Lifelong suppression of Or42a OSNs after critical-period EB exposure.

(A) Schematic of the Drosophila olfactory circuit at the level of the AL. OSN, olfactory sensory neuron, PN, projection neuron, LN, local interneuron, EG, ensheathing glia. (B) Overview of odorant-exposure experiments. Flies are exposed to 15% (v/v) ethyl butyrate (EB) or mineral oil vehicle control during the Or42a OSN critical period from 0–2 days post-eclosion (DPE). (C–F) Live imaging of Or42a OSN odor-evoked activity at 2 (C, D) or 25 (E, F) DPE, following exposure to 15% EB or mineral oil from 0–2 DPE. Or42a-GAL4>UAS-GCaMP8f flies were stimulated with three 1-second puffs (yellow bars) of 5% EB (red traces and data points) or mineral oil vehicle control (grey traces and data points) spaced 10 seconds apart. Traces (C, E) show the mean ΔF/F0 values of all trials. Data points (D, F) represent the mean peak ΔF/F0 values for each fly. (G–H) Live imaging of ASAP5 responses in Or42a OSNs in 2 DPE flies following exposure to 15% EB or mineral oil from 0–2 DPE. Flies were stimulated with odor puffs as above. Traces (G) represent the mean ΔF/F0 values of all trials. Data points (H) represent the mean peak ΔF/F0 values for each animal. Data in D, F, and H are mean ± SD. *p < 0.05, **p < 0.01, unpaired t test (D) or Mann-Whitney U-test (F, H). Genotypes, raw values, and detailed statistics are provided in Figure 1—source data 1.

Experience-dependent pruning in VM7 is restricted to the presynaptic compartment.

(A–B) Live imaging of VM7 PN odor-evoked activity at 2 DPE, following exposure to 15% EB or mineral oil from 0–2 DPE. VM7 PN split-GAL4>UAS-GCaMP8f flies were stimulated with three 1-second puffs (yellow bars) of 5% EB (red traces and data points) or mineral oil vehicle control (grey traces and data points) spaced 10 seconds apart. Traces (A) show the mean ΔF/F0 values of all trials. Data points (B) represent the mean peak ΔF/F0 values for each fly. (C) Representative dendritic arbors of VM7 PNs endogenously expressing the V5-tagged postsynaptic marker Discs large 1 (Dlg1), from flies exposed to mineral oil or 15% EB from 0–2 DPE. (D) Normalized counts of Dlg1 puncta. (E) Representative images of VM7 PNs after exposure to mineral oil or 15% EB from 0–2 DPE. Neuropil is visualized with antibody staining for the presynaptic active zone protein bruchpilot (Brp). PN dendritic arbors occupying VM7 are enclosed in white boxes. (F) Volumetric measurements of VM7 PN dendritic arbors shown in E. (G) Representative images of VM7 PN axonal arbors after exposure to mineral oil or 15% EB from 0–2 DPE. (H) Volumetric measurements of VM7 PN axonal arbors shown in G. (I) Models of VM7 PN presynapses from G after exposure to mineral oil or 15% EB from 0–2 DPE, defined as Brp puncta masked by mCD8::GFP-labeled VM7 PNs. (J) Normalized counts of VM7 PN presynapses from I. Data are mean ± SD. ns, p ≥ 0.05, *p < 0.05, Mann-Whitney U- test. Genotypes, raw values, and detailed statistics are provided in Figure 2—source data 1.

Experience-dependent pruning and inhibition are decoupled in Or43b OSNs.

(A) DoOR 2.0 profile of EB responses across all Drosophila ORs. (B–C) Representative images (bottom) and quantification of presynaptic content (B) and volume (C) of VM2 (Or43b OSNs) and VM7 (Or42a OSNs) in flies exposed to mineral oil or 15% EB from 0–2 DPE. Presynapses were visualized with Brp staining. Data are mean±SD. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, unpaired t-test. (D–E) Two-photon imaging of GCaMP8f responses in Or43b OSNs at 2 DPE following exposure to 15% EB or mineral oil from 0–2 DPE. Awake flies were stimulated with three 1 s odor puffs (yellow bars) of 5% EB (red traces and data points) or mineral oil (grey traces and data points), spaced 10 s apart. Traces (D) represent the mean ΔF/F0 values of all trials. Data points (E) represent the mean peak ΔF/F0 values for each animal. Genotypes, raw values, and detailed statistics are provided in Figure 3—source data 1.

Differential patterns of LN innervation may contribute to interglomerular differences in experience-dependent glial pruning.

(A) Numbers of neurons within the cell types included in B and C. (B–C) Sankey diagrams of the synaptic inputs and outputs of Or42a (B) and Or43b (C) OSNs. Inputs were thresholded at 99% of all synapses and outputs were thresholded at 95% of all synapses. Brackets contain the number of synapses and percentage of total synapses for that category. All connectomic analyses are derived from the FlyWire FAFB dataset. Synapse counts are provided in Figure 4—source data 1.

GABAergic inputs restrain glial pruning.

(A–B) Representative images (bottom) and quantification of VM7 (A) and VM2 (B) volume in flies exposed to mineral oil or 15% EB from 0–2 DPE. Data are mean ± SD. *p < 0.05, **p < 0.01, ****p < 0.0001, one-way ANOVA. Genotypes, raw values, and detailed statistics are provided in Figure 5—source data 1.