Article Figures & Data
Figures
- Figure 1.
MiR-92a inhibition enhances 3-h olfactory memory. A, MiR-92a sponge expression enhances 3-h aversive memory. Pan-neuronal inhibition of miR-92a by expressing a UAS-miR-92a sponge (UAS-miR-92aSP) transgene enhances 3-h memory. The UAS-scrambled transgene was used as the control. PI: performance index, PIs are the mean ± SEM with n = 20. Two-tailed, two-sample Student’s t test, *p < 0.05. B, MiR-92a heterozygous flies exhibit enhanced memory performance. Two lines of a miR-92a deficiency exhibit enhanced PIs compared with a wCS10 control. PIs are the mean ± SEM with n = 10. One-way ANOVA followed by Dunnett’s post hoc tests, *p < 0.05, **p < 0.01.
- Figure 2.
Memory enhancement from inhibiting miR-92a in MBns. A, MiR-92a inhibition in mushroom bodies enhances 3-h memory. The UAS-scrambled and UAS-miR-92aSP flies were crossed to a battery of GAL4 lines that drive expression in specific populations of neurons. Most of the drivers promote expression in neurons of the olfactory pathway. The GAL4 drivers used and abbreviations of their expression domains are shown below the bar graph. Note that there is some variation in the scores of the control genotypes because of using different GAL4 drivers. Pn: projection neurons, APLn: anterior paired lateral neuron, DPMn: dorsal paired medial neuron, CC: central complex, DAn: dopaminergic neurons, In: inhibitory neurons. PIs are the mean ± SEM with n = 12. Two-tailed, two-sample Student’s t tests for each driver expressing miR-92aSP compared with expression of the scrambled control, *p < 0.05. B, MiR-92a spatial mapping to subsets of MBns. Two independent drivers were used for each type of MBn: α’β’, γ, and αβ. The GAL4 drivers employed and their expression in the three broad types of MBns are shown below the graph. PIs are the mean ± SEM with n = 12. Two-tailed, two-sample Student’s t tests for each driver expressing miR-92aSP compared with expression of the scrambled control, *p < 0.05, **p < 0.01. C, Overexpression of miR-92a does not affect 3-h memory. MiR-92a was overexpressed in αβ and γ MBns using R25H11-GAL4. Three-hour olfactory memory performance of R25H11-GAL4>UAS-miR-92a flies was compared with GAL4-only and UAS-only controls. PIs are the mean ± SEM with n = 14. One-way ANOVA followed by Bonferroni’s post hoc tests. n.s., not significant. D, Memory enhancement occurs from miR-92a inhibition during both development and adulthood as assayed using the Gene-Switch system. Administration of RU486 during development (devel), adulthood (adult), or both (devel + adult) modulates mushroom body expression of miR-92aSP controlled by the MB-Gene-Switch driver. Note that RU treatment either during development or adulthood alters the PIs of the control genotype. PIs are the mean ± SEM with n = 6–8. Two-tailed, two-sample Student’s t tests for each RU486 feeding condition, *p < 0.05, ***p < 0.001.
- Figure 3.
Inhibiting miR-92a enhances intermediate-term memory. A, Memory decay in flies expressing miR-92aSP in the mushroom bodies. Three-minute, 1-, 2-, 3-, 6-, and 9-h memory was tested in flies expressing UAS-miR-92aSP or UAS-scrambled under the control of R13F02-GAL4. PIs are the mean ± SEM with n = 6–8. Two-tailed, two-sample Student’s t tests between the groups at each time point, *p < 0.05. B, No effect of miR-92a inhibition on memory acquisition. Three-minute memory of R13F02-GAL4>UAS-scrambled or R13F02-GAL4>UAS-miR-92aSP flies tested after 1, 2, 3, 4, 6 and 12 shock-training paradigms. PIs are the mean ± SEM with n = 6. Two-tailed, two-sample Student’s t tests for each shock group of flies expressing miR-92aSP compared with expression of the scrambled control. No significant differences were found.
- Figure 4.
Inhibiting miR-92a enhances memory consolidation. A, Three-hour memory performance of R25H11-GAL4>UAS-scrambled or R25H11-GAL4>UAS-miR-92aSP flies with or without cold shock induced amnesia applied 2-h after aversive training. Cold-shocked flies exhibited increased performance, indicating enhanced ARM. PIs are the mean ± SEM with n = 14–16. Two-tailed, two-sample Student’s t tests, *p < 0.05. B, Three-hour memory performance of R25H11-GAL4>UAS-scrambled or R25H11-GAL4>UAS-miR-92aSP flies subjected to cold shock induced amnesia at various time points after aversive training. PIs are the mean ± SEM with n = 12–16. Two-tailed, two-sample Student’s t tests, *p < 0.05.
- Figure 5.
MiR-92a mRNA target prediction. A, Design of the mRNA target prediction pipeline. Top predictions for DIANA and TargetScan have 11 genes in common, six of which are also predicted by miRecords. Five of these potential mRNA targets have human homologs that are predicted targets for human miR-92a. B, khc73 RNAi expression produces memory impairment. Only khc73 of the five candidates (Table 1) significantly impaired 3-h memory using RNAi driven by pan-neuronal nSyb-GAL4. PIs are the mean ± SEM with n = 10. Two-tailed, two-sample Student’s t tests, *p < 0.05. C, UAS-khc73 overexpression enhances 3-h memory. Expression of UAS-khc73 driven by R25H11-GAL4 significantly enhanced 3-h memory compared with UAS-only and GAL4-only controls. PIs are the mean ± SEM with n = 10. One-way ANOVA followed by Bonferroni’s post hoc tests, *p < 0.05, **p < 0.01. D, Overexpressing khc73 using R25H11-GAL4 enhances ARM. Flies were subjected to cold shock 2-h after aversive conditioning and memory was tested at 3-h. Consistent with miR-92a inhibition, overexpressing khc73 enhances ARM. PIs are the mean ± SEM with n = 12. Two-tailed, two-sample Student’s t tests, *p < 0.05, ***p < 0.001. E, Overexpressing khc73 during development (devel), adulthood (adult), or both (devel + adult) enhances adult memory as assayed using the Gene-Switch system. Administration of RU486 increases mushroom body expression of khc73 controlled by the MB-Gene-Switch driver. Similar to the effect observed in Figure 2D, RU treatment either during development or adulthood alters the PIs of the control genotype. PIs are the mean ± SEM with n = 16–20. Two-tailed, two-sample Student’s t tests for each RU486 feeding condition, *p < 0.05.
- Figure 6.
Memory performance of flies overexpressing khc73 with and without the endogenous 3’UTR. A, Three predicted miR-92a binding sites in the endogenous 3’UTR are schematized. Renilla luciferase reporter was cloned upstream of the 2.7-kb khc73 3’UTR and used for the experiment shown in C. Nucleotides with predicted miR-92a binding are shown in red font. All three miR-92a sites were altered to generate a mutant khc73 3’UTR that is unresponsive to miR-92a repression (shown with orange font) as a negative control for luciferase experiments. B, Overexpressing khc73 with the endogenous 3’UTR eliminates the memory enhancement gained without the 3’UTR. Expression of HA-tagged khc73 transgenes with or without 2.9 kb of its own 3’UTR confers normal and enhanced memory performance, respectively. PIs are the mean ± SEM with n = 14. One-way ANOVA followed by Bonferroni’s post hoc tests, *p < 0.05, ***p < 0.001. C, Overexpressing miR-92a represses expression of a khc73 3’UTR luciferase reporter. The Renilla luciferase signal was normalized to firefly luciferase for each transfection. C. elegans miR-239b served as negative and Drosophila miR-310 as positive controls. Renilla/firefly ratios were normalized to negative controls for each reporter construct. Relative repression ratios ± SEM with n = 9. One-way ANOVA followed by Bonferroni’s post hoc tests, ***p < 0.001, n.s., not significant. wt: wild-type. D, miR-92a inhibition derepresses HA-KHC73 signal. HA-Khc73 transgenes with or without the endogenous 2.9-kb 3’UTR element were overexpressed in MBn using the R13F02-GAL4 driver. The presence of the endogenous 3’UTR constraints the HA-KHC73 signal. Very weak signals in the mushroom bodies were detected (middle column) and quantified (bar plot). This weak signal was significantly derepressed when the miR-92aSP was co-expressed (right column). The bar plot shows the quantification of R13F02>scrambled, HA-khc73 + 3’UTR versus R13F02>miR-92aSP, HA-khc73 + 3’UTR brains. ROI was taken from the horizontal lobes by reference to anti-Scribble signal. HA/Scribble ratios were used to minimalize brain to brain variability. Ratios are the mean ± SEM with n = 26–30. Two-tailed, two-sample Student’s t tests, *p < 0.05.
- Figure 7.
Memory enhancement of miR-92a+/− mutant flies requires D2R, type 2 dopamine receptor. Three-hour memory scores of miR-92a+/−, D2R−/−, and D2R−/−; miR-92a+/− double mutants were tested and compared with wCS10 control using one-way ANOVA followed by Bonferroni’s post hoc tests, *p < 0.05, n.s., not significant. PIs are the mean ± SEM with n = 20–23.
