Induced pluripotent stem (iPS) cells have the potential to transform drug discovery by providing physiologically relevant cells for toxic compound identification, target validation, compound screening, and tool discovery. The technology for generating iPS cells is advancing rapidly, as is the repertoire of cell types that can be differentiated.
Tissue-specific cells derived from iPS cells are currently being evaluated by the pharmaceutical industry for their utility in identifying cardiotoxic and hepatotoxic compounds as therapeutically relevant systems for modelling cardiovascular diseases, neurodegenerative disorders and metabolic disorders, as well as for generating patient-specific cell types. In order to fully capitalise on the rapidly evolving science of iPS cell technology, pharma will need to leverage the expertise found in academia and biotechnology companies to apply iPS-derived cells in an industrial setting. This review will summarise the potential of these cells as well as highlight many of the challenges that remain.
Stem cells are a distinct self-replenishing cell population whose primary function is to generate progeny that then develop into terminally differentiated cell types, such as a cardiomyocytes, neurons or photoreceptors. Tissuespecific adult stem cells, or progenitors, are committed to producing tissue or lineage-specific cells, whereas totipotent or pluripotent stem cells can give rise to any of the 200+ cell types of the body.
There are two types of pluripotent stem cells defined by their tissue origin:
1) embryonic stem (ES) cells obtained from early embryos, typically at the blastula stage
2) induced pluripotent stem (iPS) cells derived through a reprogramming process whereby terminally differentiated somatic cells are reprogrammed or induced to a pluripotent state (Figure 1)
Successful generation of human iPS cells was reported independently in 2007 by the research groups of Drs James Thomson and Shinya Yamanaka (1,2). Both groups successfully identified a minimum number of nuclear factors that could reprogramme terminally differentiated fibroblasts to pluripotent cell lines by exogenously expressing distinct yet overlapping sets of genes: Yu and colleagues used Oct4, Sox2, Nanog and Lin28 by lentiviral gene transfer, while Yamanaka and colleagues employed Oct4, Sox2, Klf4 and c-Myc via retroviral gene transfer.
These landmark studies were a significant advance over whole nucleus reprogramming (3) and have had several important consequences. Induced pluripotent stem cell technology does not require human embryos and thus circumvents the ethical issues associated with embryonic stem cells. Of more utility is the potential to introduce population diversity into early and basic research programmes, as iPS cells can be generated from any individual.
This iPS cell capability can provide researchers with tissue from virtually any genotype, ranging from the broad spectrum of ‘healthy’ individuals to clinical and disease cohorts to individuals exhibiting specific side-effects and/or idiosyncratic toxicities. Induced pluripotent stem cells will advance the concept of personalised medicine closer to reality.
iPS cell generation: Several ways to reprogramme a cell
As noted above, the initial iPS cell derivations utilised four-independent factors delivered via viral vectors to reprogramme the starting fibroblast populations. Since those seminal reports, reprogramming of cells has been successfully accomplished with a variety of delivery systems utilising variations of the initial four-factor cocktail, related transcription factor family members, endogenously expressed factors, small molecules, and soluble factors (4,5).
The use of non-genomic, small molecule factors is an attractive method for generating iPS cells, but the specificity with which they faithfully recapitulate transcriptional pathways is unknown, and it is unclear whether they can replace traditional reprogramming factors (4).
The potential to use autologous iPS cells as therapeutic medicines holds tremendous promise. Cellular physiological integrity is, however, paramount and will require delivery systems that do not cause oncogenic transformation or alter gene function. Lentiviral- and transposon-based systems, in which the inserted transgene is subsequently excised using the Cre recombinase/lox site system or transposases respectively (6,7), remain problematic, but recent technologies that do not require genome integration look more promising. These technologies include episomal vectors (8), recombinant proteins (9), viral RNA vectors (10), and synthetic modified mRNAs (11) (Table 1).
Another important consideration for reprogramming is the parental tissue source. The most popular starting material is fibroblasts, though other cell types have been successfully used, including keratinocytes (12), mesenchymal cells (13), adipose stem cells (14), and melanocytes (15). While to date there are no reports of cell types refractory to reprogramming efforts, a more desirable source of starting material may be human peripheral blood, which can be readily obtained through routine and relatively non-invasive clinical procedures.
