Royal Society Open Science
Open AccessResearch articles

Recovery from social isolation requires dopamine in males, but not the autism-related gene nlg3 in either sex

Ryley T. Yost

Ryley T. Yost

Department of Biology, Western University, London, Ontario, Canada

Contribution: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Validation, Visualization, Writing – original draft, Writing – review and editing

Google Scholar

Find this author on PubMed

,
Andrew M. Scott

Andrew M. Scott

St Joseph’s Healthcare, Hamilton, Ontario, Canada

Contribution: Formal analysis, Investigation, Methodology, Writing – original draft

Google Scholar

Find this author on PubMed

,
Judy M. Kurbaj

Judy M. Kurbaj

Department of Biology, Western University, London, Ontario, Canada

Contribution: Formal analysis, Investigation, Writing – original draft

Google Scholar

Find this author on PubMed

,
Brendan Walshe-Roussel

Brendan Walshe-Roussel

Department of Biology, Western University, London, Ontario, Canada

Contribution: Investigation, Methodology, Supervision

Google Scholar

Find this author on PubMed

,
Reuven Dukas

Reuven Dukas

Department of Psychology, Neuroscience and Behaviour, Animal Behaviour Group, McMaster University, Hamilton, Ontario, Canada

Contribution: Funding acquisition, Resources, Supervision, Writing – original draft

Google Scholar

Find this author on PubMed

and
Anne F. Simon

Anne F. Simon

Department of Biology, Western University, London, Ontario, Canada

[email protected]

Contribution: Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing – original draft, Writing – review and editing

Google Scholar

Find this author on PubMed

    Abstract

    Social isolation causes profound changes in social behaviour in a variety of species. However, the genetic and molecular mechanisms modulating behavioural responses to social isolation and social recovery remain to be elucidated. Here, we quantified the behavioural response of vinegar flies to social isolation using two distinct protocols (social space preference and sociability, the spontaneous tendencies to form groups). We found that social isolation increased social space and reduced sociability. These effects of social isolation were reversible and could be reduced after 3 days of group housing. Flies with a loss of function of neuroligin3 (orthologue of autism-related neuroligin genes) with known increased social space in a socially enriched environment were still able to recover from social isolation. We also show that dopamine (DA) is needed for a response to social isolation and recovery in males but not in females. Furthermore, only in males, DA levels are reduced after isolation and are not recovered after group housing. Finally, in socially enriched flies mutant for neuroligin3, DA levels are reduced in males, but not in females. We propose a model to explain how DA and neuroligin3 are involved in the behavioural response to social isolation and its recovery in a dynamic and sex-specific manner.

    1. Introduction

    The environment an individual has been exposed to influence social behaviour by affecting the interactions between individuals [1]. Social isolation, or the absence of interactions with others, has profound effects on mental and physical health in humans [24]. During the COVID-19 times, forced social isolation had negative consequences on families and individuals [5]. Higher risk individuals such as children and people with neuropsychiatric disorders, like autism spectrum disorders, face more challenges to their physical and mental health because of changes to their behavioural and environment support [6,7]. It is important to understand the interactions between genetic predispositions and altered social environment, as well as the possibility to recover from social isolation.

    Social isolation causes changes to the social behaviour in several organisms, including a variety of behavioural deficits in monkeys [811]; deficits in social interactions, aggression, fear and anxiety in mice [1218] and decreased social affiliation in bees [19]. In Drosophila melanogaster, social isolation leads to physiological and behavioural changes, as described in recent reviews [2022]. For example, lifespan appears to depend on the social context in a complex manner, as social isolation has been reported to decrease [23] or increase [24] lifespan, possibly depending on the number of flies in the social environment, strains and diet. Additional changes include decreased fibre number in the mushroom bodies [25], changes to neural excitability [26], chemical communication [27,28], olfactory memory [29], courtship and courtship memory [30], sleep patterns [31,32], locomotion [33], circadian rhythm [30], aggression [26,3436], sociability [37] and social network structure [38,39]. Isolated male flies are much more territorial and defend food patches and access to mates more than socially enriched flies [40]. Our work and others’ have identified that isolated flies have increased social space [4143], a measure of inter-individual distances.

    Plastic changes in behaviour are important for an organism to adapt to a changing environment. Changes in social behaviour because of isolation can be partially or fully recovered in monkeys [10,11], mice [13,14,44] and bees [19]. Flies are no exception. In Drosophila, sleep and aggression levels were recovered after group housing [31,45]. However, more research needs to be conducted on the recovery from isolation of other Drosophila social behaviours.

    In addition to social isolation, social space is modulated by social experience [41]; physiological state such as hunger [46], age of the individual and age of the parents [46,47]; time of the day [46,48]; viral infections, as well as exposure to environmental toxins [4953] or anti-oxidant [49]. Orthologues of several candidate genes for neurodevelopmental disorders in humans also influence social space in Drosophila, as reported for example, not exhaustively, in [5461]. Finally, a number of sensory modalities, neurocircuitry and synaptic proteins underlying social spacing have been identified [42,43,4649,54,6168]. One identified synaptic protein is neuroligin3, a post-synaptic cell adhesion molecule that regulates synaptic development and function and is an orthologue to a human autism candidate gene [42,69,70]. We have previously shown nlg3 to be important for social space [42] and sociability (the tendency to engage in non-aggressive interactions with conspecifics [37]). We also showed that nlg3 is required for a typical response to social isolation, but that the NLG3 protein levels are unchanged after social isolation [42]. However, the role of nlg3 in recovery from social isolation has not been studied.

    Another important molecule for social behaviour in the fly is dopamine (DA), a monoamine neurotransmitter [7173]. We and others have previously shown that DA modulates social space [48] and is important for a response to social isolation. Specific dopaminergic neural circuity is in part responsible for the behavioural modification after isolation [43], and DA levels in males are reduced after isolation [31]. This could be because the levels of the transcripts of one of the DA biosynthesis enzymes (dopa decarboxylase) have been found to be reduced in isolated male flies’ purified dopaminergic neurons. In fact, many transcriptional and epigenetic changes occur in dopaminergic neurons in response to isolation [74]. However, the role of DA in social recovery has not been identified. In addition, evidence suggests that nlg3 and DA are part of a similar neurocircuitry or involved in a similar pathway responsible for behaviour regulation [7577]; however, their dual involvement in the regulation of social space needs more attention.

