Targeting the Cytoskeleton and Extracellular Matrix in Cardiovascular Disease Drug Discovery

Abstract

Introduction:

Currently, cardiovascular disease (CVD) drug discovery has focused primarily on addressing the inflammation and immunopathology aspects inherent to various CVD phenotypes such as cardiac fibrosis and coronary artery disease. However, recent findings suggest new biological pathways for cytoskeletal and extracellular matrix (ECM) regulation across diverse CVDs, such as the roles of matricellular proteins (e.g., tenascin-C) in regulating the cellular microenvironment. The success of anti-inflammatory drugs like colchicine, which targets microtubule polymerization, further suggests that the cardiac cytoskeleton and ECM provide prospective therapeutic opportunities.

Areas Covered:

Potential therapeutic targets include proteins such as gelsolin and calponin 2, which play pivotal roles in plaque development. This review focuses on the dynamic role that the cytoskeleton and ECM play in CVD pathophysiology, highlighting how novel target discovery in cytoskeletal and ECM-related genes may enable therapeutics development to alter the regulation of cellular architecture in plaque formation and rupture, cardiac contractility, and other molecular mechanisms.

Expert opinion:

Further research into the cardiac cytoskeleton is an area ripe for novel target discovery. Furthermore, the structural connection between the cytoskeleton and the ECM provides an opportunity to evaluate both entities as sources of potential therapeutic targets for CVDs. Refining computational analytical techniques for drug discovery over the next five, ten or so years will increase the success probability of the novel targets determined and further de-risk drug development.

Expert Opinion

The prevalence of cardiovascular disease (CVD) is rapidly rising, and it is predicted that by 2030 there will be more than 23.6 million CVD-related deaths per year, with approximately half of the United States adult population living with some form of CVD diagnosis by 2035. In the United States alone, CVDs are the cause of over $350 billion in annual spending just for management and treatment, with the most spending allocated to ischemic heart disease and hypertension health services. The growing burden of CVDs globally highlights the need for increased and sustained global prevention efforts and large-scale drug discovery approaches that can adequately address clinical unmet needs. The cytoskeletome, which we define as the complete set of cytoskeletal proteins, such as filaments and microtubules, and other associated material including the underlying extracellular matrix (ECM) that supports its architecture, present relatively new and underexplored avenues of CVD drug target discovery.

The majority of currently approved drugs for heart disease target traditional risk factors such as high cholesterol or blood pressure. Here we highlight how targeting the cytoskeletome can inform more precise therapies by tailoring treatment to cell-specific factors intrinsic to the vessel wall. These novel avenues have the potential to significantly contribute to cardioinformatics and precision cardiology initiatives that advance the use of data science methods to fight heart disease with computation by creating better drugs -- for example, by building target deconvolution algorithms that systematically identify and prioritize novel drug targets, or by assisting with biomarker-guided drug repurposing efforts from existing pharmacogenomic data. With better drugs for various cardiovascular diseases, physicians will be able to rewrite treatment guidelines for patients with diseases that do not currently have effective therapies but rather only have lifestyle changes or treating other phenotypes as part of the treatment plan and clinical practice guidelines. Additionally, improving the safety and efficacy of CVD drugs will help slow and decrease the number of CVD cases every year, which in turn will prevent more deaths and also decrease the amount of money needed to manage these diseases from a population-level healthcare perspective.

Further research into the cardiac cytoskeleton (and the underlying ECM that supports its architecture) is an area ripe for novel target discovery. There is mounting evidence suggesting that the cytoskeleton is an under-investigated and undervalued therapeutic target for CVD phenotypes and new computational advances provide an opportunity to discover the biological mechanisms driving changes in the cytoskeleton during CVD pathophysiology. Furthermore, the structural connection between the cytoskeleton and the ECM provides an opportunity to evaluate both entities as sources of potential therapeutic targets for CVDs. Incorporating data-driven computational methods to study the cytoskeleton and the ECM in the context of cardiovascular diseases will provide a more holistic perspective on the progression of CVDs such as cardiac fibrosis, hypertension, heart failure, and atherosclerosis. Additionally, further research into the cytoskeletome could lead to discoveries of related systems and pathways that drive the pathogenesis of CVDs that may lead to additional new therapeutic discoveries as well. Overall, advancing and shifting to computational drug discovery will assist future research in discovering new effective drugs for CVDs without the difficult, time-consuming, expensive, and risky process defining the current status quo of the traditional pharmaceutical industry.

Drug discovery needs to become a much more rapidly moving field as an increasing number of people are developing cardiovascular diseases, amongst others, while our current drugs are showing to not be as effective as needed and the rates of novel therapeutic development is declining. Expediting future research into computational drug discovery will potentially combat these declining rates and discover new innovative therapies for diseases that currently do not have any. The cardioinformatics drug discovery field, with continued further research and support, will hopefully evolve into being one of the forefront tactics in novel drug development as we must rely more on synthetically designing novel therapeutics and utilizing the vast amounts of data already available. Refining computational analytical techniques for drug discovery over the next five, ten or so years will increase the success probability of the novel targets determined and further de-risk drug development by further optimizing this process (e.g., saving time, money, and other resources for all parties involved) by way of computationally-derived therapeutics.

1. Introduction

The prevalence of cardiovascular disease (CVD) is rapidly rising, and it is predicted that by 2030 there will be more than 23.6 million CVD-related deaths per year, with approximately half of the United States adult population living with some form of CVD diagnosis by 2035 [1,2,3]. In the United States alone, CVDs are the cause of over $350 billion in annual spending just for management and treatment, with the most spending allocated to ischemic heart disease and hypertension health services [4]. The growing burden of CVDs globally highlights the need for increased and sustained global prevention efforts and large-scale drug discovery approaches that can adequately address clinical unmet needs.

The majority of currently approved drugs for heart disease target traditional risk factors such as high cholesterol or blood pressure. In this review, we highlight how targeting the cytoskeleton — and the associated extracellular matrix (ECM) that supports its architecture — can inform more precise therapies by tailoring treatment to cell-specific factors intrinsic to the vessel wall (Figure 1). To-date, CVD biomarker discovery and validation efforts have focused mainly on targeting the inflammation and immunopathology aspects inherent to clinical indications such as atherosclerosis and other CVDs such as heart failure. In contrast, the cytoskeletome, which we define as the complete set of cytoskeletal proteins, such as filaments and microtubules, and other associated material including the underlying ECM, present relatively new and underexplored avenues of CVD drug target discovery. These novel avenues have the potential to significantly contribute to cardioinformatics, the multidisciplinary field where bioinformatics and precision cardiovascular medicine intersect, and precision cardiology initiatives that advance and optimize the use of data science methods to fight heart disease with computation by creating better drugs -- for example, by building target deconvolution algorithms that systematically identify and prioritize novel drug targets, or by assisting with biomarker-guided drug repurposing efforts from existing pharmacogenomic data [5,6].

Figure 1: Central Illustration.

Changes in the organization of the cytoskeleton and extracellular matrix (ECM) is associated with cardiovascular disease (CVD) phenotypes. For example, in atherosclerosis, changes in ECM composition cause decreased vascular smooth muscle cell (VSMC) migration, which leads to cytoskeletal disorganization [51]. The progression of heart failure is correlated with the density of the cardiomyocyte microtubule network [42] while in the context of cardiac fibrosis, the inactivation of osteopontin (OPN) leads to high levels of ECM network disruption [12, 26]. Additionally, in pulmonary hypertension, another matricellular protein called periostin is associated with the disruption of cytoskeletal architecture in endothelial cells [80].

The dynamic cytoskeleton -- composed of proteins including, but not limited to, actin filaments, intermediate filaments, and microtubules -- plays a role in adapting the cellular architecture to molecular environmental fluctuations seen in the heart and blood vessels during changes in cardiovascular health and disease status. The intracellular cytoskeleton is essential for maintenance of cellular structure, transportation of organelles, and cellular movement. Similarly, the ECM, a dynamic network composed of diverse macromolecules such as collagen and glycosaminoglycans, has many essential cellular functions including, but not limited to, actively regulating cell structure, migration, and local cellular behavior.