Efforts to develop such methods have yielded iPS cells derived from human CD34+ blood stem cells and T lymphocytes (16-18). Of particular note is a recent report demonstrating efficient human iPS cell derivation from T-lymphocytes collected in as little as 1ml of whole blood (19).
There is considerable work ahead in the characterisation of iPS cells and their progeny. For example, recent studies indicate that iPS cells carry epigenetic memory of their source tissue and exhibit a propensity to differentiate into their parental cell type (20,21). However, this epigenetic memory appears to be transitory and can be erased through repeated passaging, additional reprogramming, or with chromatin modifying drugs. Repeated passaging may be the most straightforward approach to erasing epigenetic memory, as it is a necessary step in any scale up of iPS cells.
Stem cell differentiation and demonstrations of utility
Terminally differentiated cells in vivo are the end result of a complex set of chemical interactions, positional cues, environmental effects, and cell-tocell communications, many of which have proven difficult to reproduce under in vitro conditions. Some cell types, such as cardiomyocytes, develop spontaneously in tissue culture dishes (albeit at very low frequencies) from stem cells while other cell types, such as hepatocytes, require significant experimental intervention (22-24). Once a protocol has been elucidated, initial derivation processes and robust cell type markers can be used for method optimisation and production of industrialised quantities of cells.
What is the promise?
There are three major areas where iPS cell technology is beginning to have or will have a major impact on in vitro aspects of drug discovery: safety assessment, small molecule screening, and in vitro disease models.
Finding novel drugs that are safe and efficacious is challenging. Attrition from candidate selection to drug launch is estimated at 95%, with 20% of that due to toxicity issues (lack of efficacy accounts for the greatest single cause of attrition (24,5), therefore iPS-derived cells may have a significant impact on drug discovery if they can be used in more physiological assays for toxicity screening.
Cardiomyocytes: Cardiovascular toxicity remains the single most important targeted organ resulting in an estimated 28% of drug withdrawals (26). Multiple assay systems are currently employed to identify cardiotoxic drugs. These systems are based predominantly on cell lines such as human HEK- 293s or hamster CHOs for measuring drug effects on ion channels (including hERG) and transporters; isolated animal tissues including rabbit wedge for ECG and QT elongation studies; or whole animal experiments using rats, dogs, or monkeys for ECGs, hemodynamic studies, and pathological evaluation of repeat dose studies to identify cardiotoxic compounds.
Recent progress in manufacturing cardiomyocytes has been made using pluripotent stem cell technology. In 2009, Cellular Dynamics International (CDI) introduced the first human iPS cell-derived product, iCell™ Cardiomyocytes, while GE/Geron corporation released embryonic stem cell-derived cardiomyocytes in 2010. These human cells have the potential to reduce or replace the assay systems and animal models mentioned above. Both platforms are a pan-cardiac population consisting of ventricular, atrial and nodal cells.
Extensive characterisation of these cells has demonstrated molecular, electrophysiological, and cardiotoxic function similar to native heart cells (27) (Ma et al, in preparation). In addition to providing large numbers of reproducible cardiomyocytes for use in traditional toxicity testing, the in vivo functionality of pure populations of iPS cell-derived cardiomyocytes will enable the development of potentially more targeted, sensitive, and relevant biochemical and electrophysiological cardio-specific testing paradigms.
Hepatocytes: Liver toxicity is another major cause of drug withdrawal from the market. Current systems for measuring hepatotoxicity include primary human hepatocytes (PHH) or human hepatocyte cell lines such as HepG2s. Both of these have major draw backs. Primary human hepatocytes must be shipped either fresh immediately after isolation or frozen, and they display batch-to-batch variation. Additionally, they are not stable in culture for more than a few days and lose P450 expression. Yet PHH remain the gold standard.
Cell lines such as HepG2 express little to no P450 enzymes, the major drug metabolising enzymes, making these cell lines inappropriate model systems (28). What is needed is a reliable cell supply with batch-to-batch consistency. In addition, these cells must express appropriate markers and enzymes including the P450 enzymes and Phase I and Phase II xenobiotic metabolising enzymes, thus demonstrating intrinsic hepatocyte metabolism including oxygen consumption and ATP utilisation, exhibiting appropriate cell morphology and polarisation, and having the ability to identify known hepatotoxic compounds. Stem cell-derived hepatocytes have been reported to exhibit many, but not yet all, of these phenotypic and functional characteristics (24,29-30).
Small molecule screening
Traditional small molecule screens are performed using purified protein or recombinant cell lines over-expressing the target of interest. The industry uses such systems because they are easy to develop and configure to their platforms, are relatively inexpensive, and have a track record of success (Macarron et al, in preparation). Recombinant systems are not ideal, however. The target of interest is over-expressed in a non-native cell where splice variants or post-translational modifications may not occur; iPS cells have the potential to change that.
A recent example comes from a paper by McNeisch et al (31) where neurons expressing aamino- 3-hydroxyl-5-methyl-4-isoxazoleproprionate (AMPA) subtype glutamate receptors were generated from mouse iPS cells. This is a particularly challenging target because there are four subtypes, multiple RNA splice variants, and interactions with transmembrane regulatory proteins, all giving rise to a significant number of potentially expressed receptors.
McNeisch et al (31) successfully generated a sufficient quantity of cells to run a high-throughput screen of several million compounds and identified novel small molecule AMPA potentiators. Such an approach might readily be applied to other targets including ion channels, GPCRs, and receptor tyrosine kinases where full functionality requires multimers of subtypes, chaperones, and protein complexes.
In vitro disease models
Induced pluripotent stem cell technology provides a unique opportunity to generate ‘disease phenotypes in a dish’ for use as in vitro model systems and substrates for small molecules screens. While it is unlikely that an in vitro system can adequately represent all aspects of a multi-faceted disease, the ability to generate any human tissue cells from any genotype or disease provides unparalleled access for studies. There are an increasing number of examples where iPS cells have been derived from patients, but as of now only a few show measurable phenotypic endpoints (Table 2).
The notable examples include iPS cell-derived motor neurons from patients with spinal muscular atrophy (SMA), where the derived neurons demonstrated a significantly reduced level of survival motor neuron 1 (SMN1) gene expression, and appropriate response to drugs known to increase SMN1 gene expression accurately mimicked the phenotype observed in SMA patients (32).
Another example comes from Lee et al (33) who studied familial dysautonomia (FD), a fatal neurodegenerative disease of sensory and autonomic neurons. iPS cell-derived neural crest progenitor cells from an FD patient displayed the expected abnormal splicing of IKBKAP, exon skipping and a general reduction of protein. These neurons displayed a reduced ability to migrate, a reduction in the number of focal adhesion points, and the expected and appropriate response to drugs known to reduce the level of mutant IKBKAP splice form.
Other diseases where cellular phenotypes have been observed include Huntington’s disease (34); Parkinson’s disease; Long QT syndrome (35); Leopard syndrome (36); Pompe disease (37); hepatic glycogen storage disease (38); and ubiquitous aspects of cardiac hypertrophic phenotypes in both ES and iPS cell-derived cardiomyocytes (39); (Kattman et al, unpublished).
What needs to change for Pharma to implement iPSC-derived models?
Even though significant progress has been made, it is likely that full adoption and general acceptance will require comparative studies, scale-up and pilot screens. There remains insufficient data demonstrating that iPS cells are superior to existing primary animal cells or standard lab cell lines, though protocols for generating terminally mature, differentiated human cell types are progressing rapidly. Significant progress has been made in creating insulin-producing beta-cells (40-41), ventricular cardiomyocytes (42), P450-expressing hepatocytes (43), specific neuronal populations (44), and retinal photoreceptors (45), to name just a few.
Once these protocols are established, technology for scale up will be required that maintains a cell’s phenotype, shows batch-to-batch consistency and reproducibility, and is cost-effective for the industry. Given the intensive research efforts ongoing in academic labs to improve protocols and in biotechnology companies to scale cells, there is reason for optimism that iPS cells will be used in the lab either alongside or as replacements for traditional cell lines.
