Current status and future prospects of patient-derived induced pluripotent stem cells

Induced pluripotent stem cells (iPSCs) are produced from adult somatic cells through reprogramming, which behave like embryonic stem cells (ESCs) but avoiding the controversial ethical issues from destruction of embryos. Since the first discovery in 2006 of four factors that are essential for maintaining the basic characteristics of ESC, global researches have rapidly improved the techniques for generating iPSCs. In this paper, we review new insights into patient-specific iPSC and summarize selected “disease-in-a-dish” examples that model the genetic and epigenetic variations of human diseases. Although more researches need to be done, studies have increasingly focused on the potential utility of iPSCs. The usability of iPSC technology is changing the fields of disease modeling and precision treatment. Aside from its potential use in regenerative cellular therapy for degenerative diseases, iPSC offers a range of new opportunities for the study of genetic human disorders, particularly, rare diseases. We believe that this rapidly moving field promises many more developments that will benefit modern medicine.


Introduction
Induced pluripotent stem cells (iPSCs), which were first described in 2006 by Takahashi and Yamanaka [1], can replicate and differentiate into all kinds of somatic cell types. Importantly, iPSCs bypass the ethical concerns related to embryonic stem cells (ESCs). The iPSC approach not only facilitates the stem cell development and regenerative medicine studies, but also creates fantastic "disease-in-adish" biological models. Primary cells can be obtained from patients, reprogrammed to become pluripotent, and then induced into specific differentiated cells with the mutant gene of the patient's own genomic background. Thus, disease-specific phenotypes can be identified in these "diseasein-a-dish" models, and various potential therapies can then be tested. Owing to these characteristics, patient-specific iPSCs hold broad prospects in biomedical research, and there is an ever-increasing interest in making full use of these cells. Here, we present a brief overview of patientspecific iPSCs and highlight a few major recent studies in the application of these cells (Fig. 1). We hope that this review can offer a taste of this rapidly advancing field.

Development of reprogramming methods
The iPSC technique was first discovered in mouse fibroblasts, then implemented in human fibroblasts [1,2], using four reprogramming factors: Oct3/4, Klf4, Sox2, and c-Myc (later dubbed "Yamanaka factors"). In response to the pioneering report on iPSCs, significant global efforts have focused on reproducing and extending this work (Fig. 2). Transcriptional profiling analysis has revealed hundreds of genes correlated with ESC characteristics [3], among which Oct4 and Sox2 are the critical regulators of the early development and maintenance of ESC identity. In 2007, Yu et al. [4] successfully reprogrammed human embryonic, neonatal, and adult fibroblasts into iPSCs with a slightly different cocktail of transcription factors (Lin28, Nanog, Oct4, and Sox2) via lentivirus, in which Klf4 and c-Myc were replaced by Lin28 and Nanog. The reprogramming efficiency with this system was 0.02%, which was close to Fig. 1 The diagram of iPSC reprogramming and applications. Somatic cells from different patients can be reprogrammed into iPSCs to obtain pluripotency, and iPSCs can be induced to differentiate into corresponding cells for further treatment or research. Those iPSCs with genetic defects can be corrected by gene editing for next exploration Yamanaka's report of their first human iPSC reprogramming study (0.01-0.02%).
More recently, researchers have developed a strategy for packaging Yamanaka factors in a single lentiviral vector [5]. Benefit from the easier infection of specific cells with a single virus than with a combination of multiple viruses, the single-vector system significantly improved the efficiency of iPSCs generation (at least tenfold higher than primary methods according to various reports from different research groups). The first-generation iPSC techniques by Yamanaka and others were conducted using retroviral or lentiviral systems, which ensures efficient reprogramming. However, the transgene sequences in this system were stably integrated into the iPSC genome, thus making this method unsuitable for clinical application.
Subsequently, researchers tried to introduce these transcription factors into a donor somatic cell by non-integrative approaches using proteins or small molecules ( Table 1). One of the first non-integrative iPSC techniques involved the use of adenoviruses encoding OKSM (Oct3/4, Klf4, Sox2, and c-Myc) reprogramming factors [6]. Meanwhile, nonintegrating episomal DNA plasmids have also been adopted [7]. These methods are known as second-generation iPSC technologies, which are transgene-free and considered more desirable for clinical applications. However, these methods suffer from low efficiency (0.0001-0.001%), repetitive induction, insufficient excision of integrated vectors and laborious operation. Sendai virus (SeV), a single chain RNA virus that does not integrate into the host genome, were then applied to generate human iPSCs. The virus remains in the cytoplasm and could be diluted out of the host cells ∼10 passages after virus infection. SeV reprogramming vectors have been used to successfully reprogram neonatal and adult fibroblasts as well as blood cells with high efficiency [8]. The main concern is the remnants of the virus. More recently, direct delivery of mRNA or microRNA, which eliminates the use of DNA and viruses, has been adopted to optimize human iPSC reprogramming [9]. The engineered single RNA strand contains multiple reprogramming factors and is conveniently delivered into cells using a single transfection step, without any viral intermediates. Thus, the resulting iPSCs avoid the risk of insertional mutagenesis and possess increased chromosomal stability and reduced clonal variation.

