QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online 21 April 2009
doi: 10.1242/jcs.044354

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Fernandez-Martinez, J. Articles by Galceran, J. Search for Related Content PubMed PubMed Citation Articles by Fernandez-Martinez, J. Articles by Galceran, J. Social Bookmarking                         What's this?

Attenuation of Notch signalling by the Down-syndrome-associated kinase DYRK1A

Javier Fernandez-Martinez^*, Eva M. Vela, Mireille Tora-Ponsioen, Oscar H. Ocaña, M. Angela Nieto and Juan Galceran

Instituto de Neurociencias, CSIC-UMH, Sant Joan d'Alacant, Alicante, Spain

View larger version (108K): [in this window] [in a new window]   Fig. 1. Dyrk1a is coexpressed with components of the Notch signalling pathway in the ventricular zone of the developing cortex. In situ hybridization for elements of the Notch signalling pathway and Dyrk1a in the developing neocortex of wild-type E14 mouse embryos. The specific RNA probes used are indicated on the left, and the ventricular and subventricular zone (VZ) of the cortex, the intermediate zone (IZ) and the cortical plate (CP) are defined by the dotted lines in A. (A,B) Notch1 is expressed in the ventricular and subventricular zone (VZ) of the cortex whereas its expression in the intermediate zone (IZ) and cortical plate (CP) is negligible. (C,D) Notch ligand Dll1 is mainly restricted to the most ventricular region with some scattered positive cells in the subventricular zone. (E,F) Hes1, a target of Notch signalling is expressed in the ventricular and subventricular region. (G,H) Dyrk1a is expressed in both the cortical plate and the ventricular and subventricular zones. Expression in the ventricular zone shows also a `speckled' pattern and extends over the whole area. Right panels (B,D,F,H) are higher magnifications of the left panels (A,C,E,G). Scale bar: 100 µm (B,D,F,H) and 200 µm (A,C,E,G).

View larger version (23K): [in this window] [in a new window]   Fig. 2. (A) DYRK1A attenuates Notch signalling in cultured cells. SH-SY5Y cells were transfected with the Nrep (pGa981-6, open bars) or Hes-LUC (filled bars) luciferase reporter genes along with expression plasmids encoding DYRK1A or its K188R substitution (amounts indicated in ng). The cells were then cocultured with Jagged-1-expressing SN3T9 or parental L-tk– cells (Lindsell et al., 1995) and the relative luciferase activity was quantified. The levels of luciferase were normalized for the expression of a co-transfected CMV-driven Renilla luciferase. Relative activity is relative to the level of luciferase from cells transfected with each reporter and cultured in the presence of Jagged-1-expressing cells (100%). Values are means and errors of 2-3 experiments. (B) Removal of endogenous DYRK1A causes an increase in the strength of Notch signalling. C2C12 and SH-SY5Y cells were transfected with Nrep and the indicated siRNA with Lipofectamine 2000, and after incubation for 8 hours complexes were removed and cells were and cocultured with Jagged-1-expressing or parental L-tk– cells. Luciferase was measured as in A. Fold induction is relative to cells transfected with Nrep and incubated with control L-tk– cells. (C) Scheme of the relevant domains in Notch1 protein and the fragments used: TM, transmembrane domain; NLS, nuclear localization signal; ANK, ankyrin domain; TAD, transactivation domain. The OPA and PEST regions are represented by shaded rectangles and based on the previous description by Kurooka et al. (Kurooka et al., 1998). The extracellular domain of Notch is not represented to scale (indicated by the diagonal lines). The fragments of the NICD are represented by open rectangles; the NICD contains amino acids 1751-2531 from the mouse gene; NICD-S, 1751-2184; NDE, 1704-2184 and NICD-ANK, 1751-2098. (D) DYRK1A interacts with the RAM and ANK domains of the NICD. SH-SY5Y cells were transfected with Nrep, CMV-driven Renilla luciferase and the indicated DYRK1A and Notch expression vectors (amounts indicated in ng), and luciferase activity was quantified. Relative activity is relative to the normalized level of luciferase from cells transfected with the Nrep reporter and the corresponding NICD fragment. The fold induction for each NICD fragment over the control situation (no NICD; 100%) is indicated. Values are means, and error bars indicate the standard error from 2-3 experiments.

View larger version (40K): [in this window] [in a new window]   Fig. 3. (A) DYRK1A phosphorylation of the NICD shifts its mobility. Whole extracts of transfected HEK293 cells were treated with alkaline phosphatase in the presence or absence of inhibitors. (Upper panel) Cells transfected with the Myc-tagged NICD isoform; (lower panel) with the Myc-tagged NICD-S isoform. Open arrowheads indicate the migration of the NICD in the presence of DYRK1A, the filled arrowheads indicate the normal migration in its absence. Molecular mass marker positions are indicated on the left, in kDa. (B) Retardation of the NICD depends on the presence of the ANK and RAM domains. Whole extracts of HEK293 cells transfected with the Myc-tagged Notch isoforms, indicated at left, and the GFP linked kinases, indicated at bottom, were probed for the NICD constructs (anti-Myc upper panel in each section) and DYRK1A (anti-GFP lower panels). Retardation only occurs in the presence of DYRK1A (arrowheads on the right). Note that both NICD-S and NICD-ANK migrate as doublets (filled arrowheads). The K188R substitution shows a faster migration most probably because of its inability to autophosphorylate.

