Engineering Three-Dimensional Vascularized Cardiac Tissues

Abstract

Heart disease is one of the largest burdens to human health worldwide and has very limited therapeutic options. Engineered three-dimensional (3D) vascularized cardiac tissues have shown promise in rescuing cardiac function in diseased hearts and may serve as a whole organ replacement in the future. One of the major obstacles in reconstructing these thick myocardial tissues to a clinically applicable scale is the integration of functional vascular networks capable of providing oxygen and nutrients throughout whole engineered constructs. Without perfusion of oxygen and nutrient flow throughout the entire engineered tissue not only is tissue viability compromised, but also overall tissue functionality is lost. There are many supporting technologies and approaches that have been developed to create vascular networks such as 3D bioprinting, co-culturing hydrogels, and incorporation of soluble angiogenic factors. In this state-of-the-art review, we discuss some of the most current engineered vascular cardiac tissues reported in the literature and future directions in the field.

Impact statement

The field of cardiac tissue engineering is rapidly evolving and is now closer than ever to having engineered tissue models capable of predicting preclinical responses to therapeutics, modeling diseases, and being used as a means of rescuing cardiac function following injuries to the native myocardium. However, a major obstacle of engineering thick cardiac tissue remains to be the integration of functional vasculature. In this review, we highlight seminal and recently published works that have influenced and pushed the field of cardiac tissue engineering toward achieving vascularized functional tissues.

Introduction

Heart failure is estimated to affect at least 26 million individuals globally.¹ Ischemic heart disease is the most common cause of heart failure and one of the main causes of death worldwide, claiming the lives of over 9 million people worldwide in 2016 alone.^2–4 Even though there are treatments currently administered for ischemic heart disease^2,5 and overall heart failure,^6–10 there is no disease-modifying therapy other than organ transplantation. However, in addition to problems such as surgical complications and organ rejection, heart transplantation is inherently limited by the number of total available functional hearts.^11,12 The field of cardiac tissue engineering has opened up the possibility of overcoming these problems by reconstructing and replacing original tissues or portions of them to restore their original function. The ultimate goal of cardiac tissue engineering is to resolve the desperate need for organ donors by engineering functional hearts at a clinically relevant scale for implantation. Immune-mediated rejection of new organs and foreign objects is one of many challenges that must be overcome by further developing biocompatible and patient-specific tissues.^13–15 Recent advances in stem cell biology and biomaterials are now expected to solve many of these associated challenges by providing patient-derived pluripotent stem cells and through further refinement of advanced biomaterials.^16–19

More recently, as methodologies of tissue engineering have advanced to recapitulate cardiac microtissue structures more elaborately, the restoration of vascular networks throughout engineered tissues has emerged as a major interest and challenge within the field.^20–22 This is because the incorporated cells need to be within 100–200 μm of functional vasculature to secure oxygen and nutrient supply and diffuse out waste products, which are prerequisites to maintain cellular viability and organ-specific functionality.^23,24 In particular, cardiac-specific vasculature forms synchronously with cardiac tissue and extensively interacts with it through a variety of paracrine, autocrine, and endocrine factors for organ development and mature functionality.^25–30 It is becoming increasingly clear that vasculature in cardiac tissue is of grand importance at both a transcriptional level and in cardiac function as a whole.^31–35 In this review, we discuss techniques being used to integrate functional cardiac tissues into in vivo models. In addition, we address recent advances in engineered vascularized cardiac tissues utilizing three-dimensional (3D) printing strategies. Finally, we discuss current limitations, challenges, and the future directions and perspectives in which the field of engineering cardiac tissues progresses toward.

Engineered Biomaterials to Recapitulate the Native Extracellular Matrix

One critical challenge in the field of tissue engineering is to engineer biomaterials that provide physiologically relevant microenvironments to incorporated cells.³⁶ Cells sensitively respond to their immediate surroundings, the microenvironment, such as spatial arrangement, mechanochemical cues, nutrient availability, and gas exchange.^37–46 To achieve organ-level functionality, the biomaterial must recapitulate physiological properties of the native extracellular matrix (ECM) so that the incorporated cells exhibit their original phenotypes and the resultant tissues function as observed in vivo. In addition, the scaffold biomaterials also need to exhibit mechanical stability to support cells in a relevant 3D spatial arrangement. Compared to two-dimensional (2D) tissue culture on plastic dishes, many engineered organ systems have successfully provided tissue-specific 3D microenvironments for the incorporated cells and obtained targeted functionalities.^22,47–52

Hydrogels have been most widely used as an ECM scaffold for engineered cardiac tissues. Hydrogels are classified into three categories: natural, synthetic, and hybrid. Naturally derived hydrogels generally comprised ECM derived-components (such as decellularized ECM [dECM]), polysaccharides (such as alginate), and peptide chains (such as collagen or spider silk).^50,53–58 These are optimal for recapitulating ECM-mediated biochemical signaling as seen in vivo and are susceptible to biodegradability, which may be desirable when engineering cardiac tissue implants to aid in integration into the in vivo tissue.^59,60 Alternatively, synthetic hydrogels are often biologically inert, but allow for fine-tuning to recapitulate key aspects of the targeted native microenvironment.

