- Email: [email protected]
Laminin A is required for follicle cell–oocyte signaling that leads to establishment of the anterior–posterior axis in Drosophila
Brief Communication
683
Laminin A is required for follicle cell–oocyte signaling that leads to establishment of the anterior–posterior axis in Drosophila Wu-Min Deng and Hannele Ruohola-Baker The establishment of the anterior–posterior (AP) axis in Drosophila melanogaster requires signaling between the oocyte and surrounding somatic follicle cells during oogenesis [1,2]. First, a signal from the oocyte (Gurken α (TGFα α) homolog) (Grk), a transforming growth factor-α is received by predetermined terminal follicle cells in which the epidermal growth factor receptor (EGFR) pathway is activated and a posterior fate is induced [2–4]. Later, the posterior follicle cells send an unidentified signal back to the oocyte, which leads to the reorganization of its cytoskeletal polarity. This reorganization is required for proper localization of maternal determinants, such as oskar (osk) and bicoid (bcd) mRNAs, that determine the AP polarity of the oocyte and the subsequent embryo [2]. We show here that when the gene lanA, which encodes the extracellular matrix component laminin A, is mutated in posterior follicle cells, localization of AP determinants is disrupted in the underlying oocyte. Posterior follicle-cell differentiation and follicle cell apical–basal polarity are unaffected in the lanA mutant cells, suggesting that laminin A is required for correct signaling from the posterior follicle cells that polarizes the oocyte. This is the first evidence that the extracellular matrix is involved in the establishment of a major body axis. Address: Department of Biochemistry, University of Washington, Box 357350, Seattle, Washington 98195-7350, USA. Correspondence: Hannele Ruohola-Baker E-mail: [email protected] Received: 28 January 2000 Revised: 3 April 2000 Accepted: 3 April 2000 Published: 19 May 2000 Current Biology 2000, 10:683–686
localized at the posterior of stage-9 wild-type oocytes, was used as a marker for oocyte polarity in F1 progeny [6]. Thirteen different genes were identified from this screen, three of which encode transmembrane or secreted proteins and are therefore of great interest to us. The gene encoding laminin A (lanA) (Figure 1a) [7], was one of the genes obtained from this screen. In 109-3-9/EP(lanA) egg chambers, Kin–β-Gal is diffused from its normal position, the posterior pole of the stage-9 oocyte (data not shown). To examine whether lanA is expressed in the posterior follicle cells, whole-mount ovarian in situ hybridization was undertaken using lanA cDNA as a probe. lanA expression is first detected in all follicle cells of stage 1–4 egg chambers (Figure 1b, left panel). Starting at around stage 5, expression in the anterior follicle cells is decreased, while the rest of the follicle cells remain strongly stained (Figure 1b, middle panel). At around stage 6, lanA is expressed in follicle cells covering the posterior half of the egg chamber (Figure 1b, right panel). There is no detectable lanA expression in the germline cells. Laminin protein is present in both the basement membrane and the ECM between the posterior follicle cells and the oocyte at stage 6 ([8] and data not shown). As lanA is expressed in the early follicle cells and shows differential staining along the AP axis in the follicle-cell epithelium, we asked whether laminin A has a role in communication between posterior follicle cells and the oocyte. First, we examined the distribution pattern of an oocyte posterior marker, Staufen (Stau) [1], in lanA9–32 (a null allele Figure 1 (a) Exons 2 5′
4
1
0960-9822/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
3
6 5
8 10 7
9 11
12
14 13
15
3′
lanA9–32 break points
Results and discussion The signaling pathway by which the posterior follicle cells determine the polarization of the oocyte remains largely unclear [1,2]. In order to identify genes involved in signaling from the posterior follicle cells, we carried out a modular misexpression screen [5]. Over 2000 EP lines, each of which has a Gal4-responsive site randomly inserted in the fly genome, were crossed to a Gal4 driver line, 1093-9, which shows Gal4 expression in most follicle cells, including the posterior cells, starting at stage 6. A fusion protein, kinesin–β-galactosidase (Kin–β-Gal), which is
(b)
st1 Current Biology
st3
st5
st6
The structure of the laminin A gene (lanA) and its expression during early oogenesis. (a) The null allele lanA9–32 used in this study has a deletion between the 5′ untranslated region and the third exon. (b) lanA expression is first detected in all follicle cells from stages 1–3. lanA expression is decreased in the anterior follicle cells during stages 5–6.