    In this study, we examined the effect of social isolation on two measures of social behaviour: social space [41] and sociability [37]. We further investigated the role of nlg3 and DA in the recovery from social isolation. Finally, we propose a model for the integrated regulation of nlg3 and DA in the behavioural response to the social environment.

    2. Methods

    2.1. Fly stocks and husbandry

    All fly lines were maintained in mixed-sex groups in bottles on JazzMix media or our own food made following the same recipe (brown sugar, corn meal, yeast, agar, benzoic acid, methyl paraben and propionic acid; Fisher Scientific, Whitby, Ontario, Canada) at 25°C, 50% relative humidity on a 12 L : 12 D cycle. All flies reared in bottles for experimental use were a maximum of 14 days old to avoid variation in behaviour resulting from older parents [47]. Fly lines were obtained from the following places: CS was obtained from the laboratory of Seymour Benzer in 1998, nlg3Def1 are from Dr Brian Mozer [42], w;; TH-GAL4 was provided by Dr Serge Birman and RNAi against tyrosine hydroxylase (TH) (w;; UAS-THmiR-G) was a gift from Dr Mark Wu [43]. All lines except w;; UAS-THmiR-G were outcrossed five times to our control line, CS, to reduce variation in behaviour caused by genetic background. Crosses used to generate TH>THmiR-G and their appropriate genetic controls can be found in the electronic supplementary material, figure S1.

    2.2. Generation of isolation and recovery treatments

    Mated flies used in all experiments were collected from bottles at 1 day old and remained mixed-sex in new bottles for one more day (2 days total) to allow mating (see confirmation of mating status below) to avoid the effects of mating status on social space [41]. Following, socially isolated flies were transferred to individual vials using cold anaesthesia and remained single housed for 2, 4 or 7 days. Group-housed age-matched control flies were kept in mixed-sex bottles for the same duration as the isolated flies, such that the group-housed control flies for isolation were 4, 6 and 9-days-old, respectively. To test for a recovery after social isolation, flies isolated for 7 days were transferred to mixed-sex bottles for either 2 or 3 days. Group-housed flies used as a control for the recovery treatment were maintained in mixed-sex bottles until tested with the recovery flies, such that the group-housed control flies for isolation then recovery were 11- and 13-days-old, respectively. To test the effect of group housing density on social space, we sorted the group-housed flies into vials mixed-sex containing 2, 6 or 16 flies and a separate uncontrolled amount (random) in a bottle. The experimental design for our experiments comparing group-housed and isolated or recovered flies can be found in figure 1.

    Schematic of the experimental paradigm and parameters measured in this study.

    Figure 1. Schematic of the experimental paradigm and parameters measured in this study. Flies were group housed in mixed-sex bottles for 2 days after eclosing to adulthood, prior to being single housed or remained group housed. Flies remained single or group housed for up to 7 days (reaching the age of 9-days-old). The length of isolation is always 7 days unless indicated in the figure. After 7 days of being group or single housed, flies were tested in either the social space assay, sociability assay or used to determine DA levels. Separate populations of flies followed the same experimental design but were group housed for up to 3 days (to test recovery, reaching the age of 12-days-old) before behaviour or molecular experiments took place. All flies in each independent repeat were naive to the tests performed, and no flies were used in multiple behavioural or molecular experiments. Created using BioRender.

    To test for the effects of virginity on social isolation, pupae were collected from bottles and placed individually into vials and allowed to eclose over 24 h. Isolated virgin flies remained in individual vials, whereas group-housed virgin flies were transferred into vials containing 15 same-sex flies to maintain virginity.

    2.2.1. Females mating status at 2- and 4-days-old

    One-day-old female flies were collected and maintained with males in bottles either 1 day (24 h) or 3 days. Females were then sexed under cold anaesthesia, and each female was placed alone in a fresh vial. Females were scored as having been mated when the third instar larvae were observed.

    2.2.2. Males mating status at 2- and 4-days-old

    Single, 1-day-old male flies, sexed under cold anaesthesia, were placed with one 4-day-old virgin female in vials (based on the females results, more than 90% of females that age can mate). After either 1 day (24 h) or 3 days, the males were removed, and females remained in isolation until the third instar larvae were counted. For both sexes, vials with dead flies were not counted. We cannot ensure all flies have mated, even in the group housing treatment, so all flies exposed to the initial group housing are used in the behaviour experiments, including the small percentage of non-mated flies.

    2.3. Social behaviour assays

    Sociability tests were performed at McMaster University. All other experiments took place at Western University under similar settings.

    2.3.1. Fly handling prior to behaviour

    Twenty-four hours prior to all behaviour assays, 15–17 flies that were group housed (either lifelong or after a few days of recovery) were collected using cold anaesthesia and placed in vials. Single-housed flies were kept isolated until right before testing. The morning of the experiment, all flies were transferred to new vials and allowed to habituate to the testing conditions of 25°C and 50% relative humidity for at least 2 h. The social space assay was conducted under uniform light in the same room between 13.00 and 17.00 (ZT 4–8) to decrease behavioural variation linked to diel periodicity. Finally, we used an internal control for genetic background in each experiment, as the entire dataset display differences in performances, depending on variables that we cannot control (refer, e.g. figure 1 and Yost et al. [42]).

    2.3.2. The social space assay

    The social space assay was performed as previously described [41,42,66]. In short, once flies have settled at their preferred inter-individual distance, a photo was taken (after 20–50 min, depending on their genotype). Using the open-access software ImageJ (RRID:SCR_003070; [78]), the number of flies within four body lengths (approx. 1 cm) was determined for each fly in the chamber. The number of flies within four body lengths was averaged using all flies in the chamber, representing one individual replicate. This metric has been used in the previous studies [42,43]. The routines for image analysis are publicly available [42]. All datasets are the combination of 1–3 replicates per day and 3–5 independent days of testing. Each independent day was separated by at least a week to control for environmental factors beyond our control.

