QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online October 12, 2007
doi: 10.1242/10.1242/dev.02882

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Furrer, M.-P. Articles by Chiba, A. Search for Related Content PubMed PubMed Citation Articles by Furrer, M.-P. Articles by Chiba, A.

Slit and Robo control the development of dendrites in Drosophila CNS

Marie-Pierre Furrer, Irina Vasenkova, Daichi Kamiyama^*, Yaira Rosado and Akira Chiba^*,

Department of Cell and Developmental Biology, University of Illinois, Urbana, IL 61801, USA.

View larger version (46K): [in this window] [in a new window]   Fig. 1. Slit proteolysis during embryogenesis. (A,B) Slit proteolysis during embryogenesis. (A) Developmental western blot of whole embryo lysate probed with Slit antibody that recognizes both the full-length and carboxyl-terminus fragment, approximately 190 and 50 kDa, respectively. [Note: a previously unreported band of approximately 120 kDa was detected in tissues from early stages. A similar band was also observed in zygotic slit2/slit2 null tissues (data not shown). This 120 kDa band is thought to represent either an unknown protein that the Slit antibody recognizes or maternally supplied Slit protein of a previously unreported, truncated form. In the latter case, considering the distance between the Robo-binding domain and the antibody epitope, the protein is unlikely to activate Robo.] (B) Relative concentration of product of Slit proteolysis, i.e. [50 kDa band]/([190 kDa band]+[50 kDa band]). (C,D) Standard immunocytochemistry of wild-type (+/+) embryonic CNS with Slit (C) and Pak (D) antibodies at hour 14. As in other experiments in this study, the embryos were fillet-dissected and fixed. However, from there on they were subjected to 1 mM Triton X-100 throughout the reminder of the process. Embryos were counter-stained with HRP antibody (green). With detergent treatment, Slit can be detected abundantly in its source, the glia, with additional low-level signals outside the source (C, purple). Anti-HRP recognizes extracellular domains of neuronal cell surface proteins and labels axons in both longitudinal connectives (lc) and commissures (ac and pc; C-D, green). However, the cytoplasmic molecule Pak is detected within the entire CNS, including the longitudinal connectives (lc), commissures (ac and pc) and neuronal cell bodies (cb) (D, blue on right). (E,F) Detergent-free immunocytochemistry of wild-type (+/+) embryonic CNS with Slit (E) and Pak (F) antibodies at hour 14. Without detergent treatment, Slit is detected mainly along the longitudinal connectives (lc) and commissures (ac and pc) (E, purple on right; also see Fig. 5). This is a very different pattern from the Slit staining using standard immunocytochemistry (compare with C, purple). By contrast, Pak antibody detects very little Pak protein in either axons or neuronal cell bodies (F, blue compare with C, blue). These observations support the idea that the detergent-free method used in this study detects the pool of Slit protein that exists in extracellular space and excludes the pool that is intracellular. Scale bar: 5 µm.

View larger version (27K): [in this window] [in a new window]   Fig. 2. Quantification of Slit immunocytochemistry and the morphology of aCC. (A) Cross-sectional and dorsal views of CNS at early (left) and late (right) stages. The dimensions of the quantified volume (box) are: width (W), 30.0 µm from the midline: depth (D), 5.3±0.4, 5.2±0.2 and 7.7±0.3 µm between the dorsal and ventral sides of the midline glia cluster at hours 9, 14 and 17, respectively; and length (L) 11.3±0.3 µm from the anterior to posterior ends of the midline glia cluster at hour 9, and 4.5±0.2 and 4.7±0.1 µm from the anterior to the posterior ends of the posterior commissure at hours 14 and 17. (B) The z-stack images (1) are summed into a single 2D image (2) and then consolidated into a single line (3) (see Materials and methods). (C,D) DiI-labeled aCC (red) in the live wild-type embryo stained with anti-HRP antibody (green, C), from which the tracing of the aCC is obtained (D; see Fig. 7b bottom). Scale bar: 5 µm.

View larger version (20K): [in this window] [in a new window]   Fig. 3. aCC develops dendrites as collateral processes in the embryonic CNS. (A) Schematic of aCC motoneuron in a late-stage embryo. The aCC cell body (aCC) is present near the dorsal midline at each hemisegment in the Drosophila embryonic CNS (CNS). The aCC extends a single axon (axon) laterally that cuts across the longitudinal connective, exits the CNS, grows past proximal muscles including muscle-12, then terminates at its target, muscle-1. As its axon approaches the target, aCC begins to develop ipsilateral dendrites (dendrite) as collateral processes from the CNS portion of the axon. A medially directed dendritic growth cone (gc) extends across the midline, which later develops into contralateral dendrites. A neuropil (neuropil) emerges within the longitudinal connective. (B,C) Tracings of DiI-labeled aCC motoneuron in wild-type embryos at hours 14 (B) and 17 (C). A dotted circle surrounds the main dendrite, and an arrowhead points to the dendritic growth cone that crosses the midline. A star indicates where distal portion of the axon is cropped. Behind the aCC tracing, a shaded region indicates where there is strong anti-HRP staining (see, for example, Fig. 5C left). The width of neuropil is measured as the mean distance between the edges of the HRP-positive longitudinal connective at a point between the centers of anterior and posterior commissures (ac and pc). (D,E) Histogram of the number of aCC dendritic tips (mean±s.e.m. number per 4 µm) at hours 14 (D, n=6 neurons) and 17 (E, n=14) along the medial-lateral axis. Arrow indicates the mean position. Scale bar: 5 µm.

View larger version (14K): [in this window] [in a new window]   Fig. 4. Robo is required cell-autonomously in the aCC during dendrogenesis. (A,B) Tracings of robo4/robo4 (A) and robo1/robo4 aCCs with cell-specific genetic rescue (robo4/robo1;eve'-GAL4RN2/UAS-roboWT: B) at hour 17. (C) Relative sizes of aCC dendrites [measured as mean±s.e.m. number of dendritic tips per neuron, with the wild-type (+/+; value of 15.1) as 100%] in +/+ (n=14), cell-specific RNAi (n=15), robo4/robo4 and robo1/robo4 (n=14), rescue (n=7), robo4/+ (n=6), 1x overexpression (eve'-GAL4RN2/UAS-roboWT; n=11), and 2x overexpression (eve'-GAL4RN2/UAS-roboWT, UAS-roboWT; n=6). Asterisks indicate P<0.01 by two-tailed t-test against wild type. Scale bar: 5 µm.

View larger version (58K): [in this window] [in a new window]   Fig. 5. Slit topography spatiotemporally prefigures dendrogenesis. (A-D) Detergent-free immunocytochemistry with anti-HRP, which labels neuronal surface (green), and anti-Slit (purple) in wild-type (+/+) at hours 9 (A), 14 (B) and 17 (C), and in Slit null (slit2/slit2) embryos at hour 14 (D). Single segments are shown as projected images. The width of CNS is shown by the horizontal line beneath with gray bars indicating the width and position of bilateral neuropils. Boxes outline the region in which Slit density was quantified. (E,F) Slit concentration topography along the medial-lateral axis (mean±s.e.m.) at hour 9 (n=13 half-segments; E), and at hours 14 and 17 (n=15 and 13; F). The slit/slit tissues (n=13 and 14 for hours 9 and 14, respectively) provide the baselines. Slit densities are normalized to the midline level in the wild type at hour 9 (E) or 14 (F). The horizontal lines shows the peak of neuropilar Slit accumulation, and arrows indicate its center. Scale bar: 5 µm.

View larger version (36K): [in this window] [in a new window]   Fig. 6. Slit topography correlates to aCC dendrogenesis in robo/robo mutant. (A) Detergent-free immunocytochemistry with anti-HRP (green) and anti-Slit (purple) in robo4/robo4 embryos at hour 14. (B) Slit concentration topography (neuropilar portion, normalized to the midline level in +/+ at hour 14; see Fig. 3F) in wild type (+/+) (n=12) and robo4/robo4 (n=12) embryos at hour 14. Arrows indicate the center of the neuropilar concentration. (C) Scattergraphs (up to 10 aCCs per genotype) of dendritic tips in +/+ (n=14) and robo4/robo4 (n=14) at hour 17. Each horizontal line indicates an individual aCC neurons and circles indicate the positions of dendritic tips along its axons. Scale bar: 5 µm.

View larger version (12K): [in this window] [in a new window]   Fig. 7. Loss of Slit leads to loss of aCC dendrites in slit/slit mutant. (A,B) Tracings of wild-type (+/+; A) and slit2/slit2 (B) aCCs at hour 17. (C) Relative sizes of +/+ (n=14) and slit2/slit2 (n=9) aCC dendrites at hour 17. Scale bar: 5 µm.

View larger version (27K): [in this window] [in a new window]   Fig. 8. In comm/comm mutants neuropils and aCC dendrites shift laterally in parallel. (A) Detergent-free immunocytochemistry with anti-HRP (green) and anti-Slit (purple) in comm5/comm5 embryos at hour 14. (B) comm5/comm5 aCC at hour 17. (C) Slit concentration topography in wild-type (+/+; n=4) and comm5/comm5 (n=9) embryos at hour 14. (D) Scattergraphs (10 aCCs per genotype) of dendritic tips in +/+ and comm5/comm5 embryos at hour 17. Scale bar: 5 µm.

View larger version (51K): [in this window] [in a new window]   Fig. 9. Virtually wild-type neuropil but reduced Slit leads to small aCC dendrites in slit/+ and Sdc/Sdc mutants. (A,B) Detergent-free immunocytochemistry with anti-HRP (green) and anti-Slit (purple) in slit2/+ (A) and Sdc10608/Sdc10608 (B) embryos at hour 14. (C,D) Tracing s of slit2/+ (C) and Sdc10608/Sdc10608 (D) aCCs at hour 17. (E) Slit concentration topography in +/+ (n=12), slit2/+ (n=5), Sdc10608/Sdc10608 (n=12) and slit2/slit2 (n=3) at hour 14. Arrows and horizontal lines indicate the maximum numbers of dendritic tips per bin. (F) Histogram of the number of dendritic tips in +/+ (n=14), slit2/+ (n=12), Sdc10608/Sdc10608 (n=14) and slit2/slit2 (n=9) aCCs at hour 17. Arrows and horizontal lines indicate the neuropilar peaks. Scale bar: 5 µm.

View larger version (41K): [in this window] [in a new window]   Fig. 10. Ectopic Slit is not sufficient to induce dendrogenesis outside the CNS. (A,B) Detergent-free immunocytochemistry with anti-HRP (green) and anti-Slit (purple) in embryo with Slit overexpression in muscle-12 (GAL4M12/UAS-slitWT) at hour 16. Box in A indicates the region shown in B. Even when faced with ectopic Slit, the aCC and other axons in the ISN do not produce ectopic collateral processes (circle). However, the RP5 axon fails to innervate its target, muscle-12 (arrow). Muscle-12 extends Slit-positive myopodia to contact the RP5 axon. Scale bar: 5 µm.