684
Current Biology Vol 10 No 11
Figure 2 (a)
(e)
(f)
(j)
(k)
(b)
(c)
(d)
(g)
(h)
(i)
Current Biology
Laminin A function is required in the posterior follicle cells for the proper localization of oocyte anterior–posterior polarity determinants. (a) Stau (red fluorescence) is normally localized at the posterior of the oocyte during stages 9–10 (arrow). Green staining in (a–d) and (g–i) represents green fluorescent protein (GFP), which marks wild-type cells. (b) In a lanA9–32 germline clone, Stau localization is unaffected (arrow). (c) When the lanA mutant follicle-cell clone lies over the anterior of the oocyte, Stau is correctly localized to the posterior pole of the stage 9–10 oocyte (arrow). (d) When the mutant clone covers the posterior end of the oocyte, Stau is mislocalized to the center of the oocyte (arrow). (e) Another posterior marker, oskar (osk) RNA, is normally localized at the posterior pole of the oocyte during stages
8–10 of oogenesis (arrow). (f) In an unmarked lanA mosaic egg chamber, osk RNA is mislocalized to the center of the oocyte (arrow). (g) In a wild-type stage-10 egg chamber, Oskar protein (red fluorescence) is detected at the posterior of the oocyte (arrow). (h) When posterior follicle cells are mutant for lanA, however, Oskar is mislocalized (arrow). (i) Roughly 20 wild-type posterior follicle cells (indicated by the bracket) are sufficient to restore Oskar localization at the posterior of the oocyte (arrow). (j) The anterior determinant bcd RNA is localized to the anterior of the oocyte in a stage-9 wild-type egg chamber (arrowheads). (k) In an unmarked lanA mosaic egg chamber, bcd is mislocalized to the posterior (arrow), in addition to the normal anterior staining (arrowheads).
of lanA [7]) mosaic egg chambers. Stau is normally localized to the posterior pole of the oocyte during stages 9–10 in the wild-type (Figure 2a), and this pattern is not changed in lanA– germline clones (Figure 2b). This is consistent with the observation that lanA expression is not detected in the germline cells. Stau localization is also unaffected in lanA mosaics in which the mutant follicle-cell clones are located over the anterior part of the oocyte (Figure 2c). In 87% (n = 39) of egg chambers, however, in which lanA– folliclecell clones cover the posterior end of the oocyte, Stau localization is no longer detectable at the posterior pole of the oocyte (Figure 2d). Instead, it is mislocalized as a dot at the center of the oocyte or as a streak diffusing away from the posterior of the oocyte (Figure 2d and data not shown). As Stau is required for localization of osk RNA [1], the posterior maternal determinant, we investigated whether osk RNA and Oskar protein (Osk) are also mislocalized in lanA mosaic egg chambers. Indeed, mislocalization and diffusion of osk RNA from the posterior is observed in unmarked lanA mosaic egg chambers (Figure 2f). Anti-Oskar antibody was also used to examine the distribution of this posterior determinant in marked lanA9–32 clones. Similarly to the Stau localization experiments, Osk is mislocalized in 88% (n = 32) of the egg chambers where posterior follicle cells are mutant for lanA (Figure 2h). In order to cause the mislocalization phenotype, a patch of roughly 90 posterior-most
follicle cells must be mutant for lanA. Within a large lanA– follicle-cell clone, however, if 20 of the posterior-most cells are wild type, correct localization of Osk and Stau is restored (Figure 2i). These data indicate that lanA expression in the posterior follicle cells is critical for localization of posterior determinants in the oocyte. As signaling from follicle cell to oocyte also has a role in localizing anterior components [9–11], we examined the position of bcd mRNA in unmarked lanA mosaics. In some stage 8–9 egg chambers, mislocalization of bcd RNA in the posterior was detected (Figure 2k), similarly to egg chambers mutant for the Notch or EGFR pathways [9–11]. Localization of osk and bcd RNAs and Stau protein are microtubule dependent [2]. To determine whether cytoskeletal polarity is normal in lanA mosaics, we examined the localization of two microtubule motor proteins, dynein heavy chain (Dhc64C) and the Kin–β-Gal fusion protein [6,12]. In the wild type, these two motor proteins are localized to the posterior pole of the oocyte during stages 8 and 9 (Figure 3a and data not shown). We observed mislocalization of Dhc64C in 87% (n = 15) of lanA– posterior follicle-cell clones (Figure 3b). Kin–β-Gal is also frequently diffused away from the posterior in lanA posterior follicle-cell mosaics (data not shown). Another marker for the correct arrangement of the oocyte cytoskeleton is the
Brief Communication
685
Figure 4
Figure 3 (a)
(c)
(a)
(b)
(d)
(c)
Current Biology
Oocyte microtubule polarity is disrupted in lanA mosaics. (a) Antibody staining for dynein heavy chain (red) shows its localization at the posterior pole of the oocyte (arrows) in a stage-8 wild-type egg chamber. Green staining in (a–d) represents GFP and marks wild-type germline and follicle cells. (b) When the posterior follicle cells are mutant for lanA, dynein heavy chain is mislocalized to the center of the oocyte (arrow). (c) A stage 10 wild-type egg chamber, in which the oocyte nucleus (arrow) is located at the dorsal anterior corner of the oocyte. (d) In an egg chamber where most follicle cells, including the posterior follicle cells, are mutant for lanA, the oocyte nucleus failed to migrate from the posterior of the oocyte (arrow).
