Limiting replication stress during somatic cell reprogramming reduces genomic instability in induced pluripotent stem cells

Abstract

The generation of induced pluripotent stem cells (iPSC) from adult somatic cells is one of the most remarkable discoveries in recent decades. However, several works have reported evidence of genomic instability in iPSC, raising concerns on their biomedical use. The reasons behind the genomic instability observed in iPSC remain mostly unknown. Here we show that, similar to the phenomenon of oncogene-induced replication stress, the expression of reprogramming factors induces replication stress. Increasing the levels of the checkpoint kinase 1 (CHK1) reduces reprogramming-induced replication stress and increases the efficiency of iPSC generation. Similarly, nucleoside supplementation during reprogramming reduces the load of DNA damage and genomic rearrangements on iPSC. Our data reveal that lowering replication stress during reprogramming, genetically or chemically, provides a simple strategy to reduce genomic instability on mouse and human iPSC.

Figure 1. Expression of reprogramming factors generates replication stress.

(a,c) HTM-mediated quantification of γH2AX intensity levels in MEFs (a) or BJ human fibroblasts (c) 4 days after infection with retroviruses encoding GFP, OSK or OSKM. Uninfected cells (struck through circle) were also included as negative control. Centre lines indicate mean values, whereas squared boxes show the percentage of outliers. Data are representative of three independent experiments performed in triplicate ***P<0.001. (b,d) Western blot analyses of cell extracts obtained from MEFs (b) or BJ human fibroblasts (d) 4 days after infection with retroviruses encoding GFP, OSK or OSKM. Uninfected cells (struck through circle), human (BJ or PGP2-derived iPSC lines) or mouse iPSCs were also included as controls. See Supplementary Fig. 14 for full western blot images.

Figure 2. Increased CHK1 levels facilitate reprogramming and reduce RS on iPSC.

(a) HTM-mediated quantification of γH2AX intensity levels in wild-type or Chk1^TG/TG MEFs 4 days after infection with retroviruses encoding GFP or OSK. Centre lines indicate mean values, whereas squared boxes show the percentage of outliers. Data are representative of two independent experiments performed in triplicate *P<0.05. (b) Wt, Chk1^TG/+ or Chk1^TG/TG MEF were infected with retroviruses encoding OSK and relative reprogramming efficiencies were evaluated by counting alkaline-phosphatase-positive colonies. Data are representative of three independent experiments performed in triplicate in MEF derived from four different embryos. ***P<0.001. (c) Representative images of the analysis described in b.

Figure 3. Nucleoside supplementation decreases RS in cells expressing reprogramming factors.

(a) HTM-mediated quantification of γH2AX intensity levels in PGP2 human fibroblasts 4 days after infection with OSK or OSKM with or without daily addition of nucleosides. Centre lines indicate mean values, whereas squared boxes show the percentage of outliers. Uninfected cells () were also included as negative control. Data are representative of three independent experiments performed in triplicate ***P<0.001. (b) Graphical representation of the fork elongation rate (kb min⁻¹), measured by DNA combing, in BJ fibroblasts 4 days after infection with retroviruses encoding either GFP or OSKM with or without the addition of nucleosides. A representative image of a stained DNA fibre is shown in the lower part. Iododeoxyuridine (IdU, red) and Chlorodeoxyuridine (CIdU, green) signals are shown. (c,d) Relative reprogramming efficiency determined in RFP-infected (c) or RFP plus cMyc-infected (d) dFib-ind^OSK cells treated with doxycycline at a concentration of 100 ng ml⁻¹ with or without daily addition of nucleosides. Reprogramming efficiency was evaluated by scoring the percentage of GFP-positive colonies (resulting from the reactivation of the endogenous OCT4 promoter) and RFP-negative colonies (resulting from the silencing of exogenous transgenes). Data are representative of two independent experiments performed in duplicate.

Figure 4. Lowering reprogramming-induced RS decreases genomic instability on iPSC.

(a) HTM-mediated quantification of nuclear γH2AX intensity levels in four independent iPSC lines derived from wild-type or Chk1^TG/TG MEF. Centre lines indicate mean values, whereas squared boxes show the percentage of outliers. ***P<0.001. (b) Average number of multi-telomeric signal (MTS) per metaphase evaluated in iPSC lines derived from wild-type and Chk1^TG/TG MEF. Five different iPSC lines were used for quantification and at least 50 metaphases were evaluated per line. In the right, a representative image from the MTS analyses is shown. The white arrow indicates a chromosome with a fragmented multi-telomeric signal (MTS). (c) HTM-mediated quantification of γH2AX intensity levels in BJ-derived human iPSC lines generated with or without daily addition of different amounts of nucleosides. Centre lines indicate mean values, whereas squared boxes show the percentage of outliers. Data are representative of two independent experiments performed in triplicate ***P<0.001. (d) Average number of MTS per metaphase evaluated in iPSC lines derived from human dFib-ind^OSK cells untreated or treated with nucleosides during the whole reprogramming process. Three different iPSC lines per condition were used for quantification and at least 50 metaphases were evaluated per line. (e) SurePrint G3 Human High Resolution Microarrays were used to analyse the number of CNV in hiPSC clones untreated or treated with nucleosides during the whole reprogramming process. A one-tailed unpaired t-test was used to compare CNV data sets. *P<0.05.