Research Highlights

Formation of induced pluripotent stem cells without viral vectors

Evaluation of: Kaji K, Norrby K, Paca A, Mileikovsky M, Mohseni P, Woltjen K: Virus-free induction of pluripotency and subsequent excision of reprogramming factors. Nature doi:10.1038/nature07864 (2009) (Epub ahead of print).

The scientific community has been galvanized by the demonstration of cellular reprogramming of somatic cells to pluripotent stem cells [1]. Despite this excitement, several technical issues have dampened enthusiasm for the practical application of this technology. Most important in this regard is the fact that current reprogramming methods rely on viruses to deliver reprogramming factors. These viruses can have a negative impact upon the cells in which they are introduced. Thus, development of technology that circumvents the need to use hazardous viruses has been sought after.

In this paper, the authors explored the possibility that a single plasmid vector could be used to effectively introduce four reprogramming factors into somatic cells, and then remove them using Cre recombinase. To accomplish this they made a 2A-peptide-linked reprogramming cassette containing coding sequences of c-myc, Klf4, Oct4 and Sox2 flanked by loxP sites. DNA transfection of this construct resulted in reprogrammed colony formation between days 20 and 30. In these experiments, the authors estimated 2.5% reprogramming efficiency overall. The differential efficiency of expression for reprogramming factors was examined, as was the relationship between transgene copy number and reprogramming. However, no direct data were presented to explain the high reprogramming efficiency.

To remove the exogenous reprogramming factor gene cassette, Cre was transiently expressed, resulting in specific excision of the transgene cassette at the loxP sites. They noted at first that Cre-treated cells differentiated. To prevent this Cre-associated differentiation the colonies were treated with the FGF receptor inhibitor PD173074. Colonies that were maintained and expanded in the presence of PD173074 were stable for at least five passages. After Cre-mediated transgene excision, cells maintained expression of endogenous reprogramming factors. Furthermore, expression of other pluripotency genes also continued in the Cre-treated cells.

The reprogrammed and Cre-treated cells were examined for their ability to differentiate in an embryoid body formation assay. During embryoid body formation, pluripotency markers were downregulated, and upregulation of markers for the three primary germ layers was observed. In addition, β-tubulin-positive neurons were generated in a monolayer neural differentiation protocol. In vivo functional assays followed, in which the cells were checked for mixed teratoma formation and for high induced pluripotent stem (iPS) cell contribution chimera formation. The reprogrammed cells were successful in forming both teratomas and high percentage contribution to chimeras.

Transposon-mediated induction of pluripotent stem cells

Evaluation of: Woltjen K, Michael IP, Mohseni P et al.: piggyBac transposition reprograms fibroblasts to induced pluripotent stem cells. Nature doi:10.1038/nature07863 (2009) (Epub ahead of print).

An alternative nonviral method was reported by the same research group in which the utility of a transfection mediated piggyBac transposon gene-delivery system was examined for reprogramming murine fibroblasts. A piggyBac transposon plasmid, also containing a 2A-peptide-linked reprogramming cassette but under the control of a doxycycline-regulated promoter, was transfected into fibroblasts. Because piggyBac transposons are not silenced like retroviruses, the reprogramming cassette was linked to βgeo to monitor tightness of the doxycycline regulation. The reverse tetracycline transactivator, rtTA, was supplied in trans by mouse embryonic fibroblasts derived from a ROSA26 rtTA-IRES-GFP knockin.

After transfection murine fibroblasts formed reprogrammed colonies around days 8–10. Reprogramming was essentially dependent upon continued administration of doxycycline. An exacting bell-shaped dose-response curve was observed for doxycycline administration versus reprogramming. Doxycycline-independent growth of the colonies appeared between days 12 and 24. After day 16 alkaline phosphatase⁺ cells were observed and by days 20–22, SSEA1⁺ and Nanog⁺ colonies were visible by staining. Reverse transcription-PCR analysis confirmed expression of pluripotency markers Dax1, Eras, Fbxo, Foxd3, Nanog, Rex1 and Zfp296. Transgene copy number was determined by genomic Southern blot. Excision of the transgene was accomplished by transient expression of transposase. Variable excision was observed in multicopy transgene cell lines. Although no direct data were presented it was hypothesized that transgene location/integration site may affect transposase-mediated excision.

