Fig 2: Loss of AKAP8 protein abrogates the DNA damage effect seen in neighbouring cells.a Sanger sequencing trace of RKO cell line showing an isogenic AKAP8-/- clone with 20 nucleotide deletion (blue box) from wildtype. b Western blot validation showing loss of FBXW7 or AKAP8 protein in RKO cells. c Sanger sequencing trace of LS513 cell line showing an isogenic AKAP8-/- clone with 16 nucleotide deletion (blue box) from wildtype. d Western blot validation showing loss of FBXW7 or AKAP8 protein in LS513 cells. RKO wildtype cells were co-cultured with either single mutant FBXW7-/- cells or double mutant FBXW7-/-/AKAP8-/- cells. In RKO cells, ?H2AX (p = 0.0011; Fig. 7e, g) and 53BP1 (p < 0.0001; Fig. 7e, i) foci were decreased in double mutant cells compared with single mutant cells. Similarly, LS513 wildtype cells were co-cultured with either single mutant FBXW7-/- cells or double mutant FBXW7-/-/AKAP8-/- cells. In LS513 cells, ?H2AX (p = 0.0285; Fig. 7f, h) and 53BP1 (p = 0.0479; Fig. 7f, j) foci were decreased in double mutant cells compared with single mutant cells. Boxplots are presented as 10–90 percentiles with the median value represented by a line across the box, and an asterisk to denote the mean. Statistical significance was measured by unpaired t-test. All experiments were performed with N = 3 biological replicates.

Fig 3: AKAP8 overexpression induces DNA damage in wildtype cells.AKAP8 was overexpressed in wildtype cells and observed for DNA damage. Western blot demonstrating upregulation of AKAP8 in RKO cells (a) and LS513 cells (b). c In RKO cells, ?H2AX foci were upregulated between cells transfected with an empty vector and cells with the AKAP8 overexpression vector (p = 0.0008). 53BP1 was also upregulated in cells with the AKAP8 overexpression vector (p < 0.0001). d Similar results were observed in LS513 cells for ?H2AX (p < 0.0001) and 53BP1 (p = 0.0026). Immunofluorescent images of nuclei with ?H2AX foci in RKO cells (e) and LS513 cells (f). Immunofluorescent images of nuclei with 53BP1 foci in RKO cells (g) and LS513 cells (h). All scale bars represent 5 µm. Boxplots are presented as 10–90 percentiles with the median value represented by a line across the box, and an asterisk to denote the mean. Statistical significance was measured by unpaired t-test. All experiments were performed with N = 3 biological replicates.

Fig 4: Our FBXW7 cell line model correlates with TCGA-COAD-derived patient samples.a Analysis of Hallmark pathways using GSEA demonstrates that the oxidative phosphorylation pathway was also upregulated in TCGA-COAD-derived FBXW7 mutant cancers. The top 10 differentially expressed gene sets are represented here, with the addition of the most differentially expressed gene set which had a negative correlation in FBXW7 mutants. For Oxidative phosphorylation (highlighted with asterisk, and bar in red), the NES was 1.99, with an FDR q-value < 0.0001. b Enrichment plot depicting a significant correlation between the gene set associated with TCGA-COAD-derived FBXW7 mutant patient samples, and our own RKO cell line-derived bulk RNAseq data. c Schematic showing the interaction between FBXW7-/- and wildtype cells. FBXW7-/- cells demonstrate increased oxidative phosphorylation metabolism as well as an increased accumulation of DNA damage. FBXW7-/- cells secrete AKAP8 into its microenvironment, and this is taken up by cells which are wildtype for FBXW7. This results in increased DNA damage in wildtype cells.