In-a-dish: Induced pluripotent stem cells as a novel model for human diseases


  • P. C. B. Beltrão-Braga,

    Corresponding author
    1. Surgery Department, Stem Cell Laboratory, School of Veterinary Medicine, University of São Paulo, São Paulo, Brazil
    2. Obstetrics Department, School of Arts, Sciences and Humanities, University of São Paulo, São Paulo, Brazil
    • Surgery Department, Stem Cell Laboratory, School of Veterinary Medicine, University of São Paulo, São Paulo, Brazil
    Search for more papers by this author
  • G. C. Pignatari,

    1. Surgery Department, Stem Cell Laboratory, School of Veterinary Medicine, University of São Paulo, São Paulo, Brazil
    Search for more papers by this author
  • F. B. Russo,

    1. Surgery Department, Stem Cell Laboratory, School of Veterinary Medicine, University of São Paulo, São Paulo, Brazil
    Search for more papers by this author
  • I. R. Fernandes,

    1. Surgery Department, Stem Cell Laboratory, School of Veterinary Medicine, University of São Paulo, São Paulo, Brazil
    Search for more papers by this author
  • A. R. Muotri

    1. Surgery Department, Stem Cell Laboratory, School of Veterinary Medicine, University of São Paulo, São Paulo, Brazil
    2. Department of Pediatrics/Rady Children's Hospital San Diego, University of California San Diego, School of Medicine, Stem Cell Program, La Jolla, California 92093, MC 0695
    3. Department of Cellular and Molecular Medicine, University of California San Diego, School of Medicine, Stem Cell Program, La Jolla, California 92093, MC 0695
    Search for more papers by this author


Human pluripotent stem cells bring promise in regenerative medicine due to their self-renewing ability and the potential to become any cell type in the body. Moreover, pluripotent stem cells can produce specialized cell types that are affected in certain diseases, generating a new way to study cellular and molecular mechanisms involved in the disease pathology under the controlled conditions of a scientific laboratory. Thus, induced pluripotent stem cells (iPSC) are already being used to gain insights into the biological mechanisms of several human disorders. Here we review the use of iPSC as a novel tool for disease modeling in the lab. © 2012 International Society for Advancement of Cytometry


Pluripotency is generally defined by the ability of a stem cell to differentiate into cell types representative of all three germ layers: ectoderm, mesoderm, and endoderm (1). Pluripotent stem cells are found in the inner cell mass of a blastocyst during embryogenesis, after the process of fertilization. It is also possible to generate a blastocyst in vitro, without going through the process of fertilization by the meeting of two gametes. In 1962, John Gurdon was the first to report the reprogramming of fully differentiated intestinal epithelial cells from Xenopus tadpoles by transferring the nucleus of the somatic cells into Xenopus oocytes and obtaining a blastocyst in vitro (2). However, it took more than 30 years until in 1996, Ian Wilmut demonstrated that a somatic cell nucleus from a mammal could similarly be reprogrammed through transfer to an enucleated oocyte (3). Despite the success of Gurdon's method, this procedure was not very efficient for generating human pluripotent stem cells. The discovery of reprogramming by somatic cell nuclear transfer, together with advances in the culture of pluripotent murine and human embryonic stem cells, has led to a further understanding of the processes of self-renewal and differentiation (4, 5).

Generation of Induced Pluripotent Stem Cells from Human Somatic Cells

Takahashi and Yamanaka pioneered a method of inducing somatic cells back to the embryonic stage, creating embryonic (ES)-like cells. These cells are called induced pluripotent stem cells (iPSC) (4). For this purpose they used the ectopic expression of 4 transcription factors: Oct-4 (octamer-binding transcription factor 4, also known as POU5F1), Sox-2 (Sex determining region Y box-2), Klf-4 (Kruppel-like factor 4), and c-Myc (proto-oncogene c-Myc).