The ECM is directly connected to the intracellular cytoskeleton by various transmembrane adhesive proteins. These cell-matrix junctions (e.g., actin-linked cell-matrix junctions that bind actin filaments to the ECM) allow either the cell or ECM to influence changes on the other through both intracellular and extracellular signaling pathways (e.g.,integrin-mediated apoptosis signaling to regulate angiogenesis [7]). Also, these junctions play a key role in creating a cohesive multicellular network (e.g., cell-cell junctions allow for intracellular signaling) [8,9] while being necessary for cell survival, differentiation, and proliferation. Defects in these adhesive proteins are associated with many conditions and genetic diseases, including CVDs, as they regulate the dynamic processes of the cytoskeleton across different CVD phenotypes These proteins may be isolated as drug targets and provide a new avenue to correct dysregulated processes.

Thus far, the molecular environmental changes in the vascular cell wall that are known to accompany CVD phenotypes, and promote alterations in the regulation and interplay of the cytoskeleton and ECM, have remained largely underexplored. This review aims to summarize the current literature on the cardiac cytoskeleton, including its connection to the underlying ECM (Figure 1), and to propose new research directions supporting novel biomarker discovery and therapeutics development that alters cytoskeleton/ECM regulation and impacts cellular architecture in ways that promote the stabilization and potential reversal of cardiovascular disease and its associated co-morbidities (e.g., cardiorenal and metabolic abnormalities).

2. The Role of the Cardiac Cytoskeleton

2.1. Cardiac Fibrosis

Cardiac fibrosis and ECM remodeling (i.e., changes in the composition of the ECM through polymerization and depolymerization) have been found to be associated with various CVDs and the level of severity for adverse events such as mortality, left ventricular dilation, and even an increase in hospitalization rates [10,11]. In general, developing fibrosis is a strong indication and determinant for heart failure and cardiac function [12]. Fibrotic remodeling is seen to come with a rise in stiffness [12], which has been found to be a useful cue for predicting disease status [13]. Increased ECM stiffness is partially caused by formation of myofibroblasts, which are marked by expression of α-smooth muscle actin (αSMA) [13]. Lampi and Reinhart-King (2018) conducted a systematic review of literature ECM cues in disease and found that statins could be repurposed to target Rho and YAP/TAZ in order to prevent disease progression associated with ECM stiffening [13]. They also show examples of several therapeutics targeting ECM stiffness in clinical trials indicating future feasibility of ECM-based cardiac therapeutics. ECM stiffness, which is caused by dysregulation and remodeling, is harmful because it alters the cellular microenvironment, which has spillover effects leading to fibrosis [13]. This dysregulation of the microenvironment is also a known characteristic of vascular diseases, as cells remodel the cellular microenvironment in response to sensing mechanical signals transduced from the ECM, such as changes in elasticity [14]. Lampi and Reinhart-King (2018) concluded that understanding the contribution of ECM stiffness to the pathology of disease may improve patient outcomes. This conclusion suggests that the connection between the ECM and the cytoskeleton, and their respective roles in the cellular microenvironment, may provide an entry point for drug discovery in CVDs linked to cardiac fibrosis that are accompanied by a progression of structural changes in the cell.

Dysregulated matrix metalloproteinases (MMPs) and many matricellular proteins, which are extracellular molecules that do not have a direct structural role in the ECM (e.g., thrombospondins (TSPs), tenascin-C, and osteopontin (OPN)), have all been found to be associated with cardiac fibrosis (Table 3). MMPs, a type of collagenase, directly degrade the ECM and its components and also activate growth factors and other molecules related to stiffness processes [13,15]. Interestingly, MMPs are activated during plaque development to help with vessel wall remodeling, thereby presenting a prospective therapeutic target opportunity, some possibilities of which are further explored throughout this review [13,16]. Similarly, many studies found TSP upregulation to be linked with both cardiac fibrosis and cardiac hypertrophy [12,17,18]. In addition, TSP-1 activates TGF-β, of which high levels are linked to cardiac fibrosis by upregulating collagen type I and type III mRNA [15]. In one study, researchers found that an antagonist of TSP-1-mediated transforming growth factor TGF-β activation blocked cardiac fibrosis progression and, by reducing TGF-β activity, cardiac function was notably improved [19]. The absence of tenascin-C is also associated with a decrease in TGF-β signaling and these levels have also been linked to reduced fibrosis and effects on ECM remodeling [12]. Furthermore, the deletion of the tenascin-C gene was found to be associated with decreases in both cardiac fibrosis and inflammation (shown in studies with tenascin-C knockout mice), while upregulation of tenascin-C was seen to indirectly stimulate fibroblasts into producing excess collagen [20]. Computational techniques such as pseudotemporal analyses, use of automatic cell identification tools, or other scRNA-seq analyses provide efficient methods to isolate the effects of the ECM proteins and effectively find paths to target inflammation or plaque buildup in CVDs [21].

Table 3:

Potential Targets Relevant to Cardiac ECM Functions

Target	Function	Relevant CVD Phenotypes	References
periostin	Plays role in the disruption of cytoskeletal architecture	heart failure, pulmonary hypertension	[10,12,80]
Rho and YAP/TAZ	Contribute to ECM stiffening	cardiac fibrosis	[13]
YAP	Regulated by ECM and cytoskeletal integrity	heart failure	[10]
MMPs (collagenases)	Promotes ECM degradation; contributes to ECM stiffness by activating growth factors and other molecules; dysregulation of MMPs are associated with CVDs	cardiac fibrosis	[13]
Thrombospondins (TSPs)	Activation of TGF-β; control inflammation; participate in ECM turnover and preservation; regulates cardiac fibroblasts; inhibit MMP activity	cardiac dilation, myocardial infarction, hypertrophy, cardiomyopathy	[12,91,92]
Tenascin-C	Regulates adhesion; alters actin organization; involved in inflammation processes	myocardial infarction, myocarditis, cardiac hypertrophy	[12]
SPARC	Regulates cell function; regulates tissue remodeling; cell-cycle inhibitor	myocardial infarction, cardiac hypertrophy, cardiac fibrosis	[12]
Osteopontin (OPN)	Regulates cell adhesion; survival signal; activate macrophages; mediator in plaque formation; vascular calcification regulator	atherosclerosis, myocardial infarction, cardiac hypertrophy, cardiac fibrosis, valvular disease	[12,69,71]

Moreover, low SPARC expression is cardiometabolically associated with reduced renal fibrosis [22] and SPARC-mediated cardiac fibrosis was found to be likely caused by growth factor signaling [12]. Additionally, the absence of SPARC was associated with a significant increase in death from higher rates of cardiac rupture and heart failure [12]. On the other hand, high levels of SPARC are found in chronic fibrosis conditions [23], suggesting that a delicate balance and equilibrium in SPARC levels is predictive of health-to-disease transitions.

Furthermore, it has been shown that an increased expression of OPN is correlated with cardiac fibrosis and hypertrophy [12,24,25]. Upregulation of OPN has also been correlated with the development and severity of cardiac remodeling. In induced fibrosis models with OPN-null mice, the researchers found significantly reduced fibrosis along with high levels of ECM network disruption [12,26]. Renin-angiotensin-aldosterone system activation and reactive oxygen species production both appear to be involved in the upregulation of OPN, while inhibiting pathways such as the c-Jun N-terminal kinases pathway and ERK1/2 pathway nearly inhibits OPN upregulation entirely as mediated by angiotensin II [25]. Collectively, these signaling pathways mediate the increase of OPN levels and are prime examples of how further studies of the cardiac ECM and its related pathways can lead to potential novel drug research and development opportunities that address cardiac fibrosis.

2.2. Hypertension

Hypertension, specifically pulmonary arterial hypertension (PAH), has a high morbidity rate with a median survival rate of ~7 years and is associated with other CVDs such as arterial and myocardial fibrosis, making it one of the most significant contributors to the global CVD crisis and an even more important and pressing clinical unmet need for drug discovery efforts [15,27,28]. Hypertension onset and progression is characterized by arterial stiffening (which has been shown to be a predictor of increased mortality in hypertensive patients), an increase in collagen accumulation, and both ECM/actin cytoskeletal remodeling of the arteries and ECM rearrangement of its matricellular proteins and proteoglycans [14,28,29,30,31,32].

