Data from: Neuroligin 3 highlights the sexually dimorphic neural substrates of Drosophila social spacing
Data files
Jul 07, 2026 version files 85.34 KB
-
Excell_worksheet_for_Dryad-July_7.xlsx
67.37 KB
-
README.md
17.97 KB
Abstract
In Drosophila melanogaster, Neuroligin 3 (Nlg3), orthologous to autism-related Neuroligins in humans, supports neuronal development and regulation, but also affects social behaviour. Nlg3 is localized to the mushroom bodies, protocerebral bridge, and the optic lobes, and those structures determine social spacing differently by sex. Reducing acetylcholine release from the MB affects social space in both sexes, while dopamine receptor changes in the MB affect males. Additionally, fruitless sexually dimorphic neurons affect social behavior differently in males and females. These findings highlight how conserved synaptic proteins and sexually dimorphic neural substrates regulate social behaviour.
This data set contains all of the raw data used in the associated manuscript.
https://doi.org/10.5061/dryad.931zcrk01
Description of the data and file structure
The data in this document contain social space data, climbing data, and qRT-PCR data.
File: Excell_worksheet_for_Dryad-July_7.xlsx
Description of data organization:
The Excel file is organized in different worksheets, as follows:
On each worksheet, the data are organized by Figure number, and the variables are always as follows:
- Independent variable: genotype or temperature in the columns
- Independent variable: sex in the rows
- Dependent variables: each replicate performed is in the cells:
- Social space expressed in the number of flies within 4 body lengths (4BL)
- Climbing performance (% of flies in the upper vial)
- Gene expression expressed as a mean fold-change (± SEM) from 3 biological replicates with 2-3 technical replicates each, with Tubulin84 as the reference gene
Worksheet descriptions:
- The worksheet entitled "Main Text Data" contains the data used to create the following Figures:1J,K; 2-7; and statistical analysis reported in Supplemental Table 2
- The worksheet entitled "Sup Fig 3 – qRT-PCR-Western blots": contains the data used to create Supplemental Figure 3.
- The worksheet entitled "Sup climbing": contains the data used to create Supplemental Figures 7-10, and statistical analysis reported in Supplemental Table 3
- The worksheet entitled "Sup Social Space": contains the data used to create Supplemental Figures 5-6 and 11-14, and statistical analysis reported in Supplemental Table 4
- The worksheet entitled "Sup Tables" contains Supplemental Tables 2, 3 and 4. See description of these below the table.
Description of data presented:
Figure 1, J-K: Nlg3 expression is important for social space. Social space of flies with RNAi knockdown of nlg3 targeted specifically to all nlg3-expressing cells. There is a decrease in the average number of flies within 4 body lengths (4BL) in male (J; one-way ANOVA: F3,32 = 1.771, p = 0.0003, post-hoc Šidák’s test p<0.0001 when comparing nlg3>UAS-RNAi nlg3 to all controls) and female (K; one-way ANOVA: F3,28 = 3.967, p < 0.0001, post-hoc Šidák’s test p<0.0001 when comparing nlg3>UAS-RNAi nlg3 to all controls).
Figure 2. Knocking down Nlg3 specifically in each of the mushroom bodies, the protocerebral bridge, or the optic lobes increases fly social space in males and females. A-C: Average number of flies within 4 body lengths (4BL) in male and female flies with RNAi knockdown of nlg3 targeted specifically to the mushroom bodies (A), protocerebral bridge (B), and the optic lobes (C). A. The average number of flies within 4BL decreased in both males and females in flies with nlg3 knockdown targeted to the MB (two-way ANOVA – Effect of genotype: F1,39 = 14.71, p = 0.0004). B. The average number of flies within 4BL decreased in female flies with nlg3 knockdown targeted to the PB (two-way ANOVA – Effect of genotype: F1,40 = 19.02, p < 0.0001, with interaction between sex and genotype: F1,40 =5.607 , p < 0.0228, driven mostly by the females, post hoc Šídák' test comparing females p=0.0001). C. Knocking down nlg3 in the OL decreased the average number of flies within 4BL in both males and females compared to genetic control flies (two-way ANOVA – Effect of genotype: F1,31 = 7.191, p = 0.0116). The results of all two-way ANOVA and post-hoc tests performed can be found in Supp. Table 2.
