Scientists at the Whitehead Institute for Biomedical Research and the Massachusetts Institute of Technology have developed a new method for producing human induced pluripotent stem cells from which the lentiviral reprogramming factors have been excised.
A Whitehead researcher said this week that the cells, derived from patients with sporadic Parkinson's disease, remained pluripotent and revealed a gene-expression profile that more closely resembled human embryonic stem cells than human iPS cells still carrying the reprogramming factors.
The development is noteworthy because the use of lentiviral vectors, though a very efficient way to infect cells, can cause cellular transformation and tumor formation, said Whitehead postdoc Dirk Hockemeyer. Excising the viral reprogramming factors would reduce the risk of oncogenic transformation.
"Although this technique can be used to study most diseases where you would like to have an in vitro model, we focused on one that is of particular interest, because one reason why this disease is poorly understood is a lack of in vitro models," he said.
Frank Soldner, who is also a postdoc at Whitehead, said the Whitehead team's method, which employs basic molecular biology techniques, is available in almost any lab "around the world," and can be generalized by researchers.
"And we hope that is the case," he said. "We hope that they will use the system to make iPS cells, and to study other diseases in vitro."
The technique, described in this week's issue of Cell, "involved modifying the same lentiviruses as vectors that were used [typically]," which could be excised after integration using the enzyme cre-recombinase, said Soldner. The Whitehead Institute team's method differed in that they replaced the human ubiquitin promoter of the FUGW-loxP lentivirus with a DOX-inducible, minimal cytomegalovirus followed by the human cDNAs for OCT4, KLF4, or SOX2.
Soldner and Hockemeyer are co-authors on this publication.
Upon proviral replication, the loxP site in the 3′ long terminal repeat is duplicated into the 5′ LTR, resulting in an integrated transgene flanked by a loxP site in both LTRs. The cre recombinase enzyme then excises the loxP-flanked transgene.
They derived these iPS cells that were free of reprogramming factors from the fibroblasts of five patients with sporadic Parkinson's disease. The scientists subsequently differentiated these iPS cells into dopaminergic neurons.
The scientists chose Parkinson's as a proof of principle because of the dearth of reliable experimental models for the disease and other neurodegenerative disorders, said Soldner.
Describing a hypothetical drug-discovery application for the method, Soldner said investigators could take dopaminergic neurons without viral integrations derived from the fibroblasts of Parkinson's patients, and compare them against those derived from controls.
"Then you would try to determine the difference, if there is any, between these two cell types," he said. "That can be done on a very small scale, but once this is set up, and differences between the patient-derived cells and the control cells have been identified, [researchers] can take a chemical library and say, 'What do we have to put on a cell, in order to make a cell from a patient now behave normally?,'" said Soldner.
[ pagebreak ]
The investigators are currently trying to differentiate human iPS cells that lack a viral reprogramming factor into very specific cell types. For instance, in Parkinson's disease "we are trying to differentiate these cells into a very specific type of dopaminergic neuron," said Hockemeyer.
The next problem, which the researchers do not yet know if they will face, is that a lot of neurodegenerative diseases are long-lasting or late-onset diseases. For example, the average age of onset for Parkinson's is 40 to 50 years or older, and the maximum duration during which cells can be maintained in cell culture is several weeks.
"Therefore it may be necessary to induce a phenotype which can be used to study a Parkinson's-related cellular phenotype or screen for drugs that avoid such a phenotype," said Hockemeyer.
He added that one may need to stimulate or stress the system additionally to further study these phenotypes. "We are in the process of setting up these systems so that we can start doing drug screens, and study pathological differences in these cells that would be related to Parkinson's, to better understand the molecular mechanisms of the disease," Hockemeyer said.
Soldner and Hockmeyer's paper coincides with last week's publication in Nature of a paper by scientists at the Samuel Lunenfeld Research Institute in Canada. In their study, the researchers used piggyBac transposition and doxycycline-inducible transcription factors to reprogram murine and human embryonic fibroblasts. The iPS cells generated expressed characteristic pluripotency markers and succeeded "in a number of rigorous differentiation assays," according to the study authors.
The idea behind the Canadian study is the same as the one driving Soldner and Hockmeyer's paper, said Soldner. "Everybody agrees that removing the transgenes after you reprogram the cells is essential if you want to develop reliable model systems or if you want to use the cells for therapeutic applications," he said.
Both groups introduced the transgene, reprogrammed the cells, and removed the genes afterwards. The main difference is that the Canadian scientists used murine transcription factors to transform human cells. The Cell paper, by comparison, is a proof-of-principle study that demonstrates that human iPS cells free of the reprogramming factor can be generated.
"They did not follow their study all the way through," said Hockemeyer. "It seems that these are two complementary approaches. Our study is just that transition to the human system."
However, he stressed it is difficult to tell which one will become the method of choice. "We have molecular evidence that removal of the viral transfection factor is possible in iPS technology, and that is unique in our study," Hockemeyer said.