Linking the p53 tumour suppressor pathway to somatic cell reprogramming

Abstract

Reprogramming somatic cells to induced pluripotent stem (iPS) cells has been accomplished by expressing pluripotency factors and oncogenes, but the low frequency and tendency to induce malignant transformation compromise the clinical utility of this powerful approach. We address both issues by investigating the mechanisms limiting reprogramming efficiency in somatic cells. Here we show that reprogramming factors can activate the p53 (also known as Trp53 in mice, TP53 in humans) pathway. Reducing signalling to p53 by expressing a mutated version of one of its negative regulators, by deleting or knocking down p53 or its target gene, p21 (also known as Cdkn1a), or by antagonizing reprogramming-induced apoptosis in mouse fibroblasts increases reprogramming efficiency. Notably, decreasing p53 protein levels enabled fibroblasts to give rise to iPS cells capable of generating germline-transmitting chimaeric mice using only Oct4 (also known as Pou5f1) and Sox2. Furthermore, silencing of p53 significantly increased the reprogramming efficiency of human somatic cells. These results provide insights into reprogramming mechanisms and suggest new routes to more efficient reprogramming while minimizing the use of oncogenes.

Fig. 1. Increased generation of iPS cells by blocking p53 and p21

(a) MEFs were infected by retroviruses encoding 3 factors (Oct4/Sox2/Klf4), 2 factors (Oct4/Sox2), Klf4, c-Myc or GFP. Four days after infection, the protein levels of p53, Arf, and p21 were analyzed by Western Blotting. α-Tubulin was used as a loading control. (b) MEFs were infected by 3F (Oct4/Sox2/Klf4) in combination with mock, control shRNA (GFP) and p53 shRNA (#1 and #2). Emerging colonies of iPS cells were visualized by immunostaining with anti-Nanog antibody using an Avidin Biotin Complex (ABC) method. (c) The fold change in the number of Nanog-positive colonies compared to mock (n=4). For all figures in this study, error bars indicate s.d. p53 knockdown efficiency was examined by western blot. (d) MEFs were infected by 3F in combination with mock, p53 shRNA and p21 shRNA. Four days later, half of cells were treated with Nutlin 3a (0, 3, 10μM) and analyzed for p53 level. The remainder were stained for Nanog-positive colonies. (e) Immunostaining of Nanog positive colonies generated from p53(+/+), p53(+/−) and p53(−/−) MEFs by 3F showed p53 dose-dependent decrease of colony number. (f) Retroviral infection of p53 into p53(−/−) MEF decreased the number of Nanog-positive colonies induced by 3F. p53 and p21 levels on day 3 after infection were analyzed. (g) Nutlin 3a dramatically reduced reprogramming of p53(+/+) MEFs, but not on p53(−/−) MEF. (h) Fold change in the number of Nanog-positive colonies by p21 shRNA (n=4). p21 knockdown efficiency was examined by western blot.

Fig. 2. Modulation of p53 activity alters reprogramming efficiency

(a) and (b) Fold change in the number of 3F induced Nanog-positive colonies by Arf shRNA or by Arf/Ink4a shRNA compared to control shRNA (n=3). Protein knockdown efficiency was examined by western blot. (c) 3F induced Nanog-positive colonies from wild type (+/+) and homozygous (3SA/3SA) MEFs (n=3).

Fig. 3. Generation and characterization of 2F-p53KD-iPS cells by p53 downregulation

(a) Morphology and GFP fluorescence of 2F-p53KD-iPS cell lines. GFP expression is silenced in clone #6. (b) Alkaline phosphatase staining of 2F-p53KD-iPS cell lines. DAPI was used to visualize cell nuclei. (c) Protein levels of Nanog, Oct4, Sox2, Klf4, c-Myc, p53 in 2F-p53KD-iPS cell lines are shown. α-Tubulin was used as loading control. (d) Embryoid bodies (EBs) of 2F-p53KD-iPS cell clones on day 6 of differentiation. EBs were transferred to gelatinized dishes on day 3 to 5 for further differentiation. On day 14, EBs were subjected to immunofluorescence for α-fetoprotein (AFP)/Foxa2 (endoderm), α-sarcomeric actin/GATA4 (mesoderm) and Tuj1/GFAP (ectoderm). (e) Immunofluorescence of teratoma from 2F-p53KD-iPS cells by antibodies against AFP/Foxa2 (endoderm), α-sarcomeric actinin/Chondroitin (mesoderm), Tuj1/GFAP (ectoderm) showed spontaneous differentiation into all three germ layers. (f) Adult chimeric mice obtained from 2F-p53KD iPS lines (#1 and #6) and non-chimeric mouse in C57BL/6J host blastocysts. (g) As of the date of submission, the mating of offspring from clone #6 chimera to a C57BL/6J female generated 1 agouti pup (blue arrow), that together with PCR analysis (not shown) indicate germ line transmission of the 2F-iPS genome.

Fig. 4. Downregulation of p53 activity increases reprogramming efficiency of human somatic cells

(a) Human embryonic fibroblasts were infected with retroviruses encoding Oct4/Sox2/Klf4 (3-F) or Oct4/Sox2/Klf4/c-Myc (4-F) factors in combination with lentiviruses expressing control-or p53-shRNA. Emerging colonies of iPS cells were immunostained with anti-Nanog antibody. p53 knockdown efficiency was examined by western blot. (b) Human primary keratinocytes were co-infected with 4-F and retroviruses expressing GFP or p53-DD. Two weeks later, cells were stained for AP activity. Expression of p53-DD resulted in stabilization of wild-type p53 (lower panels). Actin was used as a loading control. (c) The bars represent the average number of iPS-like colonies obtained from 10 keratinocytes reprogrammed with 3F or 4F and retroviruses encoding GFP or p53-DD, in the absence or presence of Nutlin3a (n=3). iPS-like colonies were scored as having hES-like morphology and positive AP staining. Due to the numerous colonies generated in 4F p53-DD keratinocytes, quantification was done using 10 cells. (d, e) Colonies of human keratinocyte-derived iPS cells generated by 3F and p53-DD display strong immunoreactivity for pluripotency-associated transcription factors and surface markers (d) and differentiate in vitro into cell types that express markers of endoderm (α-fetoprotein, FoxA2), mesoderm (GATA4, sarcomeric α-actinin), and ectoderm (Tuj1, TH) (e).