Tables
Gene name Flybase ID RNAi line PI for nSyb-GAL4 screen Retest score p value CG12024 CG12024 GD-20143 0.45 ± 0.02 CG8360 CG8360 GD-41643 0.40 ± 0.08 cpr50Ca CG13338 KK-100317 0.64 ± 0.07 0.56 ± 0.08 0.15 khc-73 CG8183 KK-105984 0.38 ± 0.09 0.28 ± 0.07 0.02* crebA CG7450 KK-110650 0.37 ± 0.06 0.37 ± 0.08 0.09 Candidate mRNA targets for miR-92a were selected by applying the pipeline described in Figure 5A. Five genes were screened for a role in memory formation using an RNAi approach. The five lines from the Vienna Drosophila RNAi left (https://stockcenter.vdrc.at/control/main) that were tested are listed. RNAi lines were crossed to the nSyb-GAL4 driver and tested for 3-h memory with n = 4. Each individual RNAi line was compared with a daily nSyb-GAL4>UAS-dcr-2 control in the respective GD or KK background. The average performance index for the nSyb-GAL4>UAS-dcr-2 control was 0.36 ± 0.1 for the GD control (n = 4) and 0.49 ± 0.04 for the KK control. Three lines with a trend for a significant effect on memory were retested with n = 6. Only the khc-73 RNAi line, had a PI significantly lower than the control (n = 10). Results shown are the mean ± SEM. Two-tailed, two-sample Student’s t tests, *p < 0.05.
Extended Data Figure 1-1
Odor and shock avoidance of miR-92aSP-expressing and miR-92a heterozygous flies. A, The c155-GAL4>UAS-miR-92aSP flies avoided ben, oct and electric shock stimuli indistinguishably from c155>UAS-scrambled flies. Avoidance Index: AI, AIs are the mean ± SEM with n = 6. Two-tailed, two-sample Student’s t tests for each condition. No significant differences were detected. B, miR-92a heterozygotes for Clones D and N flies avoided ben, oct and electric shock stimuli indistinguishably from the wCS10 control. AIs are the mean ± SEM with n = 6. Two-tailed, two-sample Student’s t tests for each condition. No significant differences were detected. Download Figure 1-1, EPS file.
Extended Data Figure 2-1
Odor and shock avoidance of miR-92aSP and khc73 overexpression flies. A–D, Flies carrying UAS-miR-92aSP driven by R13F02-GAL4, R11D09-GAL4, R28H05-GAL4, or R25H11-GAL4 avoided ben, oct, and electric shock stimuli indistinguishably from flies carrying UAS-scrambled and the same drivers. AIs are the mean ± SEM with n = 6. Two-tailed, two-sample Student’s t tests for each condition. No significant differences were detected. E, R25H11-GAL4>UAS-khc73 animals avoided ben and oct indistinguishably from GAL4-only and UAS-only controls. The UAS-khc73/+ control group exhibited an increased avoidance of shock stimuli compared to the R25H11/+ control and R25H11-GAL4>UAS-khc73 experimental flies. This increased avoidance fails to explain the memory enhancement of the experimental genotype. AIs are the mean ± SEM with n = 6. One-way ANOVA followed by Bonferroni’s post hoc tests, *p < 0.05. Download Figure 2-1, EPS file.
Extended Data Figure 3-1
MiR-92aSP memory retention, acquisition and cold shock experiments with mushroom body γ and αβ drivers. A, miR-92a inhibition enhances intermediate-term memory. Inhibiting miR-92a in γ MBns using R11D09-GAL4 enhances 1- and 3-h memory. Inhibiting miR-92a in αβ MBns using the R28H05-GAL4 driver enhances 1-, 3-, and 6-h memory. The PI for UAS-miR-92aSP-expressing flies was compared to the PI for UAS-scrambled-expressing flies at each time point. PIs are the mean ± SEM with n = 12. Two-tailed, two-sample Student’s t tests for each time point, *p < 0.05, **p < 0.01. B, miR-92a inhibition does not affect acquisition. Three-minute memory of R11D09-GAL4 or R28H05-GAL4 driven miR-92aSP flies tested after receiving varying numbers of shock pulses with a 1-min CS+ odor presentation. PIs are the mean ± SEM with n = 6–12. Two-tailed, two-sample Student’s t tests. No significant differences were detected. C, Inhibiting miR-92a in γ and αβ MBns enhances ARM. Flies were subjected to cold shock 2 h after aversive conditioning, and memory was tested at 3 h. Memory of R11D09-GAL4 and R28H05-GAL4 driven miR-92aSP flies was tested with or without cold shock treatment. PIs are the mean ± SEM with n = 16–20. Two-tailed, two-sample Student’s t tests, *p < 0.05, **p < 0.01. Download Figure 3-1, EPS file.
In this issue