Vascular fibrosis is a principal part of hypertensive ECM remodeling and is characterized by the accumulation of collagen and other ECM components in the vessel wall [32]. The ECM plays a significant role in hypertension development and progression because it provides stability in the cell environment through its mechanical components and regulation of ECM protein levels [14,31]. In particular, Cai and colleagues describe how ECM remodeling due to hypertension impacts VSMCs through changes in contractility and substrate structure [31]. Additionally, ECM remodeling changes the underlying vasculature, thereby impacting the vessel wall’s structure and function in response to new environmental conditions and early disease development [28,33]. The expansion of the ECM in all three layers of the pulmonary vascular wall (intima, media, and adventitia) has been found to lead to vascular fibrosis, which causes arterial stiffening and reduced arterial compliance [28,34]. The severity of hypertension and these increased levels of collagen turnover have been linked to an imbalance in proteolytic enzymes such as MMPs, which may contribute to ECM remodeling since they mediate the degradation of cardiac matrix components, suggesting a potential ECM therapeutic target for hypertension [15,32,33,35,36].

2.3. Heart Failure

In a review of phosphoinositide 3-kinase (PI3K) studies, Zhabyeyev and colleagues (2019) suggested that PI3K plays an important role in cardiac remodeling via the cytoskeleton in the context of heart failure. Specifically, the authors observed that PI3K signaling works as a cardioprotective mechanism by promoting a polymerized cytoskeleton, since higher levels of depolymerization are associated with weaker hearts that are susceptible to heart failure [37]. Additionally, ECM expansion is associated with the outcome of adverse events in heart failure patients [38]. In patients with reduced ejection fraction, cardiac fibrosis severity can predict death and other adverse cardiac event outcomes while the expansion of interstitial ECM was frequently seen in patients with preserved ejection fraction [38]. The association between polymerization of the cytoskeleton/ECM and the function of the cardiac muscle points toward the potential for novel therapeutic targets that may serve to strengthen the cytoskeleton/ECM to enhance its cardioprotective effects in response to CVDs such as heart failure.

Investigating the activity of molecules such as gelsolin, a cytoskeletal protein that can affect cardiomyocytes and contributes to advanced heart failure via its role in mediating adverse cytoskeletal remodeling [39], can help identify potential novel therapeutic targets (Table 1). For example, sarcomeric remodeling (e.g., redistribution of α-actinin in mouse myocytes) plays a significant role in the transition from hypertrophy to heart failure through changes in sarcomeric contractility, which is associated with cytoskeletal remodeling, or rearrangement through monomer polymerization and depolymerization [40]. A focus on cytoskeletal abnormalities rather than myofilaments is preferred because previous studies have shown that contractile dysfunction in heart failure is due to changes in microtubular density that affect sarcomere motion [41]. Two types of sarcomeric remodeling are known to play an important role in the development of heart failure by altering ventricular function and contractility. In the first type of sarcomeric remodeling identified in dyssynchronous heart failure (DHF), there was a reduction in the regularity of α-actinin distribution, but the regularity could be restored by cardiac resynchronization therapy (CRT) [40]. The researchers identified a second type of sarcomeric remodeling through observed increases in longitudinal depositions of α-actinin in lateral and anterior left ventricle cells for both DHF and CRT [40]. Similarly, structural remodeling regulated by ECM-associated collagen crosslinking was seen to contribute to heart failure progression [38], while Hutchinson and colleagues (2010) found that in previous studies, a pattern of increase in ECM cell turnover was observed in heart failure.

Table 1:

Potential Gene Targets Relevant to Cardiac Cytoskeletal Function

Target	Function	Relevant CVD Phenotypes	References
HDAC6	Histone deacetylase, represses transcription, mediates tubulin deacetylase (and therefore cell motility), HDAC inhibitors block adverse cardiac remodeling in HF	Heart Failure, Heart Failure with Preserved Ejection Fraction	[138,139]
gelsolin	Actin-severing protein, contributes to increased apoptosis of VSMCs on the arterial wall, mediates adverse cytoskeletal remodeling	Heart Failure, Plaque Rupture, Heart Attack, Stroke	[39, 137]
calponin 2 (CNN2)	Regulation of the motility of macrophages and foam cells	Atherosclerosis	[49]
endoglin	Contributes to cytoskeletal organization, specifically F-actin, and vascular remodeling	Atherosclerosis	[46,47,75]
titin (TTN)	Sarcomeric protein found on the structure between the cytoskeleton and sarcomeric apparatus; contributes to flexibility and stability of the sarcomere	DCM, ARVC	[103]

The progression of heart failure is correlated with the density of the cardiomyocyte microtubule network, which suggests another relationship between CVDs and the cytoskeleton [42]. An increase in the density of this cardiomyocyte microtubule network is associated with hypertrophy leading to heart failure [42]. Ali and colleagues (2020) also noted that the density of the cardiomyocyte microtubule network indicates that the network has adaptive abilities based on the ventricular load and environment. For example, researchers found that microtubule density increases in the early postnatal period and then begins to decrease after reaching a maximum on postnatal day 5. The microtubule network density continues to decrease until it reaches a steady-state level around what is typically seen in adult hearts around postnatal day 14 [42]. This observation supports the idea that the cytoskeleton is able to adapt to varying environmental conditions and, therefore, plays a key developmental role in promoting the progression of pathophysiology during CVD development. Thus, Ali and coworkers (2020) concluded that the cytoskeleton and the sarcomere interfere with replication of cardiomyocytes after birth, thereby preventing regeneration of injured heart tissue later in life [42]. These results indicate that the mechanism by which the cytoskeleton prevents cardiomyocyte replication could be a viable therapeutic avenue to pursue for CVD drug discovery, specifically in developing regenerative therapies.

2.4. Atherosclerosis

Atherosclerosis, the main cause of coronary artery disease (CAD), is characterized by changes in blood vessel morphology that eventually cause arterial wall remodeling and plaque formation [16]. These fibrotic plaques are stabilized by the ECM produced by vascular smooth muscle cells (VSMCs), while dysregulated MMPs derived from macrophages destabilize them [43,44]. One concrete example is vulnerable plaque, which is typically characterized by large amounts of lipids and thin fibrous caps and describes plaque that may either rupture or erode depending on its biological composition [16,44,45]. Eroded plaques were found to be rich in ECM components (e.g.,proteoglycans and glycosaminoglycans) rather than lipids [44]. Plaque rupture can lead to additional complications such as myocardial infarction (MI) [43] and, although the mechanisms behind plaque erosion have not received as much attention as those that lead to rupture in the past few decades, further exploring and understanding both, especially plaque erosion, has the potential to lead to novel therapeutics [45].

The ECM could be a potential therapeutic target in order to stabilize atherosclerotic plaques to lower risk of rupture and further clinical complications, seen by the effects of proteins such as endoglin and calponin 2 (Table 1). A previous review of literature on endoglin by Nachtigal and colleagues (2012) suggested that it plays an important role in atherosclerosis, and its expression may predict the stability of the plaque phenotype [46]. More recently, Garzon-Martinez found that endoglin is associated with many cardiovascular risk factors including LDL-cholesterol and hemoglobin levels [47]. Moreover, in a study of 244 patients with atherosclerosis, soluble endoglin levels were decreased in patients with CAD, showing an inverse association between soluble endoglin levels and the severity of atherosclerosis [48]. On the other hand, calponin 2 has been seen to regulate macrophages and play a role in reducing atherosclerosis [49]. The researchers also found that deleting the calponin 2 gene leads to increased macrophage motility along with lower levels of inflammatory cytokines. These two results have been observed leading to decreases in the area of ApoE deficiency-caused atherosclerotic plaques [49]. Overall, both endoglin and calponin 2 are prime examples of how further studies into the cytoskeleton/ECM and its related molecules could potentially lead to the discovery of novel therapeutic targets for CVDs.

