Available online at www.sciencedirect.com
Materials Science and Engineering R 59 (2008) 1–37 www.elsevier.com/locate/mser
Biomaterials in cardiac tissue engineering: Ten years of research survey Qi-Zhi Chen a,b,*, Siaˆn E. Harding b, Nadire N. Ali b, Alexander R. Lyon b, Aldo R. Boccaccini a,* b
a Department of Materials, Imperial College London, Prince Consort Road, London SW7 2AZ, UK National Heart and Lung Institute, Imperial College London, Dovehouse Street, London SW3 6LY, UK
Received 24 June 2007; received in revised form 14 August 2007; accepted 14 August 2007 Available online 10 January 2008
Abstract Driven by enormous clinical need, myocardial tissue engineering has become a prime focus of research within the field of tissue engineering. Myocardial tissue engineering combines isolated functional cardiomyocytes and a biodegradable or nondegradable biomaterial to repair diseased heart muscle. The challenges in heart muscle engineering include cell related issues (such as scale up in a short timeframe, efficiency of cell seeding or cell survival rate, and immune rejection), the design and fabrication of myocardial tissue engineering substrates, and the engineering of tissue constructs in vitro and in vivo. Several approaches have been put forward, and a number of models combining various polymeric biomaterials, cell sources and bioreactors have been developed in the last 10 years for myocardial tissue engineering. This review provides a comprehensive update on the biomaterials, as well as cells and biomimetic systems, used in the engineering of the cardiac muscle. The article is organized as follows. A historic perspective of the evolution of cardiac medicine and emergence of cardiac tissue engineering is presented in the first section. Following a review on the cells used in myocardial tissue engineering (second section), the third section presents a review on biomaterials used in myocardial tissue engineering. This section starts with an overview of the development of tissue engineering substrates and goes on to discuss the selection of biomaterials and design of solid and porous substrates. Then the applications of a variety of biomaterials used in different approaches of myocardial tissue engineering are reviewed in great detail, and related issues and topics that remain challenges for the future progress of the field are identified at the end of each subsection. This is followed by a brief review on the development of bioreactors (fourth section), which is an important achievement in the field of myocardial tissue engineering, and which is also related to the biomaterials developed. At the end of this article, the major achievements and remaining challenges are summarized, and the most promising paradigm for the future of heart muscle tissue engineering is proposed (fifth section). # 2007 Elsevier B.V. All rights reserved. Keywords: Tissue engineering; Cardiac muscle; Biomaterials; Polymers; Cell therapy; Bioreactors
Contents 1. 2.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . Cells applied in myocardial tissue engineering . . 2.1. Somatic muscle cells . . . . . . . . . . . . . . . 2.1.1. Foetal or neonatal cardiomyocyte . 2.1.2. Skeletal myoblast . . . . . . . . . . . . 2.2. Angiogenic cells. . . . . . . . . . . . . . . . . . . 2.2.1. Fibroblasts. . . . . . . . . . . . . . . . . 2.2.2. Endothelial progenitor cells . . . . . 2.3. Stem cell-derived myocytes . . . . . . . . . . . 2.3.1. Basics of stem cells . . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
* Corresponding authors. Tel.: +44 20 75946723; fax: +44 20 75946757. E-mail addresses:
[email protected] (Q.-Z. Chen),
[email protected] (A.R. Boccaccini). 0927-796X/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.mser.2007.08.001
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
. . . . . . . . . .
2 5 6 6 7 7 7 7 8 8
2
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
3.
4.
5.
2.3.2. Bone marrow-derived stem cells. . . . . . . . . . . . . . 2.3.3. Adipose-derived stem cells . . . . . . . . . . . . . . . . . 2.3.4. Native cardiac progenitor cells . . . . . . . . . . . . . . . 2.3.5. Embryonic stem cells . . . . . . . . . . . . . . . . . . . . . 2.3.6. Human embryonic stem cells . . . . . . . . . . . . . . . . 2.4. Strategies to address immune rejection in cells . . . . . . . . . 2.5. Strategies of cell delivery . . . . . . . . . . . . . . . . . . . . . . . . 2.6. Summary of cell-based therapy and their limitations. . . . . . Biomaterials for myocardial tissue engineering . . . . . . . . . . . . . . 3.1. Overview of substrate development for tissue engineering . . 3.1.1. Criteria on tissue engineering substrates . . . . . . . . 3.1.2. Polymers used in soft tissue engineering . . . . . . . . 3.1.3. Fabrication of tissue engineering substrates . . . . . . 3.2. Biomaterials for myocardial tissue engineering . . . . . . . . . 3.2.1. Selection of biomaterials and design of substrates . 3.2.2. Biomaterials used in myocardial tissue engineering Biomimetic tissue engineering—bioreactors . . . . . . . . . . . . . . . . 4.1. Brief history of bioreactors . . . . . . . . . . . . . . . . . . . . . . . 4.2. Comparison of bioreactors . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Application of bioreactors in myocardial tissue engineering. 4.4. Key issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1. Introduction A historic perspective is presented in this section on the evolution of cardiac medicine and the emergence of cardiac tissue engineering as a major branch in the field of tissue engineering. It is aimed to understand the rationale of cardiac tissue engineering and related strategies. A general retrospect on the history of tissue engineering has been given recently by Vacanti [1]. Heart disease is the leading cause of death and disability in both industrialised nations and the developing world, accounting for approximately 40% of all human mortality [2]. It is estimated that 5 million Americans, 1.8 million Britons, and 25 million people worldwide suffer from heart failure, with approximately 550,000 and 120,000 new cases diagnosed each year in the United States (US) and the United Kingdom (UK), respectively [3]. Prognosis is poor with 40% mortality within 12 months of diagnosis, and a 10% annual mortality rate thereafter [4]. The economic burden imposed by this disease has reached more than $33 billion in the US and more than £700 million in the UK annually [3]. Heart failure is a condition reflecting impairment of the pumping efficiency of the heart, and it is caused by a variety of underlying diseases, including ischemic heart disease with or without an episode of acute myocardial infarction, hypertensive heart disease, valvular heart disease, and primary myocardial disease. The single most common cause of left-sided cardiac failure is ischemic heart disease (also called coronary artery disease) with an episode of acute myocardial infarction. Myocardial infarction typically results in myocyte slippage. The weakening of the collagen extracellular matrix results in heart wall thinning and ventricular dilation. The impairment of
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . .
.. 9 .. 9 .. 9 . 10 . 11 . 11 . 12 . 12 . 13 . 13 . 13 . 13 . 16 . 18 . 18 . 21 . 28 . 28 . 29 . 29 . 30 . 30 . 31 . 31
the heart wall muscle is permanent because, after a massive cell loss due to infarction, the myocardial tissue lacks significant intrinsic regenerative capability to replace the lost cells [5]. The enlargement in ventricular volume leads to progressive structural and functional changes in ventricles (called ventricular remodelling) [5]. Ventricular remodelling is initially compensatory, but adds further inefficiency to the mechanical pumping of the ventricular muscle, predisposing towards the end stage of congestive heart failure (CHF) (or just heart failure) [5], a condition in which the heart cannot pump a sufficient amount of blood to the meet the metabolic requirements of the body [6]. Pharmacological therapy focuses on reduction of work load (utilising diuretics, nitrates) and protection from the toxic humoral factors which are overactivated in heart failure [7]. These include catecholamines (b-blockers), angiotensin-converting enzyme (ACE inhibitors), and aldosterone (spironolactone). Blockade of these humoral factors represents the current standard conservative treatment for patients with mild symptoms of heart failure and slight limitation during ordinary activity [7]. Interventional therapy, such as surgery or implantation of pacing devices to control electrical/mechanical asynchrony, are now receiving more widespread application, in particular for patients with marked symptoms and marked limitation in activity [7–12]. However, both drug and interventional therapies cannot adequately control disease progression to the end stage [13]. Eventually, heart transplantation is the ultimate treatment option to end-stage heart failure. Owing to the lack of organ donors and complications associated with immune suppressive treatments, however, scientists and surgeons constantly look for new strategies to repair the injured heart [14].
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
Fig. 1. Heart–lung machine (http://heartonline.org/postpump.htm).
Historically, all these surgical strategies started with the development of the heart–lung machine (Fig. 1), which takes over the functions of the heart and lungs during an open-heart surgery [15], such as coronary bypass surgery and valve replacement [16,17]. Cardiomyoplasty was an alternative surgical approach for treating heart failure. Pre-prepared skeletal muscle, which is able to function at power levels analogous to those of the heart, was wrapped around the heart, and paced to contract with the heart, thereby improving cardiac pumping power [18,19]. Clinical studies reported that this dynamic cardiomyoplasty could improve left ventricular performance, reduce cardiac dilation, and interrupt disease progression [20,21]. However, quantitative heamodynamic analyses were not consistent, regarding the benefits of active systolic assist, and mortality from the operation was unacceptably high [22,23]. This prompted the suggestion that passive mechanical constraint by the muscle wrap might halt or even reverse the negative remodelling of the dilated ailing heart. Inspired by this hypothesis, many studies have examined
3
the use of biomaterial supports to restrain the left ventricle [24]. Marlex mesh (polypropylene) [25], Merselene mesh (knitted polyester) (26), Acorn CorCapTM heart mesh (knitted polyester) (18) and MyocorTM Myosplint1 [18] are four representative cardiac support devices that have been under investigation. A typical approach of this strategy is illustrated in Fig. 2. Although animal cardiomyoplasty showed distinct benefits of the devices [27], these have not been translated to the clinical setting, and currently evidence for their clinical benefit is absent. None of these devices has received approval of the Food and Drug Administration (FDA) [28]. Studies around the mid-1990s veered to an intriguing strategy: the application of cell transplantation. Initially, it was confirmed that diseased myocardium could be restored by the transplantation of functional cardiac myocytes [29–31]. Since then a number of research groups reproduced and refined these pioneering experiments [32–45]. Most studies support the conclusion that cell implantation in models of myocardial infarction can improve contractile function. Clinical studies are currently under way to investigate the safety and feasibility of cell implantation in patients [39]. In the cell-based therapy, isolated cells are injected to the infarct region via the pericardium, coronary arteries, or endocardium. In order to improve the site accuracy of cell delivery, an alternative approach to deliver cells to the infarct region is to rebuild 3D cell networks in vitro and to implant the cell bandage onto the infarct heart, as shown in Fig. 3 [46]. The third approach involves the usage a man-made heart patch (Fig. 4) [47–49], which is populated in vitro with cells and implanted later in vivo. In this approach, the ring- or sheetshaped heart patch serves two functions: cell delivery and mechanical support. According to theoretical simulations on the effects of injected materials (Fig. 4b), the addition of a sheet material to a damaged left ventricular wall could have important effects on cardiac mechanics, with potentially beneficial reduction of elevated myofibril stresses, as well as
Fig. 2. (a) Cardiac support device by Acorn CorCapTM, a typical approach of left ventricle restraint. http://www.sciencedaily.com/releases/2004/11/. (b) The knitted structure of the mesh [18]. Published with kind permission from Springer Science and Business Media.
4
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
Fig. 3. Schematic graph showing the transplantation of a myocardial cell-sheet graft, also called heart bandage [46]. Reproduced by permission of the MRS Bulletin.
histological and functional changes to clinical left ventricular metrics [50]. No simulation work has been reported yet, as far as the authors are aware of, regarding the effects of epicardial patches (Fig. 4a). The results may be similar in both cases. A more ambitious strategy is the implantation of myocardial tissue generated ex vivo, i.e. a 3D tissue regeneration strategy. This classic strategy of tissue engineering is established on the fact that living bodies have the potential of regeneration, and on the supposition that the employment of natural biology (e.g., cells and biomolecules) will maximise the capacity for regeneration and allow for greater success in developing therapeutic strategies aimed at the replacement and repair of tissue and the maintenance and enhancement of its function [51–53]. The successful development of tissue engineering constructs will have a profound impact in both the scientific community and the public sphere. They could be used to produce in vitro healthy cells for cell-based therapy. They may also be used for many biomedical studies, such as cell biology, organ development, functional cell differentiation from stem cells, environment–cell interaction, cancer biology, new drug treatment, and could ultimately be used for the repair of injured or diseased tissues.
Fig. 4. Schematic illustrations of (a) epicardial and (b) endoventricular heart patch approaches to deliver isolated cells to the infarct regions.
In essence, tissue engineering is a technique of imitating nature. Natural tissues consist of three components: cells, extracellular matrix (ECM), and signalling systems. The ECM is made up of a complex of cell secretions immobilised in spaces and thus forming a scaffold for its cells. Hence, it is natural that the engineered tissue construct is a triad [54], the three constitutes of which correspond to the above-mentioned three basic components of natural tissues. Fig. 5 illustrates the triad, i.e. a scaffold, living cells and signal molecules (such as growth factors and cytokines). The European Commission on Health and Consumer Protection defined tissue engineering as ‘‘the persuasion of the body to heal itself through the delivery, to the appropriate site, independently or in synergy, of cells, biomolecules and supporting structures’’ [53]. According to this definition, the surgical approaches described above could be all classified into the tissue engineering category, as listed in Table 1. It must be mentioned that that these approaches are not independent of one another. The heart patch approach, for example, is a combinatory paradigm of passive diastolic constraint and cell therapy. The classic 3D tissue engineering construction could be optimised by the incorporation of drug and gene therapies. The application of biomaterials has been mainly related with the last four approaches listed in Table 1, i.e. left ventricular constraint, scaffold-free cell sheet implantation, heart patch implantation and 3D tissue engineering construction. Cardiac tissue engineering covers heart valve, cardiovascular and myocardial tissue engineering. This review focuses on the myocardial tissue engineering. Recent reviews are available on heart valve [55–58] and cardiovascular tissue engineering [56,59–62]. In the past 10 years, many studies have been published using different cells and different biomaterials for heart muscle engineering [14,63–65], and excellent reviews focusing on different aspects of cardiac muscle engineering are also available [14,24,52,63–79]. In this article, we present a
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
5
Fig. 5. Triad of a classic tissue engineering construct.
comprehensive review on the achievements of myocardial tissue engineering, including cells, biomaterials, and biomimetic approach (i.e. bioreactors). Our intention is to identify what we consider significant challenges and the most promising approaches in the future of heart muscle tissue engineering. At the same time, we are aware of the possibility that our own biases might be embedded in the opinions expressed in the review. We also apologise in advance to any individuals whose significant effort in the field of heart tissue engineering was neglected due to our misunderstanding or oversight.