location of the oocyte nucleus (Figure 3c), as its movement from the posterior to the dorsal anterior corner during stages 7–8 reflects the microtubule polarity [3,4,10,11]. We checked whether the oocyte nucleus is correctly localized in lanA– follicle-cell clones and found that in 4% (n = 98) of egg chambers in which the posterior follicle cells are mutant, oocyte nuclei remain in the posterior after stage 8 (Figure 3d). As both microtubule motors and oocyte nuclei were mislocalized in lanA mosaics, we conclude that the defect in laminin A function in posterior follicle cells affects the organization of oocyte cytoskeletal polarity. The actin cytoskeleton is also involved in osk mRNA and Stau localization within the oocyte [13,14]. We therefore examined the overall integrity of the oocyte actin cytoskeleton network. No gross abnormalities were observed in oocytes lying under lanA– posterior follicle-cell clones (data not shown); we cannot, however, rule out the possibility that loss of lanA function produces minor defects in the oocyte actin cytoskeleton. Mutants that disrupt follicle-cell differentiation, such as Notch–, grk– and top–, also disturb the polarity of the oocyte [3,4,9–11]. In Notch mutants, the follicle cells at the termini of the egg chamber do not differentiate properly and no anterior or posterior follicle-cell fates are observed [3,4]. In grk– egg chambers, posterior follicle cells adopt the default anterior fate as a result of the lack of the germline signal. As a result, anterior follicle-cell markers such as enhancer-trap L53b and decapentaplegic
(f)
(b)
(d)
(g)
(e)
(h)
Current Biology
Differentiation of posterior follicle cells appears to be normal in lanA mosaics. (a) In wild-type egg chambers, dpp–lacZ (red) is stained in stretched and centripetal migrating follicle cells (arrowheads), which are located at the anterior of the stage-10 egg chamber. Green in (a–h) is GFP, which marks wild-type cells. In mutants of genes such as gurken, in which posterior follicle-cell differentiation is disrupted, dpp–lacZ was found in posterior follicle cells (data not shown). (b) dpp–lacZ is not, however, expressed in the posterior follicle cells (arrow) in mosaics when lanA is mutated in all follicle cells. (c) Expression of Pointed (red) marks posterior follicle cells in wild-type egg chambers (arrow). (d,e) When lanA is mutated in posterior follicle cells, Pointed expression is normal (arrows). (f–h) Follicle-cell polarity is detected by staining for Armadillo (red). In both wild-type (green) and lanA mosaics (marked by white line brackets), Armadillo is localized at the apical side of the follicle epithelium (arrowheads).
(dpp) are detected in the posterior cells, whereas posterior markers such as pointed and 12A3 are no longer expressed [2,10,11,15]. To investigate whether the involvement of lanA in establishing AP polarity is a secondary consequence of defects in posterior follicle-cell differentiation, both anterior and posterior follicle-cell markers were examined in lanA– follicle-cell clones. We found that dpp, the anterior marker that is normally expressed in the stretched and centripetally migrating follicle cells (Figure 4a), is detected in these anterior cells but not in posterior cells in those lanA mosaics that have posterior folliclecell clones (Figure 4b). On the other hand, the posterior marker pointed, which is normally expressed in a subgroup of posterior follicle cells from stage 6 to later stages (Figure 4c), is still expressed in lanA– posterior follicle-cell clones (Figure 4d,e). As pointed expression requires normal differentiation of the terminal follicle cells, this result suggests that the differentiation of the lanA– terminal follicle cells is normal. Furthermore, as pointed expression is induced by
686
Current Biology Vol 10 No 11
the Grk–EGFR pathway [15], this result indicates that EGFR signaling from the oocyte to the posterior follicle cells is not disrupted by mutation of lanA. In addition, on the basis of a normal expression pattern of the gene kekkon (data not shown), EGFR-induced dorsal–ventral polarity in follicle cells was also normal in lanA– anterior folliclecell clones. Another marker, P1333, which is stained in the majority of columnar follicle cells at around stage 10 of oogenesis, shows no difference in expression in lanA– clones (data not shown). These results indicate that posterior follicle-cell differentiation is normal and that the Notch and EGFR signaling pathways act properly in lanA mosaics. We also investigated whether the polarity of follicle-cell epithelium is normal in lanA mosaics. The Drosophila β-catenin homolog Armadillo and the transmembrane receptor Notch, which are predominantly localized to the apical side of the follicle-cell epithelium, were used as markers for follicle-cell polarity in lanA mosaics. No detectable difference in their distribution is found in lanA– follicle-cell clones (Figure 4f–h and data not shown), suggesting that follicle-cell polarity is normal. As shown above, lanA is required in posterior follicle cells for the localization of maternal determinants and reorganization of cytoskeletal polarity in the oocyte. As posterior follicle-cell differentiation appears normal in lanA mutant follicle-cell clones, we propose that lanA is required for signaling from the posterior follicle cells to the oocyte to establish the oocyte polarity and is therefore the first secreted component identified in this ill-defined pathway. Laminin A could function directly as a ligand or as a helper to provide a platform for signal delivery from the posterior follicle cells. In Xenopus, for example, laminin can convert netrin-mediated attraction into repulsion during growthcone guidance [16]. We tentatively tested the first possibility by investigating whether integrins, the potential receptors for laminin A, are involved in the formation of oocyte polarity. No mislocalization of Osk or Stau was, however, detected in the mosaics of the integrin genes analyzed so far. It will be interesting to test whether Drosophila homologs for other laminin receptors are involved in polarity formation in the oocyte [17]. In summary, our data show for the first time that laminin is a crucial player in the formation of a major body axis, and that this is achieved through its involvement in signaling between two types of cells––somatic epithelial and germline cells. It will be interesting to determine whether laminins are involved in bodyaxis formation in other organisms. Supplementary material Supplementary material including additional methodological detail is available at http://current-biology.com/supmat/supmatin.htm.