Cell sources for iPSCs
Scientists have attempted to generate iPSCs from various somatic cell types. In general, every live somatic cell type can be applied for iPSC reprogramming, and there is not currently a consensus on the preferred somatic cell source for producing iPSCs.

Fibroblasts
Fibroblasts are the most common cells in connective tissue and the first somatic cell type that was used to generate iPSCs, by both Yamanaka and Thomson in their 2007 reports [2,4]. Since then, researchers have successfully induced iPSCs from various types of fibroblasts, including adult human skin fibroblasts, newborn baby foreskin fibroblasts, fetal lung fibroblasts, as well as human fibroblasts from ESCs and mesenchymal stem cells [10,11]. To date, fibroblasts are still the most commonly used primary cell source, as they are easy to handle and are the most well studied cells in many research fields.

Blood cells
A more desirable source for iPSC induction is human peripheral blood, which is obtainable through a routine, minimally invasive clinical procedure. Studies found that iPSCs can be efficiently derived from human CD34 + blood cells and T lymphocytes from as little as 1 mL of whole blood [12,13]. More attractively, the establishment of this reprograming protocol using peripheral blood cells enables the creation of iPSCs from the large number of samples stored in the refrigerator. However, compared to other somatic cells, the reprogramming efficiency for blood cells is much lower [14]. In addition, studies have proved that human peripheral T cells-derived iPSC maintains T cell receptor (TCR) gene rearrangement of the original T cell. This epigenetic memory enables the differentiation and expansion of antigen-specific T cells from these iPSC clones; on the other hand, the TCR affinity may cause alloreactivity against the recipient's normal tissue/ cells. Taken together, blood cells-derived iPSC clones are normally very heterogeneous, and the epigenetic memory issues must be resolved before clinical application [15].

Urine cells
Voided urine contains a variety of cells from the kidney and urinary tract, providing a good cell source for iPSC reprogramming. Urine cells possess the advantages of easy access, low cost, high efficiency, and bypassing ethical issues. Moreover, it has been reported that urinary iPSCs exhibit faster reprogramming dynamics compared with human skin fibroblasts or adult-fat-derived mesenchymal stem cells [16]. The yield of iPSC colonies is generally high up to 4% using retroviral delivery of exogenous factors [17].

Keratinocytes
Keratinocytes are the primary cell type found in the epidermis, constituting 90% of human epidermal skin cells. However, skin keratinocytes are difficult to obtain. Thus, attention has been paid to those keratinocytes present in hair follicles. With this method, donors do not have to suffer from clinical or surgical investigations. Once plucked, hair samples can be stored in normal DMEM medium for several days at room temperature, without losing the proliferative ability of keratinocytes [18]. Moreover, compared to human skin fibroblasts, reprogramming of hair keratinocytes by retroviral transduction using OKSM cocktails is at least 100-fold more efficient [19].

Adipocytes
Adipose stem cells (ASCs) are localized in the stromalvascular portion of subcutaneous adipose tissue, which is a heterogenous set of mesenchymal cells [20]. Researches have demonstrated these cells not only can enhance tissue regeneration potentiality but also can differentiate into osteogenic, adipogenic, and chondrogenic lineages upon inductions in vivo [21]. More importantly, the ASCs can be readily extracted from the adipose tissue of human adults via lipoaspirate with large quantities. Thus, the ASCs have become one of the most popular cells in the field of stem cell studies and regeneration medicine, as well as studies of iPSCs.

