Constitutively active Protein kinase D acts as negative regulator of the Slingshot-phosphatase in Drosophila

Size: px

Start display at page:

Download "Constitutively active Protein kinase D acts as negative regulator of the Slingshot-phosphatase in Drosophila"

Kevin Ross
5 years ago
Views:

1 Hereditas 147: (2010) Constitutively active Protein kinase D acts as negative regulator of the Slingshot-phosphatase in Drosophila ANJA C. NAGEL, JENS SCHMID, JASMIN S. AUER, ANETTE PREISS and DIETER MAIER Institut für Genetik (240), Universität Hohenheim, Stuttgart, Germany Nagel, A. C., Schmid, J., Auer, J. S., Preiss, A. and Maier, D Constitutively active Protein kinase D acts as negative regulator of the Slingshot-phosphatase in Drosophila. Hereditas 147 : Lund, Sweden. eissn Received 16 July Accepted 27 August The mammalian protein kinase D family is involved in manifold cellular processes including cell migration and motility. Recently it was shown that human PKD1 and PKD2 phosphorylate and thereby inhibit Slingshot 1 Like (SSH1L), a phosphatase which is central to the regulation of actin cytoskeletal dynamics. We noted before that the overexpression of a constitutively active form of Drosophila PKD (PKD-SE) affects the fly retina and the resultant phenotypes suggest underlying defects in the actin cytoskeleton. Drosophila Slingshot, however, does not possess the phosphorylation site known to be targeted in SSH1L by human PKD1. Here we show that Drosophila PKD, despite this lack of conservation, nevertheless negatively regulates Slingshot. Overexpression of the active PKD-SE protein causes cellular defects that are similar to those of slingshot mutants. These include aberrant bristle morphology and positioning of photoreceptor nuclei. Interestingly, the observed nuclear mispositioning is due to a disturbance of the cytoskeleton rather than the epithelial organization. In accordance, overexpression of PKD-SE results in an accumulation of filamentous actin. This enrichment is modified by changes in slingshot gene doses, in line with an antagonistic relationship between PKD and slingshot. We conclude that similar to mammals, Drosophila PKD is a negative regulator of Ssh, with the premise of a different target phosphorylation site in Ssh. Dieter Maier, Institut f ü r Genetik (240), Universit ä t Hohenheim, Garbenstr. 30, DE Stuttgart, Germany. dieter. maier@uni-hohenheim.de The actin cytoskeleton is a highly dynamic structure that lies at the basis of cell shape and motility. There are several well documented examples that changes in cell shape provide the driving force for morphogenesis. These include the morphogenetic movements during gastrulation and the dorsal closure of the Drosophila embryo. Regular structures like the microvilli stacks within rhabdomeres of the fly retina are built up with the help of actin filaments. Moreover, actin filament based structures like bristles or fine hairs of the fly, which serve as mechano-sensors, can be extremely elaborate and stable structures (reviewed by JACINTO and B AUM 2003). The principle players that regulate actin dynamics have been identified in the past (reviewed by POLLARD and BORISY 2003). In particular, the actin depolymerising factor Cofilin promotes filament disassembly and creates new ends for elongation, thus setting the direction of cell movement (overview by DESMARAIS et al. 2005). Cofilin is inhibited by two families of kinases (LIMK1, 2 and TESK1, 2), which phosphorylate Cofilin on serine 3, thereby abolishing its actin-binding property. Notably the family of Slingshot (Ssh) phosphatases reactivates phospho-cofilin by dephosphorylation ( NIWA et al. 2002; overview by DESMARAIS et al. 2005). In human cells Slingshot 1 Like (SSH1L) phosphatase activity is dependent on actin binding ( NAGATA-OHASHI et al. 2004). However, SSH1L phosphorylated at Ser-937 and Ser-978 is bound by scaffolding proteins at these residues, thereby sequestering it from the actin/cofilin complex in the cytoplasm ( NAGATA- OHASHI et al. 2004). Recently it was shown that this C- terminal phosphorylation of SSH1L is dependent on protein kinase D 1 and 2 (PKD1, PKD2) ( EISELER et al. 2009; P ETERBURS et al. 2009). Hence, human PKD1 and PKD2 act as a negative regulator on the reactivation of the Slingshot- Cofilin signaling network which regulates actin dynamics in the cell. Albeit very similar in structure and function, mammalian SSHL phosphatases differ from Drosophila Slingshot (Ssh) in their C-terminal end ( NIWA et al. 2002; OHTA et al. 2003). Hence, Drosophila Ssh does not contain the sequence recognition site that was identified in SSH1L as target of human PKD. However, we have noted before that manipulation of Drosophila PKD activity affected the development of different organs in a way that conforms to a role of PKD in the regulation of actin dynamics in Drosophila as well ( MAIER et al. 2006, 2007). In this work we set out to determine the mode of action of Drosophila PKD on actin dynamics. We find that overexpression of a constitutively active PKD isoform, PKD-SE causes phenotypes very similar to those seen in ssh mutants. These include aberrant bristle morphology and positioning of photoreceptor nuclei. The latter phenotype is due to a disturbance of the actin cytoskeleton rather than a general defect in epithelial architecture. At the subcellular level, we note a strong enrichment of 2010 The Authors. This is an Open Access article. DOI: /j x

