**1. Introduction**

It is well-acknowledged that significant differences in hearts exist between human and model organisms. These morphological and physiological differences can lead to complex problems, such as low pathological reproducibility in clinical practice [1]. On the other hand, human pluripotent stem cells (hPSCs), including human embryonic stem cells (hESCs) and human-induced pluripotent stem cells (hiPSCs), benefitting from the property of indefinite proliferation in vitro and the capacity to differentiate into different types of somatic cells, are a promising tool in biomedical applications. Compared with the transdifferentiation of human somatic cells, the differentiation of hPSCs seems to be more efficient in terms of productivity, safety, and cost. The rapid development of hPSC research in the past few decades has made it possible to utilize hPSC-derived cardiomyocytes in large-scale cardiac tissue engineering directly.

Several effective protocols have been successfully developed to induce hPSCs to become cardiomyocytes. Embryoid body (EB) was first used in the differentiation of cardiomyocytes from hESCs and hiPSCs; however, its effectiveness and reproducibility were found to be problematic because of serum quality instabilities and heterogeneous EB sizes [2]. Differentiation protocols developed using serum-free and compound-defined media were subsequently used and improved the efficiency and reproducibility of cardiomyocytes generated from EBs. Engineering approaches to the production of homogeneous EBs emerged several years ago [3,4]. This method produced more homogenously sized EBs compared with conventional methods that used 96-well plates, and it was also appropriate to scale-up. Because of the limitations of EB protocols, monolayer three-dimensional (3D) approaches have drawn increasing attention over the past several years. Uniform hESC colonies were plated on Matrigel via a microcontact approach and the size range was optimized for maximizing mesoderm formation and cardiac induction. The method of activation of canonical Wnt signaling by the glycogen

synthase kinase-3 (GSK-3) inhibitor (CHIR99021) followed by inhibition of Wnt signaling via Inhibitor of Wnt Production 2 (IWP2) or Inhibitor of Wnt Production 4 (IWP4) were found to be sufficient to produce numerous functional cardiomyocytes from multiple human pluripotent stem cell lines in two weeks without exogenous growth factors or genetic manipulation in adherent culture or suspension culture system [5–7]. Paul et al. designed a cardiomyocytes differentiation strategy by using a medium including three components: RPMI-1640, L-ascorbic acid 2-phosphate, and rice-derived recombinant human albumin [8]. It is essential for cardiac development in vitro through an appropriate addition of different growth factors, including fibroblast growth factor-2 (FGF-2), transforming growth factor-β (TGF-β), superfamily growth factors activin A, bone morphogenetic protein-4 (BMP-4), vascular endothelial growth factor (VEGF), and dickkopf WNT signaling pathway inhibitor 1 (DKK-1). All of these factors were found to assist human pluripotent stem cells generate myocardial precursor cells and cardiomyocytes when added in order [9].

In clinical practice, patient-derived hiPSC-CMs are an optimal disease model for personalized medicine involving inherited cardiac diseases and stem cell therapies to repair or replace injured heart tissues. hiPSC-CMs can be used to model several heart diseases, including Duchenne muscular dystrophy [10], Leopard syndrome [11], long QT syndrome [12], Timothy syndrome [13], Fabry disease [14], Danon disease [15], and familial hypertrophic cardiomyopathy [16]. In addition, hiPSC-CMs from mitochondrial cardiomyopathy of Barth syndrome (BTHS) have been used to generate a platform for pathogenesis and medical therapeutics. This cardiomyopathy model shows irregular sarcomeres, abnormal myocardial contraction, and defective heart function; more importantly, it mimics mitochondrial functional impairment caused by a mature cardiolipin defect [17]. Masahide et al. constructed a Torsade de Pointes (TdP) arrhythmias model from hiPSCs to mimic a patient's disease condition and provide a chance to study the mechanisms of TdP generation and develop an anti-arrhythmias drug test [18]. Overexpression of CDK1, CDK4, cyclin B1, and cyclin D1 efficiently induced cell cycle progress in at least 15% of post-mitotic murine and human cardiomyocytes [19]. Nutlin-3a can selectively activate the p53 signaling pathway and induce cell apoptosis of DNA-damaged iPSCs except for DNA-damage-free cells. These iPSC-CMs were engrafted into an ischemic mouse heart to enhance mouse cardiac beating [20]. This technology may bring about potential benefits for patients with a cardiac disease in clinical medicine.

However, evidence indicates that cardiomyocytes differentiated by these approaches are not as mature as an adult phenotype, thus they may not be able to reflect the physiological response of the adult heart accurately. In addition, with respect to cardiac tissue engineering, cardiomyocytes that more closely resemble those of the native myocardium would contribute more to myocardial repair. For example, as cardiovascular diseases predominantly occur in elderly humans, immature hPSC-CMs may cause modeling to be imprecise and futile [21]. In this review, we discuss the state of current approaches to obtaining more mature cardiomyocytes.
