- Email: [email protected]
Planar Polarity Is Positively Regulated by Casein Kinase Iɛ in Drosophila
Current Biology 16, 1329–1336, July 11, 2006 ª2006 Elsevier Ltd All rights reserved
DOI 10.1016/j.cub.2006.04.041
Report Planar Polarity Is Positively Regulated by Casein Kinase I3 in Drosophila Helen Strutt,1 Mary Ann Price,1 and David Strutt1,* 1 Centre for Developmental and Biomedical Genetics Department of Biomedical Science University of Sheffield Western Bank Sheffield S10 2TN United Kingdom
Summary Members of the casein kinase I (CKI) family have been implicated in regulating canonical Wnt/Wingless (Wg) signaling by phosphorylating multiple pathway components [1, 2]. Overexpression of CKI in vertebrate embryos activates Wg signaling [3, 4], and one target is thought to be the cytoplasmic effector Dishevelled (Dsh), which is an in vitro target of CKI phosphorylation. Phosphorylation of Dsh by CKI has also been suggested to switch its activity from noncanonical to canonical Wingless signaling [5]. However, in vivo loss-of-function experiments have failed to identify a clear role for CKI in positive regulation of Wg signaling. By examining hypomorphic mutations of the Drosophila CKI3 homolog discs overgrown (dco)/doubletime, we now show that it is an essential component of the noncanonical/planar cell polarity pathway. Genetic interactions indicate that dco acts positively in planar polarity signaling, demonstrating that it does not act as a switch between canonical and noncanonical pathways. Mutations in dco result in a reduced level of Dishevelled phosphorylation in vivo. Furthermore, in these mutants, Dishevelled fails to adopt its characteristic asymmetric subcellular localisation at the distal end of pupal wing cells, and the site of hair outgrowth is disrupted. Finally, we also find that dco function in polarity is partially redundant with CKIa. Results and Discussion dco Interacts with fz during Planar Polarity Signaling Planar polarity is readily observed in most adult tissue of the fly. In the wing, a single trichome emerges from the distal vertex of each cell, while in the eye the ommatidia adopt a characteristic ordered array (Figure 1A). A group of core polarity genes are essential for this polarization. The best characterized is frizzled (fz), which encodes a seven-pass transmembrane protein that signals through the cytoplasmic effector Dsh [6, 7]. Both Fz and Dsh are also involved in canonical Wg signaling, but signaling diverges downstream of Dsh, which thus acts as a branchpoint between the two pathways [8, 9]. We identified the 5B2.6 allele in a genetic screen for modifiers of a hypomorphic fz phenotype in the eye [10]. In the fz19/fz20 heteroallelic combination, 10% of
*Correspondence: [email protected]
ommatidia are mispolarized (Figure 1B). The 5B2.6 mutation enhances this phenotype to 30% polarity defects (Figure 1C), while loss of one copy of dsh enhances it to 41% (Figure 1D) [10]. Recombination and deficiency mapping localized the lethal mutation on the 5B2.6 chromosome to a 280 kb region at the distal tip of chromosome 3R (Figure S1A in the Supplemental Data available with this article online). This domain is predicted to contain 22 genes, of which the most likely candidate was the fly CKI3 homolog dco [11, 12]. The identity of 5B2.6 was confirmed by its failure to complement existing dco alleles, resulting in lethality or in viable adults with broad wings and wing vein defects as previously described (Figure 1E) [12]. Sequencing of the dco gene in the 5B2.6 mutant identified a single amino acid change in a highly conserved arginine residue (R192H, Figure S1B). This residue is located at the kinase domain VIII/IX border: interestingly, it is within a 17 amino acid CKI family-specific signature motif, which may be important for protein-protein interactions [13]. Finally, the interaction of 5B2.6 with fz was implicated as being due to mutation of dco, as shown by the fact that other dco alleles strongly enhance the fz hypomorphic polarity phenotype (Figure 1D). Interestingly, it was previously reported that CKI3 negatively regulates planar polarity, on the basis of JNK kinase assays in tissue culture cells [5, 14]. However, our results show that reduced dco activity enhances a lossof-function fz phenotype, indicating that dco in fact is a positive regulator of the pathway. Gain or Loss of Function of dco Causes Polarity Defects The role of dco in planar polarity was investigated via both gain- and loss-of-function experiments. First, overexpressing dco in the eye or wing causes misorientation of ommatidia and swirling of wing hairs typical of overexpressing other polarity genes (Figures 1F and 1G). More compellingly, we also observed significant polarity defects in the eyes of the viable dco2/dco5B2.6 heteroallelic combination (Figure 1H, note that stronger alleles result in lethality, preventing analysis of their phenotype). Alteration of the polarity of wing hairs was also observed in extreme proximal regions in these flies (see Figure S2G). This phenotype is enhanced by mutation of one copy of other core polarity genes such as fz or prickle (pk) (Figure 1I), such that reproducible polarity swirls are observed in more distal regions of the wing. Hence, dco mutants show planar polarity defects in both the eye and wing. Abnormal Polarity Protein Localization in dco Mutants Prior to hair outgrowth in the pupal wing, the core polarity proteins adopt asymmetric subcellular localizations at the apicolateral plasma membrane. Thus, Fz localizes distally, where it recruits Dsh and the ankyrin repeat protein Diego (Dgo); the transmembrane protein Strabismus
Current Biology 1330
Figure 1. Polarity Phenotypes of dco Alleles (A–C, F, H) Sections through adult eyes of the indicated genotypes. In the wild-type, note that dorsal ommatidia point in the opposite direction and have opposite chirality to ventral ommatidia (A). In the cartoons, dorsal-type ommatidia are in red, ventral-type ommatidia in green, and achiral ommatidia in blue. (D) Graph of the enhancement of the fz19/fz20 polarity phenotype in the eye caused by removal of one copy of the indicated allele. Error bars show the standard deviation. (E, G, I) High- or low-magnification images of wings of adult females, showing the dorsal surface of a distal region (G) or the ventral surface of a more proximal region (I) between veins 3 and 4. dco2/dco5B2.6 flies only show polarity defects proximal to the anterior cross vein.
(Stbm) and cytoplasmic factor Prickle (Pk) are proximal; and the atypical cadherin Flamingo (Fmi) is both proximal and distal. The mechanism controlling this asymmetric distrubution is not well understood, but loss of any one of the polarity proteins results in a failure of asymmetric localisation of all the others and a subsequent loss of polarity in the adult tissue [6, 7].
We thus investigated whether the adult wing planar polarity phenotypes observed in dco mutants were a result of abnormal polarity protein localization in the pupal stage. Mutant clones of strong dco alleles fail to proliferate [12]. However, examining clones of weaker alleles revealed that in mutant tissue, the localization of Dsh at distal cell membranes is disrupted, and Dsh instead
Casein Kinase I3 in Planar Polarity 1331
Figure 2. Localization of Polarity Proteins in Pupal Wings (A and B) dcoj3B9 clones, marked by loss of GFP (green). (A) 28 hr pupal wing, stained for Dsh (red). (B) 32 hr APF pupal wing stained with Phalloidin (red) to show actin-rich prehairs and Fmi (blue). Arrows mark prehairs initiating in the center of cells. (C) Expression of Actin-Dco-EGFP (green) in a clone in a 28 hr pupal wing, costained with Fmi (red). (D) Reconstructed XZ sections stained for Dco-EGFP (green), Fmi (red), and Dsh (blue) showing that Dco-EGFP localizes apicolaterally with Fmi and Dsh. (E) Dco-EGFP localization (green) in a fmiE59 mutant clone, marked by loss of lacZ (red) and costained with DE-cadherin (blue).
