A high proliferation rate is required for cell reprogramming and maintenance of human embryonic stem cell identity

Abstract

Human embryonic stem (hES) cells show an atypical cell-cycle regulation characterized by a high proliferation rate and a short G1 phase. In fact, a shortened G1 phase might protect ES cells from external signals inducing differentiation, as shown for certain stem cells. It has been suggested that self-renewal and pluripotency are intimately linked to cell-cycle regulation in ES cells, although little is known about the overall importance of the cell-cycle machinery in maintaining ES cell identity. An appealing model to address whether the acquisition of stem cell properties is linked to cell-cycle regulation emerged with the ability to generate induced pluripotent stem (iPS) cells by expression of defined transcription factors. Here, we show that the characteristic cell-cycle signature of hES cells is acquired as an early event in cell reprogramming. We demonstrate that induction of cell proliferation increases reprogramming efficiency, whereas cell-cycle arrest inhibits successful reprogramming. Furthermore, we show that cell-cycle arrest is sufficient to drive hES cells toward irreversible differentiation. Our results establish a link that intertwines the mechanisms of cell-cycle control with the mechanisms underlying the acquisition and maintenance of ES cell identity.

Figure 1. The characteristic cell cycle signature of hES cells is acquired early in reprogramming

(A) Cell cycle distribution analyzed by EdU incorporation of the Tra-1-60 and Tra-1-81 positive and negative populations of cells obtained 12 days after the first infection (OSKC) of keratinocytes. H9 cells were included as a positive control for staining. Two independent experiments were performed and 5,000 cells are represented in each dot plot. Ke=Keratinocytes. (B) Percentage of Nanog⁺/BrdU⁺ cells in H9 and nascent hiPS cells compared to the percentage of BrdU⁺ cells in keratinocytes. A minimum of 10 Nanog⁺ colonies were evaluated in two independent experiments. Cells were incubated with 10 μM BrdU for 30 minutes. Ke=Keratinocytes. (C) The graph shows a quantification of the percentage of BrdU⁺ cells in early emerging hiPS colonies obtained 14 days after the first infection of dFib-Oct4^GFP with OSKC. H9 cells as well as the original dFib-Oct4^GFP population of cells were included for comparison. A minimum of 10 GFP⁺ colonies were evaluated in two independent experiments. Cells were incubated with 10 μM BrdU for 30 minutes. (D) BrdU immunofluorescence in hiPS cells obtained 10 days after the first infection of dFib-Oct4^GFP with OSKC. A representative example is shown. (E and F) The graphs show a quantification of the percentage of phospho-histone H3⁺ cells in hiPS colonies obtained 12 or 14 days after the first infection of keratinocytes (E) or dFib-Oct4^GFP (F) respectively, with OSKC. H9 cells as well as the original population of cells were included for comparison. A minimum of 10 GFP⁺ (E) or Tra-1-60⁺ (F) colonies were evaluated in two independent experiments. Ke=Keratinocytes.

Figure 2. Reprogramming efficiency is influenced by modulations in the expression of proteins involved in the G1 to S-phase transition

(A–D) Human cultures of keratinocytes (A–C) or BJ fibroblasts (D) were infected with retroviruses encoding the 4 factors, OSKC and either lentiviruses encoding shRNAs against the indicated proteins (A) or retroviruses encoding the noted proteins (B, C and D). Relative reprogramming efficiencies, normalized to the efficiency observed in pMX-GFP or pLVTHM-infected keratinocytes correspondingly, are shown with the fold changes indicated (upper graphs). Uninfected cells were used as a negative control for all experiments. n=number of independent experiments. All error bars depict the SEM. Sc= Scrambled shRNA. (E–G) Percentage of BrdU⁺ cells in keratinocytes (E, F) or BJ fibroblasts (G) 4 days after infection with OSKC plus either, retroviruses expressing the noted proteins or lentiviruses expressing the indicated shRNAs. Note the different percentage of BrdU⁺ cells between both experimental controls, pMX- or pLVTHM-infected keratinocytes (E, F), due to different viral delivery method used (retrovirus vs lentivirus). Two independent experiments were performed and the results are represented as averages ± SEM of at least 300 cells scored in three different fields in both experiments. Cells were incubated with 10 μM BrdU for 60 minutes. Sc= Scrambled shRNA. (H) Left graph, Number of GFP⁺ colonies per 10⁵ cells seeded following the infection of dFib-Oct4^GFP with the respective retroviruses. Two independent experiments were performed and the results represent the average ± SEM of two biological replicates per experiment. Middle and right panels, Morphological aspect as well as GFP expression or Nanog staining of colonies obtained from the different conditions 8 (middle panel) or 16 (right panel) days after the first infection. Scale bar: 100 μM.

