MOESIN Crosslinks Actin and Cell Membrane in Drosophila Oocytes and Is Required for OSKAR Anchoring

Current Biology, Vol. 12, 2060–2065, December 10, 2002, 2002 Elsevier Science Ltd. All rights reserved.

PII S0960-9822(02)01256-3

MOESIN Crosslinks Actin and Cell Membrane in Drosophila Oocytes and Is Required for OSKAR Anchoring Ferenc Jankovics,1 Rita Sinka,1 Tama´s Luka´csovich,2 and Miklo´s Erde´lyi1,3 1 Institute of Genetics 2 Institute of Biochemistry Biological Research Centre of the Hungarian Academy of Sciences P.O. Box 521 H-6726 Szeged Hungary

Summary In Drosophila, development of the embryonic germ cells depends on posterior transport and site-specific translation of oskar (osk) mRNA and on interdependent anchoring of the osk mRNA and protein within the posterior subcortical region of the oocyte [1–3]. Transport of the osk mRNA is mediated by microtubules, while anchoring of the osk gene products at the posterior pole of the oocyte is suggested to be microfilament dependent [4–7]. To date, only a single actin binding protein (TropomyosinII) has been identified with a putative role in osk mRNA and protein anchoring [4]. This communication demonstrates that mutations in the Drosophila moesin (Dmoe) gene that encodes another actin binding protein result in delocalization of osk mRNA and protein from the posterior subcortical region and, as a consequence, in failure of embryonic germ cell development. In Dmoe mutant oocytes, the subcortical actin network is detached from the cell membrane, while the polarized microtubule cytoskeleton is unaffected. In line with the earlier observations [6, 8], colocalization of ectopic actin and OSK protein in Dmoe mutants suggests that the actin cytoskeleton anchors OSK protein to the subcortical cytoplasmic area of the Drosophila oocyte. Results and Discussion DMOE is the Drosophila member of the ERM protein family, which contains three vertebrate members: ezrin, radixin, and moesin [9–12]. ERM proteins have a C-terminal actin binding domain and an N-terminal FERM domain, which is responsible for interaction with several membrane proteins [13]. Based on their structure and predominant subcortical distribution, ERM proteins have been suggested to function as crosslinkers between the cell membrane and the actin cytoskeleton. ERM proteins have been demonstrated to take part in several biological processes, such as cell-cell adhesion, maintenance of cell shape, cell motility, and internal membrane trafficking (for a review of this material, see [12]). Despite their distinct expression pattern, the vertebrate ERM protein family members show functional redundancy [14]. The Drosophila genome, however, con3

Correspondence: [email protected]

tains only a single ERM homolog: Drosophila moesin [15]. Drosophila is a valuable model organism for studying ERM functions, as the ERM homolog is not genetically redundant. In a screen for P element-induced maternal effect germ cell-less mutations, we identified a recessive allele of the Dmoe gene (Figure 1). The DmoeGT193 allele was isolated by making use of a newly designed EGFP-GT mutator P element. EGFP-GT was made by replacing the Gal4 marker gene of a dual-tagging gene trap element, pGT1, [16] with the EGFP marker gene. A collection of 200 viable, X-chromosomal EGFP-GT insertions was generated by the attached-X technique [17] and was screened for maternal effect germ cell-less phenotype by hand dissection of adult test animals. The DmoeGT193 allele was identified as a weak maternal effect germ cellless allele. Subsequent complementation analyses with P element-induced mutations obtained from the Bloomington Stock Center revealed the existence of five additional insertional alleles of Dmoe. Combinations of different Dmoe alleles resulted in germ cell-less phenotypes with a penetrance of 1%–61% (see the figure legend for Figure 1). For further analyses, we chose the DmoeEP1652/DmoeG0415 and DmoeEP1652/Df(1)KA14 combinations, which result in 60% and 61% germ cell-less phenotypes, respectively. Hereafter, the phenotype of these combinations will be referred to as the Dmoe mutant phenotype. Complementation analyses revealed several additional pleiotropic Dmoe phenotypes such as lethality, female sterility, and imperfect eye and wing development. However, in this paper, we focus exclusively on the maternal effect germ cell-less phenotype. Western blot analysis of Dmoe mutant ovaries revealed a reduction in DMOE protein levels, while precise excisions of P elements from Dmoe alleles restored wild-type protein levels (Figure 1B) and germ cell development (data not shown). These results demonstrate that the maternal effect germ cell-less phenotype is a direct consequence of P element insertions in the Dmoe gene. Since ERM proteins have been suggested as functioning as a crosslinker between the actin network and the cell membrane [12], we examined the organization of actin filaments in Dmoe oocytes. Instead of a tight, wildtype subcortical localization, in Dmoe mutants, the actin network seemed to be detached from the cortex. It intruded into the cytoplasm of the oocyte either at the posterior pole or at more lateral regions (Figures 2A and 2B). Interestingly, mislocalized subcortical actin was also found in Schizosaccharomyces pombe after being transformed with truncated Dmoe cDNA. In this heterologous transformation experiment, actin abnormalities were also coupled with cell shape changes [18]. This raised the possibility that observed actin abnormalities in Drosophila oocytes may be the result of abnormalshaped oocytes. To rule out this possibility, we visualized the cell and vitelline membranes by making use of fluorescently labeled lectins (Lycopersicon esculentum