The core components of the synthetic hydrogels used in many recent reports commonly comprised polyethylene glycol (PEG) or its modified derivatives.⁶¹ Tunable chemical/photo-crosslinkable moieties can be integrated into hydrogel polymeric networks for the fine-tuning of substrate physiochemical properties.^57,62 Biochemical cues such as growth factors (i.e., vascular endothelial growth factor: VEGF⁶³), binding domains (i.e., fibronectin⁶⁴), proteins (i.e., laminin⁶³), or peptide chains (i.e., arginine-aspartic acid [RGD]⁶⁵) can be covalently bound or blended into a synthetic hydrogel to guide cell fate and cellular organization, and increase biocompatibility and tune biodegradability.⁶⁶ While synthetic hydrogels offer greater abilities to customize specific cell-substrate interactions, it is questionable how closely synthetic hydrogels mimic native ECM. Therefore, hybrid hydrogels, a blend of both synthetic and natural hydrogels, are of interest, in that, they can encompass the advantages of both natural and synthetic hydrogels. Hybrid hydrogels have the structural flexibility of synthetic hydrogels, while providing native ECM components.^67–74 These may be the most advanced biomimetic hydrogels to date, but are still far from truly recapitulating the highly complex in vivo native ECM.

Conventional Tissue Culture-Based Methodologies for the Development of Vascularized Cardiac Tissues

Many conventional tissue culture-based methods are being utilized to develop vascularized cardiac tissues. Vascularized cardiac tissues are useful not only for understanding disease pathology but also for meeting the clinical need for transplantation. In this section, we discuss several methods of developing vascularized cardiac tissue, focusing on methods that utilize 2D and 3D conventional culture systems that integrate both physical and biochemical cell-cell and cell-ECM interactions.

Engineering transplantable tissues through coculture methods

One interesting approach is the coculturing of tissue-specific progenitor cells with endothelial cells (ECs) to facilitate cell-cell interactions and mediate integration of host vasculatures in transplanted tissues.⁷⁰ The interactions between host cells and transplanted biomaterials have significant effects on integration of host vasculatures and angiogenesis.^75–79 This has led the field to focus on engineering tissues that have biomaterial and cell types, with physiological architecture, which allow further integration into host tissue upon transplantation (Fig. 1A).^80–85 Cardiac patches have been generated by culturing cells in low adhesion plates with varying cardiomyocyte (CM):human umbilical vein endothelial cells (HUVEC):mouse embryonic fibroblast ratios of 1:0:0, 1:1:0, and 1:1:0.5.⁸⁶ These cardiac patches were transplanted into rodent hearts and observed after 1 week. Of all three ratios, it was only found that one was viable (1:1:0.5). Interestingly, this particular ratio was reported to have formed viable CD31⁺ endothelial networks resembling in vivo morphological vascular networks (Fig. 2A). In addition, the 1:1:0.5 tissues were observed to integrate into the host myocardial vasculature with lumens containing leukocytes and red blood cells, which were validated through analysis of Ter-199, an erythrocyte lineage marker, expression (Fig. 2B).

FIG. 1.

Cardiac tissues can be cultured in numerous ways, which allow for their subsequent functionalization and transplantation. (A) Hydrogels with embedded cardiac-specific cell types and angiogenic growth factors can generate functional vascular cardiac patches that can be subsequently transplanted into diseased hearts. (B) An example of cardiac cell sheeting using thermosensitive gelatin for stacking several layers of cell sheets for mechanical stability and subsequent transplantation.⁹³

FIG. 2.

Cardiac patches demonstrate the ability to integrate with host vasculatures (A) 1-week post implantation in a rat MI model vasculature within the cardiac patch stains positive for human CD31⁺ endothelial networks.⁸⁶ (B) Vessel-like structures within the cardiac patch contain leukocytes (white arrows) demonstrating functional integration into host vasculature.⁸⁶ (C) H&E staining of the cardiac patch postimplantation reveals the presence of red blood cells within the patch (blue arrows).⁸⁷ (D) Integration of host vasculature identified by the presence of cells staining positive for isolectin B4 (IB4) within the cardiac patch.⁸⁷ (E) Implanted cardiac patch (hCMP) located to the right of the white dashed line was identified by visualization of engineered iPSC-CM expressing CD31 and cardiac troponin I (cTnT).⁸⁸ (F) Maturing iPSC-CM with distinct sarcomeric structures were visualized by staining for coexpression of GFP and α-actinin.⁸⁸ iPSC-CM, induced pluripotent stem cell-derived cardiomyocyte; MI, myocardial infarct.

