Experience-dependent MAPK/ERK signaling in glia regulates critical period remodeling of synaptic glomeruli

Abstract

Early-life critical periods allow initial sensory experience to remodel brain circuitry so that synaptic connectivity can be optimized to environmental input. In the Drosophila juvenile brain, olfactory sensory neuron (OSN) synaptic glomeruli are pruned by glial phagocytosis in dose-dependent response to early odor experience during a well-defined critical period. Extracellular signal-regulated kinase (ERK) separation of phases-based activity reporter of kinase (SPARK) biosensors reveal experience-dependent signaling in glia during this critical period. Glial ERK-SPARK signaling is depressed by removal of Draper receptors orchestrating glial phagocytosis. Cell-targeted genetic knockdown of glial ERK signaling reduces olfactory experience-dependent glial pruning of the OSN synaptic glomeruli in a dose-dependent mechanism. Noonan Syndrome is caused by gain-of-function mutations in protein tyrosine phosphatase non-receptor type 11 (PTPN11) inhibiting ERK signaling, and a glial-targeted patient-derived mutation increases experience-dependent glial ERK signaling and impairs experience-dependent glial pruning of the OSN synaptic glomeruli. We conclude that critical period experience drives glial ERK signaling that is required for dose-dependent pruning of brain synaptic glomeruli, and that altered glial ERK signaling impairs this critical period mechanism in a Noonan Syndrome disease model.

Keywords: Drosophila; ERK-SPARK; Noonan syndrome; Olfactory circuit; Phagocytosis.

Figure 1:. Experience-dependent glial pruning imaged with transgenic reporters

(A) Drosophila central brain mushroom body (MB; above) and antennal lobe (AL; below) with anti-Cadherin-N (CadN, magenta) synaptic glomeruli labeling and ethyl butyrate (EB) responsive Or42a receptor driving a membrane marker (mCD8::GFP, green) in the Or42a olfactory sensory neuron (OSN) innervation of the VM7 glomeruli. Animals exposed to the mineral oil vehicle control (left) or the EB odorant (right) for 24 hours from 0–1 days post-eclosion (dpe). Experience-dependent pruning of the synaptic glomeruli during the early-life critical period. EB experience causes temporally-restricted OSN pruning only in the critical period (arrows). Scale bar: 10 µm. (B) High magnification images of the Or42a OSN innervation of a VM7 glomerulus (Or42a-mCD8::GFP, green) and the perisynaptic glia labeled with a glia-specific repo-Gal4 membrane marker (repo>mCD8::RFP, red). Right: Glia infiltrate the VM7 glomerulus to mediate EB experience-dependent pruning. Scale bar: 5 µm. (C) Schematic drawing showing the transgenic ERK-SPARK biosensor with multivalent protein-protein interactions induced by ERK-specific phosphorylation (P) of homo-oligomeric coiled coils (HOTags). Fluorescent eGFP tags (green) allow the imaging visualization of active ERK signaling via the production of GFP puncta. ERK-SPARK is cell-targeted using the transgenic Gal4-UAS binary system. Glial-specific repo-Gal4 drives UAS-ERK-SPARK to image ERK signaling puncta within glia.

Figure 2:. Experience dose-dependent activation of ERK signaling within glial cells

(A) VM7 glomerulus (white dashed circles) labeled with anti-Cadherin-N (CadN, grey) and repo-Gal4 driven ERK-SPARK in glia (repo>ERK-SPARK, green). Experience-dependent ERK-SPARK signaling (fluorescent puncta) during the critical period depends on ethyl butyrate (EB) odorant dosage. Animals exposed to mineral oil vehicle control (left), 15% EB (middle) and 25% EB (right) for 24 hours from 0–1 days post-eclosion (dpe) show increasing ERK-SPARK puncta. Scale bar: 5 µm. (B) Quantification of glial ERK-SPARK puncta within and surrounding the VM7 glomerulus in all 3 treatment conditions. The R0I is centered on the VM7 glomerulus (white dashed circles) and includes perisynaptic glia within the image frame. All individual data points shown with mean ± SEM. Significance shown based on a one-way ANOVA tests and indicated as *p<0.05 and ****p<0.0001.