Increasing improvements in proteomics-based approaches provide an avenue for more sophisticated analyses of cytoskeletal proteins. However, many of these have not yet been analyzed in a large cohort or at multiple different temporal stages of atherosclerotic plaque development. For example, the work of Stakhneva et al. (2019) provided a preliminary look at changes in cytoskeletal gene expression and explained the need for a large-scale analysis of cytoskeletal proteins related to atherosclerotic plaque. The researchers found that stable plaque was characterized by increased levels of cytoskeletal proteins such as vimentin, actin, tropomyosin, and tubulin [50]. These findings establish the need for further investigation into the biological role of increased gene expression of these respective proteins and how it may promote plaque development. In a different study, researchers observed the dynamic nature of the cytoskeleton by investigating the effects of ECM proteins on the migration distance of VSMCs and the organization of the actin cytoskeleton architecture [51]. Although the focus of the study was on the effects of changing ECM composition, specifically changes in type I collagen and expression of fibronectin that lead to changes in VSMC migration, the results showed an association between high VSMC migration distance and elevated disorganization of the actin cytoskeleton architecture [51]. These results indicate the possibility that underlying changes in gene expression contribute to the increased disorganization of the actin cytoskeleton architecture, causing faster VSMC migration and eventually increased buildup of atherosclerotic plaque. In summary, the researchers reported: “These results support our hypothesis that ECM stiffness and composition coordinate to regulate cell migration dynamics and cytoskeleton architecture. Our results also provide further insight into the effects of the heterogeneous microenvironments seen in aging and atherosclerotic plaques on disease progression”, highlighting the importance of understanding both ECM and cytoskeletal dynamics in the manifestation of CVD pathophysiology [51]. Ma and colleagues also used advanced computational tools to analyze changes in gene expression within the atherosclerotic microenvironment. Their single-cell RNA-seq analysis allowed them to gain insights into vascular cell differentiation, specifically the role of VSMC differentiation into cell types similar to fibroblasts and chondrocytes that are capable of expressing signs of inflammation and extracellular matrix degradation, along with insight on the communication pathways among those cells [21]. Of interest, this study identified several already-approved drugs as capable of disrupting communication between those smooth muscle cells and fibroblast cells. For instance, the authors identified several approved chemotherapies targeting the epidermal growth factor receptor (EGFR) signaling cascade, which hold potential for drug repurposing. Understanding how this communication or ‘crosstalk’ among cells works could help scientists better understand the complexities of atherosclerosis and how plaque begins to develop. A cardioinformatics investigation into the dynamic role of individual cytoskeletal or ECM-related proteins may provide further insight into the temporal progression of heart disease (e.g., to study whether ECM composition changes occur prior to the development of atherosclerotic plaques, which may serve as a prognostic biomarker of disease onset).

Arterial remodeling, specifically the increase in arterial stiffness, is mostly attributed to changes in both the ECM and structure of vascular smooth muscle cells (VSMCs) [52]. For example, previous research highlights the importance of the signaling molecule, NF-κB, in its involvement with VSMC remodeling in arteries and also in reactive oxygen species production that affects the regulation of the actin cytoskeleton [53,54]. Xu and colleagues (2017), however, explain that although reactive oxidative species have been involved in the results of many studies on human coronary arteriole dilation as well as vasodilator signaling, there has not been sufficient investigation into how redox regulation of the actin cytoskeleton contributes to clinical outcomes. DNA methyltransferases 1 (DNMT1) inhibition was found to lead to the calcification and cellular VSMC stiffening in vitro along with arterial stiffening, suggesting DNMT1 regulates VSMC contractility in response to environmental conditions such as arterial stiffness [52]. Xie and coworkers (2018) suggested that DNMT1 inhibition may have these effects due to its role in also regulating the promoter activities of transgelin, a TGF-β inducible gene involved in regulating human skeletal stem cell differentiation through the organization of the actin cytoskeleton [55], and αSMA. By studying the over- or underexpression of certain cytoskeletal genes like DNMT1, as well as their mutational landscape using secondary analysis of existing genetic datasets (e.g., GWAS), therapeutic solutions for arterial stiffening (and other CVD pathophysiology that modulates this regulation) are on the horizon.

In general, plaque development disrupts the cytoskeleton as gene expression levels of cytokines, such as IFN-γ and TNF-α, cause the reorganization of actin and tubulin cytoskeleton and lead to the opening of gaps between neighboring cells [43]. The cytokines also influence the permeability of macromolecules such as low-density lipoprotein (LDL), which are then retained by the accumulation of ECM at arterial sites where plaques tend to form due to disturbed blood flow in areas such as the inner curvature and branch points [43]. Investigating further the molecular specifics of abnormal cardiac cytoskeletal remodeling in atherosclerosis, especially in the context of gene expression and cellular biomarkers, can provide insight into how to manipulate remodeling using the cytoskeleton as a drug target to stabilize plaque and prevent plaque rupture and erosion. A computational investigation using a cardioinformatics framework [5] of this remodeling can provide a deeper understanding of the cytoskeleton’s role in atherosclerosis and other CVDs resulting from high arterial shear stress [56]. Specifically, the expression levels of cytoskeletal genes in cells involved in atherosclerotic plaques can be closely studied for a correlation between shear stress levels and plaque rupture risk. The use of computation within the context of atherosclerosis is relatively new but has already shown promise by aiding the derivation of an equation that allows the peak stress metric of plaque to be estimated and therefore provide insight into plaque rupture risk [56]. Doradla and coworkers (2020) hypothesized that greater stress measurements were due to plaques with irregular structures, which supports the possibility that abnormal cytoskeletal protein expression could be promoting increased stress on plaque or is associated with the structural irregularities of plaques with high rupture risk. It is clear that cytoskeletal functionality in both atherosclerosis and heart failure has the potential to be used to assess the progression of CVDs. However, previous research [6,57], which focused on the regulation of the cytoskeleton and its effects on cardiac function, points toward an unfilled gap in literature discussing the intersection of cytoskeleton/ECM and cardiovascular therapeutics.

3. Investigating Therapeutic Targets

3.1. Toxicity and Safety of Targeting the ECM/Cytoskeleton

There have already been some studies into the toxicity and safety of therapeutics that target the ECM and cytoskeleton (Table 2). For example, in the COLchicine Cardiovascular Outcomes Trial (COLCOT), colchicine (which is further explored later in this review) “showed unexpectedly little gastrointestinal intolerance. As in CANTOS, colchicine treatment was associated with a small but significant increase in infections, notably pneumonia” [58]. Additionally, the U.S. Food and Drug Administration has approved some HDAC inhibitors (HDACi), such as vorinostat, panobinostat, and belinostat, for treating some cancers [59]. As of the publishing of Gillette, 2021, there were at least 60 active clinical trials on clinicaltrials.gov and the safety of these drugs have led to preclinical testing for use in other diseases such as heart disease. Trastuzumab is another cancer drug known to interrupt the cytoskeleton which did not show significant histological changes in the hearts of patients treated with it. Trastuzumab-induced cardiotoxicity was classified as Type II due to the reversibility of cardiac dysfunction in patients [60]. Although further research is needed into the possible cytotoxicities of targeting specific aspects of the cellular architecture, it is promising to see examples of existing ECM/cytoskeleton targeting drugs that are safe enough for use and clinical trials.

Table 2:

Drugs/Therapeutics Relevant to Cardiac Cytoskeletal Function

Drug	Function	Relevant CVD Outcomes	References
Colchicine	Targets microtubules and inhibits neutrophils	Improves hypertrophy and systolic function; reduces risk of CV events	[118,121,122]
Omecamtiv Mecarbil	Myosin activator that increases cardiac contractility via increased calcium sensitivity	Lowered risk of heart failure or CVD death by 8% in patients with a mean LVEF of 26%	[125,126]
Mavacamten	Cardiac myosin inhibitor	Reduced LVOT gradient and increased pVO2 in patients with obstructive hypertrophic cardiomyopathy	[123,127]

3.2. Apoptosis

As mentioned above, abnormal cytoskeleton and ECM rearrangements can induce apoptosis within a cardiovascular disease context. The study on Taxol by Michaelis and coworkers (2005), which showed that a microtubule-stabilizing drug was able to prevent cell death, foreshadow future studies into the interface of CVD and apoptosis caused by dysfunctional cytoskeleton/ECM remodeling in search of novel therapeutic targets.