    2.3.3. Sociability assay

    The sociability chamber consisted of a circular arena (90 mm wide by 20 mm high) divided into eight compartments with a hole in the centre to allow flies to enter any compartment. We added to each compartment a patch of fresh food coated with a layer of grapefruit–yeast suspension (3 g yeast in 100 ml grapefruit juice) to enhance attractiveness. We modified the chamber and performance of the assay from Scott et al. [37]. We transferred by mouth aspiration 16 same-sex flies to the chamber through a hole in the lid and allowed them to acclimate for 1 h. Experimenters blind to treatment then counted the number of flies in each chamber and calculated an aggregation index (sample variance divided by the mean number of flies in each chamber). The variance could take values between 0 and 32. For example, the least sociable option would have eight chambers of two flies each, with a variance of 0 and the corresponding aggregation index of 0. The most sociable situation would have seven chambers of 0, and one chamber with all 16 flies. This would have a variance of 32 and therefore an aggregation index of 16 (32/2). We tested flies in sessions of four replicate arenas per treatment over three consecutive weeks, for 12 arenas total per treatment (total n = 96).

    2.4. Dopamine quantification

    Adults were separated by sex, and DA extracted from their heads using the following procedure. Extraction occurred by flash freezing flies in liquid nitrogen followed by manual decapitation and homogenization of heads in 5 mM of ammonium acetate in 90% acetonitrile using microtissue grinders (Kimble Chase, USA). The supernatant was transferred and filtered through a 0.65 µm filter (Millipore) at 4°C, and samples were stored at −80°C before detection using liquid chromatography/mass spectrometry (LC/MS).

    LC/MS analysis was performed using an Agilent 1260 Infinity LC system coupled to an Agilent 6230 Time of Flight (TOF) system. A XBridge C-18 column Rapid Resolution HT was used (4.6 Å approx. 150 mm, 3.5 μm, 600 bar, waters) at 25°C, and samples were eluted with a gradient of CH3CN (solvent B: 90% CH3CN in H2O, containing 0.1% formic acid) in H2O (solvent A: containing 0.1% formic acid). The UV lamp was set at 282 nm, and the injection volume was 10 µl. The flow rate was set to 0.4 ml min−1 and infused into an Agilent 6230 TOF-MS through a dual-spray electro ionization source with a gas temperature of 325°C flowing at 8 l min−1 and a nebulizer pressure of 35 psi. The fragmentor voltage was set to 175 V with a capillary voltage of 3500 V and a skimmer voltage of 65 V. The instrument was set in positive ESI mode, and quantification occurred using a standard curve of known DA concentrations (electronic supplementary material, figure S2). Total ion count was extracted using the Agilent MassHunter qualitative analysis software (v. B.05.00).

    2.5. Statistics

    Data were stored in Excel files and statistically analysed using GraphPad Prism (RRID:SCR_002798, v. 7.0 a for Mac, GraphPad software, La Jolla, California, USA, www. graphpad.com). We first assessed the normality and homoscedasticity of the data distribution prior to applying the appropriate parametric tests (simple t‐test or ANOVA or Welch t‐test/ANOVA when the standard deviations were different among groups). When performing statistical analysis of an experiment with only two groups, we used a t‐test. When there were more than two groups but only one variable, we used a one-way ANOVA. When we compared two variables across two or more groups, we used two-way ANOVAs. One-way and two-way ANOVAs were followed by Tukey’s or Sidak’s post hoc test (respectively) to identify significantly different points, while correcting for multiple comparisons. We used the traditional alpha level of 0.05 for all statistical tests [79]. On the graphs, we present the p-values for tests, and post hoc tests are performed only when significant. The results of all the comparisons, significant or not, can be found in tables 13.

    Table 1. ANOVA table for all two-way ANOVAs performed. (Statistically significant results are in bold.)

    experiment figure number effect d.f. F p Sidak post hoc
    social space comparison group compared with p
    days of isolation—males 2a days (2,46) 2.412 0.1009 group housed vs single housed 2 days 0.0529
    isolation (1,46) 22.01 <0.0001 4 days 0.0424
    days * isolation (2,46) 0.1534 0.8582 7 days 0.0093
    days of isolation—females 2 b days (2,47) 3.683 0.0327 group housed vs single housed 2 days 0.9992
    isolation (1,47) 6.653 0.0131 4 days 0.1404
    days * isolation (2,47) 1.453 0.2441 7 days 0.0746
    mating status—males 3 a isolation (1,52) 35.69 0.0327 group housed vs single housed mated <0.0001
    mating status (1,52) 4.223 0.0131 virgin 0.0007
    isolation * mating status (1,52) 0.2036 0.2441
    mating status—females 3b isolation (1,52) 9.924 0.0027 group housed vs single housed mated 0.0033
    mating status (1,52) 5.269 0.0258 virgin 0.4308
    isolation * mating status (1,52) 2.131 0.1504
    Cs & nlg3 Def1 isolated—males 4a genotype (1,32) 10.1 0.0033 group housed vs single housed Cs 0.0001
    isolation (1,32) 46.48 <0.0001 nlg3 Def1 <0.0001
    genotype * isolation (1,32) 0.0086 0.9267
    Cs & nlg3 Def1 recovery—males 4b genotype (1,37) 39.94 0.0002 group housed vs recovery Cs 0.9213
    recovery (1,37) 17.72 <0.0001 nlg3 Def1 <0.0001
    genotype * recovery (1,37) 13.62 0.0007
    Cs & nlg3 Def1 isolated—females 4c genotype (1,32) 13.71 0.0008 group housed vs single housed Cs <0.0001
    isolation (1,32) 27.97 <0.0001 nlg3 Def1 0.0769
    genotype * isolation (1,32) 5.37 0.027
    Cs & nlg3 Def1 recovery—females 4d genotype (1,34) 6.178 0.018 group housed vs recovery Cs 0.1326
    recovery (1,34) 0.3197 0.5755 nlg3 Def1 0.0197
    genotype * recovery (1,34) 10.6 0.0026
    TH-miR isolation—males 5a genotype (2,35) 1.633 0.2099 group housed vs single housed th-gal4/+ 0.9851
    isolation (1,35) 4.631 0.0384 UAS-th-miR-G /+ 0.0003
    genotype * isolation (2,35) 8.635 0.0009 th>th-miR-G 0.5283
    TH-miR recovery—males 5b genotype (2,37) 9.018 0.0006 group housed vs recovery th-gal4/+ >0.9999
    recovery (1,37) 4.631 0.7555 UAS-th-miR-G/+ 0.999
    genotype * recovery (2,37) 8.635 0.8263 th>th-miR-G 0.8707
    TH-miR isolation—females 5c genotype (2,37) 4.009 0.0265 group housed vs single housed th-gal4/+ 0.0188
    isolation (1,37) 21.91 <0.0001 UAS-th-miR-G/+ 0.0025
    genotype * isolation (2,37) 0.8892 0.4196 th>th-miR-G 0.2928
    TH-miR recovery—females 5d genotype (2,35) 5.124 0.0112 group housed vs recovery th-gal4/+ 0.8403
    recovery (1,35) 2.562 0.1184 UAS-th-miR-G/+ 0.0667
    genotype * recovery (2,35) 2.036 0.1457 th>th-miR-G 0.9661
    dopamine levels comparison group compared with p
    Cs 5e sex (1,27) 1.665 0.2078 males grouped housed (GH) – 9 days old (9do) males SH 9do 0.0305
    social experience (3,27) 3.043 0.0459 males GH 12do 0.2372
    sex * social experience (3,27) 1.13 0.3545 males rec 12do >0.999
    females grouped housed (GH) – 9 days old (9do) females SH 9do 0.9775
    females GH 12do 0.996
    females REC 12 do 0.5151
    Cs and nlg3 Def1 6a sex (1,7) 6.107 0.0428 Cs males Cs females 0.1531
    genotype (1,7) 17.21 0.0043 males nlg3 Def1 0.0347
    sex * genotype (1,7) 2.797 0.1384 females nlg3 Def1 0.0179
    Cs females males nlg3 Def1 0.8203
    females nlg3 Def 0.4884
    nlg3 Def males females nlg3 Def 0.9935
    nlg3 Def1 6b sex (1,16) 0.007 0.999 males grouped housed (GH) – 9 days old (9do) males SH 9do 0.9922
    social experience (3,16) 1.354 0.2922 males GH 12do 0.7168
    sex * social experience (3,16) 0.921 0.3513 males rec 12 do 0.9969
    females SH 9do 0.9798
    females GH 12do 0.6796
    females rec 12 do 0.9989