Acknowledgements We thank Danny Brower, Ed Giniger, Corey Goodman, Tom Hays, Paul Lasko, Todd Laverty, Ruth Lehmann, Stefan Luschnig, Gerry Rubin, Trudi Schüpbach, Daniel St Johnston, Beat Suter, the Bloomington stock center, and the Developmental Studies Hybridoma Bank for sending us antibodies,
cDNA clones and fly stocks. Mike Tworoger, Karin Fischer, Jennifer Blasi, Susan Kwong and Jeff Lin gave excellent technical support, and we thank the Keck Center for allowing us to use their confocal microscopy facilities. We would also like to thank Ed Giniger for critical reading of our manuscript. This work was supported by grants from the Established Investigatorship Award from the American Heart Association, March of Dimes Birth Defect Foundation, Basil O’Conner Starter Research Grant, National Institutes of Health, and Pew Memorial Trust, Pew Scholarship in the Biomedical Sciences to H.R.-B. W-M.D. was supported by a Benjamin Schultz Fellowship.
References 1. Rongo C, Lehmann R: Regulated synthesis, transport and assembly of the Drosophila germ plasm. Trends Genet 1996, 12:102-109. 2. Deng WM, Bownes M: Patterning and morphogenesis of the follicle cell epithelium during Drosophila oogenesis. Int J Dev Biol 1998, 42:541-552. 3. González-Reyes A, St Johnston D: Patterning of the follicle cell epithelium along the anterior-posterior axis during Drosophila oogenesis. Development 1998, 125:2837-2846. 4. Larkin MK, Deng-M, Tworoger M, Holder K, Clegg N, Ruohola-Baker H: Role of Notch pathway in terminal follicle cell differentiation in Drosophila oogenesis. Dev Genes Evol 1999, 209:301-311. 5. Rφrth P: A modular misexpression screen in Drosophila detecting tissue-specific phenotypes. Proc Natl Acad Sci USA 1996, 93:12418-12422. 6. Clark I, Giniger E, Ruohola-Baker H, Jan LY, Jan YN: Transient posterior localization of a kinesin fusion protein reflects anteroposterior polarity of the Drosophila oocyte. Curr Biol 1994, 4:289-300. 7. Henchcliffe C, Garcia-Alonso L, Tang J, Goodman C S: Genetic analysis of laminin A reveals diverse functions during morphogenesis in Drosophila. Development 1993, 118:325-337. 8. Gutzeit HO, Eberhardt W, Gratwohl E: Laminin and basement membrane-associated microfilaments in wild-type and mutant Drosophila ovarian follicles. J Cell Sci 1991, 100:781-788. 9. Ruohola H, Bremer KA, Baker D, Swedlow JR, Jan LY, Jan YN: Role of neurogenic genes in establishment of follicle cell fate and oocyte polarity during oogenesis in Drosophila. Cell 1991, 66:433-449. 10. González-Reyes A, Elliott H, St Johnston D: Polarization of both major body axes in Drosophila by gurken-torpedo signalling. Nature 1995, 375:654-658. 11. Roth S, Neuman-Silberberg FS, Barcelo G, Schüpbach T: cornichon and the EGF receptor signaling process are necessary for both anterior-posterior and dorsal-ventral pattern formation in Drosophila. Cell 1995, 81:967-978. 12. Li M, McGrail M, Serr M, Hays TS: Drosophila cytoplasmic dynein, a microtubule motor that is asymmetrically localized in the oocyte. J Cell Biol 1994, 126:1475-1494. 13. Erdelyi M, Michon AM, Guichet A, Glotzer JB, Ephrüssi A: Requirement for Drosophila cytoplasmic tropomyosin in oskar mRNA localization. Nature 1995, 377:524-527. 14. Manseau L, Calley J, Phan H: Profilin is required for posterior patterning of the Drosophila oocyte. Development 1996, 122:2109-2116. 15. Morimoto AM, Jordan KC, Tietze K, Britton JS, O’Neill EM, RuoholaBaker H: Pointed, an ETS domain transcription factor, negatively regulates the EGF receptor pathway in Drosophila oogenesis. Development 1996, 122:3745-3754. 16. Hopker VH, Shewan D, Tessier-Lavigne M, Poo M, Holt C: Growthcone attraction to netrin-1 is converted to repulsion by laminin-1. Nature 1999, 401:69-73. 17. Schwarzbauer J: Basement membranes: Putting up the barriers. Curr Biol 1999, 9:R242-R244.
683
Laminin A is required for follicle cell–oocyte signaling that leads to establishment of the anterior–posterior axis in Drosophila Wu-Min Deng and Hannele Ruohola-Baker The establishment of the anterior–posterior (AP) axis in Drosophila melanogaster requires signaling between the oocyte and surrounding somatic follicle cells during oogenesis [1,2]. First, a signal from the oocyte (Gurken α (TGFα α) homolog) (Grk), a transforming growth factor-α is received by predetermined terminal follicle cells in which the epidermal growth factor receptor (EGFR) pathway is activated and a posterior fate is induced [2–4]. Later, the posterior follicle cells send an unidentified signal back to the oocyte, which leads to the reorganization of its cytoskeletal polarity. This reorganization is required for proper localization of maternal determinants, such as oskar (osk) and bicoid (bcd) mRNAs, that determine the AP polarity of the oocyte and the subsequent embryo [2]. We show here that when the gene lanA, which encodes the extracellular matrix component laminin A, is mutated in posterior follicle cells, localization of AP determinants is disrupted in the underlying oocyte. Posterior follicle-cell differentiation and follicle cell apical–basal polarity are unaffected in the lanA mutant cells, suggesting that laminin A is required for correct signaling from the posterior follicle cells that polarizes the oocyte. This is the first evidence that the extracellular matrix is involved in the establishment of a major body axis. Address: Department of Biochemistry, University of Washington, Box 357350, Seattle, Washington 98195-7350, USA. Correspondence: Hannele Ruohola-Baker E-mail: [email protected] Received: 28 January 2000 Revised: 3 April 2000 Accepted: 3 April 2000 Published: 19 May 2000 Current Biology 2000, 10:683–686
localized at the posterior of stage-9 wild-type oocytes, was used as a marker for oocyte polarity in F1 progeny [6]. Thirteen different genes were identified from this screen, three of which encode transmembrane or secreted proteins and are therefore of great interest to us. The gene encoding laminin A (lanA) (Figure 1a) [7], was one of the genes obtained from this screen. In 109-3-9/EP(lanA) egg chambers, Kin–β-Gal is diffused from its normal position, the posterior pole of the stage-9 oocyte (data not shown). To examine whether lanA is expressed in the posterior follicle cells, whole-mount ovarian in situ hybridization was undertaken using lanA cDNA as a probe. lanA expression is first detected in all follicle cells of stage 1–4 egg chambers (Figure 1b, left panel). Starting at around stage 5, expression in the anterior follicle cells is decreased, while the rest of the follicle cells remain strongly stained (Figure 1b, middle panel). At around stage 6, lanA is expressed in follicle cells covering the posterior half of the egg chamber (Figure 1b, right panel). There is no detectable lanA expression in the germline cells. Laminin protein is present in both the basement membrane and the ECM between the posterior follicle cells and the oocyte at stage 6 ([8] and data not shown). As lanA is expressed in the early follicle cells and shows differential staining along the AP axis in the follicle-cell epithelium, we asked whether laminin A has a role in communication between posterior follicle cells and the oocyte. First, we examined the distribution pattern of an oocyte posterior marker, Staufen (Stau) [1], in lanA9–32 (a null allele Figure 1 (a) Exons 2 5′
4
1
0960-9822/00/$ – see front matter © 2000 Elsevier Science Ltd. All rights reserved.
3
6 5
8 10 7
9 11
12
14 13
15
3′
lanA9–32 break points
Results and discussion The signaling pathway by which the posterior follicle cells determine the polarization of the oocyte remains largely unclear [1,2]. In order to identify genes involved in signaling from the posterior follicle cells, we carried out a modular misexpression screen [5]. Over 2000 EP lines, each of which has a Gal4-responsive site randomly inserted in the fly genome, were crossed to a Gal4 driver line, 1093-9, which shows Gal4 expression in most follicle cells, including the posterior cells, starting at stage 6. A fusion protein, kinesin–β-galactosidase (Kin–β-Gal), which is
(b)
st1 Current Biology
st3
st5
st6
The structure of the laminin A gene (lanA) and its expression during early oogenesis. (a) The null allele lanA9–32 used in this study has a deletion between the 5′ untranslated region and the third exon. (b) lanA expression is first detected in all follicle cells from stages 1–3. lanA expression is decreased in the anterior follicle cells during stages 5–6.
684
Current Biology Vol 10 No 11
Figure 2 (a)
(e)
(f)
(j)
(k)
(b)
(c)
(d)
(g)
(h)
(i)
Current Biology
Laminin A function is required in the posterior follicle cells for the proper localization of oocyte anterior–posterior polarity determinants. (a) Stau (red fluorescence) is normally localized at the posterior of the oocyte during stages 9–10 (arrow). Green staining in (a–d) and (g–i) represents green fluorescent protein (GFP), which marks wild-type cells. (b) In a lanA9–32 germline clone, Stau localization is unaffected (arrow). (c) When the lanA mutant follicle-cell clone lies over the anterior of the oocyte, Stau is correctly localized to the posterior pole of the stage 9–10 oocyte (arrow). (d) When the mutant clone covers the posterior end of the oocyte, Stau is mislocalized to the center of the oocyte (arrow). (e) Another posterior marker, oskar (osk) RNA, is normally localized at the posterior pole of the oocyte during stages
8–10 of oogenesis (arrow). (f) In an unmarked lanA mosaic egg chamber, osk RNA is mislocalized to the center of the oocyte (arrow). (g) In a wild-type stage-10 egg chamber, Oskar protein (red fluorescence) is detected at the posterior of the oocyte (arrow). (h) When posterior follicle cells are mutant for lanA, however, Oskar is mislocalized (arrow). (i) Roughly 20 wild-type posterior follicle cells (indicated by the bracket) are sufficient to restore Oskar localization at the posterior of the oocyte (arrow). (j) The anterior determinant bcd RNA is localized to the anterior of the oocyte in a stage-9 wild-type egg chamber (arrowheads). (k) In an unmarked lanA mosaic egg chamber, bcd is mislocalized to the posterior (arrow), in addition to the normal anterior staining (arrowheads).
of lanA [7]) mosaic egg chambers. Stau is normally localized to the posterior pole of the oocyte during stages 9–10 in the wild-type (Figure 2a), and this pattern is not changed in lanA– germline clones (Figure 2b). This is consistent with the observation that lanA expression is not detected in the germline cells. Stau localization is also unaffected in lanA mosaics in which the mutant follicle-cell clones are located over the anterior part of the oocyte (Figure 2c). In 87% (n = 39) of egg chambers, however, in which lanA– folliclecell clones cover the posterior end of the oocyte, Stau localization is no longer detectable at the posterior pole of the oocyte (Figure 2d). Instead, it is mislocalized as a dot at the center of the oocyte or as a streak diffusing away from the posterior of the oocyte (Figure 2d and data not shown). As Stau is required for localization of osk RNA [1], the posterior maternal determinant, we investigated whether osk RNA and Oskar protein (Osk) are also mislocalized in lanA mosaic egg chambers. Indeed, mislocalization and diffusion of osk RNA from the posterior is observed in unmarked lanA mosaic egg chambers (Figure 2f). Anti-Oskar antibody was also used to examine the distribution of this posterior determinant in marked lanA9–32 clones. Similarly to the Stau localization experiments, Osk is mislocalized in 88% (n = 32) of the egg chambers where posterior follicle cells are mutant for lanA (Figure 2h). In order to cause the mislocalization phenotype, a patch of roughly 90 posterior-most
follicle cells must be mutant for lanA. Within a large lanA– follicle-cell clone, however, if 20 of the posterior-most cells are wild type, correct localization of Osk and Stau is restored (Figure 2i). These data indicate that lanA expression in the posterior follicle cells is critical for localization of posterior determinants in the oocyte. As signaling from follicle cell to oocyte also has a role in localizing anterior components [9–11], we examined the position of bcd mRNA in unmarked lanA mosaics. In some stage 8–9 egg chambers, mislocalization of bcd RNA in the posterior was detected (Figure 2k), similarly to egg chambers mutant for the Notch or EGFR pathways [9–11]. Localization of osk and bcd RNAs and Stau protein are microtubule dependent [2]. To determine whether cytoskeletal polarity is normal in lanA mosaics, we examined the localization of two microtubule motor proteins, dynein heavy chain (Dhc64C) and the Kin–β-Gal fusion protein [6,12]. In the wild type, these two motor proteins are localized to the posterior pole of the oocyte during stages 8 and 9 (Figure 3a and data not shown). We observed mislocalization of Dhc64C in 87% (n = 15) of lanA– posterior follicle-cell clones (Figure 3b). Kin–β-Gal is also frequently diffused away from the posterior in lanA posterior follicle-cell mosaics (data not shown). Another marker for the correct arrangement of the oocyte cytoskeleton is the
Brief Communication
685
Figure 4
Figure 3 (a)
(c)
(a)
(b)
(d)
(c)
Current Biology
Oocyte microtubule polarity is disrupted in lanA mosaics. (a) Antibody staining for dynein heavy chain (red) shows its localization at the posterior pole of the oocyte (arrows) in a stage-8 wild-type egg chamber. Green staining in (a–d) represents GFP and marks wild-type germline and follicle cells. (b) When the posterior follicle cells are mutant for lanA, dynein heavy chain is mislocalized to the center of the oocyte (arrow). (c) A stage 10 wild-type egg chamber, in which the oocyte nucleus (arrow) is located at the dorsal anterior corner of the oocyte. (d) In an egg chamber where most follicle cells, including the posterior follicle cells, are mutant for lanA, the oocyte nucleus failed to migrate from the posterior of the oocyte (arrow).