Oct-4 is a transcription factor initially active in the oocyte and it remains active in embryos throughout the preimplantation period (6). The Oct-4 gene expression is equated with an undifferentiated phenotype in normal as well as malignant tissue (7). Gene knockdown of Oct-4 and Nanog (another transcription factor expressed in pluripotent stem cells) (8) promotes cell differentiation, thus indicating an important role for these factors in human embryonic stem cell self-renewal (9). Sox-2 may act to maintain or preserve a developmental potential (10). Moreover, Oct-4 can form a heterodimer with Sox2, and these two factors together can drive the expression of Nanog (7). Klf4 is a transcription factor that is associated with both tumor suppression and oncogenesis (6). This factor also suppresses the expression of p53, which induces ES differentiation by suppressing Nanog gene expression (11). Myc is a transcription factor that can activate or repress gene expression (10). The Myc gene codes for a transcription factor required for normal embryonic development. In addition, the Myc protein may induce global histone acetylation (12), allowing Oct-4 and Sox2 to bind to their specific target loci (4). Thus, combining the expression of these four factors, it is possible to change the epigenetic state of the cell, leading to pluripotent stage.

Similarities Between ES and iPSC Cells

Since the first iPSC published report, scientists have asked how close iPSC are to Embryonic Stem Cells (ESC) (Table 1). In cell culture, iPSC are difficult to distinguish from ESC, as both express pluripotency genes, form teratomas, differentiate in vitro, elongate telomeres, and lastly, give rise to mouse chimeras (11). Based on the colony or cell morphology, self-renewal capacity, and developmental potential, iPSC are practically indistinguishable from ESC. However, it is at the molecular level that their similarity remains controversial (12, 13, 14, 15). The methylation status of iPSC also shows similarity to that of ES cells. Using chromatin immunoprecipitation, Pan et al. (16) analyzed the histone modification status in human iPSC and revealed the bivalent patterns of development-associated genes characteristic of hES cells, such as Gata6, Msx2, Pax6, and Hand1.

Table 1. Phenotypic markers for ES and iPSC
  1. The numbers in parentheses refer to the literature citation.

 (5, 34, 48, 49, 50)(51, 52, 53, 54)
 (5, 19, 27, 34, 48, 49, 50)(51, 52, 55)
TRA-1-60 X X
 (5, 27, 34, 48, 49, 50)(51, 52, 55, 56)
TRA-1-81 X X
 (5, 27, 34, 49, 50)(51, 52, 53)
TRA2-49/6E X X
 (27, 33, 34, 48, 38, 49, 50)(49, 53)
Oct3/4 X X
 (5, 27, 33, 34, 38, 48, 49, 50)(49, 52, 53, 57, 58)
Sox-2 X X
 (27, 33, 50)(53, 59, 60)
 (5, 61)(5, 61, 62, 63)
Rex-1 X X
 (5, 61)(5, 61)
 (5, 61)(5, 61)
 (5, 61)(5, 61)
 (5, 61)(5, 61)
 (5, 61)(5, 61)
 (5, 61)(5, 61)
 (5)(5, 61)
 (5)(5, 61)
CDy1 X X
Klf4 X X
c-Myc X

In addition, iPSC derived from different donor cells are apparently more highly similar to each other than to ESC (17). In addition, late-passage iPSC cluster more closely with ESC than with early passage iPSC, indicating that expression differences that occur between early passage iPSC and ESC can be solved upon extended passaging (18, 19). It seems that many of these differences might actually be related to the cell source used to produce iPSC, as a result of residual expression of donor cell-specific genes (18, 20, 21). These differences do not seem to significantly modify cellular function in the pluripotent state, but may potentially interfere during the differentiation process since a specific gene expression program needs to be turned on during tissue development (22).

IPSC Phenotypic Markers

IPSC colony picking, with the aim of isolating reprogrammed cells, is a time consuming and demanding procedure. Eventually, iPSC colonies that are not completely reprogrammed could be selected by mistake. Detecting and isolating true iPSC colonies is crucial to the understanding of the mechanisms of cellular reprogramming. The current methods of isolation and characterization of iPSCs depend on cell morphology in culture, or immunostaining using specific markers. Identification of certain biomarkers could be useful to track cell reprogramming during the process of iPSC colony isolation. However, these methods are also time consuming and involve the use of antibodies that may barely differentiate partially reprogrammed cells. Moreover, the treatment may often turn the cells unsuitable for further study. For example, downregulation of fibroblast markers such as Thy1 is the earliest detectable change in cells undergoing reprogramming, followed by modest upregulation of the mouse ESC marker SSEA-1, and reactivation of endogenous Oct4 and Sox2 (23). The use of Fluorescence-activated cell sorting of subpopulations defined by these markers could contribute significantly to the enrichment of cells with the potential to become true iPSC (23, 24).