Figure 3:. Experience-dependent glial ERK signaling restricted to the critical period

(A) ERK signaling in glia is experience-dependent only during the early-life critical period. Images show VM7 glomeruli (white dashed circles) labeled with anti-Cadherin-N (CadN, grey) and repo-Gal4 driven ERK-SPARK in glia (repo>ERK-SPARK, green) within the early-life critical period (0–1 dpe: top row) and in the mature adults (7–8 dpe; bottom row). At both timepoints, animals exposed to mineral oil vehicle control (left), 15% EB (middle), and 25% EB (right) for 24 hours. Scale bar: 5 µm. (B) Quantification of the number of glial ERK-SPARK puncta circuit-localized to the VM7 glomerulus domain. The same ROI is used to quantify ERK-SPARK puncta in critical period (0–1 dpe) and maturity (7–8 dpe). All individual data points shown with mean ± SEM. Two-way ANOVA tests of significance as indicated shown as not significant (ns; p>0.05) or significant at ****p<0.0001.

Figure 4:. Glial ERK signaling induction is facilitated by the glial Draper receptor

(A) Experience-dependent ERK-SPARK signaling (fluorescent puncta) during the critical period depends on the glial Draper receptor. VM7 glomeruli (white dashed circles) labeled with anti-Cadherin-N (CadN, grey) and repo-Gal4 driven ERK-SPARK in glia (repo>ERK-SPARK, green). Transgenic controls are shown in the top row (repo-Gal4 driven UAS-ERK-SPARK in the w¹¹¹⁸ background) and glial-targeted draper RNAi in the bottom row (repo>draper-RNAi). Both conditions are exposed to the mineral oil vehicle, 15% EB or 25% EB odorant for 24 hours from 0–1 dpe. Scale bar: 5 µm. (B) Quantification of puncta in the VM7 glomerulus domain in all 6 conditions. A two-way ANOVA statistical tests are displayed as not significant (ns; p>0.05) or significant at **p<0.01 and ****p<0.0001.

Figure 5:. Glial ERK loss blocks dosage experience-dependent synaptic pruning

(A) Glial-targeted RNAi for Drosophila rolled (ERK homolog) prevents synaptic glomeruli pruning in a dose-dependent manner. Or42a receptor driven mCD8::GFP (green) shows VM7 innervation in the transgenic control (Or42a-mCD8::GFP/+; repo-Gal4/+; top row) and with glial-targeted rolled RNAi (Or42a-mCD8::GFP/+; repo-Gal4>UAS-rolled-RNAi; bottom row). Animals were exposed to mineral oil vehicle (left), 15% EB (middle) or 25% EB (right) for 24 hours from 0–1 dpe. Scale bar: 5 µm. (B) Quantification of the Or42a OSN innervation volume in all 6 conditions. All individual data points are shown with the mean ± SEM. A two-way ANOVA statistical test is displayed with significance indicated as not significant (ns; p>0.05) or significant at ***p<0.001 and ****p<0.0001.

Figure 6:. Glial ERK activation and synaptic pruning in a Noonan Syndrome model

(A) Noonan Syndrome patient-derived PTPN11 gain-of-function (GoF) mutation targeted to glia induces ERK signaling and synaptic glomerulus pruning in the critical period. Top row: VM7 glomeruli (white dashed circles) labeled with anti-Cadherin-N (CadN, grey) and ERK-SPARK (repo>ERK-SPARK, green). Bottom row: Or42a driven mCD8::GFP shows VM7 innervation (green). The 3 genotypes are transgenic control (Or42a-mCD8::GFP/+; repo-Gal4/+; left) and glial-targeted PTPN11 LoF (repo-Gal4>UAS-PTPN11^Q510P) and PTPN11 GoF (repo-Gal4>UAS-PTPN11^N308D). (B) Quantification of glial ERK-SPARK puncta in the VM7 domain. (C) Quantification of VM7 innervation volume. All individual data points shown with mean ± SEM. Significance shown based on a one-way ANOVA and indicated as not significant (ns; p>0.05) or significant at ***p<0.001 and ****p<0.0001.