What’s the future vision for iPS cells?
iPS cells may have a profound impact on drug discovery, either in the generation of previously unobtainable cellular disease model systems for small molecule screening; mechanism of action studies; or highly predictable, animal-free systems for determining drug safety. One can envisage a screen of genetic diversity panels of cardiomyocytes or hepatocytes to identify rare responders, analyse drug effects on complex 3D organ systems, or perform ‘clinical trials’ using iPS-derived diseased cells where safety, efficacy, dosage studies, and the effect of genetics could be studied before initiating Phase I studies in patients.
Another potential application will be use of iPS-derived cells and/or geneticallymodified iPS cells in transplantation where there may be immunological advantages for organ engraftment. Proof of concept is now well-documented in pre-clinical models of retinal dystrophy (46), in reversal of liver damage (47), and in diabetic mouse models (48). Full adoption of stem cell technologies will require generation of iPS cells that are safe, stable and efficacious. The use of iPS cells for transplantation, potentially coupled with gene therapy, may usher in the era of true personalised medicine for patients with high unmet needs.
The authors would like to thank Blake Anson and Joleen Rau from Cellular Dynamics International, Madison, WI, for extensive discussions, helpful advice and critical review of the document; as well as Jason Gardner and Aaron Chuang (Stem Cell Discovery Performance Unit, GlaxoSmithKline) for their insight and comments. DDW
Dr Julie Holder has been involved in early discovery and development and late stage drug discovery throughout her GSK career. Julie is currently Director of Preclinical in the Stem Cell Discovery Performance Unit at GSK and has direct experience working with Mesenchymal Stem Cells (MSC) and extensive experience of compound development. She holds a PhD in Biochemistry from Leeds University, UK.
Dr Dwight Morrow is Director of Discovery in the Stem Cell Discovery Performance Unit at GSK and has extensive experience in cell-based assays, high throughput screening, and compound profiling (SAR). He received his PhD in Molecular Biology from Case Western Reserve University and did his post-doctoral work at Johns Hopkins University School of Medicine.
1 Yu, J, Vodyanik, MA, Smuga- Otto, K, Antosiewicz-Bourget, J, Frane, JL, Tian, S, Nie, J, Jonsdottir, GA, Ruotti, V, Stewart, R, Slukvin, II and Thomson, JA. (2007). Induced pluripotent stem cell lines derived from human somatic cells. Science 318(5858): 1917-20.
2 Takahashi, K, Tanabe, K, Ohnuki, M, Narita, M, Ichisaka, T, Tomoda, K and Yamanaka, S. (2007). Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell 131(5):861-72.
3 Simonsson, S, Gurdon, JB. Changing cell fate by nuclear reprogramming. (2005). Cell Cycle, 4(4):513-5.
4 Maherali, N and Hochedlinger, K. Guidelines and techniques for the generation of induced pluripotent stem cells. (2008). Cell Stem Cell 3(6):595-605.
5 Kiskinis, E and Eggan, K. (2010). Progress toward the clinical application of patientspecific pluripotent stem cells. J Clin Invest 120(1):51-9.
6 Soldner, F, Hockemeyer, D, Beard, C, Gao, Q, Bell, GW, Cook, EG, Hargus, G, Blak, A, Cooper, O, Mitalipova, M, Isacson, O and Jaenisch, R. (2009). Parkinson's disease patient-derived induced pluripotent stem cells free of viral reprogramming factors. Cell 136(5):964-77.
7 Woltjen, K, Michael, IP, Mohseni, P, Desai, R, Mileikovsky, M, Hämäläinen, R, Cowling, R, Wang, W, Liu, P, Gertsenstein, M, Kaji, K, Sung, HK and Nagy, A. (2009). piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature 458(7239):766-70.
8 Yu, J, Hu, K, Smuga-Otto, K, Tian, S, Stewart, R, Slukvin, II and Thomson, JA. (2009). Human induced pluripotent stem cells free of vector and transgene sequences. Science 324(5928):797-801.
9 Kim, D, Kim, CH, Moon, JI, Chung, YG, Chang, MY, Han, BS, Ko, S, Yang, E, Cha, KY, Lanza, R and Kim, KS. (2009). Generation of human induced pluripotent stem cells by direct delivery of reprogramming proteins. Cell Stem Cell (6):472-6.
