The Therapeutic Effect of Cell Transplantation Versus Noncellular Biomaterial Implantation on Cardiac Structure and Function Following Myocardial Infarction

Abstract

Although numerous studies demonstrated that localized delivery of either cells or biomaterials improved postinfarction cardiac function, the underlying mechanisms for this effect remain unclear. We performed a comparison of the effects of fetal, neonatal, and human embryonic stem cell-derived cardiac cell as well as mesenchymal stem cell transplantation versus biomaterial (collagen/extracellular matrix) implantation therapy in rat myocardial infarction model in our laboratory, specifically comparing their effects on infarct wall thickness, neovascularization, infarct wall motion, and left ventricular ejection fraction (LVEF). Both cell and biomaterial treatment had similar beneficial effects on cardiac structure (increasing infarct wall thickness and preventing infarct expansion) and function (preventing paradoxical LV systolic bulging and improving LVEF). In this review, we also discussed the underlying mechanisms of cell and biomaterial therapies, their advantages and disadvantages, and future research directions in the field of regenerative cardiology.

The adult mammalian heart lacks fully effective intrinsic regenerative capability to replace the lost cardiac cells following a myocardial infarction. The injured ventricle will undergo a progressive physiological and anatomical transformation, termed “ventricular remodeling,” that starts very soon after a heart attack. Necrotic myocytes induce infiltration by inflammatory cells, followed by infarct wall thinning, infarction expansion and left ventricular (LV) cavity dilation, eccentric hypertrophy, and ultimately heart failure (for review, read ¹ ).

Currently available long-term pharmacologic treatment with angiotensin-converting enzyme inhibition and β adrenergic blockade have been demonstrated to impede ventricular remodeling and improve the quality of patient’s life but the therapeutic effects are limited. Development of new strategies to improve the remodeling process and prevent progression to heart failure after myocardial infarction is critical. In recent years, numerous studies demonstrated that localized delivery of either cells or biomaterials improved postinfarction cardiac function; however, the underlying mechanisms for this effect remain unclear. In this article, we performed a comparison of the effect of fetal/neonatal cardiac cell transplantation versus biomaterial (collagen/extracellular matrix) implantation therapy on cardiac structure and function in a rat myocardial infarction model and discuss future directions in this field of regenerative cardiology.

Cell Transplantation to Repair Damaged Myocardium

Fetal/Neonatal Ventricular Cardiomyocytes Transplantation

The initial aim of cell therapy is to replace the damaged myocardium with transplanted functional cardiomyocytes. The hypothesis is that the engrafted cells will form morphological communication through functional gap junctions with the surrounding endogenous myocardial cells and will be electrically coupled with these host cells and contract in synchrony with the surrounding myocardium to improve regional and global myocardial function. However, there may be other mechanisms whereby cell therapy improves cardiac function, including passive thickening of the infarct wall with a reduction in dyskinesis and/or paracrine effects. Our research group has directly injected isolated fetal/neonatal ventricular cardiomyocytes, mesenchymal stem cells (MSCs), human embryonic stem cell-derived cardiomyocytes (hESC-CMs), and combinations of cell suspensions into the injured heart in a rat myocardial infarction model. Müller-Ehmsen et al ² isolated cardiomyocytes from male neonatal Fischer rats (1-2 days old) and directly injected cardiomyocytes (3-5 millions/50 μL) or medium (50 μL) into the infarct area of 1-week-old myocardial infarctions in female Fischer rats. Six months after cell transplantation, implanted cardiomyocytes were still present by histology and quantitative TaqMan polymerase chain reaction. Histology showed that the infarcted LV wall thickness was greater in the cell transplantation versus control group (0.91 ± 0.1 vs 0.62 ± 0.04 mm, P < .02), whereas infarct sizes (scar circumference; 31.5 ± 1.7% vs 34.2 ± 2.8%) were similar in both groups (n = 10 transplanted and n = 9 control animals). By biplane angiography, LV ejection fraction (LVEF) was greater in the cell transplantation versus control group (36 ± 3% vs 25 ± 2%, P < .01). There was significantly less infarct zone dyskinesis (30 ± 8% vs 55 ± 7%, P = .035, lateral projection) in treated animals versus control animals. The LV volume was similar in the transplanted (0.42 ± 0.02 mL) and the control rats (0.45 ± 0.03 mL, both n = 9).

