Kii I , Sumida Y , Goto T , Sonamoto R , Okuno Y , Yoshida S , Kato-Sumida T , Koike Y , Abe M , Nonaka Y , Ikura T , Ito N , Shibuya H , Hosoya T , Hagiwara M .

	Figure 1. Cell-based assay to evaluate a transitional intermediate-selective inhibitor of DYRK1A.(a) Schematic diagram of the SPHINKS assay. Doxycycline induces DYRK1A expression. Subsequently, Shield-1 stabilizes TAU fused with the destabilization domain of FKBP12 (DD-TAU). DYRK1A phosphorylates DD-TAU over the defined time period (19–21 h). Small molecules from our chemical library were added at the indicated points in Tc 1 and 2. Canonical kinase inhibitors suppress TAU phosphorylation in both Tc 1 and 2. Intermediate-selective inhibitors of DYRK1A should suppress TAU phosphorylation in Tc 1, but not in Tc 2. (b) Inducible expression of FLAG-DYRK1A and DD-TAU in HEK293 cells. Doxycycline induced FLAG-DYRK1A expression (lanes 4–6) and Shield-1 stabilized DD-TAU (lanes 2, 3, 5, 6). FLAG-DYRK1A predominantly phosphorylated Thr212 of TAU (p-TAU; lanes 5 and 6). p-TAU, TAU, FLAG and GAPDH were detected by western blot, using their corresponding antibodies. Representative data from the triplicate experiments are shown. (c) Identification of FINDY as the intermediate-selective inhibitor of DYRK1A. TAU phosphorylation was suppressed by FINDY (10 μM) in Tc 1, but not in Tc 2. In contrast, the canonical DYRK1A inhibitor INDY (10 μM) suppressed TAU phosphorylation in both Tc 1 and 2. Representative data from the triplicate experiments are shown. (d) Structure of FINDY. (e) Structure of RD0392, a canonical ATP-competitive inhibitor of DYRK1A. (f) FINDY did not inhibit the in vitro kinase activity of DYRK1A. Recombinant DYRK1A was incubated with the peptide substrate DYRKtide in the presence of FINDY, RD0392 or INDY. RD0392 and INDY inhibited the kinase activity with IC50 values of 60.2 and 139 nM, respectively. Representative dose-response curves with Hill slopes are shown. The results are presented as means±s.d. (n=4).
	Figure 2. FINDY destabilizes DYRK1A during the folding process.(a) Schematic diagram of the 2A peptide-mediated bicistronic expression of FLAG-DYRK1A and HA-EGFP. Expression is controlled by the tet operator. Effects of FINDY (b) or RD0392 (d) on destabilization of DYRK1A. FINDY (b) or RD0392 (e) was added with doxycycline to HEK293 cells harbouring the bicistronic expression vector, which was stably integrated in the genome. After incubation for 5 h, total cell lysates were subjected to SDS–PAGE followed by western blot analysis using the corresponding antibodies against FLAG, HA and GAPDH. Representative data from the pentaplicate experiments are shown. (c) The amount of endogenous DYRK1A was decreased by FINDY. Primary cortical neurons were cultured in the presence of FINDY for 3 days. DYRK1A, Tuj1 (neuron-specific class III β-tubulin) and GAPDH were detected by western blot analysis using their corresponding antibodies. Representative data from the triplicate experiments are shown. (e) FINDY did not induce the degradation of mature DYRK1A. The cells were treated with doxycycline for 16 h, then incubated with cycloheximide (CHX) with or without FINDY for the indicated time. Total cell lysates were subjected to SDS–PAGE followed by western blot analysis. Representative data from the triplicate experiments are shown. Conc., concentration.
