Advertisement

Tissue Engineering to Study and Treat Cardiovascular Calcification

  • Mark C. Blaser
  • Samantha K. Atkins
  • Elena AikawaEmail author
Living reference work entry
  • 14 Downloads
Part of the Reference Series in Biomedical Engineering book series (RSBE)

Abstract

Mineralization of the cardiovascular system is a complex condition whose presence strongly predicts cardiovascular mortality and results in a high clinical burden. Calcification is a common end-stage phenotype of many disparate vascular diseases, and attempts to therapeutically reverse calcification have been largely unsuccessful to date. These failures are largely due to our poor understanding of the underlying mechanisms that drive the initiation and progression of vascular and valvular mineralization. As such, this phenomenon represents a unifying nexus of difficult-to-treat disease, which is well suited to the novel application of tissue engineering. This chapter aims to highlight the importance of developing robust tissue-engineered approaches with which to treat cardiovascular disease, and summarizes recent advances in the manufacture, validation, and application of tissue engineering to this field. First, we present a comprehensive summary of relevant vascular diseases (including atherosclerosis, calcific aortic valve disease, and peripheral artery disease), and the pathobiology of cardiovascular calcification. We then discuss the current disease models available for these conditions, with a focus on the inadequacies and shortcomings of existing approaches. In particular, failures to recapitulate the biomechanical, biochemical, and cellular microenvironments of the tissue of interest are described. We extensively examine both current and future applications of tissue-engineered vascular constructs as they relate to (i) modeling pathobiology, (ii) drug discovery/testing, and (iii) therapeutic implantation to repair or replace compromised cardiovascular tissues. Lastly, we also describe recent work focused on how calcification initiates and progresses in tissue-engineered blood vessels and heart valves, and potential strategies for preventing these phenomena.

1 Introduction

Cardiovascular disease is the leading cause of death for both men and women worldwide (Benjamin et al. 2018). Calcification is the hallmark of cardiovascular disease, serving as an independent risk factor for cardiovascular morbidity and mortality. Calcium deposition, primarily in the form of hydroxyapatite, in the vessel wall or the aortic valve leaflets contributes to cardiovascular calcification (New and Aikawa 2011a). Historically, cardiovascular calcification was thought to be a degenerative disease caused passively by “wear and tear” on the valve leaflets and arteries over time; however, recent findings have compelled the National Heart, Lung and Blood Institute (NHLBI) Working Group on Calcific Aortic Stenosis (later evolved into the Alliance of Investigators on Calcific Aortic Valve Disease) and recognized that cardiovascular calcification is an active process that involves inflammation, remodeling, and dysregulated mineral metabolism (Rajamannan et al. 2011; Yutzey et al. 2014). Calcification of the arteries can lead to decreased elasticity, resulting in reduced arterial compliance that overtime can lead to vascular occlusion and myocardial infarction. Calcification of the heart valves leads to valvular dysfunction through decreased leaflet coaptation, hemodynamic regurgitation, ventricular remodeling, and, ultimately, heart failure. While cardiovascular calcification has vast clinical significance, the underlying mechanisms that drive the disease are still not fully understood and no pharmaceutical therapies exist to prevent the initiation and propagation of the disease. Therefore, there is a significant clinical burden with challenging unmet needs in terms of elucidating the pathobiology and preventing its progression.

Traditional methodologies for investigating the safety and efficacy of pharmaceuticals and medical devices depend largely on models that can represent normal and diseased tissue. These models can be performed in vitro and/or in vivo. In vitro models are beneficial for scalability and reduction of animal use. Historically, monolayer cell culture has been a convenient, accessible, and effective means for cell-based screening in drug development. This leads to expedited protocol writing and experimental approval, and eliminates the need for people specializing in animal care and handling. While cell culture models can be convenient, they also suffer from several disadvantages. The major disadvantages of two-dimensional (2D) cell culture are due to the loss of mechanical and biochemical cues, cell-cell/cell-matrix interactions, and tissue-specific architecture (Fang and Eglen 2017). In vivo animal models, a longstanding cornerstone of biomedical research, may underrepresent human disease due to differences in physiological, genetic, and phenotypic variances. Certain diseases, especially when it comes to conditions that may be dependent on the immune system, epigenetic regulation, or even the microbiome, will be difficult to recapitulate in an animal model (Healy 2018). In other instances, such as flow-induced remodeling, the microenvironment or disease-specific niche may be impossible to recreate in animal models, necessitating the development of 3D cell culture coupled with bioreactors that recapitulate the mechanical forces of the microenvironment.

At the center of drug development lies the disease itself, and inadequate models of disease hamper the expedition of suitable targets and drugs on the market. Tissue-engineered models of cardiovascular calcification can aid in the development of appropriate pharmaceutical targets and compliment already existing cell culture and animal models of disease. There is a strong motivation for the development of tissue-engineered models of calcification that can recapitulate the microenvironment of the vessel and/or valves in order to understand the primary contributing mechanisms to calcification, how the disease propagates, and how genetic or pharmaceutical manipulation in a 3D environment can help elucidate the mechanisms and find an appropriate solution to combat ectopic calcification. This chapter will serve as guiding principles that can be broadly applied to multiple cardiovascular cell types, multiple vascular beds (cardiac/great vessels/peripheral), and multiple tissue types (valve/vasculature) involved in cardiovascular calcification.

2 Vascular and Valvular Diseases

Many mechanisms can drive pathogenesis within the cardiovascular system which includes the heart, its valves, and vasculature: arteries, veins, and capillaries. Vascular calcification is defined as the abnormal deposition of calcium phosphate salts in the myocardium, heart valves, and blood vessels (Speer et al. 2009). Cardiovascular calcification is a primary predictor for and contributor to acute cardiovascular events (Mozaffarian et al. 2016). Atherosclerosis, valvular disease, aneurysm, and bypass graft failure all have common underlying risk factors such as aging, hyperlipidemia, hypercholesterolemia, obesity, diabetes, male gender, smoking status, and hemodynamic abnormality. In this section, we will highlight the most common vascular and valvular diseases, their clinical incidence, basic risk factors and drivers, and the mechanisms in which cardiovascular calcification can occur.

2.1 Atherosclerosis

Atherosclerosis, the leading cause of heart disease worldwide, is characterized by plaque buildup in the intimal space that leads to narrowing of the artery, causing a reduction in the amount of blood that can deliver oxygen and nutrients to the tissue bed. A combination of disturbed blood flow, endothelial dysfunction, and subendothelial lipid retention promotes a chronic inflammatory condition in the arteries known as atherosclerosis (Tabas et al. 2015). The major clinical indications include ischemic heart disease, ischemic stroke, and peripheral artery disease (PAD) (Herrington et al. 2016). The leading cause of myocardial infarction (MI) is ischemic heart disease resulting from atherosclerosis (Frostegård 2013). The main phenotypic characteristic of atherosclerosis is the presence of fatty streaks in the arterial wall that can gradually develop into plaque. The three main components of plaque are fat, cholesterol, and calcium. Because atherosclerotic plaques are full of lipid-laden foam cells, for most of the twentieth century atherosclerosis was assumed to be a cholesterol storage disease (Libby 2002). While arterial calcification is a hallmark of atherosclerosis (Benjamin et al. 2018), ectopic calcification was thought to happen passively, as calcium phosphate microscopically appeared to accumulate in regions of tissue necrosis. Research from the past two decades demonstrates that similar to endochondral bone formation, arterial calcification is an active, cell-mediated process. In fact, atherosclerosis is initiated through an inflammatory process involving monocytes, macrophages, and lymphocytes (Fung et al. 2007), and recent high-resolution techniques showed formation of microcalcifications through macrophage- and smooth muscle cell-derived calcifying extracellular vesicles (New Circ Res 2013; Hutcheson Nat Mat 2016).

The first evidence of inflammation in atherogenesis was noted in the 1850s by Rudolf Virchow, who recognized infiltrating leukocytes in areas of atherosclerotic plaque (Virchow 1858). Historically, the concept of atheroma as a passive lipid collection and aging process dominated over the previous century, and Virchow’s concept of atherogenesis was ignored for over 100 years (Libby 2002). It wasn’t until the late 1980s that the concept of inflammation as a driver for atherogenesis was revisited, after the discovery of adhesion molecules expressed by the vascular endothelium (Bevilacqua et al. 1987; Cybulsky et al. 1991; Li et al. 1993; Libby 2002). Distressed endothelial cells (ECs) expressing adhesion molecules provide a mechanism for circulating mononuclear cells to adhere, and interferon-γ (INF-γ) induced expression of other cytokines signals their diapedesis into the intima (Gu et al. 1998; Mach et al. 1999). The expression of the cytokine macrophage colony-stimulating factor (M-CSF) in the intima cues the maturation of monocytes into macrophages, which actively phagocytose lipoproteins and form foam cells (Clinton et al. 1992; Libby and Clinton 1992; Rosenfeld et al. 1992). These monocyte-derived macrophages partake in a maladaptive, nonresolving inflammatory response, actively contributing to the buildup of cells, matrix, and lipid in the subendothelial layer (Moore and Tabas 2011). Macrophages also actively contribute to plaque instability and thrombogenicity through the secretion of proteases that degrade collagen and extracellular vesicles that contain minerals like calcium and phosphate (Galis et al. 1994; New et al. 2013).

While atherosclerosis is an ongoing lipid accumulation and inflammatory process, a partial “resolution” of the lesion is characterized by the formation of the fibrous cap, which can be thought of as a type of overlying “scar” on the plaque (Libby 2008). The purpose of the fibrous cap is to lower its thrombogenic potential so that circulating platelets in the blood stream do not form blood clots when in contact with the prothrombotic plaque material (Tabas et al. 2015). Fibrous cap instability leads to “vulnerable plaques” which are prone to rupture and erosion. Plaque rupture and erosion account for nearly 75% and 25%, respectively, of all fatal coronary thromboses (Libby 2008). It is hypothesized that microcalcifications formed by trapped extracellular vesicles in the ECM contribute to plaque instability (New et al. 2013; Hutcheson et al. 2016). These microcalcifications promote high stress accumulation within the fibrous cap and lead to mechanical mismatch between the vessel wall, the atheromata, and the cap that may favor cavitation events, plaque instability, and, ultimately, plaque rupture (Hutcheson et al. 2016). The mechanism of calcification in atherogenesis has been associated with an endochondral bone formation process and involves inflammatory cells as well as the smooth muscle cells (SMCs) of the media (Rattazzi et al. 2005; Speer et al. 2009; New et al. 2013; Rogers et al. 2018). Harkening back to the studies by Rudolf Virchow, calcification in the coronary arteries was described similarly to bone formation as “an ossification, and not a mere calcification” (Virchow 1971; Doherty and Detrano 1994). Analogously to Virchow’s view on the role of inflammation in atherogenesis, the role of bone formation was considered a downstream consequence of degeneration in atherosclerosis and was disregarded over the next century. Improved imaging techniques in the late 1980s and early 1990s using scanning electron microscopy, energy-dispersive X-ray microanalysis, digital subtraction cinefluorescopy, and ultrafast computerized tomography and near-infrared molecular imaging (Aikawa et al. 2007a) reenergized the notion of calcification as an active process, likely the result of more than one mechanism (Detrano et al. 1985, 1986, 1993; Breen et al. 1992). Likely, two main mechanisms contribute to vascular calcification. The first mechanism involves calcifying extracellular vesicles (EVs) that are of both SMC and macrophage origin, which nucleate hydroxyapatite crystal similar to the mechanisms identified in chondrocytes during endochondral bone formation (Anderson 1989). The second mechanism involves osteogenesis of SMCs that switch from a contractile phenotype to a synthetic phenotype (Tyson et al. 2003; Speer et al. 2009) (Fig. 1).
Fig. 1

Calcification of the cardiovascular system (left: blood vessels; right: heart valves) is an active process that is initiated by inflammatory insults and which progresses to a mineralized end stage. (1) Pathogenic insults (altered shear stress, oxidized cholesterol, etc.) activate endothelial cells (ECs) into a state of endothelial dysfunction (2), which then recruit pro-inflammatory macrophages. (3) As macrophages accumulate, they release proteolytic enzymes such as cathepsins and matrix metalloproteases, which can induce osteogenic differentiation of valvular interstitial cells (VICs, 4) and smooth muscle cells (5). Osteogenic activity in these cell types results in generation and secretion of calcifying extracellular vesicles (7), which nucleate mineral and drive formation of microcalcification. Meanwhile, apoptosis of these same cell types drives release of much larger apoptotic bodies (8). Over time, microcalcifications grow to become macrocalcifications that are readily detected by conventional imaging techniques. (Adapted from New and Aikawa 2011b)

Inflammation-driven calcification in atherosclerotic plaques initiates as nanoparticle-sized extracellular vesicles (EVs) aggregate and form small spherical or ellipsoidal microcalcifications (<1 μm in diameter) rich in calcium and phosphate. Merging of microcalcifications leads to larger macrocalcifications that are hypothesized to stabilize the plaque (Hutcheson et al. 2016, 2017). Two different cell types, SMCs and macrophages, contribute to inflammation-driven, EV-mediated calcification. EV-mediated calcification can be subdivided into four stages. In the first stage, EVs secreted by SMCs and macrophages accumulate in the ECM (step 1: accumulation). Next, the EVs aggregate (step 2: aggregation) in the extracellular space and become trapped in the collagen network. The trapped EVs fuse together (step 3: fusion) or tether to one another through protein interactions. Finally, microcalcifications form and the mineralization matures (step 4: mineralization) (Hutcheson et al. 2016).

During atherogenesis, synthetic SMCs proliferate, migrate, and produce ECM and remodeling proteins in an attempt to stabilize the plaque (Alexander and Owens 2012; Tabas et al. 2015). In regions of human and mouse calcified atheromata, SMCs expressed the receptor sortilin, which promotes calcification by trafficking tissue nonspecific alkaline phosphatase (TNAP) into EVs. TNAP converts ATP into free phosphates that can nucleate hydroxyapatite and form mineralization crystals (Goettsch et al. 2016). Atherosclerotic plaques also contain regions of calcifying SMCs that express osteogenic factors like BMP-2 (Boström et al. 1993). Additionally, in vitro analysis of mouse aortic SMCs revealed that miR-145 may play a role in conversion of SMCs from a contractile to a synthetic phenotype by acting on its target gene, Kruppel-like Factor 5 (KLF5) and Myocardin (MYOCD); however, translation of these findings to an in vivo model has not yet been undertaken (Zhang et al. 2016). Investigating the role of SMC phenotypic switching is difficult on several levels. First, synthetic SMCs within lesions of atherosclerosis may be falsely overlooked if their markers have been lost and lineage-tracing has not been performed. Second, other cell types of myeloid lineage in atherosclerotic lesions may express SMC markers (Caplice et al. 2003). Finally, experiments on cholesterol loading of SMCs in cell culture showed a downregulation of SMC-specific markers (SM α-actin and α-tropomyosin) and upregulation of macrophage markers (CD68 and Mac2) (Rong et al. 2003). While the inability to properly trace and define the lineage in vivo and non-translatable findings in vitro convolute the role of SMC phenotypic switching in the etiopathology of atherosclerosis, this area of research is still considered a large knowledge gap that requires more thorough investigation.

Macrophages also contribute to inflammation-driven, EV-mediated calcification in atherosclerosis. For the first time ever, in vivo molecular imaging studies on apoE−/− mice with early-stage atherosclerosis revealed that inflammation precedes osteogenesis. Near-infrared molecular imaging through fluorescence reflectance mapping revealed osteogenesis concomitant with macrophage presence (Aikawa et al. 2007b). Atheromata from human carotid endarterectomy and arteries of apoE−/− mice with chronic renal disease showed macrophages released CD68-positive EVs associated with regions of microcalcifications in plaques and led to a hypothesis that macrophages themselves could contribute these calcifying EVs. These EVs not only had hydroxyapatite nucleation apparent on and in their membrane, but they also had increased levels of annexins and S100A9, a calcium binding protein. These results indicate an alternative inflammation-mediated calcification pathway in atherogenesis, independent of SMC phenotypic switching (New et al. 2013).

One other form of ectopic calcification associated with the smaller arterioles is known as calciphylaxis. Calciphylaxis describes the accumulation of calcific nodules in the small blood vessels resulting in blood clots and skin ulcers that can result in serious complications such as infection and death. While the etiology of calciphylaxis is unknown, more recent studies suggest it may be tied to blood clotting disorders (Nigwekar et al. 2015).