2 238 A. C. Nagel et al. Hereditas 147 (2010) filamentous actin as a consequence of either loss of ssh or gain of PKD-SE. Moreover we show, that there is an antagonistic genetic relationship between ssh and PKD. In summary, Drosophila PKD may act as a negative regulator of Ssh in the fly as well, suggesting that it targets a different phosphorylation site in Ssh than the ones identified earlier in human SSH1L. MATERIAL AND METHODS Fly stocks and clonal analysis The following fly stocks were used: UAS-PKD-SE GFP ( MAIER et al. 2007); FRT82B ssh 1-63 / TM3 Ser; UASssh.N 30, UAS-ssh.CS 11 (NIWA et al. 2002); y w, FRT82B ubi-gfp /TM6B; y w hs-flp; act CD2 Gal4, UAS- GFPnls / TM6B ( STRUHL and BASLER 1993; kindly obtained from K. Basler); UAS-lacZ, gmr-gal4; y 1 w 67c23. Flies were obtained from the Bloomington fly collection if not noted otherwise. Clones lacking ssh function were generated with the FLP/FRT system ( XU and RUBIN 1993) using flies of the genotype y w hs-flp; FRT82B ssh 1-63 / FRT82B Ubi-GFP. Homozygous mutant cell clones were recognized by the lack of GFP expression. Clones overexpressing PKD-SE- GFP were generated as described before ( STRUHL and B ASLER 1993; PROTZER et al. 2008) in animals of the genotype y w hs-flp, UAS-PKD-SE GFP / act CD2 Gal4, UAS-GFPnls. Overexpression clones were recognized by the expression of GFP. A given UAS-line was overexpressed in the developing eye by combining it with the gmr-gal4 driver line ( BRAND and PERRIMON 1993); UASlacZ served as control. For clonal analysis in larval or early pupal tissues, a single heat shock of 30 min was applied during second larval instar. To observe clones in adult animals, three heat shocks were given in 48 h intervals starting from late first instar larval stage. Animals were allowed to develop further into late larval or early pupal stage or to adulthood before tissues were dissected, stained by help of appropriate antibodies and mounted for microscopy. Immunohistochemistry, phenotypic analyses and imaging Larval and pupal tissues were dissected and stained with antibodies as described before ( MAIER et al. 1999, 2007). Monoclonal antibodies directed against Elav, Dlg and Arm, developed by G. M. Rubin, C. Goodman and E. Wieschaus, respectively, were obtained from Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by the Univ. of Iowa (Dept of Biological Sciences, Iowa City, IA 52242). Rhodamin coupled phalloidin was purchased from Invitrogen and secondary Cy3-coupled antibodies from Jackson Laboratory (Dianova, Hamburg). Fluorescently labelled tissues were mounted in Vectashield (Vector Laboratories) and analyzed on a Zeiss Axiophot linked to a Bio-Rad MRC1024 confocal microscope. Pictures were captured using BioRad Laser Sharp 3.1 software. Adult specimens were mounted in Hoyers medium (VAN DER MEER 1977) and analyzed as using a Zeiss Axioskop light microscope. Pictures were taken with a Pixera Camera (Optronics) and captured using the Viewfinder 2.0 software. Pictures were compiled using Corel Photo Paint and Corel Draw software. RESULTS Drosophila PKD-SE induces phenotypes comparable to the loss of Slingshot phosphatase We had noted before that overexpression of a constitutively active PKD isoform PKD-SE caused a variety of defects in the fly s retina that may have their origin in a disturbance of the actin cytoskeleton ( MAIER et al. 2007). In the course of our further examinations we observed a peculiar feature within the developing retina of larvae overexpressing PKD- SE: the nuclei of the photoreceptor cells that are normally localized at the apical side of the cell were mislocalized to a more basal position. We knew from the literature that this phenotype also occurs in photoreceptor cells lacking the phosphatase Slingshot (Ssh) ( ROGERS et al. 2005). Indeed, clones of photoreceptor cells either lacking Ssh or overexpressing the active PKD-SE isoform are characterized by a more basal position of their nuclei in comparison to their wild type sister cells (Fig. 1a d). To further compare the effects of PKD-SE overexpression and loss of ssh, we induced clones throughout development and noted a number of phenotypes affecting primarily bristle formation: mutation of ssh caused a characteristic bending, thickening and splitting of the bristles (Fig. 1g h) compared to the wild type (Fig. 1e). The latter phenotype is name giving to the mutant and arises from the failure to form correct actin bundles ( NIWA et al. 2002). Overexpression of PKD-SE resulted in similar bristle forms, albeit never as extreme (Fig. 1f). Although such bristle defects may be caused in several ways, the similarity of phenotypes is in agreement with a negative regulation of Ssh by PKD. In addition, PKD-SE clones showed conspicuous dark patches that were not seen in ssh mutant clones. The dark patches may result from secretion defects as a consequence of PKD-SE overexpression (Fig. 1f; MAIER et al. 2007). Ectopic expression of PKD-SE and loss of ssh cause an enrichment of F-actin but do not affect the overall epithelial organization The similarity between overexpression of PKD-SE and ssh mutants with regard to nuclear positioning in

