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


spacer gif
     Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents    

First published online 21 September 2005
doi: 10.1242/dev.02040


Development 132, 4587-4598 (2005)
Published by The Company of Biologists 2005


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 Yoshida, S.
Right arrow Articles by Tabata, T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Yoshida, S.
Right arrow Articles by Tabata, T.

DPP signaling controls development of the lamina glia required for retinal axon targeting in the visual system of Drosophila

Shoko Yoshida1,4, Laurent Soustelle3, Angela Giangrande3, Daiki Umetsu1,4, Satoshi Murakami1,4, Tetsuo Yasugi1,4, Takeshi Awasaki2, Kei Ito2, Makoto Sato1,4 and Tetsuya Tabata1,4,*

1 Laboratory of Pattern Formation, Institute of Molecular and Cellular Biosciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-Ku, Tokyo 113-0032, Japan
2 Laboratory of Structural Information, Institute of Molecular and Cellular Biosciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-Ku, Tokyo 113-0032, Japan
3 Institut de Génétique et Biologie Moléculaire et Cellulaire, CNRS/INSERM/ULP, BP 10142, Illkirch, C.U. de Strasbourg 67404, France
4 Graduate Program in Biophysics and Biochemistry, Graduate School of Science, The University of Tokyo, 7-3-1 Hongo, Bunkyo-Ku, Tokyo 113-0033, Japan



View larger version (78K):

[in a new window]
 
Fig. 1. Expression of dpp in the developing visual system of the third instar larvae of Drosophila. Unless otherwise noted, all specimens are viewed from a lateral perspective. (A) Developing visual system of the third instar larvae. R axons (white) and DAC-positive neural cell precursors (green) are labeled. Developing optic ganglia (the lamina, the medulla and the lobula) are shown. (B) Schematic illustration of the Drosophila visual system, including the expression domains of dpp and wg. MF, morphogenetic furrow; OS, optic stalk; lam, lamina; OPC, outer proliferation center; GPC, glial cell precursors. wg is expressed at the posterior-most domain (magenta) and dpp is expressed in the dorsal and ventral margin (green). Differentiating lamina glia migrate from the posterior domain to the lamina target region. (C) Schematic illustration of the developing visual system viewed from the horizontal perspective. Axons from R1-R6 neurons induce the differentiation of lamina neurons from lamina neuron precursors (LPCs) arrested at G1 phase in the lamina furrow (LF). Their growth cones stop between rows of epithelial (Epi) and marginal (Ma) glia. Axons from R7 and R8 proceed to the deeper target region in the medulla. (D-F) Three-dimensional image of the optic lobe in the third instar larvae. Glial cells are labeled with anti-REPO antibody (green); the dpp expression domain is visualized by anti-ß-Galactosidase (ß-GAL) antibody in the presence of dpp-lacZ (magenta). (D) Early third instar; (E,F) progressively older stages. (D) REPO-positive glial cells form the optic stalk (arrowhead). (E) The number of glial cells in the optic stalk is increased and the structure becomes thickened (arrowhead). (F) The optic stalk locates at the center of the lamina target region (arrowhead). During development, dpp-lacZ expression is observed just posterior to the margin of the lamina glia (arrows in E and F). Scale bar: 50 µm.

 


View larger version (75K):

[in a new window]
 
