- Email: [email protected]
Drosophila Anterior-Posterior Polarity Requires Actin-Dependent PAR-1 Recruitment to the Oocyte Posterior
Current Biology 16, 1090–1095, June 6, 2006 ª2006 Elsevier Ltd All rights reserved
DOI 10.1016/j.cub.2006.04.001
Report Drosophila Anterior-Posterior Polarity Requires Actin-Dependent PAR-1 Recruitment to the Oocyte Posterior He´le`ne Doerflinger,1 Richard Benton,1,2 Isabel L. Torres,1 Maarten F. Zwart,1 and Daniel St Johnston1,* 1 The Gurdon Institute and the Department of Genetics University of Cambridge Tennis Court Road Cambridge CB2 1QN United Kingdom
Summary The Drosophila anterior-posterior axis is established at stage 7 of oogenesis when the posterior follicle cells signal to polarize the oocyte microtubule cytoskeleton. This requires the conserved PAR-1 kinase, which can be detected at the posterior of the oocyte in immunostainings from stage 9. However, this localization depends on Oskar localization, which requires the earlier PAR-1-dependent microtubule reorganization, indicating that Oskar-associated PAR-1 cannot establish oocyte polarity. Here we analyze the function of the different PAR-1 isoforms and find that only PAR-1 N1 isoforms can completely rescue the oocyte polarity phenotype. Furthermore, PAR-1 N1 is recruited to the posterior cortex of the oocyte at stage 7 in response to the polarizing follicle cell signal, and this requires actin, but not microtubules. This suggests that posterior PAR-1 N1 polarizes the microtubule cytoskeleton. PAR-1 N1 localization is mediated by a cortical targeting domain and a conserved anteriorlateral exclusion signal in its C-terminal linker domain. PAR-1 is also required for the polarization of the C. elegans zygote and is recruited to the posterior cortex in an actin-dependent manner. Our results therefore identify a molecular parallel between axis formation in Drosophila and C. elegans and make Drosophila PAR-1 N1 the earliest known marker for the polarization of the oocyte. Results and Discussion The anterior-posterior axis of Drosophila becomes polarized at stage 7 of oogenesis, when the posterior follicle cells send an unknown signal that induces an anterior-to-posterior array of microtubules in the oocyte, which directs the localization of the anterior and posterior determinants bicoid and oskar mRNAs to opposite poles of the oocyte [1]. The conserved PAR-1 kinase plays a key role in this process, since microtubules are nucleated all around the cortex of par-1 mutant oocytes and their plus ends are abnormally focused in the
*Correspondence: [email protected] 2 Present address: Laboratory of Neurogenetics and Behavior, The Rockefeller University, 1230 York Avenue, Box 63, New York, New York 10021.
middle, leading to the mislocalization of oskar mRNA [2, 3]. Immunostaining with PAR-1 antibodies indicates that it localizes to the posterior of the oocyte, but only at stage 9, which is too late to play a role in the reorganization of microtubules. Furthermore, this localization depends on the posterior localization of oskar mRNA, which in turn requires the PAR-1-dependent repolarization of the microtubules [2]. Thus, the localized PAR-1 detected by antibodies cannot be responsible for the reorganization of the microtubule cytoskeleton. The par-1 gene encodes multiple isoforms that belong to three families (N1, N2, and N3) that differ in their N-terminal domains, raising the possibility that an isoform that is not detected in antibody stainings polarizes the microtubule cytoskeleton (Figure 1A). To test this possibility, we generated transgenic flies expressing N-terminally tagged GFP fusions of the major PAR-1 isoforms under the control of the maternally expressed a4-tubulin promoter (see the Supplemental Experimental Procedures in the Supplemental Data available with this article online). We first analyzed their ability to rescue the polarity defect of a viable transheterozygous combination of par-1 alleles, in which oskar mRNA is partially or completely mislocalized to the center of the oocyte in 99% of stage 9–10 egg chambers (Figures 1B–1D). Expression of either GFP-PAR-1 N1S or N1L fully rescues the posterior localization of oskar mRNA, whereas GFP-PAR-1 N2S and GFP-PAR-1 N3L rescue only partially (30% and 40% mislocalization, respectively) (Figure 1E). PAR-1 N1S is the major isoform expressed during oogenesis and is strongly reduced in the par-1 hypomorphs that give a polarity phenotype [2, 3]. These results suggest that PAR-1 N1S plays the principal role in the repolarization of the microtubule cytoskeleton, although the N2 and N3 isoforms can partially compensate for a partial loss of N1 when overexpressed. PAR-1 N1 Isoforms Localize to the Oocyte Posterior before the Microtubule Reorganization To examine whether the PAR-1 N1 isoforms show a polarized distribution, we imaged GFP fluorescence in ovaries expressing GFP-PAR1 N1S and N1L and observed that both are enriched at the posterior of the oocyte (Figures 2A–2C and data not shown). This localization differs from that of the PAR-1 detected in immunostainings in three respects. First, the posterior enrichment is much broader than the crescent of pole plasm with which PAR-1 associates. Second, PAR-1 N1 is more closely associated with the oocyte cortex than the pole plasm, as demonstrated by the lack of overlap between GFP-PAR-1 N1S and Staufen, which colocalizes with oskar mRNA at the posterior (Figures 2D and 2D0 ) [4–6]. Furthermore, in contrast to the PAR-1 detected by antibody staining, GFP-PAR-1 N1S localization is not affected by staufen mutants, which abolish the posterior localization of oskar mRNA, nor is it recruited to the anterior by the ectopic pole plasm
Actin-Dependent PAR-1 Localization 1091
Figure 1. PAR-1 N1 Isoforms Can Fully Rescue the par-1 Polarity Phenotype (A) The three families of par-1 differ mainly in their N-terminal region (exons 1 and 2) and by the presence or absence of exons 9 and 14. Within each family, short isoforms have a C terminus encoded by exons 16 to 18, whereas long isoforms have a longer C terminus containing the conserved PAR-1 domain encoded by exons 19 to 21. (B) Fluorescent in situ hybridization to oskar mRNA (pink) in a wild-type stage 10 egg chamber, showing its localization to the posterior pole of the oocyte. (C and D) In the par-16323/par-1W3 mutants, oskar mRNA is partially (C) or completely (D) mislocalized to the middle of the oocyte. (E) The rescue of the oskar mRNA localization phenotype of par-16323/par-1W3 females expressing different GFP-PAR-1 isoforms. The GFP-PAR1 N1 isoforms rescue completely, whereas GFP-PAR-1 N2S and GFP-PAR-1 N3L give only partial rescue.
