spacer gif spacer gif spacer gif spacer gif spacer gif
 QUICK SEARCH:   [advanced]


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    

doi: 10.1242/10.1242/dev.00199


This Article
Right arrow Summary Freely available
Right arrow Full Text
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Deng, W.-M.
Right arrow Articles by Ruohola-Baker, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Deng, W.-M.
Right arrow Articles by Ruohola-Baker, H.

Dystroglycan is required for polarizing the epithelial cells and the oocyte in Drosophila

Wu-Min Deng1,*, Martina Schneider2,*, Richard Frock1, Casimiro Castillejo-Lopez2,{dagger}, Emily Anne Gaman1, Stefan Baumgartner2 and Hannele Ruohola-Baker1,{ddagger}

1 Department of Biochemistry, Box 357350, University of Washington, Seattle, WA 98195, USA
2 Department of Cell and Molecular Biology, Section of Developmental Biology, Lund University, Sweden
{dagger} Present address: EBC, Jämförande Fysiologi (Comparative Physiology) Norbyv. 18A, 752 36 Uppsala, Sweden



View larger version (46K):

[in a new window]
 
Fig. 1. The Drosophila Dystroglycan gene and protein. (A) Schematic drawing of the Dystroglycan genomic region, deletion alleles and the dsDG RNAi construct. The scale bar refers to P1 clone DS03910. (B) Comparison of DG from human, Drosophila (D.m DG) and C. elegans (C.e, T21B6.1). Through Psi Blast search with D.m DG protein sequence, the following E values were obtained after the third reiteration: human DG, e-134; C.e, T21B6.1, e-112. The dark-green box indicates the region of highest amino acid identity (human DG to D.m DG, 31%; D.m DG to T21B6.1, 20%). The light-green box indicates a duplicated region in D.m DG with 25% amino acid identity to amino acids 492-733 of human DG (broken green line). Blue box, mucin-like domain; gray box, putative transmembrane domain. (C) The last 13 amino acids of the DG C terminus, including the Dystrophin-binding site, are highly conserved between human, mouse and Drosophila. (D) Alignment of the cytoplasmic domains of human DG, mouse DG, Drosophila DG and C.e hypothetical protein T21B6.1 with ClustalW program at EBI. (E) Western blot of 6-20 hour embryonic protein extracts probed with the DG antibody. Five bands at molecular weight of ~79, ~105, ~120, ~200 and >200 kDa are detected in wild-type (OrR, lane 4, marked by red dots) and the heterozygous mutant (lanes 5-7) embryos, respectively. All five bands are missing in homozygous deficiency (Df (2R)JP4 and Df(2R)JP6) embryos (lanes 1 and 2). However, this antibody detects a background band at ~110 kDa. In the homozygous Dg248 embryos (lane 3), only a remnant of the 105 kDa band is detected in addition to the background band. A similar banding pattern is observed in Dg323 embryos (data not shown). An antibody against {alpha}-Tubulin was used as a loading control, while an antibody against ß-Galatosidase was used as a control to examine the purity of the homozygous mutant embryos. (F) A stage 4-5 egg chamber with Dg323 follicle cell clones marked by lack of GFP (green). DG staining (red) is strongly reduced in the clone (broken line, arrow). (G) DG staining (red) is strongly reduced in follicle cells carrying tubP-Gal4/dsDG to target-silence DG expression by RNAi. Blue, DAPI staining in the nuclei.

 


View larger version (50K):

[in a new window]
 
Fig. 2. Dystroglycan function is required for apicobasal polarity in epithelial cells. (A) Schematic drawing of an ovariole and an eye-antennal imaginal disc. The ovariole contains egg chambers at different developmental stages (st). A layer of somatically derived follicle cells (FC), the majority of which have a typical epithelial apicobasal polarity, covers the germline cells (GC). Epithelial cells in the imaginal discs also show apical-basal polarity. (B-D) Mosaic analysis of Dg mutations. GFP in green marks wild-type cells. (E-G') RNAi analysis. (B) In a Dg mutant clone (broken lines in B, Dg248), the apicobasal polarity of the FE is disrupted, as mutant cells form a multi-layer epithelium (white arrow) and also cause discontinuance in the epithelium (yellow arrow) (red, actin; blue, DNA). White arrowheads show the wild-type region. (C,C') In a Dg323 follicle cell clone where follicle cells have not lost their columnar shape yet, an apical marker Dlt (red in C and white in C') is detected at both the apical and basal (white arrow in C and green arrow in C') side. C' is an enlarged view of the mutant clone region (broken yellow line) and vicinity shown in C. (D) In a Dg248 clone in an antennal imaginal disc (inset, the white box indicates where the mutant clone is) Dlt (red) is also mislocalized, expanding from the apical (arrowhead) to lateral side (arrow). (E) Multi-layered FE is also detected in tubPGal4/dsDG flies. (F,F',F'') An apical marker ßH-Spec (red; F shows the wild-type pattern, arrow) is mislocalized to the basal side (arrowheads in F' and F'') in dsDG follicle cells. F'' shows a more severe loss-of-polarity phenotype than does F'. (G,G') A basolateral marker, Dlg (red in G shows the wild-type pattern; arrow), is greatly reduced at the basolateral membrane (arrow in G'). DNA is shown in blue at B and E-G'.

 


View larger version (68K):

[in a new window]
 
Fig. 3. Ectopic expression of Dystroglycan interferes with epithelial cell polarity. (A) A schematic drawing of the UAS-DG construct and a Western blot to detect overexpression of UAS-DG driven by a Gal4 driver. (B) UAS-DGcyto contains the cytoplasmic and the transmembrane domain of DG linked to a FLAG-tag. Western blot analysis of UAS-DGcyto (20kDa) driven by daGal4 using anti-Flag antibody. (C) Drawing of an embryo at stage 13 to show the location of the salivary gland. The apical surface of the salivary epithelium is indicated by green line (in the center of the salivary gland). (D) A Nomarski image of a wild-type embryo at stage 12-13 stained with the Crb antibody, which shows apical staining (arrow). (E) Stage 13 embryo derived from da-GAL4 x UAS-DG cross. No Crb staining in the salivary gland can be detected (arrow). Notice that Crb staining in the pharynx is normal (arrowhead). (F) An enlarged drawing of an embryonic salivary gland at stage 13. The apical surface of the epithelial cells is shown in green. (G) A confocal image of the wild-type salivary gland around stage 13 stained with Crb (green) and DG (red) antibodies. Wild-type DG expression is below detection level. (H,H') Salivary gland of stage 13 embryo expressing UAS-DG driven by elav-GAL4. Notice the strong staining of DG (red, H) and the strong reduction of Crb (green, H; white, H') at the apical membrane. (I) Stage 10 follicle cells that overexpress UAS-DG (marked with GFP, green) accumulate high levels of DG protein (red) both in the apical (arrow) and basal surfaces. By contrast, stage 10 wild-type follicle cells express very low levels of DG (arrowhead). (J,J') Apical localization of ßH-Spec (red in J, white in J') is reduced in DG-overexpressing follicle cells (green; arrow). (K,K') Overexpression of DG in follicle cells also causes reduction (arrow) of the apical localization (arrowhead) of Baz (red in K, white in K'). (L) Summary of the localization of apical markers (red line) in the wild-type (i), Dg mutant epithelial cell clones (ii) and cells that overexpress DG (iii). Apical markers are expanded to the basolateral surface of the epithelium in Dg mutant clones, and their apical localization is substantially reduced because of DG overexpression.

 


View larger version (39K):

[in a new window]
 
Fig. 4. Dystroglycan is required for the establishment of oocyte polarity. (A) Dg germline clones are usually arrested at around stage 6 of oogenesis (arrow, a Dg248 germline clone is smaller than the two neighboring egg chambers). (B,C) Nod-ß-Gal (red), a marker that co-localizes with the MTOC in early oocytes, moves from the anterior of the oocyte (red in B) to the posterior and stays there until stage 6 (arrow in C). (D,E) In Dg248 germline clones, Nod-ß-Gal (red) is frequently mislocalized (arrows). (F) ORB (red), is also localized to the posterior of the oocyte at stages 2-6 (arrow). (G) In a Dg323 germline clone, ORB (red) is mislocalized (surrounds the oocyte nucleus, arrow) or undetectable (not shown). (H,I) DG is required for the enrichment of the actin cytoskeleton in the oocyte (marked by an asterisk). In the wild-type oocyte (H), actin (red; white in inset) is enriched at the posterior of the oocyte (arrows). In a Dg323 germline clone (I), actin (red; white in inset) failed to be enriched at the posterior of the oocyte (arrow), and ring canals accumulate tightly in the anterior of the oocyte (arrowhead). (J) Schematic drawing depicting the anterior-to-posterior migration of MTOC in an early wild-type oocyte. This movement is defective and correlates with defect in posterior enrichment of actin in Dg germline clones (K).

 


View larger version (51K):

[in a new window]
 
Fig. 5. Laminin and basal actin organization in the follicle cell epithelium is disrupted in a Dystroglycan mutant clone. (A) Drawing of a wild-type egg chamber with a basal surface view of three follicle cells. The basal actin fibers (red lines) in these cells are arrayed at a direction perpendicular to the long (AP) axis of the egg chamber from stage 8 to 12. Laminin stripes in the ECM (blue lines) shows the same orientation. (B) Drawing of a Dg follicle cell clone with misoriented basal actin fibers and Laminin stripes. (C) Phalloidin staining shows the basal actin array in the wild-type FE. Three cells are outlined. (D) In a Dg323 mutant egg chamber, the basal actin organization is disrupted. (E,F) The role of DG in basal actin organization is non-cell-autonomous, as cells adjacent to the follicle cell clone (broken outline) also frequently show a disrupted basal actin distribution (arrow in F). In wild-type egg chambers, Laminin is oriented in stripes (surface stripes in G), similar to the basal actin. This Laminin orientation is disrupted in Dg mutant clones (H). The left-hand image in H shows where the mutant follicle cell clone (black area) is located. (I) At stage 10, Laminin is mainly detected at the basal surface (arrowhead) of the wild-type FE (area lacking green). However, overexpression of DG (marked by co-expression of GFP, green) causes accumulation of Laminin ECM to the apical and lateral surfaces (arrow). (J) A model for Dystroglycan function in planar polarity (basal actin organization). In this study, we have shown that the DG function is involved in cell-cell communication, which is essential for basal actin planar polarity. This communication probably involves the cyto-architecture of the Laminin ECM. DG directs the orientation of the Laminin stripes. This information is transmitted into a neighboring cell to coordinate the orientation of the actin fibers. GFP (green) marks the wild-type cells in E and H; Actin is white in C,D,F; Laminin is turquoise in G,H.

 





© The Company of Biologists Ltd 2003