SSEA-4 expression is detected on the second day after reprogramming, whereas TRA-1-60 is detected only in a small subset of SSEA-4+ cells around days 6–10 (25). However, FC sorting disrupts individual colonies, precluding lineage tracing. Of particular interest, live staining of iPSC can be done using two keratin sulfate proteoglycan carbohydrate moieties, recognized by the surface epitopes TRA-1-60 and TRA-1-81 antibodies. These antibodies detect a neuraminidase-sensitive epitope, and a neuroaminidase-insensitive epitope, respectively. Both markers have been used successfully in live-cell immunostaining of emerging iPSC colonies (17, 25). During reprogramming, SSEA-4 seems to become upregulated earlier than TRA-1-60 or TRA-1-81, making it useful as an early marker of reprogramming. However, SSEA-4 is also present in incompletely reprogrammed iPSC as well, making it less useful for early identification of potential fully reprogrammed cells (25). In contrast, CD13, an aminopeptidase N (also known as a fibroblast marker), becomes rapidly downregulated during reprogramming and is not expressed by human pluripotent stem cells (25). mCD49e (integrin α5) is expressed by mouse embryonic fibroblasts (MEFs), and can be used to mark MEF feeder cells in human iPSC cocultures (26). Finally, CDy1 (compound of designation yellow 1) selectively marks live ESC and iPSCs without affecting growth rate, morphology, or differentiation capability (27).

One caveat on the use of antibodies for sorting is the possibility that the antibodies may have significant biological effects on the cells, causing unspecific side effects. Nonetheless, it is undoubted that flow cytometry has been increasing its use in stem cell isolation and characterization (28).

Possible iPSC Applications and Challenges

Perhaps the first obvious application of iPSC would be for autologous cell transplantation, avoiding ethical issues related to human ES use, and eventual immunological rejection after transplantation (Fig. 1). However, potential pitfalls regarding the use of iPSC for cellular therapy have been identified, such as the use of reprogramming oncogenes that may induce malignant cell transformation (4, 14, 29).

Figure 1.

Applications for pluripotent stem cells. Pluripotent stem cells have the potential to originate an entire individual or every cell types present in the body. IPSC from a patient can be used as autologous transplant, disease modeling and drug screening. IPS factors could be virus, plasmid, small molecules, or RNA.

Further difficulty hampering the use of iPSC is the relatively low efficiency of reprogramming. The most widely used method for reprogramming utilizes retroviruses for gene delivery, and takes 3–4 weeks for primary colony picking with an efficiency of 0.001–0.1% (3, 30). Beltrão-Braga et al. (2011) (31), using dental pulp stem cells that are more immature than fibroblasts, observed iPSC colonies within 2 weeks (Fig. 2) using the same protocol described by Takahashi and Yamanaka (3). This fast speed of reprogramming could be attributed to a more immature stage of cells, as dental pulp stem cells are defined as mesenchymal stem cells, expressing both mesenchymal and certain pluripotent markers (31). In addition, the finding that immature cells are easier to reprogram was also observed by others, indicating that the time of reprogramming likely depends on the donor cell type (32, 33, 34).

Figure 2.

Human iPSC colonies derived from dental pulp stem cells and neuronal differentiated cells derived from iPSC. A, B: Representative images of iPSC colonies visualized under light microscope. C: Representative image of iPSC colony expressing pluripotent markers Nanog (red) and Lin28 (green). D: Representative image of cells after neuronal differentiation, expressing neuronal markers Map2 (green), islet1 (red), and nucleus in blue (DAPI). Magnifications in A) ×10, scale bar 400 μm, B) ×40, scale bar 100 μm, C) ×40, scale bar 100 μm, and D) ×20 scale bar 200 μm.

Generating human pluripotent stem cell lines from patients affected by several disorders is a powerful strategy that can be used in disease modeling study, drug discovery, and toxicological testing protocols (35). Furthermore, making iPSCs from each patient with a diverse genetic background is important to our understanding of the variations that occur in responses to therapeutics among different patients. If iPSCs are generated from live patients, the information gained studying their differentiation potential may be applied to the patient in the future during his or her lifetime.