    Table 2. All Welch’s t-tests were performed. (Two-tailed when no hypothesis for direction and two-tailed when directions of change are hypothesized. Statistically significant results are in bold.)

    experiment figure number comparison one or two tail? d.f. t p
    sociability isolation—males 2c group-housed versus single-housed males two tailed 12.08 4.039 0.0016
    sociability isolation—females 2c group-housed versus single-housed females two tailed 13.37 2.578 0.0225
    social space recovery 2 day—males 2d group-housed versus recovery males two tailed 14.46 0.4815 0.6373
    social space recovery 2 day—females 2d group-housed versus recovery females two tailed 11.3 3.673 0.0035
    social space recovery 3 day—males 2e group-housed versus recovery males one tailed 9.545 1.542 0.0778
    social space recovery 3 day—females 2e group-housed versus recovery females one tailed 12.67 1.465 0.1672
    sociability recovery 3 day—males 2f group-housed versus recovery males one tailed 21.96 0.3738 0.7121
    sociability recovery 3 day—females 2f group-housed versus recovery females one tailed 15.01 1.954 0.0696

    Table 3. ANOVA table for all one-way ANOVAs performed. (Statistically significant results are in bold.)

    experiment figure number d.f. F p Sidak post hoc
    comparison group compared with p
    group-housed density—males 2c (4,40) 30.87 <0.0001 single housed 2 <0.0001
    6 <0.0001
    16 <0.0001
    random <0.0001
    group-housed density—females 2c (4,40) 31.35 <0.0001 single housed 2 <0.0001
    6 <0.0001
    16 <0.0001
    random <0.0001

    3. Results

    3.1. Isolation for 7 days leads to increased social space and decreased sociability

    Seven days of social isolation leads to an increased social space [41]. We wanted to determine whether we could reduce the duration of social isolation and still influence social space.

    Because mating status influences social space and virgin flies are further apart than mated flies [41], we needed to first assess how long flies should be group-housed post-emergence to avoid the confounding effect of virginity on social space. We found that 101 out of 112 (90.2%) of females were mated and 55 out of 72 (76.4%) of males were able to mate after spending 4–5 days, mixed with females (electronic supplementary material, table S1). In comparison, 94 out of 103 (91.2%) of females were mated 2 days after adult emergence spent with males, and 69 out of 81 (85.2%) of males of the same age were able to mate. The ages at which the flies were tested had no significant effect on their mating capabilities (χ² = 0.022, p = 0.88 for females and χ² = 1.92, p = 0.166 for males). We thus decided to expose our flies to 2 days of group housing post-emergence, before socially isolating them for 2, 4 or 7 days, as we now know that the same percentage of flies have mated by that age, as when they are 4–5 days. We then assessed their social space by recording the number of flies present within four body lengths.

    CS (our control line) males had fewer flies within four body lengths after social isolation compared with group-housed flies at all three time points (2, 4 or 7 days) (figure 2a ; two-way ANOVA results in table 1; effect of isolation: F 1,46 = 22.01, p < 0.0001). There was no effect of 2 days of social isolation in CS females. Only after 4 and 7 days of isolation did CS females have fewer flies within four body lengths, and the increase in social space was larger with increasing number of days flies isolated (figure 2b ; two-way ANOVA results in table 1; effect of isolation: F 1,47 = 6.653, p = 0.0131; effect of days of isolation: F 2,47 = 3.683, p = 0.0327). The effect of isolation after 7 days was similar to the effects found in earlier studies [41]. Moving forward, we used 7 days of isolation for the rest of the experiments as 7 days had the largest increase in social space in males and females.

    Isolation for 7 days leads to increased social space and decreased sociability, but flies recover after 3 days of group housing.

