofibroblasts with clinical pharmacology in mind. Moving forward, it is critical that careful consideration be given to both the pathophysiological and physiological roles of myofibroblasts in the infarcted heart. What is clear is that scleraxis is a viable target, as it may be used to attenuate selective parts of the pathological remodeling, such as EMT, without abrogating the totality of TGF-β₁ and catecholamine-mediated ECM remodeling necessary to support the cardiac macrostructure after infarction.

GRANTS

The authors acknowledge the financial support of the Saint John Regional Hospital Foundation and the National Science and Engineering Research Council.

DISCLOSURES

K. R. Brunt holds private equity in NB-BioMatrix Inc. and public stock positions in Correvio Pharma Corp. and Prometic Life Sciences Inc.

AUTHOR CONTRIBUTIONS

A.L.E. and A.J.T. prepared figures; A.L.E., A.J.T., and K.R.B. drafted manuscript; A.L.E., A.J.T., and K.R.B. edited and revised manuscript; A.L.E., A.J.T., and K.R.B. approved final version of manuscript.