Yao et al ³ isolated fetal cardiomyocytes from Fischer 344 rats (16-18 days gestation) and injected these cardiomyocytes (4 millions/70 μL) or medium (70 μL) into the infarct area of 1-week-old myocardial infarction in adult female Fischer rats. Ten months later, hematoxylin and eosin staining showed an obvious graft within the infarct scar, and these grafts appeared to “bulk-up” the wall of the infarct. These grafts demonstrated a cardiac muscle phenotype. Histologic analysis showed infarct wall thickness was greater in cell-treated at 0.69 ± 0.05 mm (n = 11) versus medium-treated hearts at 0.33 ± 0.01 mm (n = 19; P = .0001). The infarct size (35 ± 2% vs 38 ± 1%) was similar in both the groups. The LVEF assessed by LV angiography was 40 ± 2% in cell-treated (n = 16) versus 33 ± 2% in medium-treated hearts (n = 24; P < .03). Postmortem LV volume was 0.41 ± 0.04 mL in cell-treated versus 0.51 ± 0.03 mL in medium-treated hearts (P < .04). Expansion index was 1.17 ± 0.1 in cell-treated versus 2.41 ± 0.1 in medium-treated hearts (P < .05).

Reffelmann et al ⁴ isolated neonatal cardiomyocytes from hearts of 2-day-old neonates (either sex) of Fischer 344 rats and injected cardiomyocytes (4 million/50-70 μL, n = 18) or medium (50-70 μL, n = 17) into the infarct area in adult female Fischer rats with 1-week-old myocardial infarctions. At 4 weeks after cell transplantation, cardiomyocyte injection thickened the infarct wall (0.93 ± 0.07, n = 18) in comparison to medium-injected hearts (0.75 ± 0.04 mm, n = 17, P < .020). The infarct sizes (36 ± 2.5% vs 38.2 ± 2.2%, P < .26) were similar in both the groups. Clusters of engrafted cells within the scar demonstrated a high capillary density (1217 ± 114 perfused capillaries/mm² shown by blue dye); however, in the collagenous scar tissue itself lacking grafts, capillary density in the cell group (156 ± 62/mm²) did not significantly differ from the medium group (125 ± 10/mm²), suggesting that neoangiogenesis was confined to regions of successful engraftment. The LVEF assessed by LV angiography was 42.6 ± 2% in cell-treated versus 43.2 ± 1.7% in medium-treated hearts (P < .416). It may be that 4 weeks was too short period of time to observe improvement in LVEF in this study. Müller-Ehmsen et al ² showed that at 6 months, cells improved LVEF in the same model. The cell transplantation group was characterized by smaller diastolic and systolic LV volumes, as assessed by intravenous ventriculography, and lower infarct expansion indices compared to the control group (0.64 ± 0.07 in cell therapy vs 0.83 ± 0.06 in controls, P < .023).

Mesenchymal Stem Cells Transplantation

Our research group also ⁵ injected MSCs or saline into the scar of a 1-week-old myocardial infarction in Fischer rats. The transplanted MSCs survived in infarcted myocardium as long as 6 months and expressed muscle-specific markers α-actinin, myosin heavy chain, phospholamban, and tropomyosin, as well as the endothelial marker von Willebrand factor, but did not fully evolve into an adult cardiac phenotype and did not form large grafts in the infarct wall. The scar thickness, infarct expansion index, and postmortem LV volume were comparable between the 2 groups at 4 weeks and 6 months. At 4 weeks, LVEF was significantly greater in MSC-treated animals (43.8 ± 1.0%) than in the saline group (38.8 ± 1.1%, P = .0027). The MSC treatment reduced the dyskinesis and akinesis of the infarcted wall (24.7 ± 1.1%) compared to the control group (29.9 ± 2.4%, P = .06). However, LVEF and regional infarcted wall motion were similar between the 2 groups at 6 months. The underlying mechanisms of cardiac functional improvement by MSC therapy may be due to a possible early paracrine effect, because this benefit was observed at 4 weeks but was lost at 6 months and was not associated with large visible grafts of new muscle. Rather, transplanted cells were seen in small scattered clusters. In addition, the benefit at 4 weeks occurred without thickening the wall of the infarct. Kearns-Jonker et al ⁶ injected hESC-CMs alone, hESC-CMs plus MSCs, hESC-CMs plus MSCs overexpressing heme oxygenase 1 (HO-1), or saline directly into the 1-week-old infarcted wall in athymic nude rats. Four weeks later, histology showed that cell treatment increased the thickness of infarcted LV wall and significantly decreased the LV expansion index compared to the control group. Left ventriculogram demonstrated that LVEF was significantly improved by 6% to 7%, and LV akinesis and dyskinesis were prevented after stem cell treatment in all cell-treated groups compared to controls. Furthermore, coadministration of HO-1 MSCs plus hESC-CMs increased expression of prosurvival and angiogenesis-promoting genes in transplanted donor cells, transcripts of cardiac and endothelial cell markers in host heart cells, and transcripts for cell cycle regulatory genes in the host heart cells. However, cardiac function and morphology were not further improved with combination of cell therapy compared to hESC therapy alone.

