Main

Induced pluripotent stem cells (iPSCs) are pluripotent stem cells artificially generated by transiently expressing a set of exogenous transcription factors in somatic cells (Table 1). As Takahashi and Yamanaka1 originally reported the method for iPSC induction in 2006, the field has been rapidly expanding with great expectation and with some concern for their appropriate use. Essentially, the clinical implications of iPSCs are twofold; first, as a cellular resource for transplantation therapy, and second, as a system to model human diseases. Although the former direction is years away and unlikely to be an immediate concern for most experimental pathologists, the latter potential may be more relevant. We feel now is a good time to overview the present status of iPSC research for used in disease mechanism studies in this pathology-oriented journal, to discuss the potential value of iPSCs in future disease biology studies, and also to address the limitations and obstacles that need to be overcome. As many excellent reviews have been published to date (eg, see Kiskinis and Eggan,2 Stadtfeld and Hochedlinger,3 Saha and Jaenisch,4 Marchetto et al,5 and Yoshida and Yamanaka6), here we will try to avoid redundancy as much as possible and bring a newer perspective to iPSC use in modeling clinical diseases.

Table 1 Basic background of induced pluripotent stem cells (iPSCs)
Full size table

iPSCs TO MODEL CLINICAL DISEASES?

When the technology to generate human iPSCs first became available,7, 8 immediate attention was placed on their potential for use in cell-based transplantation. Using in vitro differentiation, iPSCs, like embryonic stem cells (ESCs), can provide an unlimited source of useful cell types for transplantation. The use of iPSCs in research has been largely welcomed by society because they lack the substantial ethical concern of cellular origin, which plagues ESCs. The fact that the cells are autologous for patients could be another advantage in transplantation. A major drawback of iPSCs for transplantation use is their carcinogenic potential, although recent progress in reprogramming technologies is overcoming the problem (see Table 1).

Soon after human iPSC technology was introduced, however, researchers also began to realize an additional and possibly greater value for the cells as a system to model human diseases. iPSCs can be generated from skin biopsies or blood samples of patients, and can be differentiated in vitro into cell types that are not easily accessible in patients, such as neurons and cardiomyocytes. As iPSCs retain all the genomic information from the original patients, iPSCs could be used to study how genetic aberrancies in the patient manifest in target cells in vitro.

One reason for advancement of hematopoietic disease understanding from molecular studies is ease of accessibility of blood or bone marrow samples for in vitro studies. Successful development of molecular-targeted drugs, such as imatinib, for the treatment of chronic myeloid leukemia represent a triumphal example of a successful outcome of long-term molecular study. In contrast to the blood, other patient tissues such as brain and heart are not easily accessible, which has been a substantial disadvantage for pathobiology studies in neural and cardiac disorders. Such drawbacks could be partly overcome by iPSC technology.

Skeptics can argue, of course, against this rather simple and bold scheme.4, 5 First, the cells obtained from in vitro differentiation of iPSC may be very different from equivalent cell types seen in real organs and tissues. Second, the cells will not likely fully or even closely recapitulate in vivo disease conditions, which are a consequence of complex systems with multiple cell types, and are due to the long-term effects by gene mutations. This is particularly a concern for late onset diseases. However, even when we see a part of the disease phenotypes or molecular changes in iPSC-derived cells, the system will be beneficial for defining and understanding disease mechanisms.

PIONEERING STUDIES OF DISEASE-SPECIFIC iPSCs

Despite some existing concerns, many pioneering studies have been conducted, some of which indeed demonstrate advantages to using patient iPSCs to understanding disease mechanisms and/or to identify novel therapeutic approaches. Table 2 summarizes the literature in which disease-specific iPSCs were generated. It should be noted that some papers listed in the table were not designed for disease biology study but rather intended for use in cell-based transplantation therapies in the future.