becomes localized in punctae that are uniformly distributed around the apicolateral plasma membrane (Figure 2A). Similar disruption of the asymmetric distribution of other polarity proteins was also observed (data not shown). The effects of dco mutations on trichome placement were also analyzed. Prehairs normally emerge from the distal edge of pupal wing cells [15]. In dco mutant tissue, however, trichome inititation is delayed, and the prehairs often emerge from the cell center (Figure 2B). The disruption of asymmetric localization of polarity proteins and the resulting delay in trichome initiation and defects in trichome placement are all phenotypes characteristic of mutations in the core polarity genes. Therefore, we suppose that dco acts by regulating the phosphorylation state of one or more of the core polarity proteins and by modulating their ability to localize asymmetrically. Dco Localizes to the Apicolateral Plasma Membrane but Is Not Asymmetrically Distributed As dco regulates the asymmetric localization of the core polarity genes, we wondered whether Dco itself was asymmetrically distributed. A Dco-EGFP transgene was generated, which could be expressed in clones under control of the ubiquitous Actin promoter. This transgene is functional in flies, rescuing the polarity and patterning defects of viable dco2/dco5B2.6 adults, and also rescuing lethal allele combinations to viability with no residual polarity defects (Figure S2). In pupal wings, Dco-EGFP is enriched at the apicolateral plasma membrane, with a significant proportion
also in the cytoplasm (Figures 2C and 2D). However, there is no evidence of asymmetric localization to either the proximal or distal cell edges. Furthermore, we find that the subcellular distribution of Dco-EGFP is not significantly altered in fmi mutants (Figure 2E). fmi function is required for apical recruitment of all the other core polarity proteins [16–18], so this suggests that the apicolateral plasma membrane association of Dco-EGFP is independent of core polarity gene function. Thus, Dco localizes to a subcellular domain where it can interact with the core polarity genes but is not itself asymmetrically distributed. However, this does not preclude Dco activity being asymmetrically regulated. dco Is Required for Phosphorylation of Dsh In Vivo CKI has been found to bind to and phosphorylate Dishevelled in a variety of tissue culture and in vitro assays [3–5, 19–22]. Furthermore, Dsh recruitment to the plasma membrane by Fz has been correlated with an increase in Dsh phosphorylation [17, 23]. We therefore thought it likely that the role of dco in planar polarity could be to phosphorylate Dsh. To investigate this, the phosphorylation status of Dsh in wild-type and mutant pupal wings was analyzed. In SDS-PAGE, wild-type Dsh migrates as two distinct bands: the upper band is due to phosphorylation, as shown by the fact that it is lost when samples are treated with phosphatases (data not shown; [24]). Furthermore, Dsh phosphorylation is dependent on fz activity (compare middle and last lanes of Figure 3A [23]). Significantly, the phosphorylation of Dsh is also substantially reduced in pupal
Current Biology 1332
Figure 3. Phosphorylation of Dsh by Dco In Vivo (A) Western blot of proteins from 28 hr pupal wings of the indicated genotype, probed with Dsh antibody (top) and Actin (bottom) as a loading control. (B) Western blot of proteins from 28 hr pupal wings from whsFLP; tub-GAL4/UAS>STOP>dcoK38R pupal wings that were heatshocked to induce expression at the indicated times. (C) Domain structure of Dsh and putative phosphorylation sites. The position of the DIX (Dishevelled, Axin), PDZ, and DEP (Dishevelled, Egl-10, Pleckstrin) domains are indicated, together with a basic region b. The serine/threonine-rich region between the basic domain and the PDZ domain is shown for fly Dsh and mouse Dvl1, the mutated residues in DshST5 are in red, and the additional residues mutated in DshST8 are in blue. These serine/threonine residues are conserved throughout the animal kingdom, but residue 235 is alanine in many species, and thus unlikely to be important for Dsh function. (D and E) Dorsal surfaces of a dsh1 mutant wing (affects planar polarity but not canonical Wg signaling) (D), or a dsh3 mutant wing (null for canonical Wg signaling and planar polarity) in the presence of DshST5-GFP (E). The lack of rescue previously reported [25] may be due to the different expression systems. (F) 28 hr APF DshST5-GFP pupal wing, stained with GFP antibody. (G) Western blot of wild-type Dsh (left) or DshST5 (right) transfected into HEK293 cells with or without Fz.
wings from dco2/dco5B2.6 hypomorphs (lanes 2 and 4, Figure 3A) or in flies expressing dominant-negative forms of dco during pupal development (Figure 3B). Given that Dsh can be phosphorylated by CKI in vitro, these results indicate that Dsh is the probable in vivo target of dco in planar polarity signaling. A fragment of Dsh containing the basic region and the PDZ domain has been shown to be phosphorylated by CKI [3, 5, 22]. Furthermore, mutation of a cluster of six serines/threonines in this region to alanine (residues 235–242) was reported to result in a form of Dsh that was unable to rescue the polarity phenotype of dsh1 animals [25]. These serine/threonines all correspond to CKI consensus phosphorylation sites (S/Tp-X-(X)-S/T, requiring priming phosphorylation by another kinase, [26]) and could thus be in vivo target sites for Dco. To investigate this further, we expressed a similar transgene (DshST5-GFP), in which all five of the conserved serine/threonine residues in residues 236–242 (red in Figure 3C) are mutated, under control of the endogenous dsh promoter [23]. Despite blocking any potential CKI phosphorylation at these sites, this transgene was able to rescue null dsh3 mutants to viability. Furthermore, both the ommatidial and wing hair polarity phenotypes of dsh1 and dsh3 mutants were fully rescued (Figure 3E and not shown). Consistent with this rescue, DshST5-GFP is able to localize asymmetrically in the pupal wing (Figure 3F), and this mutant form of Dsh is also hyperphosphorylated when coexpressed with Fz in tissue-culture cells (Figure 3G). These results suggest that contrary to previous reports, serine/threonine residues 236–242 of Dsh do not play an essential role in planar polarity or canonical Wnt signaling. Furthermore, a form of Dsh with eight serine/threonine residues mutated to alanine (DshST8GFP, red and blue in Figure 3C) also rescues dsh3 mutants to viability and rescues the polarity phenotypes of dsh1 mutants over most of the wing, although some regions of abnormal polarity remain (Figure 3H). While this again shows that these residues are not essential for Dsh function, this could suggest that they have a partial role in Dsh phosphorylation; however, mutation of multiple amino acids could also disrupt Dsh structure or stability. The failure to identify specific CKI phosphorylation sites that are essential for Dsh function could be explained by a number of mechanisms. First, Dco might not act by direct phosphorylation of Dsh, and the effects on planar polarity could represent a kinase-independent function of Dco. However, this is unlikely; overexpression of a kinase-dead mutant (dcoK38R) has a loss-offunction planar polarity phenotype in the wing, suggesting that the kinase activity is required (Figure 3I, compare to Figure 4C). Alternatively, Dco might act indirectly to mediate Dsh phosphorylation by acting on another substrate. However, we think it is more likely that Dsh is the direct target of Dco but that the phosphorylation
(H) Dorsal surface of a dsh1 mutant wing in the presence of DshST8GFP. Note rescue is complete below vein 4, but not between veins 3 and 4. (I) Dorsal surface of a wing expressing ptc-GAL4, UAS-dcoK38R. Wing hairs point toward the AP boundary of the wing.
Casein Kinase I3 in Planar Polarity 1333
Figure 4. Partial Redundancy between dco and CKIa (A) Wild-type wing. (B) ptc-GAL4, pWIZ-dco wing, raised at 29ºC. Loss of wing material between veins 3 and 4 (arrow) masks any polarity phenotype. (C and D) Dorsal surfaces of ptc-GAL4, pWIZ-dco (C) or ptc-GAL4, pWIZ-dco, pWIZ-CKIa (D) wings, raised at 25ºC. (E) Quantitation of ptc-GAL phenotypes. The number of wings with polarity swirls extending along the indicated length of the wing is plotted, and significance p < 0.001. (F and G) Wings discs containing dcoj3B9 mutant clones, marked by loss of lacZ (red), stained for the Wg target genes Senseless (F) and Distalless (G) in green.
sites are redundant. For example, Dsh could be capable of being phosphorylated on multiple sites, and mutation of only a subset of these sites might not compromise its function. CKIa Cooperates with dco in Planar Polarity Signaling In Xenopus, overexpression of any CKI isoform can activate Wg signaling [22], and it is unclear which CKI isoform is responsible for Dsh phosphorylation in vivo, or if they act redundantly [1]. Dominant-negative forms of CKI3 block the ability of Wnts to induce ectopic axes in Xenopus [3]; however, this may not be specific. Furthermore, similar experiments with dsRNAi against CKI3 and CKIa in tissue culture have given conflicting results [20, 27]. We examined whether any other CKI isoforms could be involved in phosphorylating Dsh during planar polarity signaling, focusing our studies on CKIa, which has recently been shown to cooperate with dco in regulating Cubitus interruptus (Ci) phosphorylation during Hedgehog signaling [28]. No mutants for CKIa are available, so we made use of inducible RNAi constructs specific to dco and CKIa. Expressing dsRNA against dco under control of the ptc-GAL4 driver at high temperatures resulted in a dramatic narrowing between veins 3 and 4 (Figures 4A
and 4B): this could be due to a failure in cell proliferation or an effect on phosphorylation of Smoothened [29–31]. At lower temperatures, wing hairs between veins 3 and 4 are mispolarized, such that in the proximal region of the wing where ptc-GAL4 expression is highest, hairs tend to point toward the anterior-posterior boundary (Figure 4C). This phenotype is similar to the effects of expressing dsRNA against fz (R. Bastock and D.S., data not shown): in both cases, hairs point from regions of high activity to low activity, consistent with dco acting positively in planar polarity. Interestingly, expression of a CKIa dsRNA alone gives no phenotype, but when both CKIa and dco dsRNAs are coexpressed, the polarity phenotype is dramatically enhanced, such that polarity swirls are also seen in more distal regions where ptc-GAL4 expression is lower (Figure 4D). Quantitation of the phenotypes reveals that dco dsRNA flies exhibit polarity swirls extending only 4% of the distance between the posterior cross vein and the distal tip of the wing, while in the double dsRNAi flies, the phenotype extends to 17% of the wing length (Figure 4E). Our results therefore show that CKIa can at least partially substitute for dco function in planar polarity signaling. Nevertheless, dco cannot be fully redundant with CKIa, because dco mutations alone cause reduced Dsh phosphorylation and polarity defects.