Clinical Trials of Cell Transplantation Therapy

In summary, most studies from our laboratory in which cells are implanted into a myocardial infarction model show that the cells can thicken the scar, improve LVEF, and in some cases reduce LV volumes, decrease dyskinesis, and improve vascularization. Our experimental therapeutic effects of cell therapy have also been observed in clinical trials. For example, Chugh et al ⁷ reported that administration of autologous cardiac stem cells improved both global and regional LV function, reduced infarct size, and increased viable tissue as assessed by cardiac magnet resonance imaging at 12 months after treatment in patients with heart failure of ischemic etiology. Makkar et al ⁸ showed that intracoronary infusion of autologous cardiosphere-derived cells after myocardial infarction reduced scar mass, increased viable heart mass and regional contractility, and improved regional systolic wall thickening compared with controls at 6 months after treatment. Williams et al ⁹ injected autologous bone marrow progenitor cells (mononuclear or MSCs) into the LV scar and border zone of patients with LV dysfunction related to remote myocardial infarction. Cardiac magnet resonance imaging at 1 year demonstrated that cell therapy decreased end diastolic volume, resulted in a trend toward decreased end systolic volume, decreased infarct size, and improved regional LV function, and these changes predicted subsequent reverse remodeling. For more clinical trial-related cardiac regeneration using cell transplantation therapy, we refer readers to other references. ^{10 -13}

Biomaterial Implantation to Limit Postinfarction Remodeling

Collagen Implantation

Ventricular remodeling is a process characterized by ventricular wall thinning and dilatation and infarct expansion. Ventricular wall thinning may play a crucial role in postmyocardial infarction remodeling according to Laplace’s law. Because thinning of the myocardial wall can cause an increase in wall stress that contributes to infarct expansion and ventricular dilation, implantation of stiffer materials to increase the thickness of the infarcted ventricular wall might decrease heart wall stress, limit adverse LV remodeling, and improve cardiac function. ¹⁴ A thickened infarct wall that is more resistant to paradoxical systolic buldging would also be expected to demonstrate less dyskinesis and improve ejection fraction and forward cardiac output. Dai et al ¹⁵ directly injected highly purified bovine dermal collagen (100 μL, n = 12) or saline (100 μL, n = 12) into the infarcted wall of 1-week-old myocardial infarctions in adult female Fischer rats. Six weeks after collagen implantation, histology showed that collagen injection significantly increased scar thickness (0.719 ± 0.026 mm) compared with the saline-treated group (0.44 ± 0.034 mm, P = 2.6 × 10⁻⁶). By LV angiography, LVEF was significantly greater in the collagen-treated group (48.4 ± 1.8%) than in the saline-treated group (40.7 ± 1.0%, P = .002). Analysis of regional wall motion demonstrated paradoxical systolic bulging in 5 of the 10 saline-treated rats that averaged 20.3 ± 2.6% of the LV diastolic circumference, but in none of the 11 collagen-treated rats (P = .012). Postmortem LV volume was comparable between the saline-treated group (0.337 ± 0.017 mL) and the collagen-treated group (0.325 ± 0.020 mL, P = .64). Expansion index was 0.78 ± 0.04 in collagen-treated versus 1.3 ± 0.12 in saline-treated hearts (P = .0003).

Cardiac Tissue Extracellular Matrix Implantation

In another recent study, ¹⁶ our research group, in collaboration with Dr Zhang at Wake Forest Institute for Regenerative Medicine, prepared cardiac tissue extracellular matrix from Fischer rat hearts by a complicated process of decellularization. The cardiac extracellular matrix (70 μL, n = 19) or saline (70 μL, n = 17) was directly injected into the 1-week-old infarcted LV wall in adult female Fischer rats. At 6 weeks after treatment, the infarct size was similar between the saline group (48.0 ± 1.2%) and the ECM group (45.7 ± 2.2%, P = .37). The average infarcted LV wall was 24.4% thicker in the ECM-treated group than in the saline-treated group (0.602 ± 0.029 vs 0.484 ± 0.03 mm, P = .0084). The ECM treatment increased LVEF by 4.3% compared to saline treatment (56.7 ± 1.4% vs 52.4 ± 1.5%, P = .043). The LV angiography demonstrated that paradoxical LV systolic bulging was significantly reduced in the ECM-treated group (6.2 ± 1.6% of the LV circumference) compared to the saline-treated group (10.3 ± 1.3%; P = .048). The density of blood vessels that contained perfused blue particles in the scar area was smaller in the matrix-treated group (165 ± 23 vessels/mm², n = 19) compared to the control group (189 ± 12 vessels/mm², n = 17; P = .077). There was a nonsignificant trend toward smaller postmortem LV volume in the matrix-treated group (349 ± 12 μL) compared to the saline group (371 ± 11 μL; P = .20). Infarct expansion index was significantly lower in the matrix-treated group (1.053 ± 0.051) than in the saline-treated group (1.382 ± 0.096, P = .0058).

Experimental Biomaterial Implantation Studies From Other Research Groups

In summary, injection of noncellular material increased infarct wall thickness, improved LVEF, and reduced expansion index. Our experimental data are consistent with the findings from other research groups using different biomaterials. Morita et al ¹⁷ injected saline or tissue filler material into the infarct within 3 hours of myocardial infarction in sheep. Eight weeks later, tissue filler material injection resulted in durable infarct thickening and stiffening, reduced LV dilatation and infarct expansion, and improved LVEF. Seif-Naraghi et al ¹⁸ developed an injectable hydrogel derived from porcine myocardial extracellular matrix and demonstrated that transendocardial injections of the myocardial matrix hydrogel at 2 weeks after myocardial infarction improved cardiac function, ventricular volumes, and global wall motion scores at 3 months in infarcted pigs compared to controls. Furthermore, a significantly larger zone of cardiac muscle was found at the endocardium of the hearts that received matrix hydrogel injection. For more information about myocardial tissue engineering, we refer readers to references. ^{19 -22}

Comparison of Therapeutic Effects of Cell and Biomaterial Injection

In the above-mentioned 7 published articles, our research group used the same 1-week-old myocardial infarction model in adult female Fischer rats or nude rats to investigate separately the therapeutic effects of intramyocardial injection of fetal/neonatal cardiac cells, stem cells, or collagen/extracellular matrix. The injection was performed under direct visualization with a 28-gauge needle attached to an insulin syringe in all of our above-mentioned studies. In order to compare the therapeutic effects of intramyocardial injection of cells versus biomaterial in the rat myocardial infarction model, we reviewed the data including LVEF, motion of the infarct wall, infarct wall thickness, LV expansion index, LV volume, and neovascularization within the infarct area, where available.

Our data indicated that both cell (fetal, neonatal, and hESC-CMs) and biomaterial treatment had similar beneficial effects on cardiac structure (in general, increasing infarct wall thickness and preventing infarct expansion) and function (preventing paradoxical LV systolic bulging and improving LVEF; Table 1). However, only cells appeared to benefit neovascularization within the infarct area. The mechanism for improvement in cardiac structure and function may largely be a passive mechanical phenomenon that is not dependent upon enhanced angiogenesis.

Table 1.

Comparison of Therapeutic Effects of Cell and Biomaterial Therapy.^a

	Cell Group (vs Control Group)	Biomaterial Group (vs Control Group)
Infarct wall thickness	↑	↑
Left ventricular ejection fraction	↑	↑
Left ventricular paradoxical systolic bulging	↓	↓
Left ventricular expansion index	↓	↓
left ventricular volume	↓	↓
Neovascularization with the infarct area	↑	—

^aCompared to control group ↑: increase; ↓: decrease; and —: no difference.

Underlying Mechanisms of Cell and Biomaterial Therapy

Recently, Shiba et al ²³ prepared hESC-CMs that stably expressed the genetically encoded fluorescent calcium sensor. These hESC-CMs exhibited robust fluorescence transients with each contractile cycle and were injected into intact and cryoinjured guinea pig hearts. After 2 or 4 weeks, epicardial fluorescent transients and host electrocardiogram showed that the engrafted cells demonstrated consistent 1:1 host–graft coupling in the intact heart. However, there were both coupled and uncoupled regions in the injured hearts. The data suggested that, in some cases, engrafted cells electromechanically couple and contract synchronously with host muscle. In other cases, the engrafted cells were isolated from the surrounding host cardiac cells by scar tissue and beat autonomously. We ²⁴ also observed that engrafted hESC-derived cardiomyocytes formed sizable graft, expressed muscle marker α-actinin and exhibited cross-striations (sarcomeres), and presented nascent gap junctions with the surrounding host cells at 4 weeks after injection into the LV wall that was subjected to ischemia–reperfusion with minimal infarction in athymic nude rats. These findings provide direct evidence that cell transplantation therapy may provide an approach for true regeneration of the heart muscle that improves cardiac function through addition of new force-generating grafts.

However, experimental cell therapy studies in our laboratory demonstrated that both cardiac and noncardiac cell transplantation therapy have similar improvement in cardiac function and lack of specificity. Therefore, the contribution of cell therapy to increased cardiac function may not be associated with active contraction of engrafted donor cells in all cases. The underlying mechanisms by which these cells improve cardiac function remain unclear. Replacement of scar tissue by any mass (which in this case was transplanted cells) induced neoangiogenesis and paracrine effects of factors released by the transplanted cells may be involved in the beneficial effects of cell transplantation therapy. ²⁵ Ii et al ²⁶ injected human multipotent adipose-derived stem cells into ischemic myocardium immediately following myocardial infarction in nude rats. Four weeks later, echocardiography demonstrated that cell treatment significantly improved cardiac function. Histology showed that cell treatment significantly reduced infarct size and increased capillary density in the peri-infarct myocardium but no transdifferentiation of cardiac or vascular lineage cells was observed. Expression of multiple proangiogenic growth factors and chemokines, such as vascular endothelial growth factor, basic fibroblast growth factor, and stromal cell-derived factor 1α, significantly increased, which resulted in endogenous bone marrow stem/progenitor recruitment to ischemic myocardium. The data suggested that stem cell therapy might improve cardiac function through paracrine effects and endogenous stem/progenitor recruitment rather than its direct contribution to tissue regeneration.

Another possible explanation is that cell therapy might improve postinfarction cardiac function in part through thickening the scar and preventing dyskinesis plus adverse LV remodeling. Regional LV wall thinning leads to mismatch between LV volume and wall thickness that results in increased wall stress. ²⁷ Increased LV wall stress worsens the LV dilation and impairs cardiac pump function. ²⁸ Wall et al ¹⁴ examined the effects of injecting material into the LV wall using a validated finite element model of an ovine left ventricle with an anteroapical infarct and demonstrated that addition of noncontractile material to a damaged LV wall reduced heart wall stress and had important effects on cardiac mechanics. Wall stress was reduced in proportion to the changes in wall volume by implanted materials. This theoretical study supported the concept that injection of cells and/or synthetic extracellular matrices to the damaged ventricle should restore cardiac function through a reduction in elevated wall stress. Injectable biomaterials can reduce wall stress by increasing the scar thickness and stabilizing the chamber size. ²⁹ Wenk et al ³⁰ also developed a 3-dimensional finite element model of the left ventricle using 3-dimensional echocardiography data from sheep with anteroapical infarcts to examine the mechanical effects of the injection of noncontractile material into the myocardium. They demonstrated that wall thickening achieved by tissue filler into the infarct helped to normalize cardiac wall stress in an injured ventricle, which could limit adverse remodeling.