10 Fusaki, N, Ban, H, Nishiyama, A, Saeki, K and Hasegawa, M. (2009). Efficient induction of transgene-free human pluripotent stem cells using a vector based on Sendai virus, an RNA virus that does not integrate into the host genome. Proc Jpn Acad Ser B Phys Biol Sci 85(8):348-62.
11 Warren, L, Manos, PD, Ahfeldt, T, Loh, YH, Li, H, Lau, F, Ebina, W, Mandal, PK, Smith, ZD, Meissner, A, Daley, GQ, Brack, AS, Collins, JJ, Cowan, C, Schlaeger, TM and Rossi, DJ. (2010). Highly efficient reprogramming to pluripotency and directed differentiation of human cells with synthetic modified mRNA. Cell Stem Cell 7(5):618-30.
12 Aasen, T, Raya, A, Barrero, MJ, Garreta, E, Consiglio, A, Gonzalez, F, Vassena, R, Bilic, J, Pekarik, V, Tiscornia, G, Edel, M, Boué, S and Izpisúa Belmonte, JC. (2008). Efficient and rapid generation of induced pluripotent stem cells from human keratinocytes. Nat Biotechnol 26(11):1276-84.
13 Park, IH, Arora, N, Huo, H, Maherali, N, Ahfeldt, T, Shimamura, A, Lensch, MW, Cowan, C, Hochedlinger, K and Daley, GQ. (2008). Diseasespecific induced pluripotent stem cells. Cell 134(5):877-86.
14 Sun, N, Panetta, NJ, Gupta, DM, Wilson, KD, Lee, A, Jia, F, Hu, S, Cherry, AM, Robbins, RC, Longaker, MT and Wu, JC. (2009). Feeder-free derivation of induced pluripotent stem cells from adult human adipose stem cells. Proc Natl Acad Sci U S A 106(37):15720-5.
15 Utikal, J, Maherali, N, Kulalert, W and Hochedlinger, K. (2009). Sox2 is dispensable for the reprogramming of melanocytes and melanoma cells into induced pluripotent stem cells. J Cell Sci 122(Pt 19):3502-10.
16 Loh, YH, Agarwal, S, Park, IH, Urbach, A, Huo, H, Heffner, GC, Kim, K, Miller, JD, Ng, K and Daley, GQ. (2009). Generation of induced pluripotent stem cells from human blood. Blood 113(22):5476-9.
17 Loh, YH, Hartung, O, Li, H, Guo, C, Sahalie, JM, Manos, PD, Urbach, A, Heffner, GC, Grskovic, M, Vigneault, F, Lensch, MW, Park, IH, Agarwal, S, Church, GM, Collins, JJ, Irion, S and Daley, GQ. (2010). Reprogramming of T cells from human peripheral blood. Cell Stem Cell 7(1):15-9.
18 Seki, T, Yuasa, S, Oda, M, Egashira, T, Yae, K, Kusumoto, D, Nakata, H, Tohyama, S, Hashimoto, H, Kodaira, M, Okada, Y, Seimiya, H, Fusaki, N, Hasegawa, M and Fukuda, K. (2010). Generation of induced pluripotent stem cells from human terminally differentiated circulating T cells. Cell Stem Cell 7(1):11-4.
19 Brown, ME, Rondon, E, Rajesh, D, Mack, A, Lewis, R, Feng, X, Zitur, LJ, Learish, RD and Nuwaysir, EF. (2010). Derivation of induced pluripotent stem cells from human peripheral blood T lymphocytes. PLoS One 5(6):e0011373.
20 Kim, K, Doi, A, Wen, B, Ng, K, Zhao, R, Cahan, P, Kim, J, Aryee, MJ, Ji, H, Ehrlich, LI, Yabuuchi, A, Takeuchi, A, Cunniff, KC, Hongguang, H, McKinney-Freeman, S, Naveiras, O, Yoon, TJ, Irizarry, RA, Jung, N, Seita, J, Hanna, J, Murakami, P, Jaenisch, R, Weissleder, R, Orkin, SH, Weissman, IL, Feinberg, AP and Daley, GQ. (2010). Epigenetic memory in induced pluripotent stem cells. Nature. 467(7313):285-90.