Current Biology 1334
dco Mutants Do Not Affect Canonical Wg Signaling In addition to their roles in planar polarity, Fz and Dsh are required for canonical Wg signaling. In the absence of Wg ligand, the transcriptional coactivator b-catenin/ Armadillo (Arm) is constitutively phosphorylated and targeted for degradation. Binding of Wg to its coreceptors Frizzled (Fz) and LRP (LDL receptor-related protein) results in recruitment of Dsh to the receptor complex. This ultimately leads to inhibition of Arm phosphorylation/degradation, allowing it to enter the nucleus and activate transcription [32]. As phosphorylation of Dsh by CKI has been suggested to positively activate Wg signaling in vertebrates, we examined the role of dco during canonical Wg signaling in flies. To do this, we generated clones of hypomorphic dco alleles that do affect planar polarity, in the wing imaginal disc. However, the expression of both high- and lowthreshold Wg target genes is unaffected in these clones (Figures 4F and 4G). Furthermore, Wg target gene expression is also unaltered in clones of cells expressing dsRNA against dco (data not shown). We also attempted to investigate Wg signaling in embryos by making dco germline clones; however, no embryos were produced, presumably due to a requirement for dco function in the germline. These results argue that dco function is less essential in this context than in polarity signaling. It is possible that this is due to the clones only partially lacking dco activity. However, we think it more likely that dco is redundant with one or more other CKI isoforms in canonical Wg signaling. As CKIa cooperates with dco in phosphorylating Dsh in polarity signaling, it is probable that CKIa is also a redundant kinase in canonical signaling. Notably, CKIa also negatively regulates canonical signaling by phosphorylating and thus stabilizing b-catenin/Arm [19, 33–35], precluding a simple analysis of its regulation of Dsh. Conclusions The membrane recruitment of Dsh during planar polarity signaling has previously been correlated with its phosphorylation [17, 23]; however, the significance of this was unclear. We have now demonstrated that this phosphorylation is likely to be due to membrane-recruited Dsh coming into proximity with Dco/CKI3. Furthermore, in the absence of this phosphorylation, Dsh is unable to adopt its proper asymmetric localization, and planar polarity patterning is disrupted. This positive regulation of Dsh by CKI3 phosphorylation contrasts with previous reports of CKI3 negatively regulating noncanonical Wnt signaling and acting as a switch to canonical signaling [5, 14]. The asymmetric proximal-distal distribution of core polarity proteins such as Dsh is thought to be the result of an initial asymmetry in protein activity induced by an upstream cue, which is then amplified by interactions between the core polarity proteins across the proximal-distal cell boundaries [18, 23, 36]. Our results suggest that phosphorylation of Dsh by Dco is essential for Dsh to participate in this stabilization of the asymmetric distribution of the core polarity proteins. Dco is not itself asymmetrically distributed within pupal wing cells, although it is seen at the apicolateral plasma membrane where it would be able to interact
with Dsh. CKI isoforms are usually thought to be constitutively active, and their ability to phosphorylate their target proteins is regulated by subcellular localization and proximity to the substrate [1, 37]. Hence, Dco may be a permissive cue in planar polarity signaling, with recruitment of Dsh to the apicolateral plasma membrane by Fz [8] bringing Dco and Dsh into proximity and permitting the constitutive phosophorylation of Dsh. Alternatively, despite the lack of asymmetric protein localization, Dco protein activity could be asymmetrically regulated such that it preferentially phosphorylates Dsh at the distal end of the cell and thus is instructive in polarity signaling. Such activation could be mediated by upstream polarity genes or other core polarity genes such as Fz. The mechanism by which this could occur is unclear, but some CKI isoforms have been shown to be regulated by inhibitory autophosphorylation of their C termini [1]. Experimental Procedures Fly Strains and Histology Mutant alleles are as described in FlyBase and mitotic clones were generated by the FLP/FRT system [38]. Pupal wings were dissected at 28 hr after prepupa formation (APF) or 32 hr APF for trichomes, as previously described [18]. Antibodies used were Fmi mouse monoclonal [39], DE-cadherin rat monoclonal [40], GFP affinity purified rabbit serum (Abcam), lacZ mouse monoclonal (Promega), Senseless guinea pig [41], and Distalless rabbit [42]. Actin was visualized with Phalloidin Texas red (Molecular Probes). The Dsh antibody is a rat serum directed against amino acids 480–623 of Dsh. DNA Constructs dco cDNA was provided by BDGP (clone number LD27173), and CKIa cDNA was made by RT-PCR from embryos. Dco-EGFP is a fusion at the C-terminal end of the full-length Dco open reading frame to the N-terminal end of EGFP. The dominant-negative variant dcoK38R has lysine 38 mutated to arginine. These variants were cloned into a pCasper4 transformation vector containing an FRT>STOP>FRT sequence downstream of the Actin5C promoter [18] for inducible ubiquitous expression. For inducible high-level expression, the pUAST vector [43] was modified by introducing an FRT>STOP>FRT sequence downstream of the promoter. dsRNAi constructs were made by inserting nucleotides 859–1258 of the coding sequence of CKIa and nucleotides 634–1033 of the coding sequence of dco into the modified pUAST vector pWIZ [44]. These sequences are outside of the conserved kinase domain and are not predicted to interfere with any other of the eight fly homologs of CKI. Dsh-GFP phosphorylation site mutants were made by mutation of the serine/threonine residues of amino acids 236–242 or 236–247 to alanine. They were inserted into a pCasper4 vector containing 2.9 kb of genomic sequences upstream of the dsh gene, which was tagged at the 30 end with EGFP [23]. For expression in tissue-culture cells, the dsh cDNA was cloned into pcDNA3.1+ (Invitrogen) and the mutated amino acids substituted. Western Blots Pupal wings were dissected either directly into sample buffer, or an extract was made in lysis buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.5% Triton X-100, protease inhibitors [Roche], and 1 mM phosphatase inhibitor microcystin). Identical results were observed with both methods. Proteins were detected by Western blotting against single pupal wing equivalents, via an affinity-purified rabbit antibody generated against amino acids 480–623 of Dsh, or Actin AC-40 mouse monoclonal (Sigma). Supplemental Data Two Supplemental Figures can be found with this article online at http://www.current-biology.com/cgi/content/full/16/13/1329/DC1/.