Others
In addition to the cell sources mentioned above, mesenchymal stem cells, hematopoietic stem cells (HSCs), prostate cells, melanocytes, hepatocytes, pancreatic β cells, and neural cells, among others, have also been used as reliable cell sources for iPSC reprogramming (Table 2). However, to prevent further discomfort or risk for patients, especially medically fragile patients who have suffered traumatic medical events, mature somatic cell sources that are noninvasive or minimally invasive, reproducible, simple, and easily accessible are preferred for broader clinical use. Thus, the most common somatic cell sources for induced iPSCs are mainly blood samples, urine samples, and skin biopsies.

Regenerative medicine and cell replacement therapy based on iPSCs
ESCs have been studied for a long time and bring hope to the treatment of many incurable diseases. Up to now, more than twenty ESCs-based clinical trials can be found in the U.S. National Library of Medicine (Clinicaltrials.gov), mainly concentrated in the following directions including neural disease, heart failure, diabetes, eye disease, etc. (Table 3). However, it has been hampered by the restricted availability of cell sources and ethical concerns. The emergence of iPSCs avoids the ethical problems and cell shortage that may arise from the use of ESCs. The advantages of iPSC have ushered in hope for a treatment and cure of a variety of degenerative diseases, congenital defects, and even injuries. So far, ten clinical trials for cell therapies using iPSCs are underway, as found in the UMIN Clinical Trials Registry, NIH National Library of Medicine, Japan Registry of Clinical Trials and JMA Clinical Trials Registry (Table 3) [22], which is promising and good news for patients suffering from diseases and disorders.

Heart diseases
Heart diseases, principally cardiac arrhythmia, cardiomyopathy, and myocardial infarction (MI), are the leading cause of global mortality and a major contributor to disability. Regenerative strategies have become attractive because endogenous regeneration is limited in human. Numerous procedures have been effectively advanced to stimulate produce cardiomyocytes from ESCs in vitro. Moreover, a series of researches based on human ESC-derived cardiomyocytes (ESC-CMs) have proved that ESC-CMs could not only enhance function of infarcted hearts in small animal models but also restore function of infarcted hearts in nonhuman primate models with server MI at a clinical scale [23][24][25][26]. iPSCs have characteristics similar to ESCs, and maintain patient-specific genomic and transcriptomic information, making it more promising to disease modeling and regenerative strategies based on personalized medicine. In early studies, scientists found that iPSC-derived cardiomyocytes (iPSC-CMs) could not accurately replicate the native function in animal models due to the complexity of structure and microenvironmental factors. To this end, groups of researchers applied various approaches to promote the iPSC-CMs development and maturation, and finally proved functional benefit after they transplanted these cells into rodent models of acute myocardial infarction (AMI) [27][28][29]. However, pre-clinical studies in large animal models of MI are necessary to fully evaluate the therapeutic potential of this approach. Indeed, Ye et al. [30] demonstrated that transplantation of human iPSC-derived endothelial cells and smooth muscle cells provide synergistic effects on cardiac function in a porcine model of ischemia reperfusion; improvements in perfusion, wall stress, and cardiac performance were observed. Weinberger et al. [31] established an engineered 3D heart tissue using human iPSC-derived cardiomyocytes and endothelial cells, and proved its therapeutic effects in a guinea pig model with human-like cardiac physiology. Gao et al. [27] developed human cardiac muscle patches with   [32] demonstrated that allogeneic transplantation of iPSC-CMs is sufficient to regenerate the infarcted heart using Filipino cynomolgus monkeys as a model system. This study represented one important step towards translation of iPSC-based cardiac regeneration to the clinic. Currently, several clinical studies were permitted to treat people who have heart disease with of pluripotent stem cell (Table 3).