As early as in 1978, Bader and Oberpriller demonstrated the regenerative capacity of amphibian hearts after autologous implantation of minced ventricular tissue samples (i.e. nonisolated cells) into injured newt hearts [80]. In this study, a partial regeneration of injured newt ventricles was observed. However, grafted tissue fragments remained morphologically Table 1 Currently applied or potential strategies for the treatment of heart failure patients 1. Pharmaceutical therapy
2. Cells applied in myocardial tissue engineering This section is devoted to provide a concise review on cells for heart tissue engineering, including key studies. Although it is not biomaterial specific, this knowledge is essential to biomaterials scientists in this field, as cell implantation plays a pivotal role in the engineering of the heart muscle that lacks significant intrinsic regenerative capability. After reading this section, material scientists hopefully will have a guide to browse the huge amounts of reports available on a variety of cell types applied in cardiac muscle regeneration and engineering.
2. Interventional therapy (1) Reduction of the heart volume (2) Implantation of a pace-maker 3. Heart transplantation 4. Tissue engineering strategy (1) Cardiomyoplasty (active systolic assist) (2) Cell-based therapy (isolated cell-delivery) (3) Left ventricular restraint (passive diastolic constraint) (4) Scaffold-free cell-sheet implantation (5) Heart patch implantation (passive diastolic constraint and cell delivery) (6) 3D tissue engineering construction (a scaffold + cells + macromolecules)
6
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
and functionally separated from the native myocardium. True tissue engineering approaches emerged in early 1990s when efforts to regenerate functional myocardial tissue were invested in grafting of isolated cell [30,31]. Since then, numerous basic studies on cells and several early-stage clinical trials have been carried out using a variety of cell types with the hope of improving myocardial function. So far, a variety of cell models have been under intensive investigation. They can be categorised into three groups (Table 2): (1) somatic muscle cells, such as foetal or neonatal cardiomyocytes [42,44,45,81– 87] and skeletal myoblasts [88–92], (2) myocardium-generating cells, such as embryonic stem cells [93–97] (possibly) bone marrow-derived mesenchymal stem cells [98–106] and adipose stem cells [107]; and (3) angiogenesis-stimulating cells, including fibroblasts [108] and endothelial progenitor cells. Each of these cell types and cell delivery approaches are reviewed in more detail in the following sections. 2.1. Somatic muscle cells 2.1.1. Foetal or neonatal cardiomyocyte Early cell transplantation studies focused on using foetal or neonatal rodent (rat or mouse) cardiomyocytes, as these cells have the inherent electrophysiological, structural and contractile properties of cardiomyocytes and still retain some proliferate capacity [31,81,87]. In their pioneering study, Soonpaa et al. established the principles of cardiac cell implantation in the heart. They demonstrated that foetal cardiomyocyte could be transplanted and integrated within the healthy myocardium of mice, and that the surviving donor cells were aligned with recipient cells and formed cell-to-cell contacts [31]. This group also reported the foetal cardiomyocyte graft in the myocardium of dystrophic mice and dogs [87]. More studies have demonstrated that cardiac myocytes from neonatal, embryonic or adult models can also be engrafted into diseased (infarcted, cryoinjured or cardiomyopathic) hearts
[30,42,44,45,81–83,85,86,109]. These results also indicate that early-stage cardiomyocytes (foetal and neonatal) were better candidates than more mature cardiac cells due to their superior in vivo survival [42]. Time course studies for the survival of grafted cardiomyocytes in the healthy heart were carried out by Muller-Ehmsen et al. [34]. They isolated and injected male donor neonatal rat cardiomyocytes into the left ventricular (LV) wall of adult female inbred rats. They demonstrated that these cells could survive and improve cardiac function for up to 6 months in a rat model of chronic myocardial infarction [33]. Murry’s group [110] showed, using syngeneic rat with cryoinjury, that cell graft survival after 7 days could be up to 33%. Cardiomyocyte transplantation, which was applied to smaller infarcts [83], has been proved effective in the prevention of cardiac dilation and remodelling following infarction [36] and the improvement of the ventricular function [44,85]. Several mechanisms have been proposed for improved heart function following cardiac myocyte transplantation [36,70,111,112]: (1) direct contribution of the transplanted myocytes to contractility; (2) attenuation of infarct expansion by virtue of the elastic properties of cardiomyocytes; (3) angiogenesis induced by growth factors secreted from the foetal cells resulting in improved collateral flow; (4) paracrine effects via the release of beneficial growth factors from the transplanted cells, which support the cardiomyocytes under strain in the failing heart, and may possibly recruit residential cardiac progenitor cells. However, the transplanted tissue decreased in size several months after transplantation [42,83]. An electron microscopy study revealed that dead cells had features of both necrosis and apoptosis. Based on the most recent experiments, it is apparent
Table 2 Potential cell sources for myocardial regeneration in human and their advantages and disadvantages for myocardial repair [63] Cell source
Autologous
Easily obtainable
Highly expandable
Cardiac myogenesis
Clinical trial
Safety
Somatic cells Foetal cardiomyocytes Skeletal myoblasts Smooth muscle cells Fibroblasts
No Yes Yes Yes
No Yes Yes Yes
No Depend on age Yes Yes
Yes Debated No No
No Yes No No
No Yes, arrhythmias No No
Yes Yes Yes No Yes
No Yes Yes Yes Yes
Depend on age Depend on age Depend on age Yes Yes
Yes Debated Debated Debated Yes
No No Yes No No
Yes, fibrosis calcification Yes, calcification Yes, calcification No Yes
No
No
Yes
Yes
No
Yes, potential teratoma if cells escape differentiation
Stem cells Somatic stem cells Mesenchymal stem cells Endothelial progenitor cells Crude bone marrow Umbilical cord cells (Hemaetopoietic stem cells) Adipose stem cells Embryonic stem cells Human embryonic stem cells
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
that cell death is rapid and extensive after cardiomyocyte grafting, with most cell deaths occurring during the first 2 days. However, after 1 week the graft is relatively stable [34,110]. Although this is important proof-of-concept work, there is no realistic possibility of human or rat neonatal cardiomyocytes coming to clinical application [110,113,114]. As a result, several alternative approaches have been developed to overcome the limitations of foetal cardiomyocyte transplantation and to obviate the need for immunosuppressants. 2.1.2. Skeletal myoblast Theoretically, skeletal muscle cells may be superior to cardiomyocytes for infarct repair, because skeletal myoblasts have almost all the properties of the ideal donor cell type except their non-cardiac origin. Skeletal myoblast satellite cells can be harvested from autologous sources, which obviate the need for immune suppression. Satellite cells are mononuclear progenitor cells found in mature muscle. In undamaged muscle, the majority of satellite cells are quiescent. Upon muscle damage, satellite cells become activated and are able to differentiate and fuse to augment existing muscle fibres and to form new fibres. They can be rapidly expanded in an undifferentiated state in vitro to clinically applicable numbers of myoblasts without a risk for tumourgenecity, and they have the capabilities to withstand ischemia better than many other cell types. Continued proliferation in vivo may be an advantage when engrafting into an injured heart, since the input of a smaller number of cells might give rise to a large graft [70,115]. Although it was originally hoped that skeletal myoblasts would adapt a cardiac phenotype, it is now clear that within heart tissue the skeletal myoblasts remain committed to form only mature skeletal muscle cells that possess completely different electromechanical properties than those of heart cells. Moreover, given the inability of myoblasts to form electromechanical connections with host cardiomyocytes (due to lack of expression of adhesion and gap junction proteins), it is not surprising that physiological studies failed to demonstrate synchronous beating of the grafted cells within the host tissue [116]. However, studies in small and large animal models of infarction demonstrated beneficial effects of grafting of these cells on ventricular performance [117,118]. The mechanisms underlying the beneficial effects of skeletal myoblasts remain to be elucidated. The improvement in heart wall motion could be achieved by contraction of the transplanted cells, a local effect on scar remodelling by mechanical support and/or paracrine influences on the remodelling process. Nevertheless, given their autologous origin, the capacity to amplify primary myoblasts from human muscle biopsies, and the encouraging preclinical results, skeletal myoblasts were the first cell type to reach clinical application [92,119–126]. The pioneering Phase I clinical trials were performed either using a direct surgery approach (during coronary artery bypass graft surgery) or using a percutaneous endocardial catheter-delivery approach and have demonstrated both the feasibility of the procedure and the ability of the cells to engraft in the infarcted myocardium [92,119–121,123–128].