Casein Kinase I3 in Planar Polarity 1335
Acknowledgments We thank BDGP for ESTs; Mike Young for fly strains; Tadashi Uemura, Hugo Bellen, and Sean Carroll for antibodies; and Jeff Axelrod and Richard Carthew for plasmids. Vickie Thomas-McArthur is thanked for generating transformants, Katrina Hofstra and Melanie Stapleton for molecular biology assistance, and Martin Zeidler for comments on the manuscript. Confocal facilities were provided by Yorkshire Cancer Research, and the work was funded by the Wellcome Trust and MRC. D.S. is a Wellcome Trust Senior Fellow in Basic Biomedical Science. Received: March 14, 2006 Revised: April 7, 2006 Accepted: April 10, 2006 Published: July 10, 2006
References 1. Knippschild, U., Gocht, A., Wolff, S., Huber, N., Lohler, J., and Stoter, M. (2005). The casein kinase 1 family: participation in multiple cellular processes in eukaryotes. Cell. Signal. 17, 675– 689. 2. Price, M.A. (2006). CKI, there’s more than one: casein kinase I family members in Wnt/Wingless and Hedgehog signaling. Genes Dev. 20, 399–410. 3. Peters, J.M., McKay, R.M., McKay, J.P., and Graff, J.M. (1999). Casein kinase I transduces Wnt signals. Nature 401, 345–350. 4. Sakanaka, C., Leong, P., Xu, L., Harrison, S.D., and Williams, L.T. (1999). Casein kinase I epsilon in the Wnt pathway: regulation of beta-catenin function. Proc. Natl. Acad. Sci. USA 96, 12548– 12552. 5. Cong, F., Schweizer, L., and Varmus, H. (2004). Casein kinase I epsilon modulates the signaling specificities of Dishevelled. Mol. Cell. Biol. 24, 2000–2011. 6. Eaton, S. (2003). Cell biology of planar polarity transmission in the Drosophila wing. Mech. Dev. 120, 1257–1264. 7. Strutt, D. (2003). Frizzled signalling and cell polarisation in Drosophila and vertebrates. Development 130, 4501–4513. 8. Axelrod, J.D., Miller, J.R., Shulman, J.M., Moon, R.T., and Perrimon, N. (1998). Differential recruitment of Dishevelled provides signaling specificity in the planar cell polarity and Wingless signaling pathways. Genes Dev. 12, 2610–2622. 9. Boutros, M., Paricio, N., Strutt, D.I., and Mlodzik, M. (1998). Dishevelled activates JNK and discriminates between JNK pathways in planar polarity and wingless signaling. Cell 94, 109–118. 10. Strutt, H., and Strutt, D. (2003). EGF signaling and ommatidial rotation in the Drosophila eye. Curr. Biol. 13, 1451–1457. 11. Kloss, B., Price, J.L., Saez, L., Blau, J., Rothenfluh, A., Wesley, C.S., and Young, M.W. (1998). The Drosophila clock gene double-time encodes a protein closely related to human Casein kinase I epsilon. Cell 94, 97–107. 12. Zilian, O., Burke, R., Brentrup, D., Gutjahr, T., Bryant, P., and Noll, M. (1999). double-time is identical to discs overgrown, which is required for cell survival, proliferation and growth arrest in Drosophila imaginal discs. Development 126, 5409–5420. 13. Longenecker, K.L., Roach, P.J., and Hurley, T.D. (1996). Threedimensional structure of mammalian casein kinase I: molecular basis for phosphate recognition. J. Mol. Biol. 257, 618–631. 14. McKay, R.M., Peters, J.M., and Graff, J.M. (2001). The casein kinase I family: roles in morphogenesis. Dev. Biol. 235, 378–387. 15. Wong, L.L., and Adler, P.N. (1993). Tissue polarity genes of Drosophila regulate the subcellular location for prehair initiation in pupal wing cells. J. Cell Biol. 123, 209–221. 16. Bastock, R., Strutt, H., and Strutt, D. (2003). Strabismus is asymmetrically localised and binds to Prickle and Dishevelled during Drosophila planar polarity patterning. Development 130, 3007– 3014. 17. Shimada, Y., Usui, T., Yanagawa, S., Takeichi, M., and Uemura, T. (2001). Asymmetric co-localisation of Flamingo, a seven-pass transmembrane cadherin, and Dishevelled in planar cell polarisation. Curr. Biol. 11, 859–863.
18. Strutt, D.I. (2001). Asymmetric localisation of Frizzled and the establishment of cell polarity in the Drosophila wing. Mol. Cell 7, 367–375. 19. Gao, Z.-H., Seeling, J.M., Hill, V., Yochum, A., and Virshup, D.M. (2002). Casein kinase I phosphorylates and destabilises the beta-catenin degradation complex. Proc. Natl. Acad. Sci. USA 99, 1182–1187. 20. Hino, S., Michiue, T., Asashima, M., and Kikuchi, A. (2003). Casein kinase I epsilon enhances the binding of Dvl-1 to Frat-1 and is essential for Wnt-3a-induced accumulation of beta-catenin. J. Biol. Chem. 278, 14066–14073. 21. Kishida, M., Hino, S., Michiue, T., Yamamoto, H., Kishida, S., Fukui, A., Asashima, M., and Kikuchi, A. (2001). Synergistic activation of the Wnt signaling pathway by Dvl and casein kinase I epsilon. J. Biol. Chem. 276, 33147–33155. 22. McKay, R.M., Peters, J.M., and Graff, J.M. (2001). The casein kinase I family in Wnt signaling. Dev. Biol. 235, 388–396. 23. Axelrod, J.D. (2001). Unipolar membrane association of Dishevelled mediates Frizzled planar cell polarity signalling. Genes Dev. 15, 1182–1187. 24. Yanagawa, S.-i., van Leeuwen, F., Wodarz, A., Klingensmith, J., and Nusse, R. (1995). The Dishevelled protein is modified by Wingless signalling in Drosophila. Genes Dev. 9, 1087–1097. 25. Penton, A., Wodarz, A., and Nusse, R. (2002). A mutational analysis of dishevelled in Drosophila defines novel domains in the Dishevelled protein as well as novel suppressing alleles of axin. Genetics 161, 747–762. 26. Flotow, H., Graves, P.R., Wang, A.Q., Fiol, C.J., Roeske, R.W., and Roach, P.J. (1990). Phosphate groups as substrate determinants for casein kinase I action. J. Biol. Chem. 265, 14264– 14269. 27. Matsubayashi, H., Sese, S., Lee, J.-S., Shirakawa, T., Iwatsubo, T., Tomita, T., and Yanagawa, S. (2004). Biochemical characterisation of the Drosophila Wingless signaling pathway based on RNA interference. Mol. Cell. Biol. 24, 2012–2024. 28. Jia, J., Zhang, L., Zhang, Q., Tong, C., Wang, B., Hou, F., Amanai, K., and Jiang, J. (2005). Phosphorylation by Double-Time/ CKIepsilon and CKIalpha targets cubitus interruptus for Slimb/ beta-TRCP-mediated proteolytic processing. Dev. Cell 9, 819– 830. 29. Apionishev, S., Katanayeva, N.M., Marks, S.A., Kalderon, D., and Tomlinson, A. (2005). Drosophila Smoothened phosphorylation sites essential for Hedgehog signal transduction. Nat. Cell Biol. 7, 86–92. 30. Jia, J., Tong, C., Wang, B., Luo, L., and Jiang, J. (2004). Hedgehog signalling activity of Smoothened requires phosphorylation by protein kinase A and casein kinase I. Nature 432, 1045–1050. 31. Zhang, C., Williams, E.H., Guo, Y., Lum, L., and Beachy, P.A. (2004). Extensive phosphorylation of Smoothened in Hedgehog pathway activation. Proc. Natl. Acad. Sci. USA 101, 17900– 17907. 32. Logan, C.Y., and Nusse, R. (2004). The Wnt signaling pathway in development and disease. Annu. Rev. Cell Dev. Biol. 20, 781–810. 33. Amit, S., Hatzubai, A., Birman, Y., Anderson, J.S., Ben-Shushan, E., Mann, M., Ben-Neriah, Y., and Alkalay, I. (2002). Axin-mediated CKI phosphorylation of beta-catenin at Ser45: a molecular switch for the Wnt pathway. Genes Dev. 16, 1066–1076. 34. Liu, C., Li, Y., Semenov, M., Han, C., Baeg, G.-H., Tan, Y., Zhang, Z., Lin, X., and He, X. (2002). Control of beta-catenin phosphorylation/degradation by a dual-kinase mechanism. Cell 108, 837– 847. 35. Yanagawa, S., Matsuda, Y., Lee, J.-S., Matsubayashi, H., Sese, S., Kadowaki, T., and Ishimoto, A. (2002). Casein kinase I phosphorylates the Armadillo protein and induces its degradation in Drosophila. EMBO J. 21, 1733–1742. 36. Tree, D.R.P., Shulman, J.M., Rousset, R., Scott, M.P., Gubb, D., and Axelrod, J.D. (2002). Prickle mediates feedback amplification to generate asymmetric planar cell polarity signalling. Cell 109, 371–381. 37. Gross, S.D., and Anderson, R.A. (1998). Casein kinase I: spatial organisation and positioning of a multifunctional protein kinase family. Cell. Signal. 10, 699–711.
Current Biology 1336
38. Xu, T., and Rubin, G.M. (1993). Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 117, 1223–1237. 39. Usui, T., Shima, Y., Shimada, Y., Hirano, S., Burgess, R.W., Schwarz, T.L., Takeichi, M., and Uemura, T. (1999). Flamingo, a seven-pass transmembrane cadherin, regulates planar cell polarity under the control of Frizzled. Cell 98, 585–595. 40. Oda, H., Uemura, T., Harada, Y., Iwai, Y., and Takeichi, M. (1994). A Drosophila homolog of cadherin associated with Armadillo and essential for embryonic cell-cell adhesion. Dev. Biol. 165, 716–726. 41. Nolo, R., Abbott, L.A., and Bellen, H.J. (2000). Senseless, a Zn finger transcription factor, is necessary and sufficient for sensory organ development in Drosophila. Cell 102, 349–362. 42. Panganiban, G., Sebring, A., Nagy, L., and Carroll, S. (1995). The development of crustacean limbs and the evolution of arthropods. Science 270, 1363–1366. 43. Brand, A.H., and Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415. 44. Lee, Y.S., and Carthew, R.W. (2003). Making a better RNAi vector for Drosophila: use of intron spacers. Methods 30, 322–329.
