The role of chromatin structure and DNA end resection in the growth state dependence of alternative end joining (alt-EJ)
The main purpose of this study was to investigate the mechanistic underpinnings of the growth state dependence of alt-EJ. While HRR is strictly cell cycle dependent due to the requirement of a sister chromatid as template, c-NHEJ operates independently of the cell cycle and growth state. C-NHEJ does not use a template and therefore, it cannot restore the sequence around the DSBs. Moreover it can also lead to translocations. Alt-EJ operates on similar basic principles as c-NHEJ but it is even more prone to errors and chromosome translocations. However, unlike c-NHEJ, its efficiency fluctuates throughout the cell cycle and it is completely abrogated when cells are in G0. In order to explore the mechanisms behind the growth state dependence of alt-EJ, we first inquired whether the chromatin structure is a determinant, as it was previously reported that G0 cells have more compact and transcriptionally inactive chromatin. Our approach was to artificially modify chromatin structure and to measure the kinetics of alt-EJ, particularly following chromatin relaxation. However, under the experimental conditions applied in this study, we were not able to show that chromatin relaxation promoted abrogated alt-EJ in G0 cells. Nevertheless, one intriguing finding was that DNA-PKcs deficient cells exhibited similar chromatin structure related signature of H3K9me2, H3K9me3, H3K27Ac, H3K27me3, H4K20me2/3, HP1 and KAP1 in both cycling and G0 phase. As DNA-PKcs deficient cells do not show typical growth state dependent fluctuations of alt-EJ, this finding suggests that chromatin structure may be the determinant in the growth state dependence of alt-EJ. However, it could be that treatments with limited effects on chromatin structure fail to revert this phenotype of alt-EJ. It is well established that chromatin structure is highly dynamic and plays a central role in DSB repair. Therefore, further investigations in this context are required. Another prime candidate to explain the growth state dependence of alt-EJ was DNA end resection, as unpublished data from our laboratory has shown that DNA-PKcs deficient human glioblastoma cells, M059J, retain DNA end resection activity in G0. While DNA end resection is typically downregulated in G0/G1 phases of the cell cycle, persistent DNA end resection activity in G0 could explain why DNA-PKcs deficient cells efficiently utilize alt-EJ in G0, in contrast to other c-NHEJ mutants. Our investigations could confirm that, in DNA-PKcs proficient status, DNA end resection is only detectable in exponentially growing cells with an enhanced activity in G2 and it is downregulated in G0 phase. Indeed, cell cycle and growth state dependent fluctuations of DNA end resection parallel the kinetics of alt-EJ and suggest that abrogation of alt-EJ in G0 may be due to lack of DNA end resection. However, unlike DNA-PKcs deficient human glioblastoma cells, we were not able to detect DNA end resection activity in DNA-PKcs deficient MEF and CHO cells in G0. Moreover, we observed that alt-EJ was abrogated in Ku80 deficient MEFs despite detectable DNA end resection activity, which is possibly a consequence of inefficient growth arrest in G0. Yet the response of the latter cell line suggests that growth factors may also be regulating alt-EJ. Here we conclude that, DNA end resection alone may not be responsible for the growth state dependence of alt-EJ. Future studies should be designed with the purpose of analyzing the combined contributions of DNA end resection and other parameters such as chromatin structure, growth factors and transcriptional regulation in the growth state dependence of alt-EJ.