Table 2 Disease-specific-induced pluripotent stem cells (iPSCs), in studies published as of January 2011
Full size table

The first phase of research focused on demonstrating that iPSCs can be successfully generated from patients, and initial studies of this type were published as early as 2008. In an inaugural paper, Park et al9 showed that they were able to generate human iPSCs from patients with a variety of genetic disorders, and these cells showed a similar pluripotent differentiation capacity equivalent to control iPSCs derived from normal individuals. Meanwhile, Dimos et al10 generated iPSCs from ALS patients and differentiated them into motor neurons, in vitro, to demonstrate the potential of iPSC technology to produce a large amount of a disease-relevant cell type for research.

The second phase was to prove the concept that disease-iPSC-derived cells can indeed recapitulate some disease-specific effects in vitro. The first paper of this kind was published in January 2009 by Ebert et al.11 Here the authors generated iPSCs from patients with spinal muscular atrophy (SMA) who have mutations in the survival motor neuron 1 (SMN1) gene. Interestingly, deletion of SMN1 is partially compensated by a redundant SMN2 gene in human patients, which can also generate a full-length SMN protein but only at a lower level. Notably, other model animals such as worms, flies and mice lack the SMN2 gene, indicating that a model system in humans is essential for fully understanding disease mechanisms. Further, as targeted SMN2 gene activation is a potential mechanism for curing the disease, only a human cell system would be useful for the research. The study demonstrated that motor neurons derived from SMA–iPSCs harbor deficits in morphology, survival and synapsin staining, which represents SMA clinical pathology in part. In addition, the authors demonstrated that drugs, which were previously known to induce SMN2-derived SMN protein, indeed increased the level of SMN protein in SMA–iPSCs, implying the iPSC system would be useful for future drug discovery. Similar to the SMA study, recapitulation of neural disease phenotypes in vitro using the iPSC system has been nicely demonstrated with familial dysautonomia12 and Rett syndrome.13 Further, iPSCs have been generated from long QT syndrome patients,14, 15 where studies demonstrated prolonged action potentials in patient-iPSC-derived cardiomyocytes and their arrhythmogenicity, recapitulating the disease phenotype in vitro. Notably, all of the studies described above also demonstrated a reversal of the observed phenotypes by previously known drugs, indicating the system is compatible with drug discovery.

The third phase would be to prove that iPSC studies will indeed lead to novel insights for disease biology, and/or identification of novel therapeutic approaches. The study by Agarwal et al16 regarding dyskeratosis congenita (DC) may be the first in this category. DC is a disorder of telomere maintenance in which DKC1 mutation leads to destabilization of telomerase RNA component (TERC). Of interest, reprogramming into a pluripotency status increased the level of TERC despite the presence of DKC1 mutation and restored the telomere length. The discovery further led the authors to study and identify previously unidentified mechanisms of TERC upregulation, which could lead to a new therapeutic approach in the future. In Freidreich's ataxia,17 an extension of GAA/TTC nucleotide repeat was seen during iPSC generation and prolonged culture, which was partially prevented by knocking down of the MSH2 gene. These data also provided some newer insights regarding the disease progression.

The fourth phase, which has not yet been achieved, will be to demonstrate that iPSC research indeed leads to disease prevention or cure by discovery of effective therapeutic approaches or drugs. This would be the time when the technology truly reaches Nobel Prize status.

WHERE WILL WE SEE THE TRUE VALUE OF iPSCs?

As discussed above, disease biology studies using iPSCs are on the way, and are progressing steadily with encouraging speed. Where will iPSCs likely show the most value in the near future? First, diseases that do not have high quality or appropriate animal models would benefit from iPSC study. In such cases, iPSC research has a great chance to facilitate disease understanding and/or drug discoveries. Indeed, we could say that iPSCs, which can be generated directly from patients relatively easily, are a fast track research tool when compared with other model systems in which we need to induce gene modifications (Figure 1). Moreover, iPSCs are a fast track research tool in clinical translational research for an additional reason. As a cellular system, disease iPSCs or iPSC-derived cells are directly applicable to drug screening. Importantly, we can achieve these schemes not only within relatively short time but also with relatively low cost when compared with the classical approaches in which we first identified the causes and then generated models to recapitulate them. These features may be particularly good news for research of rare diseases, which lack a large budget. In addition to the studies published (Table 2), a considerable number of research labs around the world are currently generating iPSCs from a variety of rare diseases. It may not be long until we hear promising discoveries of novel approaches or drugs to cure some of these diseases.