The current gold standard of treatment for coronary atherosclerosis is coronary-artery bypass grafting (CABG). CABG utilizes autologous artery or vein tissue harvested from another area of the body, most likely the internal thoracic artery or the greater saphenous vein, to bypass coronary artery obstructions. Around 400,000 CABG procedures are performed in the United States annually, and the surgery is highly invasive requiring a cardiopulmonary-bypass machine to support circulation through the 3–4 h surgery. Recovery includes a 5–7 post-operative hospital stay followed by an additional 6–12 weeks of recuperating after hospital discharge (Alexander and Smith 2016). The major problem with CABG is restenosis, especially with saphenous vein grafts that are associated with lower patency rates (Fukui et al. 2010). While the autologous tissue approach has lower patency failure rates than synthetic grafts, limited availability, donor site morbidity, and an aging population make this approach fraught with complications (Desai et al. 2011).

Due to the more recent appreciation for the role of inflammation in atherogenesis, potential inflammatory biomarkers that may signal future cardiovascular risk have begun to be investigated. One inflammatory biomarker, C-reactive protein (CRP), was identified in individuals post myocardial infarction. The development of a high-sensitivity assay for CRP (hsCRP) has aided in additional risk stratification of patients that have normal lipid levels (low-density lipoprotein cholesterol; LDLC <130 mg/dL) and high hsCRP (>2 mg/L). The first clinical trial to test the effectiveness of statins, a class of lipid lowering drugs, and their effect on CRP levels is known as JUPITER (Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin). The JUPITER trial enrolled over 17,000 patients that had high levels of inflammatory biomarker hsCRP (>2 mg/L) but were not considered candidates for statin therapy due to LDLC levels <130 mg/dL, which was the current threshold for stain treatment. The outcome of the JUPITER trial was better than anticipated, and the trial was stopped early due to the clear, significant benefits statin therapy had on all vascular events versus placebo: 54% reduction in myocardial infarction; 46% reduction in need for revascularization; 20% reduction in all-cause mortality (Ridker 2009).

3 Calcific Aortic Valve Disease

The aortic valve is located between the left ventricle and the aorta and is responsible for maintaining unidirectional blood flow to deliver oxygenated blood from the heart throughout the systemic circulation (Butcher et al. 2011). The aortic valve functions in a complex mechanical environment comprised of stretch, pressure forces, tension, and fluid shear stress imposed on the leaflets due to hemodynamics. Additionally, each side of the valve leaflet experiences a unique pattern of these forces during the cardiac cycle: the ventricularis is subjected to high-momentum flow, high pressure, and pulsatility with fluid shear stress dominating during systole, while the fibrosa experiences low-momentum recirculatory flow during systole and high tensile and bending stresses during diastole due to the high transvalvular pressure gradient (80 mmHg) which causes the leaflets to stretch radially and circumferentially (Arjunon et al. 2013).

The aortic valve leaflets are made up of three distinct layers that each has their own distinct ECM properties: the fibrosa is rich in collagen to provide bending strength; the spongiosa is rich in glycosaminoglycans (GAGs) that act as a lubricating layer; and the ventricularis is rich in elastin to aid in leaflet motion (Scott and Vesely 1995). There are two major cell types in the aortic valve: valvular endothelial cells (VECs) and valvular interstitial cells (VICs). VECs constitute the outer lining of the leaflets and are responsible for valvular homeostasis through their active role in permeability regulation, paracrine signaling, and inflammatory cellular adhesion. VICs are the most abundant cell type and are interspersed throughout the tri-layered leaflets, secreting extracellular matrix (ECM) proteins and providing added mechanical strength (Leopold 2012). VICs present in these three layers may sense biomechanical cues from their microenvironment that contribute to their activation into myofibroblast- or osteoblast-like cells that promote valvular calcification and remodeling. The ventricularis faces the left ventricle and must undergo large bending motion during the cardiac cycle; therefore, it is comprised mainly of radially-running elastin with a sparse collagen network (Vesely 1997). The middle layer, known as the spongiosa lends flexibility to the leaflets due to a high concentration of glycosaminoglycans (GAGs) (Rajamannan et al. 2011; Eckert et al. 2013; Li et al. 2013). The fibrosa, which faces the aorta, is composed mainly of circumferentially-running collagen fibers and has stiffer mechanical properties than the ventricularis in order to provide increased load-bearing capability during diastole (Stella and Sacks 2007; Rajamannan et al. 2011; Leopold 2012).

CAVD covers a spectrum of disease from initial changes in cell biology of the valve leaflets, through early calcification, tissue remodeling, and aortic sclerosis, to outflow obstruction and aortic stenosis (Otto et al. 1994; O’Brien 2006b; Rajamannan 2009; Carabello and Paulus 2009). The three main stages of CAVD are inflammation, fibrosis, and calcification. It is believed that mechanical stress in the form of abnormal wall shear stress initiates activation and inflammation on the AV fibrosa. The result is upregulation of endothelial activation proteins such as intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1), and pro-inflammatory cytokines such as bone morphogenic proteins (BMPs) and transforming growth factor beta-1 (TGF-β1). Endothelial activation accompanies macrophage accumulation in the extracellular matrix (ECM). During the fibrotic stage, remodeling through the balance of matrix metalloproteinases (MMPs) and the tissue inhibitors of matrix metalloproteinases (TIMPs) causes disorganization of ECM fibers, and VICs are switched from their normal, resting fibroblast-like phenotype to an activated myofibroblast- or osteoblast-like, calcifying phenotype. Similar to SMCs, VICs are highly plastic cells that can become activated to myofibroblasts or osteoblast-like calcifying cells under the right conditions. Using contact microradiography, staining, and micro-computerized tomography (microCT), investigators found that 2D and 3D images of calcified regions of the valve were indistinguishable from that of the human femur and had similar elemental composition of CaPO4. Analysis of mRNA transcripts showed elevated expression of the osteoblast-specific transcription factor Cbfa1 (also called Runx2), osteocalcin, osteopontin, and bone sialoprotein (Rajamannan et al. 2003). Finally, during calcification, calcifying VICs secrete chondrogenic and osteogenic growth factors that lead to calcific deposits on the leaflet fibrosa (Rajamannan et al. 2011; Victoria Gomez-Stallons et al. 2016).

Formerly considered a passive age-related disease promoted by cardiovascular risk factors and some genetic predispositions, CAVD is now recognized as an active disease process involving inflammatory, extracellular matrix remodeling, and osteogenic mediators as well as phenotypic changes in the VIC population (Mohler et al. 1991; O’Brien 2006b; Rajamannan 2009; Carabello and Paulus 2009; Rajamannan et al. 2011). The first detectable indication of CAVD is the localized thickening of the leaflet tissue as a result of valvular calcification, known as sclerosis (Sahasakul et al. 1988; O’Brien 2006a). The later stage is characterized by formation of calcium nodules preferentially on the fibrosa (i.e., leaflet aortic surface) (Sheikh and Livesey 2010) and is referred to as aortic stenosis. The stiffening of the leaflet tissue results in a reduction of the valve geometric orifice area (stenosis), thus imposing a higher transvalvular pressure gradient, which may result in ventricular hypertrophy and, ultimately, heart failure (Verma et al. 2013). Asymptomatic stenosis is often diagnosed as a heart murmur. However, once severe stenosis symptoms such as chest pain, shortness of breath, and dizziness occur, the disease is already at an advanced stage.

There are currently no pharmaceutical treatments to delay the progression of CAVD, nor advanced diagnostics to capture early stages of CAVD before the disease has reached a point of no return. A few molecules have been hypothesized to be independent risk factors for valvular calcification, including lipoprotein(a) (Lp(a)). Lp(a) is a carrier of oxidized phospholipids and is correlated with faster progression of calcification in patients with elevated serum levels. In fact, after adjusting for baseline calcium score and other cardiovascular risk factors like diabetes and hypertension, LP(a) was an independent predictor of aortic valve calcification. Additionally, VICs stimulated with 100 mg/dL of LP(a) in cell culture upregulated gene expression of inflammatory marker IL-6 and osteoblastic markers BMP2 and RUNX2 after 1 week in culture (Zheng et al. 2019).

The interplay between the complex mechanical environment and biological signaling of the AV requires additional research, and 3D tissue-engineered models of valvular inflammation, fibrosis, and calcification may aid in the development of early diagnostic tools and interventions. Creating a long-lasting, durable heart valve replacement is challenging on a number of levels. The design criteria for a valve replacement require a material that is non-thrombogenic and chemically inert. It must be able to fully open as to not obstruct blood flow, and close rapidly in response to the transvalvular pressure gradient, while providing enough mechanical force to create a seal to prevent regurgitation. In the 1920s, surgeons attempted to repair calcified valves by surgically decalcifying them or by replacing them with polymeric leaflets; however, the results of both attempts were poor. The first surgical aortic valve replacement (SAVR) utilized a mechanical valve with a ball and cage design, which prevailed until the 1970s when the bi-leaflet mechanical valve was developed; simultaneously, the continuous need for anticoagulation therapy was recognized and a more biologically and mechanically compatible replacement was sought. First, cadaver heart valves were used, but their availability was limited. This lead to the invention of the first bioprosthetic heart valve, the Carpentier-Edwards valve that consisted of decellularized, glutaraldehyde-treated porcine valve mounted on a fabric-covered metal frame. Xenogeneic bioprosthetic heart valves are still the prevailing choice over half a century later, despite limited lifespan due to calcification, cusp rupture, and structural deterioration (Carpentier 2007).

April 16, 2002 marked the first major medical advancement in aortic valve replacement in many decades with the successful implantation of the first transcatheter aortic valve replacement (TAVR) (Cribier 2014). Unlike SAVR which requires open heart surgery, coronary arrest, a coronary-bypass machine, and weeks of painful recuperation, TAVR is performed by deploying balloon expandable or self-expanding valve over the existing aortic valve. A catheter guided through the femoral artery to the left ventricular outflow tract deploys a bioprosthetic valve over the existing, calcified aortic valve. From 2002 to 2013, over 80,000 TAVRs were performed in the United States (Cribier 2014). Recently, TAVR has been shown to be as equivalently safe and effective as SAVR in both low- and high-risk patient populations (Takagi et al. 2019; Mack et al. 2019; Popma et al. 2019). While this advancement certainly improves both the patient and clinical burdens associated with SAVR, it does not address the root cause of the need for valvular replacement, and with the reduced amount of SAVRs being performed, the development of anti-calcification therapeutics may be hindered by a sudden shortage of tissue samples needed to conduct basic CAVD research. Finally, TAVR still relies on the use of bioprosthetic valves that are fraught with their own limited lifespan and need for revision surgeries. The success of TAVR ushers in a new era in which tissue-engineered valvular constructs are even more appealing in terms of solving basic biological research questions surrounding the onset and progression of CAVD, especially in the context of potential scarcities in biological tissue specimens that aid in research and development.

4 Existing Models of Vascular Disease

4.1 Models of Atherosclerotic Calcification

As described earlier in this chapter, atherosclerotic plaques are an accumulation of inflammatory cells, SMCs, lipids, ECM, and mineralization within the subendothelial space. Modeling the complex milieu of the atheroma is challenging. Compounding this challenge is the transitional gap between established animal models of calcification and the human disease, due to major differences in inflammatory mechanisms, anatomy, metabolism, and lipid profiles (Mallone et al. 2018).

One of the first in vivo models of atherosclerosis is the murine model with matrix GLA protein (MGP) deficiency. Deficiency of MGP in mice results in extensive calcium deposition in the media of the arterial wall, suggesting that the presence of MGP would inhibit mineral deposition; however, while healthy arterial walls show some expression of MGP and confirm an inhibitory role in the development of calcification, it is not the only mechanism at play and may better represent a model of passive calcification rather than active calcification (Guangbin et al. 1997). Consistent with an active model of arterial calcification, murine models with Smad6 deletion show enhanced BMP signaling resulting in potent medial arterial calcification (Galvin et al. 2000). Other in vivo murine models of medial arterial calcification include the fibrillin-1 deficient mouse and the klotho-null mouse (Kuro-o et al. 1997; Pereira et al. 1999). Today, the most widely used in vivo model of active calcification is the apoE-deficient (apoE−/−) mouse fed a Western diet (Rattazzi et al. 2005). There are three major problems complicating these in vivo models of calcification. The first major problem is that mouse models do not always translate to humans, as in the case of the MGP deficient mouse. For example, Keutel syndrome is an inherited syndrome in which the MGP gene is nonfunctional, but people with Keutel syndrome do not display the same degree of arterial calcification detected in the MGP-deficient murine models (Munroe et al. 1999). Additionally, the role of MGP in calcification is even more convoluted when compounded with conflicting results from in vitro cell culture, which show high levels of MGP and low levels of osteopontin accompanying mineralization (Proudfoot et al. 1998). The second major issue with the majority of in vivo models of calcification is that they do not necessarily recapitulate the active process of atherogenesis that includes plaque formation in the intima. For example, MGP-, Smad6-, and klotho-deficient mice all show intense calcification in the medial layer of their arteries, which is mechanistically very different than the process of atherogenesis (Doherty et al. 2003). Finally, the third major issue is that there are distinct differences in the calcification phenotype of different strains of mice on a Western diet. For example, genetically distinct inbred mouse strains such as Balb/C, C57BL/6, C3H/HeJ, DBA/2 J, SM/J, and MRL-lpr/lpr all on the same high fat, high cholesterol diet developed varying degrees of atherosclerosis suggesting that a genetic component regulates the pattern of calcification, and therefore, selecting the appropriate murine strain is non-trivial (Qiao et al. 1994).

As discussed in the earlier section, atherosclerosis is a combination of lipid accumulation and inflammation, ultimately resulting in a multifaceted disease involving inflammatory cells, endothelial cells, and SMCs. In vitro, cell-based models of atherosclerosis can be used to understand pathobiological mechanisms driving disease progression as well as the cell-cell interactions involved. These in vitro models can be used throughout the stages of drug target development from understanding initial triggers that cause disease onset, risk assessment, and toxicity of compounds and their predicted efficacy before moving to in vivo animal models and, ultimately, clinical trials. Early studies on atherogenesis utilized static culture models with a single cell type. More robust in vitro models emerged and utilize co-culture of more than one cell type under a variety of physiological and pathophysiological conditions (hyperlipidemia, hyperphosphatemia, hypercalcemia, altered hemodynamics, etc.). Unfortunately, due to the large variation in cell types, co-culture models, static versus dynamic culture, and addition of comorbidities like diabetes, obesity, and HIV, there has not been a standardized “one-size-fits-all” in vitro model of atherosclerosis. The review by Islam et al. elegantly summarizes the cell types, culture systems, focus, and key findings of over 50 in vitro atherosclerosis studies spanning from the 1980s to 2016. The first co-culture studies of atherogenesis that paved the way for future mechanistic studies included models of either endothelial cells with SMCs or endothelial cells with monocytes/macrophages (Davies et al. 1985; Navab et al. 1988). Typically, two main categories of co-culture can be used to model atherosclerosis: (1) indirect models in which cells are not in contact with one another, but coexist within the same container or experience a shared environment/secretome through conditioned media treatment; and (2) direct models in which the cells are in contact. Both static and dynamic culture have been used in these two model systems (Islam et al. 2016). Almost always, primary cells derived from human, mouse, bovine, or porcine tissues are used.

Until recently, no model of late-stage atherosclerotic plaque existed, but advances in co-culture and tissue engineering have paved the way for plaque biofabrication to better understand the pathobiology of atherogenesis. In a 2018 study, “pseudo-plaque” (ps-plaque) was developed by embedding a spheroid core of monocytes, macrophages, and dendritic cells in an EMC of lipid and collagen surrounded by a myofibroblast layer mimicking the fibrous cap. Using flow cytometry, the investigators were able to demonstrate similarities between the cellular subpopulations gene expression profiles in ps-plaque and human tissue samples from carotid endarterectomy plaque (Mallone et al. 2018). Despite this achievement, there are still several shortcomings in this model that could be improved. The ps-plaque lacked SMCs which are major communicators with immune cells. In vivo, mechanical stress is present with the addition of circulating immune cells that may replenish the signal or contribute to atherogenesis over time. Other disease factors linked to atherosclerosis like chronic kidney disease, diabetes, and hyperlipidemia were not considered in this model. While the ps-plaque model lacked the presence of calcification, it may serve as an interesting investigative tool to assess the change in gene expression and effect on calcification in a pro-calcific environment. Additionally, multicellular tissue-engineered models of atherosclerotic plaque may serve as the perfect niche microenvironment to validate pharmaceutical targets during high-throughput screening.