Brief Communication 2061

Figure 1. Molecular and Phenotypic Analysis of the Dmoe Gene (A) Schematic representation of the structure of the Dmoe gene. Triangles represent P element insertions, and black boxes represent exons. Alternative mRNA variants: Dmoe mRNA form 1 (M1) and Dmoe mRNA form 2 (M2) are indicated by bent lines; translational start points are indicated by arrows. Conceptual translation of M1 and M2 predicts the same 578 amino acid long protein. P element insertions DmoeG0404, DmoeG0067, and DmoeGT193 are not indicated since they were mapped to the same insertion point as DmoeG0415 (174,360 bp) on the AE003445 contig, while DmoeG0323 and DmoeEP1652 are inserted at 174,340 bp and 176,762 bp, respectively. (B) Western analysis of the wild-type and Dmoe mutant ovary extracts. A strong 67-kDa band was found in the wild-type (lane1), and a fainter band was found in the DmoeEP1652/DmoeG0415 (lane 2) and DmoeEP1652/Df(1)KA14 mutants (lane 3). A strong 67-kDa band was found in the following P element excision mutant combinations: DmoeEP1652EX2/DmoeEP1652EX2 (lane 4), DmoeG0415EX6/DmoeG0415EX6 (lane 5), and DmoeEP1652EX2/DmoeG0415EX6 (lane 6). The Western blot was also probed for ␥-tubulin as a loading control. (C and D) Wild-type germ cells and the maternal effect germ cellless phenotype of the DmoeEP1652/DmoeG0415 mutant. (C) Embryonic germ cells are formed at the posterior pole of a wild-type embryo and are indicated by an arrowhead. (D) A germ cell-less embryo laid by a DmoeEP1652/DmoeG0415 female. We measured the penetrance of adult germ cell-less phenotypes of the different Dmoe allele combinations. DmoeGT193/DmoeGT193, DmoeGT193/DmoeEP1652, DmoeGT193/ DmoeG041, and DmoeGT193/DmoeG0323 resulted in a 1% penetrant germ cell-less phenotype. In DmoeGT193/Df(1)KA14, DmoeEP1652/DmoeEP1652, DmoeEP1652/DmoeG0415, DmoeEP1652/DmoeG0323, and DmoeEP1652/ Df(1)KA14, we found 7%, 12.5%, 60%, 28%, and 61% penetrant phenotypes, respectively.