    Figure 2. Isolation for 7 days leads to increased social space and decreased sociability, but flies recover after 3 days of group housing. (a,b,d,e) Social space presented in terms of mean number of flies within four body lengths ± s.e.m. (c,f) Effect of 2, 4 or 7 days of isolation in males (a) and females (b) social space. (a) Male flies had fewer flies within four body lengths after 2, 4 and 7 days of isolation (two-way ANOVA and Sidak post hoc test—F 1,46 = 22.01, p < 0.0001). (b) Females had fewer flies within four body lengths after isolation and that number decreased further with increasing days of isolation (two-way ANOVA—effect of isolation: F 1,47 = 6.653, p = 0.0131; effect of days of isolation: F 2,47 = 3.687, p = 0.0367). (c) Aggregation index in males and females after 7 days of isolation. Both males (Welch’s t‐test: t 12.08 = 4.039, p = 0.0016) and females (Welch’s t‐test: t 13.37 = 2.578, p = 0.0225) had a lower aggregation index after 7 days of isolation. (d,e) Males and females after 2 (d) and 3 (e) days of group housing (recovery) following 7 days of isolation. (d) Males showed no difference in the number of flies within four body lengths after 2 days of group housing compared with males continuously group housed. However, females had fewer flies within four body lengths compared with females group housed continuously (Welch’s t‐test: t 15.13 = 2.626, p = 0.0189). (e) Both males and females were not different in the number of flies within four body lengths comparing group-housed and recovery flies (one-way ANOVA: F 3,28 = 0.9434, p = 0.4329). (f) Sociability presented as the mean aggregation index ± s.e.m. of males and females group housed and recovery. Males and females did not differ in aggregation index between group housed and flies isolated 7 days, with 3 days of recovery. n = 7–12 for all treatments. On the graph, the results of the post hoc tests are shown only when p < 0.05. The results of all the comparisons, significant or not, can be found in tables 13. Grey shade: control group housed (GH); coloured shades: treatments; blue shades: single housed (SH) and recovery from social isolation (REC) males; red shades: SH and REC females.

    We next used another measure of social behaviour, sociability [37], to examine the effect of isolation. Both males and females had a lower aggregation index in isolated flies compared with those group housed (figure 2c ; table 2; males—Welch’s t‐test: t 12.08 = 4.039, p = 0.0016; females—Welch’s t‐test: t 13.37 = 2.578, p = 0.0225).

    3.2. Males and females recover from isolation after 3 days of group housing

    As recovery from isolation has been noted in various organisms including D. melanogaster (§1), we tested whether social space and sociability could also be recovered in flies that were group housed following isolation. We started with 2 days of group housing following isolation and found that males were not different in flies within four body lengths indicating they had recovered in social space; however, females still had a lower number of flies within four body lengths in the recovery compared with group-housed flies (figure 2d ; table 2; Welch’s t‐test: t 15.13 = 2.626, p = 0.0189). Because females had not fully recovered after 2 days of group housing post isolation, we tested social space in flies that had 3 days of group housing after 7 days of isolation. Again, males were not different in flies within four body lengths; however, females were now also not different in flies within four body lengths in recovery and group-housed flies indicating that females also can recover from isolation but take longer than males (one more day in this case—figure 2e ; table 3). A period of 3 days of group housing following 7 days of isolation was then used for all other experiments investigating recovery. Finally, we tested sociability in males and females after 3 days of recovery. Males and females did not differ in aggregation index in recovery and group-housed flies, indicating that both sexes recovered (figure 2f ; table 3).

    3.3. Mating status does not alter social space in response to isolation

    We have previously reported that virgin flies have an increased social space [41] as do isolated flies ([42] and figure 1); however, we have not tested virgin flies in response to isolation, so we next tested the effect of isolation on social space in mated and virgin flies. There were less flies within four body lengths after isolation in both mated and virgin males (figure 3a ; two-way ANOVA results in table 1; effect of isolation: F 1,52 = 34.69, p < 0.0001). In addition, virgin males had less flies within four body lengths when group and single housed compared with mated group and single housed (figure 3a ; two-way ANOVA results in table 1; effect of mating status: F 1,52 = 4.223, p = 0.0449). In females, we found that in mated flies, single-housed flies had less flies within four body lengths than group-housed individuals (figure 3b ; two-way ANOVA results in table 1; effect of isolation: F 1,52 = 9.924, p = 0.0027); however, group- and single-housed virgins had similar flies within four body lengths (p = 0.4308). Similar to males, virgin females had less flies within four body lengths when group and single housed compared with mated group and single housed figure 3b ; two-way ANOVA results in table 1; effect of mating status: F 1,52 = 5.269, p = 0.0258).

    Mating status does not change social space in response to isolation and group housed with a single fly is enough to avoid a larger social space.

    Figure 3. Mating status does not change social space in response to isolation and group housed with a single fly is enough to avoid a larger social space. (a,b) Number of flies within four body lengths in males (a) and females (b) group and single housed while mated or virgin. (a) Male flies had less flies within four body lengths after isolation (two-way ANOVA—effect of isolation: F 1,52 = 34.69, p < 0.0001) regardless of mating status. Males had less flies within four body lengths when virgin compared with mated flies (two-way ANOVA—effect of mating status: F 1,52 = 4.223, p = 0.0449). (b) Female flies had less flies within four body lengths after isolation (two-way ANOVA—effect of isolation: F 1,52 = 9.924, p = 0.0027) regardless of mating status. Females had less flies within four body lengths when virgin compared with mated flies (two-way ANOVA—effect of mating status: F 1,52 = 5.269, p = 0.0258). Grey shades: control group housed (GH); coloured shades: treatments; blue shades: single-housed (SH) males; red shades: SH females. (c,d) Number of flies within four body lengths in males (c) and females (d) reared single housed or group housed with 2, 6, 16 or a random number of flies. Single-housed flies had less flies within four body lengths than group-housed flies that were reared with any number of individuals in males ((c) one-way ANOVA with Holm-Sidak post hoc: F4 ,40 = 30.87, p < 0.0001) and females ((d) one-way ANOVA with Holm–Sidak post hoc: F 4,40 = 31.35, p < 0.0001). Grey shades: random number of flies; coloured shades: varying number of flies, from 16 to one (SH: single housed); blue shades: males; red shades: females. On the graph, the results of the post hoc tests are shown only when p < 0.05. The results of all the comparisons, significant or not, can be found in tables 13. n = 9–13 for all treatments. Bars: mean ± s.e.m.

    3.4. Group-housed density does not alter social space

    We next tested what happens when group-housed flies are reared at different densities ranging from two flies up to an uncontrolled or random number of flies. We found that single-housed flies have less flies within four body lengths compared with group-housed flies at all densities in males (figure 3c ; table 3; one-way ANOVA with Holm–Sidak post hoc: F 4,40 = 30.87, p < 0.0001) and females (figure 3d ; table 3; one-way ANOVA with Holm–Sidak post hoc: F 4,40 = 31.35, p < 0.0001).