nucleated by osk-bcd 30 UTR RNA (data not shown) [7]. Third, GFP-PAR-1 N1S first becomes enriched at the posterior before the nucleus migrates from the posterior
to the anterior of the oocyte at stage 7, which is the first sign of the anterior-posterior polarization of the oocyte (Figure 2A). The localization of both GFP-PAR-1 N1
Current Biology 1092
Figure 2. GFP-PAR-1 N1S Localizes to the Posterior of the Oocyte from Stage 7 (A) GFP-PAR-1 N1S localization in an early stage 7 egg chamber. A weak crescent is visible at the posterior of the oocyte, although the oocyte nucleus has not yet migrated to the anterior (the nucleus appears as the dark circle adjacent to the posterior cortex). (B) GFP-PAR-1 N1S localization in a stage 9 egg chamber. PAR-1 N1S localizes strongly to the posterior cortex of the oocyte and around the cortex of the nurse cells but is excluded from the anterior and lateral cortex of the oocyte. (C and D) GFP-PAR-1 N1S ([C], green in [D]) forms a broader posterior crescent than Staufen (red), which marks the pole plasm (D), and is more closely associated with the oocyte cortex (D0 ).
isoforms at the posterior cortex therefore precedes the repolarization of the oocyte microtubule cytoskeleton and makes them the earliest known markers for the posterior of the oocyte. Posterior Recruitment of GFP-PAR-1 N1 Depends on the Polarizing Signal and Actin To test whether PAR-1 N1 is recruited to the posterior cortex in response to the polarizing signal from the posterior follicle cells, we analyzed its localization in gurken mutants, in which the posterior follicle cells adopt an anterior fate [8, 9]. GFP-PAR-1 N1S fails to become enriched at the posterior cortex in a strong gurken mutant combination, grk2B6/grk2E12 (Figures 3A and 3B). Furthermore, a similar loss of the posterior crescent is seen in mago nashi mutants, which block the transduction of this signal in the oocyte (Figure 3C) [10, 11]. These results indicate that the posterior recruitment of the PAR-1 N1 isoforms depends on the polarizing signal from the follicle cells. Interestingly, GFP-PAR-1 N1S usually accumulates in a single large dot close to the oocyte nucleus in mago nashi mutants (Figure 3C, arrow). This suggests that Mago nashi is required for the trafficking of PAR-1 to the cortex and that PAR-1 becomes sequestered on some internal compartment or organelle in its absence. To investigate the cytoskeletal requirements for the posterior localization of PAR1, we fed flies with drugs that depolymerize either the microtubule or actin cytoskeletons. The microtubule-destablizing drug colcemid has no effect on the posterior enrichment of GFP-PAR1 N1S, although this treatment depolymerizes the microtubules, since it abolishes all other markers for oocyte polarity, such as the anterior positioning of the nucleus (Figure 3D) [12]. In contrast, similar treatments with the actin-destabilizing drug latrunculin-A abolish the posterior enrichment of GFP-PAR-1 N1S, which becomes uniformly distributed in the oocyte cytoplasm and cortex
(Figure 3E). Therefore, the posterior enrichment of both PAR-1 N1S and N1L precedes the reorganization of the oocyte microtubule cytoskeleton and is microtubule independent, consistent with the idea that they represent the PAR-1 isoforms that are responsible for the repolarization of the microtubule cytoskeleton. The Linker Domain Drives the Cortical and Posterior Localization of PAR-1 N1S In contrast to the PAR-1 N1 isoforms, neither GFP-PAR1 N2 nor N3 become enriched at the posterior of the oocyte, which may account for their failure to fully rescue the par-1 polarity phenotype. All isoforms localize to the ring canals and along the cortex of the nurse cells, but PAR-1 N2S does not enter the oocyte, whereas PAR-1 N3L is concentrated all around the oocyte cortex (Figures 4A–4D). PAR-1 N2S differs from the other isoforms by its unique N-terminal domain and by the absence of exon 9, which encodes the five amino acids AETLR (Figure 1A), and either of these differences could prevent its entry into the oocyte. Since both PAR-1 N1S and N1L localize to the posterior of the oocyte and rescue the par-1 polarity phenotype, the highly conserved PAR-1 domain at the C terminus of PAR-1 N1L that is absent in N1S is not necessary for this localization (Figure 1A). To map the posterior localization signal of PAR-1 N1 isoforms, we generated three GFP-tagged transgenic lines expressing either the N1 domain, the N1 and kinase domains, or the linker domain alone (Figures 5A–5C and data not shown). The first two constructs become uniformly distributed in the nurse cell and oocyte cytoplasm, whereas the linker domain localizes in a posterior crescent like the full-length protein (Figures 5F–5H and data not shown). The linker domain is therefore necessary and sufficient to localize PAR-1 to the posterior cortex. Although the linker domain is only weakly conserved between PAR-1 homologs, two short regions of the
Actin-Dependent PAR-1 Localization 1093
Figure 3. Posterior Localization of GFP-PAR-1 N1S Requires the Polarizing Signal from the Follicle Cells and the Actin Cytoskeleton The localization of GFP-PAR-1 N1S in stage 9 oocytes: (A) wild-type; (B) gurken2B6/ gurken2E12; (C) mago1/Df(2R)F36; (D) wildtype treated with the microtubule-destabilizing drug colcemid; (E) wild-type treated with the actin-destabilizing drug Latrunculin-A. Note the accumulation of GFP-PAR-1 N1S in a spherical region near the nucleus in mago nashi mutant oocytes (arrow in [C]).