3 Hereditas 147 (2010) Drosophila PKD inhibits Ssh phosphatase 239 Fig. 1. Gain of PKD-SE and loss of slingshot cause similar phenotypes. (a d ) Apical (a, c) and basal (b, d) view of eye imaginal discs that were stained with anti-elav antibodies (red) to label the nuclei of the photoreceptor cells. Cell clones were induced that either overexpress PKD-SE (marked by GFP, green in a, a and b, b ) or that lack Ssh (marked by the absence of green GFP in c, c and d, d ). Note that in either cell clone, the nuclei of the developing photoreceptor cells have a more basal position than the surrounding wild type cells. (e) A wild type bristle is composed of a socket (so) and a long shaft (sh) that is regular and slender with a slight curvature, ending in a fine tip. (f) Overexpession of PKD-SE in small cell clones induces bristle abnormalities. Frequently, the shaft is bent or curved more strongly than normal (small arrows). Bristle shafts appear uneven because the centre is thicker (open arrow). The bristles appear brittle (arrowhead) and are frequently absent. In addition, PKD-SE cell clones show dark cuticle patches (also on the wing, not shown), that appear to result from superfluous cuticle (large arrow). (g) A strong ssh mutant bristle phenotype is characterized by failure of building normal actin bundles resulting in bents of the bristle shaft (sh). (h) Cell clones lacking ssh show various bristle phenotypes including greater curvature (small arrows), thick and stumpy shafts (open arrow) and less often split shafts (asterisk). Moreover, the trichomes that are present on each cell are also affected and curved more strongly. In all panels, the size bar is 20μm.