4.2 Models of Cardiac Valve Calcification

Calcification is the most common reason for bioprosthetic heart valve failure, which makes it a favorable model to study the progression of the calcification process in a dynamic setting. In the 1980s and 1990s, contemporaneous with the examination of molecular mechanisms involved in CAVD, investigators began examining bioprosthetic aortic valve calcification. As discussed earlier in this section, CAVD begins as an inflammatory process and the end-stage result is the formation of calcium phosphate nodules on the fibrosa side of the leaflet that disrupt the leaflet architecture. In preliminary studies, the development of mineralization on bioprosthetic valves was monitored in vivo and in vitro. In vivo, procalcific animal models included bioprosthetic valve replacement in cows and sheep, as well as subcutaneous leaflet implantation in rabbits or rats that showed accelerated calcification beginning in as little as 48 h and developing into large nodules in as little as 8 weeks (Levy et al. 1983; Schoen et al. 1985). In these acellular models, mineralization of hydroxyapatite crystals was mediated by “matrix-vesicle-like bodies” most likely released by non-viable cells in contact with the leaflets, supporting the idea that glutaraldehyde treatment of the valves was a driver of pathogenesis and failure of bioprosthetic valves (Gong et al. 1991). In vitro, hydroxyapatite crystallization was monitored in a bioreactor system that contained supersaturated calcium phosphate media. In this model system, investigators were able to titrate the amount of calcium and phosphate ions to compensate for the ion precipitation as hydroxyapatite was formed and in turn, were able to measure the rate of crystal formation on glutaraldehyde-treated heart valve leaflets. This system proved effective in evaluating the calcification potential of xenogeneic porcine aortic valves that would be used for surgical aortic valve replacement (SAVR) (Kapolos et al. 1997). Around the same time, glutaraldehyde-fixed bovine pericardial valves (the Glasgow Heart Valve) were being investigated for their calcification potential in a bioreactor designed on a modified Rowan Ash fatigue tester maintained under physiological conditions. The investigators found that although similar physiological conditions were maintained, there was a wide variability in calcification rates of the bioprosthetic valves, and hypothesized this may be a contributing factor to the variable success rates seen in patients (Bernacca and Mackay 1992). The early models of valvular calcification relied on decellularized, glutaraldehyde-treated valves and represented a potential passive mechanism for calcification; however, recent studies have suggested that cells actively contribute to the calcification process through EV secretion, phenotype switching, cytokine signaling, and enzymatic activity. More robust in vitro cell culture models are also needed to understand the basic biology of valvular calcification.

To understand the genetic and molecular mechanisms involved in CAVD, animal models serve as a valuable tool. The various animal models used to study valvular calcification were recently reviewed by Sider and Hjortnaes (Hjortnaes and Aikaw 2011; Sider et al. 2011a). Briefly, the most commonly used animal models to study CAVD are mouse, rabbit, and swine – of which swine is the only model to naturally develop calcification with age, albeit slowly. Rabbits require a high cholesterol diet to develop CAVD and mice require a genetic predisposition. Mice models are attractive to study the development of calcification and its mechanisms as they have a short generation time and are easy to handle and genetically modify; however, valve anatomy in mice differs from humans, and they lack the tri-layered leaflet morphology seen in humans, rabbits, and swine. The most commonly used mouse model to study valvular calcification is the C57BL/6 J strain with low-density lipoprotein (LDL) receptor deficiency fed the Harlan Teklad TD.88137 diet, or “Western” diet. The other most common mouse model is the ApoE-deficient (ApoE−/−) mouse, which can develop hypercholesterolemia without a high fat, high cholesterol diet (Aikawa et al. 2007b). Almost all animal models used to study valvular calcification rely on a hypercholesterolemic diet, and whether or not this model is a true representation of the progression of valvular calcification in humans remains to be determined. A more recent model to study aortic stenosis in the context of response-to-tissue-injury is the valvular wire injury model (Honda et al. 2014). As research progresses, it is important to judiciously reexamine animal models in the context of human disease (Sider et al. 2011b).

In vitro models have been developed to help understand the process of calcification in human heart valves and bioprosthetic valves. Generally, in vitro models are cheaper, simpler, and more controllable than in vivo models. The first in vitro studies of calcification were done in the context of endochondral bone formation. In the late 1970s and early 1980s, electron microscopy and isolation studies identified the mitochondria as the organelle source of calcium ions from matrix vesicles of the epiphyseal growth plate (Brighton and Hunt 1978; Carafoli and Crompton 1978; Cecil and Anderson 1978; Carafoli 1981). In the heart valves, calcification forms around valvular interstitial cells (VICs) in the interstitial space (Aikawa and Libby 2017). While the valvular endothelial cells (VECs) certainly contribute to the pathogenesis of CAVD and VEC-VIC interplay regulates valve osteogenesis (Hjortnaes et al. 2015), VICs are the most common choice of cell culture models of valvular calcification. Static cell culture of valvular VICs stands as a first line of investigation to determine the basic cellular biology involved in calcification, and are particularly valuable for drug development and screening purposes. Induction of calcification in VICs in vitro is achieved through media treatment that contains high levels of inorganic phosphate (Pi; ≥2 mM) and/or proinflammatory mediators like lipopolysaccharides (Mathieu et al. 2005; Babu et al. 2008; Rattazzi et al. 2008; Ortolani et al. 2010). These experiments strengthened the notion that VIC exposure to high levels of phosphate is a prerequisite for mineralization which is driven by alkaline phosphatase (ALP) activity. There are two types of calcifying medias for VIC culture, one that contains organic phosphate (“osteogenic media,” OM) and one that contains inorganic phosphate (“procalcifying media,” PM). OM contains β-glycerophosphate (organic phosphate), ascorbic acid, and dexamethasone. ALP is expressed by VICs in OM and hydrolyzes β-glycerophosphate to inorganic phosphate that forms hydroxyapatite crystals (Goto et al. 2019). PM contains elevated levels monosodium phosphate (inorganic phosphate) and ascorbic acid without the addition of dexamethasone (Schlotter et al. 2018). VIC calcification is not passage-dependent in PM; on the other hand, VICs in OM lose their ability to calcify after a number of passages (Goto et al. 2019). Because calcification in OM is TNAP dependent, it is hypothesized that cells lose TNAP over time in culture and thus lose their ability to calcify in OM. Inorganic phosphate in PM is readily available to form hydroxyapatite crystals and does not require the activity of TNAP. When choosing a media type to induce calcification in VICs, it may be important to consider which type of pathobiological condition of calcification is being modeled. OM might be more suitable for studying early passage VICs in the context of CAVD while PM might be suitable for studying the role of high phosphate generating conditions like chronic kidney disease (CKD) on valvular calcification. More work needs to be done to standardize the field of CAVD research; however, in vitro VIC culture still stands as a primary means to evaluate mechanisms driving pro- or anti-calcific effects of calcification pathways and their inhibitors.

Recent advances in cell culture techniques coupled with the growing field of tissue engineering have fostered the development of more advanced models to help expand the current understanding of the mechanisms that drive valvular calcification. To fully elucidate these mechanisms, the multicellular, dynamic biomechanical environment specific to each valve type must be recapitulated in vitro. Creating such a model is a complex and challenging undertaking that requires a multidisciplinary approach from a chemical, biological, and mechanical standpoint.

5 Applications of Tissue-Engineered Vascular Constructs

5.1 Modeling Pathobiology

The application of engineered tissues to study and treat cardiovascular disease has long been promised, but success stories remain relatively limited. One key area that is immediately amenable to the usage of engineered tissues is the development of better models of healthy and diseased tissues. As we note above, there are substantial issues with existing models of cardiovascular disease. 2D in vitro cell culture frequently lacks the cellular complexity (multiple cell types), microenvironmental cues (circulating/soluble and matrix-bound factors), and biomechanical forces (shear forces from the blood, cyclic stretch, substrate/tissue stiffness, etc.) that are present in the human cardiovascular system and which have been implicated in both tissue homeostasis and disease induction/progression (Lu and Kassab 2011; Arjunon et al. 2013). Moreover, there is a substantial body of work supporting the notion that many cell types are highly responsive to the dimensionality of their environs (Riedl et al. 2017). For example, the VIC transcriptome is significantly altered by culture in 2D or 3D conditions, where genes associated with cellular structure, polarity, and motility are highly differentially expressed (Mabry et al. 2016). While animal models of disease remain some of our best tools for understanding disease and developing translational therapies, the history of their usage is littered with translational failures. Animal studies are costly, time-consuming, all-too-frequently do not recapitulate human pathobiology, and come with inherent ethical concerns (reviewed in (Sider et al. 2011a). For all these reasons, the barrier of entry for tissue-engineered solutions is drastically reduced.

5.2 Cell Types and Sources in Tissue-Engineered Models

Over the years, scientists have called on a wide variety of cell types when constructing 3D models of vascular and valvular disease. Primary human vascular smooth muscle cells (SMCs) and VICs are the most popular options, as they come closest to the native cellular phenotypes of their respective tissues (reviewed in Jover et al. 2018). Unlike SMCs, VICs used in in vitro studies are almost exclusively isolated from diseased tissues, since normal valvular tissue is rarely, if ever, removed during surgical interventions. In experimental contexts where shear stress responses, hemolysis, or thrombosis are of interest, cardiovascular constructs are typically endothelialized by seeding endothelial cells onto the fluid-facing construct surface(s) (Ardila et al. 2019). These cells can be obtained from relatively standardized and widely available cell sources such as primary human umbilical vein endothelial cells (HUVECs) or human aortic endothelial cells. When aortic valve constructs undergo endothelialization, vascular ECs are often utilized. However, there is substantial evidence that though vascular and valvular ECs share the large majority of their identity, there are key differences: unlike vascular ECs, VECs often align perpendicular to flow (Butcher et al. 2004), are much more proliferative, and there are significant alterations in gene expression profiles (Farivar et al. 2003). As a result, many in the valve field have taken to deriving and sub-culturing primary human VECs directly from donated human tissue (Tseng et al. 2014; Hjortnaes et al. 2015); at this time VECs are largely unavailable from commercial cell sources. When extensive pre-conditioning will be employed (e.g., mechanical stimulation in a bioreactor) or when cellular differentiation after in vivo implantation is the end goal or phenomena under study, autologous or homologous cell sources such as bone marrow cells and mesenchymal stem cells are popular sources (Schmidt et al. 2010). Again, these cell types are relatively easy to obtain and culture (when compared to sources of primary human vascular SMCs or VICs), and the opportunity to exert control over their differentiation enables added flexibility when synthesizing vascular organoids. In disease, endothelium and parenchymal SMCs/VICs are no longer the sole constituents of cardiovascular tissues. When modeling disease states, care must therefore be taken to recapitulate the cellular populations that differentiate or infiltrate during pathogenesis in vivo. Macrophages (from both immortalized cell lines or differentiated from primary human monocytes) are commonly employed to serve as representatives of inflammatory cell infiltrates in engineered models of atherosclerosis (Mallone et al. 2018). When modeling late-stage cardiovascular calcification, the usage of bone-forming osteoblasts, cartilage-producing chondrocytes, and bone-resorbing osteoclasts is popular. One cell type which stands to revolutionize how tissue-engineered models of cardiovascular disease are designed and produced is the induced pluripotent stem cell (iPSC) (Musunuru et al. 2018). These cells, typically derived from primary human skin fibroblasts, peripheral blood cells, or other easily biopsiable tissues, are reprogrammed to a pluripotent stem cell state by heterologous expression of specific transcription factors (Takahashi and Yamanaka 2006). Due to their stemness, these iPSCs can be easily expanded and then differentiated down multiple lineages (smooth muscle, fibroblastic, bone-forming, bone-resorbing, inflammatory, etc.) (Musunuru et al. 2018). iPSCs can be genetically matched to individual patients: EC-iPSCs from patients with NOTCH1 mutations (linked with common congenital valvular disorders) have been utilized to study the impact of Notch deficiency on shear-regulated aortic valve calcification (Theodoris et al. 2015). They also offer a means to reduce donor-to-donor variability in models of disease that does not necessitate the use of immortalized cell lines. For example, differentiation of multiple tissue types from a single donor would enable modeling of vascular versus valvular calcification without the confounding factor of primary SMCs and primary VICs being derived from different donors.

5.3 Selection of Scaffold Materials for Engineered Valvular Tissues

The construction of engineered cardiovascular tissue models typically begins with selection of the appropriate extracellular matrix composition. In some cases, 3D collagen hydrogels that mimic aspects of human plaque ECM were used to demonstrate aggregation of calcifying extracellular vesicles into microcalcifications. By casting self-assembled layered hydrogels with varying collagen density and then seeding these acellular gels with vesicles collected from cultured vascular SMCs, the authors modeled the physicochemical interactions that underlie mineral formation in atheromata and identified links between calcification morphology, collagen content, and plaque stability (Hutcheson et al. 2016). With respect to the aortic valve, the mechanosensitive nature of VICs and VECs, as well as their complex biomechanical microenvironment, has driven efforts to engineer “valve-on-a-chip” organoids as models of CAVD. Most simplistically, VICs encapsulated in a variety of 3D hydrogel systems have been utilized. Poly(ethylene-glycol (PEG) hydrogels are popular, as they preserve VIC viability, can be engineered to be degradable by matrix metalloproteinases (MMPs), and their elastic moduli can be easily varied within a physiological range between ~0.24 kPa and 12 kPa (Mabry et al. 2015). Others have used electrospun PEG-dimethacrylate-poly(L-lactide) (PEGdma-PLA) to mimic the tri-layer structural and mechanical properties of the native aortic valve, with VIC/VEC co-culture and incorporation of key homeostatic collagen and proteoglycan isoforms (Hinderer et al. 2014). Elsewhere, mechanically tunable methacrylated gelatin and methacrylated hyaluronic acid (GelMA/HAMA) scaffolds have been frequently employed: casting of VICs into these gels supports both maintenance of fibroblastic VIC quiescence (Duan et al. 2016) or differentiation towards pathological synthetic, myofibrogenic, and mineralizing phenotypes (Hjortnaes et al. 2014, 2016). Such models have been utilized to implicate key components of RhoA/ROCK signaling in CAVD pathogenesis and elucidate how a quiescent-myofibrogenic-osteogenic transition underpins disease development. GelMA/HAMA is also well suited to 3D bioprinting, and was utilized to fabricate the first multi-layered tissue-engineered aortic valve constructs that mimic elastic moduli of the native leaflet layers (Valk et al. 2018). Frequently, hydrogel-based models are designed with components that are degradable by matrix metalloproteinases. These gels ensure that homeostatic ECM turnover, and the morphology and proliferation of encapsulated cells are not inhibited by the artificial matrix (Gonzalez Rodriguez et al. 2018). Constructs are frequently impregnated with or exposed to growth factors or other biological molecules in order to examine their impact on valvular health, as in the case where FGF-2 or TGF-B1 exposure differentially modulated myofibrogenesis and matrix remodeling by encapsulated valvular cells. In addition to GelMA/HAMA hydrogels, the incorporation of methacrylated glycosaominoglycans (GAGs) has been employed in order to model the proteoglycan-rich lesions known to compose the earliest known features of CAVD (Porras et al. 2018). GelMA/GAGMA gels strongly retained oxidized low-density lipoproteins which in turn drove myofibrogenesis of encapsulated VICs, secretion of inflammatory cytokines, and VIC production of GAGs which acted as the earliest constituents of endochondral ossification in this disease-inspired model (Porras et al. 2018).

Others have taken steps to engineer polymeric artificial valves that mimic the calcific burden present in human CAVD. A calcium phosphate/polyurethane mixture was developed and cured in anatomically-accurate silicone molds, resulting in a tri-leaflet valve with randomly-distributed calcified nodules (Falahatpisheh et al. 2017). Under echocardiographic testing in a heart flow simulator, both moderate and severe stenosis could be modeled by varying the degree of these hydroxyapatite inclusions. One key usage of such valves is as a phantom-like model for testing the performance of transcatheter valvular delivery systems and procedures in vitro. Despite their rise in popularity and recent demonstrations of safety, many questions remain regarding the long-term durability, paravalvular regurgitation, (re-)positioning accuracy, and the long-term impact of leaving heavily-calcified leaflets in the aortic root after TAVR (Mack et al. 2019; Popma et al. 2019); calcified models such as this are well positioned to answer some of these outstanding questions under highly-controlled testing conditions. The incorporation of mineral into in vitro models has shed light on disease pathogenesis in other contexts as well: when hydroxyapatite nanoparticles are embedded within 3D collagen hydrogels, the degree of hydroxyapatite crystallinity modulated VIC differentiation (Richards et al. 2018). Low crystallinity induced myofibrogenesis and osteogenesis of VICs, and abrogated the anti-calcific effects of VEC co-culture. In short, this approach enabled modeling of the VIC response to a mineral-rich extracellular environment for the first time, thereby enabling investigation of VIC responses in the later stages of CAVD.