    3.5. The lack of nlg3 does not prevent from recovering from social isolation

    We have previously shown that nlg3 is important for a typical social space response to isolation [42]. We wondered whether nlg3 is also important for the recovery from social isolation. In males, both CS and nlg3Def1 had less flies within four body lengths in single-housed compared with group housed-flies and nlg3Def1 had less flies within four body lengths than CS (figure 4a ; two-way ANOVA results in table 1; effect of social experience: F 1,32 = 46.48, p < 0.0001; effect of genotype: F 1,32 = 10.10, p = 0.0033). In females, CS had less flies within four body lengths when single housed; however, nlg3Def1 females had no difference in the number of flies within four body lengths compared with group-housed flies (figure 4b ; two-way ANOVA results in table 1; effect of social experience: F 1,32 = 27.97, p > 0.0001; effect of genotype: F 1,32 = 13.71, p = 0.0008; interaction of social experience and genotype: F 1,32 = 5.370, p = 0.0270).

    nlg3Def1 flies have decreased social space after a recovery from isolation.

    Figure 4. nlg3Def1 flies have decreased social space after a recovery from isolation. (a, b) Number of flies within four body lengths in group- and single-housed males (a) and females (b) group housed. (a) Single-housed male flies had decreased flies within four body lengths compared with group-housed flies in both CS and nlg3Def1 (two-way ANOVA—effect of social experience: F 1,32 = 46.48, p < 0.0001), and nlg3Def1 flies had less flies within four body lengths in both group-housed and single-housed (two-way ANOVA—effect of genotype: F 1,32 = 10.10, p = 0.0033). (b) CS females had less flies within four body lengths in single-housed compared with group-housed flies; however, nlg3Def1 flies were not different between group- and single-housed flies (two-way ANOVA—effect of social experience: F 1,32 = 27.97, p < 0.0001; effect of genotype: F 1,32 = 13.71, p = 0.0008; interaction of social experience and genotype: F 1,32 = 5.370, p = 0.0270). Grey shades: control group housed (GH); coloured shades: treatments; blue shades: single housed (SH); males; red shades: SH females. (c,d) Number of flies within four body lengths in group and recovered males (c) and females (d). (c) The number of flies within four body lengths for CS males was not different in group housed versus recovery flies; however, nlg3Def1 flies had an increased number of flies within four body lengths in recovery compared with group-housed flies (two-way ANOVA—effect of social experience: F 1,37 = 17.72, p = 0.0002; effect of genotype: F 1,37 = 34.94, p < 0.0001; interaction of social experience and genotype: F1,37 = 13.62, p = 0.0007). (d) The number of flies within four body lengths for CS females was not different in group-housed versus recovery flies; however, nlg3Def1 females had an increased number of flies within four body lengths in recovery compared with group-housed flies (two-way ANOVA—effect of social experience: F 1,34 = 0.3197, p = 0.5755; effect of genotype: F 1,34 = 6.178, p = 0.0180; interaction of social experience and genotype: F 1,34 = 10.60, p = 0.0026). Grey shades: group housed (GH); coloured shades: recovery from social isolation (REC); blue shades: males; red shades: females. On the graph, the results of the post hoc tests are shown only when p < 0.05. The results of all the comparisons, significant or not, can be found in tables 13. n = 7–12 for all treatments. GH: group housed, SH: single housed, REC: recovery.

    When testing the recovery, CS males did not differ in the number of flies within four body lengths; however, nlg3Def1 males had increased flies within four body lengths in the recovery treatment compared with group-housed nlg3Def1 flies (figure 4c ; two-way ANOVA results in table 1; effect of social experience: F 1,37 = 17.72, p = 0.0002; effect of genotype: F 1,37 = 34.94, p > 0.0001; interaction of social experience and genotype: F 1,37 = 13.62, p = 0.0007). In females, CS flies recovered from isolation did not have a different number of flies within four body lengths compared with flies always group housed; however, nlg3Def1 flies had increased number of flies within four body lengths in the recovery treatment compared with group-housed flies (figure 4d ; two-way ANOVA and Tukey post hoc results in table 1; effect of social experience: F 1,34 = 0.3197, p = 0.5755; effect of genotype: F 1,34 = 6.178, p = 0.0180; interaction of social experience and genotype: F 1,34 = 10.60, p = 0.0026).

    This recovery experiment is repeated in the electronic supplementary material, figure S3 (those data were in fact collected at the same time as the isolation treatments published by Yost et al. [42]). Combined, the data indicate that nlg3 is not required for recovery to occur.

    3.6. Dopamine is required for a response to isolation and decreases after isolation in a sex-specific manner

    DA is important not only for social space in both sexes [48] but also for the response to social isolation in males [43]. We tested to see whether we could recapitulate these results after isolation and determine if DA was important for the recovery from isolation. We drove a RNAi against the gene encoding for TH, the rate-limiting enzyme for DA biosynthesis (TH-RNAi, specifically the UAS-THmiR-G) using a TH-gal4 driver.

    In males, TH>THmiR G, there was no effect of social experience. We did observe a significant reduction in the number of flies within four body lengths with isolation in TH-Gal4/+ and UAS-THmiR-G/+ (figure 5a ; two-way ANOVA results in table 1; effect of isolation: F 1,42 = 19.13, p < 0.0001; genotype and isolation interaction: F 2,35 = 8.635, p = 0.0009). In females, we observed a decrease in the number of flies within four body lengths for all three genotypes after isolation and a decrease in TH>THmiR-G flies compared with their genetic controls (figure 5b ; two-way ANOVA results in table 1; effect of genotype: F 2,37 = 4.009, p = 0.0265; effect of isolation: F 1,37 = 21.91, p < 0.0001).

    Dopamine is important for a response to social isolation and decreases after isolation in males but not females.

