Serum starvation induced cell cycle synchronization facilitates human somatic cells reprogramming

Abstract

Human induced pluripotent stem cells (iPSCs) provide a valuable model for regenerative medicine and human disease research. To date, however, the reprogramming efficiency of human adult cells is still low. Recent studies have revealed that cell cycle is a key parameter driving epigenetic reprogramming to pluripotency. As is well known, retroviruses such as the Moloney murine leukemia virus (MoMLV) require cell division to integrate into the host genome and replicate, whereas the target primary cells for reprogramming are a mixture of several cell types with different cell cycle rhythms. Whether cell cycle synchronization has potential effect on retrovirus induced reprogramming has not been detailed. In this study, utilizing transient serum starvation induced synchronization, we demonstrated that starvation generated a reversible cell cycle arrest and synchronously progressed through G2/M phase after release, substantially improving retroviral infection efficiency. Interestingly, synchronized human dermal fibroblasts (HDF) and adipose stem cells (ASC) exhibited more homogenous epithelial morphology than normal FBS control after infection, and the expression of epithelial markers such as E-cadherin and Epcam were strongly activated. Futhermore, synchronization treatment ultimately improved Nanog positive clones, achieved a 15-20 fold increase. These results suggested that cell cycle synchronization promotes the mesenchymal to epithelial transition (MET) and facilitates retrovirus mediated reprogramming. Our study, utilization of serum starvation rather than additional chemicals, provide a new insight into cell cycle regulation and induced reprogramming of human cells.

Figure 1. Serum starvation induced cell cycle synchronization.

(A) and (B): Representative FACS data showed cell cycle arrested at G0/G1 phase (0 h) after serum starvation for 18 h. Re-feeding with FBS resulted in a clear accumulation of S and G2/M phase at the indicated time. (C): Modfit LT analysis revealed a time dependant increase in G0/G1 fraction upon serum deprivation. *P<0.01 vs control (n = 3). (D) and (E): A summary of cell cycle distribution after release from starvation. (F): Immunofluorescence revealed HDF and ASC increased BrdU incorporation after synchronization. Green fluorescence represents BrdU; the nuclei were labeled with DAPI (blue). Scale Bars = 20 µm. (G): Quantitative analysis indicated starvation and re-feed procedure increased BrdU positive cells in HDF. Counting was based on Brdu and DAPI staining. Positive cells were counted from a total number of at least 300 cells per well (n = 3) in randomly selected field. *P<0.01 vs unsynchronized control (n = 3). (H): Apoptosis assay of HDF after serum deprivation by Hoechst 33258 staining. *P<0.05 vs control (n = 3).

Figure 2. Cell cycle synchronization increased retrovirus mediated transduction.

(A): Fluorescence microscope figures showed the improved infection of GFP in HDF after synchronization. To avoid misleading results from excessive virus infection, 0.1 ml concentrated GFP retrovirus was added to HDF in one well of a 12-well plate. The samples were analyzed at 72 h after infection. Scale Bars = 100 µm. (B): About 46% cells were GFP positive in unsynchronized HDF. A minimum of 30,000 events were acquired for each sample. (C): Representative FACS results showed that GFP positive HDF cells distinctly increased after synchronization and release for 14 h (left), 18 h (middle), and 22 h (right). (D) and (E): Statistical analysis of FACS data identified that the infection efficiency was enhanced about 1.9-fold in synchronized HDF and ASC. **P<0.01 vs unsynchronized control (n = 3). (F): Quantitative RT-PCR for transgene expression of Oct4 and Sox2 on day 5 after infection. **P<0.05 vs 22 h release (n = 3). (G): Quantitative RT-PCR for endogenous expression of Oct4 and Nanog on day 6 after OSKM (Oct4, Sox2, Klf4, and c-Myc) infection. **P<0.01 vs 22 h release (n = 3).

Figure 3. Synchronization procedure promoted MET and reprogramming.

(A) and (B): Microscope images of HDF and ASC on day 6 after reprogramming factors infection. Higher magnification images of the boxed regions in (A) and (B) are shown in right. Scale Bars = 200 µm (left), 100 µm (right). (C): Quantitative RT-PCR revealed the up-regulation of epithelial markers in infected HDF. **P<0.05 vs 22 h release (n = 3). (D): The expression of mesenchymal associated genes was suppressed. *P<0.01 vs FBS control, **P<0.05 vs 22 h release (n = 3). (E) and (F): Synchronization increased TRA-1–60 and Nanog positive clones both in HDF and ASC. Counting was based on TRA-1–60 or Nanog plus DAPI staining; the data was collected from 3 independent experiments.

Figure 4. Exhibition of hESCs related characters in hiPSCs.

(A): Quantitative PCR assay revealed the relative expression of endogenous Oct4, Sox2 in iPSC lines and ESCs. The data are shown as relative averages ± SEM calculated from triplicate samples. (B): Expression of endogenous Nanog and Rex1 in HDF and ASC derived iPS cell lines. The parental cells of origin and hESCs were used as controls. (C): Quantitative PCR for relative expression of FGF4 and Lin28 in hiPS cell lines. (D): Results from quantitative PCR indicated that exogenous Oct4 and Sox2 transgenes were silenced in hiPSCs. Original cells and cultures on day 5 after retroviral infection were used as controls. (E) and (F): Reprogrammed hiPSCs exhibited hESC morphology and expressed AP (alkaline phosphatase). Scale bar = 200 µm. Immunofluorescence analysis demonstrated that HDF-hiPSCs (E) and ASC-iPSCs (F) expressd TRA-1–81, TRA-1–60, Nanog, and E-Cadeherin. DAPI was used to mark nuclei. Scale Bars = 50 µm. (G): Bisulfite genomic sequencing analysis of Oct4 and Nanog promoters in HDF, ASC, iPSCs, and hESCs.

Figure 5. Differentiation of iPSCs in vitro and in vivo.

(A): EBs were generated from HDF or ASC derived iPSCs by suspension culture, and auto-differentiation appeared after withdrawing bFGF (right). Scale Bars = 100 µm. (B): Quantitative PCR for specific differentiation markers of the three germ layers. Differentiated iPSCs express Pax6 (ectodermal), GATA6 (mesodermal), and Sox17 (endodermal). (C): Immunostaining for the indicated markers of three germ layers after EB differentiation. Scale Bars = 50 µm. (D): H&E staining showed that the teratomas induced by iPSCs contained tissues of three germ layers. HDF and ASC derived iPSCs differentiation into various tissues including epithelium (endoderm), cartilage (mesoderm), muscles (mesoderm) and neural rosettes (ectoderm). Scale Bars = 50 µm.