Cardiac patches with multiple cell types have also been fabricated using fibrin gel scaffolds containing both induced pluripotent stem cell-derived CMs (iPSC-CM) and vascular pericytes in a myocardial infarct (MI) model.⁸⁷ MI was achieved by permanent ligation of the left anterior descending coronary artery LAD, and cardiac patches containing either iPSC-CM only or iPSC-CM and vascular pericytes were transplanted into the infarct area. It was observed that the iPSC-CM and vascular pericyte patches were able to recruit host vasculature and remain viable over a 4-week period (Fig. 2C, D). In comparison to the iPSC-CM-only cardiac patch, the iPSC-CM and vascular pericyte patch treatment group had an increase in ejection fraction at the 1-week mark and fractional shortening at both the 1- and 4-week time points, indicating rescue of cardiac function. The concept of a tricultured transplantable cardiac patch was explored in a study using human CM, EC, and embryonic fibroblasts in biodegradable porous scaffolds composed of PLLA/PLGA.⁸² Following 2 weeks of culture, it was reported that the engineered cardiac tissue was able to spontaneously contract, at which point, it was transplanted into the anterior wall of the left ventricle of rats. Two weeks following transplantation, there was significantly more host vascularization in the tricultured cardiac tissues compared to a control CM-only tissue quantified by average blood vessels present. Injection of florescent microspheres into coronary circulation revealed perfusion into both rat heart and the transplanted scaffold, demonstrating integration of host vasculature into transplanted tissues. Most interestingly, fibroblasts within the transplanted cardiac tissue were identified within the vascular wall and stained positively for αSMA, indicating differentiation into vascular mural cells, such as smooth muscle cells, exhibiting a more mature vessel morphology.

In a recent study, a tricultured cardiac patch was developed using human iPSC-CM, iPSC-EC, and iPSC-smooth muscle cells (iPSC-SMC) within a fibrin-cased scaffold.⁸⁸ Following 1 day of implantation into the scaffold, it was observed that iPSC-CM were able to synchronously depolarize within the scaffold. Functionality of the cardiac patch in vitro was assessed 7 days following seeding in the fibrin scaffold, and it was observed that intracellular coupling of the iPSC-CM was similar to that of the left ventricle of rabbits and could reliably be placed in a broad range of intervals (450–1000 ms). To validate the efficacy of the cardiac patch in regenerative medicine applications, the authors moved toward engrafting the cardiac patch into a porcine model of MI. In this model, the LAD was occluded for 60 minutes, reperfused, and the cardiac patch was subsequently engrafted into the region of infarct. Following 4 weeks of observation, the recipient group had statistically significantly improved cardiac function compared to controls, along with reduced wall stress and infarct size. In addition, endogenous factors secreted by the patch improved cell proliferation and neovascularization of the embedded iPSC-EC (Fig. 2E, F). Proteomic analysis of the cardiac tissue following 4 weeks with the tricultured cardiac patch revealed an interesting stabilization of sarcomeric protein phosphorylation, suggesting that the cardiac patch had either prevented or reversed phosphorylation of key sarcomeric regulatory proteins that occur following MI. This unexpected, but interesting finding was hypothesized to have had a role in the improved cardiac function in the cardiac patch recipient group. As evidence of the works discussed in this section, cardiac patches containing multiple cell types found in native myocardium show improved functionality and rescue cardiac function in MI models. In addition, the inclusion of multiple cell types has improved recruitment of host vasculatures, which may aid in rescuing cardiac function along with ensuring long-term viability of the implanted grafts. iPSC-derived tissues seem to have the greatest effect on both vascularization of the region of infarct and overall cardiac function, which may be attributed to cell signaling pathways (e.g., angiogenesis, vasculogenesis, and cell cycle regulation)^89,90 in relatively immature tissues.