	Figure 3. FINDY suppresses autophosphorylation of Ser97 in DYRK1A.(a) The bicistronic expression of intact DYRK1A (WT) and the substituted mutants of Lys188 to Arg (K188R), Ser97 to Ala (S97A), Tyr111 to Phe (Y111F), Tyr319 to Phe (Y319F), Tyr321 to Phe (Y321F), Tyr319 and Tyr321 to Phe (Y319F/Y321F) and Ser529 to Ala (S529A) with EGFP in HEK293 cells. The expression vectors were stably integrated in the genome of the cells. Each protein was detected with the corresponding antibody against FLAG, EGFP or GAPDH. Representative data from the quadruplicate experiments are shown. (b) The alignment of Ser97 and its surrounding amino-acid sequences from DYRK1A of the indicated species. The conserved serine residues corresponding to Ser97 of Homo sapiens DYRK1A are highlighted by the red rectangular box. Underlined portions indicate the phospho-peptide used in the immunization. (c) The polyclonal antibody against phosphorylated Ser97 (p-Ser97) specifically detected FLAG-DYRK1A (WT), but not the kinase-dead (K188R) or S97A mutants. The cells were transiently transfected with the expression vectors of the indicated constructs and the proteins were overexpressed. Total cell lysates were subjected to SDS–PAGE followed by western blot analysis. Representative data from the triplicate experiments are shown. (d) FINDY suppressed Ser97 autophosphorylation. FLAG-DYRK1A was produced in HEK293 cells incubated with epoxomicin and FINDY for 5 h, then purified with anti-FLAG antibody-conjugated beads. Each protein was detected with the corresponding antibodies against p-Ser97, phospho-Tyr (p-Tyr) or FLAG. Representative data from the triplicate experiments are shown. (e) FINDY did not suppress Ser97 autophosphorylation in Tc 2. FLAG-DYRK1A was produced in HEK293 cells for 16 h in the presence of doxycycline, after which cells were treated with 10 μM of FINDY or RD0392 for 5 h. Total cell lysates were subjected to immunoprecipitation (IP) with anti-FLAG antibody-conjugated beads. Each protein was detected with the corresponding antibodies against p-Ser97 and FLAG. Representative data from the duplicate experiments are shown. Conc., concentration.
	Figure 4. FINDY affects the folding of DYRK1A, targeting the ATP-binding pocket.(a) Schematic diagram of the biotin-transfer experiment. Expression of FLAG-DYRK1A is induced by doxycycline (Dox) with epoxomicin (Epo) and kinase inhibitors (10 μM). The resulting FLAG-DYRK1A proteins are immunopurified and reacted with the biotinylated acyl phosphate of ATP (20 μM). If the inhibitor occupies the ATP-binding pocket, then biotin transfer is prevented. (b) FINDY competes with the biotin-ATP analogue for the DYRK1A intermediate. FINDY, INDY and RD0392 were used as described in a. As a positive control for the competition experiment, the FLAG-DYRK1A protein was pre-treated with ATP (100 μM) before the reaction with the biotin-ATP analogue. The labelled proteins were blotted with streptavidin and anti-FLAG antibody. Representative data from the triplicate experiments are shown. IP, immunoprecipitation.
	Figure 5. FINDY suppresses Ser97 autophosphorylation of DYRK1A in an in vitro transcription-translation system.(a) Expression of DYRK1A in a cell-free E. coli-based coupled transcription-translation system. Representative data from the duplicate experiments are shown. (b,c) In vitro expression of DYRK1A in the presence of FINDY (b) or RD0392 (c). Representative data from the triplicate experiments are shown. (d) Relative activities of Ser97 autophosphorylation in the in vitro expression of DYRK1A in the presence of FINDY or RD0392. The band intensities in b,c were quantified. The graph shows the means±s.d. (n=3). (e) DYRK1A produced in the in vitro system was incubated with the peptide substrate DYRKtide in the presence of FINDY or RD0392 (up to 32 μM). The results are presented as means±s.d. (n=3). (f) DYRK1A produced in the in vitro system was purified with anti-FLAG antibody-conjugated beads, then reacted with lambda protein phosphatase for 2 h. After washout of phosphatase, the dephosphorylated DYRK1A was allowed to autophosphorylate in the presence of ATP and sodium orthovanadate (1 mM) for 2 h. To check the residual phosphatase activity, the non-treated FLAG-DYRK1A protein was added into the autophosphorylation reaction (lane 5). Representative data from the duplicate experiments are shown. Conc., concentration.