That cell therapy may improve cardiac function by enhancing the passive properties of the infarcted wall is supported by biomaterial therapy studies. Besides the collagen and decellularized heart tissue extracellular matrix that we have implanted into the infarcted myocardium in the rat model, many other biomaterials, such as alginate, chitosan, fibrin, hyaluronic acid, keratin, matrigel, and variations of poly(N-isopropylacrylamide), and poly(ethylene glycol), have been developed and evaluated as injectable biomaterial therapies for myocardial infarction. ^22,31 Most of the studies demonstrated that implantation of injectable biomaterial increased scar thickness, provided physical support to the damaged cardiac tissue, prevented adverse cardiac remodeling, and preserved LV geometry and prevented a deterioration of cardiac function following myocardial infarction.

Advantages and Disadvantages of Cell Therapy

Both cell and biomaterial therapies display advantages and limitations. Compared to biomaterial implantation, cell therapy may repair injured myocardium by regenerating functional cardiomyocytes that may contract in synchrony with the surrounding host myocardium to improve cardiac structure and function. Theoretically, this active contraction should generate better cardiac output than simply preventing dyskinesis with injection of noncellular material. However, there are around 40 million myocytes/1 g of adult myocardium. ³² About 50 g of the heart muscle will be lost in a typical human myocardial infarction. ³³ Therefore, to regenerate the damaged myocardium to a similar size (50 g approx 1-2 billion cells), the amount of transplanted cells needed is huge. On the other hand, if the injected cells proliferate or attract homing of endogenous cardiac stem cells, these factors could enhance regeneration of myocardium. Complicating matters are the limitations of cell therapy including poor retention and survival rate of injected cells within the scar area. In addition, cardiac myocytes as donor cells are difficult to be obtained, expanded, and must be allogeneic cells in order to be “off the shelf.” Once cells that are differentiated are injected, there is limited potential for proliferation and full replacement of infarcted myocardial mass. Other cell sources, such as stem cells, have the problems of differentiation into the cardiac cell phenotype, arrythmogenicity, immunogenicity, and tumorigenicity. Lengthy processing of the cells could result in the potential for contamination, mutations, or infections.

Potential Advantages and Disadvantages of Biomaterials Therapy

Although biomaterial therapy itself can’t replace lost myocardial mass with new, contractile myocardium, the therapeutic effects of cell therapy, such as improvement in postmyocardial infarction cardiac structure and function, can be achieved at least in part by biomaterial therapy. As mentioned, in our experiments, such materials consistently thickened the infarcted wall and improved LVEF. There are 2 kinds of biomaterials used for tissue engineering and regeneration: natural materials derived from animal tissues or plants or synthetic materials. Natural materials may create a microenvironment to facilitate exogenous or endogenous cells adhesion, proliferation, and differentiation. However, potential limitations of natural materials might be batch-to-batch variations and animal-derived tissue might have potential contamination. Some natural materials lack sufficient mechanical strength and degrade rapidly in the heart. Neovascularization within the implanted graft might be difficult or delayed for some biomaterials. ¹⁵ Synthetic materials have precise molecular weight and degradation time, can be manufactured with predictable and reproducible mechanical and physical properties, and can ensure off-the-shelf availability. ²¹ Because the biomaterial is easy to prepare, direct injection of biomaterials into injured myocardium becomes an attractive approach for in situ cardiac tissue engineering. In Europe and Israel, ongoing clinical studies are investigating the effects of alginate injections on cardiac function in patients with myocardial infarction (clinicaltrials.gov, NCT01226563 and NCT01311791).

Combination of Cells Plus Biomaterial

Combining cell transplantation with biomaterials may offer an advantage of improving the local retention of cells and preventing their migration to remote organs. Dai et al ³⁴ labeled bone marrow-derived rat MSCs with isotopic colloidal nanoparticles containing europium and used collagen matrix as a delivery vehicle to transplant the cells to 1-week-old infarcted rat myocardium. Four weeks later, collagen matrix as a delivery vehicle significantly reduced the relocation of transplanted MSCs to remote organs (lung, liver, spleen, and kidney) and noninfarcted myocardium. Yu et al ³⁵ encapsulated human MSCs in alginate microspheres and injected them into infarcted myocardium at 1 week postmyocardial infarction in female nude rats. Ten weeks postinjection, cell survival was significantly increased in the encapsulated human MSC group compared to cells alone. The results suggested that surface modification and microencapsulation of cells with biomaterials increased stem cell retention and survival within the damaged myocardium.

