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

First published online 19 January 2005
doi: 10.1242/dev.01614

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 Marin, E. C. Articles by Luo, L. Search for Related Content PubMed PubMed Citation Articles by Marin, E. C. Articles by Luo, L.

Developmentally programmed remodeling of the Drosophila olfactory circuit

¹ Department of Biological Sciences, Stanford University, Stanford, CA 94305, USA
² Institute of Molecular and Cellular Biosciences, The University of Tokyo, Tokyo 113-0032, Japan

View larger version (87K): [in a new window]   Fig. 1. Embryonic-born projection neurons exhibit characteristic connectivity in the adult olfactory system. (A-D) Many projection neurons innervating glomeruli in the adult AL are born during embryogenesis. (A) The GAL4-GH146 enhancer trap is expressed by 90 PNs innervating approximately 40 of the 50 glomeruli in the adult AL. (B) A lateral neuroblast clone labeled by early larval heatshock innervates 11 of these glomeruli. (C) An anterodorsal neuroblast clone labeled by early larval heatshock innervates a non-overlapping subset of 12 glomeruli. (D) An anterodorsal neuroblast clone labeled by early embryonic heatshock innervates at least 15 additional non-overlapping glomeruli. Asterisks indicate superficial glomeruli that are innervated by embryonic but not larval-born PNs. (E-L) Each embryonic-born PN generally innervates one glomerulus in adult AL (E1-L1) except VL2p+ (F1, see Table 1). Each glomerular class of PN exhibits a characteristic axon branching pattern in the MB calyx and LH (E2,E3-L2,L3 show two examples of each glomerular class). In this and subsequent figures, all brains were stained with anti-mCD8 (green, reporter for GAL4 or MARCM clones) and nc82 (magenta, marking neuropiles), and presented as flattened confocal stacks oriented with midline to the left and dorsal to the top unless otherwise noted. aD, anterodorsal; e HS, embryonic heatshock; GAL4-GH146, whole GH146 driver expression in AL; Lat, lateral; l HS, larval heatshock; Nb, neuroblast.

View larger version (104K): [in a new window]   Fig. 2. Embryonic-born projection neurons participate in the larval olfactory system. (A) GAL4-GH146 staining pattern in the larval brain as visualized by UAS-mCD8-GFP reporter. PN dendrites innervate the larval AL and their axons travel via the inner and outer antennocerebral tracts to innervate ipsilateral higher brain centers, the MB calyx and the presumptive LH. (B) Single-cell clones of embryonic-born PNs (one on each hemisphere) show that each innervates one glomerulus in the larval AL and sends its axon to the MB calyx and LH. (C1-C3) High magnification of larval AL for three single-cell clones. Rarely, two glomerulus-like structures are innervated by a single PN (C3). (D1-D3) High magnification of axon projections for three single-cell clones. (E1-E3) Higher magnification of MB for three single-cell clones; each PN innervates one or occasionally two glomeruli in the larval MB calyx. Note: images C1-E3 are from nine different animals. Arrows indicate dendrites in larval AL glomeruli. Arrowheads indicate axon termini in MB calyx glomeruli. iACT, inner antennocerebral tract; LAL, larval antennal lobe.

View larger version (85K): [in a new window]   Fig. 3. Persistent projection neurons prune processes locally during early metamorphosis. (A-D) Timecourse for PPN dendritic pruning at the larval AL. (A) At the onset of puparium formation, each PPN labeled using GH146 MARCM has larval morphology, with dense dendrites in one glomerulus of the larval AL (score of 1). (B) As metamorphosis proceeds, dendrites in the larval AL start to disappear (score of 2). (C) Soon, only a few dendritic remnants are visible in the larval AL (score of 3). (D) Eventually, all dendritic remnants vanish from the larval AL (score of 4). (E-H) Timecourse for PPN axonal pruning at the larval MB and LH. (E) At puparium formation, each PPN has an axon with one or two large synaptic densities in MB calyx and branches in LH (score of 1). (F) The synaptic bouton in MB calyx shrinks as branches in the LH disappear (score of 2). (G) Soon only a slight swelling in the MB calyx indicates the former bouton and the main axon branch has begun to fragment (score of 3). (H) At last, no trace of the bouton(s) in MB calyx can be seen and the axon has pruned back to the edge of the MB calyx (score of 4). (I-K) Quantification of pruning scores of PPN dendrites in the (I) larval AL, (J) MB calyx and (K) LH every 2 hours APF. 0 hour APF brains were stained with anti-mCD8 and nc82, all others with anti-mCD8 and anti-synaptotagmin. Slender arrows indicate dendrites in the larval AL. Arrowheads indicate synaptic boutons in the MB calyx glomeruli. Blunt arrows indicate axon termini in the LH.

View larger version (390K): [in a new window]   Fig. 5. Ultrastructure analysis of MB calyx. (A) Ultrastructure analysis of larval (A2,A3) and 6 hours APF calyx (A4,A5) using GH146 driven EM marker CD2::HRP labeling PPN axons. (A1) Illustration of electron micrograph (EM) section plane using confocal Z projection of larval GH146-GAL4 driving mCD8::GFP, which labels the AL, inner antennocerebral tract and calyx with a dashed box showing the approximate location of EM section through the calyx. (A2) 4000x EM of the larval calyx showing the punctate labeling pattern with labeled boutons toward the outside of the calyx. Asterisks denote large presynaptic boutons. (A3) 40,000x EM of labeled larval PPN presynaptic terminals (filled arrowheads) apposing small unlabeled dendritic profiles postsynaptically. Open arrowheads point to the profile of the entire presynaptic bouton. (A4) 4000x EM of 6 hours APF degenerating PPN presynaptic boutons in the calyx. (A5) 40,000x EM at 6 hours APF with open arrowheads pointing to the border of a labeled presynaptic bouton and surrounding glia cytoplasm. (B) Ultrastructure analysis of larval (B2,B3) and 6 hours APF calyx (B4,B5) using GAL4-201Y driving CD2::HRP labeling MB neuron dendrites. (B1) Illustration of EM section plane using confocal Z projection of larval 201Y-GAL4 driving mCD8::GFP, which labels dendrites in the calyx, the axon peduncle, and medial and dorsal axon lobes. Dashed box shows approximate location of EM section through the calyx. (B2) 4000x EM of the larval calyx showing the concentric labeling pattern, with labeled dendrites toward the outside of the calyx. Asterisks denote large presynaptic boutons. (B3) 40,000x EM of labeled larval dendritic calyx showing unlabeled PPN presynaptic terminals (filled arrowheads) apposing small labeled dendritic profiles postsynaptically. (B4) 4000x EM of 6 hours APF MB neuron labeled calyx showing degeneration of larval concentric labeling. (B5) 40,000x EM at 6 hours APF with an arrow specifying a degenerating synapse full of synaptic vesicles surrounded by labeled dendritic profiles, which appear to be engulfed by a glia (open arrowheads pointing to the border of engulfed synaptic profiles and glia cytoplasm). Dotted lines in A2,A4,B2,B4 denote the border of the calyx. Scale bars: 50 µm in A1,B1; 10 µm in A2,A4,B2,B4; 1 µm in A3,A5,B3,B5. c, calyx; d, dorsal axon lobe; iACT, inner antennocerebral tract; m, medial axon lobe; p, axon peduncle.

View larger version (83K): [in a new window]   Fig. 4. MZ612+ persistent projection neurons prune processes locally during early metamorphosis. (A-H) Timecourse of pruning of (A-D) dendrites and (E-H) axons of the GAL4-MZ612-positive PPN. (I-K) Quantification of pruning scores of MZ612-positive PPN dendrites in the (I) larval AL, (J) MB calyx and (K) LH every 2 hours APF. Experimental conditions and figure labeling are as in Fig. 3. In addition to strongly labeling a single PPN that eventually innervates VA6 in adult, GAL4-MZ612 also labels other cells and processes near the larval AL, MB and LH. However, confocal tracing allows unambiguous distinction of PPN dendritic and axonal processes.

View larger version (51K): [in a new window]   Fig. 6. Persistent projection neuron pruning requires ultraspiracle and baboon. Anti-EcR-B1 staining (magenta) in single-cell MARCM clones (green) that are (A1) wild-type or homozygous for (A2) baboFd4 or (A3) usp3. At the onset of puparium formation, (A2) baboFd4 PPNs fail to express EcRB1, as indicated by the lack of nuclear magenta in the MARCM single-cell clone. At third instar larva, single-cell MARCM clones of PPNs homozygous for (B2) baboFd4 or (B3) usp3 exhibit apparent normal dendritic projections in larval AL compared with (B1) wild-type control. At third instar larva, single-cell MARCM clones of PPNs homozygous for (C2) baboFd4 or (C3) usp3 exhibit apparent normal axonal projections in larval MB and LH compared with (C1) wild-type control. At 8 hours APF, (D1) the majority of wild type PPNs have completely pruned their dendrites in the larval AL. However, many (D2) baboFd4 and (D3) usp3 PPNs still exhibit larval-like dense dendrites in the larval AL. At 8 hours APF, the majority of wild-type PPNs have largely eliminated their boutons in the MB calyx and pruned their axons back to the edge of the calyx (E1). By contrast, most (E2) baboFd4 and (E3) usp3 PPN axons still retain larval-like morphology, with large boutons in the MB calyx and terminal branches in the LH. (F) Cumulative pruning scores in larval AL, MB calyx and LH at 8 hours APF. L3 brains were stained with anti-mCD8 and nc82. At 8 hours APF, brains were stained with anti-mCD8 and anti-synaptotagmin. Slender arrows indicate dendrites in larval AL. Arrowheads indicate synaptic boutons in MB calyx glomeruli. Blunt arrows indicate axon termini in LH. L3, third instar larva; LAL, larval antennal lobe; WT, wild type.

View larger version (126K): [in a new window]   Fig. 7. Adult phenotypes of ultraspiracle and baboon PPN clones. (A1) Wild-type PPNs labeled by GH146 MARCM typically innervate a single glomerulus in the adult AL and (A2) have several collateral branches with terminal boutons in the MB calyx and a characteristic branching pattern in the LH (see also Fig. 1). (B1,C1) usp3 PPN dendrites generally target an appropriate glomerulus in the AL but also exhibit a few ectopic processes (arrows), while (B2,C2) their axons often retain large larval-like synaptic structures directly on the main trunk (arrowheads). The axon trunk sometimes makes a detour from the main inner antennocerebral tract within MB to connect these larval-like boutons (C2 only). Axon termini in the LH are morphologically normal, although they may include possible extra branches, which might be putative larval remnants (B2, blunt arrow). (D1-F1) baboFd4 PPN dendrites often exhibit more severe phenotypes than that of usp3; they can be sparse and localized to a few areas in the AL without targeting a specific glomerulus (D1), leave the AL entirely (E1, arrow), or be broadly distributed throughout the AL (F1). (D2-F2) All baboFd4 PPN axons exhibit larval-like boutons directly on their main trunks in the MB calyx (arrowheads) and occasionally detour from the inner antennocerebral tract to connect them (F2 only). Additional defects seen in the LH include unusually profuse swellings along the branches (D2), failure to enter the LH (E2), and failure to elaborate axon termini (F2). Note: All AL and MB/LH images are presented in pairs from the same animal, but this is not meant to imply consistent correlations between the various dendritic and axonal phenotypes. (G) Quantification of usp3 and baboFd4 phenotypes in the adult AL. Categories are `WT' (uniglomerular dense dendrites), `glom + extra' (uniglomerular dendrites plus extra processes in the AL), `sparse local' (sparse dendrites localized to a region of the AL but not innervating any particular glomerulus), `sparse broad' (sparse dendrites distributed broadly throughout the AL) and `leaves AL' (dendrites arborize outside the AL proper). Note that although the first four categories are mutually exclusive, the fifth can overlap with any of the others except for `WT'. (H) Quantification of usp3 and baboFd4 phenotypes in the adult MB calyx. Categories are `WT' (adult-like collateral branches with terminal swellings), `leaves iACT' (main axon trunk detours from the inner antennocerebral tract, usually to connect with a larval-like bouton), `larval + adult boutons' (both larval-like boutons on the main axon trunk and adult-like collateral branches), and `larval boutons only' (larval-like boutons on the main axon trunk but no adult-like collateral branches). Note that `leaves iACT' can overlap with either of the last two categories; the others are mutually exclusive. (I) Quantification of usp3 and baboFd4 phenotypes in the adult LH. Categories are `WT' (terminal branches of varying complexity in LH), `swellings' (unusually profuse swellings along the terminal branches), `LH entry' (failure to enter the LH proper), and `elaboration' (failure to elaborate second order terminal branches in the LH). Note that there is overlap between all categories except for `WT'.