Cardiomyocytes undergoing apoptosis have been found in patient tissue samples of those suffering from myocardial infarction, dilated cardiomyopathy, and heart failure, suggesting that apoptosis holds a role in the pathogenesis of various CVD phenotypes and inhibiting apoptosis could be cardioprotective, especially in regards to heart failure [61,62]. Additionally, a number of papers point towards multiple stressors like cytokine production, oxidative stress, and DNA damage in CVDs that likely induce apoptosis in cardiomyocytes [61–63,64], along with cytoskeletal disruption [65]. Specifically, Suria and coworkers (1999) found that the integrity of the cytoskeleton potentially regulates the compartmentalization of death effector molecules, such as cytochrome c and apoptosis-inducing factor, and so disruptions may lead to their release or activation. The composition of the ECM has also been linked to apoptosis rates [66,67]. For example, Richter and Kostin (2015) found that the quantity of telocytes, interstitial cells with very long and thin extensions, was influenced by ECM composition and there were higher rates of telocyte apoptosis in diseased and dying hearts compared to normal ones and those telocytes showed extreme degenerative structural changes. Overall, apoptosis is such a highly regulated process [61] that studying the interface of apoptosis, the cytoskeletome, and CVDs could potentially lead to the discovery of novel therapeutic targets for various CVD-relevant phenotypes.

3.3. Gelsolin

Potential therapeutic targets can be discovered by investigating the activity of molecules such as gelsolin, an actin-severing protein that affects cardiomyocytes and contributes to advanced heart failure by mediating adverse cytoskeletal remodeling [39]. As previously discussed, recent cardiovascular disease drug discovery efforts have traditionally focused on targeting the inflammatory processes that contribute to plaque buildup. The actin cytoskeleton provides a relatively new perspective on the progression of CVDs such as CAD. For instance, Patel and colleagues investigated the suppression of gelsolin and found that it leads to decreased actin polymerization because gelsolin caps the barbed ends of actin filaments and is also an actin-severing protein [39]). The researchers also discovered that phosphatidylinositol (3,4,5)-trisphosphate (PIP3) and PI3Kα, a PIP3-producing enzyme a part of the PI3K family, are negative regulators of gelsolin and prevent the disruption of actin cytoskeleton in cardiomyocytes, thereby preventing reduced cardiac contractility [39]. In addition to the suppression of gelsolin, PI3K signaling is involved in other signaling pathways that contribute to cardiac cytoskeleton remodeling. Specifically, Fan and colleagues (2014) demonstrated that PI3K signaling increases phosphorylation of AKT, which leads to the downregulation of α-SM actin, and that phosphorylation could be inhibited by LY294002, a chemical compound able to inhibit a wide range of proteins [68]. The researchers concluded that this pathway would be a good therapeutic target for PAH, which is notable considering the aforementioned role of the PI3K/AKT pathway in cytoskeletal organization [68]. Taken together, these findings suggest that gelsolin is a potentially tractable therapeutic target for CVDs.

3.4. Osteopontin

The matricellular protein osteopontin (OPN) is known to be a survival signal by inhibiting apoptosis [12] and it is also known to be involved in tissue remodeling and inflammation [69]. These proteins play an important role in ECM remodeling and researchers have found that low levels of OPN cause LV dilation [70]. Additionally, OPNs interact with fibronectin and collagen types I-V while also modulating adhesion to such ECM proteins [70]. OPNs regulate inflammatory cells by activating macrophages and studies have shown that OPNs are an important mediator in the development of plaque formation [12]. Specifically, OPNs may be a regulator of TGF-β [71], which suggests that OPNs could be an effective ECM drug target through association with the ECM via TGF-β (as seen with TSP-1 (see previous discussion)).

OPNs were also found to act on ECM formation mechanisms, which have not been studied precisely to-date. Some of these mechanisms include protecting the cells from adverse remodeling and possibly promoting matrix deposition [12]. Myers and colleagues (2003) found in their studies that OPN-null mice had altered ECM remodeling, exhibiting worse dilative remodeling post-MI relative to wild-type mice. The carotid arteries of OPN-null mice were much smaller, indicating dysregulated remodeling, compared to wild-type mice. Overall, the researchers found that OPN regulates vascular remodeling responses and indirectly affects the ECM, which suggests that OPNs could be another class of promising cardiac ECM therapeutic targets.

3.5. Calponin 2

Calponin 2, a gene which encodes a protein associated with actin filaments and myosin ATPase, is another potential cytoskeletal therapeutic target because of the role it plays in reducing atherosclerosis through regulation of macrophages [49] as briefly mentioned above. Deletion of the calponin 2 gene leads to increased motility of macrophages and foam cells as well as decreased levels of inflammatory cytokines. Previous studies have demonstrated that the combination of these two effects leads to an observed decrease in the area of ApoE deficiency-caused atherosclerotic plaques [49]. The exact mechanism by which the cytoskeleton is involved in the increased motility of macrophages and foam cells remains unknown, but Liu and Jin suggest that the deletion of calponin 2 is able to unblock the myosin II motor, which connects calponin 2 to the regulation of actin cytoskeleton functions [49]. The deletion of calponin 2 in the experiment by Liu and Jin (2016) is significant because it raises the question of whether the plaque environment contributes to changes in the expression of calponin 2 in CAD patients.

More recent research further supports the potential for calponin 2 to be an effective cytoskeletal target for cardiovascular diseases. Given the shared characteristics between atherosclerosis and calcific aortic valve disease (CAVD), the most common heart valve disease in the western world, Plazyo and colleagues (2018) investigated the role of calponin 2 in CAVD [72]. CAVD is characterized by fibrotic thickening along with both aortic sclerosis and stenosis, which eventually leads to congestive HF, but despite its common occurrence, there were no non-surgical treatments, hence the researchers’ focus on calponin 2 [72]. The results of this study showed that deletion of the calponin 2 gene reduced calcification in ApoE knockout mice [72]. Plazyo and colleagues concluded that calponin 2’s role in the differentiation of aortic valve interstitial cells (AVICs) is a major step in the development of CAVD [72]. These findings provide additional support to the hypothesis that the cytoskeleton is dysregulated during CVD progression. In summary, Plazyo and coworkers (2018) concluded: “Because calponin 2 acts as a direct modulator of actin and actin-dependent cellular functions, targeting calponin 2 may be a way to manipulate cytoskeleton activity for altering mechanoregulation in the treatment of diseases involving dysregulated mechanical signaling, such as CAVD. Thus, this therapeutic promise merits further investigation”. The aforementioned findings are notable because of calponin 2’s direct role in cytoskeletal function. Hossain and coworkers (2005) suggested that calponin 2 interacts with actin cytoskeleton function through its association with tropomyosin, which is colocalized with calponin 2 in actin stress fibers [73]. The study also showed that calponin 2 protects actin filaments from cytochalasin B, which inhibits polymerization by binding to the barbed end, and demonstrates the stability of the actin filaments [73]. Overall, these studies suggest that calponin 2 is a viable cytoskeletal drug target that merits further research in other CVDs.

3.6. Endoglin

Endoglin, a protein involved in cytoskeletal organization and vascular remodeling, can be modulated to prevent dysregulation of plaque morphology. It also encourages the expression of extracellular matrix proteins (e.g., type I collagen) and promotes cardiac hypertrophy and cardiac fibrosis [74]. In one study, researchers found that F-actin was bundled in the absence of endoglin expression, while endoglin expression with ZRP-1 in endothelial cells led to an organized cytoskeleton and no accumulation of F-actin [75]. Additionally, a common feature of hereditary hemorrhagic telangiectasia (HHT) cells with low endoglin levels were the disorganization of the F-actin cytoskeleton (depolymerization of actin fibers) when compared to healthy blood outgrowth endothelial cell cultures [76] which further suggests an association between endoglin and the cytoskeleton. In other studies reviewed by López-Novoa and Bernabeu (2010), endoglin was found to “oppose TGF-β1-dependent responses such as the inhibition of cellular proliferation, the expression of the ECM proteoglycan lumican, as well as the increased expression of ECM components, including… collagen”. Endoglin has also been found to inhibit the accumulation of ECM components such as basal and TGF-β1- induced collagen, and α2 (I) collagen mRNA expression [77]. In conclusion, these observations suggest that endoglin is a potentially effective cytoskeletal target given its documented contributions to cytoskeletal organization.