	Figure 6. FINDY inhibits Ser97 autophosphorylation of recombinant DYRK1A protein.(a) Recombinant GST-DYRK1A-TS protein (WT) and its pseudo-phosphorylation mutants (Y319D and Y319E), produced at 6 °C in E. coli, were reacted with ATP (100 μM), then analysed by SDS–PAGE followed by western blot analysis. Representative data from the quadruplicate experiments are shown. (b) The band intensities in a were quantified. The graph shows the means±s.d. (n=4). Statistical significance was calculated compared with the value without ATP. NS indicates not significant. *P<0.05. Statistical analysis was performed with the Mann–Whitney U-test. (c) The Y319E mutant was reacted with ATP (100 μM) in the presence of FINDY. Representative data from the triplicate experiments are shown. (d) Relative activities of Ser97 and tyrosine phosphorylation in the presence of ATP (100 μM and 1 mM) and FINDY, which were calculated from the ATP-dependent increase in the intensities of p-Ser97 and p-Tyr (see the Methods for details). The graph represents means±s.d. (n=3). Statistical significance was calculated compared with the value of the relative activity of p-Tyr at the same concentration of FINDY. *P<0.05 and **P<0.01. Data were analysed with the paired t-test. (e) RD0392 inhibited the phosphorylation of Ser97 and tyrosine residues. The Y319E mutant was reacted with ATP (100 μM) in the presence of RD0392. Representative data from the triplicate experiments are shown. (f) Relative activities of Ser97 and tyrosine phosphorylation in the presence of ATP (100 μM) and RD0392, which were calculated from the ATP-dependent increase in the intensities of p-Ser97 and p-Tyr (see the Methods for details). The graph represents means±s.d. (n=3). (g) The recombinant DYRK1A proteins were incubated with the peptide substrate DYRKtide in the presence of FINDY or RD0392 (up to 16 μM). The results are presented as means±s.d. (n=3).
	Figure 7. FINDY strengthens the interaction between DYRK1A and the co-chaperone CDC37.(a) 293T cells stably expressing CDC37-nanoKAZ were transiently transfected with expression vectors of 3xFLAG-DYRK1A, as were mutants of S97A, Y319F, Y321F, Y319F/Y321F and K188R. Total cell lysates were subjected to the CDC37-nanoKAZ interaction assay (see the Methods for details). Luminescence intensities are shown as fold-changes relative to that of DYRK1A, normalized to the amount of bound protein on a 96-well plate. The graphs represent means±s.d. (n=7). Statistical significance was calculated compared with the value of the intact DYRK1A. ***P<0.001. Statistical analysis was performed with the Mann–Whitney U-test. (b) 293T cells stably expressing CDC37-nanoKAZ were transiently transfected with an expression vector of 3xFLAG-DYRK1A. Twenty-four hours after the transfection, cells were treated with FINDY and RD0392, and incubated for an additional 24 h. Luminescence intensities are shown as fold-changes relative to that at 0 μM, normalized to the amount of 3xFLAG-DYRK1A bound on a 96-well plate. Points on the graph are means±s.d. (n=8). Statistical significance was calculated compared with the value at 0 μM. *P<0.05, **P<0.005 and ***P<0.001. Statistical analysis was performed with the Mann–Whitney U-test.
	Figure 8. FINDY selectively inhibits DYRK1A.(a,b) Recombinant DYRK1B (a) and DYRK2 (b) were incubated with the peptide substrate DYRKtide in the presence of FINDY, RD0392 or INDY. RD0392 and INDY inhibit the kinase activity with IC50 values of 53.3 and 69.2 nM, respectively, for DYRK1B, and with IC50 values of 45.6 and 27.7 nM, respectively, for DYRK2. The results are presented as means±s.d. (n=4). (c,d) Bicistronic expression of FLAG-DYRK1B (c) and FLAG-DYRK2 (d) with EGFP in HEK293 cells. FINDY administered at the indicated concentrations (Conc.) did not affect the expression of FLAG-DYRK1B and FLAG-DYRK2. Representative data from the triplicate experiments are shown. (e) Intact HEK293 cells were cultured in the presence of FINDY for 3 days. The indicated kinases and GAPDH were detected by western blot analysis using the corresponding antibodies. The categories of the indicated kinases in the kinome are shown. Representative data from the duplicate experiments are shown.
