Glenewinkel F , Cohen MJ , King CR , Kaspar S , Bamberg-Lemper S , Mymryk JS , Becker W .

MOP-111173 CIHR

	Figure 1. DCAF7 interacts with class 1 DYRKs and HIPK2.(a) Co-IP of human DCAF7 with rat DYRK1A, human DYRK1B and human HIPK2. - HeLa cells were transfected to co-express FLAG-hDCAF7 and the indicated GFP-fused protein kinases or GFP as a negative control. Total cell lysates were subjected to IP with GFP-trap beads and bound proteins were detected by immunoblotting with the antibodies directed against DCAF7 and GFP (αDCAF7 and αGFP). Aliquots of the whole cell lysates are shown as input. (b) HIPK1 does not bind DCAF7. – GFP, GFP-HIPK1 and GFP-HIPK2 were immunoprecipitated from transiently tranfected HeLa cells and analysed by Western blotting. (c) Co-IP of endogenous DYRK1A and DCAF7. – The lysate of untransfected HeLa cells was split before overnight IP with either goat anti DYRK1A (αD1A) or an unrelated goat antibody (αGFP). HeLa cells expressing FLAG-DCAF7 served as a positive control. (d) Co-IP of human DCAF7 with Xenopus laevis DYRK1B. – FLAG-hDCAF7 was co-expressed in HeLa cells with untagged Xenopus DYRK1B (xDYRK1B), human DYRK1B (hDYRK1B) as a positive control or empty vector as a negative control (Co). Binding of DCAF7 to immunoprecipitated hDYRK1B and xDYRK1B was detected by Western blot analysis. Note that the amounts of human and Xenopus DYRK1B cannot be directly compared because the epitope is only partially conserved in the Xenopus ortholog. (e) Co-IP of rat DYRK1A, human DYRK1B and Xenopus DYRK1B with zebrafish DCAF7. - GFP-tagged zebrafish DCAF7 was co-expressed in HeLa cells with murine FLAG-DYRK1A, hDYRK1B or xDYRK1B. GFP-trap was used to precipitate zDCAF7 and binding partners. The vertical lines indicate where irrelevant lanes were deleted from the final images.
	Figure 2. Mapping of the DCAF7-interacting sequence in DYRK1A.(a) Schematic representation of DYRK1A domain structure and deletion clones. - Deletion constructs tested for DCAF7 binding are illustrated in a close-up of the N-terminal domain. ALS, alternatively spliced segment of 9 amino acids that is absent in the short splicing variant; NLS, nuclear localization sequence; DH, DYRK homology box; cat, catalytic domain; PEST, proline-, glutamic acid-, serine-, threonine-rich region; polyHis, poly-histidine tract. (b) Co-IP of FLAG-DCAF7 with GFP-DYRK1A deletion constructs. - HeLa cells co-expressing FLAG-DCAF7 and the indicated GFP-DYRK1A deletion constructs or unfused GFP were used for GFP-IP. Binding of DCAF7 was analysed by immunoblotting. (c) Alignment of the minimal DCAF7 binding region in vertebrate class 1 DYRKs and the Drosophila ortholog MNB. - A deletion was introduced in the full length kinase to narrow down the binding region (Δ93–104). (d) FLAG-DCAF7 does not bind to GFP-DYRK1A-Δ93–104. – Binding of FLAG-DCAF7 to GFP-DYRK1A constructs was assayed by co-IP as in panel (b). GFP-DYRK1A-ΔN lacks the N-terminal domain (Δ1–134) and was used as a negative control.