linker show a much higher degree of homology: a UBA domain near the N terminus of the linker and a stretch of 17 amino acids near the C terminus (Figure 5A and Figure S1). Replacement of five conserved amino acids in the UBA domain with alanines has no effect on the posterior enrichment of GFP-PAR-1 N1S (Figure S2). In contrast, mutation of five conserved amino acids at either the left or the right end of the second conserved motif leads to high levels of GFP-PAR-1 N1S localization all around the oocyte cortex, without any specific enrichment at the posterior (Figures 5D and 5I). Thus, this region serves as an anterior and lateral exclusion motif (AEM). Since mutants in both the UBA domain and the AEM still localize cortically, some other region of the PAR-1 linker must function as a cortical localization signal. Deletion of the C-terminal 37 amino acids from PAR1 N1S abolishes cortical localization and leads to a uniform distribution in the cytoplasm (Figures 5E and 5J). Recruitment to the cortex therefore requires the region immediately after the AEM of the linker. The mapping of the PAR-1 N1 localization signals also explains why GFP-PAR-1 N3L does not become enriched at the
posterior cortex. The cDNA used for this construct is a splice variant that lacks exon 14, which encodes 38 amino acids, including the AEM (Figure 5P). This form of PAR-1 therefore contains the C-terminal cortical localization signal, but not the anterior and lateral exclusion signal. Interestingly, both constructs with mutations in the AEM cause a penetrant oocyte polarity phenotype, as shown by the mispositioning of the oocyte nucleus in more than half of the egg chambers (Figure 5I, arrow). This is presumably a consequence of ectopic PAR-1 activity at the anterior and lateral cortex, demonstrating that the normal localization of PAR-1 to the posterior is essential for the polarization of the oocyte. PAR-1 is also required for apical-basal polarity in epithelial cells, where it localizes to the lateral membrane [2, 13–15]. As previously reported, the linker domain of PAR-1 is responsible for its targeting to the lateral membrane (Figures 5K–5M) [16]. Furthermore, mutants in the AEM lead to ectopic apical localization, whereas the deletion of the C-terminal 37 amino acids blocks all localization to the cortex (Figures 5N and 5O). Thus, the same
Figure 4. Localization of Different PAR-1 Isoforms in the Oocyte (A) GFP-PAR-1 N1S; (B) GFP-PAR-1 N1L; (C) GFP-PAR-1 N2S; (D) GFP-PAR-1 N3L. All GFP-PAR-1 isoforms localize to the ring canals and along the cortex of the nurse cells. However, PAR-1 N2S (C) does not accumulate in the oocyte, and PAR-1 N3L (D) is concentrated all around the oocyte cortex, whereas PAR-1 N1S and PAR-1 N1L localize to the posterior cortex.
Current Biology 1094
Figure 5. Posterior Localization of GFP-PAR N1S Is Mediated by an Anterior Exclusion Signal and a Cortical Targeting Signal in the Linker Domain (A–E) The domain structure of the GFP-tagged constructs used in this study: (A) full-length PAR-1 N1S, showing the unique N-terminal N1 domain (green), the kinase domain (gray), and the linker domain (orange), which contains a conserved UBA domain (dark blue) and a conserved AEM (red); (B) the linker domain alone; (C) PAR-1 N1S without the linker domain; (D) full-length PAR-1 N1S with mutations in the AEM; and (E) linker domain lacking the C-terminal 37 amino acids (linkerDC). (F–J) Localization of the GFP-tagged PAR-1 N1S constructs in the oocyte. Full-length PAR1 N1S (F) and the linker domain alone (G) become enriched at the posterior cortex, whereas the N1 and kinase domain are cytoplasmic (H). Mutation of the AEM leads to localization of PAR-1 N1S around the entire oocyte cortex (I), causing a mispositioning of the oocyte nucleus (arrow), while cortical localization is abolished by the deletion of the C terminus of the linker (J). (K–O) Localization of GFP-tagged PAR-1 N1S constructs in the follicular epithelium. Fulllength PAR-1 N1S (K) and the linker domain (L) localize to the lateral membrane, whereas the PAR-1N1S lacking the linker domain is cytoplasmic (M). Mutation of the AEM leads to ectopic localization of PAR-1 N1S at the apical membrane (N), and deletion of the C-terminal 41 amino acids prevents recruitment to the cortex (O). (P) Sequence of the C terminus of PAR-1 N1S and the corresponding region in PAR-1 N3L, showing the AEM motif (red) and the cortical targeting signal (yellow). PAR-1 N3L does not contain exon 14, encoding the AEM motif, which accounts for its lack of exclusion from the anterior and lateral cortex of the oocyte.
cis-acting signals mediate the lateral localization of PAR-1 in epithelial cells and its posterior localization in the oocyte, strongly suggesting that it is targeted by the same mechanism in each case. Conclusions In contrast to the other PAR-1 isoforms, PAR-1 N1S and N1L localize to the posterior of the oocyte at stage 7, as an early response to the polarizing signal from the follicle cells. This localization is microtubule independent and begins before the reorganization of the oocyte microtubule-cytoskeleton that defines the anteriorposterior axis. Furthermore, only the N1 isoforms fully rescue the anterior-posterior polarity defects of par-1 mutants, whereas mutant forms of PAR-1 that localize all around the cortex disrupt oocyte polarity. Thus, anterior-posterior polarity depends on the localization of PAR-1 N1 to the posterior cortex of the oocyte and its exclusion from other regions of the cortex. These results indicate that the actin-dependent posterior recruitment of PAR-1 N1 is essential for the reorganization of the oocyte cytoskeleton. One question raised by our results is why the localization of the PAR-1 N1 isoforms to the posterior cortex has not been detected by antibody staining, given that our antibody was raised against the linker domain and should recognize all forms of PAR-1 [2]. This may reflect
the higher sensitivity of GFP tagging compared to antibody staining and the low levels of endogenous protein at this site. It seems more likely, however, that the epitopes recognized by this antibody are masked at the cortex, since a new antibody against a different region of PAR-1 shows a very similar posterior cortical localization of endogenous PAR-1 to that seen with GFP-PAR-1 N1S (M. Bettencourt-Dias, personal communication). Conversely, none of our GFP-PAR-1 constructs localize to the pole plasm like the PAR-1 detected in immunostainings. Thus, either the fusion of GFP to the N termini of these isoforms inhibits their recruitment to the pole plasm, or this staining reflects the localization of another splice variant that has not yet been analyzed. The PAR-1 N1 isoforms are targeted to the posterior by a cortical localization signal at the C terminus of the protein and an adjacent antero-lateral exclusion motif (AEM), and these two signals are also necessary and sufficient to target PAR-1 to the lateral cortex in epithelial cells. Mammalian PAR-1b is restricted in a similar way to the lateral domain of polarized MDCK cells, and this is mediated by the phosphorylation of a conserved threonine in the AEM by apically localized atypical Protein kinase C (aPKC), which excludes PAR-1 from the apical cortex and inhibits its activity (Figure S2) [17, 18]. Drosophila aPKC shows a similar localization to the subapical region of epithelial cells, suggesting that