    Figure 5. DA is important for a response to social isolation and decreases after isolation in males but not females. Social space measured in terms of number of flies within four body lengths. Bars: mean ± s.e.m. (a,b) Social space in isolated male (a) and female (b) flies. (a) The social space in male TH>THmiR-G was not different in isolated versus group-housed flies; however, TH-GAL4/+ and UAS-THmiR-G/+ males had lower flies within four body lengths when in isolation (two-way ANOVA—effect of isolation: F 1,43 = 19.13, p < 0.0001; interaction between isolation and genotype: F 2,42 = 8.254, p = 0.0009). (b) In females, all three genotypes had a decrease in the number of flies within four body lengths after isolation and TH>THmiR G flies had the lowest number of flies within four body lengths compared with their genetic controls (two-way ANOVA—effect of isolation: F 1,37 = 21.91, p < 0.0001; effect of genotype: F2 ,37 = 4.009, p = 0.0265). (c,d) Social space in recovered male (c) and female (d) flies. (c) Male recovery flies in all genotypes were not different compared with group-housed males. (d) In females, all three genotypes were similar when comparing group-housed versus recovery treatments; however, TH>THmiR-G females had a less flies within four body lengths compared with the controls (two-way ANOVA—effect of genotype: F 2,35 = 5.124, p = 0.0112). (e) DA levels (pg head−1) in males decreased in isolated compared with group-housed flies but do not differ in recovery flies compared with group housed in males or females (two-way ANOVA—effect of social experience: F 3,27 = 3.043, p = 0.0459, but the only significant difference after correcting for multiple comparison through a Dunnett test is indicated on the graph, p = 0.0132). Appropriate age comparisons were made (refer §2 for details). *p < 0.05. ****p < 0.0001. Bars: mean ± s.e.m. On the graph, the results of the post hoc tests are shown only when p < 0.05. The results of all the comparisons, significant or not, can be found in tables 13. n = 5–9 for all treatments in social space. n = 4 for DA quantification. GH: group housed, light grey shade; SH: single housed, dark grey shade; REC: recovery from social isolation, medium grey shade.

    When investigating the recovery, we observed no difference in any genotype between group-housed and recovery treatments in males (figure 4c ; two-way ANOVA results in table 1) or females (no genotype was different comparing group housed to recovery treatments—figure 5d ; two-way ANOVA results in table 1; effect of genotype: F 2,35 = 5.124, p = 0.0112).

    To establish the importance of DA in a response to isolation and recovery, we performed LC/MS on the heads of males and females to quantify changes in DA levels after isolation and recovery. After isolation, males had decreased DA levels, whereas females’ DA levels remained similar to those of group-housed flies. Furthermore, male DA levels returned to a similar level as group housed after a recovery period. Females’ DA levels remained unchanged (figure 5e ; two-way ANOVA results in table 1; effect of social experience: F 3,27 = 3.043, p = 0.0459).

    Taken together, these results indicate that DA is important for social space in response to isolation and recovery in males but not females and that DA levels are influenced by previous social experience in a sex-specific manner.

    3.7. nlg3Def1 flies have less dopamine, but its levels are unchanged after isolation

    To determine whether an interaction was occurring between DA and nlg3, we tested DA levels in group-housed CS and nlg3Def1. In both males and females, DA was decreased in nlg3Def1 compared with CS, and females had lower DA levels compared with males (figure 6a ; two-way ANOVA results in table 1; effect of sex: F 1,7 = 6.107, p = 0.0428; effect of genotype: F 1,7 = 17.21, p = 0.0043). However, when we tested the effect of social experience, no difference in DA levels was observed between nlg3Def1 group housed, isolated and recovery males and females (figure 6b ; two-way ANOVA results in table 1).

    nlg3Def1 flies have less dopamine, and those levels are not altered by social experience.

    Figure 6. nlg3Def1 flies have less DA, and those levels are not altered by social experience. (a) DA levels (pg head−1) in CS and nlg3Def1 males (blue shade) and females (red shade). nlg3Def1 flies had less DA when group-housed and isolated compared with group-housed CS, with females showing a lesser amount overall (two-way ANOVA—effect of genotype: F 1,7 = 17.21, p = 0.0043 and effect of sex: F 1,7 = 6.107, p = 0.0428). (b) DA levels (pg head−1) in male (blue shade) and female (red shade) group-housed and recovery nlg3Def1 flies are unchanged with social experience. Appropriate age comparisons were made (refer §2 for details). Bars: mean ± s.e.m. Grey shades: group housed (GH); coloured shades: social isolation (SH) or recovery from social isolation (REC), as indicated; blue shades: males; red shades: females. On the graph, the results of the post hoc tests are shown only when p < 0.05. The results of all the comparisons, significant or not, can be found in tables 13. n = 3–6 for all treatments. GH: group housed; SH: single housed; REC: recovery from social isolation.

    4. Discussion

    Here, we report the effects of social isolation on social space and sociability. In addition, for the first time to our knowledge, we report the effect of social recovery on the fly social behaviour. Social space was increased, and sociability decreased after isolation but was recovered after 3 days of group housing following isolation. We also show that mating status affected the response to isolation in a sex-specific manner and that as soon as two flies or more are present, the size of the group prior to assaying social space has no effect on social spacing. In addition, we show that an autism candidate gene, nlg3, important for the response to social experience, is not required for recovery from social isolation to occur. We also further confirm the role of DA in social space. However, we show for the first time to our knowledge, the sex-specific role of DA in social spacing: DA is not only required for a response to isolation but also for recovery from isolation in males, but not females. Furthermore, DA levels also respond to the social environment by decreasing in males after isolation and returning to normal levels when recovery occurs, but not in females. Finally, we show that a nlg3 loss of function mutant, nlg3Def1 , inhibits a change in DA in males and females, with no further changes added with social isolation.

    We first looked at how long flies needed to be isolated to effect social space. We saw a sex-specific response to isolation where males only needed 2 days of isolation while females needed closer to 7 days. The females were less effected in the early days of isolation and seemed to be more resilient to the effects of isolation than males (figure 2a,b ). Our results for both males and females isolated for 7 days phenocopy the results of Simon et al. [41]. With regard to recovery, males also responded faster and recovered in 2 days versus 3 days in females (figure 2d,e ). In summary, although males were affected sooner by isolation, they were also able to recover faster than females. The males’ quick response to their social environment may be driven by their need to be quickly adaptable and ready for the possibility of sperm competition, which is affected by perception of a rival [80,81]. Finally, we showed that sociability is reduced in males and females but recovers after group housing (figure 2c,f ), indicating that isolation is affecting multiple social behaviours in the flies. Similarly, for males, increasing isolation length from 1 to 4 days leads to increased effect on aggression [36].