Figure 3. Manipulation of neuronal transmission within the brain structures enriched with Nlg3 increases fly social space. A-C: Average number of flies within 4 body lengths (4BL) in male and female flies when genetically hyperactivating or silencing neural transmission within the MB (A), PB (B), and nlg3-containing neurons (C). A. Hyperactivation of MB neurons decreases the number of flies within 4BL in both males and females compared to basal transmission of MB neurons (two-way ANOVA – Effect of neuronal transmission: F1,47 = 7.395, p = 0.0091). Silencing MB neurons also decreases the average number of flies within 4BL in both males and females compared to basal transmission (two-way ANOVA – Effect of neuronal transmission: F1,30 = 14.40, p = 0.0007). B. The average number of flies within 4BL was not different in males and females when the PB neurons are hyperactivated (two-way ANOVA - Effect of neuronal transmission: F1,74 = 17.72, p = 0.2018) or silenced (two-way ANOVA – Effect of neuronal transmission: F1,31 = 3.443, p = 0.0730). C. Hyperactivation of nlg3-containing neurons decreases the average number of flies within 4BL in both males and females compared to basal transmission (two-way ANOVA – Effect of neuronal transmission: F1,28 = 12.70, p = 0.0013). There is an interaction between sex and silencing nlg3-containing neurons where females have a decrease in the average number of flies within 4BL (two-way ANOVA – Effect of interaction: F1,22 = 6.169, p = 0.0211, post hoc Šídák' test comparing females p=0.0042). The results of all two-way ANOVA and post-hoc tests performed can be found in Supp. Table 2.
Figure 4. Knocking down the acetylcholine rate-limiting biosynthesis enzyme ChAT in the MB decreases fly social space in both sexes. Average number of flies within 4 body lengths (4BL) in male and female flies with RNAi knockdown of ChAT for acetylcholine synthesis specifically in the mushroom bodies. Both male and female flies have an increased average number of flies within 4BL with ChAT knockdown when compared to the genetic control flies (two-way ANOVA – Effect of genotype: F1,32 = 11.19, p = 0.0021). The results of all two-way ANOVA and post-hoc tests performed can be found in Supp. Table 2.
Figure 5. RNAi manipulation of all dopamine receptors in the MB increases social space in males. A-D: Average number of flies within 4 body lengths (4BL) in male and female flies RNAi manipulations of dopamine receptors Dop1R1 (A), Dop1R2 (B), Dop2R (C), and DopEcR (D) specifically in the mushroom bodies. All receptors’ knockdowns in the MB result in a decreased average number of flies within 4BL compared to genetic controls, with a large effect size in males, and a small to no effect in females. A. Dop1R1 (Two-way ANOVA – Effect of genotype: F1,32 = 68.23, p < 0.0001). B. Dop1R2 (Two-way ANOVA – Effect of genotype: F1,32 = 7.401, p = 0.0105). C. Dop2R (Two-way ANOVA – Effect of genotype: F1,32 = 41.10, p < 0.0001). D. DopEcR (Two-way ANOVA – Effect of genotype: F1,32 = 8.905, p = 0.0054). The results of all two-way ANOVA and post-hoc tests performed can be found in Supp. Table 2.
Figure 6. Hyperactivation of fruP1 neurons increases social space in males and females while silencing fruP1 neurons only decreases social space in males. Average number of flies within 4 body lengths (4BL) in males and females when genetically hyperactivating neuronal transmission in fruP1 neurons. Hyperactivating fruP1 neurons decreases the average number of flies within 4BL in both males and females compared their basal transmission (two-way ANOVA – Effect of neuronal transmission: F1,33 = 22.65, p < 0.0001). Silencing fruP1 neurons did not alter female social space but decreased the average number of flies within 4BL in males (two-way ANOVA with Šídák's post-hoc test: p = 0.0453). The results of all two-way ANOVA and post-hoc tests performed can be found in Supp. Table 2.
Figure 7. Knocking down nlg3 in fruP1 neurons decreases only females’ social space, while a reduction of acetylcholine and GABA in fruP1 neurons decrease social space in both sexes. A-D: Average number of flies within 4 body lengths (4BL) in male and female flies with RNAi knockdown of nlg3 (A), TH for dopamine synthesis (B), ChAT for acetylcholine synthesis (C) or GAD1 for GABA synthesis (D) specifically in fruP1 neurons. A. Knocking down nlg3 in fruP1 increases the average number of flies within 4BL in females, with no effect in males (two-way ANOVA – interaction between sex and genotype F1,57 = 4.09, p = 0.0478; Šídák's post hoc comparing the genotypes in females p<0.0001). B. TH knockdown in fruP1 neurons does not change social space in either sex (two-way ANOVA – Effect of genotype: F1,29 = 0.413, p = 0.5255). C. ChAT knockdown in fruP1 neurons increases the average number of flies within 4BL in both males and females (two-way ANOVA – Effect of genotype: F1,47 = 99.11, p < 0.0001). D. Similarly, GAD1 knockdown in fruP1 neurons increases the average number of flies within 4BL in both males and females (two-way ANOVA – Effect of genotype: F1,42 = 8.478, p = 0.0057). The results of all two-way ANOVA and post-hoc tests performed can be found in Supp. Table 2.