4 240 A. C. Nagel et al. photoreceptor cells prompted us to compare more closely the effects of either mutation of the fly. The mislocalization of the photoreceptor nuclei could be a result of a disturbance of the cellular cytoskeleton like in ssh mutants or of the epithelial organization. We investigated this possibility further by comparing clones lacking ssh with those overexpressing PKD-SE in the larval retina. The cytoskeleton was monitored directly by staining for filamentous actin (F-actin). We noted extreme changes in ssh mutant cell clones that generally showed a strong disturbance of the cell shape. As expected, the ssh mutant cells were strongly enriched in F-actin which frequently accumulated along scars rather than within the entire clone (Fig. 2b b ). The scars may arise from a contraction of the apical cell surface as a consequence of Ssh loss and also from cell death of mutant cells (NIWA et al. 2002; ROGERS et al. 2005). The epithelial organization was analyzed by monitoring several proteins that mark different apical-basal positions within the cell including Armadillo (Arm) that is present at the zonula adherens and Disc large (Dlg) that marks the septate junctions (MÜLLER and BOSSINGER 2003). Typically, the markers appeared enriched in the mutant clones, however, were all present at the expected position (Fig. 2d d, 2f f ). Overexpression of PKD-SE in cell clones had similar effects, that were however considerably Hereditas 147 (2010) weaker (Fig. 2a a, 2c c, 2e e ). Notably, F-actin was strongly enriched, whereas the epithelial organization remained intact. We conclude that activation of PKD influences the dynamics of the actin cytoskeleton in a similar way as does the loss of Ssh. Hence PKD might function as a negative regulator of Ssh. Antagonistic relationship between PKD-SE and ssh In order to address the genetic relationship between PKD and ssh, we made use of the F-actin accumulation in PKDSE overexpressing cells and asked, whether it was dependent on ssh gene dosis. To combine PKD-SE overexpression with either gain or loss of ssh gene activity, we made use of the Gal4/UAS system that allows a tissue specific overexpression of the desired genes (BRAND and PERRIMON 1993). We utilized a dominant negative form of Ssh (ssh. CS; NIWA et al. 2002) and a wild type form of Ssh (ssh.n; NIWA et al. 2002) to generate a loss and a gain of function situation, respectively. Indeed, we observed a strong accumulation of F-actin in the retina of pupae in which PKDSE was overexpressed. F-actin accumulation was extremely strong in the homozygous animals compared to the heterozygotes and the controls (Fig. 3a c). A combined overexpression of the dominant negative ssh.cs together with PKD-SE resulted in an enhanced F-actin Fig. 2. Sub-cellular effects of PKD-SE overexpression match those of Ssh loss. Cell clones were induced in eye imaginal discs. Clones overexpressing PKD-SE are marked by GFP (green; a, a, c, c, e, e ), whereas clones lacking ssh are marked by the absence of GFP (b, b, d, d, f, f ). The discs were stained with either rhodamine-coupled phalloidin to visualize F-actin (a b ), or antibodies directed against Armadillo (Arm) to mark the zonula adherence (c d ) or against Disc large (Dlg) to mark the septate junctions (e f ). The markers are shown in red. (a f) shows the double staining; a f the single markers (red) and a f the clones (green and absence of green, respectively; clones are outlined). Size bar is 20μm.

Hereditas 147 (2010) Drosophila PKD inhibits Ssh phosphatase 241 enrichment (Fig. 3g). Since ssh.cs overexpression alone did not elicit this effect (Fig.

