Long-term culture of mesenchymal stem cells impairs ATM-dependent recognition of DNA breaks and increases genetic instability

Abstract

Background: Mesenchymal stem cells (MSCs) are attracting increasing interest for cell-based therapies, making use of both their immuno-modulating and regenerative potential. For such therapeutic applications, a massive in vitro expansion of donor cells is usually necessary to furnish sufficient material for transplantation. It is not established to what extent the long-term genomic stability and potency of MSCs can be compromised as a result of this rapid ex vivo expansion. In this study, we investigated the DNA damage response and chromosomal stability (indicated by micronuclei induction) after sub-lethal doses of gamma irradiation in murine MSCs at different stages of their in vitro expansion.

Methods: Bone-marrow-derived tri-potent MSCs were explanted from 3-month-old female FVB/N mice and expanded in vitro for up to 12 weeks. DNA damage response and repair kinetics after gamma irradiation were quantified by the induction of γH2AX/53BP1 DSB repair foci. Micronuclei were counted in post-mitotic, binucleated cells using an automated image analyzer Metafer4. Involvement of DNA damage response pathways was tested using chemical ATM and DNA-PK inhibitors.

Results: Murine bone-marrow-derived MSCs in long-term expansion culture gradually lose their ability to recognize endogenous and radiation-induced DNA double-strand breaks. This impaired DNA damage response, indicated by a decrease in the number of γH2AX/53BP1 DSB repair foci, was associated with reduced ATM dependency of foci formation, a slower DNA repair kinetics, and an increased number of residual DNA double-strand breaks 7 h post irradiation. In parallel with this impaired efficiency of DNA break recognition and repair in older MSCs, chromosomal instability after mitosis increased significantly as shown by a higher number of micronuclei, both spontaneously and induced by γ-irradiation. Multifactorial regression analysis demonstrates that in vitro aging reduced DNA damage recognition in MSCs after irradiation by a multiplicative interaction with dose (p < 0.0001), whereas the increased frequency of micronuclei was caused by an additive interaction between in vitro aging and radiation dose.

Conclusion: The detrimental impact of long-term in vitro expansion on DNA damage response of MSCs warrants a regular monitoring of this process during the ex vivo growth of these cells to improve therapeutic safety and efficiency.

Keywords: Adult stem cells; DNA repair; Genetic instability; In vitro aging; Ionizing radiation; Mesenchymal stem cells; Micronuclei.

Fig. 1

Induction of γH2AX- and 53BP1-foci formation after exposure to irradiation in aging MSCs in vitro. a MSCs were cultured for 1, 4, and 8 weeks in vitro. Repair foci formation is shown in MSCs 90 min after sham-irradiation (0 Gy) or after 2 Gy of γ-irradiation by immunofluorescence staining for γH2AX (red) and 53BP1 (green). Nuclear counterstaining was done using DAPI. b Quantification γH2AX- and 53BP1-foci formation in MSCs was done 90 min after sham-irradiation (0 Gy) or after γ-irradiation with 0.05, 0.5, 2, and 6 Gy using the Keyence BZ-II Analyzer software (see Additional file 6: Figure S1). Mean number of foci ± SEM (n = 3) are calculated from at least 50 analyzed nuclei

Fig. 2

Effect of inhibition of ATM and DNA-PK on γH2AX- and 53BP1-foci formation in aging MSCs after exposure to irradiation. a One-week and b 8-week cultured MSCs were pre-incubated with DMSO and a chemical ATM (ATMi) or DNA-PK (DNA-PKi) inhibitor and irradiated with 6 Gy before γH2AX- and 53BP1-foci formation was detected 90 min post irradiation. Foci formation in at least 50 nuclei was automatically quantified using Keyence BZ-II Analyzer software, and the relative amount of repair foci was normalized to the 90 min value (mean values ± SEM, n = 3,)

Fig. 3

Kinetics of γH2AX- and 53BP1-foci formation in aging MSCs after exposure to 2 Gy. Aging MSCs were irradiated with 2 Gy before γH2AX- and 53BP1-foci formation was detected 90, 180, 300, and 420 min post irradiation. Foci formation in at least 50 nuclei was automatically quantified using Keyence BZ-II Analyzer software and the relative amount of repair foci was normalized to the 90 min value. Data plotted on a linear scale can be found in the Additional file 7: Figure S5. Mean repair half time in phase 1 and phase 2 and fraction of unrepaired DNA breaks is given in the Additional file 4: Table S1

Fig. 4

Frequency of micronuclei after exposure to irradiation in aging MSCs in vitro. a MSCs were cultured for 1, 4, 8, and 12 weeks under hypoxia in vitro. Mean number of micronuclei per 100 binuclear MSCs was detected after sham-irradiation (0 Gy) and 0.05, 0.5, 2, and 6 Gy irradiation and incubation for 5 days to allow repair of DNA damage in the presence of cytochalasin B. b Representative images of MSCs 5 days after 6 Gy γ-irradiation and DAPI staining showing cells with 1, 2, and 4 micronuclei. c Mean percentage of irradiated binucleated MSCs with two or more micronuclei (mean values ± SEM, n = 3)