7
The clinical application of autologous skeletal myoblasts is currently limited by several concerns [115,129,130]. (1) Lack of myocardial phenotype. This has been blamed for the disturbingly high incidence of life-threatening ventricular arrhythmias noted in the initial post transplantation phase in these trials [119,120]. (2) Low recovery of satellite cells. The recovery of satellite cells from muscle biopsies of elderly patients is low. (3) Efficiency. Grafting methods need be developed to improve the efficiency of cell engraftment and survival. (4) Questionable beneficial effects. The efficacy of skeletal myoblast therapy is still uncertain because of the following facts. Firstly, the Phase I clinical trials were not randomised, and might be placebo controlled. Secondly, widely different cell transplantation protocols were used in a relatively small number of patients. Finally, the trials were associated with other confounding factors such as concomitant left ventricle assistant device (LVAD) implantation or revascularisation. Ongoing Phase II clinical trials will hopefully address these concerns and thoroughly evaluate the safety and efficacy of myoblast transplantation. 2.2. Angiogenic cells 2.2.1. Fibroblasts Vascularisation is a key step in tissue repair. At sites of injury, pheno-transformed fibroblast-like cells are responsible for fibrous tissue formation. These cells are termed myofibroblasts because they contain alpha-smooth muscle actin microfilaments and are contractile. In vivo studies of injured rat cardiac tissues and in vitro cell culture studies [131] have shown that such fibroblast-like cells contain requisite components for angiotensin peptide generation and angiotensin II receptors. Such locally generated angiotensin II acts in an autocrine/paracrine manner to regulate collagen turnover and thereby tissue homeostasis in injured tissue. Human dermal fibroblasts have been applied for myocardial regeneration to stimulate revascularisation and preserve left ventricular (LV) function of the infarcted LV in mice [132]. It has been shown that dermal fibroblasts functioned to attenuate further loss of LV function accompanying acute myocardial infarct and that this might be related in part to myocardial revascularisation. 2.2.2. Endothelial progenitor cells Endothelial progenitor cells (EPCs) are present in the bone marrow and the peripheral blood and exhibit phenotypical markers of mature endothelial cells [133]. It has been found that rats with inflammatory-mediated cardiomyopathy exhibited a significant mobilization of EPCs from the bone marrow to the periphery and their ability to adhere to fibronectin, mature endothelial cells and cultured cardiomyocytes was significantly reduced when compared to healthy rats [134]. This result prompted studies in the application of EPCs in attenuating remodelling followed by acute myocardial infarction. Transfer of EPCs resulted in a functional improvement in cardiac
8
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
performance. EPC transfer is effective in attenuating myocardial damage in a model of non-ischemic dilated cardiomyopathy [134], and probably exert their beneficial effects via new vessel growth and improved blood supply to the failing heart. 2.3. Stem cell-derived myocytes 2.3.1. Basics of stem cells Stem cells are primal undifferentiated, and thus unspecialized, cells that retain the ability to differentiate into multiple cell lineages [135]. In principle stem cells are the optimal cell source for tissue regeneration, including myocardium. Firstly, they are capable of self-replication throughout life such that an unlimited number of stem cells of similar properties can be produced via expansion in vitro. Secondly, the stem cells are clonogenic, and thus each cell can form a colony in which all the cells are derived from this single cell and have identical genetic constitution. Thirdly, they are able to differentiate into one or more specialised cell types. Hence, after expansion stem cells can be directed to differentiate into cardiomyogenic lineage [94,136–138]. For these reasons, stem cell-based therapy for cardiac muscle regeneration has been under intensive research during the last decade. Stem cells can be categorised according to their potency (totipotent, pluripotent, multipotent, and unipotent) (Fig. 6), anatomic source (adult, embryonic, foetus, cord blood, or cancer), or by cell surface markers, transcription factors, and proteins they express. Four basic stem cell types classified according to their potency are briefly introduced as follows [135].
(Fig. 6). Blastocyst embryonic stem cells are cultured cells obtained from the undifferentiated inner cell mass of an early stage pre-implantation embryo. Unlike somatic stem cells, embryonic stem cells can differentiate into any one of the body’s more than 200 cell types. Foetal stem cells. After eighth week of development, the human embryo is referred to as a foetus. By this time it has developed a human-like form. Stem cells in the foetus are responsible for the initial development of all tissues before birth. Like embryonic stem cells, foetal stem cells are pluripotent. Cord blood stem cells are derived from the blood of the placenta and umbilical cord after birth. Umbilical cord stem cells are multipotent. Cancer stem cells arising through malignant transformation of adult stem cells are proposed to be the source of some or all tumours and cause metastasis and relapse of the disease. The common features of all stem cells include: (1) Ability to self renew, which means they can divide and produce stem cell progeny with similar properties.
(1) Totipotent stem cells are produced from the fusion of an egg and sperm cell. These cells can differentiate into embryonic and extra-embryonic cell types. (2) Pluripotent stem cells are the descendants of totipotent cells and can differentiate into cells derived from the three germ layers. (3) Multipotent stem cells can produce only cells of a closely related family of cells (e.g. haematopoietic stem cells differentiate into red blood cells, white blood cells, platelets, etc.). (4) Unipotent cells can produce only one cell type, but have the property of self-renewal which distinguishes them from non-stem cells. There are five sources for stem cells, as described below [139]. Adult stem cells are undifferentiated cells found among differentiated cells of a specific tissue and are mostly multipotent cells. They are more accurately called somatic stem cells because foetal and umbilical cord stem cells also fall into this category. They are present in all tissues and seem to survive long time periods and harsh conditions. Important sources of somatic stem cells include bone marrow-derived stem cells, such as mesenchymal stem cells, and adipose stem cells. Embryonic stem cells include early embryonic stem cells (totipotent) and blastocyst embryonic stem cells (pluripotent)
Fig. 6. Categories of stem cells in terms of potency [140].