5 Hereditas 147 (2010) Drosophila PKD inhibits Ssh phosphatase 241 enrichment (Fig. 3g). Since ssh.cs overexpression alone did not elicit this effect (Fig. 3f), we conclude that PKD and Ssh act antagonistically. Accordingly, F-actin enrichment appeared reduced when the wild type ssh.n gene copy was overexpressed together with PKD-SE (Fig. 1d e). These results are indicative of a negative relationship between PKD and Ssh also in Drosophila. DISCUSSION Drosophila PKD-SE acts as negative regulator of Slingshot phosphatase In this study, we have used Drosophila as an in vivo model system to investigate the role of PKD in the regulation of actin dynamics. Our results suggest that a constitutively active Drosophila PKD acts as negative regulator Slingshot phosphatase. Expression of PKD-SE results in the accumulation of F-actin, thus affecting the cytoskeleton. The observed phenotypes include defects in cell morphology, e.g. positioning of photoreceptor nuclei or rhabdomere formation in the ommatidia ( MAIER et al. 2007). Malformation of bristles and hairs point to a major role of PKD in the formation of parallel actin-filament bundles, resembling mutations in ssh ( NIWA et al. 2002). However, the phenotypes that arise from overactivation of Drosophila PKD are always much weaker than those induced by the loss of Ssh activity. This is not unexpected since it is already known that several players regulate Ssh activity apart from PKD. One example is PAK4 (p21 activated kinase 4), which acts as a dual regulator in activating LIMK1 and inhibiting SSH1L by phosphorylation ( SOOSAIRAJAH et al. 2005). Drosophila dpak has been shown to activate LIMK-1, acting as an essential regulator of actin dynamics as well ( CONDER et al. 2004; NG and L UO 2004). Drosophila Ssh harbors several presumptive PAK-consensus sites ( RENNEFAHRT et al. 2007). Hence, it is conceivable that dpak may phosphorylate and thereby inactivate Ssh in Drosophila like in mammals. Fig. 3. PKD-SE induces F-actin accumulation that is modified by Ssh activity. Upper panels, apical view and lower panels, basal view of a pupal retina stained with rhodaminecoupled Phalloidin to visualize filamentous actin. Note strong enrichment of F-actin upon overexpression of active PKD-SE (a, a wild type vs b, b one and c, c two copies of PKD-SE overexpression construct, respectively). Concomitant overexpression of the dominant negative Ssh form (ssh.cs) slightly increases F-actin accumulation (g, g ) compared with the overexpression of ssh-cs alone (f, f ). In contrast, overexpression of the wild type Ssh (ssh.n), which on its own has no effect (d, d ), decreased the F-actin accumulation by PKD-SE to more normal levels (e, e ). Size bar is 20 μm. The following genotypes are shown: (a a ) gmr-gal4 UAS-lacZ control retina. (b b ) gmr-gal4 UAS-PKD-SE / ; UAS-lacZ /. (c c ) gmr-gal4 UAS-PKD-SE / gmr-gal4 UAS-PKD-SE. (d d ) gmr-gal4/ ; UAS-ssh.N /. (e e ) gmr-gal4 UAS-PKD-SE / ; UASssh.N /. (f f ) gmr-gal4/ ; UAS-ssh.CS/. (g g ) gmr-gal4 UAS-PKD-SE / ; UAS-ssh.CS/. Molecular differences between Drosophila Slingshot and human Slingshot like proteins SSH1L, SSH2L and SSH3L From studies in mammals it is known that Slingshot like proteins (SSH1L, SSH2L) bind to F-actin with the C-terminal domain, which is a prerequisite for the activation of the phosphatase ( OHTA et al. 2003; NAGATA-OHASHI et al. 2004; SOOSAIRAJAH et al. 2005). Hence, the C-terminal domain acts auto-inhibitory to SSH activity. In SSH1L the C-terminal domain contains two Serine residues (Ser 937 and Ser 978 ) that are phosphorylated by human PKD1 and PKD2 ( EISELER et al. 2009; PETERBURS et al. 2009). Phosphorylation of Ser 937 and Ser 978 triggers the binding of scaffolding proteins, and the sequestration of SSH1L to the cytoplasm, thereby limiting the amount of actin-bound active SSH1L ( NAGATA-OHASHI et al. 2004). Interestingly, these phosphorylation sites are neither conserved in SSH3L nor in the Drosophila Ssh protein ( OHTA et al. 2003). Instead, there is a potential PKD target sequence that is located next to the phosphatase catalytic centre and that is well conserved in human SSH1L, SSH2L and Drosophila Slingshot proteins. This raises the possibility of an alternative mechanism of Slingshot inhibition by PKD kinases: phosphorylation of Slingshot at the catalytic centre may directly interfere with its enzymatic activity. It is tempting to speculate that such an inhibitory mechanism is crucial

