Histone Deacetylase Inhibitors Promote Abnormal Binding of DNA Repair Proteins to DNA Double Strand Breaks: Consequences for DNA Repair and Genomic Instability?.

Abstract 186

Histone Deacetylase inhibitors (HDi) affect gene expression through induction of histone acetylation and lead also to the acetylation of many other proteins which could affect their cellular activity. We have previously shown that HDi trigger in hematopoietic cells not only widespread histone acetylation and DNA damage responses, but actual DNA Double Strand Breaks (DSBs) which are significantly increased and persist for long periods of time compared with normal cells _{(Gaymes TJ. and al., Mol Cancer Res. 2006)}. This raises the hypothesis that HDi regulate the capacity of leukemic cells to repair DSBs, and the explanation for the increased and persistent DNA damage in leukemic cells may be that HDi directly acetylates proteins involved in DSB repair, thus decreasing repair activity. Non Homologous End-Joining (NHEJ) is one of the main pathways for the repair of DSBs in mammalian cells. While normal cells use NHEJ that is Ku and DNA-PKcs dependent, an alternative (Alt) NHEJ pathway (DNA-PKcs and Ku independent) involving Poly-ADP-Ribose Polymerase-1 (PARP-1), Werner syndrome helicase (WRN) and DNA LigaseIIIa proteins, has been identified and is responsible for deletions and translocations in cancer. We have recently reported that myeloid leukemia cells repair DSBs using this Alt NHEJ pathway _{(Sallmyr A. and al., Blood, 2008)}. Here we show that HDi treatment by Trichostatin A (300nM) results in differential acetylation of main NHEJ protein Ku70 in acute leukemia K562 cell line. In addition, PARP-1, active in several repair pathways, including single strand break repair and Alt NHEJ is also hyperacetylated in K562 cells after 1 and 6 hours of Trichostatin A treatment compared with control treatment. To investigate whether Trichostatin A treatment alters the binding of DNA repair proteins to DSBs, we used a chromatin immunoprecipitation (CHIP) assay in K562 cell line stably transfected with the DRNeo construct that can be induced to express a single DSB. Strikingly, CHIP analysis shows that PARP-1 is increased at the DSB after 1 hour of Trichostatin A treatment, compared with controls. Preliminary CHIP analysis for the protein XRCC1, necessary for the final step of Alt NHEJ repair, shows that it is decreased at the DSB site. Importantly, AML patients treated with the HDi MS-275 in vivo show significantly increased colocalization of PARP-1 and gH2A.x, a marker for DSBs, compared with pretreatment controls, confirming our in vitro data in leukemia cell lines. Altogether, these data suggest that HDi treatment leads to an increased presence of PARP-1 at DSBs, and that this may prevent subsequent critical repair steps, providing a possible explanation for the persistence of DNA damage. Finally, to determine whether DSB repair activity is indeed decreased with HDi treatment, we used an in vivo NHEJ repair assay in K562 and HL60 acute leukemia cell lines before and after treatment with Trichostatin A for 1 hour. Both leukemia cell lines demonstrate a significant decrease in the capacity of the cells to repair DSBs following Trichostatin A treatment. These results suggest that HDi result in both a physical and functional alteration of proteins participating in DNA repair pathways, leading to a decrease in NHEJ activity. The decrease in Alt NHEJ activity may have implications for genomic instability, diminishing abnormal repair following HDi treatment.