Antioxidant supplementation reduces genomic aberrations in human induced pluripotent stem cells

Abstract

Somatic cells can be reprogrammed to induced pluripotent stem cells (iPSCs) using oncogenic transcription factors. However, this method leads to genetic aberrations in iPSCs via unknown mechanisms, which may limit their clinical use. Here, we demonstrate that the supplementation of growth media with antioxidants reduces the genome instability of cells transduced with the reprogramming factors. Antioxidant supplementation did not affect transgene expression level or silencing kinetics. Importantly, iPSCs made with antioxidants had significantly fewer de novo copy number variations, but not fewer coding point mutations, than iPSCs made without antioxidants. Our results suggest that the quality and safety of human iPSCs might be enhanced by using antioxidants in the growth media during the generation and maintenance of iPSCs.

Graphical abstract

Figure 1

Antioxidant Reduces ROS and DNA Damage to Promote Cell Survival during Reprogramming (A) Flow-cytometry analysis of ROS levels during reprogramming. Cells at day 6 of infection with 4F were analyzed and the mean ± SEM is represented (n = 3 independent experiments, Student’s t test, p = 0.01 for the difference between fibroblasts (Fib) and Fib+4F, and p = 0.035 for the difference between Fib+4F and Fib+4F+Nac). (B) Immunostaining of γH2AX during reprogramming. Cells at day 5 of infection with 4F were stained and 100 cells of each condition were counted for gH2AX-positive foci. Mean ± SEM is represented (n = 3 independent experiments, Student’s t test); scale bar: 5 μm. (C) Flow-cytometry analysis of cell death during reprogramming. Cells at day 7 of reprogramming were analyzed and the mean ± SEM is represented (n = 3 independent experiments, Student’s t test).

Figure 2

Antioxidant Does Not Change Transgene Expression and Silencing during Reprogramming (A–D) qPCR analysis of exogenous OCT4 (A), SOX2 (B), KLF4 (C), and c-MYC (D) expression at days 7, 10, and 17 of reprogramming with or without NAC treatment. Mean ± SEM is represented (n = 3 independent experiments, Student’s t test). An iPSC line generated with 4F without NAC treatment was used to assess transgene silencing after reprogramming.

Figure 3

Antioxidants Reduce Genomic Aberrations in iPSCs (A and B) Reprogramming efficiency with or without NAC. Cells were stained with alkaline phosphatase (AP) at day 27 of reprogramming with 4F. The number of AP⁺ colonies per 100,000 cells was quantified. Mean ± SEM was represented (n = 4 independent experiments, Student’s t test). (C) Characterization of iPSCs. iPSCs of different conditions were stained with NANOG, SSEA4, and TRA-160, or SOX17, A2B5, and alpha-smooth muscle actin (alpha-SMA) before or after differentiation, respectively. Scale bar: 5 μm. (D) CNV analysis of iPSCs. CNV profiling was performed using Affymetrix SNP 6.0 and copy number regions were called using Affymetrix Genotype Console (v 4.0). The numbers of parental CNVs present in control (n = 5), NAC (n = 3), or Vc (n = 4) are not significantly different from each other (Mann-Whitney U test, p > 0.05). (E) CNV analysis of iPSCs. Data were analyzed as in (D) except that the numbers of nonparental and novel (i.e., not documented in the Database of Genomic Variants) CNVs present are displayed. The numbers of nonparental CNVs are significantly greater in control (n = 5) than in NAC (n = 3) and Vc (n = 4) combined (Mann-Whitney U test, p < 0.02).