location of the oocyte nucleus (Figure 3c), as its movement from the posterior to the dorsal anterior corner during stages 7–8 reflects the microtubule polarity [3,4,10,11]. We checked whether the oocyte nucleus is correctly localized in lanA– follicle-cell clones and found that in 4% (n = 98) of egg chambers in which the posterior follicle cells are mutant, oocyte nuclei remain in the posterior after stage 8 (Figure 3d). As both microtubule motors and oocyte nuclei were mislocalized in lanA mosaics, we conclude that the defect in laminin A function in posterior follicle cells affects the organization of oocyte cytoskeletal polarity. The actin cytoskeleton is also involved in osk mRNA and Stau localization within the oocyte [13,14]. We therefore examined the overall integrity of the oocyte actin cytoskeleton network. No gross abnormalities were observed in oocytes lying under lanA– posterior follicle-cell clones (data not shown); we cannot, however, rule out the possibility that loss of lanA function produces minor defects in the oocyte actin cytoskeleton. Mutants that disrupt follicle-cell differentiation, such as Notch–, grk– and top–, also disturb the polarity of the oocyte [3,4,9–11]. In Notch mutants, the follicle cells at the termini of the egg chamber do not differentiate properly and no anterior or posterior follicle-cell fates are observed [3,4]. In grk– egg chambers, posterior follicle cells adopt the default anterior fate as a result of the lack of the germline signal. As a result, anterior follicle-cell markers such as enhancer-trap L53b and decapentaplegic
(f)
(b)
(d)
(g)
(e)
(h)
Current Biology
Differentiation of posterior follicle cells appears to be normal in lanA mosaics. (a) In wild-type egg chambers, dpp–lacZ (red) is stained in stretched and centripetal migrating follicle cells (arrowheads), which are located at the anterior of the stage-10 egg chamber. Green in (a–h) is GFP, which marks wild-type cells. In mutants of genes such as gurken, in which posterior follicle-cell differentiation is disrupted, dpp–lacZ was found in posterior follicle cells (data not shown). (b) dpp–lacZ is not, however, expressed in the posterior follicle cells (arrow) in mosaics when lanA is mutated in all follicle cells. (c) Expression of Pointed (red) marks posterior follicle cells in wild-type egg chambers (arrow). (d,e) When lanA is mutated in posterior follicle cells, Pointed expression is normal (arrows). (f–h) Follicle-cell polarity is detected by staining for Armadillo (red). In both wild-type (green) and lanA mosaics (marked by white line brackets), Armadillo is localized at the apical side of the follicle epithelium (arrowheads).
(dpp) are detected in the posterior cells, whereas posterior markers such as pointed and 12A3 are no longer expressed [2,10,11,15]. To investigate whether the involvement of lanA in establishing AP polarity is a secondary consequence of defects in posterior follicle-cell differentiation, both anterior and posterior follicle-cell markers were examined in lanA– follicle-cell clones. We found that dpp, the anterior marker that is normally expressed in the stretched and centripetally migrating follicle cells (Figure 4a), is detected in these anterior cells but not in posterior cells in those lanA mosaics that have posterior folliclecell clones (Figure 4b). On the other hand, the posterior marker pointed, which is normally expressed in a subgroup of posterior follicle cells from stage 6 to later stages (Figure 4c), is still expressed in lanA– posterior follicle-cell clones (Figure 4d,e). As pointed expression requires normal differentiation of the terminal follicle cells, this result suggests that the differentiation of the lanA– terminal follicle cells is normal. Furthermore, as pointed expression is induced by
686
Current Biology Vol 10 No 11
the Grk–EGFR pathway [15], this result indicates that EGFR signaling from the oocyte to the posterior follicle cells is not disrupted by mutation of lanA. In addition, on the basis of a normal expression pattern of the gene kekkon (data not shown), EGFR-induced dorsal–ventral polarity in follicle cells was also normal in lanA– anterior folliclecell clones. Another marker, P1333, which is stained in the majority of columnar follicle cells at around stage 10 of oogenesis, shows no difference in expression in lanA– clones (data not shown). These results indicate that posterior follicle-cell differentiation is normal and that the Notch and EGFR signaling pathways act properly in lanA mosaics. We also investigated whether the polarity of follicle-cell epithelium is normal in lanA mosaics. The Drosophila β-catenin homolog Armadillo and the transmembrane receptor Notch, which are predominantly localized to the apical side of the follicle-cell epithelium, were used as markers for follicle-cell polarity in lanA mosaics. No detectable difference in their distribution is found in lanA– follicle-cell clones (Figure 4f–h and data not shown), suggesting that follicle-cell polarity is normal. As shown above, lanA is required in posterior follicle cells for the localization of maternal determinants and reorganization of cytoskeletal polarity in the oocyte. As posterior follicle-cell differentiation appears normal in lanA mutant follicle-cell clones, we propose that lanA is required for signaling from the posterior follicle cells to the oocyte to establish the oocyte polarity and is therefore the first secreted component identified in this ill-defined pathway. Laminin A could function directly as a ligand or as a helper to provide a platform for signal delivery from the posterior follicle cells. In Xenopus, for example, laminin can convert netrin-mediated attraction into repulsion during growthcone guidance [16]. We tentatively tested the first possibility by investigating whether integrins, the potential receptors for laminin A, are involved in the formation of oocyte polarity. No mislocalization of Osk or Stau was, however, detected in the mosaics of the integrin genes analyzed so far. It will be interesting to test whether Drosophila homologs for other laminin receptors are involved in polarity formation in the oocyte [17]. In summary, our data show for the first time that laminin is a crucial player in the formation of a major body axis, and that this is achieved through its involvement in signaling between two types of cells––somatic epithelial and germline cells. It will be interesting to determine whether laminins are involved in bodyaxis formation in other organisms. Supplementary material Supplementary material including additional methodological detail is available at http://current-biology.com/supmat/supmatin.htm.