Single copy transgenic lines were selected for analysis of transgene excision following transient transposase expression. Genomic PCR confirmed that cell lines were negative for transgene sequences, and sequence analysis confirmed that excision of the transgene resulted in wild-type genomic structure in ten out of 11 lines. After excision these cell lines remained reprogrammed and expressed endogenous pluripotency genes. No deleterious effect of transposase was mentioned and no use of FGF receptor inhibitor was described.

Teratoma assays confirmed the ability of iPS cells to differentiate in vivo, resulting in the formation of mixed tissue teratomas containing cells from all three primary germ layers. Cell lines were tested for chimera contribution using the tetraploid embryo complementation assay. Reprogrammed cells showed 100% contribution to chimera development, including the germ cells of the derived chimera.

These experiments were extended to the human system by introducing mouse reprogramming factors into human embryonic fibroblasts. Human iPS cells appeared and were picked between days 14 and 28 following transfection. Four out of five alkaline phosphatase⁺ clones became doxycycline independent.

The sequences of events surrounding these reprogramming events are similar. Transient expression of the reprogramming factors c-myc, Klf4, Oct4 and Sox2 induce a iPS phenotype within 7–10 days after expression. Subsequently, endogenous expression of reprogramming factors is activated and maintained by reprogrammed cells in a stable fashion. Further exogenous expression is not required after reactivation of endogenous reprogramming factors, which are coordinated with expression of other markers of the pluripotent stem cell compartment. iPS cells derived from the virus-free approach participate in both in vitro and in vivo differentiation programs. These iPS cells are indistinguishable from iPS cells derived from virus-mediated approaches.

Generation of β cells by reprogramming other cellular compartments of the pancreas

Evaluation of: Zhou Q, Brown J, Kanarek A, Rajagopal J, Melton DA: In vivo reprogramming of adult pancreatic exocrine cells to β cells. Nature 455(7213), 627–632 (2008).

Most cells in adult organisms are considered to be terminally differentiated, a state that until recently was generally considered to be irreversible. With a similar approach to that originally reported to reprogram fibroblasts to pluripotent cells in vitro[1], Zhou et al. have reprogrammed murine pancreatic exocrine cells into another pancreatic cell type: insulin-producing β cells. Using adenoviruses, the authors ectopically expressed a pool of nine transcription factors previously implicated in the development of β cells and their precursors. Subsequently, they removed genes from this pool to end up with three key factors that, when injected into the pancreas, were essential to the reprogramming: Ngn3, Pdx1 and Mafa.

Using green fluorescent protein (GFP) expression and exocrine markers to track the progress of infected cells in the pancreata of adult mice, 95% of infected cells were exocrine cells. Subsequently, 20% of those cells expressed insulin at levels comparable to endogenous β cells within 10 days. Induced cells were similar to endogenous β cells both morphologically and by molecular marker expression profiling (positive for Glut2, GCK, PC1/3, NeuroD, Nkx2.2 and Nkx6.1). Expression of the three transgenes diminished after 1 month. Ngn3, which is required to induce endocrine differentiation but not expressed in mature β cells, disappeared. On the other hand, endogenous expression of Pdx1 and Mafa was detected in induced β cells. In conjunction with the absence of molecular markers for exocrine cells, ductal cells and other endocrine cell types, these results indicate that a genetic program specific for β cells had become active.

Functionality of induced β cells was tested by killing endogenous β cells with streptozotocin prior to adenoviral infections. Increased serum insulin levels and normalization of glucose tolerance in streptozotocin-treated animals resulted after reprogramming of infected cells. This study marks the most thorough reprogramming of adult cells into insulin-producing cells. The reprogrammed cells are nearly indistinguishable from endogenous β cells in both appearance and function. One obvious exception is that induced cells do not form islet clusters with other endocrine cells, but rather exist singly or in small clusters. They were, however, able to establish contacts with blood vessels amongst the remaining exocrine cells.