5.4 Scaffold Composition of Engineered Vascular Grafts

When engineering vascular tissue constructs, many of the same principles and components are employed: collagen and fibrin hydrogels, though collagen-based vessels suffer from a relative lack of collagen cross-linking (and therefore tensile strength) as well as lower cell density (Weinberg and Bell 1986). This is likely due to the disorganized nature of the denatured collagen in use, as well as the well-documented phenomenon that fibroblasts and SMCs downregulate collagen secretion when cultured in collagen gels (Clark et al. 1995; Thie et al. 1991). Crosslinking reagents such as glutaraldehyde, lysyl oxidase, or transglutaminase improve this shortcoming, but can induce cytotoxicity (Charulatha and Rajaram 2003; Elbjeirami et al. 2003). In contrast, fibrin gels drive more production of endogenous collagen and elastin from encapsulated SMCs (Grassl et al. 2002; Long and Tranquillo 2003) and can be successfully implanted in the vasculature of animal models (Swartz et al. 2005). Additional fabrication techniques have been developed, including rolling concentric layers of cultured 2D SMC/fibroblast sheets (McAllister et al. 2009) or electrospinning aligned fibers into dense scaffolds (Bergmeister et al. 2012). PLA, PGA, and other such biopolymers have been utilized in conjunction with bone marrow cells, and frequently demonstrate repopulation with host-derived monocytes, SMCs, and ECs upon implantation in mice (Roh et al. 2010). Sequential seeding of primary human myofibroblasts and HUVECs on PGA, PLA, and PCL scaffolds generates the formation of collagen-rich, endothelialized neotissue with a tight endothelial barrier. This system was used to successfully study and model transport of lipoproteins (e.g., apoE, HDL) across the luminal surface – a key component of atherosclerotic plaque development (Robert et al. 2017). Polymeric models have been used to describe a relationship between scaffold porosity, neointimal organization, and mineralization in tissue-engineered vascular grafts. In large-pore grafts composed of PLA coated with poly(L-lactide-co-ε-caprolactone) (PLCL), a well-organized neointima containing SMCs and few macrophages was present after 12 months of implantation in mice; these grafts were largely free of calcification (Tara et al. 2014). In contrast, small-pore grafts composed exclusively of electrospun PLA developed significant macrophage infiltration (with few SMCs present) and substantial neointimal calcification. Degradation rates of these polymeric scaffolds have also been demonstrated to strongly affect calcific potential of engineered vasculature. Graft calcification is inhibited in rapidly-degrading PLA/PGA nanofiber-spun grafts implanted in mice, whereas slow-degrading tissues made of PLA alone rapidly calcified after 8 weeks of implantation (Sugiura et al. 2017). Decreased calcification coincided with significantly elevated levels of cellular infiltrates. Others have demonstrated similar findings where polymeric degradation rates impact the composition and mechanical properties of neotissue formation (van Haaften et al. 2019). Lastly, decellularized xenografts from cows, pigs, and other species have been utilized as a scaffold for construct development. Typically, these tissues are decellularized through detergent-based means or physical disruption and then treated with cross-linking agents such as glutaraldehyde in order to mitigate immune responses upon implantation (Schmidt and Baier 2000). Small intestinal submucosa and bovine carotid arteries have commonly been employed in this manner. Decellularized xenografts, in conjunction with autologous bone marrow cells, have been successfully implanted into animals and demonstrate long-term preservation of tissue morphology and function (Cho et al. 2005).

5.5 Incorporation of Microenvironmental Cues

In both vascular and valvular disease, constituents of the surrounding tissue microenvironment are known to be intimately involved in disease initiation and progression. As such, incorporation of microenvironmental cues into tissue-engineered constructs has been a point of focus for some time. Bioreactors are a key component of this approach, as they allow the incorporation of fluid flow, cyclic stretch, and cyclic compression into in vitro culture. Fluid shear stresses are typically administered to engineered tissues with the use of a pulsatile flow bioreactor, where culture media is pumped through the valve orifice in a cyclical manner. Physiological fluid shear rates support rapid maturation of engineered tissues, as they increase collagen synthesis (Ramaswamy et al. 2014). Perfusion of 3D scaffold cultures at physiological shear stresses mediates angiogenic sprouting, endothelial cell migration, vessel network morphogenesis, and ECM protein production versus static controls (Zohar et al. 2018). In vascular constructs, preconditioning culture in shear bioreactors modulates smooth muscle cell proliferation, homogeneous SMC distribution, increased collagen expression, and an improved aerobic cellular metabolism (Engbers-Buijtenhuijs et al. 2006). Besides being used as a pre-conditioning step, shear bioreactors are important for modeling the in vivo pathophysiology of vascular disease. For example, roles for shear-mediated endothelial cell and SMC cross-talk in modulating vessel integrity and endothelial dysfunction have been studied in tissue-engineered blood vessels composed of peripheral blood mononuclear cells seeded on polycaprolactone conduits and exposed to cyclic shear stress (Ragaseema et al. 2013). Production of nitric oxide and cyclic GMP were accelerated by administration of shear stress, independent of vessel wall morphology, while both factors modulate the onset of a contractile SMC phenotype. Cyclic stretch and flexion of engineered valvular and vascular tissues has also been employed in culture, both to speed/improve maturation of constructs and to model normal/disease-induced biomechanics. In the vasculature, elastic arteries are particularly exposed to high degrees of pulsatile cyclic stretch throughout the cardiac cycle; the eventual transmission of pulsatile stretch to the smaller resistance arteries (upon disease-induced stiffening of the larger elastic ones) is believed to play an important role in microvascular dysfunction and downstream organ damage (Thorin-Trescases and Thorin 2016). Rings of tissue-engineered vessels have undergone circumferential cyclic stretch in pneumatic (Adams et al. 2011) or hydraulic (Cooper et al. 2014) bioreactors. These systems are amenable to long-term culture (>1 week) at physiological frequencies (~1 Hz) and over range of physiological to pathological stretch (static-15% strain). Cyclic distention for up to 7 days drives reductions in stiffness and tensile strength, consistent with greater vessel compliance. Such approaches have also been used to induce differentiation and maturation of non-SMC cell types (e.g., mesenchymal stem cells) into contractile SMCs (Gong and Niklason 2008). Tissue oxygenation (or lack thereof) is believed to play a key role in the thickening and neovascularization of aortic valve leaflets during disease development. Furthermore, oxidative stress is known to be elevated in the initial and end stages of vascular and valvular calcification, though the specific downstream mechanisms that modulate this pathogenic effect are not fully clear. Hypoxic bioreactors have been used to demonstrate loss of valvular layer stratification and altered angiogenic signaling in engineered valvular constructs (Sapp et al. 2016); others have demonstrated the impact of hypoxia on myofibroblast-laden PGA meshes intended for tissue-engineered heart valves: reduced oxygen levels drove glycosaminoglycan production and collagen cross-linking, resulting in tissue constructs whose biomechanics more closely matched those of the native tissue (Balguid et al. 2009).

5.6 Integration with Computational Approaches to Model Disease

One growing application of these tissue-engineered in vitro models is as inputs to computational modeling of disease. As computational power grows and becomes less expensive, in silico approaches are gaining popularity due to their highly controllable and infinitely variable nature. Such models are, however, only as good as the boundary conditions within which they operate. When faced with a lack of baseline data (e.g., the “normal” condition) in humans or knowledge of clinical disease states, data from tissue-engineered models can act as a suitable substitute in order to define the boundaries of the simulation. For example, levels of the anti-angiogenic isoform of VEGF-A measured from human and mouse muscle biopsies in peripheral arterial disease have been utilized to computationally model the distribution of this VEGF isoform across multiple compartments of the body (Chu et al. 2016). Others have mathematically modeled inflammatory, lipid, and macrophage dynamics within atherosclerotic plaques using data from in vitro culture models (Thon et al. 2018). Multi-level models of the human circulation that incorporate data from 3D in vitro models as input to pulsatile, patient-specific models of the vasculature have been used for the purposes of surgical planning (Eslahpazir et al. 2017). These approaches have been miniaturized as well: through the use of stereolithography 3D printing, molds for 3D blood vessel constructs, reminiscent of normal or stenotic morphologies were created, then cast in PDMS and lined with human endothelial cells to create microfluidic chips of miniature human blood vessels (Costa et al. 2017). Computational fluid dynamics was then used to scale flow fields and wall shear rates between the original human artery and that on the microfluidic chip, and the devices were perfused with human blood at physiologically-relevant shear rates in order to assess the thrombotic potential of these vessel geometries. Others have used computational modeling to systematically tissue-engineered heart valve morphology and biomechanics while predicting leaflet remodeling upon surgical implantation, based upon initial valve geometry prior to implantation (Emmert et al. 2018).

Regardless of the specific approach utilized to produce the model of interest, the usage of next-generation quantitative techniques such as transcriptomics (Schlotter et al. 2018) and proteomics (Goto et al. 2019) in combination with cutting-edge systems biology (Lee et al. 2019) promises to further enable insight into the in vivo condition. Integration of these varied data types along with advanced imaging techniques and other sources of “big data” (e.g., electronic health records) has lagged behind fields such as oncology; the field is now moving towards the use of artificial intelligence and machine learning to integrate these data in such a way that provides novel mechanistic insights and potentiates drug discovery (Rogers and Aikawa 2019).

5.7 Drug Discovery/Testing

Once engineered models have been successfully developed and validated to recapitulate the relevant aspects of native tissue, the next logical step is to translate these models into the drug discovery/testing space. Currently, the vast majority of drug discovery occurs via high-throughput screening (HTS), where massive pre-existing libraries containing millions of compounds are screened for the excitatory/inhibitory capability of particular interest (Thorne et al. 2010). These screens are performed in either cell-free systems (e.g., to identify binding partners for a particular receptor), or in simple 2D cell culture. Because well-to-well variability must be highly controlled for valid hit selection (efficient HTS relies in part on limited technical replication), cell culture-based HTS is almost exclusively performed on immortalized cell lines, not primary cells. It should therefore be unsurprising that the vast majority of hits generated by current HTS systems do not result in therapeutically-effective compounds (Sink et al. 2010). This high false-positive rate is frequently attributed to the extreme deviations between screening in 2D cultured cell lines and human organs (Nierode et al. 2016). One promising application for tissue engineering is thus the development of construct arrays that enable HTS to be performed on 3D tissues that mimic the native microenvironment with high fidelity (Nierode et al. 2016). As is often the case with drug discovery, the field of cancer therapeutics has been at the forefront of this push. Multiple recent efforts have, in general, focused on combining bioprinted cell-repellent surfaces with self-assembled homogeneous 3D spheroids of tumor cell lines in 96–1536-well microtiter plates (Madoux et al. 2017; Kota et al. 2018; Hou et al. 2018).

5.8 High-Throughput Fabrication of Cardiovascular Models

3D-bioprinting utilizes controlled extrusion and deposition of cells and biomaterials to fabricate organoids, provides low construct-to-construct variability, and enables unsupervised fabrication (Murphy and Atala 2014; van der Ven et al. 2017). In a move towards scalable and reproducible production of complex cardiovascular organoids, 3D-bioprinted multi-layered aortic valve constructs with encapsulated primary VICs were recently developed (Valk et al. 2018). These constructs were comprised of GelMA/HAMA scaffolds whose stiffnesses were tuned to mimic the static biomechanics of the native aortic valve leaflet layers (as measured by microdissection and nanoindentation). Bioprinted constructs supported viability of primary human VICs over long-term culture (>20 days) and demonstrated in vivo-like collagen-rich remodeling of the construct ECM. Importantly, they recapitulated the layer-specific induction of calcification found in vivo, as VICs in the fibrosa-like layer formed microcalcifications under osteogenic stimulation, while those in the spongiosa-like layer did not calcify (Valk et al. 2018). Gelatin hydrogels support the growth of large-scale valvular microtissues in a format that is compatible with 3D bioprinting. In this context, encapsulated VICs remained quiescent and remodeled the surrounding ECM through enhanced expression of collagens, elastins, and proteoglycans. VIC proliferation and migration were also enhanced in these microtissues (Roosens et al. 2019). Other groups have utilized 3D bioprinting to fabricate complete aortic root-like conduits, composed of alginate/gelatin tri-leaflet structures and containing heterogenous mixtures of both VSMCs and VICs and whose mechanical properties mimick those of the native valve. Over long-term static culture, these VICs within these conduits remain viable, express vimentin and αSMA, and remodel their extracellular surroundings (Duan et al. 2013). When it comes to engineered vasculature for drug screening, 3D bioprinting has again been the consensus approach to date. Some groups have printed multi-layered, endothelialized blood vessels (intima) surrounded by an elastic layer of SMCs (media) and a collagenous matrix of fibroblasts (adventitia) that supports long-term culture, physiological flow, and high viability (Schöneberg et al. 2018). Others have employed supportive bioprinted scaffolds, onto which HUVECs, SMCs, and dermal fibroblasts were sequentially deposited by perfusion – mechanical testing confirmed that tissue moduli mimicked that of the native aorta after 48 h in culture (Xu et al. 2018). Notably, while 3D bioprinting approaches have the potential for scaling to multiwell arrays (simply print scaled-down versions of the full-sized model into multiwell plates), the reality is much less clear. Well-by-well bioprinting tends to result in high print times when scaled beyond 96-well plates, and care must be taken to ensure within-plate consistency between the first and last well printed. Bioprinting extrusion parameters and printhead hardware itself often require significant re-optimization, beyond the simple adjustment of scaling factors within the bioprinting code. At a smaller scale of throughput, micro-contact printing by stamping of silicone-based polydimethylsiloxane (PDMS) has been employed to generate arrays of cell-encapsulated constructs (Cirka et al. 2017). This technique generates highly reproducible calcification and cellular aggregation, and has been used to demonstrate induction of substrate–stress-dependent apoptosis alongside construct mineralization. Its promise lies in an inherent ability to allow for independent control of how biomechanics, endogenous cytokines, and pharmacologic compounds drive calcification. Other approaches for manufacturing HTS-compatible 3D constructs include hanging drop spheroid culture, self-assembling peptide hydrogels that do not require covalent or photo-crosslinking, and shear-thinning hydrogels that can be injection-molded into custom microplates (Worthington et al. 2017).

5.9 Automated Visualization and Quantitation of 3D Tissue-Engineered Models

One large challenge in 3D HTS is the visualization of organoid structures with automated imaging systems. Automated fluorescent microscopy or microplate readers are typically used to acquire functional readouts in HTS; however, thick 3D constructs suffer from scattering and absorption of light as well as poor light penetration, and lengthy image acquisition times needed to acquire Z-stacks through engineered tissues (Langhans 2018). Commonly used hydrogels (e.g., collagens) or artificial polymers suffer from these disadvantages. Emerging techniques such as tissue clearing and rapid 4D light sheet imaging may solve some of the acquisition time concerns associated with fluorescent or visible-light microscopy (Ding et al. 2018); other techniques such as multiphoton microscopy can dramatically improve light penetration depths in 3D spheroid culture (König et al. 2011). Others are pursuing more novel approaches, such as a unique micropillar culture chip that enables measurement of toxicity-associated parameters in alginate/fibrin-encapsulated 3D culture (Joshi et al. 2018). Optical coherence tomography (OCT), a near-infrared interferometry technique that enables rapid and deep tissue penetration, has been utilized on engineered blood vessels in long-term culture to assess real-time construct morphology (e.g., wall thickness) under pulsatile stimulation. OCT measurement of polymer degradation was well correlated with those of scanning electron microscopy and polarized light microscopy, offering the promise of online assessment of scaffold integrity during construct synthesis, fabrication, pre-conditioning, and modeling (Chen et al. 2018).