Induced Pluripotent Stem Cells for Disease Modeling and Drug Testing

The use of the iPSC protocol to generate pluripotent cells derived from patients' cells represents a breakthrough to study biological mechanisms of diseases. Dimos et al. generated iPSCs from a patient with amyotrophic lateral sclerosis (ALS) (36). The disease-specific iPSCs were successfully differentiated into motor neurons, the cell type affected in ALS. However, there is no report on a disease phenotype in ALS motor neurons from this publication. In 2009, other investigators reprogrammed cells from others diseases such as Familial Dysautonomia (37), Parkinson's disease (38), and type 1 diabetes (39). Also, Chamberlain et al. suggested that iPSC could be used for studying genomic imprinting disorders, reprograming cells from patients with Angelman and Prader-Willi syndrome (40) (Fig. 3).

Figure 3.

Time line for iPSC disease modeling.

The first comparison between affected and non-affected cells derived from iPSCs was a study published by Ebert et al. in 2009. The researchers reprogrammed fibroblasts from a patient with spinal muscular atrophy (SMA), and differentiated these iPSC into motor neurons, the cells affected in this syndrome. Using this strategy, they observed the low survival level of motor neurons in a patient with this disease when comparing with motor neurons derived from iPSC from a non affected relative (41). Through comparison of normal and pathologic cells, and evaluation of the effects of drug treatment in vitro, iPSC lines generated from patients offer an unprecedented opportunity to recapitulate pathologic mechanisms of disease in vitro, a new technology platform for drug screening and customized therapy for each patient.

In 2009, an important contribution in the application of disease-specific iPSC was published by Raya et al. The group generated disease-corrected, patient-specific cells with potential value for cell therapy applications. They first corrected the genetic defect from Fanconi anemia patients, and then reprogrammed to pluripotency to generate patient-specific iPSC. Reprogrammed cells gave rise to haematopoietic progenitors of the phenotypically normal myeloid and erythroid lineages (42).

Another important contribution of the iPSC technology to modeling of neuronal disease was made by Marchetto et al. in 2010, using Rett syndrome (RTT) as a syndromic autistic prototype (43). Using the iPSC strategy, the authors observed that neurons derived from RTT-iPSCs had fewer synapses, reduced spine density, smaller soma size, altered calcium signaling, and electrophysiological defects when compared to controls. In addition, with this “dish-disease” model, they tested the effects of drugs on rescuing synaptic defects (43).

More recently, Pasca et al. studied neurons derived from iPSC of Timothy syndrome patients (44). Using these neurons, they found defects in action potential firing and [Ca2+]i signaling (45). As a result of calcium imbalance, communication between neurons was defective. These imbalanced cells were producing an excess of an enzyme necessary for the production of the catecholamines norepinephrine and dopamine, which have a key role in sensorial neurons and social behavior. They also showed that roscovitine could block the defective calcium channel resulting in reduction of such an enzyme implicated in cathecholamine production. This effect was observed only in human iPSC-Timothy neurons, and not in transgenic mice expressing a point mutation in an alternatively spliced exon of CACNA1C gene, reflecting intrinsic differences from the genetic background of humans and mice.

However, some concerns still need to be addressed for the full impact of iPSC on disease modeling. For example, it seems important to correctly choose the best donor cell type to reprogram in order to model specific diseases. Somatic cells can retain a cellular memory even after reprogramming (33), affecting cellular differentiation and dowstream data interpretation. Our group has recently opted to use human dental pulp stem cells as a source for cellular reprogramming (31), since they have the same embrionic origin as cells from the nervous system, and our interest is to ultimatley produce cells from the nervous system in a plate.


We believe that the iPSC technology can be very useful for modeling human diseases, especially complex neurodevelopmental disorders. Initial results are now emerging from modeling monogenetic diseases (43, 44). Finally, using iPSC technology in combination with genomic approaches will lead to novel molecular and cellular pathways affected in several human diseases, increasing our perpectives on therapeutical targets.


The authors are thankful to Caroline P. Winck, Silvia A.F. Lima, Cassiano Carromeu, and Marianna Yusupova for their critical reading and editing of the manuscript.