ABSTRACT
The advent of human induced pluripotent stem cells (hiPSC) is revolutionizing many research fields including cell-replacement therapy, drug screening, physiopathology of specific diseases and more basic research such as embryonic development or diseases modeling. Despite the large number of reports on reprogramming methods, techniques in use remain globally inefficient. We present here a new optimized approach to improve this efficiency. After having tested different monocistronic vectors with poor results, we adopted a polycistronic cassette encoding Thomson's cocktail OCT4, NANOG, SOX2 and LIN28 (ONSL) separated by 2A peptides. This cassette was tested in various vector backbones, based on lentivirus or retrovirus under a LTR or EF1 alpha promoter. This allowed us to show that ONSL-carrier retrovectors reprogrammed adult fibroblast cells with a much higher efficiency (up to 0.6%) than any other tested. We then compared the reprogramming efficiencies of two different polycistronic genes, ONSL and OCT4, SOX2, KLF4 and cMYC (OSKM) placed in the same retrovector backbone. Interestingly, in this context ONSL gene reprograms more efficiently than OSKM but OSKM reprograms faster suggesting that the two cocktails may reprogram through distinct pathways. By equally mixing RV-LTR-ONSL and RV-LTR-OSKM, we indeed observed a remarkable synergy, yielding a reprogramming efficiency of >2%. We present here a drastic improvement of the reprogramming efficiency, which opens doors to the development of automated and high throughput strategies of hiPSC production. Furthermore, non-integrative reprogramming protocols (i.e. mRNA) may take advantage of this synergy to boost their efficiency.