3.7. Periostin

Periostin, an ECM protein typically expressed very early in embryogenesis and not in normal adult myocardium, is expressed in certain diseased adult hearts and functions in regulating cardiac remodeling and hypertrophy [78,79]. The researchers found that a lack of periostin was associated with short-term issues such as greater ventricular rupture after a myocardial infarction event and long-term benefits such as improvement in cardiac function within the animal subjects that survived [79]. The greater risk in ventricular rupture, at least for acute myocardial infarction, was attributed to a few abnormal characteristics such as a reduced number of αSMA–positive cells and impaired collagen fibril formation [78]. Oka and colleagues (2007) concluded: “… inhibition of cardiac fibrosis benefits the long-term remodeling process and preserves cardiac function to a greater extent; in the short term, it can predispose to rupture by compromising scar formation.” Additionally, an experiment investigating the role of periostin in pulmonary arterial hypertension found that it disrupted cytoskeletal architecture in endothelial cells as well [80,81]. Periostin is involved with post-injury tissue remodeling and the researchers concluded that it may be a useful potential therapeutic target given that increased periostin expression was associated with worsened clinical outcomes and increased severity of PAH, aligning with the results Oka and colleagues found [80]. Overall, these results affirm the importance of investigating the connection between cytoskeletal remodeling and disruption in various CVDs while showing the potential for periostin to be a viable CVD therapeutic target.

Periostin has also been shown to induce cardiac regeneration through switching differentiated cardiomyocytes back into the cell cycle [82,83]. From a developmental biology and tissue renewal perspective, progenitor cells influence and affect ECM changes in diseased hearts [84]. The ECM’s composition and properties are known to play a role in cardiac development and regeneration [82], which significantly changes in diseased hearts as observed during the scarring and stiffening of myocardial tissue [84–86]. For example, Bayomy and coworkers (2012) concluded after a systematic review of multiple studies that the injured heart’s limited regeneration ability may derive from processes inhibiting progenitor cell differentiation rather than simply a lack of cells [84]. Qiu and colleagues (2015) also found that not only did injured hearts have a normal quantity of cardiac progenitor cells compared to healthy hearts, a higher quantity was seen in some injured cases, further strengthening this biological hypothesis. Additionally, Ozcebe and coworkers (2021) found that a younger ECM supported proliferation and stress-coping abilities, even in aged cells, while an aged ECM led to a decrease in cardiac function. These findings agree with how age is considered one of the top risk factors of CVD [87], as it lowers the threshold for CVD manifestation [88]. Therefore, further exploring periostin and its involvement in cardiac tissue regeneration along with the cellular microenvironments of healthy and aging hearts would potentially help understand the relationship between the ECM and the behavior of cardiomyocytes, paving the way for more and improved regenerative cardiac therapies [85,86,89,90] that are motivated by precision medicine approaches focusing on longevity pathways.

3.8. Thrombospondins

Various ECM-specific targets can be discovered by studying matricellular proteins such as the thrombospondins (TSPs). TSPs are known to bind to structural components in the ECM network and have de-adhesion properties by inducing the loss of focal adhesions [12]. A study found that TSP-1 may exhibit cardioprotective effects since TSP-1 levels are significantly higher in pressure-overloaded myocardium through preventing cardiac remodeling and chamber dilation [91]. TSP-1 null cases were found to have higher incidence rates of early hypertrophy; for example, Xia and coworkers (2011) found that TSP-1 null mice had a larger increase in left ventricular mass compared to the wild-type mice. These findings support the role of TSP-1 as a potential drug target to explore in clinical indications such as cardiac hypertrophy and fibrosis.

TSP-1 activates TGF-β [92], which holds an important role in inflammatory processes and inhibits MMP-9 [91]. Xia and colleagues (2011) found that introducing TSP-1 in the cardiac ECM protects against adverse matrix remodeling and thereby preserves the ECM, most likely due to mutual synergistic biology between TSP-1 and both TGF-β and MMPs. TSP-1 activates TGF-β which then binds to another protein, latent TGF-β binding protein (LTBP), to form a large latent complex that is known to covalently associate with the ECM [12]. In addition to its association with TGF-β, TSP-1 inhibits MMP activity, which was seen to increase ECM preservation and therefore prevent chamber dilation [91]. Inversely, an increase in MMP-2 expression and MMP-9 activity was detected in TSP-1 null pressure-overloaded hearts, which possibly led to adverse remodeling by promoting matrix degradation [91]. The aforementioned findings suggest that TSP-1 could be an effective therapeutic target through indirectly targeting the ECM.

In another study, TSP-2 null animals were found to have an increase in the incidence rate of cardiac rupture, demonstrating that TSP-2 may have an important role in the structure of the ECM, specifically in preserving it [91]. This suggests that other members of the TSP family, such as TSP-2, could be viable ECM therapeutic targets for cardiovascular diseases.

3.9. Tenascin-C

In the group of tenascin proteins, only tenascins C and X are considered to be matricellular proteins [12] and are expressed during certain conditions such as tissue repair and inflammation [93]. Tenascin-C, specifically, binds to many ECM molecules and is also stimulated by angiotensin II, which regulates cardiac remodeling [12]. It is also significantly expressed in CVDs and inflammatory and cardiac tissue remodeling processes (12,94,95]. Similar to TSP-1, tenascin-C has a role in regulating adhesion and promotes de-adhesion and weak cell attachment by inducing focal adhesion loss [12]. Another similarity to TSP-1, tenascin-C interacts with MMP-2 by being cleaved by it [12]. This suggests relationships between these matricellular proteins and intersections of biological pathways and molecular mechanisms, which may increase the number of viable cardiac ECM therapeutic targets not yet discovered to-date. It has been known for some time that tenascin-C alters actin organization in fibroblasts by suppressing Rho activation as a response to fibronectin [96]. More recently, tenascin-C was seen to also affect the ECM, as cells within a tenascin-C containing matrix were not able to cause ECM contraction [12], further strengthening the possibility of tenascin-C being a viable ECM target and a CVD biomarker [94–95]. It was also found that the loss of tenascin-C led to the loss of dilative remodeling in an infarcted heart [12] and, in contrast, high tenascin-C levels were able to predict later development of dilative remodeling [97]. Taken together, tenascin-C is another matricellular protein that appears to be a promising cardiac ECM therapeutic target to treat cardiovascular diseases.

4. Further Exploration of Cytoskeletal Genes

The importance of research into early diagnostic markers of CVD progression has been emphasized in recent publications, but there are still relatively few studies that take a data-science oriented approach [98,99]. The cellular environment of plaque formation is dynamic and supports the notion that plaque formation can be accelerated by changes in cytoskeletal protein expression or associated ECM remodeling. For instance, Martin-Lorenzo et al. (2016) observed changes in proteins involved in cytoskeletal organization, including downregulation of vinculin isoform 2 (VCL) and upregulation of integrin linked kinase (ILK). The combined effect of VCL and ILK contributed to the disorganization of actin cytoskeleton as well as to changes in intermediate filaments and microtubules that led to increased cell migration in rabbit models, suggesting there is significant evidence that cytoskeletal dysregulation is linked to atherosclerosis [98].

Another large-scale analysis included multiple cytoskeletal and lipid metabolism-related proteins, including other proteins associated with atherosclerosis, collectively representing a systematic review of existing proteomic studies that resulted in a selection of 76 proteins to be used as mechanism-related biomarkers for at-risk plaque prognosis [100]. Specifically, Eslava-Alcon and coworkers (2020) analyzed actin cytoskeleton proteins involved with organization in the atherosclerotic plaque tissue and secretome and found that these proteins, which included vimentin, gelsolin, filamin A, tubulin-beta, vinculin, HSP27, calponin, and transgelin, were down-regulated in secretome and atherosclerotic tissues [100]. The researchers also provided proteomic evidence in support of the potential role of the previously discussed actin protein, calponin, in hyperplasia, suggesting that: “genetic restoration of Calponin in the artery wall may be of potential value for CVD” [100]. Many types of CVD phenotypes occur as a complication of atherosclerosis, so the uncertainty of the role that these cytoskeletal proteins play in atherosclerosis, as well as the potential for cytoskeletal proteins to be useful therapeutic targets, establishes the need for a closer investigation of actin cytoskeleton gene expression levels in atherosclerotic tissues.

