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

First published online 22 November 2006
doi: 10.1242/dev.02653

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development 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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Garbe, D. S. Articles by Bashaw, G. J. Search for Related Content PubMed PubMed Citation Articles by Garbe, D. S. Articles by Bashaw, G. J.

ß-Spectrin functions independently of Ankyrin to regulate the establishment and maintenance of axon connections in the Drosophila embryonic CNS

¹ Department of Neuroscience, University of Pennsylvania School of Medicine, 421 Curie Boulevard, Philadelphia, PA 19104, USA.
² Program in Cell and Developmental Biology, and Department of Biological Sciences, University of Illinois Chicago, Chicago, IL 60607, USA.

View larger version (111K): [in this window] [in a new window]   Fig. 1. ß-Spectrin is required in neurons for midline axon guidance. (A, top) Expanded 5' region of the ß-spectrin genomic locus showing the location of each p-element insert. (Bottom) 1, Wild-type ß-Spectrin (ßSpecWT) protein including functional domains (ABD, actin-binding domain; ank, Ankyrin-binding domain; PH, pleckstrin homology domain); 2, ßSpecank replaces segment 16 (the 15th repeat) of ß-Spectrin with segment 13 of -Spectrin; 3, ßSpecPH places a stop codon at the beginning of the PH domain in ß-Spectrin. (B-D) Stage 16 embryos stained with monoclonal antibody (Mab) BP102 to reveal the CNS axon scaffold. Anterior is up. (B) ß-spectrin (em6) heterozygous embryos exhibit a wild-type ladder-like CNS architecture. (C) ß-spectrin (em6) hemizygous embryos display fused anterior and posterior commissures reflecting a reduction in midline repulsion. (D) Rescue of ß-spectrin (em6) phenotype by expressing wild-type ß-Spectrin in all post-mitotic neurons with elavGAL4. (E-H) Late stage 16 embryos stained with anti-Fas2 Mab to reveal longitudinal axon pathways. Anterior is up. (E) ß-spectrin heterozygous embryos have no Fas2-positive axon bundles crossing the midline. (F) ß-spectrin (em6) hemizygous embryo. Medial ectopic Fas2-positive bundles cross the midline (arrows) and are closer together [distance between medial fasicles: wild-type, 13.7±1.37 µm; ßspec(G0108), 8.9±1.27 µm; compare E and F, dashed line]. Lateral Fas2-positive bundles contain longitudinal breaks (arrowheads). (G) ß-spectrin mutant embryo expressing full-length wild-type ß-Spectrin in all neurons using elavGAL4. Note complete rescue of medial and lateral longitudinal axon guidance defects. (H) ß-spectrin em6 embryo expressing full-length wild-type ß-Spectrin in midline glia using simGAL4. ß-spectrin mutant phenotypes in the longitudinal pathways (arrows and arrowheads) are still observed. (I) Quantification of ectopic crosses per embryo in genotypes: a, em6/+ (n=15); b, em6/Y (n=18); c, em6/Y,UAS-ßSpec/+,elavGAL4/+ (n=15); d, em6/Y,UAS-ßSpec/+,simGAL4/+ (n=11); e, em21/Y (truncation) (n=18); f, strong p-element allele G0108/Y (n=14); g, presumptive hypomorphic p-element allele G0074/Y (n=13); h, presumptive hypomorphic p-element allele G0198/Y (n=14). Asterisk denotes significant difference in genotypes b and a, and b and c (P=8.027x10-7; two-sample Student's t-test). There is no significant difference between genotypes b and d. Error bars indicate s.e.m. (J,K) Stage 15 embryos stained with polyclonal anti-ß-Spectrin antibody. Anterior is up. (J) Wild-type ß-Spectrin protein localizes to the plasma membrane around every cell of the CNS neuropil and to the axon scaffold (brackets). (K) ß-Spectrin protein levels are severely reduced in ß-spectrin (em6) mutant embryos. (L) ß-spectrin (G0108) mutant embryo stained with an anti-wrapper antibody showing the appropriate number and location of the midline glia.

View larger version (134K): [in this window] [in a new window]   Fig. 2. -spectrin weakly contributes to midline axon guidance. (A-E) Late stage 16 embryos stained with anti-Fas2 Mab. Anterior is up. (A) Heteroallelic -spectrin (lm88/rg41) mutant embryos on average contain less than one Fas2-positive ectopic crossover (arrow). (B) -spectrin (Im88) heterozygous embryos do not exhibit midline guidance errors. (C) Similar to the heteroallelic combination, -spectrin (Im88) homozygous embryos display mild midline guidance defects. (D) Removing one copy of -spectrin (Im88) mildly enhances the ß-spectrin (G0198) hypomorphic phenotype. (E) Stronger genetic interactions are observed when two copies of -spectrin are removed in ß-spectrin hypomorphic mutants. Note the increase in Fas2-positive longitudinal axon bundles crossing the midline. (F) Quantification of single and double mutants. a, -spectrin(rg41)/ -spectrin(lm88) (n=19); b, lm88/TM3ß (n=7); c, lm88/lm88 (n=11); d, ß-spectrin(G0198)/Y (n=14); e, G0198/Y; lm88/TM3ß (n=6); f, G0198/Y; lm88/lm88 (n=8). Asterisk denotes a significant difference between genotypes f and c, and f and d (P=2.86x10-5 and 0.000154 respectively; two-sample Student's t-test). Error bars indicate s.e.m.

View larger version (127K): [in this window] [in a new window]   Fig. 3. Co-dependence of - and ß-Spectrin for proper protein localization. (A-I) Stage 15 embryos stained with polyclonal anti-ß-Spectrin antibody (A,D,G) or with monoclonal anti--Spectrin antibody (B,E,H). A merged composite image of - and ß-Spectrin staining for each genotype is shown in C, F and I. (A-C) An -spectrin heterozygous embryo. (A) ß-Spectrin protein localizes to the plasma membrane surrounding all CNS cells and to the axon scaffold (white brackets). (B) -Spectrin shows similar localization. (C) A merged image (yellow indicates colocalization). (D-F) A ß-spectrin (em6) mutant embryo. (D) ß-spectrin mutant embryos have significantly reduced ß-Spectrin protein levels. (E) -Spectrin protein levels are also reduced and are almost undetectable when confocal settings identical to those used for the image in B are applied, suggesting that ß-Spectrin is required to maintain normal levels of -Spectrin. When the Photo Multiplier Tube (PMT) gain is increased, low levels of -Spectrin can be seen at the plasma membrane (starred arrowheads). (F) A merged image showing no colocalization. (G-I) An -spectrin mutant embryo. (G) In -spectrin mutants, ß-Spectrin is redistributed to axons (white brackets). (H) -spectrin mutants have low levels of -Spectrin. When PMT gain is increased, residual -Spectrin can be seen at the plasma membrane (arrowheads) but not on axons (arrows). (I) A merged image of - and ß-Spectrin localization in an -spectrin mutant. Arrowheads depict cells in which - and ß-Spectrin colocalize (yellow). For all panels, the genotypes are listed on the left and antibodies are listed on top.

View larger version (91K): [in this window] [in a new window]   Fig. 4. The Ankyrin-binding and PH domains are not required for axon guidance. (A-E) Late stage 16 embryos stained with the anti-Fas2 Mab. Anterior is up. (A) ß-spectrin heterozygous embryos have no Fas2-positive longitudinal bundles crossing the midline. (B) ß-spectrin (em6) hemizygous embryos display Fas2-positive axons crossing the midline (arrows). (C) ß-spectrin null hemizygous embryos ubiquitously expressing full-length ß-Spectrin are rescued for both the medial ectopic crossing defect and lateral longitudinal breaks. (D) Ubiquitously expressing a form of ß-Spectrin lacking the Ankyrin-binding domain (ßSpecank) also rescues the midline axon guidance defects seen in ß-spectrin mutant embryos. (E) Ubiquitously expressing a form of ß-Spectrin lacking the PH domain (ßSpecPH) also rescues ß-spectrin mutant midline guidance errors. (F) Quantification of ectopic Fas2-positive midline crossovers in ß-spectrin heterozygous, hemizygous mutant, and transgenic rescued backgrounds. a, em6/+ (n=15); b, em6/Y (n=13); c, em6/Y+Ub-ßSpecWT (n=9); d, em6/Y+Ub-ßSpecank (n=15); e, em6/Y + Ub-ß-SpecPH (n=20). Asterisk denotes a significant difference between genotype b and genotypes a, c, d and e (P=1.22x10-6; two-sample Student's t-test). Error bars indicate s.e.m. (G-M) Same embryos as in B-E, stained with the polyclonal anti-Myc antibody. Anti-Myc stainings were performed at the same time and images were taken at the same confocal settings. (G) Sibling ß-spectrin mutant embryos not expressing a transgene do not stain with the Myc antibody. Ubiquitously expressed wild-type ß-Spectrin localizes to the axons (H) and to the plasma membrane at sites of cell contact (K). Ubiquitously expressed ßSpecank localizes similarly to full-length transgenic protein to the axon scaffold (I) and to the plasma membrane (L). Ubiquitously expressed ßSpecPH localizes to axons (J), although at reduced levels, but is no longer localized to the plasma membrane nor at sites of cell contact (M).

View larger version (94K): [in this window] [in a new window]   Fig. 5. Mutations in ß-spectrin genetically interact with the Slit-Robo pathway. All embryos were dissected at late stage 16 and were stained with anti-Fas2 Mab. Anterior is up. In all panels, arrows indicate sites of medial axons ectopically crossing the midline and arrowheads point to lateral longitudinal bundle breaks. (A) ß-spectrin null embryos exhibit ectopic Fas2-postitive axons crossing the midline and lateral longitudinal breaks. (B) slit, robo/+; elavGal4/+ (SRE) heterozygous embryos contain approximately two to three Fas2-positive bundles crossing the midline. (C) Removing ß-spectrin dramatically and specifically enhances the SRE defect. Note the effect is synergistic and not simply additive. (D) The genetic interaction seen in C can be suppressed by expressing wild-type UAS-ß-Spectrin in all postmitotic neurons with elavGal4. Note that approximately two to three ectopic crosses (arrows) are still observed as a result of the SRE background; however, the distance between medial longitudinal bundles has widened and the lateral breaks have disappeared. (E) ß-spectrin hypomorphic embryos have mild axon guidance defects. In this embryo, one medial longitudinal bundle crosses the midline (arrow) and lateral breaks are still observed (arrowheads). (F) Similar to the null allele, hypomorphic ß-spectrin alleles also show specific dose-dependent genetic interactions in a slit,robo/+ heterozygous background. (G) Quantification of single mutant phenotypes and genetic interactions. a, em6/+ (n=15); b, em6/Y (n=18); c, slit, robo/+; elavGal4/+ (n=19); d, em6/Y; slit,robo/+; elavGal4/+ (n=9); e, em6/Y; slit,robo/UAS-ßSpec; elavGal4/+ (n=23); f, G0198/Y (n=14); g, G0198/Y; slit,robo/+ (n=10); h, G0074/Y (n=13); i, G0074/Y; slit,robo/+ (n=11). Asterisk denotes a significant difference between genotypes d and b, and d and e (P=2.71x10-8 and 1.13x10-11, respectively; two-sample Student's t-test). Double asterisk denotes a significant difference between genotypes g and f (P=1.73x10-6; two-sample Student's t-test). Triple asterisk denotes a significant difference between genotypes i and h (P=2.60x10-6; two-sample Student's t-test). There is no significant difference between genotypes c and e. Error bars indicate s.e.m.

View larger version (137K): [in this window] [in a new window]   Fig. 6. The level and localization of Robo are not altered in ß-spectrin null embryos. (A,D) Embryos stained with anti-HRP. (B,E) Embryos stained with anti-Robo. (C,F) Merged image for each genotype. ß-spectrin mutant embryos (D-F) do not show reduction or mislocalization of Robo, although the distance between opposing sides of the midline is reduced when compared with wild-type embryos (A-C). Embryos were processed in parallel and images were taken at the same confocal settings.

View larger version (108K): [in this window] [in a new window]   Fig. 7. ß-Spectrin affects the guidance of the Apterous (Ap) neurons. (A-E) All embryos have AptGal4, UAS-TauMycGFP in the background. (F,G) Embryos have AptGal4; UAS-TauGFP in the background. (A) Wild-type embryos have no `thick' bundles of Ap neurons crossing the midline. (B) slit/+ heterozygous embryos have less than one thick bundle of Ap neurons crossing the midline (arrow). (C) ß-spectrin mutant embryos have approximately one to two thick Ap bundles crossing the midline (arrow). (D) Ap neuron guidance defects in a slit/+ heterozygous background are enhanced by removing ß-spectrin (arrows). (E) The genetic interaction seen in D cannot be rescued by expressing UAS-ß-Spectrin in the Ap neurons (arrows). (F) In robo mutants, Ap neurons in every segment cross and recross the midline (starred arrowheads). (G) Unlike what is observed in the ß-spectrin background, Ap neuron defects seen in robo1 mutants can be rescued by expressing UAS-RoboMyc specifically in the Ap neurons themselves (arrowheads). (H) Quantification of genotypes: a, AptTMG/+ (n=12); b, slit, AptTMG/+ (n=14); c, em6/Y; AptTMG (n=10); d, em6/Y; slit, AptTMG/+ (n=16); e, em6/Y; slit, AptTMG/UAS-ßSpec (n=20); f, G0198/Y; AptTMG/+ (n=14); g, G0198/Y; slit, AptTMG/+ (n=13). Asterisk denotes a significant difference between genotypes d and c (P=0.0144; two-sample Student's t-test). There is no significant difference between genotypes d and e. Double asterisk denotes a significant difference between genotypes g and f (P=0.00139; two-sample Student's t-test). AptTMG, ApterousGal4, UAS-TauMycGFP.

View larger version (87K): [in this window] [in a new window]   Fig. 8. ß-Spectrin contributes to the maintenance of axonal connections. All embryos were stained with anti-Fas2 and anti-GFP, and contain aptGAL4 and UAS-TauMyc GFP in the background except for D-F; these embryos contain UAS-TauGFP. Anterior is up in all panels. Embryos of each genotype (left) were fixed, stained and dissected at the indicated stages (top). (A-C) yw embryos. (A) At stage 14, the Ap neurons in each segment approach the midline and make a stereotypical turn anteriorly (arrowhead). (B) At stage 15, neurons extend growth cones further anteriorly almost reaching the next segment. (C) At late stage 16, axons make one continuous bundle spanning the length of the embryo. Note that in wild-type embryos, the Ap neurons remain ipsilateral. (D-F) robo mutant embryos. In robo mutant embryos at stage 14 (D), Ap neurons do not respect the midline boundary and cross over to the other side (starred arrowhead). Later in development, axons continue to recross the midline as they extend anteriorly (E,F). (G-I) ß-spectrinem6/Y hemizygous embryos. (G) Similar to wild-type embryos at stage 14, the Ap neurons in ß-spectrin mutants approach the midline, make an anterior turn, and do not cross the midline (arrowhead). Interestingly, in some segments, the Ap neurons also begin to extend processes slightly posteriorly (arrows). (H) At stage 15, most of the Ap neurons still respect the midline boundary, although we occasionally saw a stray axon crossing over (starred arrowhead). (I) Later in development, the Ap neurons lose sensitivity to the midline, fail to maintain their appropriate ipsilateral connections and cross the midline (starred arrowheads). Additionally, in some segments, we noticed that some axons do not extend all the way to the next segment (feathered arrowhead). (J-L) ß-spectrinem6 hemizygous plus slit heterozygous embryos. (J) Even when one copy of slit is removed in a ß-spectrin null background, stage 14 Ap neurons still do not cross the midline but begin to extend anteriorly. (K) By stage 15, some axons already lose sensitivity to the midline (starred arrowheads). (L) By late stage 16, Ap neurons in many segments now cross the midline (starred arrowheads).