View larger version (35K): [in this window] [in a new window]   Fig. 4. (A) In vitro phosphorylation of the NICD by DYRK1A. Immunopurified GFP-fused kinases (indicated at the bottom) and Myc-tagged substrates (plasmids indicated on the left) were isolated from whole cell extracts of transfected HEK293 cells. Beads loaded with the immunopurified proteins were mixed and subjected to a kinase reaction in presence of radioactive [-32P]ATP. Enzymes and substrates were separated by immunoprecipitation (indicated on the right) and were assessed separately. Autoradiograms of the resulting immunoprecipitations (IP) indicate the phosphorylation levels of the Notch (-Myc IP) and DYRK1A isoforms (-GFP IP). Note the lack of phosphorylation of NICD-S and NICD-ANK incubated with DYRK1A-K188R or GFP immunocomplexes. Only DYRK1A shows autophosphorylation. (B) 2-D electrophoresis analysis of migration of the NICD-S in the presence of DYRK1A. Plasmids and siRNAs, indicated on the right, were transfected in the cell lines indicated on the left; endogenous expression of DYRK1A is given below. The presence of exogenous DYRK1A induces a shift towards the acidic region of the gradient labelled with a + over the panels. siRNA against DYRK1A prevents the conversion of the NICD into more acidic protein spots.

View larger version (49K): [in this window] [in a new window]   Fig. 5. (A) DYRK1A and the NICD interact in vitro. The two upper samples of each panel are the same western blots probed for the NICD (anti-Myc) and DYRK1A (anti-GFP). The lower part of each panel had one tenth of the input used for immunoprecipitation. Whole cell extract from HEK293 cells transfected with Myc-tagged NICD (upper panel) or Myc-tagged NICD-S (lower panel) and the indicated form of DYRK1A, as shown at the bottom, were immunoprecipitated with -GFP antibodies and immunoblots were probed with the indicated antibodies. Molecular mass standards are indicated in kDa on the right. (B) Effect of DYRK1A expression on Notch stability. HEK293 cells were transfected with the indicated plasmids and 16 hours after transfection were incubated with cycloheximide (100 µg/ml) or the solvent (water) for the indicated time. Note the change in mobility of the NICD in the presence of DYRK1A. No significant changes were observed between the stability of NICD in the presence or absence of DYRK1A.

View larger version (54K): [in this window] [in a new window]   Fig. 6. (A) DYRK1A and the NICD colocalize in the cell nucleus. HEK293 cells were transfected with the expression plasmids encoding the proteins indicated. Single confocal image sections are shown for each channel. GFP detection is of the GFP moiety of the DYRK1A fusions (a,e,i,m), -Myc antibodies detected the tagged NICD protein (b,f,j,n), -Sc35 antibodies detect the endogenous Sc35 splicing factor (c,g,k,o); the Merge images (d,h,l,p) are of the whole row. Insets in i-l are magnifications of the regions indicated to illustrate the colocalization of the NICD and DYRK1A in puncta close to the Sc35 speckles. (B) Endogenous NICD and DYRK1A are localized in the nucleus. Endogenously cleaved Notch1 was detected in C2C12 cells in nuclear puncta, in close vicinity to Sc35 speckles (a-c), and some cytoplasmic puncta are also evident. Endogenous DYRK1A distribution is both cytoplasmic and nuclear. Most of the nuclear puncta are located in close association with Sc35 speckles (d-f). Dotted lines delimit the nuclei.

View larger version (53K): [in this window] [in a new window]   Fig. 7. DYRK1A interferes with Notch signalling in vivo. (A) DYRK1A attenuates the NICD repression on neuritogenesis. Neuro2a cells were transfected with Nrep-RFP, a reporter of Notch activity driving mRFP, plus the plasmids indicated in each panel. Red cells are to those that have received Notch signal. Differentiating Neuro2a cells produce neurites and the NICD represses this process (lower left panel). Co-transfection of DYRK1A relieves this repression (lower right panel) as quantified (mean and s.e.m. of three experiments, Student's t-test: *P0.02, **P0.05). (B) DYRK1A reduces Hes 5-1 expression in the developing neural tube. The right hand side of the neural tube of stage HH12 chick embryos was electroporated with a vector encoding GFP plus control empty vectors (pCX) or those encoding wild-type or inactive forms of DYRK1A. The embryos were subsequently allowed to develop until HH17 before assessing the expression of the Notch signalling readout Hes5-1. The middle panel shows the expression of the co-electroporated GFP-expressing vector on the right half of the neural tube. Dotted lines in left and middle panels indicate the level of the sections shown in the right panels, and white lines in the middle panels outline the neural tube. When the vector encoding DYRK1A was electroporated, Hes 5-1 expression was highly reduced in the electroporated side. Electroporation of both the control empty vector (pCX) or of the inactive DYRK1A-K188R mutant form failed to downregulate Hes 5-1 expression.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter