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

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 HighWire Citing Articles via Google Scholar Google Scholar Articles by Sharrow, M. Articles by Tiemeyer, M. Search for Related Content PubMed PubMed Citation Articles by Sharrow, M. Articles by Tiemeyer, M.

Gliolectin-mediated carbohydrate binding at the Drosophila midline ensures the fidelity of axon pathfinding

Department of Cell Biology, Yale University School of Medicine, 333 Cedar St, New Haven, CT 06510, USA

View larger version (83K): [in a new window]   Fig. 1. Gliolectin is expressed in the embryonic Drosophila nerve cord during commissural and longitudinal pathfinding. Expression of gliolectin in the midline glia at early stage 13. (A) In wild-type embryos, the midline glia stain with the anti-Gliolectin monoclonal antibody 1B7 (arrowhead, ventral view of a whole mount embryo). (B) A wild-type embryo dissected after staining with 1B7 (black) and BP102 (brown) demonstrates the spatial and temporal relation between Gliolectin expression and axon scaffold formation at early stage 13. (C) In 3013 homozygotes, Gliolectin expression is completely absent at the midline by 1B7 antibody staining (arrowhead). Scale bar: 64 µm in A; 25 µm in B; 16 µm in C.

View larger version (98K): [in a new window]   Fig. 2. Loss of Gliolectin perturbs commissure formation. BP102 staining of embryos from late stage 12 through late stage 13 (A,F, stage 12/1; B,G, stage early 13; C,H, stage mid 13; D,I, stage late 13; E,J, stage early 14) shows the progressive development of distinct, well separated commissures in wild type (OreR, A-E) and the effects of loss of Gliolectin in 3013 homozygotes (-/-, F-J). In OreR embryos at stage 12/1, commissural axons (A, arrowhead) are diffusely spread among the midline glia but by early stage 13 (B, arrowhead) commissural separation is becoming apparent. In 3013 homozygotes, however, the commissures are tightly fasciculated into a single bundle in late stage 12 and early stage 13 embryos (F,G, arrowhead) and remain poorly separated into mid-stage 13 (H, arrowhead). By mid-stage 13, wild-type commissures (C, arrowhead) have separated into anterior and posterior bundles and the longitudinal pathways (arrow) are evident within each segment. In 3013 homozygotes at mid-stage 13, the forming longitudinals possess less axon density than wild-type (compare arrows in C and H). Partial commissural fusions and distortions remain common in late-stage 13 and early-stage 14 deletion homozygotes (I,J, arrowheads). Scale bar: 11 µm.

View larger version (73K): [in a new window]   Fig. 3. In the absence of Gliolectin, commissural axons cross the midline at the dorsal boundary of the nerve cord rather than interdigitating between the midline glial cells. All panels are transverse sections through the ventral nerve cord (20 µm, ventral towards the bottom) of stage 12/1 embryos (equivalent to Fig. 2A,F) at the position of maximal commissural axon density. Axons are labeled with mAb BP102 (brown) and the entire width of the nerve cord is presented. (A) Wild-type embryo double stained with mAb 1B7 (black) to visualize the midline glial cells (MG) demonstrates the intimate association between diffuse commissural axons and Gliolectin-expressing glial cells. The forming longitudinal comprises a distinct lateral mass of BP102 staining axons (arrows). (B) Without 1B7 staining, the fine distribution of commissural axons between the midline glia is apparent in a wild-type embryo as is the density of fibers in the forming longitudinal (arrows). (C) However, in 3013/3013 embryos, commissural axons fasciculate into a dense bundle that crosses the midline at the dorsal boundary of the nerve cord (arrowhead) and longitudinal mass is greatly reduced (arrows). Scale bar: 5 µm.

View larger version (128K): [in a new window]   Fig. 4. EMS-induced gliolectin mutations affect commissure formation. Appearance of commissures stained with mAb BP102 in wild-type embryos (A) and in 3013 homozygotes (B) at mid-stage 13. Commissure formation is abnormal in the EMS mutant lines designated glecm24 (C) and glecm98 (D) but only when either is in combination with the deletion chromosome (3013) or with each other (E). Both the glecm24/3013 and glecm98/3013 embryos (C,D) are double-stained with BP102 in black and 1B7 (anti-Gliolectin) in brown, demonstrating that neither mutation abolishes the expression of the Gliolectin antigen. Incomplete commissure separation (B-E, arrowheads) is apparent in the deletion and in the EMS mutants, while decreased longitudinal density is less pronounced in the EMS mutants than in the deletion (B-E, arrows). Scale bar: 13 µm.

View larger version (132K): [in a new window]   Fig. 5. Loss of Gliolectin function delays pioneering of the longitudinal pathway. Pioneering of a longitudinal pathway is visualized with the mAb 1D4 (anti-Fasciclin II, nickel-enhanced) in wild-type (A-F), 3013 homozygous (G-I) and glecm98/3013 (J-L) embryos at late stage 12 (A,D,G,J), early stage 13 (B,E,H,K) and late state 13 (C,F,I,L). In all panels, anterior is towards the left. The schematic diagrams across the top (A-C) depict the progressive completion of the pCC/MP1 pathway. At late stage 12, the pCC growth cone extends anteriorly towards the SP1 neuron (A, in all panels the arrow indicates site at which pCC meets the SP1 neuron). By early stage 13, the pCC growth cone has contacted the SP1 neuron (B, arrow) and continues to grow anteriorly to form a continuous longitudinal pathway by late stage 13 (C, arrow). The intensity of 1D4 staining increases from stage 12 through stage 13 in a pair of neurons at the midline (the dMP2/vMP2 pair, A-L, arrowhead), providing an independent assessment of nerve cord maturation. In the schematic diagram, the level of 1D4 staining is indicated by the shade of blue that colors the dMP2/vMP2 neuron cluster (A-C, arrowhead). In all other panels, the arrowhead indicates the position of the dMP2/vMP2 pair, although only one of each of these neurons is seen in the relevant focal plane. While dMP2/vMP2 staining is just barely detectable in stage 12 wild-type embryos (A,D, arrowheads), it progressively defines distinct neuronal boundaries by late 13 (C,F,I,L, arrowheads). In late stage 13 wild-type embryos, the pCC growth cone displays the streamlined morphology characteristic of a rapidly extending axonal process as it extends towards the point at which it will contact the SP1 neuron (D, arrow). In 3013 homozygotes at late stage 12, however, pCC axons do not extend as far as in wild type and frequently possess growth cones with spread morphology (G). By early stage 13, pCC frequently remains stalled within its segment of origin and continues to exhibit growth cone morphology more consistent with exploration than with fasciculation in 3013 homozygotes (H). In late stage 13 deletion homozygotes, pCC has extended into the next segment, although breaks in the pCC/MP1 pathway are frequently seen (I, arrow). In glecm98/3013 embryos, the absence of pCC extension is striking (J-L, arrow) even into late stage 13 (L, arrowhead indicates strong staining in vMP2/dMP2, a characteristic of the assigned stage). Scale bar: 5 µm in D-L.

View larger version (125K): [in a new window]   Fig. 6. Embryonic CNS phenotypes found in 3013 homozygotes are rescued by transgenic expression of Gliolectin. Forming commissures are visualized in black with mAb BP102 (A-D, mid-stage 13 embryos) and the pCC longitudinal pioneer is also stained black with mAb 1D4 (E-H, early stage 13 embryos). Compared with wild-type embryos (A,E), commissures and longitudinals are distorted (B, arrowhead and arrow, respectively) and pCC outgrowth is delayed (F, arrow indicates position of SP1 neuron) in 3013 homozygotes. rho-Gal4 driven expression of a single UAS-glec element (expression detected with mAb 1B7 in brown in C,D,G,H) partially rescues both the commissural (C, arrowhead) and longitudinal (G) phenotypes of 3013 homozygotes. In the rescued embryo, commissures are more clearly separated and organized than in the null. Likewise, pCC extension is comparable with wild type; the most severe abnormality observed in rescued embryos is indicated (G, arrow indicates a remaining gap between pCC and SP1 in a single segment). When rho-Gal4 drives expression from two copies of UAS-glec in a 3013/3013 background, commissures (D, arrowhead) and longitudinals are more completely rescued (arrow in D and H). Scale bar: 12 µm in all panels.

View larger version (31K): [in a new window]   Fig. 7. Loss of Gliolectin expression affects commissure and longitudinal formation. Phenotypes associated with loss of Gliolectin indicate that this carbohydrate binding protein functions to capture axons at the surface of the midline glial cells (shown in gray) in which it is expressed. For longitudinal axons (shown in blue), capture facilitates the transmission of signals that keep them from crossing the midline (Robo, Slit). In the absence of Gliolectin, longitudinal axon extension is delayed. For commissural axons (shown in red), capture ensures the transmission of signals that allow midline crossing (Commissureless, Netrins) and contributes to the formation of a favorable growth substrate. In the absence of Gliolectin, early commissural axons grow upon other axons rather than in close association with the midline glia.