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
(2) Clonogenic, which means that each cell can form a colony in which all the cells are derived from this single cell and have identical genetic constitution. (3) Broad differentiation profile, which means they can grow into one or more mature cell types. In this section, we focus on the applications of four types of stem cells in myocardial tissue engineering: bone marrowderived stem cells, adipose stem cells, native cardiac progenitor cells, and embryonic stem cells. 2.3.2. Bone marrow-derived stem cells Bone marrow stem cells are the most primitive cells in the marrow. These cells can be classified into: (1) bone marrowderived mesenchymal stem cell (MSC), and (2) haematopoietic stem cell (HSC). 2.3.2.1. Bone marrow-derived mesenchymal stem cells. Bone marrow-derived mesenchymal stem cells are a subset of bone marrow stromal cells (the term ‘‘mesenchymal stem cell’’ is now used to include multipotent cells that are derived from either bone marrow or other tissues, such as adult muscle or the Wharton’s jelly present in the umbilical cord). This potential multipotent stem cell is derived from the non-haematopoietic, stromal compartment of the bone marrow, which can grow into non-marrow cells, such as bone, cartilage, tendon, adipose, and endothelial cells [141]. A number of studies suggested that bone marrow-derived MSCs could differentiate into cardiomyocytes both in vitro and in vivo [38,40,136,137,142–145]. Makino et al. [137] treated murine mesenchymal stem cells with 5-azacytidine and isolated a cardiomyogenic cell line after repeated screening of spontaneous beating cells. This result was confirmed by Tomita et al. [146]. Later Orlic et al. [38] reported that a subpopulation of bone marrow stem cells were capable of generating myocardium in vivo in mice. More recently, it was reported that transplantation of mesenchymal stem cells into the infarcted myocardium of rats and pigs resulted in improved myocardial performance [144,147]. One possible advantage of mesenchymal stem cells is their ability to be either autotransplanted or allotransplanted, as some reports suggested that they may be relatively privileged in terms of immune compatibility [148]. 2.3.2.2. Haematopoietic stem cells. In addition to the initial hypothesis that bone marrow stem cells might be able to differentiate into myocardium in vivo, another major rationale behind the research of bone marrow stem cells for cardiac muscle regeneration was the essential roles of vascularisation and angiogenesis in tissue regeneration. Studies in the animal models of ischemia and Phases I and II clinical trials suggested that delivery of haematopoietic stem cells and circulating endothelial progenitor cells, both originating from bone marrow stem cells, may result in improvement in the ventricular function in ischemic heart disease patients [115]. Furthermore, since bone marrow stem cells reside in the bone marrow of all patients, they can be obtained by a relatively simple procedure
9
of bone marrow aspiration, expanded in vitro with or without differentiation, and re-transplanted into the patient, thus eliminating the need for immunosuppressants [70]. The initial assumption, regarding the capability of bone marrow-derived stem cells to regenerate the heart by transdifferentiation into cardiomyocytes, has been challenged by a number of recent studied. Balsam et al. [149,150] and Murry et al. [151] demonstrated that the haematopoietic stem cells continued to differentiate along the haematopoietic lineage, suggesting the functional improvement observed may not be related to transdifferentiation into the cardiac lineage, but rather from indirect mechanisms. A considerable body of data indicates that a specific subset of bone marrowderived angioblasts, expressing endothelial precursor markers, is responsible for neovascularisation and angiogenesis [99,152– 157]. Kocher et al. [99] for example, demonstrated that an intravenous injection of human bone marrow donor cells to the infarcted myocardium of rats resulted in a significant increase in neovascularisation of post-infarction myocardial tissue, attenuation of cardiomyocyte apoptosis and left ventricular remodelling. The potential of bone marrow stem cells to heal a damaged heart by inducing vasculogenesis in the injured myocardium, thereby increasing heart viability and restoring cardiac function has promoted the studies on bone marrow stem cells quickly from small animals to clinical trials [115,158]. At present, the results of three medium size clinical trials (100–200 patients) show a variable and modest healing function of autologous bone marrow stem cells in cardiac function [159– 161]. The application of bone marrow cells for cardiac disease is still in its preliminary phase, as optimal cell type, delivery route, dose and timing require further optimisation. The application of bone marrow-derived stem cells is also limited by a safety issue: obtaining adequate autologous cells from a patient with myocardium infarction in time to prevent postinfraction remodelling may be difficult. In addition, the presence of stem cells for cardiomyocytes in other parts of the body, including bone marrow, has not been widely accepted yet. Caspi and Gepstein [115] have given an excellently tabulated overview on the clinical trial results of using bone marrow stem cells in the treatment of acute and chronic heart diseases. 2.3.3. Adipose-derived stem cells Human adipose tissue provides a uniquely abundant and accessible source of adult stem cells for applications in tissue engineering and regenerative medicine [162–165]. Adiposederived stem cells have the ability to differentiate along multiple lineage pathways. The cardiomyocyte phenotype from adipose-derived cells has been reported [107,166]. Animal trial with rats showed that adipose tissue-derived regenerative cells improved heart function following myocardial infarction [167]. 2.3.4. Native cardiac progenitor cells It had long been believed that the adult mammalian heart, a terminally differentiated organ, had no self-renewal potential. This notion about the adult heart, however, has been challenged by accumulated evidence that myocardium itself contains a
10
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
resident progenitor cell population capable of giving rise to new cardiomyocytes [168–172]. There are scientists who have hypothesised that cardiac progenitor stem cells reside in the hearts of neonatal animals, and that these progenitor stem cells (if any) could eventually serve as the basis for cardiac cell lineage formation and thus its application in the treatment of cardiac disease in humans. Recently, cardiac progenitor cells were found in the hearts of neonate humans, rats, and mice by a multi-institution group of the United States and German researchers [172]. Nonetheless, given the limited regeneration ability of the adult heart, it is apparent that the existence of the above mentioned cells within the adult heart do not translate to a functionally significant cardiac differentiation following myocardial infarction [115]. The role of these cells in the normal adult heart is still to be elucidated. They may represent an intrinsic repair system capable of replacing cells lost in the normal process of ageing, or may simply reflect remnant cells from organ development in early life. The existence of these progenitor cells will no doubt open new opportunities for myocardial repair, though many issues still need to be addressed. 