and Datura stramonium lectins, respectively, Vector Laboratories) and immunostaining for DE-cadherin, a transmembrane protein that is present in all cell membranes of egg primordia [19, 20]. In developing mutant eggs, however, normal distribution of the lectins and DE-cadherin stainings were detected (Figures 2D and 2G, data not shown). Furthermore, by simultaneous visu-

alization of the cell membrane-specific lectin and actin, we showed that, at actin intrusion sites, the cell membrane was normal (Figures 2C–2H). These results demonstrate that, in Dmoe mutants, the shape of the oocyte is normal and the observed actin abnormality is due to detachment of subcortical actin from the cell membrane. We concluded, therefore, that one of the functions of Dmoe in developing oocytes is to crosslink the subcortical actin network and the cell membrane. Embryonic germ cell formation is initiated during oogenesis by posterior localization of a highly specialized cytoplasmic region, the pole plasm [21]. Since the assembly of the pole plasm depends on the anterior-toposterior transport and subsequent anchoring of osk mRNA, we examined the distribution of osk mRNA in Dmoe oocytes. Instead of the characteristic wild-type posterior localization, we found several different types of abnormal osk mRNA distribution in the mutants (Figures 3A–3D, Table 1). Most frequently, the mislocalized osk mRNA appeared in a scattered pattern concentrated near the posterior pole. Mislocalization of osk mRNA to the central regions of oocytes was also observed. In some stage-10 oocytes, ectopic osk mRNA was found to be localized tightly to the lateral cell cortex. In order to gain insight into the time course of osk mRNA localization in mutants, we measured and compared the penetrance of these phenotypes in stages 9 and 10. We observed a slight, but convincing, decrease of the wildtype osk mRNA localization in stage 10, and this decrease suggests that Dmoe mutations do not interfere with the early posterior transport of osk mRNA but rather with its anchoring to the posterior pole (Table 1). Another pole plasm component, the STAUFEN (STAU) protein, which always colocalizes with osk mRNA [22], consistently showed a distribution identical to that of osk mRNA in mutant oocytes (data not shown). Since correct localization of osk mRNA requires proper functioning of both the oocyte and follicle cells, and Dmoe was reported to be expressed predominantly in the follicle cells [15], we generated Dmoe mutant germline clones to identify the cell type in which Dmoe mutations exert their effect. In developing eggs composed of the homozygous germline and heterozygous follicle cells for DmoeG0415, DmoeG0404, DmoeG0067, and DmoeG0323 mutations, we found osk mRNA mislocalization phenotypes identical to those of DmoeEP1652/ DmoeG0415 and DmoeEP1652/Df(1)KA14 mutant oocytes (Figures 3E–3G). This result demonstrates that Dmoe is required in the germ cells for posterior localization of osk mRNA. To investigate whether Dmoe mutations affect other localized determinants in the Drosophila oocyte, we examined the distribution of gurken (grk) and bicoid (bcd) mRNAs. In mutant oocytes, normal grk and bcd mRNA localization was found, which demonstrates that the Dmoe mutant phenotype is osk specific (Figures 4A and 4B). Similarly to Dmoe mutations, par-1 and Rab11 mutant alleles cause abnormal osk mRNA localization and normal distribution of the bcd and grk mRNAs [23–26]. In these mutants, a plus end microtubule marker molecule, Kin:␤-Gal, is localized to the central region of the oocyte, while the minus end-specific Nod:␤-Gal remains at the anterior pole [27, 28]. In contrast to these mutants, how-

Current Biology 2062

Figure 2. Dmoe Is Required for Actin Organization (A and B) Confocal analysis of the actin cytoskeleton. (A) The subcortical actin network in a wild-type oocyte. In DmoeEP1652/DmoeG0415 mutant oocytes, actin is detached from the oocyte cortex in various foci. (B) An extreme example is shown. (C–H) Visualization of (C and F) the actin network and (D and G) cell membranes. (E and H) A merged picture of actin and membrane stainings. (E) In the wild-type oocyte, actin and fluorescein-Lycopersicon esculentum lectin are colocalized. (H) In the DmoeEP1652/ DmoeG0415 oocyte beneath the actin intrusion, which is marked by an arrow, lectin distribution is normal.

ever, normal Nod:␤-Gal and Kin:␤-Gal localization was observed in Dmoe oocytes (Figures 4C and 4G, Table 1). Furthermore, a normal arrangement of the microtubule network was revealed both by immunostaining and by direct in vivo visualization of the microtubules by using Tubulin:GFP [29] or Tau:GFP fusion proteins [30] (Fig-

ures 4D and 4E, data not shown). By a simultaneous visualization of STAU protein and microtubule plus ends, we showed that STAU mislocalization is not a consequence of an abnormal microtubule network. In Dmoe mutant oocytes, the plus ends of the microtubules normally point to the posterior pole, as demonstrated by the

Figure 3. Dmoe is Required for osk mRNA Localization (A–G) In situ osk mRNA hybridizations. Instead of a (A) wild-type tight posterior localization, (B) osk mRNA is mislocalized in a scattered pattern concentrated near the posterior pole, (C) or it is mislocalized to the more central region of Dmoe mutant oocytes. (D) In some stage-10 Dmoe mutant oocytes, osk mRNA is also detectable in lateral regions attached to the cell cortex. Similarly, in homozygous germline clones for the DmoeG0415 mutation, (E) scattered, (F) central, and (G) lateral osk mRNA mislocalization patterns were observed.

Brief Communication 2063

Table 1. osk mRNA, OSK, and Kin:␤-Gal Protein Localization in the Dmoe Mutant Oocytes Abnormal Localization in Stage-9/10 Oocytes (%) Localized Molecule

Genotype (n ⫽ Number of Scored Oocytes at Stage 9/10)

Posterior Localization in Stage-9/10 Oocytes (%)

No Staining

Scattered Posterior Pattern

Laterala

Central

osk mRNA

Wild-type (n ⫽ 50/87) DmoeEP1652/DmoeG0415 (n ⫽ 70/134) DmoeEP1652/Df(1)KA14 (n ⫽ 46/96) Wild-type (n ⫽ 67/57) DmoeEP1652/DmoeG0415 (n ⫽ 104/97) DmoeEP1652/Df(1)KA14 (n ⫽ 95/70) Wild-type (n ⫽ 123) DmoeEP1652/Df(1)KA14 (n ⫽ 129)

80/67 44/28 52/20 88/81 42/15 11/11 78 74

20/33 27/36 39/32 12/19 30/37 68/60 22 26

0 17/29 4/46 0 26/26 20/23 0 0

0 0/4 0/2 0 0/16 0/4 0 0

0 11/2 5/1 0 2/5 1/1 0 0

OSK

Kin:␤-Galb a b

Lateral mislocalization was exclusively found in stage-10 oocytes. Kin:␤-Gal localization was determined exclusively in stage 9 [35].

correct localization of Kin:␤-Gal. In contrast, however, STAU is mislocalized (Figures 4F and 4G), suggesting again that the mislocalization of osk mRNA is not a consequence of abnormal transport; rather, it appears to be caused by abnormal anchoring.

In order to further investigate the role of DMOE in osk regulation, we examined the level and distribution of OSK protein in Dmoe oocytes. In mutant ovaries, a reduced level of OSK was found by Western analysis (data not shown). Consistent with this observation, no OSK

Figure 4. Polarity of the Microtubule Network Is Normal in Dmoe Oocytes (A and B) Proper (A) grk and (B) bcd mRNA localization was detected by in situ mRNA hybridization in DmoeEP1652/Df(1)KA14 mutant oocytes. (C) Nod:␤-Gal normally localizes to the anterior margin of the DmoeEP1652/DmoeG0415 mutant oocyte. (D and E) Direct visualization of the microtubules of (D) wild-type and (E) DmoeEP1652/ DmoeG0415 mutant oocyte by a Tubulin:GFP fusion protein. An anterior to posterior gradient of the fluorescence indicates normal microtubule arrangement in stage-8 oocytes. (F and G) Simultaneous immunostaining of Kin:␤-Gal (shown in green) and STAU (shown in red). (F) In wild-type oocytes, both STAU and Kin:␤-Gal are localized at the posterior pole. (G) In DmoeEP1652/DmoeG0415 mutant oocytes, Kin:␤-Gal is localized correctly; however, STAU accumulates ectopically.

Figure 5. Dmoe Mutations Result in OSK Delocalization (A–E) Double staining for OSK protein and actin (OSK in red, actin in green). (A) In wildtype oocytes, actin is subcortical and OSK is tightly localized at the posterior pole. (B) In DmoeEP1652/DmoeG0415 mutants, OSK is found in a scattered pattern near the posterior pole. (C and D) Actin is detached from the posterior pole and is colocalized with the ectopic OSK. The framed section of Figure 5D is also shown with high-power magnification. (E) In DmoeEP1652/ DmoeG0415, OSK protein is localized to the lateral subcortical actin network in two foci. Arrows indicate the lateral OSK localizations. Similar phenotypes were found in the DmoeEP1652/Df(1)KA14 oocytes.

Current Biology 2064

was detected by immunostaining in the majority of the mutant oocytes, while, in the remaining cases, the pattern of the mislocalized OSK protein was found in a spatial and temporal distribution similar to that of osk mRNA (Table 1). OSK was mostly found in a scattered pattern at the posterior region of the oocytes (Figure 5B). In the majority of Dmoe oocytes in which OSK was delocalized, the actin network appears to be attached normally to the cell membrane, at least as far as can be visualized with light microscopy. This indicates that Dmoe might have a role in OSK anchoring, which is separable from its actin-cell membrane crosslinking function. In the rare cases when detachment of actin from the cell membrane occurred at the posterior pole, we observed that OSK protein was colocalized with the ectopic actin network (Figures 5C and 5D). This ectopic colocalization of the subcortical actin and OSK protein in Dmoe mutants reveals that the actin network has OSK anchoring capacity and strongly suggests that the actin network anchors OSK at the normal place, at the posterior subcortical region, too. In Dmoe mutants, we also observed the recently published [6, 8] abnormal osk mRNA and protein localization phenotype when osk products are tightly localized to lateral segments of the subcortical actin (Figures 3D, 3G, and 5E). Our results confirm the model by Cha et al. [6] that proposes that the focused posterior localization of osk is not defined by a special segment of the posterior subcortical actin; rather, it is determined by the microtubule-based posterior transport of osk mRNA to this region. In this paper, we demonstrate that the partial loss of Dmoe activity weakens the ability of the actin network to anchor osk mRNA and protein, resulting in different degrees of their delocalization. We also show that Dmoe has actin-cell membrane crosslinking activity in the developing oocyte. Finally, we present cell biological evidence that factors other than actin define the site of osk mRNA and protein localization.