5.10 Microenvironmental Cues in High-Throughput Model Systems

As array fabrication and automated analysis tools have matured, so too have approaches designed to incorporate cardiovascular microenvironmental cues into tissue-engineered arrays. In this way, as 3D culture systems become popularized in the drug discovery and HTS field, so too will the ability to better recapitulate native in vivo conditions within HTS-compatible culture. Using precision machining, Parrish et al. developed a 96-channel perfusion microplate that accommodates fluorescent imaging, CT scanning, and particle tracing (for assessment of flow patterns) measurements (Parrish et al. 2018). These microplates support culture of soft (gelatin-based) and hard (calcium chloride) tissues, as well as vascular co-culture of HUVECs and MSCs – the latter of which were utilized to perform proof-of-concept HTS. In other studies, 24-well custom-fabricated microplates with embedded channels and pneumatic fluid delivery have been engineered to study platelet activation and thrombus formation under a range of physiological to pathological shear stresses (0.1–200 dyn/cm2) (Conant et al. 2011). Alternative approaches to apply shear to tissue-engineered constructs by stacking 96-well-microplate-sized arrays of culture wells atop aligned arrays of microchannels (Chebotarev and Simmons 2015). Culture medium is then drawn from a reservoir in a controllable manner (thereby enabling regulation of shear stresses, flow pulsatility, etc.) and passed through the microchannels, through the culture wells within which vascular/valvular constructs have been seeded. These stacks are compatible with standard liquid handling systems, and again, the incorporation of viewing ports enables quantification of gross construct parameters by plate reader systems. Along with shear stresses and pulsatile flow, devices have also been developed that support the delivery of cyclic stretch, compression, or flexion to arrays of tissue-engineered cardiovascular constructs. Early work by Moraes et al. produced arrays of bulging membranes that applied cyclic equibiaxial substrate strains to adherent cell populations, and demonstrated activation of pathogenic Wnt/β-catenin signaling in cultured primary porcine VICs (Moraes et al. 2010a). Liu and colleagues utilized cell-laden PEG-norbornene hydrogels bonded to deformable PDMS membranes to produce arrays of miniatured cyclic stretch bioreactors. By deforming the membranes using a pneumatic system, myofibroblastic differentiation of embedded MSCs was demonstrated in a context that was imageable and under a degree of stretch consistent with that present in the aortic valve and elastic arteries (Liu et al. 2016). In a complementary approach to the bulging membrane-induced cyclic stretch described above, cyclic compression arrays have been achieved by simply confining the upper surface (and optionally, the sides) of hydrogels bonded to the aforementioned PDMS membranes using a custom-fabricated faceplate that fits over the membrane array (Moraes et al. 2010b). In this way, when the membrane is bulged, the hydrogel is compressed against the faceplate; the degree of compression can be fine-tuned by modulating membrane bulge (Moraes et al. 2011). Compressive cyclic strains of ~6–45% have been achieved using these platforms, are generally uniform within the central portion of the hydrogel, and have enabled modeling of stress/strain transfer phenomena between the hydrogel, cell, and nuclear components of cultured MSCs (Moraes et al. 2010b) – an important means of studying mechanobiological signal transduction (Fig. 2).
Fig. 2

Idealized tissue-engineered models of cardiovascular calcification will seek to recapitulate the native in vivo microenvironment to the highest possible degree. Functionalized native matrix protein, hydrogels, or other polymeric substrates would be utilized to mimic the extracellular matrix. Multiple cell types may be incorporated, such as ECs, SMCs, VICs, macrophages, osteoblasts, chrondrocytes, etc. Primary cells may be obtained directly from human donors, or immortalized cell lines could be employed to ensure batch-to-batch model consistency. Alternatively, iPS cell sources could ensure robust repeatability by eliminating the donor-to-donor variability inherent in primary cells while avoiding the altered biology inherent to immortalized cell lines. These models would, as determined by the particular (patho)biology of interest, undergo dynamic conditioning before and during experimentation in bioreactors that enable carefully regulated delivery of biomechanical forces (e.g., cyclic stretch, cyclic flexure, blood shear stress), oxygenation (hypoxia, oxidative stress), circulating factors, and cell types (e.g., oxidized cholesterol, hyperphosphatemia, drugs of interest, inflammatory cells). Such an idealized model would be compatible with 3D-bioprinting or other manufacturing techniques that enable easily scalable production and allow spatial control over the location of multiple cell types, biomechanics, ECM structure/composition, and soluble factors within the model

5.11 Therapeutic Usage

Beyond the lower-hanging fruit of usage as research models and in drug discovery, the holy grail of cardiovascular tissue engineering is, of course, direct implantation for therapeutic use in humans. At the moment, traditional bioprosthetic valves are unsuitable for pediatric populations or younger adults, since these existing devices are incapable of growing along with the patient. Such prostheses cannot undergo self-repair, nor can they readily adapt to changes in the valvular microenvironment. Tissue engineering offers a promising alternative approach.

The first tissue-engineered heart valves were composed of PGA fibers seeded with fibroblasts and endothelial cells and were non-regurgitant 2 months after implantation in sheep (Shinoka et al. 1995). These early designs were frequently struck by stenotic remodeling (reviewed in (Blum et al. 2018)). Subsequent efforts with more supple scaffolds found more success: PGA-based scaffolds seeded with myofibroblasts and endothelial cells functioned normally for over 5 months in sheep, while ECM and DNA content rose to match those of the native tissue (Hoerstrup et al. 2000). Later incarnations employed PGA-based meshes and autologous bone marrow or peripheral blood cells. Initial attempts in this area used sheep as both the cell source and recipient, and tissue-engineered valves underwent bioreactor conditioning prior to implantation (Schmidt et al. 2010). Eight weeks after transapical implantation of valves on self-expanding stents, valves remained competent and histology found layered deposition of ECM and intact neo-endothelium. There was, however, a noted absence of elastin and an accumulation of myofibroblastic cells in the valvular interstitium. Later, rapid one-step cell harvesting and seeding in non-human primates provided proof-of-principle that extensive and lengthy pre-conditioning might be avoidable in humans (Weber et al. 2011), as the elapsed time from harvesting of bone marrow to valve implantation was less than 20 min. Primates were followed for up to 4 weeks, and scaffolds underwent rapid and layered accumulation of ECM along with cellular ingrowth, as evidenced by CD31 (endothelium), αSMA (myofibroblasts), CD68 (monocytic infiltration), and an absence of the originally-seeded bone marrow cells. Several animals displayed moderate regurgitation, perhaps indicative of the specific added benefit that dynamic pre-conditioning may provide. Despite the success of polymer-based valves in large animal models, they remain unstudied in clinical trials. At the moment, the field suffers from a lack of mechanistic understanding as to the drivers of recellularization, ECM synthesis, leaflet remodeling pre- and post-implantation (Blum et al. 2018).

Tissue engineered valve replacements moved into humans as pulmonary valve replacements or as important components of the Ross procedure, where a pulmonary valve is taken to replace a diseased aortic valve. To date, there has been only very limited delivery of tissue-engineered valves into the aortic position, due primarily to the large pressure differential and more challenging hemodynamic environment found around the aortic valve when compared with cardiac valves on the right side of the heart. The Tranquillo group has demonstrated success in this position in an ovine model, using a decellularized tissue produced by dermal fibroblasts seeded on fibrin gel and pre-conditioned for a total of 7 weeks in static and pulsatile conditions (Syedain et al. 2015). The decellularized tissue was collagen rich, with mechanical properties nearly identical to those of the native aortic valve. By 24 weeks after implantation, these valves had been endothelialized and repopulated by VIC-like cells. The recent expansion of transcatheter aortic valve replacement (TAVR) into the low risk patients has dramatically altered the treatment landscape. In the future, commercially-successful tissue-engineered heart valves will need to be compatible with TAVR-based delivery. One key concern is the impact of leaflet crimping on tissue-engineered valves prior to delivery (Alavi et al. 2014). Crimpled bioprosthetic leaflets suffer from significant increases in leaflet calcification and leaflet thickness when compared with un-crimped counterparts when exposed to calcifying stimuli under accelerated wear testing. This early calcification is likely due to microstructural ECM damage during crimping (Zareian et al. 2019).

In the field of vascular grafts, early trials in humans utilized allogeneic cell sheet constructs generated by fibroblasts or vascular SMCs that were implanted as arteriovenous shunts for dialysis (Wystrychowski et al. 2014). In a small number of patients, these constructs induced no immune reactions, construct degradation, or aneurysm, though there was a relatively high incidence of thrombosis – due perhaps to the absence of endothelialization before implantation. Small-diameter vascular grafts composed of acellular PGA scaffolds seeded with vascular SMCs, pre-conditioned under pulsatile flow for 8 weeks, then decellularized prior to implantation in humans did not develop aneurysms nor immune responses. While patency was >60% after 6 months, it dropped sharply to <30% after 1 year (Lawson et al. 2016). Other approaches have utilized bone marrow cells and PCL/PLA co-polymers, which were successfully implanted into patients as extracardiac total cavopulmonary connection grafts for up to 32 months with 100% patency and no appearance of aneurysm, thrombosis, or calcification (Hibino et al. 2010). Notably, anticoagulation therapy was employed for only up to 6 months after surgery, implying re-endothelialization occurred post-implantation. Decellularized allogeneic iliac veins reseeded with autologous bone marrow cells have seen limited (but successful use) in pediatric patients, without immunogenicity nor thrombosis (Olausson et al. 2012). To date, unlike in the field of tissue-engineered valve replacement, there remains little effort to engineer calcification resistance into vascular grafts destined for PAD or CABG; non-thrombogenicity and patency remain the primary concerns in these applications. It is also notable that therapeutic applications in humans have, to date, been limited principally to venous shunts and hemodialysis access – these constructs are not being utilized in the high-pressure, higher-stakes environment of the coronary artery or peripheral vasculature.

6 Conclusion

Tissue engineering is poised to revolutionize our understanding of and approaches to treat cardiovascular disease. Currently, the absence of imaging modalities with the sensitivity to detect the earliest stages of vascular and valvular calcification, as well as a lack of reliable animal models of CAVD that faithfully recapitulate the natural history of this disease in humans both dramatically hinder any progress that might otherwise be made in our understanding of pathological cardiovascular mineralization. The development and manufacture of cardiovascular constructs, organoids, and artificial tissues and their usage in combination with genetic engineering approaches (e.g., CRISPR), novel bioreactors, multimodal imaging, and next-generation omics holds the promise to allow careful and step-wise dissection of the factors and mechanisms that drive disease initiation, progression, mineral nucleation, and growth of macrocalcifications. Once the complex pathobiology can be robustly and reliably modeled in tissue-engineered constructs, that will more closely reflect the in vivo ground truth; it is likely that the bench-to-bedside translational drug/device discovery pipeline can be shortened and streamlined, since the use of models and screening systems which are more representative of the in vivo environment should result in fewer false-positives or false-negatives (Fig. 3). In terms of direct clinical applicability, one key area of this field will be the engineering of inherent calcification resistance into artificial tissues destined for surgical replacement or repair of diseased vessels and valves (e.g., AVR, CABG, or PABG). It remains to be seen whether such functionality is most efficiently and safely achieved by incorporating protein- or small-molecule inhibitors of mineralization into tissue-engineered scaffolds, by implanting tissues that are pre-seeded with exogenous cells which recapitulate the homeostatic functionality of now-dysfunctional endogenous resident cells, by introducing scaffolds that provide a niche which promotes the migration, capture, and engraftment of autologous cells, or some unique combination thereof. As this exciting field progresses, important questions continue to emerge: Can artificial tissues truly replace animal models in the mechanistic study of disease? What microenvironmental cues (biomechanical/chemical) are necessary? What degree/magnitude of these cues effectively recapitulates the native tissue? When designing and testing constructs, how can researchers deal with the extreme donor-to-donor variability often found in primary human cell cultures? Are iPS cells, cell lines, or xenobiotic cell sources more appropriate? How does the disruption of traditional cell-free or cell monolayer HTS by the incorporation of dramatically more complex engineered models impact the current economic paradigm of drug discovery? Inevitably, screening costs will rise as models become more intricate – presumably this will be offset by significant improvements in overall cost and more efficient lead identification, but this remains to be seen. Can calcification inhibitors be safely incorporated into vascular grafts without compromising either the stability of existing plaques or skeletal integrity? In the long run, tissue engineering holds the promise to enable us to overcome these and other roadblocks which have for decades vexed those working to understand and treat vascular calcification.
Fig. 3

The optimal model for a translational pipeline that incorporates tissue-engineered models of cardiovascular calcification. Clockwise from top-left: (1) Robust in vitro/in vivo models of the tissue/pathobiology of interest are engineered, with care taken to incorporate all relevant cell types, biomechanical stresses, biochemical factors, etc. These models are then utilized to carefully study the molecular mechanisms which regulate initiation and progression of disease. (2) Clinical imaging (e.g., ultrasound, MRI, PET-CT) of the pathology of interest in humans is performed to obtain functional human data on disease pathogenesis. (3) Such studies are then increasingly taking advantage of next-generation “omics” techniques that enable collection and study of the entire genome, miRNAome, transcriptome, and proteome from human and model system tissues. (4) Once targets have been identified by omics, the massive amount of data generated by this multi-omics approach is analyzed by systems biology techniques (including artificial intelligence and machine learning algorithms), which identify the most promising drug targets by combining omics data with clinical imaging and prior knowledge from in vitro models. (5) Once druggable targets have been identified, high-throughput screening can be employed in 3D models of disease that more accurately recapitulate the native tissue than do cell-free or 2D culture assays, thereby reducing the false-positive hit rate. Target hits can then undergo initial validation in robust in vitro models prior to more expensive and time-consuming animal and human clinical studies. (We gratefully acknowledge Dr. Sasha Singh for designing this figure)

Notes

Acknowledgments

This work was supported by the National Institutes of Health (NIH) grants R01HL136431, R01HL141917, and R01HL147095.