2.3.5. Embryonic stem cells Embryonic stem cells are thought to have much greater translational potential than other stem cells because of their several advantages over other stem cells, in addition to the common features shared by all stem cells as mentioned above. First, they are pluripotent, which means they have a broader multilineage expressing profile. Unlike adult stem cells, which can differentiate to a relatively limited number of cell types, embryonic stem cells have the potential to contribute to all adult tissues. Second, they are robust. They have the long-term proliferation ability with a normal karyotype, and can be cryopreserved. Third, they can be genetically manipulated [173]. Hence, research using embryonic stem cells remains at the zenith of stem cell science. The embryos, from which embryonic stem cells are derived, are a hollow microscopic ball of cells called blastocyst. At this blastocyst stage, a group of cells begins to separate from the outer cell mass (or trophoblast) and forms the inner cell mass (ICM) (also called embryoblast) (Fig. 6). While the outside layer of cells of the blastocyst goes on to form the placenta, the inner cell mass forms the embryo and will ultimately develop into all the tissues in the body. Embryoblasts are, therefore, truly pluripotent. In 1981, the inner embryoblasts were isolated from mouse blastocysts and were successfully used to generate pluripotent stem cell lines, which were termed embryonic stem cells (ESC) [174,175]. However, the documentary world had not seen a human embryonic stem cell line until 1998 when two independent teams, Thomson et al. [94] and Shamblott et al. [176], described the generation of human ESC lines. The embryonic stem cell is capable of continuous proliferation and self-renewal in vitro but also retains the ability to differentiate into derivatives of all three germ layers both in vitro and in vivo. Thus, following cultivation in suspension, the ESCs tend to spontaneously create 3D
aggregates of differentiating tissue known as embryoid bodies (EBs) [177]. Upon aggregation, differentiation is initiated and the cells begin to a limited extent to recapitulate embryonic development. Though they cannot form trophectodermal tissue (which includes the placenta), cells of virtually every other type present in the organism can develop. The aggregate at first appears as a simple ball of cells, and then grow into an increasingly more complex appearance. After a few days a hollow ball (cystic embryoid body) forms, followed by the appearance of internal structures, such as a yolk sac and heart muscle cells (i.e. cardiomyocytes) which beat in a rhythmic pattern to circulate nutrients within the increasingly larger embryoid body. The availability of the embryonic stem cell system has boosted the hope of heart regeneration. The first study using embryonic stem cell as a source for cell transplantation into the myocardium was reported by Klug et al. [93]. By using genetically selected mouse embryonic stem cell-derived cardiomyocytes, Klug et al. showed that the differentiated cells developed myofibrils and gap junctions between adjacent cells and performed synchronous contractile activity in vitro for up to 7 weeks. This study proved the feasibility to guide an unlimited number of embryonic stem cells into cardiomyogenic cell linage and to utilise them for myocardial regeneration. Later studies [178–181], utilising the infarcted rat heart model, demonstrated that transplantation of differentiated mouse embryonic stem cell-derived cardiomyocytes can result in short- and long-term improvement of myocardial performance. Theoretically, the undifferentiated ESCs, which might be able to differentiate in vivo into cardiomyocytes in the host microenvironment containing cardiac-specific differentiation signalling, could improve the heart function. However, controversial results have been reported, regarding the in vivo differentiation and outcome of the transplantation of undifferentiated ESCs. Puceat’s group [182,183] showed that in infarcted myocardium, grafted stem cells differentiated into functional cardiomyocytes integrated with surrounding tissue, improving contractile performance. However, Nussbaum et al. [184] discovered that undifferentiated mouse ESCs consistently formed cardiac teratomas in nude or immunocompetent syngeneic mice, and that cardiac teratomas contained no more cardiomyocytes than hind-limb teratomas, suggesting lack of guided differentiation. Hence the authors concluded that undifferentiated ESCs did not differentiate toward a cardiomyocyte fate in either normal or infarcted hearts [184]. As regards immunogenicity of the transplantation of undifferentiated ESCs, Menard et al. [185,186] reported that cardiac committed mouse ESCs, which were transplanted to the infarcted sheep heart following incubation with BMP-2, differentiated to mature cardiomyocytes, and that cell transplantation resulted in a significant improvement in cardiac function independent of whether the sheep were immunosuppressed or not. However, another group [187] reported the increased immunogenicity (i.e. rejection) of mouse ESCs upon in vivo differentiation after transplantation into ischemic myocardium of allogeneic animals, implying that clinical transplantation of allogeneic ESCs or ESC derivatives for
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
treatment of cardiac failure might require immunosuppressive therapy. This result was confirmed by Nussbaum’s group [184], who found no evidence for allogeneic immune tolerance of cell derivatives. Hence, successful cardiac repair strategies involving ESCs will need to control cardiac differentiation, avoid introducing undifferentiated cells, and will likely require immune modulation to avoid rejection. 2.3.6. Human embryonic stem cells The vast biomedical potential of human ESCs has stirred enthusiasm in the field of tissue engineering. From human ESCs, scientists hope to grow replacement tissues for people with various diseases, including bone marrow for cancer patients, neurons for people with Alzheimer’s disease, pancreatic cells for people with diabetes, and cardiomyocytes for patients with heart damage. Furthermore, establishment of a tissue-specific differentiation system may have significant impact on the study of early human tissue differentiation, functional genomics, pharmacological testing, and cell therapy. Given these reasons, we think that the utilisation of human ESCs in cardiac tissue engineering is worthy of a separate section. An overview of the relevant reports is given in Table 3. After the establishment of human ESC lines in 1998 [94,176], other research groups have been able to develop a reproducible cardiomyocyte differentiation system from the human ESCs [138,188–191]. The detailed protocols can be found in these reports. It has been demonstrated that the human ESC-derived cardiomyocytes displayed structural properties of early-stage cardiomyocytes [138,188,189,192]. The presence cardiacspecific proteins and the absence of skeletal muscle markers have also been confirmed [192,193]. These studies also demonstrated the progressive maturation from an irregular myofilament distribution to a more mature (i.e. organized) sarcomeric pattern [138,192,193]. The human ESC-derived cardiomyocytes were also shown to display functional properties, consistent with an early-stage cardiac phenotype [138,189,190,194,195]. Functional improvement of heart following cell transplantation would require structural, electrophysiological, and mechanical coupling of donor cells to the existing network of host cardiomyocytes [115]. Hence, it is important to investigate whether cells derived from human ESCs can restore myocardial electromechanical properties. A study of Gepstein group [196,197] demonstrated tight electrophysiological coupling between the engrafted human ESCs and host rat (or swine) cardiomyocytes both in vitro and in vivo. In spite of the exciting potential of human ESCs, the cells are also giving rise to daunting legal and ethical concerns. ESCs are controversial because they are obtained from the destruction of a potential human embryo. Technologies for therapeutic and, especially, reproductive cloning add further ethical problems. In addition, a number of technical issues need to be addressed prior to clinical application [115], as discussed below. (1) Scale up. One of drawbacks of cell therapy is the difficulty in scaling up to meet larger production needs in clinical applications.