Supplemental Figure 3: Efficacy of UAS-nlg3 RNAi. (A) Relative expression of nlg3 mRNA in male (M) and female (F) fly heads (3–5 days old) of either UAS-RNAi nlg3/+ or nlg3-Gal4>UAS-RNAi nlg3 measured by qPCR. Data represent mean fold-change (± SEM) from 3 biological replicates with 2-3 technical replicates each, with Tubulin84 as the reference gene; statistical evidence was assessed using an unpaired Welch t-test. Number of replicates indicated in the bars. (B) Nlg3 Protein abundance in mixed sex fly heads (3–5 days old) of either UAS-RNAi nlg3/+ or nlg3-Gal4>UAS-RNAi nlg3. Data represent abundance relative to total protein levels, from only 1 biological replicate, as we ran out of the anti-body.
Supplemental Figure 5: Effect of temperature on social space in the heterozygous driver lines used. A-C: Average number of flies within 4 body lengths (4BL) in male and female flies when testing at 25oC or 28oC drivers of the MB (A), OL (B), PB (C), or fruP1 (D). Interestingly, in the PB/+ we do not detect an effect of sex, probably because of the influence of that genetic background. Finally, only in the MB/+ do we see a change in social space, with females becoming closer at higher temperature (effect of temperature depending on sex 2-way ANOVA, Interaction between sex and temperature p=0.0366, driven by the females: p=0.04 in a Šidák’s post-hoc). However, in our manipulations (Fig. 3), we saw that the females driving either shits or TrpA1 were further apart, so this effect of temperature on MB/+ is not underlying the changes we report in Fig. 3.
Supplemental Figure 6: Elevated temperature on Cs (+/+), Trp1A/+, and shits/+ does not affect social space or climbing. The genetic controls of the temperature sensitive constructs were tested in social space with the average number of flies within 4 body lengths A-C: and climbing with a performance of percentage of flies reaching the top vial D-F: in both male and female which reveals the construct alone does not affect social space or climbing.
Supplemental Figure 7: Knocking down nlg3 specifically in the protocerebral bridge and optic lobes decreases fly climbing in males and females but not when knocking down in the mushroom bodies. A-C: Startle-induced climbing represented by the mean percentage of flies reaching the top vial (performance index). Genetic constructs include RNAi knockdown of nlg3 targeted specifically to the mushroom bodies (A), protocerebral bridge (B), and the optic lobes (C). Grey colour: genetic control RNAi/+; pink colour: driver>RNAi.
Supplemental Figure 8: Hyperactivation of all the brain structures enriched with Nlg3 decreases fly climbing. A-C: Startle-induced climbing represented by the mean percentage of flies reaching the top vial (performance index). Hyperactivation or silencing neurons specifically to the mushroom bodies (A), protocerebral bridge (B), and all nlg3-expressing neurons (C).
Supplemental Figure 9: Hyperactivation and silencing of fruP1 neurons decreases climbing in males and females. Startle-induced climbing represented by the mean percentage of flies reaching the top vial (performance index). Hyperactivation or silencing of neurons specifically in the fruP1 neuron subset. Grey colour bars: inactivated genetic construct and basal transmission of neurons; green colour bars: activation of genetic construct and hyperactivation of neurons with Gal4 driver; pink colour bars: activation of genetic construct and silencing of neurons with Gal4 driver.
Supplemental Figure 10: Knocking down nlg3 in fruP1 neurons decreases male climbing, while a reduction of dopamine, acetylcholine, or GABA in fruP1 neurons all decrease climbing in male and female flies. A-D: Startle-induced climbing represented by the mean percentage of flies reaching the top vial (performance index). Genetic constructs include RNAi knockdown targeted specifically to the fruP1 subset of neurons of nlg3 (A), knocking down dopamine (B), acetylcholine (C), and GABA (D).