approximately one ovary was loaded per lane on 10% SDS-PAGE gels. Proteins were transferred to PVDF membrane (Amersham) in transfer buffer (20% methanol, 25 mM Tris-Cl, 192 mM Glycine). BioRad Kaleidoscope Prestained Standards were used as molecular weight markers. Blots were blocked overnight with 5% dry milk in TTBS and were incubated with anti-OSK antibody (1:1000), antiDMOE antibody (1:10000), or anti-␥-tubulin (Sigma) antibody (1:3000) in 5% dry milk in TTBS. The membrane was washed in dH2O and TTBS and was incubated with peroxidase-conjugated goat anti-rabbit secondary antibody in 5% dry milk in TTBS (Jackson Immuno Research Laboratories). Signals were detected with enhanced chemiluminescence. Whole-Mount Antibody and Lectin Stainings and In Situ Hybridization For antibody and lectin stainings, ovaries were dissected in PBS and fixed for 15 min in 4% paraformaldehyde in PBS. Ovaries were washed twice (20 min per wash) in PBT and were blocked for 4 hr in blocking solution (2% BSA, 0.1% Triton X-100, 5% Normal Goat Serum, and 0.02% NaN3 in PBS). Ovaries were then incubated overnight with rabbit anti-OSK (1:500), rabbit anti-STAU (1:2000), rat antiDE-cadherin (1:20), rat anti-␣-tubulin (1:150, GeneTex), and mouse anti-␤-Gal (1:20, Sigma) primary antibodies in blocking solution. After four 30-min washes in PBT, ovaries were incubated for 4 hr with a Cy3-conjugated anti-rabbit secondary antibody (for OSK and STAU), Cy3-conjugated anti-rat secondary antibody (for DE-cadherin), Cy2-conjugated anti-rat secondary antibody (for ␣-tubulin), or Cy2-conjugated anti-mouse secondary antibodies (for Kin:␤-Gal) (Jackson Immuno Research Laboratories), diluted 1:200 in blocking solution. After two 30-min washes in PBT, ovaries were stained for 4 hr with OregonGreen-phalloidin (1:20) (Molecular Probes). For lectin staining, paraformaldehyde-fixed (see above) ovaries were incubated overnight in blocking solution containing 150 ␮g/ml cell membrane-specific fluorescein-Lycopersicon esculentum or vitelline membrane-specific fluorescein-Datura stramonium lectin [19] (Vector Laboratories) and rhodamine-phalloidin (1:20) (Molecular Probes). Preparations were mounted in 80% glycerol and 4% n-propyl gallate. In situ hybridizations were carried out by using Digoxigeninlabeled DNA probes (Boehringer Mannheim). The osk probe corresponded to the 2.1-kb SacI fragment of the osk cDNA [32], the bcd probe corresponded to the 2.5-kb EcoRI fragment of the bcd cDNA [33], and the grk probe corresponded to the XhoI-NotI fragment of the grk cDNA [34]. In situ hybridization was carried out as described in [32]. Acknowledgments

Experimental Procedures Cloning and Western Blot Analysis The EGFP-GT mutator element was designed as follows. The BamHI-SacII fragment (containing the Gal4 gene, followed by the hsp transcription terminator signal and the hs-neo gene on the complementer strand) of the pGT1 plasmid [16] was replaced with the EGFP coding sequence of plasmid pEGFP-N1 (Clontech) completed by an SV40 transcription terminator. As a result of this cloning step, NotI and XbaI restriction sites were created between the EGFP and SV40 sequences, and the SacII site was transformed into PstI. Due to this replacement, the EGFP gene was placed into the same context as the Gal4 gene in the original pGT1 vector. Flanking sequences of the P elements in the Dmoe gene were amplified by inverse PCR [31], sequenced with a Perkin Elmer sequencer, and compared to the Berkley Drosophila Genome Project public sequence repository. Dmoesin cDNA M1 corresponding to the EST LD42363 was amplified with the following specific primers: forward, 5⬘-GGGGGATCCG CGGCAATTGGATAAGAAACG-3⬘; reverse, 3⬘-CCTCTAGACTTTCTG GAGTCTCACCGCTCCC-5⬘. Dmoesin cDNA M1 was verified by restriction analyses. For Western analysis, hand-dissected ovaries were sonicated in 10 volumes extraction buffer (6 mM Tris-Cl [pH ⫽ 6.8], 6.4% glycerol, 2% SDS, 100 mM DTT and bromophenol blue), containing 1 mM PMSF, 0.2 mM leupeptin, and 15 ␮M aprotinin. The equivalent of

We would like to thank Anne Ephrussi for the anti-OSK, Daniel St. Johnston for the anti-STAU, Masatoshi Takeichi for the DE-Cadherin, Daniel Kiehart for the anti-DMOE antisera, and Francois Payre for sharing unpublished results. We are grateful to Daniel St. Johnston for providing us a Tau:GFP-expressing Drosophila line. We would also like to thank Ira Clark for the Kin:␤-Gal and Nod:␤-Gal transgenic Drosophila strains. We are also grateful to Mim Bower for critically reading the manuscript. This work was supported by a grant (T038135) from the Hungarian National Science Foundation (OTKA). Received: May 15, 2002 Revised: September 30, 2002 Accepted: September 30, 2002 Published: December 10, 2002 References 1. Johnstone, O., and Lasko, P. (2001). Translational regulation and RNA localization in Drosophila oocytes and embryos. Annu. Rev. Genet. 35, 365–406. 2. Rongo, C., Gavis, E.R., and Lehmann, R. (1995). Localization of oskar RNA regulates oskar translation and requires Oskar protein. Development 121, 2737–2746. 3. Markussen, F.H., Michon, A.M., Breitwieser, W., and Ephrussi,

Brief Communication 2065

4.