Actin-Dependent PAR-1 Localization 1095
apical exclusion by aPKC phosphorylation of the AEM domain may be a conserved mechanism for restricting PAR-1 to the lateral cortex. It is more difficult to account for the posterior localization of PAR-1 in the oocyte by this mechanism, however, since DaPKC null mutant germline clones block the specification of the oocyte in the germarium, but a proportion of the mutant egg chambers escape this early defect and go on to develop a normal anterior-posterior axis at stage 9. This suggests that aPKC is not required for the repolarization of the oocyte at stage 7, although it is possible that residual aPKC perdures in these escaper clones and can still function to restrict PAR-1 to the posterior. Alternatively, PAR-1 may be excluded from the anterior and lateral cortex by a different mechanism or by another kinase that phosphorylates the same site in the AEM. The polarization of the anterior-posterior axis in C. elegans also depends on the localization of PAR-1 to the posterior cortex of the one-cell zygote [19]. However, the upstream steps that led to the posterior recruitment of PAR-1 appeared to be quite different in the two systems, because the posterior localization of PAR-1 in C. elegans requires the actin cytoskeleton [20–23], whereas the polarization of the Drosophila oocyte is microtubule dependent. Our results indicate that the mechanisms are more similar than previously thought, since the posterior recruitment of the PAR-1 N1 isoforms is upstream of microtubule polarity and depends on actin. The localization of C. elegans PAR-1 requires the cortical myosin, NMY2, which associates directly with PAR-1 [20, 24]. It will therefore be important to determine whether a myosin is also required for the cortical recruitment of Drosophila PAR-1. Supplemental Data Supplemental Data include two figures and Supplemental Experimental Procedures and can be found with this article online at http:// www.current-biology.com/cgi/content/full/16/11/1090/DC1/. Acknowledgments This work was supported by the Wellcome Trust. We would like to thank Mo´nica Bettencourt-Dias for sharing her unpublished results and Sophie Martin, Vincent Mirouse, and Uwe Irion for help and advice. Received: December 16, 2005 Revised: March 30, 2006 Accepted: April 3, 2006 Published: June 5, 2006 References 1. Ray, R., and Schu¨pbach, T. (1996). Intercellular signaling and the polarization of body axes during Drosophila oogenesis. Genes Dev. 10, 1711–1723. 2. Shulman, J.M., Benton, R., and St Johnston, D. (2000). The Drosophila homolog of C. elegans PAR-1 organizes the oocyte cytoskeleton and directs oskar mRNA localisation to the posterior pole. Cell 101, 1–20. 3. Tomancak, P., Piano, F., Riechmann, V., Gunsalus, K.C., Kemphues, K.J., and Ephrussi, A. (2000). A Drosophila melanogaster homologue of Caenorhabditis elegans par-1 acts at an early step in embryonic-axis formation. Nat. Cell Biol. 2, 458–460. 4. Ephrussi, A., Dickinson, L.K., and Lehmann, R. (1991). oskar organizes the germ plasm and directs localization of the posterior determinant nanos. Cell 66, 37–50.
5. Kim-Ha, J., Smith, J.L., and Macdonald, P.M. (1991). oskar mRNA is localized to the posterior pole of the Drosophila oocyte. Cell 66, 23–35. 6. St Johnston, D., Beuchle, D., and Nu¨sslein-Volhard, C. (1991). Staufen, a gene required to localize maternal RNAs in the Drosophila egg. Cell 66, 51–63. 7. Ephrussi, A., and Lehmann, R. (1992). Induction of germ cell formation by oskar. Nature 358, 387–392. 8. Gonza´lez-Reyes, A., Elliott, H., and St Johnston, D. (1995). Polarization of both major body axes in Drosophila by gurkentorpedo signalling. Nature 375, 654–658. 9. Roth, S., Neuman-Silberberg, F.S., Barcelo, G., and Schu¨pbach, T. (1995). cornichon and the EGF receptor signaling process are necessary for both anterior-posterior and dorsal-ventral pattern formation in Drosophila. Cell 81, 967–978. 10. Micklem, D.R., Dasgupta, R., Elliott, H., Gergely, F., Davidson, C., Brand, A., Gonza´lez-Reyes, A., and St Johnston, D. (1997). The mago nashi gene is required for the polarisation of the oocyte and the formation of perpendicular axes in Drosophila. Curr. Biol. 7, 468–478. 11. Newmark, P.A., Mohr, S.A., Gong, L., and Boswell, R.E. (1997). mago nashi mediates the posterior follicle cell-to-oocyte signal to organize axis formation in Drosophila. Development 124, 3197–3207. 12. Theurkauf, W.E., Smiley, S., Wong, M.L., and Alberts, B.M. (1992). Reorganization of the cytoskeleton during Drosophila oogenesis: implications for axis specification and intercellular transport. Development 115, 923–936. 13. Bo¨hm, H., Brinkmann, V., Drab, M., Henske, A., and Kurzchalia, V. (1997). Mammalian homologues of C. elegans PAR-1 are asymmetrically localized in epithelial cells and may influence their polarity. Curr. Biol. 7, 603–606. 14. Cox, D.N., Lu, B., Sun, T., Williams, L.T., and Jan, Y.N. (2001a). Drosophila par-1 is required for oocyte differentiation and microtubule organization. Curr. Biol. 11, 75–87. 15. Doerflinger, H., Benton, R., Shulman, J.M., and St Johnston, D. (2003). The role of PAR-1 in regulating the polarised microtubule cytoskeleton in the Drosophila follicular epithelium. Development 130, 3965–3975. 16. Vaccari, T., Rabouille, C., and Ephrussi, A. (2005). The Drosophila PAR-1 spacer domain is required for lateral membrane association and for polarization of follicular epithelial cells. Curr. Biol. 15, 255–261. 17. Hurov, J.B., Watkins, J.L., and Piwnica-Worms, H. (2004). Atypical PKC phosphorylates PAR-1 kinases to regulate localization and activity. Curr. Biol. 14, 736–741. 18. Suzuki, A., Hirata, M., Kamimura, K., Maniwa, R., Yamanaka, T., Mizuno, K., Kishikawa, M., Hirose, H., Amano, Y., Izumi, N., et al. (2004). aPKC acts upstream of PAR-1b in both the establishment and maintenance of mammalian epithelial polarity. Curr. Biol. 14, 1425–1435. 19. Guo, S., and Kemphues, K.J. (1995). par-1, a gene required for establishing polarity in C. elegans embryos, encodes a putative Ser/ Thr kinase that is asymmetrically distributed. Cell 81, 611–620. 20. Guo, S., and Kemphues, K.J. (1996). A non-muscle myosin required for embryonic polarity in Caenorhabditis elegans. Nature 382, 455–458. 21. Cuenca, A.A., Schetter, A., Aceto, D., Kemphues, K., and Seydoux, G. (2003). Polarization of the C. elegans zygote proceeds via distinct establishment and maintenance phases. Development 130, 1255–1265. 22. Munro, E., Nance, J., and Priess, J.R. (2004). Cortical flows powered by asymmetrical contraction transport PAR proteins to establish and maintain anterior-posterior polarity in the early C. elegans embryo. Dev. Cell 7, 413–424. 23. Cheeks, R.J., Canman, J.C., Gabriel, W.N., Meyer, N., Strome, S., and Goldstein, B. (2004). C. elegans PAR proteins function by mobilizing and stabilizing asymmetrically localized protein complexes. Curr. Biol. 14, 851–862. 24. Shelton, C.A., Carter, J.C., Ellis, G.C., and Bowerman, B. (1999). The nonmuscle myosin regulatory light chain gene mlc-4 is required for cytokinesis, anterior-posterior polarity, and body morphology during Caenorhabditis elegans embryogenesis. J. Cell Biol. 146, 439–451.
