Fig 2: RNF169 is a DYRK1A substrate. (A) FLAG-tagged RNF169 expressed in HEK-293T cells depleted of DYRK1A by siRNA transfection was purified using an anti-FLAG Ab and subsequently used in an IVK assay in the presence of radioactive labeled ATP. Proteins were analyzed by autoradiography and in WBs probed with an anti-FLAG Ab to check for equal amounts of RNF169 protein. DYRK1A depletion was assessed in WBs of total lysates from parallel samples. Quantification is shown in Fig. S5E. (B) A radioactive IVK assay was performed using bacterially produced MBP-RNF169 in the absence or presence of GST-DYRK1A. The background DYRK1A autophosphorylation was determined by incubation of the protein alone. The Coomassie blue staining demonstrates equal loading and the arrow points to MBP-RNF169. (C) MBP-fused RNF169 was used as a substrate in IVK assays with GST-DYRK1A and the phosphorylated peptides were identified by MS analysis. The position of the phosphorylated aa is in violet and the validated residues are in blue. The peptide coverage of RNF169 is also shown (identified peptides in green), rendering a coverage of 87%. See Fig. S6E,F for validation experiments. (D) Evolutionary conservation of the DYRK1A-dependent phosphosites in RNF169. Alignment of the RNF169 proteins from different the species and human RFN168: Dr, Danio rerio; Hs, Homo sapiens; Md, Monodelphis domestica; Mm, Mus musculus; Xl, Xenopus laevis. The phosphosites validated in the IVK assays are shown in blue, with residues at P - 3 and P + 1 in the same color if matching the DYRK1A consensus phosphorylation site. The putative phosphosites assayed in IVK assays but not validated are shown in violet. The arrows indicate residues important for the interaction with the ubiquitinated nucleosome41.

Fig 3: DYRK1A co-localizes with RNF169 at DSBs. (A) HeLa cells were transiently transfected with plasmids for the indicated proteins (GFP-DYRK1A and FLAG-RNF169 versions) and 36 h later, they were subjected to IR (3 Gy; +IR) or left untreated (-IR). Cells were fixed after a 1 h recovery and immunostained with the Abs indicated, counter-staining the nuclei with DAPI (blue). Scale bar, 10 µm. Some of the images include an inset to show the overlap in staining. (B) HeLa cells expressing FLAG-RNF169 or FLAG-RNF169?1 (see Fig. 4A) were transfected with a plasmid encoding GFP-DYRK1A. Following 10 Gy irradiation, the cells were pre-extracted using 0.5% Triton X-100 prior to PFA fixation. The cells were subsequently immunolabeled for RNF169 (anti-FLAG) or ?H2AX 1 h after recovery, and those cells expressing DYRK1A were detected by direct GFP fluorescence. Only irradiated cells are shown (+IR). (C) The U2-OS mCherry-Fok1 reporter cell line was used to detect recruitment of endogenous DYRK1A to DSBs. The DSB site was detected by direct mCherry fluorescence, while the DYRK1A Ab-C1 was used to detect endogenous DYRK1A. Nuclei were counterstained with DAPI. (D) DYRK1A was allowed to interact with RNF169 and DCAF7 from untreated or irradiated (3 Gy) HeLa cells and collected in RIPA-buffer 1 h after recovery. (E) HeLa cells ectopically expressing FLAG-DYRK1A (WT) or a kinase-mutant version (KR) were irradiated (3 Gy), and subjected to immunofluorescence staining 1 h later with Abs to FLAG (green) and 53BP1 (red). The nucleus of the cells was counterstained with DAPI (blue). Scale bar, 10 µm. (F,G) HeLa cells expressing the different FLAG-tagged RNF169 proteins were irradiated (3 Gy) and processed for immunofluorescence analysis with anti-FLAG and 53BP1 Abs 1 h after recovery. Cells with more than 10 foci were quantified as positive. The graph shows the percentage of cells positive for RNF169 (F) or 53BP1 (G) in control cells and after IR (mean ± SEM of two independent experiments). See Fig. S7B for the analysis of the interaction between DYRK1A and the RNF169 proteins.

Fig 4: A central region in RNF169 and a conserved region in the N-terminus of DYRK1A are necessary for the DYRK1A-RNF169 interaction. (A,B) Interaction experiments using HEK-293T cells transiently transfected with the RNF169 and DYRK1A expression constructs indicated, and immunoprecipitated with anti-HA (A) or anti-FLAG (B) Abs. Schematic illustrations of the RNF169 (A) or DYRK1A (B) primary structures and the deletion mutants used: HIS, histidine tract; KD, kinase domain; LR, leucine-arginine domain; MIU, motif interacting with ubiquitin; NLS, nuclear localization signal; PEST, PEST region; RING, Really Interesting New Gene domain; S/T, region rich in serine and threonine. (C) Soluble extracts from HEK-293T cells transiently expressing the proteins indicated were used in anti-FLAG co-immunoprecipitation experiments (see Fig. S4D for the sequence of the deletions). (D) Soluble extracts from HEK-293T cells expressing either aa 79–113 of DYRK1A fused to GST-GFP or GST-GFP alone were subjected to affinity purification using glutathione-Sepharose beads. Both the lysates and the GST-bound fractions were analyzed in WBs probed with Abs to the proteins indicated, and to GFP to detect the fusion proteins: *non-specific band. (E) WBs of anti-FLAG immunoprecipitates of HEK-293T soluble extracts expressing FLAG-RNF169. (F) Pull-down experiment with MBP-RNF169 expressed in bacteria as bait. HEK-293T extracts expressing FLAG-DYRK1A alone or together with HA-DCAF7 were used as preys, and both the lysates and the bound proteins were analyzed in WBs probed with antibodies to the tags.