Aside from analyses of cytoskeletal protein levels, previous studies [6,47] point towards the need for investigation of SNPs in cytoskeletal genes [47]. Specifically, changes in cytoskeletal proteins associated with different stages of atherosclerosis or heart failure can be investigated through studies of cytoskeleton gene expression quantitative trait loci (eQTLs). A previous eQTL analysis on left atrial appendage tissue from patients who had surgery for atrial fibrillation or other CVDs found multiple GWAS SNP eQTLs in genes associated with the cytoskeleton. The third strongest eQTL found was for a gene associated with the nuclear envelope and the cytoskeleton, while the strongest eQTL observed was for the genes synaptopodin 2-like (SYNPO2L) and myozenin-1 (MYOZ1). Interestingly, both of these genes interact with cytoskeletal proteins, but SYNPO2L specifically encodes cytoskeletal heart-enriched actin-associated protein (CHAP), which affects cardiac contractility and sarcomere function [101].

Likewise, the study of genetic variants in dilated cardiomyopathy (DCM) has led to the discovery of mutations in over 33 genes that encode for proteins involved in sarcomere, Z-disc, and cytoskeleton function [102]. One example is the mutations associated with the gene TTN, which encodes the protein titin [103]. These mutations have been shown to cause DCM, because titin directly contributes to sarcomere structure and has decreased thick filament binding sites when it is truncated due to genetic variation [103]. Similar to the contribution of the newly discovered variants implicated in DCM, close inspection of variants in cytoskeletal genes of individuals with atherosclerotic plaques can provide clues toward effective therapeutic drugs that may directly prevent plaque rupture and erosion [102]. In conclusion, close studies of genes that affect cardiac cytoskeletal remodeling, for example in monogenic disorders involving cardiomyopathies, is promising because it allows for an understanding of how the cytoskeleton and the sarcomere contribute to normal functioning of cardiac muscle [104].

Overall, the use of data science-oriented computational techniques to understand changes in cytoskeletal gene expression that can lead to a more conducive environment for developing atherosclerotic plaque has the potential to impact the drug discovery field for CVDs at a broader scale. For example, the use of genomic studies previously led to the discovery of novel therapeutic targets for atherosclerosis [105], demonstrating the importance of investigating lesser known cytoskeletal genes for potentially more impactful and heretofore undiscovered targets and undervalued assets. Some studies have suggested that a computational approach has great potential to aid in relevant biomarker discovery and repurposing efforts that advance precision cardiology [5,106] Similarly, a large-scale cardioinformatics analysis of cytoskeletal gene expression and mutational landscape at all stages of plaque development may provide another metric to detect plaque rupture risk or diagnose unstable plaque in its earliest stages, when it can still be intervened upon with preventative measures such as physical exercise, healthy diet changes, and smoking cessation [107].

Although there have been systematic reviews published that discuss the role of the actin cytoskeleton in inflammatory cells within a CVD environment, there have been few studies involving detailed meta-analyses, large cohorts, or profiling of cellular gene expression, despite the fact that a general hypothesis regarding cytoskeleton drugs has been put forth before [6]. Thomas and Advani (2006) draw attention to it by explaining that: “As understanding of the mechanisms of cell responses in atheromatous plaque develops there may be potential to develop more sophisticated therapeutic agents to target malfunctioning actin filament proteins.” Fortunately, recent advances in the emerging field of cardioinformatics now allow for meta-analysis studies to span a large enough cohort size to discern measurable differences in plaque rupture risk with sufficient statistical power. Cytoskeletal changes provide a novel angle to analyze the environment of atherosclerotic plaque, and gene expression is a parameter that can be studied across diverse patient scenarios. A close analysis of cytoskeletal proteins in a large cohort provides the data scale necessary to establish a set of robust biomarkers to diagnose plaque at risk of rupture at its early stages based on transient but temporally consistent changes in gene expression levels. As of this writing, there have been no studies focused on analyzing protein levels simultaneously in patients who have different stages of plaque development and are influenced by different factors such as gender or race [100].

5. Further Exploration of Epigenetics

Cardiac development and homeostasis are known to be influenced by epigenetic changes and miRNAs, non-coding RNA molecules involved in post-transcriptional gene silencing [108]. Epigenetics are heritable changes in gene expression that are not caused by DNA sequence changes, such as histone modifications (e.g., acetylation, phosphorylation, and ubiquitylation) and DNA methylation [5,108]. Dysregulated epigenetic processes have been frequently associated with the pathogenesis of many diseases (e.g., dysregulated miRNA expression in CVD pathogenesis, chromatin structure alterations were shown to lead to heart failure [109]), which suggests that determining an individual’s epigenetic makeup and targeting both miRNAs and related networks could lead to the discovery of novel immunotherapies and therapeutic targets [5,108,110].

Although it is not yet clear how to target miRNAs in both an effective and safe way [110], there are a number of studies that point towards miRNAs as strong potential therapeutic targets for CVDs. For example, miR-155–5p, miR-33, and miR-223–3p all have studies backing up their roles in CVD pathogenesis (e.g., inhibiting miR-33 stimulated processes associated with plaque remodeling leading to contraction) [110–112]. However, the exact mechanisms behind how miR-33 affects the plaque microenvironment is yet not clearly understood [111]. Additionally, Liu and coworkers (2018) examined how overexpressing miR-223–3p and lower levels of ITGB3 (as miR-223–3p suppresses ITGB3 expression) improved PAH symptoms and attenuated pulmonary vascular remodeling. Overall, these findings further suggest how future studies into miRNAs and epigenetic gene expression could lead to the discovery of potential novel targets.

In a study of genetic changes associated with carotid atherosclerosis based on transcriptomic (RNA-seq) data, researchers found increased expression of ITGA11, ITGB2, and ITGB1 in the actin cytoskeleton, suggesting its involvement in developing carotid atherosclerosis. The results of this genomic analysis led to initial observations of miR-19B and miR-19A, implicating these microRNAs as biomarkers of carotid atherosclerosis [99]. Thus, closer studies focused on changes to the cytoskeleton seen in atherosclerotic tissue, rather than genetic changes associated with atherosclerosis in general, have the potential to reveal additional specific cytoskeletal protein targets.

Additionally, many studies into the interactions between histone deacetylases (HDACs) and miRNAs are revealing HDACs’ roles in CVDs as some miRNAs are known to target epigenetic modulators such as HDACs [108]. For example, Travers and coworkers (2021) found that inhibiting HDACs with ITF2357/Givinostat, a clinical-stage HDACi, blocks ECM remodeling while increasing left ventricular compliance in a context of diastolic dysfunction with a preserved ejection fraction. Additionally, other HDACis were found to inhibit both inflammation and fibrosis along with regulating the composition of the ECM [59]. Li and colleagues (2021) discovered that HDAC6 activation disrupted the microtubule network which is associated with structural remodeling and the promotion of atrial fibrillation. Although the molecular mechanisms underlying how this disruption leads to contractile dysfunction is not clear [82], this suggests that HDAC6 is a potential therapeutic drug to target the cytoskeleton in CVDs. There are many other examples of regulatory relationships between miRNAs and HDACs, especially within the context of atherosclerosis [108,113–115]. In conclusion, studying miRNAs and its regulatory gene expression pathways within the context of CVDs could lead to the discovery of novel therapeutic targets such as members of the HDAC family.