DOI 10.1016/j.cub.2006.04.041
Report Planar Polarity Is Positively Regulated by Casein Kinase I3 in Drosophila Helen Strutt,1 Mary Ann Price,1 and David Strutt1,* 1 Centre for Developmental and Biomedical Genetics Department of Biomedical Science University of Sheffield Western Bank Sheffield S10 2TN United Kingdom
Summary Members of the casein kinase I (CKI) family have been implicated in regulating canonical Wnt/Wingless (Wg) signaling by phosphorylating multiple pathway components [1, 2]. Overexpression of CKI in vertebrate embryos activates Wg signaling [3, 4], and one target is thought to be the cytoplasmic effector Dishevelled (Dsh), which is an in vitro target of CKI phosphorylation. Phosphorylation of Dsh by CKI has also been suggested to switch its activity from noncanonical to canonical Wingless signaling [5]. However, in vivo loss-of-function experiments have failed to identify a clear role for CKI in positive regulation of Wg signaling. By examining hypomorphic mutations of the Drosophila CKI3 homolog discs overgrown (dco)/doubletime, we now show that it is an essential component of the noncanonical/planar cell polarity pathway. Genetic interactions indicate that dco acts positively in planar polarity signaling, demonstrating that it does not act as a switch between canonical and noncanonical pathways. Mutations in dco result in a reduced level of Dishevelled phosphorylation in vivo. Furthermore, in these mutants, Dishevelled fails to adopt its characteristic asymmetric subcellular localisation at the distal end of pupal wing cells, and the site of hair outgrowth is disrupted. Finally, we also find that dco function in polarity is partially redundant with CKIa. Results and Discussion dco Interacts with fz during Planar Polarity Signaling Planar polarity is readily observed in most adult tissue of the fly. In the wing, a single trichome emerges from the distal vertex of each cell, while in the eye the ommatidia adopt a characteristic ordered array (Figure 1A). A group of core polarity genes are essential for this polarization. The best characterized is frizzled (fz), which encodes a seven-pass transmembrane protein that signals through the cytoplasmic effector Dsh [6, 7]. Both Fz and Dsh are also involved in canonical Wg signaling, but signaling diverges downstream of Dsh, which thus acts as a branchpoint between the two pathways [8, 9]. We identified the 5B2.6 allele in a genetic screen for modifiers of a hypomorphic fz phenotype in the eye [10]. In the fz19/fz20 heteroallelic combination, 10% of
*Correspondence: [email protected]
ommatidia are mispolarized (Figure 1B). The 5B2.6 mutation enhances this phenotype to 30% polarity defects (Figure 1C), while loss of one copy of dsh enhances it to 41% (Figure 1D) [10]. Recombination and deficiency mapping localized the lethal mutation on the 5B2.6 chromosome to a 280 kb region at the distal tip of chromosome 3R (Figure S1A in the Supplemental Data available with this article online). This domain is predicted to contain 22 genes, of which the most likely candidate was the fly CKI3 homolog dco [11, 12]. The identity of 5B2.6 was confirmed by its failure to complement existing dco alleles, resulting in lethality or in viable adults with broad wings and wing vein defects as previously described (Figure 1E) [12]. Sequencing of the dco gene in the 5B2.6 mutant identified a single amino acid change in a highly conserved arginine residue (R192H, Figure S1B). This residue is located at the kinase domain VIII/IX border: interestingly, it is within a 17 amino acid CKI family-specific signature motif, which may be important for protein-protein interactions [13]. Finally, the interaction of 5B2.6 with fz was implicated as being due to mutation of dco, as shown by the fact that other dco alleles strongly enhance the fz hypomorphic polarity phenotype (Figure 1D). Interestingly, it was previously reported that CKI3 negatively regulates planar polarity, on the basis of JNK kinase assays in tissue culture cells [5, 14]. However, our results show that reduced dco activity enhances a lossof-function fz phenotype, indicating that dco in fact is a positive regulator of the pathway. Gain or Loss of Function of dco Causes Polarity Defects The role of dco in planar polarity was investigated via both gain- and loss-of-function experiments. First, overexpressing dco in the eye or wing causes misorientation of ommatidia and swirling of wing hairs typical of overexpressing other polarity genes (Figures 1F and 1G). More compellingly, we also observed significant polarity defects in the eyes of the viable dco2/dco5B2.6 heteroallelic combination (Figure 1H, note that stronger alleles result in lethality, preventing analysis of their phenotype). Alteration of the polarity of wing hairs was also observed in extreme proximal regions in these flies (see Figure S2G). This phenotype is enhanced by mutation of one copy of other core polarity genes such as fz or prickle (pk) (Figure 1I), such that reproducible polarity swirls are observed in more distal regions of the wing. Hence, dco mutants show planar polarity defects in both the eye and wing. Abnormal Polarity Protein Localization in dco Mutants Prior to hair outgrowth in the pupal wing, the core polarity proteins adopt asymmetric subcellular localizations at the apicolateral plasma membrane. Thus, Fz localizes distally, where it recruits Dsh and the ankyrin repeat protein Diego (Dgo); the transmembrane protein Strabismus
Current Biology 1330
Figure 1. Polarity Phenotypes of dco Alleles (A–C, F, H) Sections through adult eyes of the indicated genotypes. In the wild-type, note that dorsal ommatidia point in the opposite direction and have opposite chirality to ventral ommatidia (A). In the cartoons, dorsal-type ommatidia are in red, ventral-type ommatidia in green, and achiral ommatidia in blue. (D) Graph of the enhancement of the fz19/fz20 polarity phenotype in the eye caused by removal of one copy of the indicated allele. Error bars show the standard deviation. (E, G, I) High- or low-magnification images of wings of adult females, showing the dorsal surface of a distal region (G) or the ventral surface of a more proximal region (I) between veins 3 and 4. dco2/dco5B2.6 flies only show polarity defects proximal to the anterior cross vein.
(Stbm) and cytoplasmic factor Prickle (Pk) are proximal; and the atypical cadherin Flamingo (Fmi) is both proximal and distal. The mechanism controlling this asymmetric distrubution is not well understood, but loss of any one of the polarity proteins results in a failure of asymmetric localisation of all the others and a subsequent loss of polarity in the adult tissue [6, 7].
We thus investigated whether the adult wing planar polarity phenotypes observed in dco mutants were a result of abnormal polarity protein localization in the pupal stage. Mutant clones of strong dco alleles fail to proliferate [12]. However, examining clones of weaker alleles revealed that in mutant tissue, the localization of Dsh at distal cell membranes is disrupted, and Dsh instead
Casein Kinase I3 in Planar Polarity 1331
Figure 2. Localization of Polarity Proteins in Pupal Wings (A and B) dcoj3B9 clones, marked by loss of GFP (green). (A) 28 hr pupal wing, stained for Dsh (red). (B) 32 hr APF pupal wing stained with Phalloidin (red) to show actin-rich prehairs and Fmi (blue). Arrows mark prehairs initiating in the center of cells. (C) Expression of Actin-Dco-EGFP (green) in a clone in a 28 hr pupal wing, costained with Fmi (red). (D) Reconstructed XZ sections stained for Dco-EGFP (green), Fmi (red), and Dsh (blue) showing that Dco-EGFP localizes apicolaterally with Fmi and Dsh. (E) Dco-EGFP localization (green) in a fmiE59 mutant clone, marked by loss of lacZ (red) and costained with DE-cadherin (blue).