Neurological diseases
Neurodegeneration leading to Parkinson's disease (PD), Alzheimer's disease (AD), and Huntington's disease (HD) has become a major health burden worldwide. Current treatments mainly target the control of symptoms, there are no therapeutics available in clinical practice to prevent neurodegeneration or induce neuronal repair. Recently developed protocols have allowed induction of iPSCs to differentiate into dopaminergic neurons with ventral midbrain identity [33]. In 2013, Morizane et al. [34] proved in a non-human primate model that autologous transplantation of iPSCderived neuron cells is advantageous for minimizing the immune response in the brain compared with allogeneic grafts. Then, Hallett et al. [35] proved that transplantation of iPSC-derived midbrain dopaminergic neurons into a cynomolgus monkey model of PD provides a gradual functional motor improvement contralateral to the side of dopamine neuron transplantation, increasing motor activity without need for immunosuppression. Yoon et al. [36] transplanted human iPSC-derived neural progenitor cells (NPCs) into model rats with spinal cord injury (SCI) and showed significant behavioral improvements for up to 12 weeks in these received cell transplants according to various tests, including the staircase, rotarod, stepping, apomorphine-induced rotation, and cylinder tests. After sacrificing the animals, the authors also found that the transplanted neural progenitor cells had partially replaced the lost neurons and reconstituted the damaged neuronal connections. Lately, accumulating evidence suggested the function recovery after transplantation of human iPSC-derived dopaminergic progenitor cells in mouse, rat and no-human primate models of PD [37][38][39], and showed that MHC (major histocompatibility complex) matching improves engraftment of iPSC-derived neurons [40]. In addition to replacing damaged cells, certain diseases caused by congenital dysplasia of functional cells, such as congenital hypomyelination, can also be treated with iPSCderived associated cells. Neonatal engraftment by oligodendrocyte progenitor cells (OPCs) permits the myelination of congenitally dysmyelinated brain [41]. Neonatally engrafted human iPSC-OPCs robustly myelinated the brains of myelin-deficient shiverer mice and substantially increased their survival. The speed and efficiency of myelination by iPSC-OPCs are higher than that previously observed using fetal tissue-derived OPCs, and no tumors from these grafts were noted as long as nine months post-transplantation. These results proved that the utility of human iPSC-OPCs in treating disorders associated with myelin loss. Apart from the basic research on neural diseases, clinical research studies have also been carried out using iPSCs. A clinical trial started in Japan in 2018 uses allogeneic human iPSC-derived dopaminergic neuronal precursor cells for patients with PD (UMIN000033564, JMA-IIA00384). In addition, clinical trials of ESC-based cellular therapy for PD are also underway in Australia (NCT02452723) and China (NCT03119636, ChineseASZQ-003) [33].

Diabetes
Type 1 diabetes (T1D), a very common disease, is caused by the damage of pancreatic β cells, which prevents them from properly producing and secreting insulin. T1D can be effectively treated by cellular therapy utilizing allogeneic islet transplantation from an organ donor postmortem. Owing to the advantages of iPSCs, generating pancreatic β cells from iPSCs for T1D treatment is becoming increasingly feasible. The directed differentiation of pancreatic lineage cells from iPSCs has been vigorously studied. iPSCs differentiate through multiple developmental stages, such as definitive endoderm, primitive gut tube, posterior foregut, pancreatic endoderm and endocrine precursors, into insulin-expressing β cells. In 2014, Pagliuca et al. [42] first demonstrated the in vitro generation of glucose-responsive β cells from human iPSCs. These generated cells express mature β cell markers and secrete insulin in response to multiple sequential glucose challenges. Furthermore, after transplantation into a diabetes mouse model, these iPSC-derived β cells could ameliorate hyperglycemia of model mice. In 2016, Millman et al. [43] generated iPSC-derived β cells from a type 1 diabetes patient and showed these cells function as normal β cells following transplantation to model mice. In 2020, Yoshihara et al. [44] reported that human islet-like organoids from iPSCs contain endocrine-like cell types that, upon transplantation, rapidly reestablish glucose homeostasis in diabetic NOD/ SCID mice.