Fig. 2. Medea is required for R axon projection patterning. Mutant clones can be identified by the absence of GFP (green); outlined in white in G-J,L,M. R axons are detected with mAb24B10 (white) and developing lamina neurons with an anti-DAC antibody (magenta) except where noted. (A) Three-dimensional image of a wild-type optic lobe with the characteristic crescent-shaped lamina (lam). (B,C) Confocal image of an optic lobe with Medea4 mutant clones. (B) A clone in the posterior-ventral domain (arrow). R axons (white dots) are not observed in mutant clones. (C) Three-dimensional image of the specimen shown in B. There is no R-axon projection in the posterior-ventral region where the clone is present, and the lamina appears to be truncated (arrows in B and C). (D,E) Confocal image of an optic lobe with Medea8 mutant clones. (D) Confocal image showing a clone in the posterior-dorsal region (arrow). R axon projection is not seen in the mutant clone. (E) Three-dimensional image of the specimen shown in D. The R-axon innervation pattern and the lamina morphology are compromised near the border of the mutant clone (arrow). (F) Confocal image of a wild-type optic lobe with anti-CUT antibody staining (magenta). (G,H) Confocal image of an optic lobe with Medea4 clones in the medulla stained with anti-CUT antibody (magenta). CUT expression is not affected in the mutant clone (shown alone in H). (I,J) Confocal image of an optic lobe with Medea4 clones in the OPC and the lamina. DAC expression is not affected in the mutant clone (shown alone in J). (K) Confocal image of a wild-type optic lobe stained with anti-Cyclin A. (L,M) Confocal image of an optic lobe with Medea4 clones in the OPC, stained with anti-Cyclin A antibody (magenta). No clear defect is observed in Cyclin A expression in the mutant clone (shown alone in M). Scale bar: 50 µm.

 


View larger version (97K):

[in a new window]
 
Fig. 3. Inhibition of DPP signaling causes defects in R axon projection patterns and lamina morphology. (A-D) Three-dimensional images of a wild-type optic lobe. (A) Expression of the omb-Gal4 driver, visualized by UAS-GFP (green). (B) R axons visualized by mAb24B10 (white). (C) Developing lamina neurons visualized by anti-DAC antibody (magenta). (D) Merged image of A and C. (E-G) Three-dimensional images of an optic lobe in which UAS-Dad and UAS-GFP expression are induced by omb-Gal4. (E) Expression pattern of omb-Gal4 visualized with UAS-GFP. (F) R axon projections visualized by mAb24B10; the normal crescent-shaped pattern is disrupted. (G) Lamina neurons visualized by anti-DAC antibody; the normal crescent shape is compromised. (H) Merged image of E and G. (I) Schematic illustration of the visual system viewed from the coronal perspective. Epi glia, epithelial glia; Ma glia, marginal glia; Me glia, medulla glia. (J) Confocal image of the wild-type optic lobe carrying omb-Gal4, UAS-GFP and UAS-GFPnls, viewed from the coronal perspective. R axons are visualized by mAb24B10 (red) and lamina neurons are visualized by anti-DAC antibody (magenta). GFP-positive glial cells are shown alone in J'. The growth cones of R axons form the lamina plexus (arrowheads) between the rows of epithelial and marginal glial cells (two asterisks at the top in J,J'). (K,K') Confocal image of the optic lobe, in which Dad expression is induced by omb-Gal4. Alignment of the growth cones of R axons is irregular and growth cones do not form a clear plexus structure (K, arrowheads). Epithelial, marginal and medulla glia layers are not clearly distinguishable and the shape of the cells is irregular. GFP-positive glial cells are shown alone in K'. Scale bars: 50 µm.

 


View larger version (98K):

[in a new window]
 
Fig. 4. Expression of the gcm-lacZrA87 reporter during development of the optic lobe and its dependence on DPP signaling. (A) Three-dimensional image of the expression of the gcm-lacZrA87 reporter in the visual system (revealed by anti-ß-GAL antibody, magenta), together with REPO expression (revealed by anti-REPO antibody, green). (A') Magnified image of the boxed area of A. Expression of ß-GAL is seen in the glial precursor cells just before the entry into the lamina target region and the onset of REPO expression (arrow in A,A'). (B) Expression of gcm-lacZ (magenta), and omb-Gal4 and UAS-GFPnls. (C) Confocal image of the optic lobe with gcm-lacZ, viewed from the coronal perspective. gcm-lacZ expression is visualized by anti-ß-GAL antibody (green) and glia are shown by anti-REPO antibody (magenta). Expression of ß-GAL can be observed in the lamina region and in the epithelial and marginal glia, but not in the medulla glia (arrows in C). (D) Three-dimensional view showing the expression of gcm-lacZ (green) in lamina neurons, revealed by co-labeling with DAC (anti-DAC antibody, magenta). (E) Three-dimensional image of so1 mutant brain carrying gcm-lacZ, visualized by anti-ß-GAL antibody (green). A significant population of ß-GAL-positive cells is observed. (F) Glial cells visualized by anti-REPO antibody (magenta) in the same specimen shown in E; many ß-GAL-positive cells have migrated and express REPO, However, some REPO-negative (and DAC-negative; H) cells are also present. (G) Confocal image of the coronal view of the same specimen as in E and F. REPO-positive glia fail to form the normal three-layer structure (arrow). (H) Anti-DAC antibody staining of the same specimen shown in E-G. No DAC-positive cells are detected, and thus no R axon innervations, in this specimen. (I) Three-dimensional image of gcm-lacZ expression in the brain from dppd12/dppd6 mutant animals (anti-ß-GAL antibody, green). A severe reduction of ß-GAL-positive cells is observed compared with in the so1 mutant brain (yellow arrow). (J) Glial cells (visualized by anti-REPO antibody, magenta) in the same specimen as in I. Most of the cells expressing gcm-lacZ are REPO-positive glia (yellow arrow). (K) Three-dimensional image of gcm-lacZ expression in a brain mutant for hhbar3 (visualized by anti-ß-GAL antibody, green). gcm-lacZ expressing cells form an almost normal crescent shape; however, expression in the presumptive lamina region is decreased. (L) Anti-DAC antibody staining of the hhbar3 mutant brain (magenta), shown together with gcm-lacZ (green). No DAC-positive cells are detected. (M) Double staining with anti-ß-GAL (green) and anti-REPO (magenta) antibody in the same specimen as in K. Most of the ß-GAL-positive cells are glia; however, REPO- and DAC-negative cells were also present. Scale bar: 50 µm.

 


View larger version (135K):

[in a new window]
 
Fig. 5. Medea is required autonomously for the expression of gcm. (A-C) Confocal image of an optic lobe expressing the gcm-lacZ rA87 reporter and containing Medea8 mutant clones at the posterior-dorsal domain. Reporter expression visualized with anti-ß-GAL antibody (magenta); lamina neurons visualized with anti-DAC antibody (blue). (A) The position of the clone (arrow) and the expression pattern of gcm-lacZ revealed by anti-ß-GAL antibody (magenta) are shown. (B,C) ß-GAL (magenta, B) and DAC (blue, C) expression in the same specimen as in A. Expression of gcm-lacZ and DAC are compromised near the border of the clone (arrows in A-C). (D,E) Confocal image from another specimen. (D) The border of wild-type and Medea-/- cells coincides the border of gcm-expressing cells (arrows in D). (D') Magnified image of the boxed area in D. (E) Double labeling with anti-DAC (blue) and anti-ß-GAL (magenta) antibodies in the same specimen as in D. The migrating ß-GAL-positive cells (arrow) are negative for DAC, indicating these are glial cells. (F) Confocal image of the optic lobe with Medea8 clones inside the lamina target region. A small clone is visible in the lamina plexus region (arrow in the boxed area in F, a magnified image is shown in F'). gcm-lacZ is expressed (anti-ß-GAL, magenta) in the clone inside of the lamina target region (arrow in F'). Scale bar: 50 µm.

 


View larger version (68K):

[in a new window]
 
Fig. 6. Activation of DPP signaling in the posterior domain results in the ectopic induction of gcm and REPO-positive glial cells. (A-C) Three-dimensional image of a wild-type optic lobe carrying the omb-Gal4 driver, with UAS-GFPnls (green, A) and gcm-lacZ expression visualized using anti-ß-GAL antibody (magenta, B). (C) Merged image of A and B. (D) Confocal image of the wild-type optic lobe showing gcm-lacZ (anti-ß-GAL, blue), UAS-GFPnls driven by omb-Gal4 (green), and REPO expression (anti-REPO, red). (E-G) Three-dimensional image of the optic lobe carrying the omb-Gal4 driver, with UAS-GFPnls (green, E), and gcm-lacZ visualized by anti-ß-GAL antibody (magenta, F). (G) Merged image of E and F. An ectopic cluster of ß-GAL-expressing cells is seen in the posterior domain (arrow in F). (H) Confocal image of the optic lobe with tkvQ253D expression by omb-Gal4. REPO expression (anti-REPO, red), gcm-lacZ (anti-ß-GAL, blue), and UAS-GFP driven by omb-Gal4 (green) are shown. REPO-positive glial cells are visible inside of the domains that ectopically express gcm-lacZ (arrows). Scale bar: 50 µm.

 


View larger version (89K):

[in a new window]
 
Fig. 7. gcm controls differentiation of the lamina glia and lamina neurons. (A-C) Three-dimensional image of a wild-type optic lobe. (A) The expression pattern of omb-Gal4 visualized by UAS-GFPnls (green). (B) R axons are revealed by mAb24B10 (white). (C) The lamina neurons are visualized by anti-DAC antibody (magenta). (D) Confocal image of the lamina target region in the wild-type optic lobe. Lamina neurons are visualized with anti-DAC (magenta), mature glial cells in the target region are visualized with anti-REPO (red) and omb-Gal4 expression is visualized by UAS-GFPnls (green). Mature glial cells are doubly positive for REPO and GFP, shown in yellow. (E-H) Three-dimensional image of the optic lobe with expression of UAS-glideDN driven by omb-Gal4. (E) Expression of the driver is visualized with UAS-GFPnls (green). (F) mAb24B10 staining in the same specimen (white); the R-axon projection pattern is compromised. (G) Anti-DAC staining in the same specimen (magenta). The lamina is smaller than in wild type and the morphology is defective. (H) Confocal image showing the lamina target region in the optic lobe expressing glideDN with the omb-Gal4 driver. Lamina neurons, glial cells and omb-expressing cells are visualized as in D. There is a smaller number of glial cells in the target region than in wild type. (I) Three-dimensional image of the wild-type optic lobe, in which the expression pattern of NP6099-Gal4 is visualized by UAS-GFPnls. (J) Anti-DAC staining of the same specimen as in I. (K) Merged image of I and J, showing that Gal4 expression coincides with the developing lamina neurons. (L) Three-dimensional image of the optic lobe where UAS-glideDN expression is induced in the developing lamina neurons by NP6099-Gal4. A reduction of DAC-positive cells is observed (arrow). (M) The same specimen as in L, shown with expression of the driver (green). (N) mAb24B10 staining (white) in the same specimen as in L and M. R axons project to the presumptive lamina target region with an irregular pattern. Scale bar: 50 µm.

 


View larger version (22K):

[in a new window]
 
Fig. 8. Model showing how DPP signaling controls differentiation of the lamina glia. (A) dpp expression (green) is induced by wg (magenta) in the lamina glia progenitor domains, then dpp induces the expression of gcm in GPCs (purple), triggering the lamina glia (blue) differentiation program. (B) As glia migrate, they contact R axons, leading to the formation of the lamina plexus by the R1-R6 growth cones. Signals from R axons facilitate the migration and maturation of the lamina glia, leading to formation of the correctly layered structure. (C) In the absence of DPP signaling, gcm expression is not induced and lamina glia fail to differentiate. R axons fail to find their intermediate partner and mis-target, resulting in an irregular induction of lamina neurons and thus the characteristic crescent-shaped lamina fails to form.

 





© The Company of Biologists Ltd 2005