becomes localized in punctae that are uniformly distributed around the apicolateral plasma membrane (Figure 2A). Similar disruption of the asymmetric distribution of other polarity proteins was also observed (data not shown). The effects of dco mutations on trichome placement were also analyzed. Prehairs normally emerge from the distal edge of pupal wing cells [15]. In dco mutant tissue, however, trichome inititation is delayed, and the prehairs often emerge from the cell center (Figure 2B). The disruption of asymmetric localization of polarity proteins and the resulting delay in trichome initiation and defects in trichome placement are all phenotypes characteristic of mutations in the core polarity genes. Therefore, we suppose that dco acts by regulating the phosphorylation state of one or more of the core polarity proteins and by modulating their ability to localize asymmetrically. Dco Localizes to the Apicolateral Plasma Membrane but Is Not Asymmetrically Distributed As dco regulates the asymmetric localization of the core polarity genes, we wondered whether Dco itself was asymmetrically distributed. A Dco-EGFP transgene was generated, which could be expressed in clones under control of the ubiquitous Actin promoter. This transgene is functional in flies, rescuing the polarity and patterning defects of viable dco2/dco5B2.6 adults, and also rescuing lethal allele combinations to viability with no residual polarity defects (Figure S2). In pupal wings, Dco-EGFP is enriched at the apicolateral plasma membrane, with a significant proportion
also in the cytoplasm (Figures 2C and 2D). However, there is no evidence of asymmetric localization to either the proximal or distal cell edges. Furthermore, we find that the subcellular distribution of Dco-EGFP is not significantly altered in fmi mutants (Figure 2E). fmi function is required for apical recruitment of all the other core polarity proteins [16–18], so this suggests that the apicolateral plasma membrane association of Dco-EGFP is independent of core polarity gene function. Thus, Dco localizes to a subcellular domain where it can interact with the core polarity genes but is not itself asymmetrically distributed. However, this does not preclude Dco activity being asymmetrically regulated. dco Is Required for Phosphorylation of Dsh In Vivo CKI has been found to bind to and phosphorylate Dishevelled in a variety of tissue culture and in vitro assays [3–5, 19–22]. Furthermore, Dsh recruitment to the plasma membrane by Fz has been correlated with an increase in Dsh phosphorylation [17, 23]. We therefore thought it likely that the role of dco in planar polarity could be to phosphorylate Dsh. To investigate this, the phosphorylation status of Dsh in wild-type and mutant pupal wings was analyzed. In SDS-PAGE, wild-type Dsh migrates as two distinct bands: the upper band is due to phosphorylation, as shown by the fact that it is lost when samples are treated with phosphatases (data not shown; [24]). Furthermore, Dsh phosphorylation is dependent on fz activity (compare middle and last lanes of Figure 3A [23]). Significantly, the phosphorylation of Dsh is also substantially reduced in pupal
Current Biology 1332
Figure 3. Phosphorylation of Dsh by Dco In Vivo (A) Western blot of proteins from 28 hr pupal wings of the indicated genotype, probed with Dsh antibody (top) and Actin (bottom) as a loading control. (B) Western blot of proteins from 28 hr pupal wings from whsFLP; tub-GAL4/UAS>STOP>dcoK38R pupal wings that were heatshocked to induce expression at the indicated times. (C) Domain structure of Dsh and putative phosphorylation sites. The position of the DIX (Dishevelled, Axin), PDZ, and DEP (Dishevelled, Egl-10, Pleckstrin) domains are indicated, together with a basic region b. The serine/threonine-rich region between the basic domain and the PDZ domain is shown for fly Dsh and mouse Dvl1, the mutated residues in DshST5 are in red, and the additional residues mutated in DshST8 are in blue. These serine/threonine residues are conserved throughout the animal kingdom, but residue 235 is alanine in many species, and thus unlikely to be important for Dsh function. (D and E) Dorsal surfaces of a dsh1 mutant wing (affects planar polarity but not canonical Wg signaling) (D), or a dsh3 mutant wing (null for canonical Wg signaling and planar polarity) in the presence of DshST5-GFP (E). The lack of rescue previously reported [25] may be due to the different expression systems. (F) 28 hr APF DshST5-GFP pupal wing, stained with GFP antibody. (G) Western blot of wild-type Dsh (left) or DshST5 (right) transfected into HEK293 cells with or without Fz.
wings from dco2/dco5B2.6 hypomorphs (lanes 2 and 4, Figure 3A) or in flies expressing dominant-negative forms of dco during pupal development (Figure 3B). Given that Dsh can be phosphorylated by CKI in vitro, these results indicate that Dsh is the probable in vivo target of dco in planar polarity signaling. A fragment of Dsh containing the basic region and the PDZ domain has been shown to be phosphorylated by CKI [3, 5, 22]. Furthermore, mutation of a cluster of six serines/threonines in this region to alanine (residues 235–242) was reported to result in a form of Dsh that was unable to rescue the polarity phenotype of dsh1 animals [25]. These serine/threonines all correspond to CKI consensus phosphorylation sites (S/Tp-X-(X)-S/T, requiring priming phosphorylation by another kinase, [26]) and could thus be in vivo target sites for Dco. To investigate this further, we expressed a similar transgene (DshST5-GFP), in which all five of the conserved serine/threonine residues in residues 236–242 (red in Figure 3C) are mutated, under control of the endogenous dsh promoter [23]. Despite blocking any potential CKI phosphorylation at these sites, this transgene was able to rescue null dsh3 mutants to viability. Furthermore, both the ommatidial and wing hair polarity phenotypes of dsh1 and dsh3 mutants were fully rescued (Figure 3E and not shown). Consistent with this rescue, DshST5-GFP is able to localize asymmetrically in the pupal wing (Figure 3F), and this mutant form of Dsh is also hyperphosphorylated when coexpressed with Fz in tissue-culture cells (Figure 3G). These results suggest that contrary to previous reports, serine/threonine residues 236–242 of Dsh do not play an essential role in planar polarity or canonical Wnt signaling. Furthermore, a form of Dsh with eight serine/threonine residues mutated to alanine (DshST8GFP, red and blue in Figure 3C) also rescues dsh3 mutants to viability and rescues the polarity phenotypes of dsh1 mutants over most of the wing, although some regions of abnormal polarity remain (Figure 3H). While this again shows that these residues are not essential for Dsh function, this could suggest that they have a partial role in Dsh phosphorylation; however, mutation of multiple amino acids could also disrupt Dsh structure or stability. The failure to identify specific CKI phosphorylation sites that are essential for Dsh function could be explained by a number of mechanisms. First, Dco might not act by direct phosphorylation of Dsh, and the effects on planar polarity could represent a kinase-independent function of Dco. However, this is unlikely; overexpression of a kinase-dead mutant (dcoK38R) has a loss-offunction planar polarity phenotype in the wing, suggesting that the kinase activity is required (Figure 3I, compare to Figure 4C). Alternatively, Dco might act indirectly to mediate Dsh phosphorylation by acting on another substrate. However, we think it is more likely that Dsh is the direct target of Dco but that the phosphorylation
(H) Dorsal surface of a dsh1 mutant wing in the presence of DshST8GFP. Note rescue is complete below vein 4, but not between veins 3 and 4. (I) Dorsal surface of a wing expressing ptc-GAL4, UAS-dcoK38R. Wing hairs point toward the AP boundary of the wing.
Casein Kinase I3 in Planar Polarity 1333
Figure 4. Partial Redundancy between dco and CKIa (A) Wild-type wing. (B) ptc-GAL4, pWIZ-dco wing, raised at 29ºC. Loss of wing material between veins 3 and 4 (arrow) masks any polarity phenotype. (C and D) Dorsal surfaces of ptc-GAL4, pWIZ-dco (C) or ptc-GAL4, pWIZ-dco, pWIZ-CKIa (D) wings, raised at 25ºC. (E) Quantitation of ptc-GAL phenotypes. The number of wings with polarity swirls extending along the indicated length of the wing is plotted, and significance p < 0.001. (F and G) Wings discs containing dcoj3B9 mutant clones, marked by loss of lacZ (red), stained for the Wg target genes Senseless (F) and Distalless (G) in green.
sites are redundant. For example, Dsh could be capable of being phosphorylated on multiple sites, and mutation of only a subset of these sites might not compromise its function. CKIa Cooperates with dco in Planar Polarity Signaling In Xenopus, overexpression of any CKI isoform can activate Wg signaling [22], and it is unclear which CKI isoform is responsible for Dsh phosphorylation in vivo, or if they act redundantly [1]. Dominant-negative forms of CKI3 block the ability of Wnts to induce ectopic axes in Xenopus [3]; however, this may not be specific. Furthermore, similar experiments with dsRNAi against CKI3 and CKIa in tissue culture have given conflicting results [20, 27]. We examined whether any other CKI isoforms could be involved in phosphorylating Dsh during planar polarity signaling, focusing our studies on CKIa, which has recently been shown to cooperate with dco in regulating Cubitus interruptus (Ci) phosphorylation during Hedgehog signaling [28]. No mutants for CKIa are available, so we made use of inducible RNAi constructs specific to dco and CKIa. Expressing dsRNA against dco under control of the ptc-GAL4 driver at high temperatures resulted in a dramatic narrowing between veins 3 and 4 (Figures 4A
and 4B): this could be due to a failure in cell proliferation or an effect on phosphorylation of Smoothened [29–31]. At lower temperatures, wing hairs between veins 3 and 4 are mispolarized, such that in the proximal region of the wing where ptc-GAL4 expression is highest, hairs tend to point toward the anterior-posterior boundary (Figure 4C). This phenotype is similar to the effects of expressing dsRNA against fz (R. Bastock and D.S., data not shown): in both cases, hairs point from regions of high activity to low activity, consistent with dco acting positively in planar polarity. Interestingly, expression of a CKIa dsRNA alone gives no phenotype, but when both CKIa and dco dsRNAs are coexpressed, the polarity phenotype is dramatically enhanced, such that polarity swirls are also seen in more distal regions where ptc-GAL4 expression is lower (Figure 4D). Quantitation of the phenotypes reveals that dco dsRNA flies exhibit polarity swirls extending only 4% of the distance between the posterior cross vein and the distal tip of the wing, while in the double dsRNAi flies, the phenotype extends to 17% of the wing length (Figure 4E). Our results therefore show that CKIa can at least partially substitute for dco function in planar polarity signaling. Nevertheless, dco cannot be fully redundant with CKIa, because dco mutations alone cause reduced Dsh phosphorylation and polarity defects.