Rare diseases
The discovery and manufacture of iPSC-based therapies hold great promise for developing effective treatments for many rare diseases. For instance, Takahashi's group conducted the first clinical study involving an iPSC derivative in 2014, using a sheet of autologous iPSC-derived retinal pigment epithelial (RPE) cells in a patient suffering from wet-type age-related macular degeneration (AMD) [45]. According to their reports, the transplanted iPSC-RPE was intact, and there was no sign of immune rejection during the 1-year monitoring period after transplantation. And besides, a team from the National Eye Institute of the U.S. National Institutes of Health lead by Dr. Bharti is working on promoting another clinical trial for treating dry-type AMD patients with autologous iPSC-derived RPE cells.
Limb-girdle muscular dystrophy 2D (LGMD2D) is a hereditary muscle disorder caused by genetic defects in α-sarcoglycan gene. Tedesco et al. [46] generated LGMD2D patient-specific iPSCs using fibroblasts and myoblasts, which were replicated and genetically corrected. More attractively, these genetically corrected human iPSCderived mesoangioblasts generated healthy muscle fibers that expressed normal levels of α-sarcoglycan when they were transplanted back to a mouse model of LGMD2D.

Patient-specific iPSCs: "disease-in-a-dish"
In addition to being a potentially unlimited cell source for regenerative medicine or cell replacement therapy, the derivation of iPSCs offers new opportunities for basic research, providing a disease model as well as drug discovery tool, especially for individuals suffering from genetic syndromes. Researchers can reprogram a patient's somatic cells to iPSCs, then induce them to differentiate into disease-related cell types (Table 4). Thus, the iPSC-derived cells are genetically matched to the person from whom they were derived, making them ideally suited for the study of diseases with a strong underlying genetic cause. Of note, it is crucial to choose the right control in phenotypic analysis of monogenic inherited diseases. Normally, iPSC controls were generated from the cell of healthy family-members to get rid of the influence of different genetic backgrounds. In recent years, CRISPR/ Cas9-based gene editing emerged as a better choose which enables the preparation of an isogenic control by normalizing a disease-relevant mutation in patient-specific iPSCs or by inducing the mutation in wild-type iPSCs, so that diseased and control iPSCs with the same genetic background are obtained.

Heart diseases
The generation of iPSCs from patients with familial cardiomyopathies has already put new molecular insights into the pathogenesis of these disorders. Long QT syndrome (LQTS) is a potentially severe arrhythmogenic disorder associated with a prolonged QT interval and sudden death. It is caused by mutations in key genes regulating cardiac electrophysiology. In 2010, LQTS patient-specific iPSCs were established for the first time [47], allowing scientists to conduct in vitro studies of the disease traits using the iPSC-CMs. iPSC-based studies of LQTS1, LQTS2, and LQTS3 have been reported, and the key mutations corresponding to the respective types of LQTS were identified.
In 2012, Sun et al. [48] reported a model for familial dilated cardiomyopathy, which is the most common type of cardiomyopathy. iPSC lines were generated from two patients bearing the R173W mutation in TNNT2, and from unaffected family members as controls. The mutant iPSC-CMs exhibited abnormal calcium handling, reduced contractility, and myofibrillar disarray. Using these iPSC-CMs, authors from same group revealed that epigenetic activation of PDE2/3 is a key molecular event during the pathogenesis of Dilated Cardiomyopathy (DCM). Hinson et al. [49] established iPSCs with either truncating or missense mutations in TTN, then induced them to differentiate into iPSC-CMs and assembled them into cardiac microtissues. They found that these patient-specific iPSC-derived cells and microtissues displayed obvious defects, including sarcomere insufficiency and impaired responses to mechanical and β-adrenergic stress. Moreover, patient-specific iPSC-CMs also have been used in drug screening, which can elucidate individual propensities toward drug-induced cardiotoxicity.