References

  1. Adams K, Bishop K, Casey E, et al (2011) Design of a bioreactor to cyclically strain tissue engineered blood vessel rings. In: 2011 IEEE 37th annual northeast bioengineering conference (NEBEC). IEEE, Troy, NY, pp 1–2Google Scholar
  2. Aikawa E, Libby P (2017) A rock and a hard place: chiseling away at the multiple mechanisms of aortic stenosis. Circulation 135:1951–1955.  https://doi.org/10.1161/CIRCULATIONAHA.117.027776CrossRefGoogle Scholar
  3. Aikawa E, Nahrendorf D, Sosnovik D et al (2007a) Multimodality molecular imaging identifies proteolytic and osteogenic activities in early aortic valve disease. Circulation 115:377CrossRefGoogle Scholar
  4. Aikawa E, Nahrendorf M, Figueiredo J et al (2007b) Osteogenesis associates with inflammation in early-stage atherosclerosis evaluated by molecular imaging in vivo. Circulation 116(24):2841–2850.  https://doi.org/10.1161/CIRCULATIONAHA.107.732867CrossRefGoogle Scholar
  5. Alavi SH, Groves EM, Kheradvar A (2014) The effects of transcatheter valve crimping on pericardial leaflets. Ann Thorac Surg 97:1260–1266.  https://doi.org/10.1016/j.athoracsur.2013.11.009CrossRefGoogle Scholar
  6. Alexander MR, Owens GK (2012) Epigenetic control of smooth muscle cell differentiation and phenotypic switching in vascular development and disease. Annu Rev Physiol 74:13–40.  https://doi.org/10.1146/annurev-physiol-012110-142315CrossRefGoogle Scholar
  7. Alexander JH, Smith PK (2016) Coronary-artery bypass grafting. N Engl J Med 374:1954–1964CrossRefGoogle Scholar
  8. Anderson H (1989) Mechanism of mineral formation in bone. Lab Investig 60:320–330Google Scholar
  9. Ardila D, Liou J-J, Maestas D et al (2019) Surface modification of electrospun scaffolds for endothelialization of tissue-engineered vascular grafts using human cord blood-derived endothelial cells. J Clin Med 8:185.  https://doi.org/10.3390/jcm8020185CrossRefGoogle Scholar
  10. Arjunon S, Rathan S, Jo H, Yoganathan A (2013) Aortic valve: mechanical environment and mechanobiology. Ann Biomed Eng 41:1331–1346CrossRefGoogle Scholar
  11. Babu AN, Meng X, Zou N et al (2008) Lipopolysaccharide stimulation of human aortic valve interstitial cells activates inflammation and osteogenesis. Ann Thorac Surg 86:71–76.  https://doi.org/10.1016/j.athoracsur.2008.03.008CrossRefGoogle Scholar
  12. Balguid A, Mol A, van Vlimmeren MAA et al (2009) Hypoxia induces near-native mechanical properties in engineered heart valve tissue. Circulation 119:290–297.  https://doi.org/10.1161/CIRCULATIONAHA.107.749853CrossRefGoogle Scholar
  13. Benjamin E, Virani S, Callaway C, et al (2018) American heart association 2018 statistics at glance. Circulation – Lippincott Williams & Wilkins, Philadelphia, PA, USAGoogle Scholar
  14. Bergmeister H, Grasl C, Walter I et al (2012) Electrospun small-diameter polyurethane vascular grafts: ingrowth and differentiation of vascular-specific host cells. Artif Organs 36:54–61.  https://doi.org/10.1111/j.1525-1594.2011.01297.xCrossRefGoogle Scholar
  15. Bernacca G, Mackay T (1992) A dynamic in vitro method for studying bioprosthetic heart valve calcification. J Mater Sci 3:293–298Google Scholar
  16. Bevilacqua M, Pober J, Mendrick D et al (1987) Identification of an inducible endothelial-leukocyte adhesion molecule. Proc Natl Acad Sci U S A 84:9238–9242CrossRefGoogle Scholar
  17. Blum KM, Drews JD, Breuer CK (2018) Tissue-engineered heart valves: a call for mechanistic studies. Tissue Eng Part B Rev 24:240–253.  https://doi.org/10.1089/ten.teb.2017.0425CrossRefGoogle Scholar
  18. Boström K, Watson KE, Horn S et al (1993) Bone morphogenetic protein expression in human atherosclerotic lesions. J Clin Invest 91:1800–1809.  https://doi.org/10.1172/JCI116391CrossRefGoogle Scholar
  19. Breen J, Sheedy P, Schwartz R et al (1992) Coronary artery calcification detected with ultrafast CT as an indication of coronary artery disease. Radiology 185:435–439CrossRefGoogle Scholar
  20. Brighton C, Hunt R (1978) Electron microscopic pyroantimonate studies of matrix vesicles and mitochondria in rachitic growth plate. Metab Bone Dis 1(3):199–204CrossRefGoogle Scholar
  21. Butcher JT, Penrod AM, García AJ, Nerem RM (2004) Unique morphology and focal adhesion development of valvular endothelial cells in static and fluid flow environments. Arterioscler Thromb Vasc Biol 24:1429–1434.  https://doi.org/10.1161/01.ATV.0000130462.50769.5aCrossRefGoogle Scholar
  22. Butcher J, Mahler G, Hockaday L (2011) Aortic valve disease and treatment: the need for naturally engineered solutions. Adv Drug Deliv Rev 63:242–268CrossRefGoogle Scholar
  23. Caplice NM, Bunch TJ, Stalboerger PG et al (2003) Smooth muscle cells in human coronary atherosclerosis can originate from cells administered at marrow transplantation. Proc Natl Acad Sci 100:4754–4759.  https://doi.org/10.1073/pnas.0730743100CrossRefGoogle Scholar
  24. Carabello BA, Paulus WJ (2009) Aortic stenosis. Lancet 373:956–966.  https://doi.org/10.1016/S0140-6736(09)60211-7CrossRefGoogle Scholar
  25. Carafoli E (1981) The uptake and release of calcium from mitochondria. In: Lees C, Shatz G, Dallner G (eds) Mitochondria and microsomes. Addison-Wesley Publ. Co, London, p 357Google Scholar
  26. Carafoli E, Crompton M (1978) The regulation of intracellular calcium by mitochondria. Ann N Y Acad Sci 307:269–284CrossRefGoogle Scholar
  27. Carpentier A (2007) The development of prosthetic heart valves – lessons in form and function. N Engl J Med 357:1368–1371CrossRefGoogle Scholar
  28. Cecil R, Anderson H (1978) Freeze-fracture studies of matrix vesicle calcification in epiphyseal growth plate. Metab Bone Dis 1(2):89–95CrossRefGoogle Scholar
  29. Charulatha V, Rajaram A (2003) Influence of different crosslinking treatments on the physical properties of collagen membranes. Biomaterials 24:759–767CrossRefGoogle Scholar
  30. Chebotarev O, Simmons CA (2015) Microfluidic device for cell-based assays. US Patent# US10481150B2: patents.google.com/patent/US10481150B2/enGoogle Scholar
  31. Chen W, Liu S, Yang J, et al (2018) Nondestructive monitoring of degradable scaffold-based tissue-engineered blood vessel development using optical coherence tomography. J Vis Exp:e58040.  https://doi.org/10.3791/58040
  32. Cho S-W, Lim SH, Kim I-K et al (2005) Small-diameter blood vessels engineered with bone marrow???Derived cells. Ann Surg 241:506–515.  https://doi.org/10.1097/01.sla.0000154268.12239.edCrossRefGoogle Scholar
  33. Chu L-H, Ganta VC, Choi MH et al (2016) A multiscale computational model predicts distribution of anti-angiogenic isoform VEGF165b in peripheral arterial disease in human and mouse. Sci Rep 6:37030.  https://doi.org/10.1038/srep37030CrossRefGoogle Scholar
  34. Cirka HA, Uribe J, Liang V et al (2017) Reproducible in vitro model for dystrophic calcification of cardiac valvular interstitial cells: insights into the mechanisms of calcific aortic valvular disease. Lab Chip 17:814–829.  https://doi.org/10.1039/C6LC01226DCrossRefGoogle Scholar
  35. Clark RA, Nielsen LD, Welch MP, McPherson JM (1995) Collagen matrices attenuate the collagen-synthetic response of cultured fibroblasts to TGF-beta. J Cell Sci 108(Pt 3):1251–1261Google Scholar
  36. Clinton S, Underwood R, Hayes L et al (1992) Macrophage colony-stimulating factor gene expression in vascular cells and in experimental and human atherosclerosis. Am J Pathol 1(140):301–316Google Scholar
  37. Conant CG, Schwartz MA, Beecher JE et al (2011) Well plate microfluidic system for investigation of dynamic platelet behavior under variable shear loads. Biotechnol Bioeng 108:2978–2987.  https://doi.org/10.1002/bit.23243CrossRefGoogle Scholar
  38. Cooper JL, Favreau JT, Gaudette GR, Rolle MW (2014) Effects of cyclic stretch on three-dimensional vascular smooth muscle cell rings. In: 2014 40th annual northeast bioengineering conference (NEBEC). IEEE, Boston, MA, pp 1–2Google Scholar
  39. Costa PF, Albers HJ, Linssen JEA et al (2017) Mimicking arterial thrombosis in a 3D-printed microfluidic in vitro vascular model based on computed tomography angiography data. Lab Chip 17:2785–2792.  https://doi.org/10.1039/C7LC00202ECrossRefGoogle Scholar
  40. Cribier AG (2014) The odyssey of TAVR. Texas Hear J 41(2):125–130CrossRefGoogle Scholar
  41. Cybulsky M, Gimbrone M Jr, Libby P (1991) Endothelial expression of a mononuclear leukocyte adhesion molecule during atherogenesis. Science 251(80):788–791CrossRefGoogle Scholar
  42. Davies PF, Truskey GA, Warren HB et al (1985) Metabolic cooperation between vascular endothelial cells and smooth muscle cells in co-culture: changes in low density lipoprotein metabolism. J Cell Biol 101:871–879.  https://doi.org/10.1083/jcb.101.3.871CrossRefGoogle Scholar
  43. Desai M, Seifalian AM, Hamilton G (2011) Role of prosthetic conduits in coronary artery bypass grafting. Eur J Cardio-thoracic Surg 40:394–398.  https://doi.org/10.1016/j.ejcts.2010.11.050CrossRefGoogle Scholar
  44. Detrano R, Markovic D, Simpfendorfer C et al (1985) Digital subtraction fluoroscopy: a new method of detecting coronary calcifications with improved sensitivity for the prediction of coronary disease. Circulation 71:725–732CrossRefGoogle Scholar
  45. Detrano R, Salcedo E, Hobb R, Yiannikas J (1986) Cardiac cinefluoroscopy as an inexpensive aid in the diagnosis of coronary artery diease. Am J Cardiol 56:1041–1046CrossRefGoogle Scholar
  46. Detrano R, Wong N, French W et al (1993) Prevalence of coronary calcific deposits in 1,000 high risk, asymptomatic individuals. Eur Heart J 13:P1065Google Scholar
  47. Ding Y, Ma J, Langenbacher AD, et al (2018) Multiscale light-sheet for rapid imaging of cardiopulmonary system. JCI Insight 3.  https://doi.org/10.1172/jci.insight.121396
  48. Doherty T, Detrano R (1994) Coronary arterial calcification as an active process: a new perspective on an old problem. Calcif Tissue Int 54(3):224–230CrossRefGoogle Scholar
  49. Doherty TM, Asotra K, Fitzpatrick LA et al (2003) Calcification in atherosclerosis: bone biology and chronic inflammation at the arterial crossroads. Proc Natl Acad Sci U S A 100:11201–11206.  https://doi.org/10.1073/pnas.1932554100CrossRefGoogle Scholar
  50. Duan B, Hockaday LA, Kang KH, Butcher JT (2013) 3D Bioprinting of heterogeneous aortic valve conduits with alginate/gelatin hydrogels. J Biomed Mater Res Part A 101A:1255–1264.  https://doi.org/10.1002/jbm.a.34420CrossRefGoogle Scholar
  51. Duan B, Yin Z, Hockaday Kang L et al (2016) Active tissue stiffness modulation controls valve interstitial cell phenotype and osteogenic potential in 3D culture. Acta Biomater 36:42–54.  https://doi.org/10.1016/j.actbio.2016.03.007CrossRefGoogle Scholar
  52. Eckert CE, Fan R, Mikulis B et al (2013) On the biomechanical role of glycosaminoglycans in the aortic heart valve leaflet. Acta Biomater 9:4653–4660CrossRefGoogle Scholar
  53. Elbjeirami WM, Yonter EO, Starcher BC, West JL (2003) Enhancing mechanical properties of tissue-engineered constructs via lysyl oxidase crosslinking activity. J Biomed Mater Res 66A:513–521.  https://doi.org/10.1002/jbm.a.10021CrossRefGoogle Scholar
  54. Emmert MY, Schmitt BA, Loerakker S et al (2018) Computational modeling guides tissue-engineered heart valve design for long-term in vivo performance in a translational sheep model. Sci Transl Med 10:eaan4587.  https://doi.org/10.1126/scitranslmed.aan4587CrossRefGoogle Scholar
  55. Engbers-Buijtenhuijs P, Buttafoco L, Poot AA et al (2006) Biological characterisation of vascular grafts cultured in a bioreactor. Biomaterials 27:2390–2397.  https://doi.org/10.1016/j.biomaterials.2005.10.016CrossRefGoogle Scholar
  56. Eslahpazir BA, Beggs KW, Lauters ZM et al (2017) Translating in-vitro and computational multi-scale models for the vascular surgeon. Ann Vasc Surg 38:7–8.  https://doi.org/10.1016/J.AVSG.2016.10.021CrossRefGoogle Scholar
  57. Falahatpisheh A, Morisawa D, Toosky TT, Kheradvar A (2017) A calcified polymeric valve for valve-in-valve applications. J Biomech 50:77–82.  https://doi.org/10.1016/J.JBIOMECH.2016.11.027CrossRefGoogle Scholar
  58. Fang Y, Eglen RM (2017) Three-dimensional cell cultures in drug discovery and development. SLAS Discov 22:456–472.  https://doi.org/10.1177/1087057117696795CrossRefGoogle Scholar
  59. Farivar RS, Cohn LH, Soltesz EG et al (2003) Transcriptional profiling and growth kinetics of endothelium reveals differences between cells derived from porcine aorta versus aortic valve. Eur J Cardiothorac Surg 24:527–534.  https://doi.org/10.1016/s1010-7940(03)00408-1CrossRefGoogle Scholar
  60. Frostegård J (2013) Immunity, atherosclerosis and cardiovascular disease. 1–8.  https://doi.org/10.1186/1741-7015-11-117
  61. Fukui T, Tabata M, Manabe S et al (2010) Graft selection and one-year patency rates in patients undergoing coronary artery bypass grafting. Ann Thorac Surg 89:1901–1905.  https://doi.org/10.1016/j.athoracsur.2010.02.016CrossRefGoogle Scholar
  62. Fung E, Tang SMT, Canner JP et al (2007) Delta-like 4 induces Notch signaling in macrophages: implications for inflammation. Circulation 115:2948–2956.  https://doi.org/10.1161/CIRCULATIONAHA.106.675462CrossRefGoogle Scholar
  63. Galis Z, Sukhova G, Lark M, Libby P (1994) Increased expression of matrix metalloproteinases and matrix degrading activity in vulnerable regions of human atherosclerotic plaques. J Clin Invest 94:2493–2503CrossRefGoogle Scholar
  64. Galvin KM, Donovan MJ, Lynch CA et al (2000) A role for Smad6 in development and homeostasis of the cardiovascular system. Nat Genet 24:171–174.  https://doi.org/10.1038/72835CrossRefGoogle Scholar
  65. Goettsch C, Hutcheson JD, Aikawa M et al (2016) Sortilin mediates vascular calcification via its recruitment into extracellular vesicles. J Clin Invest 126:1323–1336.  https://doi.org/10.1172/JCI80851CrossRefGoogle Scholar
  66. Gong Z, Niklason LE (2008) Small-diameter human vessel wall engineered from bone marrow-derived mesenchymal stem cells (hMSCs). FASEB J 22:1635–1648.  https://doi.org/10.1096/fj.07-087924CrossRefGoogle Scholar
  67. Gong G, Ling Z, Seifter E et al (1991) Aldehyde tanning: the villain in bioprosthetic calcification. Eur J Cardio-thoracic Surg 5:288–293.  https://doi.org/10.1016/1010-7940(91)90037-KCrossRefGoogle Scholar
  68. Gonzalez Rodriguez A, Schroeder ME, Walker CJ, Anseth KS (2018) FGF-2 inhibits contractile properties of valvular interstitial cell myofibroblasts encapsulated in 3D MMP-degradable hydrogels. APL Bioeng 2:046104.  https://doi.org/10.1063/1.5042430CrossRefGoogle Scholar
  69. Goto S, Rogers MA, Blaser MC et al (2019) Standardization of human calcific aortic valve disease in vitro modeling reveals passage-dependent calcification. Front Cardiovasc Med 6:1–12.  https://doi.org/10.3389/fcvm.2019.00049CrossRefGoogle Scholar
  70. Grassl ED, Oegema TR, Tranquillo RT (2002) Fibrin as an alternative biopolymer to type-I collagen for the fabrication of a media equivalent. J Biomed Mater Res 60:607–612CrossRefGoogle Scholar
  71. Gu L, Okada Y, Clinton S et al (1998) Absense of monocyte chemoattractant protein-1 reduces atherosclerosis in low density lipoprotein receptor-deficient mice. Mol Cell 2:275–281CrossRefGoogle Scholar