Figure 1
figure 1

Fast track iPSC studies may facilitate clinical translational research.

Full size image

Since the first transgenic and knockout mouse studies were published in 1980 and 1989, respectively, mice have been extensively used to model human diseases. There is no doubt their contributions to medicine are and will be countless, and indeed the development of knockout mouse technology was awarded the 2007 Nobel Prize. However, we also know that many human diseases are complex polygenic diseases, which are not easily recapitulated by gene modifications in mice. In the era of genome-wide association study (GWAS), when we accumulate our knowledge of polygenetic linkage to diseases, an alternative model to recapitulate the polygenic modifications is highly desired. In other words, in order to truly understand mechanisms of how polygenic modifications identified through GWAS lead to disease conditions, we need an experimental interface to study their pathobiological effects. In this context, iPSCs, retaining all the genetic information from patients, should have another indispensable value (Figure 1).

Personalized medicine is another field in which iPSCs are expected to make a contribution. GWAS information and clinical databases should be sufficient to predict drug efficacy and toxicity for individuals with drugs already widely used in clinics. However, for new drugs that have not been tested in clinics yet, or those tested in a small number of patients, the information would not be sufficient. Here a large iPSC library accompanying GWAS information will be very powerful. Toxicity or efficacy of new drugs on liver, eg, can be tested using hepatocytes derived from an iPSC library of normal and diseased individuals (eg, α1-antitrypsin deficiency).

Several issues should be overcome if we truly want to advance the field quickly. First, it will be critical to network iPSC labs around the world to create an iPSC library of both normal and diseased cells using a common quality standard. Second, a systematic approach to develop an iPSC library in conjunction with a clinical database, tissue bank and GWAS would be most useful. Third, further development of efficient in vitro iPSC differentiation protocols into many more cell types is essential for progress in the field. Forth, continuous effort to recapitulate phenotypes of late-onset diseases in vitro, at least partly, would be critical to extend their applications. Lastly, reducing complexity of culture methods will be important to make the system more easily applicable to high throughput screening.

CONCLUDING REMARKS

Here we discussed the potential and progress of iPSC research for disease biology studies and drug discovery. Admittedly, the major strength of iPSCs is likely the fact that these cells are derived directly from human patients. In this sense, we may be able to say that iPSCs are an alternative or additional resource for clinical tissue banks that are becoming increasingly valuable in clinical translational research particularly in this GWAS era. Although iPSCs are not comparable at all with tissues for their ability to give us histological information, iPSC provide an unlimited source of live cells from patients, even cell types that cannot be easily or frequently obtained alive. In addition, although tissue banks provide more static disease information, an iPSC system can allow for the dynamic study of gene aberrations during the process of development or cell differentiation. Lastly, as a live cell system, it is feasible to apply iPSCs to drug discovery, efficacy and toxicity testing.

Disease-specific iPSCs are a new system to model human diseases, which can become very powerful in multiple directions as discussed above. We have no intention here, however, to conclude that iPSCs are a superior model for human diseases compared with others. Animal models such as mice, rats, fruit flies, yeast etc have contributed enormously to, and will remain crucial for understanding disease biology and/or drug discovery without a doubt. Each model has its strengths and weaknesses, or advantages and drawbacks. As has always been true, the combination of multiple model systems would be the most powerful way to understand human disease biology. However, the value of iPSCs as a first universal system to use human cells for modeling a variety of human diseases should not be overlooked. Further progress in patient iPSC research may lead us to remember an old yet fundamental truth in medicine; ‘we can learn best from patients’.