Neurological diseases
Neurodegenerative diseases (NDDs) are the leading causes of death worldwide, yet no disease-modifying therapies exist. NDDs include AD, HD, frontotemporal dementia (FTD), PD, and amyotrophic lateral sclerosis (ALS). Each disease is characterized by the dysfunction and death of a specific subtype of neurons, but the mechanisms are not fully understood. In recent years, increasing efforts have been dedicated to generating different kinds of neural cells from patient-specific iPSCs. These cells provide a valuable tool to model specific molecular phenotypes of neurodegenerative diseases in vitro, which may significantly benefit the better understanding of disease mechanisms, pathology, and progression, as well as development of effective therapies.
Various protocols have been developed to differentiate iPSCs to neural cells. Abud et al. [50] reported the effective and robust generation of human iPSC microglial-like cells (iMGLs) resembling fetal and adult microglia, demonstrating their utility in investigating neurological diseases like AD. HD is a devastating autosomal-dominant inherited disorder marked by the progressive loss of medium spiny neurons and corticostriatal connections in the brain. Mehta et al. [51] differentiated iPSCs from an HD patient into functional cortical neurons that display altered transcriptomes, as well as morphological and functional phenotypes indicative of altered corticogenesis in HD. Almeida et al. [52] [120] demonstrated that key neuropathological features of FTD with GGG GCC repeat expansions can be recapitulated in iPSC-derived human neurons and also suggested that compromised autophagy may represent a novel underlying pathogenic mechanism. Nguyen et al. [53] found that LRRK2 mutant iPSC-derived dopaminergic neurons demonstrate increased susceptibility to oxidative stress. Kim et al. [54] found that 3D midbrain organoids from LRRK2 mutant iPSCs recapitulated the pathological features that observed in LRRK2-associated sporadic PD patients. Fujimori et al. [55] established iPSC-based cellular models of sporadic ALS (SALS), and confirmed that ropinirole could be selected as a potential therapeutic candidate. However, there is a need for more suitable protocols to generate ageand region-specific neuronal cells and, also, improved differentiation efficiency.

Diabetes
T1D is a disorder characterized by the loss of pancreatic β cell mass and β cell function, leading to insulin deficiency. Hua et al. [56] showed that pancreatic β cells derived from iPSCs with a gene-encoding glucokinase (GCK) mutation show more phenotypes of maturity-onset diabetes of the young type 2 (MODY2), compared to the healthy control. Burkart et al. [57] compared mitochondrial metabolism in iPSCs from five healthy individuals and four type 2 diabetes (T2D) patients with genetic insulin resistance. Hosokawa et al. [58] generated β-like cells from iPSCs derived from patients with fulminant type 1 diabetes. These cells showed upregulated expression of apoptotic markers after cytokine treatment. Additionally, expression changes were recorded for several apoptosis-and immunoregulation-related genes. In general, these iPSC models should allow for further understanding of the pathophysiology and the establishment of therapeutic methods for diabetes.

Liver diseases
The liver is a complex organ, being structurally and functionally heterogeneous and comprising different cell types, including epithelial cells, Kupffer cells, circulating monocytes, mesenchymal cells, and sinusoidal endothelial cells. It has been a challenging task for understanding the molecular and cellular mechanisms of liver development and pathogenesis. Recent technological advances in the stem cell field have shown the potentiality of iPSCs to model liver diseases in vitro. Studies confirmed the successful differentiation of patient-specific iPSCs into hepatocytes, cholangiocytes, and nonparenchymal cells, and researchers are now modeling genetic and complex liver diseases. Rashid et al. [59] generated iPSCs from patients with inherited metabolic conditions, including alpha1-antitrypsin deficiency, familial hypercholesterolemia, and glycogen storage disease type 1a, and induced them to differentiate into mature hepatocytes. They proved that these iPSC-derived cells recapitulate vital pathological features, such as abnormal folding of alpha1-antitrypsin, low density LDL receptor-mediated cholesterol deficiency, and elevated lipid and glycogen accumulation. Guan et al. [60] developed an in vitro model system where Alagille syndrome (ALGS) iPSCs differentiate into 3D human hepatic organoids consisting of hepatocytes and cholangiocytes, through stages that resemble the human liver during its embryonic development under the influence of different JAG1 mutations.