	Figure 9. FINDY selectively represses excessive DYRK1A activity in an animal model.(a) FINDY rescued the developmental malformation of Xenopus laevis embryos as induced by the overexpression of DYRK1A, but did not rescue that induced by DYRK1B. proINDY was effective in both cases. Embryos were treated with the vehicle alone (−), FINDY (2.5 μM) or proINDY (2.5 μM), then two dorsal blastomeres were injected with 750 pg of β-globin, DYRK1A or DYRK1B mRNA at the eight-cell stage. Uninjected embryos were used as controls. Development of the embryos was allowed to proceed to stage 40 in the dark. Representative images are shown. Arrows indicate small eyes. (b) Summary of the rescue rates of small eyes as induced by the overexpression of DYRK1A or DYRK1B. The results are the means±s.e.m. of triplicate experiments (n=55). (c) RT–PCR analysis of the neural marker genes in the head regions of the embryos at stage 25. Each of the marker genes indicates a specific region: NCAM for the pan-neural region, Brain Factor-1 (BF-1) for the forebrain, Pax-6 and Rx-1 for the eye, Otx-2 for the fore-midbrain, Engrailed-2 (En-2) for the mid-hindbrain and Krox-20 for the hindbrain. Ornithine decarboxylase (ODC) is an internal control.
	Figure 10. Model of the folding intermediate-selective inhibition by FINDY.In the folding process of DYRK1A, a folding intermediate autophosphorylates Tyr319/321 and subsequently Ser97 in an intramolecular manner, which prevents degradation of DYRK1A. Autophosphorylated DYRK1A takes on a mature conformation. FINDY interferes only with the folding intermediate by preventing the incorporation of ATP, leading to its degradation.

Adayev, Dual-specificity tyrosine phosphorylation-regulated kinase 1A does not require tyrosine phosphorylation for activity in vitro. 2007, Pubmed

Adayev, Dual-specificity tyrosine phosphorylation-regulated kinase 1A does not require tyrosine phosphorylation for activity in vitro. 2007, Pubmed
Altafaj, Neurodevelopmental delay, motor abnormalities and cognitive deficits in transgenic mice overexpressing Dyrk1A (minibrain), a murine model of Down's syndrome. 2001, Pubmed
Alvarez, DYRK1A autophosphorylation on serine residue 520 modulates its kinase activity via 14-3-3 binding. 2007, Pubmed
Arron, NFAT dysregulation by increased dosage of DSCR1 and DYRK1A on chromosome 21. 2006, Pubmed
Baell, New substructure filters for removal of pan assay interference compounds (PAINS) from screening libraries and for their exclusion in bioassays. 2010, Pubmed
Banaszynski, A rapid, reversible, and tunable method to regulate protein function in living cells using synthetic small molecules. 2006, Pubmed
Becker, Activation, regulation, and inhibition of DYRK1A. 2011, Pubmed
Caplan, Molecular chaperones and protein kinase quality control. 2007, Pubmed
Chico, Targeting protein kinases in central nervous system disorders. 2009, Pubmed
Crews, Mouse Erk-1 gene product is a serine/threonine protein kinase that has the potential to phosphorylate tyrosine. 1991, Pubmed
Daggett, The present view of the mechanism of protein folding. 2003, Pubmed
Dowjat, Trisomy-driven overexpression of DYRK1A kinase in the brain of subjects with Down syndrome. 2007, Pubmed
Edelman, Protein serine/threonine kinases. 1987, Pubmed
Greggio, The Parkinson disease-associated leucine-rich repeat kinase 2 (LRRK2) is a dimer that undergoes intramolecular autophosphorylation. 2008, Pubmed
Gwack, A genome-wide Drosophila RNAi screen identifies DYRK-family kinases as regulators of NFAT. 2006, Pubmed
Himpel, Identification of the autophosphorylation sites and characterization of their effects in the protein kinase DYRK1A. 2001, Pubmed
Ho, Site-directed mutagenesis by overlap extension using the polymerase chain reaction. 1989, Pubmed
Horton, Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. 1989, Pubmed
Hyodo-Miura, Involvement of NLK and Sox11 in neural induction in Xenopus development. 2002, Pubmed , Xenbase
Kamikawaji, Identification of the autophosphorylation sites of LRRK2. 2009, Pubmed
