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

First published online 19 September 2007
doi: 10.1242/dev.011577

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 Related articles in Development 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 Ruel, L. Articles by Thérond, P. P. Search for Related Content PubMed PubMed Citation Articles by Ruel, L. Articles by Thérond, P. P. Social Bookmarking                         What's this?

Phosphorylation of the atypical kinesin Costal2 by the kinase Fused induces the partial disassembly of the Smoothened-Fused-Costal2-Cubitus interruptus complex in Hedgehog signalling

Institute of Signaling, Developmental Biology and Cancer Research, CNRS UMR 6543, Université de Nice-Sophia Antipolis, Parc Valrose, 06108 Nice Cedex 02, France.

View larger version (46K): [in this window] [in a new window]   Fig. 1. One of the Hh-dependent Cos2 phosphorylation sites is located within the Fu-binding domain of Cos2. (A) Western blot analysis showing the differences in the electrophoretic changes of the Fu and Cos2 proteins in S2 cells treated with Hh-N-conditioned medium or okadaic acid (OA). (B) In fu-targeted RNAi cells, the Hh-dependent electrophoretic shift of Cos2 is impaired, whereas it is not impaired upon pka or sgg RNAi treatment. Cellular extracts were subjected to western analysis using Cos2, Fu, Sgg and PKA antibodies. (C) ptc RNAi induced a Cos2 electrophoretic shift, whereas double RNAi treatment targeting ptc and fu blocked this shift. (D) Structure of wild-type Cos2 protein and Myc-tagged deleted constructs. All constructs contain a Myc tag at the C-terminal end that preserved Cos2 function when expressed in transgenic animals. (E) Cells were transfected with fu-V5 and with different cos2-myc constructs (described in D). Immunoprecipitates were analysed for the presence of Fu with an anti-V5 antibody (upper panel) or for the presence of Cos2 with an anti-Myc antibody (lower panel). The large arrow indicates the smallest cos2-myc (543-605) construct capable of interacting with the Fu protein. (F) S2 cells expressing cos2-WT-myc or cos2-572A-myc were treated with OA. Arrows indicate the different phosphorylated states of the proteins. WB, western blot.

View larger version (45K): [in this window] [in a new window]   Fig. 2. Specific immunodetection of the Hh-induced phosphorylated form of Cos2 on Ser572. (A) On western blot analysis, the anti-S572P antibody specifically detects the Hh-induced phosphorylated form of Cos2 on Ser572. Notice the absence of signal in the non-Hh-expressing cells and the strong decrease in the Hh-expressing cells treated with fu RNAi (lane 2 in right panel). (B) After transfection with cos2 and fu constructs, S2 cells were treated with okadaic acid (OA) to mimic Hh pathway activation. Notice that phosphorylated Cos2 was absent when cells were transfected with Fu-KD. (C) S2 cells were transfected with cos2-WT-myc, fu-WT-HA and different variants of smo. Phosphorylated Cos2 on Ser572 was increased in the presence of the SmoDDD variant but was absent in the presence of SmoAAA or Fu-KD.

View larger version (24K): [in this window] [in a new window]   Fig. 3. Activities of Cos2 are affected by mutations in Ser572. (A) Hh-expressing cl8 cells were transfected with increasing amounts of smo and with different amounts of cos2-WT. Ci reporter activity increased in the presence of low levels of transfected cos2-WT. (B-D) smo was transfected at a fixed concentration (10 ng) together with fixed amounts of fu-WT or fu-KD and with different amounts (0-100 ng) of cos2-WT (B), cos2-572A (C) or cos2-572D (D). Notice that Fu-WT potentiates Ci activity, whereas Fu-KD abolishes the positive effect of Cos2. (E) Fixed amounts of ci and ptc-luc vectors were co-transfected with cos2 constructs into S2 cells. At high, but not at low, concentrations, the various Cos2 proteins similarly repressed Ci transactivation. (F) The levels of the three different Cos2 proteins (500 ng) and the Ci proteins were similar in each pool of transfected cells. (G-I) At fixed quantities of ci and ptc-luc vectors, different amounts of the cos2-WT (G), cos2-572A (H) and cos2-572D (I) constructs were co-transfected into S2 cells, together with fixed amounts of fu-WT or fu-KD and in the presence or absence (control) of wild-type Hh. In Hh-stimulated cells, the repressive activity of Cos2-WT is regulated by Fu kinase activity, but it is not in the cos2 mutants in which the phosphosite at position 572 has been affected.