Compared with recent reports of reprogramming adult cells into pluripotent cells, the current study resulted in relatively high efficiency and rapid rate of reprogramming. This was apparently due, in part, to the fact that exocrine cells and β cells are closely related. Interestingly, the reprogrammed cells in this study appeared not to replicate. Moreover, lack of expression of Sox9 and Hnf6, markers of precursors for both exocrine and endocrine cells, indicates that the reprogrammed cells did not dedifferentiate or transition through a precursor stage before activating the β-cell program.

The in vivo nature of this study complements a previous report describing transplantation of human embryonic stem cell-derived pancreatic precursors into adult mice, which was apparently necessary for complete development into functional endocrine cells [2]. Both provide strong evidence that functional β cells can be generated given the proper environment, and the current study demonstrates increased specificity towards β cells.

The need to create large quantities of cells is still a significant obstacle for cell-replacement therapy for Type I diabetes. However, the current report poses the possibility of inducing the formation of new β cells within an organism, without having to transition through a stem-cell stage. Before this method could be translated into clinical applications, a few issues must be addressed. First, the use of nonviral methods to induce reprogramming is desirable. Second, the fact that reprogrammed cells in this study did not replicate indicates that extensive epigenetic reprogramming might not have been necessary. It remains to be determined whether the human pancreas also exists in such a permissive state. Hopefully a combination of methods from recent discoveries will prove fruitful in the near future for clinical applications.

Pancreatic β cells can be reprogrammed into pluripotent cells

Evaluation of: Stadtfeld M, Brennand K, Hochedlinger K: Reprogramming of pancreatic β cells into pluripotent stem cells. Curr. Biol. 18(12), 890–894 (2008).

To further define the utility of reprogramming somatic cells it is important to understand which cell types can be reprogrammed. In this paper, a murine experimental system was established to test the hypothesis that terminally differentiated β cells can be reprogrammed by the same factors used to reprogram fibroblasts [1]. Expression of GFP under control of the Pdx1 promoter system was used to evaluate whether isolated islets containing β cells can be cultured in iPS media. After 12 days, isolated islets did not spontaneously acquire a reprogrammed phenotype. Subsequently, these cells were shown to be infectable with a lentivirus encoding the tdTomato marker.

To isolate adult β cells for reprogramming, transgenic mice were crossed to form the experimental strain harboring ROSA26-LacZ and RIP-CRE under control of the Ins promoter. This experimental strain was shown to express LacZ exclusively in the β cells. Islets from RIP-Cre/lacZ mice were then isolated and infected with lentivirus encoding Oct4, Sox2, c-myc and Klf4 under tetracycline-inducible control. Out of these infected and induced cultured cells emerged colonies with the morphological characteristics of reprogrammed pluripotent cells. Two distinct populations of colonies were identified based on β-galactosidase staining. The reasons for lacZ^- colonies were not identified. LacZ⁺ colonies were evaluated for expression of stem cell-specific molecular markers, revealing expression of SSAE1, Nanog, Sox2 and Oct4 protein, and c-myc, Klf4, Oct4 and Sox2 RNA.

Reprogrammed cells were tested for pluripotency by two classical assays. First, reprogrammed cells were introduced into ectopic sites in nude mice, and exhibited the formation of mixed tissue teratomas. In the second assay the cells were injected into day-4.5 blastocysts, and subsequently neonatal mice were scored for a high percentage of iPS chimeras.

These experiments suggest that not only replication-competent cells like fibroblasts can be reprogrammed, but also that terminally differentiated cell lineages are also amenable to reprogramming. The use of a terminally differentiated tissue was expressly intended as a proof of principle. If β cells can be reprogrammed then any terminally differentiated tissue might also follow the same fate. This observation should ensure that it is highly possible to develop disease model-specific iPS cells. Thus, tissue source should not be a limitation or impediment for production and isolation. Furthermore, these papers together make a strong case that protocols can be developed to reprogram many somatic cells directly into other somatic compartments.

Financial & competing interests disclosure

The authors are supported by grants from the Larry L Hillblom Foundation. A Hayek is a consultant for Novocell, Inc. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.