Cardiac cellular sheets

Cell sheeting capitalizes on cell-cell interactions in monolayer culture systems to obtain vascularized cardiac tissues through stacking monolayers (Fig. 1B).^91–95 In this system, CM monolayers are cultured on a temperature-responsive polymer, poly(N-isopropylacrylamide). Cells adhere to this polymer at physiological temperature, but detach as a sheet at lower temperatures without the need for enzymatic processes that disrupt cell-cell and cell-surface adhesion.⁹⁶ This method reliably creates cardiac cell sheets that contain elongated sarcomeres and established gap junctions, and recruit vasculatures when implanted in vivo, while maintaining long-term viability and cardiac-specific functionality. Engineered layered cardiac cell sheets showed promise in morphological integration into host myocardium and improved cardiac function and recovery in in vivo MI models (Fig. 3A–C).^92,97,98 While these studies proved the potential for cardiac sheets in regenerative medicine, they still had yet to mimic the microenvironment of the native myocardium or integrate neovasculature. In an effort to address this shortcoming, cardiac cell sheets were cocultured in varying relative ratios with EC (Fig. 3D, E).⁹⁹ When analyzed in vitro, cardiac sheets co-cultured with EC had significantly elevated secretion of VEGF, basic fibroblast growth factor, and hepatocyte growth factor in comparison to cell sheets composed of CM alone. When the cocultured cardiac sheets were transplanted into an MI rat model in vivo, there was a significant increase in fractional shortening and a decrease of cardiac fibrosis at the region of infarct. In addition, embedded EC were able to connect with host capillaries at the periphery of the cardiac cell sheet. In a more recent study, iPSC-CM, vascular EC, and vascular mural cells were cocultured within cell sheets.¹⁰⁰ In this study, cells were cultured on gelatin-coated thermosensitive culture dishes as opposed to poly(N-isopropylacrylamide) where cell sheets could release at physiological temperature. These sheets were transplanted into porcine MI models and it was observed that there was integration of host vasculature and formation of capillary beds within the grafted cell sheet. The cell sheet recipient group exhibited significantly larger ejection fractions and fractional shortening in comparison to the nontreated control group, indicating rescue of cardiac function. In addition, there was significantly less fibrosis in the group treated with cell sheets compared to the control. These data indicate that the cardiac cellular sheets are useful for improvement of cardiac function, vascularization, and potential long-term viability.

FIG. 3.

Cell sheet engineering has been explored as a potential option for generating transplantable vascular cardiac tissues. (A) Cardiac cell sheets have been generated, layered, and cultured within vascular bioreactors to influence prevascularization before transplantation.⁹² (B, C) H&E and Azan staining show evidence of cardiac sheets containing perfusable vasculature capable (white arrows) of supporting the engineered tissue.⁹² (D) Transplanted cardiac cell sheet containing GFP expressing ECs stained with isolectin B4 demonstrates fused vasculature from cardiac cell sheet to host MI model.⁹⁹ (E) Cardiac patch containing GFP expressing ECs stained for troponin T (red) shows transplanted cardiac cell sheet vascularizing host myocardium in rat MI model.⁹⁹

Angiogenic Factors and Their Incorporation into Cardiac Tissue Models

During development, tissues become vascularized de novo through angiogenesis, where EC remodel the ECM and form new capillary sprouts, which gives rise to a new functional circulatory system from existing blood vessels.^101,102 As discussed in Section 3, this process can be influenced by cell-cell interactions, but is also subject to signaling through angiogenic growth factors (Table 1).^85,102–105 EC will respond to gradients of one or a combination of growth factors and directionally sprout into the ECM, forming new capillaries toward the source of growth factors.⁴⁹

Table 1.

Various Growth Factors Involved in Angiogenesis and Their Method of Action

Growth factor	Activity
VEGF	A signaling protein involved in vasculogenesis and angiogenesis through binding of VEGF receptors.102,103
PDGF	A growth factor that regulates blood vessel formation, blood vessel growth, vascular smooth muscle cell proliferation, and chemotaxis through receptor tyrosine kinase signals.103,149
SDF-1	A strongly chemotactic growth factor that directs formation of large blood vessels in fetal development, as well as in adult angiogenesis through recruitment of endothelial progenitor cells from bone marrow.102
IGF-1	A prosurvival growth factor that stabilizes nascent vessel formation, enhancing angiogenesis.150
HGF	A factor that regulates cell growth, motility, and morphogenesis. It acts as a mitogen for vascular EC, but not vascular smooth muscle cells, furthering angiogenesis.102
bFGF	The bFGF stimulates EC to secrete proteases to degrade the vessel basement membrane, allowing cells to invade the surrounding matrix, proliferate, and form neovessels.102,151
MCP-1	An angiogenic chemokine that activates VEGF downstream to further angiogenesis.152
PMA	The PMA increases production of collagenase and allows EC to invade through the basement membrane and form further vessels.153
Substance P	The Substance P is mitogenic for endothelial and vascular smooth muscle cells, promoting angiogenesis.105,154
S1P	A blood-borne lipid mediator that acts on G protein-coupled receptors, such as S1P receptors. It stimulates EC proliferation and migration, and formation of lumen-containing vessels.155

bFGF, basic fibroblast growth factor; EC, endothelial cell; HGF, hepatocyte growth factor; IGF-1, insulin-like growth factor 1; MCP-1, monocyte chemoattractant protein-1; PDGF, platelet-derived growth factor; PMA, phorbol 12-myristate 13-acetate; S1P, sphingosine-1-phophate; SDF-1, stromal cell-derived factor-1; VEGF, vascular endothelial growth factor.