6. Drug Development

6.1. Colchicine

Previous studies have shown that inflammation is a major driver of atherosclerotic plaque formation and rupture [6,57]. Inflammation is caused by endothelial injury that begins the process of atherosclerosis, and the inflammation process itself is associated with the cytoskeleton through the activation of NADPH oxidase, which regulates the actin cytoskeleton [6]. Furthermore, low shear stress has also been shown to promote inflammation processes and endothelial cell proliferation that is conducive to atherosclerosis and the formation of vulnerable plaque [116]. In addition, Thomas and Advani (2006) suggested that polymorphisms in actin filament protein genes may contribute to differences in plaque development across populations, supporting the hypothesis that the structure of the cytoskeleton may impact the level of shear stress in endothelial cells differently across different racial/ethnic groups. Although multiple studies have found that shear stress promotes inflammation and endothelial cell proliferation through various pathways such as WNT, Notch, BMP-TGFβ, and many others [116], there has been no investigation into whether cytoskeletal regulation is also altered by changes in gene expression due to the pro-inflammatory environment.

Investigation of cytoskeletal remodeling is important within the CAD context because it will allow for the discovery of specific targets that can alter and help stabilize the plaque environment. Recent research supports the hypothesis that the cytoskeleton is a potentially important drug target for atherosclerosis because the microtubule-SMC pathway can be targeted to prevent the contractile and mitochondrial dysfunction that leads to atrial fibrillation. The results revealed that stabilization of the microtubule network prevented SMC reduction, mitochondrial dysfunction, and tachypacing-induced contractile dysfunction [117].

Currently, the most effective therapeutic target for microtubules is colchicine, which is an anti-inflammatory drug that alleviates hypertrophy and improves systolic function [118]. Specifically, colchicine targets polymerization of microtubules through binding to tubulin -- at low doses, it is able to prevent polymerization, but at high doses, it actively causes depolymerization -- and treats atherosclerosis by lowering inflammation while used in conjunction with statins [119,120]. In one study where participants were recruited within 30 days after a myocardial infarction, 0.5 mg of daily colchicine significantly reduced the risk of ischemic cardiovascular events compared to the placebo [121]. Furthermore, in the COLCOT, colchicine showed to be most effective when administered within the first 3 days after a myocardial infarction. The study also found a 48% reduction in the risk of adverse cardiovascular events including CV death, resuscitated cardiac arrest, MI, stroke, or urgent hospitalization for angina requiring coronary revascularization compared with placebos [122]. Furthermore, studies of patients with acute and chronic coronary disease showed that colchicine treatment resulted in a reduction in the median of serum expression of proteins, including those not directly involved in inflammatory pathways (e.g., NLRP3 inflammasome pathway), and in rates of adverse events such as MI and cardiovascular death compared to placebos [58,123,124]

6.2. Other Drugs and Therapeutics Relevant to the Cardiac Cytoskeleton and ECM

Besides colchicine, there are currently a few other drugs such as omecamtiv mecarbil and spironolactone that target the cardiac ECM or are relevant to cardiac cytoskeleton function (Table 4). Omecamtiv mecarbil is a myosin activator that has been shown to increase cardiac contractility through increased calcium sensitivity and slightly lower HF risk or CVD death in patients with a mean LVEF of 26% [125,126]. Additionally, mavacamten is a cardiac myosin inhibitor and studies have shown it to reduce LVOT gradient and increase pVO2 in patients with obstructive hypertrophic cardiomyopathy [123,127] (Table 2).

Table 4:

Cardiac ECM Targeting Drugs/Therapeutics

Drug	Function	Relevant CVD Outcomes	References
Agrin	ECM proteoglycan that causes acetylcholine receptor aggregation; involved in formation of neuromuscular junction	Improve cardiac function post myocardial infarction; retention of wall thickness; protection from dilated cardiomyopathy	[128]
Spironolactone	Decreases cardiac fibrosis by limiting ECM turnover	Improve outcomes of congestive heart failure patients; lowers levels of cardiac fibrosis synthesis markers	[129]

Agrin and spironolactone are two cardiac ECM-targeting drugs that strongly suggest that therapeutically modulating the ECM provides beneficial relevant CVD outcomes. Specifically, agrin is an ECM proteoglycan that causes the aggregation of acetylcholine receptors and is involved in the formation of the neuromuscular junction [128]. It has been shown to improve cardiac function after incidents of myocardial infarction and protection from DCM [128] Spironolactone, similarly, has shown beneficial effects of targeting the ECM in CVDs. Spironolactone limits ECM turnover and, therefore, decreases cardiac fibrosis [129]. It has improved outcomes of congestive HF patients along with lowering levels of cardiac fibrosis synthesis markers [129]. Studying these handful of current drugs and therapeutics that target the cytoskeleton/ECM and have also shown positive outcomes in CVDs strengthens the idea that the cellular architecture can provide a source of novel targets for drug development. However, there is the possible risk of a lack of specificity of targeting ECM proteins that may lead to adverse outcomes due to the fundamental roles that the extracellular matrix and the cytoskeleton play within our cells, but the previously discussed examples of drugs that target the ECM/cytoskeleton and have shown positive outcomes show us that this is still a field worth delving into especially when backed by computational analyses.

7. Conclusions

Further research into the cardiac cytoskeleton (and the underlying ECM that supports its architecture) is an area ripe for novel target discovery. There is mounting evidence that the cytoskeleton is an under-investigated and undervalued therapeutic target for CVD phenotypes and new computational advances provide an opportunity to discover the biological mechanisms driving changes in the cytoskeleton during CVD pathophysiology. Furthermore, the structural connection between the cytoskeleton and the ECM provides an opportunity to evaluate both as potential therapeutic targets for CVDs such as their role in myocardial disease through changes in the stiffness of the myocardium which is dependent on cellular components such as the cytoskeleton and ECM [135]. There has also been evidence supporting the direct use of the ECM as a drug, ex. the delivery of ECM biomaterials, for encouraging cardiac regeneration which further goes to show the importance of the ECM in cardiac repair [136]. Incorporating data-driven computational methods to study the cytoskeleton and the ECM in the context of cardiovascular diseases will provide a more holistic perspective on the progression of CVDs such as cardiac fibrosis, hypertension, heart failure, and atherosclerosis.

Article Highlights.

• Previous CVD studies indicate relationships between ECM proteins and the disruption of the cytoskeletal architecture.

• Observed changes in cellular architecture during cardiovascular health-to-disease transitions indicate the cytoskeleton/ECM is likely dysregulated.

• A cardioinformatics analysis of cytoskeletal/ECM-related genes may reveal new therapeutic modalities.

• Mechanisms behind existing therapeutics provide support for targeting cytoskeletal/ECM biology at different CVD development stages.

• Refining computational analytical techniques for drug discovery over the next five, ten or so years is essential for increasing the probability of success and help to decrease risk in drug development.

Abbreviations:

CAD

coronary artery disease

CAVD

calcific aortic valve disease

CVD

cardiovascular disease

CRT

cardiac resynchronization therapy

ECM

extracellular matrix

DHF

dyssynchronous heart failure

CRT

cardiac resynchronization therapy

MMP

matrix metalloproteinase

TSP

thrombospondin

OPN

osteopontin

TGF

transforming growth factor

ACS

acute coronary syndrome

DGC

dystrophin glycoprotein complex

EGFR

epidermal growth factor receptor

eQTL

expression quantitative trait loci

SNP

single nucleotide polymorphism

GWAS

genome-wide association study

LVOT

left ventricular outflow tract

LTBP

Latent TGF-β binding protein

pVO₂

mixed venous oxygen tension

VSMC

vascular smooth muscle cell

VCL

vinculin isoform 2

ILK

integrin linked kinase

CHAP

cytoskeletal heart-enriched actin-associated protein

OM

omecamtiv mecarbil

PAH

pulmonary arterial hypertension

mRNA

messenger ribonucleic acid

mRNA-seq

messenger ribonucleic acid sequencing

ApoE

apolipoprotein E

MI

myocardial infarction

HDAC

histone deacetylase

PI3K

phosphoinositide 3-kinase

PIP3

phosphatidylinositol (3,4,5)-trisphosphate

BMP

bone morphogenetic protein

HDACi

HDAC inhibitors

DNMT1

DNA methyltransferases 1

αSMA

α-smooth muscle actin

HHT

hereditary hemorrhagic telangiectasia

HDACi

HDAC inhibito