5.

6.

7.

8.

9.

10.

11.

12. 13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24.

A. (1995). Translational control of oskar generates short OSK, the isoform that induces pole plasma assembly. Development 121, 3723–3732. Erdelyi, M., Michon, A.M., Guichet, A., Glotzer, J.B., and Ephrussi, A. (1995). Requirement for Drosophila cytoplasmic tropomyosin in oskar mRNA localization. Nature 377, 524–527. Tetzlaff, M.T., Jackle, H., and Pankratz, M.J. (1996). Lack of Drosophila cytoskeletal tropomyosin affects head morphogenesis and the accumulation of oskar mRNA required for germ cell formation. EMBO J. 15, 1247–1254. Cha, B.J., Serbus, L.R., Koppetsch, B.S., and Theurkauf, W.E. (2002). Kinesin I-dependent cortical exclusion restricts pole plasm to the oocyte posterior. Nat. Cell Biol. 4, 592–598. Brendza, R.P., Serbus, L.R., Duffy, J.B., and Saxton, W.M. (2000). A function for kinesin I in the posterior transport of oskar mRNA and Staufen protein. Science 289, 2120–2122. Vanzo, N.F., and Ephrussi, A. (2002). Oskar anchoring restricts pole plasm formation to the posterior of the Drosophila oocyte. Development 129, 3705–3714. Bretscher, A., Chambers, D., Nguyen, R., and Reczek, D. (2000). ERM-Merlin and EBP50 protein families in plasma membrane organization and function. Annu. Rev. Cell Dev. Biol. 16, 113–143. Bretscher, A. (1999). Regulation of cortical structure by the ezrin-radixin-moesin protein family. Curr. Opin. Cell Biol. 11, 109–116. Tsukita, S., and Yonemura, S. (1999). Cortical actin organization: lessons from ERM (ezrin/radixin/moesin) proteins. J. Biol. Chem. 274, 34507–34510. Louvet-Vallee, S. (2000). ERM proteins: from cellular architecture to cell signaling. Biol. Cell 92, 305–316. Chishti, A.H., Kim, A.C., Marfatia, S.M., Lutchman, M., Hanspal, M., Jindal, H., Liu, S.C., Low, P.S., Rouleau, G.A., Mohandas, N., et al. (1998). The FERM domain: a unique module involved in the linkage of cytoplasmic proteins to the membrane. Trends Biochem. Sci. 23, 281–282. Doi, Y., Itoh, M., Yonemura, S., Ishihara, S., Takano, H., Noda, T., and Tsukita, S. (1999). Normal development of mice and unimpaired cell adhesion/cell motility/actin-based cytoskeleton without compensatory up-regulation of ezrin or radixin in moesin gene knockout. J. Biol. Chem. 274, 2315–2321. McCartney, B.M., and Fehon, R.G. (1996). Distinct cellular and subcellular patterns of expression imply distinct functions for the Drosophila homologues of moesin and the neurofibromatosis 2 tumor suppressor, merlin. J. Cell Biol. 133, 843–852. Lukacsovich, T., Asztalos, Z., Awano, W., Baba, K., Kondo, S., Niwa, S., and Yamamoto, D. (2001). Dual-tagging gene trap of novel genes in Drosophila melanogaster. Genetics 157, 727–742. Gans, M., Audit, C., and Masson, M. (1975). Isolation and characterization of sex-linked female-sterile mutants in Drosophila melanogaster. Genetics 81, 683–704. Edwards, K.A., Montague, R.A., Shepard, S., Edgar, B.A., Erikson, R.L., and Kiehart, D.P. (1994). Identification of Drosophila cytoskeletal proteins by induction of abnormal cell shape in fission yeast. Proc. Natl. Acad. Sci. USA 91, 4589–4593. Bretscher, M.S. (1996). Expression and changing distribution of the human transferrin receptor in developing Drosophila oocytes and embryos. J. Cell Sci. 109, 3113–3119. Oda, H., Uemura, T., and Takeichi, M. (1997). Phenotypic analysis of null mutants for DE-cadherin and Armadillo in Drosophila ovaries reveals distinct aspects of their functions in cell adhesion and cytoskeletal organization. Genes Cells 2, 29–40. Mahowald, A.P. (2001). Assembly of the Drosophila germ plasm. Int. Rev. Cytol. 203, 187–213. Micklem, D.R., Adams, J., Grunert, S., and St. Johnston, D. (2000). Distinct roles of two conserved Staufen domains in oskar mRNA localization and translation. EMBO J. 19, 1366–1377. 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 localization to the posterior pole. Cell 101, 377–388. Tomancak, P., Piano, F., Riechmann, V., Gunsalus, K.C., Kemphues, K.J., and Ephrussi, A. (2000). A Drosophila melano-