	Figure 3. Mapping of DCAF7-binding sequences in Dictyostelium DYRK1 and human HIPK2.(a) Structure of the DYRK1 ortholog in Dictyostelium discoideum (DdDYRK1A, Uniprot Q76NV1). - The sequence encoded by the first exon shows similarity with the DCAF7 binding region of human DYRK1A (boxed). (b) Co-IP of human DCAF7 with DdDYRK1A. - HeLa cells expressing a GFP fusion protein of the N-terminal region DdDYRK1A (amino acids 1–40) and human FLAG-DCAF7 were used for anti GFP IP. DYRK1A deletion constructs served as negative and positive controls. (c) Structure of human HIPK2 (Uniprot Q9H2X6). - The N-terminal domain of HIPK2 harbours a motif weakly similar to the DCAF7 binding region of DYRK1A. The corresponding sequence in HIPK1 is shown for comparison. The arrowhead points to a proline in the HIPK1 sequence that is not found in HIPK2. (d) Co-IP of DCAF7 with GFP-HIPK2 deletion constructs and point mutants. - GFP, GFP-HIPK1 and GFP-HIPK2 were immunoprecipitated from transiently tranfected HeLa cells and analysed by Western blotting for the binding of endogenous DCAF7. GFP and GFP-DYRK1A served as negative and positive controls. HIPK2-D324N is a catalytically inactive point mutant of HIPK2. DH, DYRK homology box; cat, catalytic domain; poly Q/N, polyglutamine/polyasparagine tracts; PEST, region rich in proline, glutamic acid, serine and threonine residues; YH, tyrosine/histidine-rich region. Putative nuclear localization signals are highlighted in bold print.
	Figure 4. DCAF7 mediates binding of E1A to DYRK1A, DYRK1B and HIPK2.(a) Co-IP of myc-E1A with DYRK1A, DYRK1B, HIPK2 and DCAF7. - HeLa cells were transfected to co-express myc-E1A (289 amino acid form) GFP-DYRK1A, DYRK1B or HIPK2 and either FLAG-DCAF7 or a control vector. The vertical line indicates where irrelevant lanes were deleted from the final image. Note that GFP-HIPK2 is difficult to reveal on the blots due to its large size and could not be detected in the cell lysates (input). (b) Co-IP of endogenous DYRK1A/DCAF7 with E1A. - The lysate of untransfected HEK293 cells or HeLa cells was subjected to IP with either goat anti DYRK1A (αD1A) or mouse anti-E1A (αE1A). An unrelated goat antibody (Ctrl) was used as a negative control for the IP with αD1A. HeLa cells (which lack E1A) were used as background control for the αE1A IP. The heavy chain of the immunoprecipitating E1A antibody is marked by an asterisk (IgG). (c) DYRK1A, DCAF7 and E1A are components of a common complex. - HEK293-(GFP-DYRK1A-tetOn) cells were transfected with a FLAG-DCAF7 expression vector or empty control plasmid and induced with doxycyclin to express GFP-DYRK1A. Lysates were subjected to sequential IP with anti FLAG and anti GFP. The two E1A bands may correspond to the major protein forms of 289 and 243 amino acids that are expressed in HEK293 cells. (d) GST pulldown assay. - HeLa cells were transfected to express GFP or GFP-DYRK1A either with FLAG-DCAF7 or alone. Aliquots of cell lysates were used for pulldown assays with agarose-bound GST or a GST-tagged construct of the exon2-encoded portion of E1A (E1A-X2) as bait. A deletion mutant (X2Δ, deletion of amino acids 255–270) that lacks the DYRK1A/DCAF7 binding region of E1A served as negative control. (e) Outline of the pulldown assay (f,g) In vitro interaction assays. – GFP-DCAF7 was in vitro-translated in rabbit reticulocyte lysate (RRL) and used as a prey for GST-pulldown assays with immobilized GST fusion proteins as indicated below the bottom panel. In parallel control samples (Co), in vitro transcription was driven by the empty vector.
	Figure 5. DCAF7 is an essential adaptor protein for the association of E1A with DYRK1A in vivo.Human HT1080 cells were treated with control siRNA, siRNA specific to DYRK1A (a) or siRNA specific to DCAF7 (b). Cells were subsequently co-transfected with the blank vector, GFP-E1A (WT, 289 amino acid form) or the E1A point mutant R262/263E (Mu) and vectors expressing FLAG-DCAF7 (a) or HA-DYRK1A (b). The E1A-R262/263E mutant does not interact with DYRK1A or DCAF726 and served as a specificity control. Lysates were immunoprecipitated using anti-FLAG antibodies (a) or anti-HA antibodies (b) and immunoblotted using anti-GFP antibodies to detect the presence of co-precipitating E1A.
	Figure 6. Structural basis for E1A binding to DYRK1A and HIPK2.(a,b) Co-IP of myc-E1A wild type DYRK1A, deletion mutants of DYRK1A and a kinase-negative point mutant of DYRK1A (K188R). - Lysates of HeLa cells expressing myc-E1A, FLAG-DCAF7 and the indicated GFP-DYRK1A constructs were subjected to anti GFP IP. The dashed lines indicate where irrelevant regions of the blots were deleted from the final image. (c,d) GST pulldown assays. - HeLa cells were transfected to co-express FLAG-DCAF7 with GFP-DYRK1A constructs or GFP-HIPK2 constructs as indicated. Cell lysates were subjected to GST-pulldown assay with immobilized GST or GST-E1A-X2 and bound proteins were analysed by immunoblotting. GST-E1A-X2Δ lacks the DYRK1A/DCAF7 binding region of E1A and served as negative control. To reveal the GFP-DYRK1A1–176 construct, a polyclonal goat antibody directed against an N-terminal epitope was used in panel (c). This antibody crossreacts with an unidentified band (marked by an asterisk) that is not detected by the monoclonal DYRK1A antibody used in Fig. 4d. Endogenous DYRK1A is detected as a double band at ~90 kDa.
	Figure 7. Phosphorylation of E1A by DYRK1A and HIPK2.(a) In vitro kinase assay. - Bacterially expressed GST-E1A-X2 was incubated with recombinant DYRK1A at 30 °C in the presence of 1 mM ATP. Aliquot of the reaction were taken at variable times and phosphorylation of E1A was detected by Western blot analysis with a phosphospecific antibody directed against pSer219. (b) Phosphorylation of E1A in HEK293 cells. - HEK293-(GFP-DYRK1A-tetOn) cells were transfected with a FLAG-DCAF7 expression vector or empty control plasmid and either induced with doxycyclin (dox) to express GFP-DYRK1A or not induced. Two days after transfection, total cellular lysates were analyzed for Ser219 phosphorylation. Detection of E1A by the monoclonal antibody M58 is independent of the phosphorylation state. The asterisks mark an upshifted band in DYRK1A overexpressing samples. (c–e) Phosphorylation of E1A exon2 by DYRK1A and HIPK2. - HeLa cells were co-transfected with expression plasmids for myc-E1A-X2, FLAG-DCAF7 and DYRK1A, HIPK1, HIPK2 or mutant kinase constructs as indicated. In c, myc-E1A-X2 was immunoprecipitated and either dephosphorylated by calf intestinal phosphatase (CIP) or not treated before SDS-PAGE. The asterisks mark the light chain bands of the immunoprecipitating antibody. In (d,e), total cellular lysates were analysed for phosphorylation of E1A-X2 2 days after transfection.
	Figure 8. Subcellular localization of the E1A/DCAF7/DYRK1A complexes.HeLa cells transiently transfected to co-express either wild type GFP-DYRK1A (WT) or GFP-DYRK1A-Δ93–104 together with E1A (243 amino acid form) and/or FLAG-DCAF7. Proteins were detected by autofluorescence (GFP) or immunofluorescence (E1A, FLAG-DCAF7). Nuclei were stained with DAPI. Scale bar, 100 μm.
	Figure 9. Evolutionary conservation of the DYRK/DCAF7 interaction.(a) Consensus sequence of the DCAF7 binding motif in class 1 DYRKs. - The sequence logo was created from an alignment9 of 19 representative sequences of class 1 DYRKs in the animal kingdom, including sponges, jellyfish, sea urchin, insects and different worms using the WebLogo application52. (b) Phylogenetic relationship of the DYRKs and HIPKs that interact with DCAF7. – Conservation of the DCAF7 binding sequence is illustrated for kinases that are known to bind DCAF7 or orthologous proteins. Mammalian DYRK2-4 and HIPK1 do not bind DCAF7. It is unknown whether Drosophila HIPK binds to the wap protein, but the DCAF7 binding site from HIPK2 is not conserved in invertebrates. Kinase branches from top to bottom: class 1 DYRKs, class 2 DYRKs, YAK branch of DYRKs, HIPKs. Non-mammalian kinases are from Danio rerio (D.r.), Drosophila melanogaster (D.m.), Dictyostelium discoideum (D.d.), Saccharomyces cerevisiae (S.c.), and Arabidopsis thaliana (A.t.).
	Figure 10. Proposed model of the DYRK1A-DCAF7-E1A complex.Multiple binding sites allow for the simultaneous interaction of DCAF7 with E1A and DYRK1. E1A is an intrinsically disordered protein that is known to interact with many cellular proteins including the pocket proteins (RB1, RBL1). NLS, nuclear localization signal.

Adams, RACK1, A multifaceted scaffolding protein: Structure and function. 2011, Pubmed

Adams, RACK1, A multifaceted scaffolding protein: Structure and function. 2011, Pubmed
Aranda, DYRK family of protein kinases: evolutionary relationships, biochemical properties, and functional roles. 2011, Pubmed
Becker, Activation, regulation, and inhibition of DYRK1A. 2011, Pubmed
Becker, Emerging role of DYRK family protein kinases as regulators of protein stability in cell cycle control. 2012, Pubmed
Breitkreutz, A global protein kinase and phosphatase interaction network in yeast. 2010, Pubmed
Chen, Dosage of Dyrk1a shifts cells within a p21-cyclin D1 signaling map to control the decision to enter the cell cycle. 2013, Pubmed
Cohen, Functional analysis of the C-terminal region of human adenovirus E1A reveals a misidentified nuclear localization signal. 2014, Pubmed
Cohen, Dissection of the C-terminal region of E1A redefines the roles of CtBP and other cellular targets in oncogenic transformation. 2013, Pubmed
Crooks, WebLogo: a sequence logo generator. 2004, Pubmed
Degoutin, Riquiqui and minibrain are regulators of the hippo pathway downstream of Dachsous. 2013, Pubmed
Durocher, The FHA domain is a modular phosphopeptide recognition motif. 1999, Pubmed
Fonseca, Adenovirus evasion of interferon-mediated innate immunity by direct antagonism of a cellular histone posttranslational modification. 2012, Pubmed
Frisch, Adenovirus-5 E1A: paradox and paradigm. 2002, Pubmed
Guedj, DYRK1A: a master regulatory protein controlling brain growth. 2012, Pubmed
Harlow, Monoclonal antibodies specific for adenovirus early region 1A proteins: extensive heterogeneity in early region 1A products. 1985, Pubmed
Himpel, Identification of the autophosphorylation sites and characterization of their effects in the protein kinase DYRK1A. 2001, Pubmed
Ho, Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. 2002, Pubmed
Jin, A family of diverse Cul4-Ddb1-interacting proteins includes Cdt2, which is required for S phase destruction of the replication factor Cdt1. 2006, Pubmed , Xenbase
Kii, Selective inhibition of the kinase DYRK1A by targeting its folding process. 2016, Pubmed , Xenbase
Komorek, Adenovirus type 5 E1A and E6 proteins of low-risk cutaneous beta-human papillomaviruses suppress cell transformation through interaction with FOXK1/K2 transcription factors. 2010, Pubmed
Leder, Alternative splicing variants of dual specificity tyrosine phosphorylated and regulated kinase 1B exhibit distinct patterns of expression and functional properties. 2003, Pubmed
Litovchick, DYRK1A protein kinase promotes quiescence and senescence through DREAM complex assembly. 2011, Pubmed
Lochhead, Activation-loop autophosphorylation is mediated by a novel transitional intermediate form of DYRKs. 2005, Pubmed
Mazmanian, The zebrafish dyrk1b gene is important for endoderm formation. 2010, Pubmed
Miyata, DYRK1A binds to an evolutionarily conserved WD40-repeat protein WDR68 and induces its nuclear translocation. 2011, Pubmed
Miyata, The molecular chaperone TRiC/CCT binds to the Trp-Asp 40 (WD40) repeat protein WDR68 and promotes its folding, protein kinase DYRK1A binding, and nuclear accumulation. 2014, Pubmed
Morita, HAN11 binds mDia1 and controls GLI1 transcriptional activity. 2006, Pubmed
Morriss, The Drosophila wings apart gene anchors a novel, evolutionarily conserved pathway of neuromuscular development. 2013, Pubmed
Nissen, A zebrafish screen for craniofacial mutants identifies wdr68 as a highly conserved gene required for endothelin-1 expression. 2006, Pubmed
Odho, Characterization of a novel WDR5-binding site that recruits RbBP5 through a conserved motif to enhance methylation of histone H3 lysine 4 by mixed lineage leukemia protein-1. 2010, Pubmed
Papadopoulos, Splice variants of the dual specificity tyrosine phosphorylation-regulated kinase 4 (DYRK4) differ in their subcellular localization and catalytic activity. 2011, Pubmed
Pelka, Intrinsic structural disorder in adenovirus E1A: a viral molecular hub linking multiple diverse processes. 2008, Pubmed
Piccoli, Leucine-rich repeat kinase 2 binds to neuronal vesicles through protein interactions mediated by its C-terminal WD40 domain. 2014, Pubmed
Ritterhoff, The WD40-repeat protein Han11 functions as a scaffold protein to control HIPK2 and MEKK1 kinase functions. 2010, Pubmed
Schmitz, Integration of stress signals by homeodomain interacting protein kinases. 2014, Pubmed
Sitz, The Down syndrome candidate dual-specificity tyrosine phosphorylation-regulated kinase 1A phosphorylates the neurodegeneration-related septin 4. 2008, Pubmed
Skurat, Phosphorylation of Ser640 in muscle glycogen synthase by DYRK family protein kinases. 2004, Pubmed
Soppa, The Down syndrome-related protein kinase DYRK1A phosphorylates p27(Kip1) and Cyclin D1 and induces cell cycle exit and neuronal differentiation. 2014, Pubmed
Stirnimann, WD40 proteins propel cellular networks. 2010, Pubmed
Ubersax, Mechanisms of specificity in protein phosphorylation. 2007, Pubmed
Varjosalo, The protein interaction landscape of the human CMGC kinase group. 2013, Pubmed
Walte, Mechanism of dual specificity kinase activity of DYRK1A. 2013, Pubmed
Wang, Wdr68 requires nuclear access for craniofacial development. 2013, Pubmed
Wong, AKAP signalling complexes: focal points in space and time. 2004, Pubmed
Woods, Charting the landscape of tandem BRCT domain-mediated protein interactions. 2012, Pubmed
Wu, Identifying the hotspots on the top faces of WD40-repeat proteins from their primary sequences by β-bulges and DHSW tetrads. 2012, Pubmed
Zhang, Interaction of the E1A oncoprotein with Yak1p, a novel regulator of yeast pseudohyphal differentiation, and related mammalian kinases. 2001, Pubmed
Zhang, Homeodomain interacting protein kinase 2 promotes apoptosis by downregulating the transcriptional corepressor CtBP. 2003, Pubmed
de Vetten, The an11 locus controlling flower pigmentation in petunia encodes a novel WD-repeat protein conserved in yeast, plants, and animals. 1997, Pubmed
van Bon, Disruptive de novo mutations of DYRK1A lead to a syndromic form of autism and ID. 2016, Pubmed
van der Laden, Effect of tyrosine autophosphorylation on catalytic activity and subcellular localisation of homeodomain-interacting protein kinases (HIPK). 2015, Pubmed