  72. Guangbin L, Patricia D, McKee MD et al (1997) Spontaneous calcification of arteries and cartilage in mice lacking matrix GLA protein. Nature 386:78–81CrossRefGoogle Scholar
  73. Healy K (2018) Tissue-engineered disease models. Nat Biomed Eng 2:879–880.  https://doi.org/10.1038/s41551-018-0339-2CrossRefGoogle Scholar
  74. Herrington W, Lacey B, Sherliker P et al (2016) Epidemiology of atherosclerosis and the potential to reduce the global burden of atherothrombotic disease. Circ Res 118:535–546.  https://doi.org/10.1161/CIRCRESAHA.115.307611CrossRefGoogle Scholar
  75. Hibino N, McGillicuddy E, Matsumura G et al (2010) Late-term results of tissue-engineered vascular grafts in humans. J Thorac Cardiovasc Surg 139:431–436.e2.  https://doi.org/10.1016/j.jtcvs.2009.09.057CrossRefGoogle Scholar
  76. Hinderer S, Seifert J, Votteler M et al (2014) Engineering of a bio-functionalized hybrid off-the-shelf heart valve. Biomaterials 35:2130–2139.  https://doi.org/10.1016/J.BIOMATERIALS.2013.10.080CrossRefGoogle Scholar
  77. Hjortnaes J, Aikawa E (2011) Calcific aortic valve disease. In: Ying-Fu Chen, Chwan-Yau Luo (eds) Aortic valve. InTech Publishing, Chicago, IL, USA, ISBN: 9789533075617Google Scholar
  78. Hjortnaes J, Camci-Unal G, Hutcheson JD et al (2014) Directing valvular interstitial cell myofibroblast-like differentiation in a hybrid hydrogel platform. Adv Heal Mater 4(1):121–130.  https://doi.org/10.1002/adhm.201400029CrossRefGoogle Scholar
  79. Hjortnaes J, Shapero K, Goettsch C et al (2015) Valvular interstitial cells suppress calcification of valvular endothelial cells. Atherosclerosis 242:251–260.  https://doi.org/10.1016/j.atherosclerosis.2015.07.008CrossRefGoogle Scholar
  80. Hjortnaes J, Goettsch C, Hutcheson JD, et al (2016) Simulation of early calcific aortic valve disease in a 3D platform: a role for myofibroblast differentiation. J Mol Cell Cardiol.  https://doi.org/10.1016/j.yjmcc.2016.03.004CrossRefGoogle Scholar
  81. Hoerstrup SP, Sodian R, Daebritz S et al (2000) Functional living trileaflet heart valves grown in vitro. Circulation 102:III44–III49CrossRefGoogle Scholar
  82. Honda S, Miyamoto T, Watanabe T et al (2014) A novel mouse model of aortic valve stenosis induced by direct wire injury. Arterioscler Thromb Vasc Biol 34:270–278.  https://doi.org/10.1161/ATVBAHA.113.302610CrossRefGoogle Scholar
  83. Hou S, Tiriac H, Sridharan BP et al (2018) Advanced development of primary pancreatic organoid tumor models for high-throughput phenotypic drug screening. SLAS Discov Adv life Sci R D 23:574–584.  https://doi.org/10.1177/2472555218766842CrossRefGoogle Scholar
  84. Hutcheson JD, Goettsch C, Bertazzo S, et al (2016) Genesis and growth of extracellular-vesicle-derived microcalcification in atherosclerotic plaques. Nat Mater.  https://doi.org/10.1038/nmat4519CrossRefGoogle Scholar
  85. Hutcheson J, Blaser M, Aikawa E (2017) Giving calcification its due: recognition of a diverse disease a first attempt to standardize the field. Circ Res 120:270–273.  https://doi.org/10.1161/CIRCRESAHA.116.310060.GivingCrossRefGoogle Scholar
  86. Islam K, Timraz SB, Nasser R et al (2016) Co-culture methods used to model atherosclerosis in vitro using endothelial, smooth muscle and monocyte cells. SM J Biomed Eng 1:1–15CrossRefGoogle Scholar
  87. Joshi P, Datar A, Yu K-N et al (2018) High-content imaging assays on a miniaturized 3D cell culture platform. Toxicol Vitr 50:147–159.  https://doi.org/10.1016/J.TIV.2018.02.014CrossRefGoogle Scholar
  88. Jover E, Fagnano M, Angelini G, Madeddu P (2018) Cell sources for tissue engineering strategies to treat calcific valve disease. Front Cardiovasc Med 5:155.  https://doi.org/10.3389/fcvm.2018.00155CrossRefGoogle Scholar
  89. Kapolos J, Mavrilas D, Missirlis Y, Koutsoukos PG (1997) Model experimental system for investigation of heart valve calcification in vitro. J Biomed Mater Res 38(3):183–190CrossRefGoogle Scholar
  90. König K, Uchugonova A, Gorjup E (2011) Multiphoton fluorescence lifetime imaging of 3D-stem cell spheroids during differentiation. Microsc Res Tech 74:9–17.  https://doi.org/10.1002/jemt.20866CrossRefGoogle Scholar
  91. Kota S, Hou S, Guerrant W et al (2018) A novel three-dimensional high-throughput screening approach identifies inducers of a mutant KRAS selective lethal phenotype. Oncogene 37:4372–4384.  https://doi.org/10.1038/s41388-018-0257-5CrossRefGoogle Scholar
  92. Kuro-o M, Nabeshima Y, Matsumura Y et al (1997) Mutation of the mouse klotho gene leads to a syndrome resembling ageing. Nature 390:45–51.  https://doi.org/10.1038/36285CrossRefGoogle Scholar
  93. Langhans SA (2018) Three-dimensional in vitro cell culture models in drug discovery and drug repositioning. Front Pharmacol 9:6.  https://doi.org/10.3389/fphar.2018.00006CrossRefGoogle Scholar
  94. Lawson JH, Glickman MH, Ilzecki M et al (2016) Bioengineered human acellular vessels for dialysis access in patients with end-stage renal disease: two phase 2 single-arm trials. Lancet 387:2026–2034.  https://doi.org/10.1016/S0140-6736(16)00557-2CrossRefGoogle Scholar
  95. Lee LH, Halu A, Morgan S et al (2019) XINA: a workflow for the integration of multiplexed proteomics kinetics data with network analysis. J Proteome Res 18:775–781.  https://doi.org/10.1021/acs.jproteome.8b00615CrossRefGoogle Scholar
  96. Leopold JA (2012) Cellular mechanisms of aortic valve calcification. Circ Cardiovasc Interv 5:605–614.  https://doi.org/10.1161/CIRCINTERVENTIONS.112.971028CrossRefGoogle Scholar
  97. Levy R, Schoen F, Levy J et al (1983) Biologic determinants of dystrophic calcification and osteocalcin deposition in glutaraldehyde-preserved porcine aortic valve leaflets implanted subcutaneously in rats. Am J Pathol 113:143–155Google Scholar
  98. Li H, Cybulsky M, Gimbrone M Jr, Libby P (1993) An atherogenic diet rapidly induces VCAM-1, a cytokine-regulatable mononuclear leukocyte adhesion molecule, in rabbit aortic endothelium. Arterioscler Thromb 13:197–204CrossRefGoogle Scholar
  99. Li C, Xu S, Gotlieb AI (2013) The progression of calcific aortic valve disease through injury, cell dysfunction, and disruptive biologic and physical force feedback loops. Cardiovasc Pathol 22:1–8.  https://doi.org/10.1016/j.carpath.2012.06.005CrossRefGoogle Scholar
  100. Libby P (2002) Inflammation in atherosclerosis. Nature 420:868–874.  https://doi.org/10.1038/nature01323CrossRefGoogle Scholar
  101. Libby P (2008) The molecular mechanisms of the thrombotic complications of atherosclerosis. J Intern Med 263:517–527.  https://doi.org/10.1111/j.1365-2796.2008.01965.x.TheCrossRefGoogle Scholar
  102. Libby P, Clinton S (1992) Cytokines as mediators of vascular pathology. Nouv Rev Fr Hematol 34(Suppl):S47–S53Google Scholar
  103. Liu H, Usprech J, Sun Y, Simmons CA (2016) A microfabricated platform with hydrogel arrays for 3D mechanical stimulation of cells. Acta Biomater 34:113–124.  https://doi.org/10.1016/j.actbio.2015.11.054CrossRefGoogle Scholar
  104. Long JL, Tranquillo RT (2003) Elastic fiber production in cardiovascular tissue-equivalents. Matrix Biol 22:339–350CrossRefGoogle Scholar
  105. Lu D, Kassab GS (2011) Role of shear stress and stretch in vascular mechanobiology. J R Soc Interface 8:1379–1385.  https://doi.org/10.1098/rsif.2011.0177CrossRefGoogle Scholar
  106. Mabry KM, Lawrence RL, Anseth KS (2015) Dynamic stiffening of poly(ethylene glycol)-based hydrogels to direct valvular interstitial cell phenotype in a three-dimensional environment. Biomaterials 49:47–56.  https://doi.org/10.1016/j.biomaterials.2015.01.047CrossRefGoogle Scholar
  107. Mabry KM, Payne SZ, Anseth KS (2016) Microarray analyses to quantify advantages of 2D and 3D hydrogel culture systems in maintaining the native valvular interstitial cell phenotype. Biomaterials 74:31–41.  https://doi.org/10.1016/j.biomaterials.2015.09.035CrossRefGoogle Scholar
  108. Mach F, Sauty A, Iarossi A et al (1999) Differential expression of three T lymphocyte-activating CXC chemokines by human atheroma-associated cells. J Clin Invest 104:1041–1050CrossRefGoogle Scholar
  109. Mack MJ, Leon MB, Thourani VH et al (2019) Transcatheter aortic-valve replacement with a balloon-expandable valve in low-risk patients. N Engl J Med 380:1695–1705.  https://doi.org/10.1056/NEJMoa1814052CrossRefGoogle Scholar
  110. Madoux F, Tanner A, Vessels M et al (2017) A 1536-well 3D viability assay to assess the cytotoxic effect of drugs on spheroids. SLAS Discov Adv Life Sci R&D 22:516–524.  https://doi.org/10.1177/2472555216686308CrossRefGoogle Scholar
  111. Mallone A, Stenger C, Von Eckardstein A et al (2018) Biofabricating atherosclerotic plaques: in vitro engineering of a three-dimensional human fibroatheroma model. Biomaterials 150:49–59.  https://doi.org/10.1016/j.biomaterials.2017.09.034CrossRefGoogle Scholar
  112. Mathieu P, Voisine P, Pepin A et al (2005) Calcification of human valve interstitial cells is dependent on alkaline phosphatase activity. J Heart Valve Dis 14:353–357Google Scholar
  113. McAllister TN, Maruszewski M, Garrido SA et al (2009) Effectiveness of haemodialysis access with an autologous tissue-engineered vascular graft: a multicentre cohort study. Lancet 373:1440–1446.  https://doi.org/10.1016/S0140-6736(09)60248-8CrossRefGoogle Scholar
  114. Mohler ER, Sheridan MJ, Nichols R (1991) Development and progression of aortic valve stenosis: atherosclerosis risk factors--a causal relationship? A clinical morphologic study. Clin Cardiol 14:995–999CrossRefGoogle Scholar
  115. Moore K, Tabas I (2011) Macrophages in the pathogenesis of atherosclerosis. Cell 145:341–355CrossRefGoogle Scholar
  116. Moraes C, Chen J-H, Sun Y, Simmons CA (2010a) Microfabricated arrays for high-throughput screening of cellular response to cyclic substrate deformation. Lab Chip 10:227–234.  https://doi.org/10.1039/B914460ACrossRefGoogle Scholar
  117. Moraes C, Wang G, Sun Y, Simmons CA (2010b) A microfabricated platform for high-throughput unconfined compression of micropatterned biomaterial arrays. Biomaterials 31:577–584.  https://doi.org/10.1016/J.BIOMATERIALS.2009.09.068CrossRefGoogle Scholar
  118. Moraes C, Zhao R, Likhitpanichkul M et al (2011) Semi-confined compression of microfabricated polymerized biomaterial constructs. J Micromechanics Microengineering 21:054014.  https://doi.org/10.1088/0960-1317/21/5/054014CrossRefGoogle Scholar
  119. Mozaffarian D, Benjamin EJ, Go AS, et al (2016) AHA statistical update heart disease and stroke statistics – 2016 update a report from the American Heart Association WRITING GROUP MEMBERS. Circulation – Lippincott Williams & Wilkins, Philadelphia, PA, USAGoogle Scholar
  120. Munroe PB, Olgunturk RO, Fryns JP et al (1999) Mutations in the gene encoding the human matrix Gla protein cause Keutel syndrome. Nat Genet 21:142–144.  https://doi.org/10.1038/5102CrossRefGoogle Scholar
  121. Murphy SV, Atala A (2014) 3D bioprinting of tissues and organs. Nat Biotechnol 32:773–785.  https://doi.org/10.1038/nbt.2958CrossRefGoogle Scholar
  122. Musunuru K, Sheikh F, Gupta RM et al (2018) Induced pluripotent stem cells for cardiovascular disease modeling and precision medicine: a scientific statement from the American Heart Association. Circ Genomic Precis Med 11:e000043.  https://doi.org/10.1161/HCG.0000000000000043CrossRefGoogle Scholar
  123. Navab M, Hough GP, Stevenson LW et al (1988) Monocyte migration into the subendothelial space of human aortic endothelial and smooth muscle cells coculture of adult. J Clin 82:1853–1863Google Scholar
  124. New SEP, Aikawa E (2011a) Cardiovascular calcification. Circ J 75:1305–1313.  https://doi.org/10.1253/circj.cj-11-0395CrossRefGoogle Scholar
  125. New SEP, Aikawa E (2011b) Molecular imaging insights into early inflammatory stages of arterial and aortic valve calcification. Circ Res 108:1381–1391.  https://doi.org/10.1161/CIRCRESAHA.110.234146CrossRefGoogle Scholar
  126. New SEP, Goettsch C, Aikawa M et al (2013) Macrophage-derived matrix vesicles: an alternative novel mechanism for microcalcification in atherosclerotic plaques. Circ Res 113:72–77.  https://doi.org/10.1161/CIRCRESAHA.113.301036CrossRefGoogle Scholar
  127. Nierode G, Kwon PS, Dordick JS, Kwon S-J (2016) Cell-based assay design for high-content screening of drug candidates. J Microbiol Biotechnol 26:213–225.  https://doi.org/10.4014/jmb.1508.08007CrossRefGoogle Scholar
  128. Nigwekar SU, Kroshinksy D, Nazarian RM et al (2015) Calciphylaxis: risk factors, diagnosis, and treatment. Am J Kidney Dis 66:133–146.  https://doi.org/10.1053/j.ajkd.2015.01.034.CalciphylaxisCrossRefGoogle Scholar
  129. O’Brien KD (2006a) Pathogenesis of calcific aortic valve disease: a disease process comes of age (and a good deal more). Arterioscler Thromb Vasc Biol 26:1721–1728.  https://doi.org/10.1161/01.ATV.0000227513.13697.acCrossRefGoogle Scholar
  130. O’Brien KD (2006b) Pathogenesis of calcific aortic valve disease a disease process comes of age (and a good deal more). ATVB 26:1721–1728.  https://doi.org/10.1161/01.ATV.0000227513.13697.acCrossRefGoogle Scholar
  131. Olausson M, Patil PB, Kuna VK et al (2012) Transplantation of an allogeneic vein bioengineered with autologous stem cells: a proof-of-concept study. Lancet 380:230–237.  https://doi.org/10.1016/S0140-6736(12)60633-3CrossRefGoogle Scholar
  132. Ortolani F, Rigonat L, Bonetti A et al (2010) Cells in a novel in vitro model simulating dystrophic. Ital J Anat Embryol 115:135–139Google Scholar
  133. Otto CM, Kuusisto J, Reichenbach DD (1994) Characterization of the early lesion of “degenerative” valvular aortic stenosis. Histological and immunohistochemical studies. Circulation 90:844–853.  https://doi.org/10.1161/01.CIR.90.2.844CrossRefGoogle Scholar
  134. Parrish J, Lim KS, Baer K et al (2018) A 96-well microplate bioreactor platform supporting individual dual perfusion and high-throughput assessment of simple or biofabricated 3D tissue models. Lab Chip 18:2757–2775.  https://doi.org/10.1039/C8LC00485DCrossRefGoogle Scholar
  135. Pereira L, Lee S, Gayraud B et al (1999) Pathogenetic sequence for aneurysm revealed in mice underexpressing fibrillin-1. Proc Natl Acad Sci U S A 96:3819–3823CrossRefGoogle Scholar
  136. Popma JJ, Deeb GM, Yakubov SJ et al (2019) Transcatheter aortic-valve replacement with a self-expanding valve in low-risk patients. N Engl J Med 380:1706–1715.  https://doi.org/10.1056/NEJMoa1816885CrossRefGoogle Scholar
  137. Porras AM, Westlund JA, Evans AD, Masters KS (2018) Creation of disease-inspired biomaterial environments to mimic pathological events in early calcific aortic valve disease. Proc Natl Acad Sci 115:E363–E371.  https://doi.org/10.1073/PNAS.1704637115CrossRefGoogle Scholar
  138. Proudfoot D, Skepper JN, Shanahan CM, Weissberg PL (1998) Calcification of human vascular cells in vitro is correlated with high levels of matrix Gla protein and low levels of osteopontin expression. Arterioscler Thromb Vasc Biol 18:379–388.  https://doi.org/10.1161/01.ATV.18.3.379CrossRefGoogle Scholar
  139. Qiao JH, Xie PZ, Fishbein MC et al (1994) Pathology of atheromatous lesions in inbred and genetically engineered mice. Genetic determination of arterial calcification. Atheroscler Thromb Vasc Biol 14:1480–1497Google Scholar
