Fig. 4
From: Defective DNA damage repair leads to frequent catastrophic genomic events in murine and human tumors
DNA repair pathways in brain tumors with catastrophic events. a, b Normalized expression levels for repair factors involved in cNHEJ, HRR and Alt-EJ. a Red, expression levels in medulloblastoma cells of XRCC4/p53 mice (n = 8), yellow, expression levels in the healthy cerebellum and granule neural progenitors (n = 4) for age and genotype-matched animals (expression levels for the non-neoplastic cerebellum and for granule neural progenitors were not significantly different and therefore shown as one box). The horizontal line depicts the threshold for which genes are considered to be expressed. b Normalized expression levels for repair factors in gliomas for BRCA2/p53 (blue, n = 6) or Lig4/p53 (green, n = 6) deficient animals. Unpaired t-tests were used to test for significance in a and b. c Analysis of the microhomology length at the breakpoints due to catastrophic events in tumors developing in XRCC4/p53, Lig4/p53 and BRCA2/p53 animals. The analysis focuses on chromosome regions affected by the catastrophic events. Beta-regression analysis was used to test for significance. d Mutational processes are dominated by signature 3 (associated with HR deficiency). The main contributing mutational signatures are very similar in tumors based on inactivation of factors essential for HR or for cNHEJ. A human MB from a BRCA2 compound heterozygous patient is shown for comparison. e Synthetic lethality approaches targeting medulloblastoma cells from XRCC4/p53-deficient mice. Medulloblastoma cells (MB 594, MB 794, MB 1197) from Nestin-Cre p53−/− Xrcc4c/− and Nestin-Cre p53−/− Xrcc4c/c mice as well as control neural progenitor cells (NSPC1, NSPC2) from Nestin-Cre p53−/− Xrcc4+/− mice were treated with RAD51 inhibitors (B02 or Ri1, 10 µM), PARP inhibitor (Olaparib, 5 µM), topotecan (100 nM) or with combinations. f Combination of PARP inhibitor with topoisomerase inhibitors reduces cell viability in human medulloblastoma (DAOY) and high-grade glioma cells (A172) with complex genome rearrangements. Cell viability is shown as a percentage of the viability for vehicle control cells in e and f (mean values and standard deviations for at least three independent experiments). ANOVA was used to test for significance in e and f. The medians for boxplots are indicated by center lines, rectangles span the interquartile ranges (whiskers show maximum and minimum)