Supplemental Figure 11: Knocking down nlg3 specifically in the optic lobes increases fly social space in both sexes but no changes are seen when knocking down in the protocerebral bridge or mushroom bodies comparing to both genetic controls with different genetic backgrounds. Average number of flies within 4 body lengths (4BL) in male and female flies with RNAi knockdown of nlg3 targeted specifically to the mushroom bodies (A), protocerebral bridge (B), and the optic lobes (C). A. The number of flies within 4BL does not change in both males and females in flies with nlg3 knockdown targeted to the MB. Comparing the knockdown line to both control lines shows an intermediate phenotype. B. The average number of flies within 4BL does not change in both males and females in flies with nlg3 knockdown targeted to the PB. Comparing the knockdown line to both control lines shows an intermediate phenotype. C. Knocking down nlg3 in the OL decreases the number of flies within 4BL in both males and females compared to genetic control flies (two-way ANOVA – Effect of genotype: p = 0.0318).
Supplemental Figure 12. Knocking down the acetylcholine rate-limiting biosynthesis enzyme ChAT in the MB decreases female social space when comparing to both genetic controls with different genetic backgrounds. Average number of flies within 4 body lengths (4BL) in male and female flies with RNAi knockdown of ChAT for acetylcholine synthesis specifically in the mushroom bodies. The interaction between sex and ChAT knockdown shows that female flies have an increased number of flies within 4BL when compared to both the genetic control flies (two-way ANOVA – Effect of interaction: p = 0.0079). Comparing the female ChAT knockdown line to both control lines shows an intermediate phenotype.
Supplemental Figure 13: RNAi manipulation of all dopamine receptors in the MB increases male social space when comparing with both genetic controls with different genetic backgrounds. The main conclusions remain unchanged when comparing with both genetic controls from the V10 background even with different genetic backgrounds. Average number of flies within 4 body lengths (4BL) in male and female flies knocking down dopamine receptors show that Dop receptors knockdown in the MB of males has a decreased number of flies within 4BL compared to genetic controls. A. Dop1R1 (Two-way ANOVA – Effect of genotype: F1,32 = 19.61, p < 0.0001). B. Dop1R2 (Two-way ANOVA – Effect of genotype: F1,32 = 41.10, p < 0.0001). C. Dop2R (Two-way ANOVA – Effect of genotype: F1,32 = 41.10, p < 0.0001). D. DopEcR (Two-way ANOVA – Effect of genotype: F1,32 = 8.905, p = 0.0054).
Supplemental Figure 14: Knocking down nlg3 in fruP1 neurons decreases female social space, while a reduction of acetylcholine in fruP1 neurons decreases social space in both sexes when comparing with both genetic controls with different genetic backgrounds. Average number of flies within 4 body lengths (4BL) in male and female flies with RNAi knockdown of nlg3 (A), TH for dopamine synthesis (B), ChAT for acetylcholine synthesis (C) or GAD1 for GABA synthesis (D) specifically in fruP1 neurons. A. The interaction between sex and nlg3 knockdown shows that female flies have a decreased number of flies within 4BL when compared to both the genetic control flies (two-way ANOVA – Effect of interaction: p = 0.0204). Comparing the male nlg3 knockdown to both control lines shows an intermediate phenotype. B. TH knockdown in fruP1 neurons does not change social space in males and females. C. ChAT knockdown in fruP1 neurons increases the number of flies within 4BL in both males and females compared to the genetic control flies (two-way ANOVA – Effect of genotype: p = 0.0030). D. The interaction between sex and GAD1 knockdown shows that female flies have an increased number of flies within 4BL when compared to both the genetic control flies (two-way ANOVA – Effect of interaction: p = 0.0041). Male GAD1 knockdown in fruP1 neurons shows an intermediate phenotype when compared to both genetic controls.
Code/software
Not applicable
Access information
Other publicly accessible locations of the data:
- N/A
Data was derived from the following sources:
- N/A
Social Space Assay
The assay was performed as previously described (McNeil et al. 2015; Yost et al. 2020). In short, flies were collected after eclosion and group housed until aged to 3-4-days old. Yost et al. (2024) have shown that most of the flies are no longer virgin at this age, matching the treatment conditions used for these experiments. Twenty-four hours prior to the assay, flies were sexed under cold anesthesia and separated into groups of 12-17 in vials with food for each treatment tested. Two hours before the assay, each treatment was transferred into new vials and acclimated to standard conditions (25°C and 50% humidity). At the time of testing, flies were loaded into the triangular chamber and knocked to the bottom of the chamber to induce a startle response that forces the flies into a group at the top of the chamber within the first minute. Flies were then given 20-40 minutes to settle at their preferred social space. Each treatment was loaded into the social space apparatus and photos were taken as soon as flies had settled (McNeil et al. 2015). Social space assay photos were processed with ImageJ (National Institute of Health, Bethesda, MA, USA) to calculate the average number of flies within a four-body length radius (4BL or ~1cm) of each fly in the chamber using macros (available in Yost et al., 2020 and McNEIL et al. 2015). Image analysis data were imported to GraphPad Prism 10 software for statistical analysis and graphical representation.
