Parkinson’s disease (PD) is the second most common neurodegenerative disorder worldwide (Cooper et al., 2006). It is caused by the progressive loss of dopaminergic (DA) neurons in the substantia nigra and monoaminergic neurons in the brainstem, as well as increased microglial activation and the accumulation of characteristic proteinaceous inclusion bodies (Lewy bodies). Although several genes contribute to rare familial forms of PD, most cases of PD are sporadic and probably reflect a combination of genetic susceptibility and exposure to environmental toxins. Investigations of gene mutations linked to familial PD have revealed clues about the interplay between genetic and environmental factors in sporadic PD.
Genomic information and bioinformatics enable experiments in model organisms that provide insight into human disease. The conservation of genes and pathways that contribute to PD makes model systems valuable to understanding the mechanisms that underlie this complex disease (Meulener et al., 2005; Cooper et al., 2006; Mosharov et al., 2009) and for identifying potential susceptibility factors and therapeutic targets (Hamamichi et al., 2008). However, the development of effective therapeutic strategies has been limited by the absence of adequate in vitro, and particularly mammalian, models that accurately reproduce the pathophysiology and genetic program of PD.
Recent advances in stem cell technology might provide some traction for understanding and eventually treating PD. Takahashi and Yamanaka developed a technique that reprograms cells from fully differentiated tissues into induced pluripotent stem cells (iPSCs) through viral expression of four transcription factors (termed ‘reprogramming factors’) (Takahashi and Yamanaka, 2006). Similar to embryonic stem cells (ESCs), these iPSCs can differentiate into any cell type. This discovery suggested that patient-specific stem cell lines (Park et al., 2008) could be generated to study patient-specific disease progression and to treat patients with personalized, tissue-matched transplants. However, a major drawback of iPSC technology is that it requires viral vectors for reprogramming, which can result in residual transgene expression that can cause malignant tumors. iPSCs have great therapeutic potential for PD and other diseases if safer alternatives for reprogramming can be identified.
Building on this technology, Soldner, Hockemeyer and colleagues (Soldner et al., 2009) report a novel, virus-free method for creating iPSCs from skin biopsies of PD patients. First, the researchers set out to reprogram fibroblasts from five patients with sporadic PD (table 1 in Soldner et al., 2009). Using doxycycline (DOX)-inducible lentiviral vectors, they successfully generated iPSCs by transducing fibroblasts with either three (OCT4, SOX2, KLF4) or four (OCT4, SOX2, KLF4, MYC) reprogramming factors. Analysis of several parameters (e.g. morphology, gene expression and methylation) demonstrated that these cells are similar to pluripotent ESCs (figure 1 in Soldner et al., 2009). Injection of the iPSCs into SCID (severe combined immunodeficient) mice produced teratomas from tissues of all three embryonic germ layers, confirming their pluripotency (figure 2 in Soldner et al., 2009). The authors analysed the lentiviral integrations by Southern blotting to confirm that each of the iPSC lines was derived from independently infected fibroblasts, carrying a total of three to ten proviral copies. They further characterized the usefulness of this system by determining the reprogramming efficiencies for one of the iPSC lines, and compared transduction efficiencies of three versus four factors. This analysis revealed that despite being less efficient, reprogramming by three factors is more practical because it resulted in a more specific induction of fibroblasts to iPSCs. Importantly, these results demonstrate the efficient generation of iPSCs with a low number of proviral integrations from skin biopsies obtained from PD patients in the absence of MYC.
Next, the researchers generated DA neurons from the PD-patient-derived iPSCs, using two established protocols to induce neural differentiation. Both methods successfully produced DA neurons, as shown by immunofluorescence staining (figure 3 in Soldner et al., 2009). Importantly, they showed that there was no difference in the ability of PD-derived iPSCs, non-PD-derived iPSCs, and ESCs to generate DA neurons; another key point is that DA neurons were generated from all the iPSC lines, regardless of whether three or four reprogramming factors were used.
To create PD-patient-derived iPSCs free of viral reprogramming factors, the authors engineered loxP-flanked DOX-inducible lentiviral vectors that could be removed after integration using Cre recombinase (figure 4 in Soldner et al., 2009). Using these viruses, they transduced fibroblasts with three reprogramming factors and efficiently produced iPSCs with a low number of viral integrations. As before, the researchers performed multiple tests to prove that the cells were similar to pluripotent ESCs. Next, to investigate whether Cre-mediated vector excision could produce transgene-free cells, they transfected these iPSCs with an expression vector encoding Cre recombinase (figure 5 in Soldner et al., 2009). Remarkably, Southern blot analysis for proviral integrations revealed that this method successfully produced iPSCs free of all three reprogramming factors, and cytogenetic analysis demonstrated normal karyotype after Cre-mediated excision. Tests for pluripotency markers and teratoma formation confirmed that factor-free iPSCs maintained their pluripotent ESC-like state (figure 6 in Soldner et al., 2009). Importantly, this demonstrates that iPSCs can be reprogrammed into a self-sustaining proliferative state that does not depend on exogenous reprogramming factors for its maintenance.
Finally, the authors examined whether the removal of the viral vectors affected the molecular properties of the iPSCs. They used genome-wide microarrays to compare gene expression between ESCs and patient-derived iPSCs before and after transgene excision (figure 6 in Soldner et al., 2009). Upon excision of the viral vector, iPSCs exhibited an 80% reduction in the number of genes that were differentially expressed in comparison with ESCs (i.e. iPSCs were much more similar to ESCs after transgene removal). These results provide strong evidence that the expression of viral vectors affects the molecular characteristics of iPSCs, which further emphasizes the importance of developing strategies that efficiently remove these vectors after reprogramming. Vector-free cells are essential for the application of iPSC technology to create reliable patient-specific in vitro disease models and potential therapeutics.
Although patient-specific iPSCs offer an extremely promising new angle for the PD community, there are several challenges in applying this approach that remain before the disease can be understood. For example, the authors’ method for transgene excision can leave a long terminal repeat (LTR) footprint at the integration sites, which could cause gene disruption. However, transposons like piggyBac have recently been used to overcome this problem (Woltjen et al., 2009). There is also an obvious challenge to reproduce a disease pathology that takes years to develop in vivo, and primary neurons in vitro have a short lifespan. Modeling the physiological effects of aging might be possible through the use of environmental or genetic factors to accelerate the disease progression. Differences between the environments of cultured cells and cells in vivo can also complicate studies. For example, iPSCs cannot recapitulate the complex interactions between various types of neurons and neuronal networks that are formed in the brain and their contribution towards pathophysiology of PD; therefore, it would not be possible to mimic disease progression precisely in vitro. Each research system offers advantages and limitations, so results culminating from a variety of approaches are likely to be the most informative.
This breakthrough in iPSC technology has an immediate impact on stem cell research. In the near future, it can be applied towards the study of diseases including, but certainly not limited to, PD. This type of research holds the exciting potential to rapidly transform the biomedical field and bridges the gap between bench research and bedside medicine. Using virus-free iPSCs should help create model organisms to accelerate identification of the root causes and cures for many diseases.