Another challenging approach in the field of cardiovascular tissue engineering is the creation of a contracting heart muscle patch. In an early study, Leor et al ³⁶ made cell constructs by culturing fetal cardiac cells within 3-dimensional porous alginate scaffolds. Histology showed that the cardiac cells formed distinctive, multicellular contracting aggregates within the scaffold pores within 2 to 3 days in culture. The scaffolds with cells were implanted into the surface of 1-week-old myocardial infarction in rats. Nine weeks later, intensive neovascularization from the neighboring coronary network was observed in the implanted graft, which was well integrated with the host tissue. Treatment with scaffolds attenuated LV dilatation and prevented progressive deterioration in LV contractility in this experimental rat model. Tulloch et al ³⁷ cultured hESC and human-induced pluripotent stem cell-derived cardiomyocytes in a 3-dimensional collagen matrix and showed that uniaxial mechanical stress conditioning promoted cardiomyocyte proliferation in vitro. After transplantation onto epicardial surfaces of uninjured athymic rat hearts, 1 week later, the implanted human myocardium formed microvessels that were perfused by the host coronary circulation and survived and developed connections to the surrounding host myocardium. Although various biomaterials and cell sources have been used for the creation of engineered cardiac tissue grafts in recent years, there are several critical limitations that remain to be resolved. One problem is that the maximum size of the graft is limited by maximum diffusion distances for nutrients and oxygen. Others include vascularization after implantation in vivo, the electrically and mechanically coupling to the surrounding host cardiac cells (for review, read ³⁸ ).

Biomaterials as a Delivery Vehicle

Biomaterials can be used as a vehicle for intramyocardial sustained delivery of growth factor. Binsalamah et al ³⁹ intramyocardially injected chitosan–alginate nanoparticles containing placental growth factor into an acute myocardial infarction model in rats. Placental growth factor was released at the site of action for an extended time period and significantly decreased scar area formation and stimulated myocardial angiogenesis at the infarction border and improved cardiac function at 8 weeks after treatment. Ruvinov et al ²⁹ intramyocardially injected affinity-binding alginate biomaterial loaded with insulin-like growth factor 1 and hepatocyte growth factor into the infarction in rat. This treatment preserved scar thickness, attenuated infarct expansion, reduced scar fibrosis, increased angiogenesis, and mature blood vessel formation, prevented cell apoptosis, and induced cardiomyocyte cell cycle reentry after 4 weeks.

Future Research Directions

The underlying mechanisms of the therapeutic effects of cell transplantation remain undetermined and many of their therapeutic effects could be achieved by biomaterial implantation to improve the passive properties of damaged myocardium. However, cell therapy has a potential advantage to replace damaged myocardium with a contractile graft, as well as the newer concept that cells may have paracrine effects that improve the surrounding tissue, possibly by recruiting endogenous stem cells. The regeneration of new contractile myocardium in the diseased heart depends on the further understanding of stem cell biology. Considering the lost mass of myocardial infarction, the ideal donor cells should maintain the ability to proliferate within the recipient heart, could be induced to differentiate into mature cardiac cells with mature sarcomeres, form gap junctions and be aligned appropriately with the native myocardium, and be capable of synchronous longitudinal shortening. ⁴⁰ Replacing damaged myocardium with active functional synchronic contractile tissue remains the optimal goal.

Cumulative experimental studies demonstrate that biomaterial implantation can improve the passive properties of myocardium of the diseased heart. The ideal biomaterials for cardiac tissue engineering remain to be determined and should (1) be nonimmunogenic, nontoxic, biocompatible, and not induce a foreign body reaction; (2) be sterilizable; (3) match biomechanical characteristics of tissue it is replacing, and have stable structure to withstand the shearing forces and transmit contractile forces; (4) can be modified to release growth factors, gene signals, and other proteins, in a time-dependent manner, if necessary; (5) can create a microenvironment for cell adhesion, proliferation, and differentiation, as well as an environment that is favorable for vascularization; and (6) be injectable for catheter delivery at minimally invasive surgery. ^{41 -43}

A novel approach for cardiac regeneration is the concept of injecting a combination of cells and biomaterial, as well as the more combinations with growth factors (for review, read ^21,42 ). The progress in the understanding of developmental biology of cardiomyocytes and the searching of suitable engineering biomaterials will be crucial for successful cardiac regeneration.