DOI 10.1016/j.cub.2006.04.001
Report Drosophila Anterior-Posterior Polarity Requires Actin-Dependent PAR-1 Recruitment to the Oocyte Posterior He´le`ne Doerflinger,1 Richard Benton,1,2 Isabel L. Torres,1 Maarten F. Zwart,1 and Daniel St Johnston1,* 1 The Gurdon Institute and the Department of Genetics University of Cambridge Tennis Court Road Cambridge CB2 1QN United Kingdom
Summary The Drosophila anterior-posterior axis is established at stage 7 of oogenesis when the posterior follicle cells signal to polarize the oocyte microtubule cytoskeleton. This requires the conserved PAR-1 kinase, which can be detected at the posterior of the oocyte in immunostainings from stage 9. However, this localization depends on Oskar localization, which requires the earlier PAR-1-dependent microtubule reorganization, indicating that Oskar-associated PAR-1 cannot establish oocyte polarity. Here we analyze the function of the different PAR-1 isoforms and find that only PAR-1 N1 isoforms can completely rescue the oocyte polarity phenotype. Furthermore, PAR-1 N1 is recruited to the posterior cortex of the oocyte at stage 7 in response to the polarizing follicle cell signal, and this requires actin, but not microtubules. This suggests that posterior PAR-1 N1 polarizes the microtubule cytoskeleton. PAR-1 N1 localization is mediated by a cortical targeting domain and a conserved anteriorlateral exclusion signal in its C-terminal linker domain. PAR-1 is also required for the polarization of the C. elegans zygote and is recruited to the posterior cortex in an actin-dependent manner. Our results therefore identify a molecular parallel between axis formation in Drosophila and C. elegans and make Drosophila PAR-1 N1 the earliest known marker for the polarization of the oocyte. Results and Discussion The anterior-posterior axis of Drosophila becomes polarized at stage 7 of oogenesis, when the posterior follicle cells send an unknown signal that induces an anterior-to-posterior array of microtubules in the oocyte, which directs the localization of the anterior and posterior determinants bicoid and oskar mRNAs to opposite poles of the oocyte [1]. The conserved PAR-1 kinase plays a key role in this process, since microtubules are nucleated all around the cortex of par-1 mutant oocytes and their plus ends are abnormally focused in the
*Correspondence: [email protected] 2 Present address: Laboratory of Neurogenetics and Behavior, The Rockefeller University, 1230 York Avenue, Box 63, New York, New York 10021.
middle, leading to the mislocalization of oskar mRNA [2, 3]. Immunostaining with PAR-1 antibodies indicates that it localizes to the posterior of the oocyte, but only at stage 9, which is too late to play a role in the reorganization of microtubules. Furthermore, this localization depends on the posterior localization of oskar mRNA, which in turn requires the PAR-1-dependent repolarization of the microtubules [2]. Thus, the localized PAR-1 detected by antibodies cannot be responsible for the reorganization of the microtubule cytoskeleton. The par-1 gene encodes multiple isoforms that belong to three families (N1, N2, and N3) that differ in their N-terminal domains, raising the possibility that an isoform that is not detected in antibody stainings polarizes the microtubule cytoskeleton (Figure 1A). To test this possibility, we generated transgenic flies expressing N-terminally tagged GFP fusions of the major PAR-1 isoforms under the control of the maternally expressed a4-tubulin promoter (see the Supplemental Experimental Procedures in the Supplemental Data available with this article online). We first analyzed their ability to rescue the polarity defect of a viable transheterozygous combination of par-1 alleles, in which oskar mRNA is partially or completely mislocalized to the center of the oocyte in 99% of stage 9–10 egg chambers (Figures 1B–1D). Expression of either GFP-PAR-1 N1S or N1L fully rescues the posterior localization of oskar mRNA, whereas GFP-PAR-1 N2S and GFP-PAR-1 N3L rescue only partially (30% and 40% mislocalization, respectively) (Figure 1E). PAR-1 N1S is the major isoform expressed during oogenesis and is strongly reduced in the par-1 hypomorphs that give a polarity phenotype [2, 3]. These results suggest that PAR-1 N1S plays the principal role in the repolarization of the microtubule cytoskeleton, although the N2 and N3 isoforms can partially compensate for a partial loss of N1 when overexpressed. PAR-1 N1 Isoforms Localize to the Oocyte Posterior before the Microtubule Reorganization To examine whether the PAR-1 N1 isoforms show a polarized distribution, we imaged GFP fluorescence in ovaries expressing GFP-PAR1 N1S and N1L and observed that both are enriched at the posterior of the oocyte (Figures 2A–2C and data not shown). This localization differs from that of the PAR-1 detected in immunostainings in three respects. First, the posterior enrichment is much broader than the crescent of pole plasm with which PAR-1 associates. Second, PAR-1 N1 is more closely associated with the oocyte cortex than the pole plasm, as demonstrated by the lack of overlap between GFP-PAR-1 N1S and Staufen, which colocalizes with oskar mRNA at the posterior (Figures 2D and 2D0 ) [4–6]. Furthermore, in contrast to the PAR-1 detected by antibody staining, GFP-PAR-1 N1S localization is not affected by staufen mutants, which abolish the posterior localization of oskar mRNA, nor is it recruited to the anterior by the ectopic pole plasm
Actin-Dependent PAR-1 Localization 1091
Figure 1. PAR-1 N1 Isoforms Can Fully Rescue the par-1 Polarity Phenotype (A) The three families of par-1 differ mainly in their N-terminal region (exons 1 and 2) and by the presence or absence of exons 9 and 14. Within each family, short isoforms have a C terminus encoded by exons 16 to 18, whereas long isoforms have a longer C terminus containing the conserved PAR-1 domain encoded by exons 19 to 21. (B) Fluorescent in situ hybridization to oskar mRNA (pink) in a wild-type stage 10 egg chamber, showing its localization to the posterior pole of the oocyte. (C and D) In the par-16323/par-1W3 mutants, oskar mRNA is partially (C) or completely (D) mislocalized to the middle of the oocyte. (E) The rescue of the oskar mRNA localization phenotype of par-16323/par-1W3 females expressing different GFP-PAR-1 isoforms. The GFP-PAR1 N1 isoforms rescue completely, whereas GFP-PAR-1 N2S and GFP-PAR-1 N3L give only partial rescue.