  140. Ragaseema VM, Columbus S, Ramesh R, Krishnan L (2013) Potential of tissue engineered blood vessel as model to study effect of flow and wall thickness on cellular communication. Curr Tissue Eng 3:39–46.  https://doi.org/10.2174/2211542002666131209233849CrossRefGoogle Scholar
  141. Rajamannan NM (2009) Calcific aortic stenosis: lessons learned from experimental and clinical studies. Arterioscler Thromb Vasc Biol 29:162–168.  https://doi.org/10.1161/ATVBAHA.107.156752CrossRefGoogle Scholar
  142. Rajamannan NM, Subramaniam M, Rickard D et al (2003) Human aortic valve calcification is associated with an osteoblast phenotype. Circulation 107:2181–2184.  https://doi.org/10.1161/01.CIR.0000070591.21548.69CrossRefGoogle Scholar
  143. Rajamannan NM, Evans FJ, Aikawa E et al (2011) Calcific aortic valve disease: not simply a degenerative process: a review and agenda for research from the National Heart and Lung and Blood Institute Aortic Stenosis Working Group. Executive summary: calcific aortic valve disease-2011 update. Circulation 124:1783–1791.  https://doi.org/10.1161/CIRCULATIONAHA.110.006767CrossRefGoogle Scholar
  144. Ramaswamy S, Boronyak SM, Le T et al (2014) A novel bioreactor for mechanobiological studies of engineered heart valve tissue formation under pulmonary arterial physiological flow conditions. J Biomech Eng 136:121009.  https://doi.org/10.1115/1.4028815CrossRefGoogle Scholar
  145. Rattazzi M, Bennett B, Bea F et al (2005) Calcification of advanced atherosclerotic lesions in the innominate arteries of apoe-deficient mice: potential role of chondrocyte-like cells. Atheroscler Thromb Vasc Biol 25:1420–1425CrossRefGoogle Scholar
  146. Rattazzi M, Iop L, Faggin E et al (2008) Clones of interstitial cells from bovine aortic valve exhibit different calcifying potential when exposed to endotoxin and phosphate. Arterioscler Thromb Vasc Biol 28:2165–2172.  https://doi.org/10.1161/ATVBAHA.108.174342CrossRefGoogle Scholar
  147. Richards JM, Kunitake JAMR, Hunt HB et al (2018) Crystallinity of hydroxyapatite drives myofibroblastic activation and calcification in aortic valves. Acta Biomater 71:24–36.  https://doi.org/10.1016/j.actbio.2018.02.024CrossRefGoogle Scholar
  148. Ridker PM (2009) The JUPITER trial. Circ Cardiovasc Qual Outcomes 2:279–285.  https://doi.org/10.1161/CIRCOUTCOMES.109.868299CrossRefGoogle Scholar
  149. Riedl A, Schlederer M, Pudelko K et al (2017) Comparison of cancer cells in 2D vs 3D culture reveals differences in AKT–mTOR–S6K signaling and drug responses. J Cell Sci 130:203–218.  https://doi.org/10.1242/jcs.188102CrossRefGoogle Scholar
  150. Robert J, Button EB, Yuen B et al (2017) Clearance of beta-amyloid is facilitated by apolipoprotein E and circulating high-density lipoproteins in bioengineered human vessels. elife 6:e29595.  https://doi.org/10.7554/eLife.29595CrossRefGoogle Scholar
  151. Rogers MA, Aikawa E (2019) Cardiovascular calcification: artificial intelligence and big data accelerate mechanistic discovery. Nat Rev Cardiol 16:261–274.  https://doi.org/10.1038/s41569-018-0123-8CrossRefGoogle Scholar
  152. Rogers MA, Aikawa M, Aikawa E (2018) Macrophage heterogeneity complicates reversal of calcification in cardiovascular tissues. Circ Res 121:5–7.  https://doi.org/10.1161/CIRCRESAHA.117.311219.MacrophageCrossRefGoogle Scholar
  153. Roh JD, Sawh-Martinez R, Brennan MP et al (2010) Tissue-engineered vascular grafts transform into mature blood vessels via an inflammation-mediated process of vascular remodeling. Proc Natl Acad Sci 107:4669–4674.  https://doi.org/10.1073/pnas.0911465107CrossRefGoogle Scholar
  154. Rong JX, Shapiro M, Trogan E, Fisher EA (2003) Transdifferentiation of mouse aortic smooth muscle cells to a macrophage-like state after cholesterol loading. Proc Natl Acad Sci 100:13531–13536.  https://doi.org/10.1073/pnas.1735526100CrossRefGoogle Scholar
  155. Roosens A, Handoyo YP, Dubruel P, Declercq H (2019) Impact of modified gelatin on valvular microtissues. J Tissue Eng Regen Med 13:771–784.  https://doi.org/10.1002/term.2825CrossRefGoogle Scholar
  156. Rosenfeld M, Yla-Herttuala S, Lipton B et al (1992) Macrophage colony-stimulating factor mRNA and protein in atherosclerotic lesions of rabbits and humans. Am J Pathol 140:291–300Google Scholar
  157. Sahasakul Y, Edwards WD, Naessens JM, Tajik AJ (1988) Age-related changes in aortic and mitral valve thickness: implications for two-dimensional echocardiography based on an autopsy study of 200 normal human hearts. Am J Cardiol 62:424–430CrossRefGoogle Scholar
  158. Sapp MC, Krishnamurthy VK, Puperi DS et al (2016) Differential cell-matrix responses in hypoxia-stimulated aortic versus mitral valves. J R Soc Interface 13:20160449.  https://doi.org/10.1098/rsif.2016.0449CrossRefGoogle Scholar
  159. Schlotter F, Halu A, Goto S et al (2018) Spatiotemporal multi-omics mapping generates a molecular atlas of the aortic valve and reveals networks driving disease. Circulation 138:377–393.  https://doi.org/10.1161/CIRCULATIONAHA.117.032291CrossRefGoogle Scholar
  160. Schmidt CE, Baier JM (2000) Acellular vascular tissues: natural biomaterials for tissue repair and tissue engineering. Biomaterials 21:2215–2231CrossRefGoogle Scholar
  161. Schmidt D, Dijkman PE, Driessen-Mol A et al (2010) Minimally-invasive implantation of living tissue engineered heart valves: a comprehensive approach from autologous vascular cells to stem cells. J Am Coll Cardiol 56:510–520.  https://doi.org/10.1016/J.JACC.2010.04.024CrossRefGoogle Scholar
  162. Schoen F, Levy R, Nelson A, Al E (1985) Onset and progression of experimental bioprosthetic heart valve calcification. Lab Investig 52:523–532Google Scholar
  163. Schöneberg J, De Lorenzi F, Theek B et al (2018) Engineering biofunctional in vitro vessel models using a multilayer bioprinting technique. Sci Rep 8:10430.  https://doi.org/10.1038/s41598-018-28715-0CrossRefGoogle Scholar
  164. Scott M, Vesely I (1995) Aortic valve cusp microstructure: the role of elastin. Ann Thorac Surg 60(2 Suppl):S391–S394CrossRefGoogle Scholar
  165. Sheikh AM, Livesey SA (2010) Surgical management of valve disease in the early 21st century. Clin Med 10:177–181CrossRefGoogle Scholar
  166. Shinoka T, Breuer CK, Tanel RE et al (1995) Tissue engineering heart valves: valve leaflet replacement study in a lamb model. Ann Thorac Surg 60:S513–S516CrossRefGoogle Scholar
  167. Sider KL, Blaser MC, Simmons CA (2011a) Animal models of calcific aortic valve disease. Int J Inflam 2011:1–18.  https://doi.org/10.4061/2011/364310CrossRefGoogle Scholar
  168. Sider KL, Blaser MC, Simmons CA (2011b) Animal models of calcific aortic valve disease. Int J Inflam 2011:1–18.  https://doi.org/10.4061/2011/364310CrossRefGoogle Scholar
  169. Sink R, Gobec S, Pečar S, Zega A (2010) False positives in the early stages of drug discovery. Curr Med Chem 17:4231–4255CrossRefGoogle Scholar
  170. Speer MY, Yang H, Brabb T, et al (2009) Smooth muscle cells give rise to osteochondrogenic precursors and chondrocytes in calcifying arteries. pp 733–741.  https://doi.org/10.1161/CIRCRESAHA.108.183053CrossRefGoogle Scholar
  171. Stella JA, Sacks MS (2007) On the biaxial mechanical properties of the layers of the aortic valve leaflet. J Biomech Eng 129:757–766CrossRefGoogle Scholar
  172. Sugiura T, Tara S, Nakayama H et al (2017) Fast-degrading bioresorbable arterial vascular graft with high cellular infiltration inhibits calcification of the graft. J Vasc Surg 66:243–250.  https://doi.org/10.1016/j.jvs.2016.05.096CrossRefGoogle Scholar
  173. Swartz DD, Russell JA, Andreadis ST (2005) Engineering of fibrin-based functional and implantable small-diameter blood vessels. Am J Physiol Circ Physiol 288:H1451–H1460.  https://doi.org/10.1152/ajpheart.00479.2004CrossRefGoogle Scholar
  174. Syedain Z, Reimer J, Schmidt J et al (2015) 6-month aortic valve implantation of an off-the-shelf tissue-engineered valve in sheep. Biomaterials 73:175–184.  https://doi.org/10.1016/j.biomaterials.2015.09.016CrossRefGoogle Scholar
  175. Tabas I, García-Cardeña G, Owens GK (2015) Recent insights into the cellular biology of atherosclerosis. J Cell Biol 209:13–22.  https://doi.org/10.1083/jcb.201412052CrossRefGoogle Scholar
  176. Takagi H, Hari Y, Kawai N, et al (2019) Meta-analysis of transcatheter aortic valve implantation for bicuspid versus tricuspid aortic valves. J Cardiol.  https://doi.org/10.1016/j.jjcc.2019.03.018CrossRefGoogle Scholar
  177. Takahashi K, Yamanaka S (2006) Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell 126:663–676.  https://doi.org/10.1016/j.cell.2006.07.024CrossRefGoogle Scholar
  178. Tara S, Kurobe H, Rocco KA et al (2014) Well-organized neointima of large-pore poly(l-lactic acid) vascular graft coated with poly(l-lactic-co-ε-caprolactone) prevents calcific deposition compared to small-pore electrospun poly(l-lactic acid) graft in a mouse aortic implantation model. Atherosclerosis 237:684–691.  https://doi.org/10.1016/j.atherosclerosis.2014.09.030CrossRefGoogle Scholar
  179. Theodoris CV, Li M, White MP et al (2015) Human disease modeling reveals integrated transcriptional and epigenetic mechanisms of NOTCH1 haploinsufficiency. Cell 160:1072–1086.  https://doi.org/10.1016/j.cell.2015.02.035CrossRefGoogle Scholar
  180. Thie M, Schlumberger W, Semich R et al (1991) Aortic smooth muscle cells in collagen lattice culture: effects on ultrastructure, proliferation and collagen synthesis. Eur J Cell Biol 55:295–304Google Scholar
  181. Thon MP, Ford HZ, Gee MW, Myerscough MR (2018) A quantitative model of early atherosclerotic plaques parameterized using in vitro experiments. Bull Math Biol 80:175–214.  https://doi.org/10.1007/s11538-017-0367-1MathSciNetCrossRefzbMATHGoogle Scholar
  182. Thorin-Trescases N, Thorin E (2016) Lifelong cyclic mechanical strain promotes large elastic artery stiffening: increased pulse pressure and old age-related organ failure. Can J Cardiol 32:624–633.  https://doi.org/10.1016/j.cjca.2015.12.022CrossRefGoogle Scholar
  183. Thorne N, Auld DS, Inglese J (2010) Apparent activity in high-throughput screening: origins of compound-dependent assay interference. Curr Opin Chem Biol 14:315–324.  https://doi.org/10.1016/j.cbpa.2010.03.020CrossRefGoogle Scholar
  184. Tseng H, Balaoing LR, Grigoryan B et al (2014) A three-dimensional co-culture model of the aortic valve using magnetic levitation. Acta Biomater 10:173–182.  https://doi.org/10.1016/j.actbio.2013.09.003CrossRefGoogle Scholar
  185. Tyson KL, Reynolds JL, Mcnair R, et al (2003) Osteo/chondrocytic transcription factors and their target genes exhibit distinct patterns of expression in human arterial calcification. pp 489–494.  https://doi.org/10.1161/01.ATV.0000059406.92165.31CrossRefGoogle Scholar
  186. Valk DC Van Der, Ven CFT Van Der, Blaser MC, et al (2018) Engineering a 3D-bioprinted model of human heart valve disease using nanoindentation-based biomechanics. pp 1–21.  https://doi.org/10.3390/nano8050296CrossRefGoogle Scholar
  187. van der Ven CF, Wu PJ, Tibbitt MW et al (2017) In vitro 3D model and miRNA drug delivery to target calcific aortic valve disease. Clin Sci 131:181–195.  https://doi.org/10.1042/cs20160378CrossRefGoogle Scholar
  188. van Haaften EE, Duijvelshoff R, Ippel BD, et al (2019) The degradation and performance of electrospun supramolecular vascular scaffolds examined upon in vitro enzymatic exposure. Acta Biomater.  https://doi.org/10.1016/j.actbio.2019.05.037CrossRefGoogle Scholar
  189. Verma S, Yanagawa B, Kalra S et al (2013) Knowledge, attitudes, and practice patterns in surgical management of bicuspid aortopathy: a survey of 100 cardiac surgeons. J Thorac Cardiovasc Surg 146:1033–1040.e4.  https://doi.org/10.1016/j.jtcvs.2013.06.037CrossRefGoogle Scholar
  190. Vesely I (1997) The role of elastin in aortic valve mechanics. J Biomech 31:115–123CrossRefGoogle Scholar
  191. Victoria Gomez-Stallons M, Wirrig-Schwendeman EE, Hassel KR et al (2016) BMP signaling is required for aortic valve calcification HHS public access. Arter Thromb Vasc Biol 36:1398–1405.  https://doi.org/10.1161/ATVBAHA.116.307526CrossRefGoogle Scholar
  192. Virchow R (1858) Cellular pathology. John Churchill, LondonGoogle Scholar
  193. Virchow R (1971) Cellular pathology as based upon physiological and pathological histology (English translation of second German edition). JB Lippincott, PhiladelphiaGoogle Scholar
  194. Weber B, Scherman J, Emmert MY et al (2011) Injectable living marrow stromal cell-based autologous tissue engineered heart valves: first experiences with a one-step intervention in primates. Eur Heart J 32:2830–2840.  https://doi.org/10.1093/eurheartj/ehr059CrossRefGoogle Scholar
  195. Weinberg CB, Bell E (1986) A blood vessel model constructed from collagen and cultured vascular cells. Science 231:397–400CrossRefGoogle Scholar
  196. Worthington P, Drake KM, Li Z et al (2017) Beta-hairpin hydrogels as scaffolds for high-throughput drug discovery in three-dimensional cell culture. Anal Biochem 535:25–34.  https://doi.org/10.1016/j.ab.2017.07.024CrossRefGoogle Scholar
  197. Wystrychowski W, McAllister TN, Zagalski K et al (2014) First human use of an allogeneic tissue-engineered vascular graft for hemodialysis access. J Vasc Surg 60:1353–1357.  https://doi.org/10.1016/j.jvs.2013.08.018CrossRefGoogle Scholar
  198. Xu Y, Hu Y, Liu C, et al (2018) A novel strategy for creating tissue-engineered biomimetic blood vessels using 3D bioprinting technology. Mater (Basel, Switzerland) 11.  https://doi.org/10.3390/ma11091581CrossRefGoogle Scholar
  199. Yutzey KE, Demer LL, Body SC et al (2014) Calcific aortic valve disease: a consensus summary from the alliance of investigators on calcific aortic valve disease. Arterioscler Thromb Vasc Biol 34:2387–2393.  https://doi.org/10.1161/ATVBAHA.114.302523CrossRefGoogle Scholar
  200. Zareian R, Tseng J-C, Fraser R, et al (2019) Effect of stent crimping on calcification of transcatheter aortic valves. Interact Cardiovasc Thorac Surg.  https://doi.org/10.1093/icvts/ivz024CrossRefGoogle Scholar
  201. Zhang Y-n, Xie B-d, Sun L et al (2016) Phenotypic switching of vascular smooth muscle cells in the “normal region” of aorta from atherosclerosis patients is regulated by miR-145. J Cell Mol Med 20:1049–1061.  https://doi.org/10.1111/jcmm.12825CrossRefGoogle Scholar
  202. Zheng KH, Tsimikas S, Pawade T et al (2019) Lipoprotein(a) and oxidized phospholipids promote valve calcification in patients with aortic stenosis. J Am Coll Cardiol 73:2150–2162.  https://doi.org/10.1016/j.jacc.2019.01.070CrossRefGoogle Scholar
  203. Zohar B, Blinder Y, Mooney DJ, Levenberg S (2018) Flow-induced vascular network formation and maturation in three-dimensional engineered tissue. ACS Biomater Sci Eng 4:1265–1271.  https://doi.org/10.1021/acsbiomaterials.7b00025CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Mark C. Blaser
    • 1
  • Samantha K. Atkins
    • 1
  • Elena Aikawa
    • 1
    • 2
    Email author
  1. 1.Center for Interdisciplinary Cardiovascular Sciences, Division of Cardiovascular Medicine, Department of MedicineBrigham and Women’s Hospital, Harvard Medical SchoolBostonUSA
  2. 2.Center for Excellence in Vascular Biology, Division of Cardiovascular Medicine, Department of MedicineBrigham and Women’s Hospital, Harvard Medical SchoolBostonUSA

Personalised recommendations