Rare diseases
Rare diseases include a wide variety of often-serious diseases suffered by a large number of patients. Almost all rare diseases are genetic diseases, which usually mean patients face lower chances of a complete cure. However, thanks to the specificity of iPSC technology, anyone suffering from a rare disease can provide cells to generate individualized iPSCs, no matter how many other patients share the same disease, for subsequent research. With the help of iPSC technology, we could unveil the mechanisms of rare diseases one by one. For instance, LGMD2D, which is caused by mutations in the gene-encoding α-sarcoglycan, is a muscle disease that has been neglected by research. Tedesco et al. [46] reprogrammed fibroblasts and myoblasts from LGMD2D patients to generate iPSCs for deriving mesoangioblasts, which were replicated and genetically corrected in vitro with a lentiviral vector carrying the gene-encoding human α-sarcoglycan and a promoter that would ensure expression only in striated muscle. When these genetically corrected human iPSC-derived mesoangioblasts were transplanted into a mouse model of LGMD2D, they generated muscle fibers that expressed α-sarcoglycan. Finally, the authors proved that transplantation of mouse iPSC-derived mesoangioblasts into LGMD2D mice resulted in functional amelioration of the dystrophic phenotype and restoration of the depleted progenitors.

Challenges and future perspectives
The introduction of iPSCs has greatly enhanced the possibility and potential of personalized cell therapy and provided new approaches to regenerative medicine, disease modeling, and drug testing. The iPSC technology has evolved rapidly since its conception, and the scientific community is joining efforts to push iPSC studies into clinical application. The first clinical study based on iPSC-derived retinal pigmented epithelium was authorized and conducted at the RIKEN institute in Japan in 2014. Since then, several clinical studies based on allogeneic iPSCs have been developed and approved, mostly nationally approved in Japan, with indications including PD, AMD, severe cardiac failure, aplastic anemia, and SCI [22]. Nevertheless, iPSCs derived from adult human cells still face ethical and scientific hurdles. Except for debates focused on the embryo problem, many ethical issues are applied to both ESCs and iPSCs. For instance, there is the possibility of introducing these infinitely stem cells into an embryo. Another issue is genetic privacy, since the iPSCs derived from any individual inherently contain the private information (DNA). The solution is not to condemn the technology but rather to enact logical, universal policy with severe penalties to discourage potential misuse of iPSCs. Ethics aside, the more pressing concern is that whether iPSCs are truly equivalent to ESCs. Series of publications in past years have revealed the differences between the two types of stem cells, such as epigenetic imprints of the source somatic cells [61]. Researchers are hoping that such problems arise from the existing reprogramming technology, and many teams are still working on modes of producing the iPSCs. And besides, another vital concern related to the application of iPSCs in regeneration medicine is about their potential tumorigenicity. There might be small numbers of tumorigenic cells in the iPSC-based products. Therefore, quality control with gold standard, reproducibility of protocols, and immunogenicity are of great important. Indeed, a quality control guideline for clinical-grade iPSC production was published in 2018 by the Global Alliance for iPSC Therapies [62]. However, variability is still observed in the levels of tumorigenicity, genome instability, epigenetic status, and differentiation potentials within inter-and intra-iPSC lines. Moreover, many studies have reported genetic and epigenetic variations, which may lead to the possibility of tumorigenicity. These unexpected mutations or modifications could be introduced and accumulate during iPSC reprogramming, colony selection, expansion, and purification, potentially affecting cell differentiation and/or function. Thus, routine validation of iPSCs is required. In recent years, rapid progress has been made in the creation of in vitro systems using patient-specific iPSCdifferentiated cells for studying cellular mechanisms and the pathophysiology of certain diseases; however, significant hurdles remain in this field. The major concern is that the purified cultures of individual cell types never fully recapitulate the complexity of the human microenvironments. More recently, scientists induced phenotypic cells from patientspecific iPSCs and inoculated in a microfluidic cell culture system, known as organs-on-chips models, which could better mimic complex tissue architecture and the physiochemical microenvironment. These methods are increasingly used as physiologically relevant pre-clinical disease models.
In spite of these challenges, the potential value of patientspecific iPSCs remains enormous. New strategies are being developed to solve these pitfalls, such as a proposed iPSCbased multi-cell-type organized 3D model that can be realistically established on a timescale representing the development of disease in patients. Furthermore, a number of emerging areas may expand on the unlimited potential of iPSCs, such as iPSC-based vaccines, iPSC-derived exosomes, and others. In summary, the continued development of patient-specific iPSCs has highlighted the potential of precision medicine and provided exciting opportunities for clinical translation of this technology. It can be expected that clinically use of patient-specific iPSCs may soon be more widely available.