nucleated by osk-bcd 30 UTR RNA (data not shown) [7]. Third, GFP-PAR-1 N1S first becomes enriched at the posterior before the nucleus migrates from the posterior
to the anterior of the oocyte at stage 7, which is the first sign of the anterior-posterior polarization of the oocyte (Figure 2A). The localization of both GFP-PAR-1 N1
Current Biology 1092
Figure 2. GFP-PAR-1 N1S Localizes to the Posterior of the Oocyte from Stage 7 (A) GFP-PAR-1 N1S localization in an early stage 7 egg chamber. A weak crescent is visible at the posterior of the oocyte, although the oocyte nucleus has not yet migrated to the anterior (the nucleus appears as the dark circle adjacent to the posterior cortex). (B) GFP-PAR-1 N1S localization in a stage 9 egg chamber. PAR-1 N1S localizes strongly to the posterior cortex of the oocyte and around the cortex of the nurse cells but is excluded from the anterior and lateral cortex of the oocyte. (C and D) GFP-PAR-1 N1S ([C], green in [D]) forms a broader posterior crescent than Staufen (red), which marks the pole plasm (D), and is more closely associated with the oocyte cortex (D0 ).
isoforms at the posterior cortex therefore precedes the repolarization of the oocyte microtubule cytoskeleton and makes them the earliest known markers for the posterior of the oocyte. Posterior Recruitment of GFP-PAR-1 N1 Depends on the Polarizing Signal and Actin To test whether PAR-1 N1 is recruited to the posterior cortex in response to the polarizing signal from the posterior follicle cells, we analyzed its localization in gurken mutants, in which the posterior follicle cells adopt an anterior fate [8, 9]. GFP-PAR-1 N1S fails to become enriched at the posterior cortex in a strong gurken mutant combination, grk2B6/grk2E12 (Figures 3A and 3B). Furthermore, a similar loss of the posterior crescent is seen in mago nashi mutants, which block the transduction of this signal in the oocyte (Figure 3C) [10, 11]. These results indicate that the posterior recruitment of the PAR-1 N1 isoforms depends on the polarizing signal from the follicle cells. Interestingly, GFP-PAR-1 N1S usually accumulates in a single large dot close to the oocyte nucleus in mago nashi mutants (Figure 3C, arrow). This suggests that Mago nashi is required for the trafficking of PAR-1 to the cortex and that PAR-1 becomes sequestered on some internal compartment or organelle in its absence. To investigate the cytoskeletal requirements for the posterior localization of PAR1, we fed flies with drugs that depolymerize either the microtubule or actin cytoskeletons. The microtubule-destablizing drug colcemid has no effect on the posterior enrichment of GFP-PAR1 N1S, although this treatment depolymerizes the microtubules, since it abolishes all other markers for oocyte polarity, such as the anterior positioning of the nucleus (Figure 3D) [12]. In contrast, similar treatments with the actin-destabilizing drug latrunculin-A abolish the posterior enrichment of GFP-PAR-1 N1S, which becomes uniformly distributed in the oocyte cytoplasm and cortex
(Figure 3E). Therefore, the posterior enrichment of both PAR-1 N1S and N1L precedes the reorganization of the oocyte microtubule cytoskeleton and is microtubule independent, consistent with the idea that they represent the PAR-1 isoforms that are responsible for the repolarization of the microtubule cytoskeleton. The Linker Domain Drives the Cortical and Posterior Localization of PAR-1 N1S In contrast to the PAR-1 N1 isoforms, neither GFP-PAR1 N2 nor N3 become enriched at the posterior of the oocyte, which may account for their failure to fully rescue the par-1 polarity phenotype. All isoforms localize to the ring canals and along the cortex of the nurse cells, but PAR-1 N2S does not enter the oocyte, whereas PAR-1 N3L is concentrated all around the oocyte cortex (Figures 4A–4D). PAR-1 N2S differs from the other isoforms by its unique N-terminal domain and by the absence of exon 9, which encodes the five amino acids AETLR (Figure 1A), and either of these differences could prevent its entry into the oocyte. Since both PAR-1 N1S and N1L localize to the posterior of the oocyte and rescue the par-1 polarity phenotype, the highly conserved PAR-1 domain at the C terminus of PAR-1 N1L that is absent in N1S is not necessary for this localization (Figure 1A). To map the posterior localization signal of PAR-1 N1 isoforms, we generated three GFP-tagged transgenic lines expressing either the N1 domain, the N1 and kinase domains, or the linker domain alone (Figures 5A–5C and data not shown). The first two constructs become uniformly distributed in the nurse cell and oocyte cytoplasm, whereas the linker domain localizes in a posterior crescent like the full-length protein (Figures 5F–5H and data not shown). The linker domain is therefore necessary and sufficient to localize PAR-1 to the posterior cortex. Although the linker domain is only weakly conserved between PAR-1 homologs, two short regions of the
Actin-Dependent PAR-1 Localization 1093
Figure 3. Posterior Localization of GFP-PAR-1 N1S Requires the Polarizing Signal from the Follicle Cells and the Actin Cytoskeleton The localization of GFP-PAR-1 N1S in stage 9 oocytes: (A) wild-type; (B) gurken2B6/ gurken2E12; (C) mago1/Df(2R)F36; (D) wildtype treated with the microtubule-destabilizing drug colcemid; (E) wild-type treated with the actin-destabilizing drug Latrunculin-A. Note the accumulation of GFP-PAR-1 N1S in a spherical region near the nucleus in mago nashi mutant oocytes (arrow in [C]).
linker show a much higher degree of homology: a UBA domain near the N terminus of the linker and a stretch of 17 amino acids near the C terminus (Figure 5A and Figure S1). Replacement of five conserved amino acids in the UBA domain with alanines has no effect on the posterior enrichment of GFP-PAR-1 N1S (Figure S2). In contrast, mutation of five conserved amino acids at either the left or the right end of the second conserved motif leads to high levels of GFP-PAR-1 N1S localization all around the oocyte cortex, without any specific enrichment at the posterior (Figures 5D and 5I). Thus, this region serves as an anterior and lateral exclusion motif (AEM). Since mutants in both the UBA domain and the AEM still localize cortically, some other region of the PAR-1 linker must function as a cortical localization signal. Deletion of the C-terminal 37 amino acids from PAR1 N1S abolishes cortical localization and leads to a uniform distribution in the cytoplasm (Figures 5E and 5J). Recruitment to the cortex therefore requires the region immediately after the AEM of the linker. The mapping of the PAR-1 N1 localization signals also explains why GFP-PAR-1 N3L does not become enriched at the
posterior cortex. The cDNA used for this construct is a splice variant that lacks exon 14, which encodes 38 amino acids, including the AEM (Figure 5P). This form of PAR-1 therefore contains the C-terminal cortical localization signal, but not the anterior and lateral exclusion signal. Interestingly, both constructs with mutations in the AEM cause a penetrant oocyte polarity phenotype, as shown by the mispositioning of the oocyte nucleus in more than half of the egg chambers (Figure 5I, arrow). This is presumably a consequence of ectopic PAR-1 activity at the anterior and lateral cortex, demonstrating that the normal localization of PAR-1 to the posterior is essential for the polarization of the oocyte. PAR-1 is also required for apical-basal polarity in epithelial cells, where it localizes to the lateral membrane [2, 13–15]. As previously reported, the linker domain of PAR-1 is responsible for its targeting to the lateral membrane (Figures 5K–5M) [16]. Furthermore, mutants in the AEM lead to ectopic apical localization, whereas the deletion of the C-terminal 37 amino acids blocks all localization to the cortex (Figures 5N and 5O). Thus, the same
Figure 4. Localization of Different PAR-1 Isoforms in the Oocyte (A) GFP-PAR-1 N1S; (B) GFP-PAR-1 N1L; (C) GFP-PAR-1 N2S; (D) GFP-PAR-1 N3L. All GFP-PAR-1 isoforms localize to the ring canals and along the cortex of the nurse cells. However, PAR-1 N2S (C) does not accumulate in the oocyte, and PAR-1 N3L (D) is concentrated all around the oocyte cortex, whereas PAR-1 N1S and PAR-1 N1L localize to the posterior cortex.
Current Biology 1094
Figure 5. Posterior Localization of GFP-PAR N1S Is Mediated by an Anterior Exclusion Signal and a Cortical Targeting Signal in the Linker Domain (A–E) The domain structure of the GFP-tagged constructs used in this study: (A) full-length PAR-1 N1S, showing the unique N-terminal N1 domain (green), the kinase domain (gray), and the linker domain (orange), which contains a conserved UBA domain (dark blue) and a conserved AEM (red); (B) the linker domain alone; (C) PAR-1 N1S without the linker domain; (D) full-length PAR-1 N1S with mutations in the AEM; and (E) linker domain lacking the C-terminal 37 amino acids (linkerDC). (F–J) Localization of the GFP-tagged PAR-1 N1S constructs in the oocyte. Full-length PAR1 N1S (F) and the linker domain alone (G) become enriched at the posterior cortex, whereas the N1 and kinase domain are cytoplasmic (H). Mutation of the AEM leads to localization of PAR-1 N1S around the entire oocyte cortex (I), causing a mispositioning of the oocyte nucleus (arrow), while cortical localization is abolished by the deletion of the C terminus of the linker (J). (K–O) Localization of GFP-tagged PAR-1 N1S constructs in the follicular epithelium. Fulllength PAR-1 N1S (K) and the linker domain (L) localize to the lateral membrane, whereas the PAR-1N1S lacking the linker domain is cytoplasmic (M). Mutation of the AEM leads to ectopic localization of PAR-1 N1S at the apical membrane (N), and deletion of the C-terminal 41 amino acids prevents recruitment to the cortex (O). (P) Sequence of the C terminus of PAR-1 N1S and the corresponding region in PAR-1 N3L, showing the AEM motif (red) and the cortical targeting signal (yellow). PAR-1 N3L does not contain exon 14, encoding the AEM motif, which accounts for its lack of exclusion from the anterior and lateral cortex of the oocyte.
cis-acting signals mediate the lateral localization of PAR-1 in epithelial cells and its posterior localization in the oocyte, strongly suggesting that it is targeted by the same mechanism in each case. Conclusions In contrast to the other PAR-1 isoforms, PAR-1 N1S and N1L localize to the posterior of the oocyte at stage 7, as an early response to the polarizing signal from the follicle cells. This localization is microtubule independent and begins before the reorganization of the oocyte microtubule-cytoskeleton that defines the anteriorposterior axis. Furthermore, only the N1 isoforms fully rescue the anterior-posterior polarity defects of par-1 mutants, whereas mutant forms of PAR-1 that localize all around the cortex disrupt oocyte polarity. Thus, anterior-posterior polarity depends on the localization of PAR-1 N1 to the posterior cortex of the oocyte and its exclusion from other regions of the cortex. These results indicate that the actin-dependent posterior recruitment of PAR-1 N1 is essential for the reorganization of the oocyte cytoskeleton. One question raised by our results is why the localization of the PAR-1 N1 isoforms to the posterior cortex has not been detected by antibody staining, given that our antibody was raised against the linker domain and should recognize all forms of PAR-1 [2]. This may reflect
the higher sensitivity of GFP tagging compared to antibody staining and the low levels of endogenous protein at this site. It seems more likely, however, that the epitopes recognized by this antibody are masked at the cortex, since a new antibody against a different region of PAR-1 shows a very similar posterior cortical localization of endogenous PAR-1 to that seen with GFP-PAR-1 N1S (M. Bettencourt-Dias, personal communication). Conversely, none of our GFP-PAR-1 constructs localize to the pole plasm like the PAR-1 detected in immunostainings. Thus, either the fusion of GFP to the N termini of these isoforms inhibits their recruitment to the pole plasm, or this staining reflects the localization of another splice variant that has not yet been analyzed. The PAR-1 N1 isoforms are targeted to the posterior by a cortical localization signal at the C terminus of the protein and an adjacent antero-lateral exclusion motif (AEM), and these two signals are also necessary and sufficient to target PAR-1 to the lateral cortex in epithelial cells. Mammalian PAR-1b is restricted in a similar way to the lateral domain of polarized MDCK cells, and this is mediated by the phosphorylation of a conserved threonine in the AEM by apically localized atypical Protein kinase C (aPKC), which excludes PAR-1 from the apical cortex and inhibits its activity (Figure S2) [17, 18]. Drosophila aPKC shows a similar localization to the subapical region of epithelial cells, suggesting that
Actin-Dependent PAR-1 Localization 1095
apical exclusion by aPKC phosphorylation of the AEM domain may be a conserved mechanism for restricting PAR-1 to the lateral cortex. It is more difficult to account for the posterior localization of PAR-1 in the oocyte by this mechanism, however, since DaPKC null mutant germline clones block the specification of the oocyte in the germarium, but a proportion of the mutant egg chambers escape this early defect and go on to develop a normal anterior-posterior axis at stage 9. This suggests that aPKC is not required for the repolarization of the oocyte at stage 7, although it is possible that residual aPKC perdures in these escaper clones and can still function to restrict PAR-1 to the posterior. Alternatively, PAR-1 may be excluded from the anterior and lateral cortex by a different mechanism or by another kinase that phosphorylates the same site in the AEM. The polarization of the anterior-posterior axis in C. elegans also depends on the localization of PAR-1 to the posterior cortex of the one-cell zygote [19]. However, the upstream steps that led to the posterior recruitment of PAR-1 appeared to be quite different in the two systems, because the posterior localization of PAR-1 in C. elegans requires the actin cytoskeleton [20–23], whereas the polarization of the Drosophila oocyte is microtubule dependent. Our results indicate that the mechanisms are more similar than previously thought, since the posterior recruitment of the PAR-1 N1 isoforms is upstream of microtubule polarity and depends on actin. The localization of C. elegans PAR-1 requires the cortical myosin, NMY2, which associates directly with PAR-1 [20, 24]. It will therefore be important to determine whether a myosin is also required for the cortical recruitment of Drosophila PAR-1. Supplemental Data Supplemental Data include two figures and Supplemental Experimental Procedures and can be found with this article online at http:// www.current-biology.com/cgi/content/full/16/11/1090/DC1/. Acknowledgments This work was supported by the Wellcome Trust. We would like to thank Mo´nica Bettencourt-Dias for sharing her unpublished results and Sophie Martin, Vincent Mirouse, and Uwe Irion for help and advice. Received: December 16, 2005 Revised: March 30, 2006 Accepted: April 3, 2006 Published: June 5, 2006 References 1. Ray, R., and Schu¨pbach, T. (1996). Intercellular signaling and the polarization of body axes during Drosophila oogenesis. Genes Dev. 10, 1711–1723. 2. Shulman, J.M., Benton, R., and St Johnston, D. (2000). The Drosophila homolog of C. elegans PAR-1 organizes the oocyte cytoskeleton and directs oskar mRNA localisation to the posterior pole. Cell 101, 1–20. 3. Tomancak, P., Piano, F., Riechmann, V., Gunsalus, K.C., Kemphues, K.J., and Ephrussi, A. (2000). A Drosophila melanogaster homologue of Caenorhabditis elegans par-1 acts at an early step in embryonic-axis formation. Nat. Cell Biol. 2, 458–460. 4. Ephrussi, A., Dickinson, L.K., and Lehmann, R. (1991). oskar organizes the germ plasm and directs localization of the posterior determinant nanos. Cell 66, 37–50.
5. Kim-Ha, J., Smith, J.L., and Macdonald, P.M. (1991). oskar mRNA is localized to the posterior pole of the Drosophila oocyte. Cell 66, 23–35. 6. St Johnston, D., Beuchle, D., and Nu¨sslein-Volhard, C. (1991). Staufen, a gene required to localize maternal RNAs in the Drosophila egg. Cell 66, 51–63. 7. Ephrussi, A., and Lehmann, R. (1992). Induction of germ cell formation by oskar. Nature 358, 387–392. 8. Gonza´lez-Reyes, A., Elliott, H., and St Johnston, D. (1995). Polarization of both major body axes in Drosophila by gurkentorpedo signalling. Nature 375, 654–658. 9. Roth, S., Neuman-Silberberg, F.S., Barcelo, G., and Schu¨pbach, T. (1995). cornichon and the EGF receptor signaling process are necessary for both anterior-posterior and dorsal-ventral pattern formation in Drosophila. Cell 81, 967–978. 10. Micklem, D.R., Dasgupta, R., Elliott, H., Gergely, F., Davidson, C., Brand, A., Gonza´lez-Reyes, A., and St Johnston, D. (1997). The mago nashi gene is required for the polarisation of the oocyte and the formation of perpendicular axes in Drosophila. Curr. Biol. 7, 468–478. 11. Newmark, P.A., Mohr, S.A., Gong, L., and Boswell, R.E. (1997). mago nashi mediates the posterior follicle cell-to-oocyte signal to organize axis formation in Drosophila. Development 124, 3197–3207. 12. Theurkauf, W.E., Smiley, S., Wong, M.L., and Alberts, B.M. (1992). Reorganization of the cytoskeleton during Drosophila oogenesis: implications for axis specification and intercellular transport. Development 115, 923–936. 13. Bo¨hm, H., Brinkmann, V., Drab, M., Henske, A., and Kurzchalia, V. (1997). Mammalian homologues of C. elegans PAR-1 are asymmetrically localized in epithelial cells and may influence their polarity. Curr. Biol. 7, 603–606. 14. Cox, D.N., Lu, B., Sun, T., Williams, L.T., and Jan, Y.N. (2001a). Drosophila par-1 is required for oocyte differentiation and microtubule organization. Curr. Biol. 11, 75–87. 15. Doerflinger, H., Benton, R., Shulman, J.M., and St Johnston, D. (2003). The role of PAR-1 in regulating the polarised microtubule cytoskeleton in the Drosophila follicular epithelium. Development 130, 3965–3975. 16. Vaccari, T., Rabouille, C., and Ephrussi, A. (2005). The Drosophila PAR-1 spacer domain is required for lateral membrane association and for polarization of follicular epithelial cells. Curr. Biol. 15, 255–261. 17. Hurov, J.B., Watkins, J.L., and Piwnica-Worms, H. (2004). Atypical PKC phosphorylates PAR-1 kinases to regulate localization and activity. Curr. Biol. 14, 736–741. 18. Suzuki, A., Hirata, M., Kamimura, K., Maniwa, R., Yamanaka, T., Mizuno, K., Kishikawa, M., Hirose, H., Amano, Y., Izumi, N., et al. (2004). aPKC acts upstream of PAR-1b in both the establishment and maintenance of mammalian epithelial polarity. Curr. Biol. 14, 1425–1435. 19. Guo, S., and Kemphues, K.J. (1995). par-1, a gene required for establishing polarity in C. elegans embryos, encodes a putative Ser/ Thr kinase that is asymmetrically distributed. Cell 81, 611–620. 20. Guo, S., and Kemphues, K.J. (1996). A non-muscle myosin required for embryonic polarity in Caenorhabditis elegans. Nature 382, 455–458. 21. Cuenca, A.A., Schetter, A., Aceto, D., Kemphues, K., and Seydoux, G. (2003). Polarization of the C. elegans zygote proceeds via distinct establishment and maintenance phases. Development 130, 1255–1265. 22. Munro, E., Nance, J., and Priess, J.R. (2004). Cortical flows powered by asymmetrical contraction transport PAR proteins to establish and maintain anterior-posterior polarity in the early C. elegans embryo. Dev. Cell 7, 413–424. 23. Cheeks, R.J., Canman, J.C., Gabriel, W.N., Meyer, N., Strome, S., and Goldstein, B. (2004). C. elegans PAR proteins function by mobilizing and stabilizing asymmetrically localized protein complexes. Curr. Biol. 14, 851–862. 24. Shelton, C.A., Carter, J.C., Ellis, G.C., and Bowerman, B. (1999). The nonmuscle myosin regulatory light chain gene mlc-4 is required for cytokinesis, anterior-posterior polarity, and body morphology during Caenorhabditis elegans embryogenesis. J. Cell Biol. 146, 439–451.