In a seminal work, an approach was taken to engineer vascular cardiac tissue utilizing an alginate scaffold seeded with a mixture comprising Matrigel, mouse left ventricular CM, and growth factors stromal cell-derived factor-1, insulin-like growth factor-1 (IGF-1), and VEGF.¹⁰⁶ The patch was cultured for 48 h and then implanted into rat omentum for 7 days. Seven days following implantation, the cardiac patch was removed, and it was observed that there was extensive vascularization of the patch in comparison to control patches lacking growth factors. The prevascularized cardiac patches were then evaluated in an in vivo rat MI model and the degree of dyskinesis (irregular movement of the myocardium outward during systole¹⁰⁷) was reduced in comparison to nontreated controls. This validated the functionality of the cardiac patch because increasing degrees of dyskinesis is a predictor of mortality following MI in patients.¹⁰⁸

In a more direct approach, endogenous cell signaling was controlled by modulating VEGF expression in transduced myoblasts engrafted directly into an area of MI in an in vivo murine model.¹⁰⁹ Skeletal myoblasts transfected to express VEGF were cultured with rat CM within porous poly(glycerol sebacate) scaffolds. The patches were then engrafted into areas of MI and examined 4 weeks after implantation. It was reported that the VEGF expressing cardiac patches showed significantly more vasculatures in the area of infarct in comparison to non-VEGF expressing controls and were also able to induce angiogenesis into the underlying myocardium. This concept has been expanded on in more recent studies that have engineered biocompatible materials capable of controlled release of angiogenic factors over extended periods of time to recruit vascular ECs and stem cells.^110,111 While angiogenic factors have shown great promise in being therapeutic for MI and ischemic heart disease, their effects on single cell and systemic pathology are less understood. In an effort to engineer a system capable of safely releasing growth factors over extended periods of time to induce neovascularization in a localized region, VEGF and platelet-derived growth factor were encapsulated in poly(D,L-lactide-co-glycolide) acid and porous silica nanoparticles that were subsequently integrated into an electrospun gelatin scaffold.¹¹¹ When the scaffold with embedded growth factors was implanted subcutaneously in a murine model for 21 days, it was observed that there was an increase in αSMA and CD31⁺ cells in the implant area, indicating enhanced neovascularization and vessel maturation. In addition, in the growth factor-embedded scaffold, qPCR revealed an increase in proangiogenic factors, further providing evidence that the scaffold was capable of influencing vascularization in a localized region. In a similar study, electrospun scaffolds containing embedded IGF-1 and substance P were fabricated as cardiac patches to influence neovascularization in an MI murine model.¹¹⁰ In the electrospun cardiac patches, it was observed that patches containing both IGF-1 and substance P implanted in infarcted myocardium contained more CD31⁺ vessels and isolectin B4-positive capillaries in comparison to other controls containing only one growth factor or none at all. These findings may suggest that this combination of growth factors may also act a synergistic pair when attempting to revascularize and heal damaged myocardium.

3D Bioprinting for Reconstructing Cardiac Tissue Constructs

In recent years, 3D printing technology has become increasingly advanced and reliable as an option for manufacturing complex tissues. 3D printing allows one to precisely control deposition of materials layer by layer. It provides unique advantages when reconstructing hierarchical human tissue structures composed of various types of cells. Thus, 3D printing has been used as a suitable method to engineer thick tissues with complex physiologically relevant architectures and/or organ-specific functionalities.^112–116 Many approaches have been proposed to mimic spatial patterning of organs with tissue-specific cell types, while simultaneously mimicking biological and mechanical properties of the tissue.¹¹⁷ However, the integration of functional vasculatures remains to be one of the largest challenges when engineering thick tissues using 3D printing methods.⁵² To this end, we discuss the most recent state-of-the-art advances made in 3D printing of vascularized cardiac tissues for the purpose of disease modeling and regenerative medicine.