11
(2) Purification of the differentiating cardiomyocyte population is necessary. (3) In vivo delivery and efficiency of grafting. Cell-delivering techniques should be developed to enable proper alignment of the grafted tissue, high seeding rate of the transplanted cells, and minimal damage to the host tissue. (4) Immune rejection. Human ESCs (hESCs) are derived from cell lines genetically distinct to the potential human recipient. Therefore the potential for immune rejection following hESC transplantation into an immunocompetent adult recipient exists, as is observed with conventional solid organ transplantation. As hESCs are therefore allogenic, there is the potential requirement for co-administration of immunosuppression [198–203]. Encouragingly there is some evidence that hESCs have an immunopriviledged status. Under certain conditions both in vitro and in vivo it they display limited immunogenicity and are tolerated in cross species (xeno-) transplantation without immunosuppression [204,205]. Reduced cell surface expression of both major histocompatibility complex and accessory proteins is believed to underlie this immunopriviledged status. However this is mainly observed in undifferentiated hESCs, and with differentiation into specialised tissues such as cardiomyocytes hESCs may acquire a greater immunogenic phenotype. Strategies aimed at preventing immunological rejection of the cells, such as genetic modification or graftrecipient tissue type matching, should be explored. 2.4. Strategies to address immune rejection in cells As mention above, the major limitation of cell therapy is immune rejection associated with most used cell types. Successful application of tissue engineering in man will depend on the utilisation of an autologous or nonimmunogeneic cell source, as well as synthetic scaffold materials, to avoid life long immunosuppression. Several strategies aimed at achieving immunological tolerance are being developed. These strategies include [115]: (1) Establishing banks of major histocompatibility complex (MHC) antigen-typed human ESC lines. (2) Genetically altering the human ESC to suppress the immune response (e.g., by knocking out the major histocompatibility complexes). (3) The concept of haematopoietic chimerism. (4) Generating immune compatible ESC-derived cardiomyocytes with the patient’s own genetic information, known as somatic cell nuclear transplantation or therapeutic cloning. The successful application of nuclear transfer techniques to a range of mammalian species has brought the possibility of human therapeutic cloning significantly closer. The objective of therapeutic cloning is to produce pluripotent stem cells that carry the nuclear genome of the patient and then induce them to differentiate into replacement cells, such as cardiomyocytes to replace damaged heart tissue. In the process, a somatic cell nucleus from a patient is transferred into an enucleated oocyte
12
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
Table 3 Differentiation of human ESCs towards cardiomyocytes Method of hES differentiation
Major results
Reference
In vitro: via EBs in suspension
ESCs differentiated into cardiomyocytes, even after long-term culture. Upon differentiation, beating cells were observed after one week, increased in numbers with time, and retained contractility for >70 days. The beating cells expressed markers of cardiomyocytes ESCs showed consistence in phenotype with early-stage cardiomyocytes, and expression of several cardiac-specific genes and transcription factors ESCs showed a progressive ultrastructural development from an irregular myofibrillar distribution to an organized sarcomeric pattern at late stages ESC-derive myocytes at mid-stage development demonstrated the stable presences of functional receptors and signalling pathways, and the presence of cardiac-specific action potentials and ionic currents Tight electrophysiological coupling between the engrafted hESC-derived cardiomyocytes and rat cardiomyocytes was observed. The transplanted hES cell-derived cardiomyocytes paced the hearts of swine
Carpenter and coworkers [188]
In vitro: via EBs in suspension In vitro: via EBs in suspension In vitro: via EBs in suspension
In vitro: co-culture of differentiated rat cardiomyocyte and hESCderived cardiomyocytes; In vivo: transplantation of hESC-derived cardiomyocytes into swine In vitro: co-culture of undifferentiated hESC with mouse endoderm-like cells In vitro: via EBs in suspension In vitro: co-culture of hESC-derived cardiomyocytes with rat myocytes; In vivo: differentiated hESCderived cardiomyocytes transplanted into guinea pig In vivo: differentiated cardiacenriched hESC progeny
ESCs differentiated to beating muscle. Sarcomeric marker proteins, chronotropic responses, and ion channel expression and function were typical of cardiomyocytes. Electrophysiology demonstrated that most cells resembled human foetal ventricular cells ES cells differentiated into cardiomyocytes. Upon differentiation, beating cells were observed after 9 days, and retained contractility for longer than 6 months Electrically active, hESC-derived cardiomyocytes are capable of actively pacing quiescent, recipient, ventricular cardiomyocytes in vitro and ventricular myocardium in vivo
hESCs can form human myocardium in the rat heart, permitting studies of human myocardial development and physiology and supporting the feasibility of their use in myocardial repair
(i.e. its nucleus has been removed), as described in the cloning of Dolly the sheep [207]. The oocyte containing the new nucleus carries the genetic information of the patient. Using a tiny pulse of electricity to cause the new nucleus to fuse with the enucleated oocyte’s cytoplasm, this manipulated oocyte can develop in vitro into a blastocyst. From this blastocyst, embryonic stem cells with the genetics of the patient can be isolated and expanded in vitro, and then differentiate in vitro into genetically matched cardiomyocytes for transplantation. Obviously, cloning would eliminate the critical problem of immune incompatibility. 2.5. Strategies of cell delivery In cell-based therapy, isolated cell suspensions are directly injected into injured heart via the pericardium, coronary arteries, or endocardium. Direct injection of isolated cells avoids an open-heart surgery. However, it is difficult to control the location of the grafted cells in the transplantation. To address these problems, 2D and 3D cell delivery vehicles have been under development using biomaterials. These strategies are the core of the present review and will be discussed in great detail in Section 3. 2.6. Summary of cell-based therapy and their limitations Cell-based cardiac therapy represents an exciting strategy in, for example, heart tissue regeneration and heart tissue
Gepstein and coworkers [138] Snir et al. [193] Satin et al. [195]
Kehat et al. [196]
Mummery et al. [189]
Harding et al. [206] Xue et al. [197]
Laflamme et al. [192]
engineering. Huge efforts have been invested in the development of cell sources for myocardial regeneration, including foetal cardiomyocytes, skeletal myoblasts, bone marrow stem cells, adipose stem cells, endothelial progenitors, native cardiac progenitor cells, and embryonic stem cells (ESC) (Table 2). A large amount of studies has also been carried out to assess the roles of these potential cell types in the regeneration of myocardial tissue, both in vitro and in vivo. The translation of these basic scientific studies from bench to bedside has been progressing. Among these cell sources, embryonic stem cells possess the greatest potential because of their intrinsic pluripotency. These pluripotent cells can self-replicate tirelessly in the undifferentiated state in vitro and be induced to differentiate into cell derivatives of all three germ layers, including cardiomyocytes. The ability to generate human cardiac tissue in vitro using embryonic stem cells provides therefore the most exciting approach in the field of cardiac tissue regenerative medicine and tissue engineering. Despite the enormous potential of cell-based therapy, a number of technical issues need to be addressed prior to clinical application. Key technical issues include (1) scaling up of cells, (2) cell delivery, (3) efficiency of grafting, (4) suppression of alternative unwanted cell phenotypes when ES cells are applied, and (5) immune rejection. These issues, in particular those associated with the efficiency of cell delivery and grafting, could be potentially addressed by the strategies of tissue engineering involving biomaterials.