6 242 A. C. Nagel et al. Hereditas 147 (2010) to the regulation of Slingshot and it may be conserved from fly to human. Acknowledgements We are indebted to I. Wech for excellent technical help. We acknowledge K. Basler, the Bloomington stock centre, the DSHB and DGRC for fly stocks, monoclonal antibodies and cdna clones, respectively. This work was supported by the German Research Foundation (DFG) through a grant to DM (MA 1328/8-1). REFERENCES Brand, A. H. and Perrimon, N Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118: Conder, R., Yu, H., Ricos, M. et al dpak is required for integrity of the leading edge cytoskeleton during Drosophila dorsal closure but does not signal through the JNK cascade. Dev. Biol. 276: DesMarais, V., Ghosh, M., Eddy, R. et al Cofilin takes the lead. J. Cell Sci. 118: Eiseler, T., D ö ppler, H., Yan, I. K. et al Protein kinase D1 regulates cofilin-mediated F-actin reorganization and cell motility through slingshot. Nat. Cell Biol. 11: Jacinto, A. and Baum, B Actin in development. Mech. Dev. 120: Maier, D., Nagel, A. C., Johannes, B. et al Subcellular localization of Hairless protein shows a major focus of activity within the nucleus. Mech. Dev. 89: Maier, D., Hausser, A., Nagel, A. C. et al Drosophila protein kinase D is broadly expressed and a fraction localizes to the Golgi compartment. Gene Expr. Patterns 6: Maier, D., Nagel, A. C., Gloc, H. et al Protein kinase D regulates several aspects of development in Drosophila melanogaster. BMC Dev. Biol. 25; 7:74. M ü ller, H.-A. and Bossinger, O Molecular networks controlling epithelial cell polarity in development. Mech. Dev. 120: Nagata-Ohashi, K., Ohta, Y, Goto, K. et al A pathway of neuregulin-induced activation of cofilin-phosphatase Slingshot and cofilin in lamellipodia. J. Cell. Biol. 165: Ng, J. and Luo, L Rho GTPases regulate axon growth through convergent and divergent signalling. Neuron 44: Niwa, R., Nagata-Ohashi, K., Takeichi, M. et al Control of actin reorganization by Slingshot, a family of phosphatases that dephosphorylate ADF/cofilin. Cell 108: Ohta, Y., Kousaka, K., Nagata-Ohashi, K. et al Differential activities, subcellular distribution and tissue expression patterns of three members of Slingshot family phosphatases that dephosphorylate cofilin. Genes Cells 8: Peterburs, O., Heering, J., Link, G. et al Protein kinase D (PKD) regulates cell migration by direct phosphorylation of the cofilin phosphatase Slinghot 1 like (SSH1L). Cancer Res. 14: Pollard, T. D. and Borisy, G. G Cellular motility driven by assembly and disassembly of actin filaments. Cell 112: Protzer, C. E., Wech, I. and Nagel, A. C Hairless induces cell death by downregulation of EGFR signalling activity. J. Cell Sci. 121: Rennefahrt, U. E., Deacon, S. W., Parker, S. A. et al Specificity profiling of PAK kinases allows identification of novel phosphorylation sites. J. Biol. Chem. 282: Rogers, E. M., Hsiung, F., Rodrigues, A. B. et al Slingshot cofilin phosphatase localization is regulated by Receptor Tyrosine Kinases and cytoskeletal structure in the developing Drosophila eye. Mech. Dev. 122: Soosairajah, J., Maiti, S., Wiggan, O N. et al Interplay between components of a novel LIM kinase slingshot phosphatase complex regulates cofilin. EMBO J. 24: Struhl, G. and Basler, K Organizing activity of wingless protein in Drosophila. Cell 72: van der Meer, J Optical clean and permanent whole mount preparations for phase contrast microscopy of cuticular structures of insect larvae. Drosophila Inf. Serv. 52: 160. Xu, T. and Rubin, G. M Analysis of genetic mosaics in the developing and adult Drosophila tissues. Development 117:

Baz, Par-6 and apkc are not required for axon or dendrite specification in Drosophila

Baz, Par-6 and apkc are not required for axon or dendrite specification in Drosophila Melissa M. Rolls and Chris Q. Doe, Inst. Neurosci and Inst. Mol. Biol., HHMI, Univ. Oregon, Eugene, Oregon 97403 Correspondence