Development of connectivity in a motoneuronal network in Drosophila larvae

Abstract

Background: Much of our understanding of how neural networks develop is based on studies of sensory systems, revealing often highly stereotyped patterns of connections, particularly as these diverge from the presynaptic terminals of sensory neurons. We know considerably less about the wiring strategies of motor networks, where connections converge onto the dendrites of motoneurons. Here, we investigated patterns of synaptic connections between identified motoneurons with sensory neurons and interneurons in the motor network of the Drosophila larva and how these change as it develops.

Results: We find that as animals grow, motoneurons increase the number of synapses with existing presynaptic partners. Different motoneurons form characteristic cell-type-specific patterns of connections. At the same time, there is considerable variability in the number of synapses formed on motoneuron dendrites, which contrasts with the stereotypy reported for presynaptic terminals of sensory neurons. Where two motoneurons of the same cell type contact a common interneuron partner, each postsynaptic cell can arrive at a different connectivity outcome. Experimentally changing the positioning of motoneuron dendrites shows that the geography of dendritic arbors in relation to presynaptic partner terminals is an important determinant in shaping patterns of connectivity.

Conclusions: In the Drosophila larval motor network, the sets of connections that form between identified neurons manifest an unexpected level of variability. Synapse number and the likelihood of forming connections appear to be regulated on a cell-by-cell basis, determined primarily by the postsynaptic dendrites of motoneuron terminals.

Figure 1

Genetic Tools to Differentially Label Partner Neurons Reveal a Motoneuron-Specific Pattern of Connectivity during Development (A and D) Intersection of the split-Gal4 line BF29^VP16.AD with Cha(7.4kb)^Gal4.DBDJ8A1 targets Gal4 activity to two pairs of descending interneurons, namely a lateral (IN_lateral) and a medial (IN_medial) interneuron on each side of the midline, as well as to the segmentally repeated dda sensory neurons (ddaD and ddaE). (B–D) Two motoneurons (aCC and RP2) that make connections with these interneurons are visualized with a LexA-Flpout system. (E) Synaptic contacts between partner neurons (arrowheads) are identified by co-localization of the presynaptic active zone marker Bruchpilot (Brp) and the cell-cell contact reporter (GRASP). The arrow points to a contact region between partner neurons that is devoid of presynaptic Brp::mRFP and therefore not considered a putative synaptic site. Insets show the synapse area as separate channels. (F–G′) Cross section views of nerve cords showing the location of aCC and RP2 motoneurons, relative to their presynaptic sites from their common partners, IN_lateral (filled arrowheads) and dda terminals (open arrowheads). (F′) and (G′) show the GRASP and Brp channels separately. (H and I) Quantification of the development of connections between the aCC and RP2 motoneurons and the IN_lateral (H) and ddas (I) during larval life. Boxplots show the median of the distribution (middle line), the 75^th (upper limit of box) and 25^th (lower limit of box) percentile; whiskers indicate the highest and lowest value of each dataset. Each of the dendritic arbors shows a significant increase in connectivity with the IN_lateral over time: ANOVA (p values < 0.01) with post tests for linear trend: p < 0.0001 for RP2 and aCC ipsilateral; p = 0.0025 for aCC contralateral arbor. RP2 arbors had a significantly different number of synapses with the IN_lateral at 24 hr and 48 hr as compared to aCC contralateral arbors: ANOVA (p values = 0.06 and 0.01) with uncorrected Fisher’s least significant difference test, ^∗p = 0.0415 and ^∗∗p = 0.0083, respectively. (I) Motoneuron-dda connectivity or absence thereof is shown. RP2 and aCC differ significantly in their sensory-motor connectivity, at 24 hr and 48 hr (Fisher’s exact test, ^∗∗p = 0.0023 and ^∗∗p = 0.0098, respectively). Anterior is up in (A)–(E); dorsal is up in (F)–(G′). Dashed line represents the midline. Scale bars represent 10 μm, except in (E), where each inset is 5 × 5 μm. See Figures S1–S3 for tests of GRASP and Brp::mRFP for reporting on synapses.

Figure 2

Variable Connectivity between Intersegmental Interneurons and Motoneurons (A–B′′) Comparison of connectivity between an interneuron from the split-Gal4 expression line BF29^VP16.AD (IN_lateral) and two motoneuron types: RP3, innervating longitudinal muscles, and the motor neuron of the segment border muscle (MN-SBM), innervating the transverse segment border muscle (SBM). Z projections of confocal image sub-stacks show presynaptic sites reported by UAS-brp::mRFP (magenta) and motoneurons manually labeled with DiD (green). (A′), (B′), and (B′′) show the insets in (A) and (B) in more detail, as single confocal planes. The gray level insets in (A′) show the Brp::mRFP and DiD motorneuron channels separately. Although RP3 and the MN-SBM cover similar dendritic areas around the IN_lateral, RP3 shows putative contacts (arrowheads), whereas the MN-SBM does not. (C) First instar larval nerve cord with neuropil visualized with alpha-Bungarotoxin-Alexa Fluor 488 (yellow) and two reconstructions of intersegmental interneurons from the split-Gal4 expression line BF59^VP16.AD intersected with Cha(7.4kb)^{Gal4.DBD J8A1}, shown in thoracic segments T2 (magenta) and T3 (green). (D and E) Putative synaptic contacts between motoneurons and these intersegmental interneurons (IN_BF59) at larval hatching. Interneurons expressing UAS-brp::mRFP (red) were manually labeled with DiO (blue), and motoneurons were visualized with DiD (green). (D) Single confocal optical section. (E) Reconstructed interneuron (blue) and partially reconstructed dendritic arbor of motoneuron (MN, green) where a site of likely physical overlap coinciding with a presynaptic site is highlighted red (arrow). The inset in (D) is a single confocal section showing the overlap in more detail in the motoneuron (green) and presynaptic marker (red) channels. (F) Table of frequencies at which synaptic contacts with BF59^VP16.AD interneurons were observed for different motoneuron types. Anterior is up. Ventral midline is indicated by dashed line. Scale bars of (A), (B), and (C) represent10 μm; scale bars of (A′), (B′), (B′′), (D), and (E) represent 5 μm; the scale bar of (D) inset represents 1 μm.

Figure 3

aCC Motor-Sensory Neuron Connections Are Achieved through Different Routes (A–D′) Cross-section views of nerve cords showing the different configurations of aCC-dda connections (arrowheads): no connection (A), contralateral (B), ipsilateral (C), or bilateral (D). In order to make the distinction among pseudo-colored motoneuron (cyan), GRASP (green), and Brp::mRFP (magenta) easier, the GRASP and Brp channels are also displayed separately in (A′), (B′), (C′), and (D′). In (B) and (C), aCC neurites connecting to the ddas are outlined with dotted lines to show that synaptic dendritic segments do not always originate from the main dendritic tree. Dorsal is up. All scale bars represent 10 μm. Dashed line indicates midline. All data were collected 24 hr ALH.

Figure 4

Presynaptic Sites Appear Randomly Distributed along the IN_lateral Axon and Do Not Predetermine Connectivity Outcomes (A) Two or more motoneurons of the same kind (RP2, aCC ipsilateral, or aCC contralateral dendritic arbors) can receive different numbers of synaptic connections from the same presynaptic IN_lateral axon. The number of synapses does not correlate with the antero-posterior location of the motoneuron (see Figure S4). Observations from different larval stages are displayed. (B) Distribution of UAS-brp::mRFP-labeled presynaptic sites along the IN_lateral axon traversing segments A2 to A8. Data are shown for 17 neurons from 11 individuals with axons normalized for length. Brp::mRFP puncta distribution along axons was indistinguishable from random (Kolmogorov-Smirnov two-sample test, p = 0.89). (C) Distribution of distances between adjacent presynaptic sites shown in (B) is also indistinguishable from random. Inset in (C): distribution of distances between adjacent presynaptic sites in the IN_lateral homologs within one specimen (left is indicated by open bars; right is indicated by filled bars). Differences in local Brp::mRFP puncta densities do not correlate with the number of synapses made onto adjacent motoneurons (see Figure S5).

Figure 5

Positioning of Postsynaptic Motoneuron Dendrites Is an Important Factor in Determining the Number of Connections (A) Changing the medio-lateral positioning of motoneuron dendrites by targeted overexpression of the guidance cue receptor Frazzled/DCC shifts the distribution of the dendritic arbor away from the lateral neuropil (where the IN_lateral partner axon is located) and toward the midline, closer to the IN_medial axon. Magenta indicates BF29^VP16.AD > brp::mRFP; cyan indicates motoneurons; green indicates GRASP signal, with arrowheads highlighting contacts with the IN_medial (open arrowhead) and IN_lateral (filled arrowhead), respectively. (B) Diagram of motoneuron dendrites (cyan) contacting ipsilateral (magenta) and contralateral (gray) IN_lateral axons (darker shade) and IN_medial axons (lighter shade). (C and D) In thoracic segments, the proportion of synaptic sites that motoneurons make with the IN_lateral as compared to the IN_medial changes with the repositioning of motoneuron dendrites (t test; RP2: p = 0.0005; aCCi: p = 0.0194). Scale bars represent 10 μm. Dashed lines in (A) and (B) indicate midline. Anterior is up in (A) and left in (B).