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
13
3. Biomaterials for myocardial tissue engineering
Table 5 Selected polymeric biomaterials for tissue engineering [79,211–214]
3.1. Overview of substrate development for tissue engineering
Biomaterial
Abbreviation
1. Synthetic polymers Bulk biodegradable polymers Aliphatic polyesters Poly(lactic acid) Poly(D-lactic acid) Poly(L-lactic acid) Poly(D,L-lactic acid) Poly(glycolic acid) Poly(lactic-co-glycolic acid) Poly(e-caprolactone) Poly(hydroxyalkanoate) Poly(3 or 4-hydroxybutyrate) Poly(3-hydroxyoctanoate) Poly(3-hydroxyvalerate) Poly ( p-dioxanone) Poly(propylene fumarate) Poly (1,3-trimethylene carbonate) Poly(glycerol-sebacate) Poly (ester urethane)
PLA PDLA PLLA PDLLA PGA PLGA PCL PHA PHB PHO PHV PPD or PDS PPF PTMC PGS PEU
The aim of this section is to provide an overview on the development of substrates (scaffolds, matrices) for tissue engineering, the principles of which also prevail in the development of myocardial tissue engineering matrices. 3.1.1. Criteria on tissue engineering substrates Being a very much fledgling discipline, tissue engineering encounters a variety of challenges, which can be grouped into three categories associated with the science and technology of cells, materials, and interaction between them. The challenges that the material scientists encounter are caused by the strict requirements on the tissue engineering substrates, as listed in Table 4. The first three requirements are directly associated with the chemical and physical properties of biomaterials, and the last three criteria are set for the fabrication technologies of substrates. The criterion list is not exhaustive. Specific requirements on a specific tissue engineering construct, such as osteoconductivity of bone tissue engineering scaffolds and electric conductivity for nerve tissue engineering, are not included. Some listed criteria might not be essential in certain approaches. Porosity, for example, is not essentially required in the heart patch approach (Fig. 4), although a porous heart patch can be used initially to grow vascular cells before seeding cardiac muscle cells. Biodegradability is not demanded in the treatment of congenital heart diseases either. 3.1.2. Polymers used in soft tissue engineering Polymers are the major type of materials used in soft tissue engineering. Selected biopolymers are listed in Table 5. They can be naturally occurring or synthetic. Detailed reviews are available in literature [79,211–214]. Table 4 Criteria for tissue engineering substrates [208–210] 1. Ability to deliver and foster cells The material should not only be biocompatible (i.e. nontoxic), but also foster cell attachment, differentiation, and proliferation 2. Biodegradability The composition of the material should lead biodegradation in vivo at rates appropriate to tissue regeneration 3. Mechanical properties The substrate should provide mechanical support to cells until sufficient new extracellular matrix is synthesised by cells 4. Porous structure The scaffold should have an interconnected porous structure for cell penetration, tissue ingrowth and vascularisation, and nutrient delivery 5. Fabrication The material should possess desired fabrication capability, e.g., being readily produced into irregular shapes of scaffolds that match the defects in bone of individual patients 6. Commercialisation The synthesis of the material and fabrication of the scaffold should be suitable for commercialisation
Surface bioerodible polymers Poly(ortho ester) Poly(anhydride) Poly(phosphazene) Polyurethane
POE PA PPHOS PU
Nondegradable polymers Poly(tetrafluoroethylene) Poly(ethylene terephthalate) Poly(propylene) Poly(methyl methacrylate) poly(N-isoproplylacrylamide
PTFE PET PP PMMA PNIPAAm
2. Natural degradable polymers Polysaccharides Hyaluronan Alginate Chitosan Starch
HyA
Proteins Collagen Gelatin Fibrin
3.1.2.1. Naturally occurring polymers. Natural extracellular matrices of soft tissues are composed of various collagens. It is not surprising that much research effort has been focused on naturally occurring polymers such as collagen [215,216] and chitosan [217] for tissue engineering applications. Theoretically, naturally occurring polymers should not cause foreign materials response when implanted in humans. They provide a natural substrate for cellular attachment, proliferation, and differentiation in its native state. For the above-mentioned reasons, natural occurring polymers could be a favourite substrate for tissue engineering [211,218]. Table 6 presents major naturally occurring polymers, their sources and applications. Among them, collagen, fibrin, gelatin and alginate have been extensively investigated for myocardial tissue engineering [24,64]. Each of these polymers will be reviewed in detail in Section 3.2, in terms of their applications in myocardial tissue engineering.
14
Q.-Z. Chen et al. / Materials Science and Engineering R 59 (2008) 1–37
Table 6 List of naturally occurring polymers, their sources and main application fields [219] Polymer
Source
Main application fields
Collagen Collagen-GAG (alginate) copolymers Albumin Hyaluronic acid
Tendons and ligament
Multi-applications, including cardiac tissue engineering Artificial skin grafts for skin replacement
In blood In the ECM of all higher animals
Transporting protein, used as coating to form a thromboresistant surface An important starting material for preparation of new biocompatible and biodegradable polymers that have applications in drug delivery and tissue engineering, including cardiac tissue engineering Multi-applications, including cardiac tissue engineering Multi-applications, including cardiac tissue engineering
Fibrinogen-fibrin Gelatin Chitosan MatrigelTM (gelatinous protein mixture) Alginate Polyhydroxyalkanoates
Purified from plasma in blood Extracted from the collagen inside animals’ connective tissue Shells of shrimp and crabs Mouse tumor cells
Multi-applications, including cardiac tissue engineering Myocardial tissue regeneration
Abundant in the cell walls of brown algae By fermentation
Multi-applications, including cardiac tissue engineering Cardiovascular and bone tissue engineering
3.1.2.2. Synthetic polymers. Although naturally occurring polymers possess the above-mentioned advantages, their poor mechanical properties and variable physical properties with different sources of the protein matrices have hampered the progress with these approaches. Concerns have also arisen regarding immunogenic problems associated with the introduction of foreign collagen [220]. Following the developmental efforts using naturally occurring polymers as scaffold materials, much attention has been paid to synthetic polymers. Synthetic polymers are essential materials for tissue engineering not only due to their excellent processing characteristics, which can ensure the offthe-shelf availability; but also because of their advantage of being biocompatible and biodegradable [220,221]. Synthetic polymers have predictable and reproducible mechanical and physical properties (e.g., tensile strength, elastic modulus, and degradation rate), and they can be manufactured with great precision. Although they are unfamiliar to cells and many suffer some shortcomings, such as eliciting persistent inflammatory reactions, being eroded, not be compliant or able to integrate with host tissues, they may be replaced in vivo in a timely
fashion by native extracellular matrices built by the cells seeded into them. It has become widely realised that an ideal tissue engineered substitute should be made from a synthetic scaffold. Table 5 has listed most synthetic polymers used in tissue engineering. Table 7 provides selected properties of synthetic, biocompatible and biodegradable polymers that have been intensively investigated as substrate materials for tissue engineering, type I collagen fibres being included for comparison. Among them, aliphatic polyesters (PLA, PGA, and PCL) have widely been applied as scaffolding materials for 3D tissue engineering constructs, an approach listed in Table 1. Degradability is generally a desired characteristic in tissue engineering substrates because the second surgery to remove them (such as a heart patch) would be averted if the substrate could be removed by the physiological system of the host body. The biodegradable suture is a successful example of the applications of biodegradable polymers. Moreover, in many applications 3D tissue engineering constructs cannot be removed, and the non-degradable biomaterials would act as barriers to new tissue ingrowth and blood flow. Nonetheless, a few nondegradable polymers, including PTFE and PET, have
Table 7 Selected properties of synthetic, biodegradable polymers investigated as scaffold materials Polymers
Melting point, Tm (8C)
1. Bulk degradable polymers PDLLA Amorphous PLLA 173–178 PGA 225–230 PLGA Amorphous PPF