Figure 3. Cell cycle arrest in hES cells induces differentiation

(A) Western blot analysis of GFP, p15, p16 and p21-inducible H9 cell line protein extracts 24 hours after treatment with 200ng/ml of doxycyclin. (B) Cumulative BrdU labeling curves of the GFP, p15, p16 and p21-inducible H9 cell lines. After 24 hours of culture with or without doxycyclin, fresh BrdU was added to the media every two hours and the BrdU labeling index was analyzed at different time points as indicated. The BrdU labeling index indicates the proportion of DAPI-stained nuclei that were positive for BrdU incorporation. Each time point represents the average and standard deviation of triplicate counts of at least 500 nuclei in different fields in 2 biological replicates. The upper dashed line represents the maximum labeling index which was similar independent of doxycyclin treatment. The time necessary to reach the maximum labeling index corresponds to the total cell cycle length minus the length of the S-phase (T^G2+M+G1) [21]. (C) Cell morphology of the same cell lines in the absence of doxycyclin (Day 0) or 2, 6 and 10 days after p15, p16 or p21 induction. The small hES cell-like colonies observed in the case of p15 and p16 expressing lines are outlined by dashed lines. Scale bar: 100μM. (D) Immunofluorescence analysis of BrdU and Oct4 in the p15, p16 and p21-inducible H9 cell lines after 10 days of induction with doxycyclin. Oct4⁺ colonies are outlined with dashed lines. Scale bar: 100μM. Cells were incubated with 10 μM BrdU for 30 minutes. (E) Western blot analysis of the indicated proteins in the GFP and p21-expressing H9 cell lines with or without doxycyclin induction for 10 days. (F) Immunofluorescence analysis of the differentiation markers for the mesodermal (SMA), trophectodermal (Cdx2), ectodermal (Tuj1) and endodermal (GATA4, AFP, Albumin, FoxA2) germ layers in the p15, p16 and p21-inducible H9 cell lines following 10 days of induction with 200ng/ml of doxycyclin. Scale bar: 100μM.

Figure 4. Differentiation mediated by p21 expression in hES cells is an irreversible process

(A) Schematic representation of the experiment performed. Briefly, p21 expression was induced and maintained as indicated by adding doxycyclin to the media for the indicated number of days, followed by its withdrawal at each indicated time point. All cells were analyzed following a total of 10 days in culture. (B) Cell morphology of the GFP and p21-inducible cell lines maintained for 10 days in the indicated conditions. The small hES cell-like colonies observed after a single day of p21 induction are outlined by dashed lines. Scale bar: 100μM. (C) Immunofluorescence analysis of the differentiation markers for the mesodermal (SMA), trophectodermal (Cdx2), ectodermal (Tuj1) and endodermal (AFP, Albumin, FoxA2) germ layers in the p21-inducible H9 cell line after 3 days of induction with 200ng/ml of doxycyclin followed by 7 days of continued culture after its withdraw. Scale bar: 100μM. (D) Real time PCR analysis of the pluripotent marker (Oct4) and the specific differentiation markers for mesodermal (Msx1), trophectodermal (Cdx2), ectodermal (Pax6, FGF5) and endodermal (GATA4, GATA6, AFP, Albumin, FoxA2) germ layers in the p21-inducible H9 cell line after the induction conditions indicated. Data are shown as relative averages ± SEM of two biological replicates analyzed in triplicate.