Acknowledgements We thank Danny Brower, Ed Giniger, Corey Goodman, Tom Hays, Paul Lasko, Todd Laverty, Ruth Lehmann, Stefan Luschnig, Gerry Rubin, Trudi Schüpbach, Daniel St Johnston, Beat Suter, the Bloomington stock center, and the Developmental Studies Hybridoma Bank for sending us antibodies,
cDNA clones and fly stocks. Mike Tworoger, Karin Fischer, Jennifer Blasi, Susan Kwong and Jeff Lin gave excellent technical support, and we thank the Keck Center for allowing us to use their confocal microscopy facilities. We would also like to thank Ed Giniger for critical reading of our manuscript. This work was supported by grants from the Established Investigatorship Award from the American Heart Association, March of Dimes Birth Defect Foundation, Basil O’Conner Starter Research Grant, National Institutes of Health, and Pew Memorial Trust, Pew Scholarship in the Biomedical Sciences to H.R.-B. W-M.D. was supported by a Benjamin Schultz Fellowship.
References 1. Rongo C, Lehmann R: Regulated synthesis, transport and assembly of the Drosophila germ plasm. Trends Genet 1996, 12:102-109. 2. Deng WM, Bownes M: Patterning and morphogenesis of the follicle cell epithelium during Drosophila oogenesis. Int J Dev Biol 1998, 42:541-552. 3. González-Reyes A, St Johnston D: Patterning of the follicle cell epithelium along the anterior-posterior axis during Drosophila oogenesis. Development 1998, 125:2837-2846. 4. Larkin MK, Deng-M, Tworoger M, Holder K, Clegg N, Ruohola-Baker H: Role of Notch pathway in terminal follicle cell differentiation in Drosophila oogenesis. Dev Genes Evol 1999, 209:301-311. 5. Rφrth P: A modular misexpression screen in Drosophila detecting tissue-specific phenotypes. Proc Natl Acad Sci USA 1996, 93:12418-12422. 6. Clark I, Giniger E, Ruohola-Baker H, Jan LY, Jan YN: Transient posterior localization of a kinesin fusion protein reflects anteroposterior polarity of the Drosophila oocyte. Curr Biol 1994, 4:289-300. 7. Henchcliffe C, Garcia-Alonso L, Tang J, Goodman C S: Genetic analysis of laminin A reveals diverse functions during morphogenesis in Drosophila. Development 1993, 118:325-337. 8. Gutzeit HO, Eberhardt W, Gratwohl E: Laminin and basement membrane-associated microfilaments in wild-type and mutant Drosophila ovarian follicles. J Cell Sci 1991, 100:781-788. 9. Ruohola H, Bremer KA, Baker D, Swedlow JR, Jan LY, Jan YN: Role of neurogenic genes in establishment of follicle cell fate and oocyte polarity during oogenesis in Drosophila. Cell 1991, 66:433-449. 10. González-Reyes A, Elliott H, St Johnston D: Polarization of both major body axes in Drosophila by gurken-torpedo signalling. Nature 1995, 375:654-658. 11. Roth S, Neuman-Silberberg FS, Barcelo G, Schüpbach T: cornichon and the EGF receptor signaling process are necessary for both anterior-posterior and dorsal-ventral pattern formation in Drosophila. Cell 1995, 81:967-978. 12. Li M, McGrail M, Serr M, Hays TS: Drosophila cytoplasmic dynein, a microtubule motor that is asymmetrically localized in the oocyte. J Cell Biol 1994, 126:1475-1494. 13. Erdelyi M, Michon AM, Guichet A, Glotzer JB, Ephrüssi A: Requirement for Drosophila cytoplasmic tropomyosin in oskar mRNA localization. Nature 1995, 377:524-527. 14. Manseau L, Calley J, Phan H: Profilin is required for posterior patterning of the Drosophila oocyte. Development 1996, 122:2109-2116. 15. Morimoto AM, Jordan KC, Tietze K, Britton JS, O’Neill EM, RuoholaBaker H: Pointed, an ETS domain transcription factor, negatively regulates the EGF receptor pathway in Drosophila oogenesis. Development 1996, 122:3745-3754. 16. Hopker VH, Shewan D, Tessier-Lavigne M, Poo M, Holt C: Growthcone attraction to netrin-1 is converted to repulsion by laminin-1. Nature 1999, 401:69-73. 17. Schwarzbauer J: Basement membranes: Putting up the barriers. Curr Biol 1999, 9:R242-R244.