Spatial patterning of cardiac and vascular tissues

Human organs and tissues have complex architectures in 3D that are necessary to fulfill their function. These architectures must be recapitulated in engineered tissues. For example, the use of multiple bioinks is a common strategy to construct the hierarchal structure of vascularized cardiac tissues. 3D bioprinters with incorporated microfluidic channels are capable of printing multiple bioinks both homogenously and heterogeneously with high fidelity (Fig. 4A).^118,119 For example, alginate bioinks can be crosslinked through electrostatic interactions¹²⁰ with Ca²⁺ so that coextrusion of the Ca²⁺ solution with bioink in a microfluidic system instantaneously forms hydrogel networks upon exiting the nozzle, ensuring structural stability during the printing process.¹²¹ With multiple reservoirs incorporated into this microfluidic system, bioinks containing cardiac-specific cell types can be extruded either individually or simultaneously to achieve unique, spatially patterned hydrogels. The mechanical properties of the printed construct can be precisely tuned to the target tissue by incorporating ultra violet crosslinking sites, such as gelatin methacroyl.^122–124 Fine tuning of these properties for cardiac tissues and incorporating varying components (i.e., gold nanoparticles) can improve cardiac cell viability and electrical propagation, and promote overall functionality.¹²⁵ In a recent study, iPSC-CM were printed with EC in an alginate bioink infused with polyethylene glycol monoacrylate-fibrinogen (PF).¹²⁶ The mechanical properties of the printed construct were fine-tuned to the target tissue by modulating the UV crosslinking density of the infused PF. In addition, the alginate was partially removed using a chelating agent (ethylenediaminetetraacetic acid) to ensure sufficient diffusion of the culture media and improve overall cell viability.¹²⁷ They observed that the side-by-side coextrusion of the same numbers of iPSC-CM and HUVEC exhibited the highest degree of CM alignment and viability for both cell types in vitro. Also, coextrusion patterning exhibited maximum recruitment of host vasculature into the cardiac patch in murine MI models.

FIG. 4.

3D printing provides a platform for constructing detailed architectures of cardiac tissues (A) Bioinks composed of various cell types can be spatially patterned to give rise to functional cardiac tissues that can be subsequently evaluated both in vitro and in vivo. (B) Sacrificial inks are 3D printed into cardiac tissues and subsequently removed to create embedded luminal structures within engineered cardiac tissues that support thick tissue perfusion. (C) Support baths used 3D print soft materials with high fidelity that exhibit complex hierarchal structures of in vivo heart anatomy. 3D, three dimensional.

3D printed sacrificial inks

In contrast to spatially patterning cardiac-specific cell types, some have looked toward using sacrificial inks, materials that are removable after the printing process, to directly imbed luminal structures into cardiac tissues (Fig. 4B). Sacrificial inks are printed in 3D at a desirable geometry together with tissue-specific, cell-laden biomaterials that can be subsequently crosslinked to support structural integrity. Following the removal of the sacrificial inks, open lumens remain and serve as physical conduits for EC to be seeded to form blood vessels within a thick tissue, enabling diffusive mass transport of nutrients, oxygen, and metabolic waste.^128–130 Carbohydrate glass is one of the common sacrificial inks used to form perfusable vascular networks in biomaterials.^131–133 Carbohydrate glass can be printed at an elevated temperature (110°C), but rapidly solidifies at room temperature leaving behind a mechanically stable structure.^131,134 Once established, the printed lattice can be coated with poly(D-lactide-co-glycolide) (PDLGA) to facilitate the clearance of carbohydrate glass through the printed network and prevent osmotic damage to cells within the bulk construct due to the lack of media transport.¹³¹ Carbohydrate glass constructs can be cured within a wide variety of hydrogels, including agarose, alginate, PEG, fibrin, Matrigel, and collagen, among others.¹³¹ This method was utilized to engineer a vascular cardiac patch through printing of a fibrin hydrogel with various luminal structures. This was transplanted into a region of infarct in a rat MI model, where it successfully recruited host vasculatures and induced a substantial increase in host capillary beds within the region of infarct. This served to increase the ejection fraction and cardiac output along with a decrease in left ventricular internal dimension.⁶²

Thermosensitive bioinks are also commonly used to create sacrificial lumen structures in engineered tissues.^{129,135–137} These bioinks undergo a thermally reversible gelation process as the temperature changes and can be dissolved in an aqueous solvent or evacuated from the tissue. This technology was utilized to print the major coronary arteries of the left ventricle in a cardiac patch in 3D (Fig. 5A–G).¹³⁸ To more closely mimic vascular patterns in the heart, they utilized computerized tomography (CT) to scan the 3D distribution of vasculatures in vivo, generate computer-aided designs (CAD), and print 3D tissue constructs accordingly in a layer-by-layer process. In a bottom-up approach, they printed EC-laden thermosensitive gelatin in parallel with decellularized omentum containing iPSC-CM to fabricate a vascularized cardiac patch. At physiological temperature, the Young's modulus of gelatin markedly drops, allowing it to be washed away, leaving open lumens embedded into the tissue.¹³⁹ Following 7 days in culture, lumens within the cardiac patch were lined with confluent monolayers of EC and the neighboring cardiac tissue was capable of depolarization. Therapeutic potential of the cardiac patches was also assessed in vivo when transplanted into rat omentum for a 7-day period, where it was observed that the CM were elongated and aligned, and had an exceptional degree of striation, demonstrating further maturation of the cardiac patch following implantation.

FIG. 5.