Current Biology 1334
dco Mutants Do Not Affect Canonical Wg Signaling In addition to their roles in planar polarity, Fz and Dsh are required for canonical Wg signaling. In the absence of Wg ligand, the transcriptional coactivator b-catenin/ Armadillo (Arm) is constitutively phosphorylated and targeted for degradation. Binding of Wg to its coreceptors Frizzled (Fz) and LRP (LDL receptor-related protein) results in recruitment of Dsh to the receptor complex. This ultimately leads to inhibition of Arm phosphorylation/degradation, allowing it to enter the nucleus and activate transcription [32]. As phosphorylation of Dsh by CKI has been suggested to positively activate Wg signaling in vertebrates, we examined the role of dco during canonical Wg signaling in flies. To do this, we generated clones of hypomorphic dco alleles that do affect planar polarity, in the wing imaginal disc. However, the expression of both high- and lowthreshold Wg target genes is unaffected in these clones (Figures 4F and 4G). Furthermore, Wg target gene expression is also unaltered in clones of cells expressing dsRNA against dco (data not shown). We also attempted to investigate Wg signaling in embryos by making dco germline clones; however, no embryos were produced, presumably due to a requirement for dco function in the germline. These results argue that dco function is less essential in this context than in polarity signaling. It is possible that this is due to the clones only partially lacking dco activity. However, we think it more likely that dco is redundant with one or more other CKI isoforms in canonical Wg signaling. As CKIa cooperates with dco in phosphorylating Dsh in polarity signaling, it is probable that CKIa is also a redundant kinase in canonical signaling. Notably, CKIa also negatively regulates canonical signaling by phosphorylating and thus stabilizing b-catenin/Arm [19, 33–35], precluding a simple analysis of its regulation of Dsh. Conclusions The membrane recruitment of Dsh during planar polarity signaling has previously been correlated with its phosphorylation [17, 23]; however, the significance of this was unclear. We have now demonstrated that this phosphorylation is likely to be due to membrane-recruited Dsh coming into proximity with Dco/CKI3. Furthermore, in the absence of this phosphorylation, Dsh is unable to adopt its proper asymmetric localization, and planar polarity patterning is disrupted. This positive regulation of Dsh by CKI3 phosphorylation contrasts with previous reports of CKI3 negatively regulating noncanonical Wnt signaling and acting as a switch to canonical signaling [5, 14]. The asymmetric proximal-distal distribution of core polarity proteins such as Dsh is thought to be the result of an initial asymmetry in protein activity induced by an upstream cue, which is then amplified by interactions between the core polarity proteins across the proximal-distal cell boundaries [18, 23, 36]. Our results suggest that phosphorylation of Dsh by Dco is essential for Dsh to participate in this stabilization of the asymmetric distribution of the core polarity proteins. Dco is not itself asymmetrically distributed within pupal wing cells, although it is seen at the apicolateral plasma membrane where it would be able to interact
with Dsh. CKI isoforms are usually thought to be constitutively active, and their ability to phosphorylate their target proteins is regulated by subcellular localization and proximity to the substrate [1, 37]. Hence, Dco may be a permissive cue in planar polarity signaling, with recruitment of Dsh to the apicolateral plasma membrane by Fz [8] bringing Dco and Dsh into proximity and permitting the constitutive phosophorylation of Dsh. Alternatively, despite the lack of asymmetric protein localization, Dco protein activity could be asymmetrically regulated such that it preferentially phosphorylates Dsh at the distal end of the cell and thus is instructive in polarity signaling. Such activation could be mediated by upstream polarity genes or other core polarity genes such as Fz. The mechanism by which this could occur is unclear, but some CKI isoforms have been shown to be regulated by inhibitory autophosphorylation of their C termini [1]. Experimental Procedures Fly Strains and Histology Mutant alleles are as described in FlyBase and mitotic clones were generated by the FLP/FRT system [38]. Pupal wings were dissected at 28 hr after prepupa formation (APF) or 32 hr APF for trichomes, as previously described [18]. Antibodies used were Fmi mouse monoclonal [39], DE-cadherin rat monoclonal [40], GFP affinity purified rabbit serum (Abcam), lacZ mouse monoclonal (Promega), Senseless guinea pig [41], and Distalless rabbit [42]. Actin was visualized with Phalloidin Texas red (Molecular Probes). The Dsh antibody is a rat serum directed against amino acids 480–623 of Dsh. DNA Constructs dco cDNA was provided by BDGP (clone number LD27173), and CKIa cDNA was made by RT-PCR from embryos. Dco-EGFP is a fusion at the C-terminal end of the full-length Dco open reading frame to the N-terminal end of EGFP. The dominant-negative variant dcoK38R has lysine 38 mutated to arginine. These variants were cloned into a pCasper4 transformation vector containing an FRT>STOP>FRT sequence downstream of the Actin5C promoter [18] for inducible ubiquitous expression. For inducible high-level expression, the pUAST vector [43] was modified by introducing an FRT>STOP>FRT sequence downstream of the promoter. dsRNAi constructs were made by inserting nucleotides 859–1258 of the coding sequence of CKIa and nucleotides 634–1033 of the coding sequence of dco into the modified pUAST vector pWIZ [44]. These sequences are outside of the conserved kinase domain and are not predicted to interfere with any other of the eight fly homologs of CKI. Dsh-GFP phosphorylation site mutants were made by mutation of the serine/threonine residues of amino acids 236–242 or 236–247 to alanine. They were inserted into a pCasper4 vector containing 2.9 kb of genomic sequences upstream of the dsh gene, which was tagged at the 30 end with EGFP [23]. For expression in tissue-culture cells, the dsh cDNA was cloned into pcDNA3.1+ (Invitrogen) and the mutated amino acids substituted. Western Blots Pupal wings were dissected either directly into sample buffer, or an extract was made in lysis buffer (50 mM Tris-HCl [pH 7.5], 150 mM NaCl, 0.5% Triton X-100, protease inhibitors [Roche], and 1 mM phosphatase inhibitor microcystin). Identical results were observed with both methods. Proteins were detected by Western blotting against single pupal wing equivalents, via an affinity-purified rabbit antibody generated against amino acids 480–623 of Dsh, or Actin AC-40 mouse monoclonal (Sigma). Supplemental Data Two Supplemental Figures can be found with this article online at http://www.current-biology.com/cgi/content/full/16/13/1329/DC1/.