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

gaster homologue of Caenorhabditis elegans par-1 acts at an early step in embryonic-axis formation. Nat. Cell Biol. 2, 458–460. Jankovics, F., Sinka, R., and Erdelyi, M. (2001). An interaction type of genetic screen reveals a role of the Rab11 gene in oskar mRNA localization in the developing Drosophila melanogaster oocyte. Genetics 158, 1177–1188. Dollar, G., Struckhoff, E., Michaud, J., and Cohen, R.S. (2002). Rab11 polarization of the Drosophila oocyte: a novel link between membrane trafficking, microtubule organization, and oskar mRNA localization and translation. Development 129, 517–526. Clark, I.E., Jan, L.Y., and Jan, Y.N. (1997). Reciprocal localization of Nod and kinesin fusion proteins indicates microtubule polarity in the Drosophila oocyte, epithelium, neuron and muscle. Development 124, 461–470. Clark, I., Giniger, E., Ruohola-Baker, H., Jan, L.Y., and Jan, Y.N. (1994). Transient posterior localization of a kinesin fusion protein reflects anteroposterior polarity of the Drosophila oocyte. Curr. Biol. 4, 289–300. Grieder, N.C., de Cuevas, M., and Spradling, A.C. (2000). The fusome organizes the microtubule network during oocyte differentiation in Drosophila. Development 127, 4253–4264. Micklem, D.R., Dasgupta, R., Elliott, H., Gergely, F., Davidson, C., Brand, A., Gonzalez-Reyes, A., St. Johnston, D., et al. (1997). The mago nashi gene is required for the polarization of the oocyte and the formation of perpendicular axes in Drosophila. Curr. Biol. 7, 468–478. Huang, A.M., Rehm, E.J., and Rubin, G.M. (2000). Recovery of DNA sequences flanking P element insertions. In Drosophila Protocols, W. Sullivan, M. Ashburner, and R.S. Hawley, eds. (Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press), pp. 429–437. 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. Berleth, T., Burri, M., Thoma, G., Bopp, D., Richstein, S., Frigerio, G., Noll, M., Nusslein-Volhard, C., et al. (1988). The role of localization of bicoid RNA in organizing the anterior pattern of the Drosophila embryo. EMBO J. 7, 1749–1756. Neuman-Silberberg, F.S., and Schupbach, T. (1993). The Drosophila dorsoventral patterning gene gurken produces a dorsally localized RNA and encodes a TGF alpha-like protein. Cell 75, 165–174. Clark, I., Giniger, E., Ruohola-Baker, H., Jan, L.Y., and Jan, Y.N. (1994). Transient posterior localization of a kinesin fusion protein reflects anteroposterior polarity of the Drosophila oocyte. Curr. Biol. 4, 289–300.