3D printing enables the generation of functional vascular cardiac tissues and complex cardiovascular structures. (A–C) Overview of bottom-up 3D printing process for the generation of a cardiac patch.¹³⁸ (D) Image of excised omentum following 7 days of in vivo transplantation with white dashed lines highlighting the border of the cardiac patch.¹³⁸ (E–G) Imaging of excised 3D printed cardiac patch following 7 days of in vivo transplantation for sarcomeric actinin (red) exhibiting a matured cardiac morphology at increasing magnification.¹³⁸ (H, I) Cardiac organoid organ building block staining positive for cardiac specific markers cardiac troponin T (green) and α-actinin (red).¹⁴⁰ (J, K) Viable SWIFT 3D printed perusable cardiac tissues imaged from top-down staining positive for cardiac troponin T and α-actinin following 24 h of perfusion.¹⁴⁰ (L) SWIFT printed CAD rendering of left anterior descending artery into left ventricle-shaped mold.¹⁴⁰ CAD, computer-aided designs; SWIFT, sacrificial writing into functional tissue.

In a recent work, Skylar-Scott and Uzel et al. demonstrated a technique for 3D printing vasculature into iPSC-CM tissues using a thermally sacrificial gelatin ink, termed “sacrificial writing into functional tissue” (SWIFT) (Fig. 5H–L).¹⁴⁰ In their work, gelatin sacrificial inks were directly printed into a viscoelastic slurry of iPSC-CM organoids and engineered ECM. The gelatin sacrificial ink was printed at 2°C so that the sacrificial ink could maintain its mechanical stiffness at an order of magnitude higher than the ECM/CM organoid slurry. Once brought to physiological temperature, the thick ECM/CM organoid slurry crosslinks, while the gelatin sacrificial ink melts leaving behind large perfusable lumens. The vascularized cardiac tissue was able to beat synchronously over an 8-day period and showed increased overall cardiac displacement over time, indicating cardiac tissue maturation. In addition, the authors were able to demonstrate that SWIFT was capable of utilizing CT imaging-inspired CAD to 3D print the hierarchical architecture of the LAD within the ECM/CM slurry.

3D printing of cardiac structures in support materials

While sacrificial inks are used to provide a material that can be subsequently removed, allowing for the development of lumens in the printed tissue, some tissues additionally need more structural support during printing. This is due to the poor mechanical stability of many printable hydrogels and biomaterials of physiological stiffness. Sacrificial constructs have been printed within support materials, which allows for higher print resolution and fidelity of soft materials (Fig. 4C).^141–144 In one approach, sacrificial inks were printed into modified photocurable support gels to engineer omnidirectionally pervasive vascular networks.¹²⁹ Sacrificial inks printed into support gels composed of Pluronic F127 diacrylate generally maintain the shape and position at the point of extrusion. Because the support material easily deforms as the extrusion needle moves through it, creating a void in the gel, low viscosity Pluronic F127 diacrylate gel was poured in to fill voids created during the printing procedure. The resulting printed constructs were crosslinked with UV light and the sacrificial ink was removed, allowing for embedded vasculature within the tissue construct.

Building on this concept in a seminal work, Hinton et al. described a method allowing 3D printing of soft materials within a support bath termed freeform reversible embedding of suspended hydrogels (FRESH).¹⁴¹ This methodology relies on the mechanical stability of gelatin at specific temperatures and the pH range at which collagen crosslinks. In this technique, a gelatin support bath acts as a Bingham plastic at room temperature, allowing a moving nozzle to flow through it, while depositing material without disrupting the structural integrity of the support bath. The collagen is kept in a highly acidic environment preventing crosslinking and allowing for extrusion from the print nozzle. Once extruded from the nozzle, the collagen is deposited into the gelatin bath kept at a neutral pH of 7.4 where the collagen rapidly crosslinks, leaving behind a solid structure within the gelatin. Subsequently, the construct and bath are warmed to 37°C, causing the gelatin to melt, while the printed collagen maintains structural integrity and is released by the support bath. In this study, the gelatin support bath incorporated irregularly shaped microparticles, which limited print resolution to ∼65 μm. To overcome this obstacle in a more recent work, the group developed FRESH v2.0, which incorporates gelatin microparticles of ∼25 μm diameter, improving the print resolution.¹⁴⁵ The resolution of the prints was showcased by printing CAD-rendered CT scans of high-resolution architectures of the heart, including tri-leaflet heart valves, LAD branching, and a full-sized neonatal-scale human heart containing features such as atrial and ventricular chambers, trabeculae, and pulmonary and aortic valves (Fig. 6A–E). In addition, the authors also reported using FRESH v2.0 to print a scaffold of a ventricle that was populated with cardiac cells that exhibited directional wave propagation during spontaneous contraction, which maintained viability for up to 28 days. In a more recently published article, the group reported using FRESH v2.0 support baths to print a full-sized human heart composed of alginate based on MRI CAD renderings to train for and plan out cardiothoracic surgeries (Fig. 6F).¹⁴⁶ Other groups have also expanded on freeform reversible printing to afford engineered vascular cardiac tissues capable of both perfusion and depolarization.^144,147

FIG. 6.