Casein Kinase I3 in Planar Polarity 1335
Acknowledgments We thank BDGP for ESTs; Mike Young for fly strains; Tadashi Uemura, Hugo Bellen, and Sean Carroll for antibodies; and Jeff Axelrod and Richard Carthew for plasmids. Vickie Thomas-McArthur is thanked for generating transformants, Katrina Hofstra and Melanie Stapleton for molecular biology assistance, and Martin Zeidler for comments on the manuscript. Confocal facilities were provided by Yorkshire Cancer Research, and the work was funded by the Wellcome Trust and MRC. D.S. is a Wellcome Trust Senior Fellow in Basic Biomedical Science. Received: March 14, 2006 Revised: April 7, 2006 Accepted: April 10, 2006 Published: July 10, 2006
References 1. Knippschild, U., Gocht, A., Wolff, S., Huber, N., Lohler, J., and Stoter, M. (2005). The casein kinase 1 family: participation in multiple cellular processes in eukaryotes. Cell. Signal. 17, 675– 689. 2. Price, M.A. (2006). CKI, there’s more than one: casein kinase I family members in Wnt/Wingless and Hedgehog signaling. Genes Dev. 20, 399–410. 3. Peters, J.M., McKay, R.M., McKay, J.P., and Graff, J.M. (1999). Casein kinase I transduces Wnt signals. Nature 401, 345–350. 4. Sakanaka, C., Leong, P., Xu, L., Harrison, S.D., and Williams, L.T. (1999). Casein kinase I epsilon in the Wnt pathway: regulation of beta-catenin function. Proc. Natl. Acad. Sci. USA 96, 12548– 12552. 5. Cong, F., Schweizer, L., and Varmus, H. (2004). Casein kinase I epsilon modulates the signaling specificities of Dishevelled. Mol. Cell. Biol. 24, 2000–2011. 6. Eaton, S. (2003). Cell biology of planar polarity transmission in the Drosophila wing. Mech. Dev. 120, 1257–1264. 7. Strutt, D. (2003). Frizzled signalling and cell polarisation in Drosophila and vertebrates. Development 130, 4501–4513. 8. Axelrod, J.D., Miller, J.R., Shulman, J.M., Moon, R.T., and Perrimon, N. (1998). Differential recruitment of Dishevelled provides signaling specificity in the planar cell polarity and Wingless signaling pathways. Genes Dev. 12, 2610–2622. 9. Boutros, M., Paricio, N., Strutt, D.I., and Mlodzik, M. (1998). Dishevelled activates JNK and discriminates between JNK pathways in planar polarity and wingless signaling. Cell 94, 109–118. 10. Strutt, H., and Strutt, D. (2003). EGF signaling and ommatidial rotation in the Drosophila eye. Curr. Biol. 13, 1451–1457. 11. Kloss, B., Price, J.L., Saez, L., Blau, J., Rothenfluh, A., Wesley, C.S., and Young, M.W. (1998). The Drosophila clock gene double-time encodes a protein closely related to human Casein kinase I epsilon. Cell 94, 97–107. 12. Zilian, O., Burke, R., Brentrup, D., Gutjahr, T., Bryant, P., and Noll, M. (1999). double-time is identical to discs overgrown, which is required for cell survival, proliferation and growth arrest in Drosophila imaginal discs. Development 126, 5409–5420. 13. Longenecker, K.L., Roach, P.J., and Hurley, T.D. (1996). Threedimensional structure of mammalian casein kinase I: molecular basis for phosphate recognition. J. Mol. Biol. 257, 618–631. 14. McKay, R.M., Peters, J.M., and Graff, J.M. (2001). The casein kinase I family: roles in morphogenesis. Dev. Biol. 235, 378–387. 15. Wong, L.L., and Adler, P.N. (1993). Tissue polarity genes of Drosophila regulate the subcellular location for prehair initiation in pupal wing cells. J. Cell Biol. 123, 209–221. 16. Bastock, R., Strutt, H., and Strutt, D. (2003). Strabismus is asymmetrically localised and binds to Prickle and Dishevelled during Drosophila planar polarity patterning. Development 130, 3007– 3014. 17. Shimada, Y., Usui, T., Yanagawa, S., Takeichi, M., and Uemura, T. (2001). Asymmetric co-localisation of Flamingo, a seven-pass transmembrane cadherin, and Dishevelled in planar cell polarisation. Curr. Biol. 11, 859–863.
18. Strutt, D.I. (2001). Asymmetric localisation of Frizzled and the establishment of cell polarity in the Drosophila wing. Mol. Cell 7, 367–375. 19. Gao, Z.-H., Seeling, J.M., Hill, V., Yochum, A., and Virshup, D.M. (2002). Casein kinase I phosphorylates and destabilises the beta-catenin degradation complex. Proc. Natl. Acad. Sci. USA 99, 1182–1187. 20. Hino, S., Michiue, T., Asashima, M., and Kikuchi, A. (2003). Casein kinase I epsilon enhances the binding of Dvl-1 to Frat-1 and is essential for Wnt-3a-induced accumulation of beta-catenin. J. Biol. Chem. 278, 14066–14073. 21. Kishida, M., Hino, S., Michiue, T., Yamamoto, H., Kishida, S., Fukui, A., Asashima, M., and Kikuchi, A. (2001). Synergistic activation of the Wnt signaling pathway by Dvl and casein kinase I epsilon. J. Biol. Chem. 276, 33147–33155. 22. McKay, R.M., Peters, J.M., and Graff, J.M. (2001). The casein kinase I family in Wnt signaling. Dev. Biol. 235, 388–396. 23. Axelrod, J.D. (2001). Unipolar membrane association of Dishevelled mediates Frizzled planar cell polarity signalling. Genes Dev. 15, 1182–1187. 24. Yanagawa, S.-i., van Leeuwen, F., Wodarz, A., Klingensmith, J., and Nusse, R. (1995). The Dishevelled protein is modified by Wingless signalling in Drosophila. Genes Dev. 9, 1087–1097. 25. Penton, A., Wodarz, A., and Nusse, R. (2002). A mutational analysis of dishevelled in Drosophila defines novel domains in the Dishevelled protein as well as novel suppressing alleles of axin. Genetics 161, 747–762. 26. Flotow, H., Graves, P.R., Wang, A.Q., Fiol, C.J., Roeske, R.W., and Roach, P.J. (1990). Phosphate groups as substrate determinants for casein kinase I action. J. Biol. Chem. 265, 14264– 14269. 27. Matsubayashi, H., Sese, S., Lee, J.-S., Shirakawa, T., Iwatsubo, T., Tomita, T., and Yanagawa, S. (2004). Biochemical characterisation of the Drosophila Wingless signaling pathway based on RNA interference. Mol. Cell. Biol. 24, 2012–2024. 28. Jia, J., Zhang, L., Zhang, Q., Tong, C., Wang, B., Hou, F., Amanai, K., and Jiang, J. (2005). Phosphorylation by Double-Time/ CKIepsilon and CKIalpha targets cubitus interruptus for Slimb/ beta-TRCP-mediated proteolytic processing. Dev. Cell 9, 819– 830. 29. Apionishev, S., Katanayeva, N.M., Marks, S.A., Kalderon, D., and Tomlinson, A. (2005). Drosophila Smoothened phosphorylation sites essential for Hedgehog signal transduction. Nat. Cell Biol. 7, 86–92. 30. Jia, J., Tong, C., Wang, B., Luo, L., and Jiang, J. (2004). Hedgehog signalling activity of Smoothened requires phosphorylation by protein kinase A and casein kinase I. Nature 432, 1045–1050. 31. Zhang, C., Williams, E.H., Guo, Y., Lum, L., and Beachy, P.A. (2004). Extensive phosphorylation of Smoothened in Hedgehog pathway activation. Proc. Natl. Acad. Sci. USA 101, 17900– 17907. 32. Logan, C.Y., and Nusse, R. (2004). The Wnt signaling pathway in development and disease. Annu. Rev. Cell Dev. Biol. 20, 781–810. 33. Amit, S., Hatzubai, A., Birman, Y., Anderson, J.S., Ben-Shushan, E., Mann, M., Ben-Neriah, Y., and Alkalay, I. (2002). Axin-mediated CKI phosphorylation of beta-catenin at Ser45: a molecular switch for the Wnt pathway. Genes Dev. 16, 1066–1076. 34. Liu, C., Li, Y., Semenov, M., Han, C., Baeg, G.-H., Tan, Y., Zhang, Z., Lin, X., and He, X. (2002). Control of beta-catenin phosphorylation/degradation by a dual-kinase mechanism. Cell 108, 837– 847. 35. Yanagawa, S., Matsuda, Y., Lee, J.-S., Matsubayashi, H., Sese, S., Kadowaki, T., and Ishimoto, A. (2002). Casein kinase I phosphorylates the Armadillo protein and induces its degradation in Drosophila. EMBO J. 21, 1733–1742. 36. Tree, D.R.P., Shulman, J.M., Rousset, R., Scott, M.P., Gubb, D., and Axelrod, J.D. (2002). Prickle mediates feedback amplification to generate asymmetric planar cell polarity signalling. Cell 109, 371–381. 37. Gross, S.D., and Anderson, R.A. (1998). Casein kinase I: spatial organisation and positioning of a multifunctional protein kinase family. Cell. Signal. 10, 699–711.
Current Biology 1336
38. Xu, T., and Rubin, G.M. (1993). Analysis of genetic mosaics in developing and adult Drosophila tissues. Development 117, 1223–1237. 39. Usui, T., Shima, Y., Shimada, Y., Hirano, S., Burgess, R.W., Schwarz, T.L., Takeichi, M., and Uemura, T. (1999). Flamingo, a seven-pass transmembrane cadherin, regulates planar cell polarity under the control of Frizzled. Cell 98, 585–595. 40. Oda, H., Uemura, T., Harada, Y., Iwai, Y., and Takeichi, M. (1994). A Drosophila homolog of cadherin associated with Armadillo and essential for embryonic cell-cell adhesion. Dev. Biol. 165, 716–726. 41. Nolo, R., Abbott, L.A., and Bellen, H.J. (2000). Senseless, a Zn finger transcription factor, is necessary and sufficient for sensory organ development in Drosophila. Cell 102, 349–362. 42. Panganiban, G., Sebring, A., Nagy, L., and Carroll, S. (1995). The development of crustacean limbs and the evolution of arthropods. Science 270, 1363–1366. 43. Brand, A.H., and Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415. 44. Lee, Y.S., and Carthew, R.W. (2003). Making a better RNAi vector for Drosophila: use of intron spacers. Methods 30, 322–329.