Climbing Assay
The assay was performed as previously described (McNeil et al. 2015; Yost et al. 2020). In this assay we are testing flies’ stamina, ability to respond to startle, and locomotion. Any fly line with individuals able to reach the top vial (performance index, PI, more than zero) in 15 seconds does not have impairments preventing testing in the social space assays (McNeil et al. 2015; Yost et al. 2020). Flies were aged to 2-4-days old and 24 hours prior to conducting the assay were sexed under cold anesthesia and separated in vials with food containing 40 flies for each treatment. Two hours before the assay, each treatment was transferred into new vials and acclimated to testing conditions (25⁰C and 50% humidity). Each treatment was transferred into a testing tube and inserted into the climbing assay. Flies were forced to the bottom of the tube and given 15 seconds to climb to the tube on top of the apparatus. This short timeframe was chosen to allow approximately 80% of control flies to reach the top vial. After 15 seconds, the apparatus was closed and the number of flies in the top and bottom tubes were counted to calculate a performance index representing the percentage of flies that reached the top vial.
Temperature sensitive experiments
For these experiments, we used temperature-activated genetic constructs to compare the effect of altered neuronal transmission on social space. Temperature-sensitive treatment groups were assayed at 50% humidity and their respective activation temperatures: 27°C for TrpA1 flies to hyperactivate neurons, and 29°C for shits flies to inhibit neurons. Simultaneously, control flies of the same genotype were assayed at 25°C so their neuronal transmission remained unaltered. Social spacing of all genetic background lines used to create the experimental lines was not affected by temperature changes (Supp Fig. 3 & 4), or in one case, an effect in the opposite direction from what was observed in the experiments (females MB/+ are sightly closer at 28°C Fig. 3A, but further apart when used to drive shits or TrpA1 Fig 3B).
Statistical Analysis
Statistical analysis and graphical representation were performed in GraphPad Prism (RRID:SCR_002798, version 7.0a for Mac, GraphPad Software, La Jolla, CA, USA, www.graphpad.com). Prior to performing statistical tests, normality and homoscedasticity tests were used to assess whether the data followed a gaussian distribution, and the variance differed between the groups compared. When normality was confirmed, the appropriate two-way ANOVA tests were performed, with data that had two independent variables and were collected simultaneously. Following the two-way ANOVA, additional statistical tests were performed, Šídák’s post-hoc tests, to assess the weight of the statistical evidence, especially relevant when interactions between the independent variables is found, while correcting for multiple comparisons (Supp. Table 2). Supp. Tables 3 and 4 report the ANOVA results and subsequent tests of the Supplemental figures. In the rare cases where the same controls were used for two experiments (Fig. 2A and B, as well as Sup. Fig. 5A and B), and that p values were <0.01, we added a p- value corrected for multiple comparisons in the corresponding tables, using the Bonferroni method (dividing the p value by the number of times the control was used for comparison, in all cases it was 2) Those corrections did not affect the interpretations. P-values are reported on the figures when there is statistical support for evidence of effect (Greenland et al. 2016).
RT-qPCR to confirm effect of the UAS-nlg3 RNAi line
Heads from ~20 male or female flies (3–5 days old) were collected separately. Control and RNAi treatment groups were selected to match the genotypes and ages used in the behavioural assays, ensuring consistency across experiments. Total RNA was extracted using TRIzol reagent (#15596026- Invitrogen) according to the manufacturer’s instructions. RNA samples were treated with DNase (TURBOTM DNase-Invitrogen) to eliminate genomic DNA contamination, and cDNA was synthesized using the cDNA Synthesis Kit (#1708841-BIO-RAD). RT-qPCR was performed using the PowerUpTM SYBRTM Green Master Mix (Thermo Fisher Scientific) on a CFX Connect Real-Time PCR Detection System (BIO-RAD). Relative expression levels were calculated using the ΔΔCq method, with Tubulin84 serving as the internal control gene. The qPCR analysis confirmed that nlg3 transcript levels were significantly reduced in RNAi lines compared to controls. Knockdown efficiency was observed in both sexes, with females showing stronger reduction than males, as shown in Supp. Fig. 2.