View larger version (37K): [in this window] [in a new window]   Fig. 4. Phosphorylation of Cos2 on Ser572 strongly reduces its interaction with Smo and Ci. (A) Cos2-WT-Myc, Cos2-572A-Myc and Cos2-572D-Myc were revealed with specific antibodies in the initial cl8 cell lysate before immunoprecipitation (IP). Notice that endogenous Smo, Cos2 and Fu proteins were shifted and Ci was stabilised upon Hh treatment. Notice also the presence of a phospho-Ser signal on endogenous Cos2 and on Cos2-WT-Myc in Hh-expressing cells. (B) In Cos2 IP, Fu associated similarly with the three different Cos2 variants. The presence of Smo and Ci was strongly reduced in Cos2-572D-Myc IP (lanes 7 and 8) compared with Cos2-WT-Myc and Cos2-572A-Myc IP (lanes 3-6). Two western blots performed from two independent IP experiments are presented for Smo. Notice also that a portion of Ci was dissociated from the protein complex in Hh-treated cells expressing Cos2-WT. This dissociation was less apparent in Cos2-572A-Myc or Cos2-572D-Myc immunoprecipitates. (C,D) The graphs represent the amount of Smo (C) or Ci (D) associated with the different Cos2 variants after IP, expressed as a percentage relative to the amount of Smo (C) or Ci (D) associated with Cos2-WT in non-Hh-treated cells. The mean of three different experiments is shown. (E) Structure of the wild-type Cos2 protein and the relative localisation of the epitopes recognised by the three different Cos2 antibodies used in the IP experiments. (F,G) Left panels: initial lysates of cl8 cells analysed with specific antibodies. (F) The same extracts were immunoprecipitated with an anti-Cos2 or anti-S572P antibody recognising, respectively, the epitopes shown in blue and in red in E. Under similar conditions, lysates (G) were also submitted to IP with anti-Fu, anti-Smo and three different anti-Cos2 antibodies. Notice the strong decrease in Smo and Ci levels in the IP performed with the S572P antibody. Notice also that phosphorylated Cos2 is absent following IP performed with the anti-Smo antibody, but is present following IP performed with the anti-Fu antibody.

View larger version (110K): [in this window] [in a new window]   Fig. 5. Activation of the Hh pathway is revealed with the phospho-Cos2 antibody. All pictures are wild-type embryos at stage 10-11. (A) En is in red and S572P-Cos2 is revealed in green. Notice that the phospho-Cos2 labelling is only present in Hh-producing cells (in red) and in adjacent cells. (B) Staining with the antibody recognising the heptad repeats of Cos2 (red, single channel in D); S572P-Cos2 is green (single channel in C). (E) Magnification (x1260) of one abdominal segment (delimited in by the square in B) showing the complementary Cos2 (red) and S572P Cos2 (green) staining. (F-H) Confocal x/y sub-apical sections (magnification x1890) of one segment from wild-type stage 11 embryos. S572P-Cos2 (F-H) is green, Ci (G) is red, Smo (H) is pink and Nrt (F) is blue. No overlap of S572P-Cos2 with Ci (G), Smo (H) or with the plasma membrane marker Nrt (F) could be detected. S572P-Cos2 labelling is mainly cytoplasmic. White line delimits the Hh-expressing and -receiving cells.

View larger version (77K): [in this window] [in a new window]   Fig. 6. Cos2-572A and Cos2-572D do not rescue cos2-null phenotypes in embryos. (A-C) Wild-type embryos. (D-O) cos2w1 germline clones (glc; D-F) expressing UAS-cos2-WT (G-I), UAS-cos2-572A (J-L) or UAS-cos2-572D (M-O) under the control of the ubiquitous 69BGal4 driver. (A,D,G,J,M) Cuticle views of first instar larva. (Other panels) In blue, mRNA in situ hybridisation for wg (B,E,H,K,N) and rho (C,F,I,L,O). In brown, immunostaining against En. The wild-type embryo displays an alternating segment pattern of naked cuticle and denticle belts composed of six types of denticles numbered from 1 to 6 (A). In cos2-null embryos, an anterior expansion of wg (E) and a posterior expansion of rho (F) are observed, and only naked cuticle and denticle types 1 and 2 are made (D). Expression of a cos2-WT transgene fully rescues the cos2w1 glc phenotype defects with restored cuticle pattern and correct Hh target gene expression (G-I). By contrast, the rescue is much weaker with the cos2-572A transgene, and is barely observable with cos2-572D. Embryos were raised at similar temperatures (18°C). Two to three different transgenic lines were tested for each variant in rescue experiments and showed comparable results. (P) Western blot for Cos2-Myc and Arm proteins from embryos expressing cos2-WT-myc, cos2-572A-myc or cos2-572D-myc driven by the armGal4 driver. (Q) Bar graph showing the quantity of Cos2-Myc as analysed by western blot as in P, expressed as a percentage relative to the amount of Arm. The mean of two independent experiments is shown.