Karaman, A quantitative analysis of kinase inhibitor selectivity. 2008, Pubmed
Karnitz, Cdc37 regulation of the kinome: when to hold 'em and when to fold 'em. 2007, Pubmed
Kentrup, Dyrk, a dual specificity protein kinase with unique structural features whose activity is dependent on tyrosine residues between subdomains VII and VIII. 1996, Pubmed
Keshwani, Cotranslational cis-phosphorylation of the COOH-terminal tail is a key priming step in the maturation of cAMP-dependent protein kinase. 2012, Pubmed
Kimura, The DYRK1A gene, encoded in chromosome 21 Down syndrome critical region, bridges between beta-amyloid production and tau phosphorylation in Alzheimer disease. 2007, Pubmed
Kinstrie, Characterization of a domain that transiently converts class 2 DYRKs into intramolecular tyrosine kinases. 2010, Pubmed
Laguna, Triplication of DYRK1A causes retinal structural and functional alterations in Down syndrome. 2013, Pubmed
Liu, Overexpression of Dyrk1A contributes to neurofibrillary degeneration in Down syndrome. 2008, Pubmed
Lochhead, Protein kinase activation loop autophosphorylation in cis: overcoming a Catch-22 situation. 2009, Pubmed
Lochhead, Activation-loop autophosphorylation is mediated by a novel transitional intermediate form of DYRKs. 2005, Pubmed
Lochhead, A chaperone-dependent GSK3beta transitional intermediate mediates activation-loop autophosphorylation. 2006, Pubmed
Lochhead, dDYRK2: a novel dual-specificity tyrosine-phosphorylation-regulated kinase in Drosophila. 2003, Pubmed
Malinge, Increased dosage of the chromosome 21 ortholog Dyrk1a promotes megakaryoblastic leukemia in a murine model of Down syndrome. 2012, Pubmed
Mendgen, Privileged scaffolds or promiscuous binders: a comparative study on rhodanines and related heterocycles in medicinal chemistry. 2012, Pubmed
Muraki, Manipulation of alternative splicing by a newly developed inhibitor of Clks. 2004, Pubmed , Xenbase
Noble, CRAF autophosphorylation of serine 621 is required to prevent its proteasome-mediated degradation. 2008, Pubmed
Ogawa, Development of a novel selective inhibitor of the Down syndrome-related kinase Dyrk1A. 2010, Pubmed , Xenbase
Patricelli, Functional interrogation of the kinome using nucleotide acyl phosphates. 2007, Pubmed
Polier, ATP-competitive inhibitors block protein kinase recruitment to the Hsp90-Cdc37 system. 2013, Pubmed
Ryoo, DYRK1A-mediated hyperphosphorylation of Tau. A functional link between Down syndrome and Alzheimer disease. 2007, Pubmed
Satoh, Nemo-like kinase-myocyte enhancer factor 2A signaling regulates anterior formation in Xenopus development. 2007, Pubmed , Xenbase
Shen, Inhibition of DYRK1A and GSK3B induces human β-cell proliferation. 2015, Pubmed
Sonamoto, Identification of a DYRK1A Inhibitor that Induces Degradation of the Target Kinase using Co-chaperone CDC37 fused with Luciferase nanoKAZ. 2015, Pubmed
Soundararajan, Structures of Down syndrome kinases, DYRKs, reveal mechanisms of kinase activation and substrate recognition. 2013, Pubmed
Szymczak, Correction of multi-gene deficiency in vivo using a single 'self-cleaving' 2A peptide-based retroviral vector. 2004, Pubmed
Taipale, Chaperones as thermodynamic sensors of drug-target interactions reveal kinase inhibitor specificities in living cells. 2013, Pubmed
Taipale, Quantitative analysis of HSP90-client interactions reveals principles of substrate recognition. 2012, Pubmed
Tejedor, MNB/DYRK1A as a multiple regulator of neuronal development. 2011, Pubmed
Thomas, Full deacylation of polyethylenimine dramatically boosts its gene delivery efficiency and specificity to mouse lung. 2005, Pubmed
Walte, Mechanism of dual specificity kinase activity of DYRK1A. 2013, Pubmed
Wang, A high-throughput chemical screen reveals that harmine-mediated inhibition of DYRK1A increases human pancreatic beta cell replication. 2015, Pubmed
Wegiel, The role of DYRK1A in neurodegenerative diseases. 2011, Pubmed
Zhang, Targeting cancer with small molecule kinase inhibitors. 2009, Pubmed