3D printing into support material (A, B) CAD-generated renderings of cardiac structures used for FRESH v2.0 3D printing.¹⁴⁵ (C, D) FRESH v2.0 printed neonatal scale human heart exhibiting complex cardiac morphology and adult size tri-leaflet heart valve, respectively.¹⁴⁵ (E) FRESH v2.0 printed construct based on MRI-derived CAD rendering of the left anterior descending artery with computationally generated microvasculature capable of perfusion.¹⁴⁵ (F) FRESH v2.0 printed full-size adult human heart.¹⁴⁶ (G, H) Heart containing independent chambers, vasculature, and imbedded CMs and ECs printed into alginate microparticle support bath.¹³⁸ FRESH, freeform reversible embedding of suspended hydrogels.

Support baths comprising alginate microparticles have also been used in lieu of gelatin.¹³⁸ This can be advantageous when printing hydrogels that crosslink at 37°C due to the fact that alginate will not melt at this temperature. The alginate support bath can be brought to 37°C, allowing for the printed hydrogel to crosslink in place and maintain shape within the support bath. Once printed, the alginate support bath can be dissolved through an enzymatic process rather than a thermal process. In a recent work, this method was used to print a model of the heart containing CM and EC.¹³⁸ CM were embedded into a dECM hydrogel, which were printed as cardiac walls with distinct chambers on both the left and right side of the heart (Fig. 6G, H). In addition, a major blood vessel was printed using EC embedded into a dECM hydrogel that wrapped around the periphery of the cardiac wall. Once printed, the bath was brought to 37°C allowing for crosslinking of the dECM hydrogel and subsequent enzymatic removal. It was reported that 1 day after printing, there was homogenous distribution of CM within the printed construct, but no further structural or functional findings were reported.

Conclusion and Perspective

In the rapidly evolving field of engineered cardiac tissues, there have been remarkable advances and discoveries made in both basic and applied sciences. Biomaterials are continually investigated, but biocompatibility of the materials is often compromised to provide improved structural integrity and vice versa. In addition, the ability to provide the correct physiochemical cues for all cell types native to the myocardium has yet to be fully addressed. While these challenges still need to be addressed, the greatest challenges in the field are finding ways to not only introduce vasculature but also to engineer pervasive capillary networks within functional cardiac tissues. Cardiac patches and cell sheets show potential in the ability to recruit host vasculature when transplanted in vivo, but their full integration with the host, abilities to synchronously contract with host myocardium, and functional restoration following MI are not yet suitable as clinical therapeutics. 3D printing shows great promise in the ability to pattern hierarchal architectures of the heart and their vasculature, but is ultimately hindered by the material properties of bioinks and the resolution of the print when printing connective capillary beds. Another limitation in the field is that the majority of engineered vascularized cardiac tissues use EC types not specific to cardiac tissue. The identification of signaling pathways that regulate cardiac-specific endothelial differentiation and expansion in culture would significantly contribute to advancing cell-based therapies for cardiac regeneration. It should also be noted that in addition to blood vessels, lymphatic vessels in the heart play a critical role in cardiac health and disease. The importance of lymphatic vasculature in cardiac development and repair is reviewed in detail by Klaourakis et al.¹⁴⁸ However, to our best knowledge, there have been no attempts to incorporate tissue-engineered lymphatic vessels into engineered cardiac tissues. Since lymphatic vessels help wound healing in the heart, preventing fibrosis, more complex models containing tissue-engineered lymphatic vessels in the vascularized cardiac tissues would be a strong improvement to existing constructs. As we further explore materials, exploit biological pathways, and refine methodologies, we can more closely study human cardiac disease pathology and eventually engineer functional, vascularized cardiac tissues as therapeutic implants in clinic.

Authors' Contributions

M.A.C.W. wrote main body of text and generated figures. D.M. wrote portions of text, edited, generated figures, and generated tables. E.L. wrote portions of text and provided editing. W.L. wrote portions of text and provided editing. D.-H.K. devised the main conceptual ideas and proof outline, oversaw writing, and provided editing.

Disclosure Statement

D.-H.K. is a scientific founder and equity holder of Curi Bio. The other authors report no conflicts.

Funding Information

This work was supported by grants from the National Institutes of Health for R01HL135143, R01HL146436, UG3TR003271, UH3TR003519 (to D.-H.K), R25GM109441 (to M.A.C.W.), and K25CA201545 (to W.L.). This work was also supported by the Cornell Nancy and Peter Meinig Family Investigator funds (to E.L.).