View larger version (94K): [in this window] [in a new window]   Fig. 7. Phosphorylation of Cos2 on Ser572 impairs its regulation of Ci stability. (A) 71BGal4;UAS-GFP; (B) yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ, (C) yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ; 71BGAL4/UAS cos2-WT-myc, (D) yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ; 71BGAL4/UAS cos2-572A-myc and (E) yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ; 71BGAL4/UAS cos2-572D-myc imaginal discs are stained for Ci (2A1; red), Myc (blue) and ß-galactosidase (green), except in A where GFP is in green. (A,A') Ci is stabilised in 10-15 cells in the anterior compartment near the anteroposterior (AP) border, as in a wild-type disc. The 71B line drives the expression of Gal4 in a large domain centred in the wing pouch indicated by a broken line. (B,B') In cos2W1 clones, the Ci155 isoform is stabilised, as revealed by the 2A1 antibody (arrow in B'). (C-F) Examples of rescue experiments with UAS-cos2-WT (C), UAS-cos2-572A (D) and UAS-cos2-572D (E,F) driven by 71BGal4. The rescue domain is visualised by Myc staining (encircled). cosWI loss-of-function clones are identified by a loss of ß-galactosidase staining (C'-F'). In C''-F'', all the anterior clones outside of the wing pouch are encircled by an unbroken line. The inhibition efficiencies of the wild-type (WT) and mutant cos2 transgenes were quantified in each disc by comparing Ci155 levels in cos2W1 clones situated inside the rescue domain (marked by arrows) with clones outside of this domain (asterisk). Inhibition efficiencies were calculated using data obtained from 8-12 different clones for each Cos2 variant and are indicated as percentages in C''-F''. When compared with the surrounding tissues within the driver domain, the Ci level was increased in mutant clones expressing Cos2-WT, Cos2-572A and Cos2-572D by 10.6, 22 and 38.5%, respectively.

View larger version (109K): [in this window] [in a new window]   Fig. 8. Phosphorylation of Cos2 on Ser572 is necessary for anterior Engrailed expression. (A) 71BGAL4/UAS cos2-WT-myc, (B) 71BGAL4/UAS cos2-572A-myc, (C) 71BGAL4/UAS cos2-572D-myc and (D) yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ imaginal discs are stained for En (4D9) in blue and Myc in green, except in D,D' in which ß-gal is in green. (A'-D') En is in white and an enlargement of the squared region is shown; the bracket indicates the anterior region at the anteroposterior (AP) boundary. (E,E') yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ;ac71BGAL4/UAS cos2-WT-myc, (F-G') yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ; 71BGAL4/UAS cos2-572A-myc and (H,H') yw,hs-flp; FRT42D cos2W1/FRT42D armlacZ; 71BGAL4/UAS cos2-572D-myc imaginal discs are stained for Myc (red), ß-gal (green) and En (blue). (E''-H'') En is in white and enlargements of the cos2W1 clones pointed to by an arrow are shown. (A) In the anterior region, Cos2-WT transgene expression (18°C) is sufficient to strongly inhibit the endogenous expression of En. (B) The two mutant Cos2 variants are less efficient, because en expression is only slightly reduced. (D) Cos2 is necessary for the high level of Hh signalling, because En expression is lost (arrow in D') in anterior cos2W1 loss-of-function clones situated at the AP boundary. (E-H) En expression in anterior clones situated inside the 71B domain at the AP boundary was observed in different discs from distinct experiments (n clones=11 to 15). The rescue domain is visualised by Myc staining and encircled in E-H. cos2WI loss-of-function clones are identified by loss of ß-gal staining, and anterior clones near the AP boundary are encircled by an unbroken line (E'-H'), or arrows (E''-H''). En expression is absent in cos2W1 loss-of-function clones expressing the cos2-WT and cos2-572D transgenes. Expression of the cos2-572A transgene gave similar results in 60% of the discs (F-F''), but in the other 40%, a low level of En was detected in cos2W1 loss-of-function clones in which Cos2-572A was expressed at a low level (G-G''). Notice that the level of En is much higher in the few anterior cells outside of the cos2 mutant clone (arrowhead in G''). Notice also that some cells in the cos2W1 clones (G'') showed no En expression due to the higher expression of the transgene in these cells.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter