Fig 1.
Johnston’s Organ inputs onto the Drosophila giant fiber escape circuit.
(A) Overview of the giant fiber (GF) escape circuit, showing the GF axons descending from the brain to the thoracic ganglion, where they contact neurons that excite the tergotrochanteral ‘jump’ muscle (TTM) and the dorsal longitudinal indirect flight muscle (DLM). The antennae, one of the sources of input to the GF, are indicated (ant). (B) Side view of the pair of Giant Fiber (GF) neurons in the brain (cyan), showing their axons projecting posteriorly in the cervical connective to the thorax (left), and their main dendrites projecting anteriorly (right) to contact axons of Johnston’s Organ neurons (JONs) originating in the second antennal segment, or pedicel (ped). Two subsets of JONs are indicated here, the JO-A (red or magenta) and JO-B (red) sound-sensitive types. The third and fourth antennal segments, i.e. the funiculus (fun) and arista (ari), are also indicated. (C) Dorsal view of the right pedicel with the top removed to show the JONs within, with their dendrites attached to the funicular stalk (asterisk). Two subpopulations of neurons are shown, the JO-A (magenta) and JO-B JONs (red). (D) Dorsal view of the right GF neuron, also showing the neighboring GCI neurons (dark blue). The main GF dendrite projects anteriorly within a cylindrical group of JO-A axons, which are dye-coupled to it. The dashed box indicates the region shown in subsequent figures.
Fig 2.
Quantification of GF synaptic inputs and branching.
(A) Diagrammatic view of the GF dendrite (cyan) and JON axons (red spots), similar to that shown in the box in Fig 1D. The JO-A group of axons (magenta) is NB-coupled to the GF. The GCI neurons are also NB-coupled to the GF. The posterodorsally-projecting GF neurite (n) is cropped out of subsequent images. Anterior (A) is towards the top, lateral (L) to the right. (B-C). Confocal views of a typical preparation. The right GF axon was injected with a mixture of Neurobiotin (NB: blue) and Alexa Fluor 488 –coupled dextran (DA488: green), making it cyan in color. JO15-GAL4 was used to drive expression of Strawberry-tagged Brp-short (Brp: red) in JO-A and JO-B subgroups of JON axons, labeling putative active zones at chemical synapses. Blue NB passes retrogradely into a subgroup of the JO-A axons, so their red active zones appear magenta, or white when apposed to the GF. (B) Dorsal 3D view of the complete image stack of frontal slices, with anterior (A) towards the top, lateral (L) to the right. The subsequent panels represent single frontal slices taken at the antero-posterior positions indicated by the arrows: 1, at the anterior end of the GF dendrite, 2, midway along the dendrite in the region of medial branches, and 3, at the posterior end of the dendrite, where it bends dorsally and extends a ventral branch. In these panels, dorsal is to the top and lateral to the right. (C) Single confocal slices taken at the indicated regions, with dorsal (D) towards the top, lateral (L) to the right. Regions of overlap of Brp signal with the cyan GF dendrites (ie. putative synaptic contacts) appear white. (D) Thresholded signal. (E) NB signal removed to show only putative active zones (AZ: pink) on the GF. The total AZ volume over the length of the GF dendrite was subsequently quantified. (F) Single pixel outline of the GF dendrites, used as an approximation for surface area. Dendrites projecting medially more than 5 um were masked (pale green) and quantified separately. Scale bars: 20 μm in A, 10 μm in B-E.
Fig 3.
ShakB(N+16) increases, and ShakB(N) decreases, NB coupling.
(A) Diagrammatic dorsal view of the same GF shown in Fig 1D, showing only the analyzed portion of the main dendrite. JO-A and JO-B axons are labeled with red Brp spots, while JO-A axons are also magenta indicating NB-coupling. (B) Examples of experimental preparations, showing dorsal views of cropped image stacks of the GF dendrite and JON axons, with anterior towards the top, lateral to the right. The analyzed portion of the right GF dendrite is shown (cyan), along with Brp labeling in JON axons (red) and NB coupling (blue). Panels 1–4 show control animals (Con, column 1), UAS-shakB(N+16) animals (N+16, column 2), UAS-shakB(N) (N, column 3), and, as a positive control, UAS-en (En, column 4). Panels 5–8 show the corresponding GF and NB channels only and indicate the 10 μm-thick region at the tip of the dendrite from which the area of NB coupling was averaged. (C) Cross-sectional area of NB-coupled axons. Scatter plots combined with bar charts showing mean ± SEM. Asterisks or “ns” above each column indicate the significance when compared to preceding columns with post-hoc Tukey tests. Expression of N+16 caused a significant increase in coupling area, similar to that caused by En, whereas the ShakB(N) isoform abolishes NB coupling.
Fig 4.
ShakB immunostaining is increased by ShakB(N+16) overexpression.
(A) Diagrammatic dorsal view of the same region of JON afferents shown in previous figures. Both JO-A and JO-B axons are labeled with green GFP and red Brp spots, while the cylindrical cluster of JO-A axons is highlighted for clarity (lilac). (B) Typical preparations, in which JO15-GAL4 was used to drive UAS-brp-sh-strawberry (Brp) and also UAS-CD8::GFP (GFP) in the JO-A and JO-B neurons. The GF was not labeled. ShakB protein was stained with antibody against all isoforms (ShB). Panels 1–3 show examples of experimental preparations, showing dorsal views of cropped image stacks of the JON axons, with anterior towards the top, lateral to the right. To the right of each is the ShB channel alone. Control animal (Con, column 1), UAS-shakB(N+16) animal (N+16, column 2), and UAS-shakB(N) (N, column 3). Arrows indicate the approximate position of the single frontal slices shown in subsequent panels. (B4-6) Single thresholded frontal slices, dorsal towards the top, lateral to the right. Plaques of ShB labeling (blue) are shown in the two subsets of axons, JO-A (light gray) and JO-B (dark gray). (C) Total volume of ShakB staining in the two subsets of JO axons. Scatter plots combined with bar charts showing mean ± SEM. Asterisks or “ns” above each column indicate the significance when compared to preceding columns with post-hoc Tukey tests. Control JO-A axons have more ShakB staining than JO-B. The N+16 isoform increases ShakB staining in both JO-A and JO-B axons, while the N isoform does not.
Fig 5.
ShakB misexpression and active zone distribution on GF.
Examples of experimental preparations, showing dorsal views of cropped image stacks of the GF dendrite and JON axons, with anterior towards the top, lateral to the right. The analyzed portion of the right GF dendrite is shown (cyan), along with Brp labeling in JON axons (red) and NB coupling (blue). Panels 1–4 show control animals (Con, column 1), UAS-shakB(N+16) animals (N+16, column 2), UAS-shakB(N) (N, column 3), and UAS-en (En, column 4). Panels 5–8 show the GF and Brp appositions (putative AZ) on the GF dendrites, with insets showing representative single frontal slices taken at the indicated antero-posterior positions. Medial GF dendrites are indicated by a pale green color. Panels 9–12 show putative AZ on only the medial GF branches (pale green).
Fig 6.
Quantification of effects of ShakB misexpression on chemical synapses and dendritic branches of GF.
(A-D) Scatter plots combined with bar charts showing mean ± SEM. Asterisks or “ns” above each column indicate the significance when compared to preceding columns with post-hoc Tukey tests. (A) Total volume of AZ apposed to the GF. Neither ShakB(N+16) nor ShakB(N) affected the amount of AZ, compared to the increase seen with En overexpression. (B) Total surface area of GF dendrites. There is no effect of ShakB(N+16) or En misexpression but ShakB(N) causes a 23% decrease in branching. (C) Surface area of GF dendrites that project more than 5μm medially. There is no effect of ShakB(N+16) or ShakB(N) misexpression but En results in an increase in branching. (D) Total volume of AZ apposed to the medial GF dendrites. There is no effect of ShakB misexpression but En results in a large increase in medial branch AZ.
Fig 7.
(A) Typical preparations, in which JO15-GAL4 was used to drive UAS-brp-sh-strawberry (Brp) and also UAS-CD8::GFP (GFP) in the JO-A and JO-B neurons. The GF was not labeled. ShakB protein was stained with antibody against all isoforms (ShB). Panels A1-3 show examples of experimental preparations, showing dorsal views of cropped image stacks of the JON axons, with anterior towards the top, lateral to the right. To the right of each the GFP channel is omitted to show the overlap between Brp and ShB. Control animal (Con, column 1), UAS-shakB(N+16) animal (N+16, column 2), and UAS-shakB(N) (N, column 3). Arrows indicate the approximate position of the single frontal slices shown in subsequent panels. (B) Plaques of ShB labeling in the two subsets of axons, JO-A (light gray) and JO-B (dark gray). (C) Putative AZ (Brp) in the JO-A and JO-B axons. (D) Overlap of Brp and ShB in the JO-A and JO-B axons. (E-F) Scatter plots combined with bar charts showing mean ± SEM. Asterisks or “ns” above each column indicate the significance when compared to preceding columns with post-hoc Tukey tests. (E) Total volume of putative AZ in the JO axons. More are present in JO-B than JO-A but ShakB isoform expression has no effect on either. (F) Colocalization of ShakB and AZ is significantly increased by the N+16 isoform in both A and B JO axons.