    As we have previously shown that mating status and isolation are important for social space [41,42], we investigated the role of mating status in social space in response to isolation. In both males and females, social space was increased in virgin group-housed flies compared with mated group-housed flies as previously reported ([41]; figure 3a,b ). Both virgin and mated male flies had increased social space after isolation. We then wanted to know whether the number of flies present while group-housed affected social space. We tested single-housed flies and flies reared group housed with 2, 6, 16 and an uncontrolled random number of flies. The isolated flies had increased social space as expected (figure 3c,d ). Group-housed flies had lower social space than isolated flies; however, the social space was similar among all the densities indicating the presence of even one other fly is enough to avoid the negative consequences of isolation on social space. Similarly, the presence of another fly is also sufficient to almost eliminate the effects of isolation on aggression in flies [36].

    Next, we looked at social space in the Drosophila homologue of an autism candidate gene, nlg3, after isolation and recovery. We showed that males CS and nlg3Def1 flies had increased social space after isolation. We have previously reported a diminished response to isolation in nlg3Def1 flies [42]; however, this time, we observed that males responded to isolation similarly to their genetic controls (figure 4a ). However, isolated nlg3Def1 females did not, similar to what we had reported previously (figure 3b ; [42]). When testing the recovery after isolation, male and female CS flies had similar social space to flies always group housed, indicating recovery had occurred (figure 3c,d ). Interestingly, in nlg3Def1 flies, both males and females had decreased social space in the recovery treatment, which appears as an over compensatory response. In an independent repeat of those data, both CS and nlg3Def1 flies recovered similarly from isolation (electronic supplementary material, figure S3). In both cases, the lack of nlg3 did not prevent the recovery from social isolation.

    Using an RNAi against TH, we were able to show that DA is important for a response to isolation in males since the isolated flies did not respond to isolation and even had a slightly closer social space in isolated compared with group-housed flies (figure 5a,c). A similar result was reported by Xie et al. [43], where group-housed flies acted as if isolated and isolated flies acted as if group housed. We do not see as strong of effect as those authors reported, but we also used a pan-TH driver, while Xie et al. [43] focused on a few TH-expressing cells only. Our work and that of Xie et al. [43] demonstrate the importance of DA in males in the response to the social environment. However, we show for the first time to our knowledge, that females do not require DA for a response to isolation and recovery to occur (figure 5b,d ). Another neurotransmitter or neuromodulator could be at play in females, in response to social experience. It has been previously shown that both males and females respond similarly to reduced DA levels, when group housed, with an increase in social space [48]. However, DA might not be the neurotransmitter involved in responding to social experience in females.

    When we isolated our control line, CS, we confirmed that there is a decrease in DA in males but not in females and that the DA levels in males returned to the level of group-housed flies after recovery (figure 5e ). Ganguly-Fitzgerald et al. [31] also saw a decrease in DA in males after isolation. Taken together, these results indicate that DA in males is extremely important for social space in response to isolation, but not in females. Finally, we report that DA levels in nlg3Def1 remain unchanged after isolation, providing further evidence that nlg3 and DA are both contributing to the modulation of social space after isolation (figure 6a,b ). In both CS and nlg3Def1, decreases in DA were observed between 9- and 12-day-old flies regardless of social experience, consistent with age-related decreases in DA reported previously [82]. By contrast, the social space of group-housed CS flies has not been reported to change before they reach 14 days of age [46,47]; and those ages of 12- and 14- days-old might even be quite advanced, for a fly in its natural habitat. Indeed, D. melanogaster typically have a lifespan averaging only 6 days [83], considerably shorter than their lifespan in captivity (up to three months—for review see [84]; which is why we decided to limit our study to periods of isolation of no more than 7 days).

    In the past decade, Drosophila has emerged as a model to study the underlying mechanisms of neuropsychiatric disorders [8594], despite its limitations [84]. In that context, what does recovering from social isolation mean for DA and an autism-related gene? DA in males is required for a response to the environment; however, the flies need nlg3 for DA levels to change in response to the social environment and subsequently modulate behaviour. Furthermore, DA levels decrease in response to social isolation in males, but the NLG3 protein itself is not responding to social experience [42]. We predict that nlg3 and DA are part of a pathway responsible for the modulation of social behaviour after isolation in males. nlg3 would be required downstream of DA, for proper DA signalling in response to the social environment, with some feedback regulation since DA did not change after isolation when nlg3 was absent. Females do require nlg3 for a typical response to isolation [42], but other neurotransmitters or neuromodulators, responsive to social experience in females, may be interacting with nlg3 (figure 7). For example, nlg3 has been shown to affect glutamate receptor GluRIIA recruitment at the larval neuromuscular junction [95]; however, the involvement of glutamate in social spacing seems limited, at least in males, and not reported for females [64]. Similarly, neuromodulators of social space in flies only studied in males, such as acetylcholine [64]), while others have been studied in both males and females such as serotonin [63,65] and GABA [62]. These neurotransmitters should further be investigated for their role in a response to social isolation in females. Also, whether other genes, neurotransmitters and neuropeptides involved in the response to social experience, in both males and females, such as those recently reviewed by Yadav et al. [20], also modulate changes in social space and sociability is still unknown.

    (a) Manipulations performed in this study. (b) Working model: nlg3 and DA are part of a pathway responsible for the modulation of social behaviour after isolation in males.

    Figure 7. (a) Manipulations performed in this study. (b) Working model: nlg3 and DA are part of a pathway responsible for the modulation of social behaviour after isolation in males. nlg3 would be required downstream of DA, for proper DA signalling in response to the social environment, with some feedback regulation. Females do require nlg3 for a typical response to isolation, but other neurotransmitters or neuromodulators, responsive to social experience in females, may be interacting with nlg3. Blue outline: Drosophila brain outline. Created with BioRender.

    A role of Nlgs in DA signalling, and their common influence on behaviour, including social behaviour, has been observed in mice [77,96], the northern swordtail fish [97] and the worm Caenorhabditis elegans [75]. What remains unknown is whether nlg3 is directly involved in the post-synaptic recruitment of one of the DA receptors in flies, as has been suggested, based on work in mice, by Uchigashima et al. [76].

    Beyond nlg3 and DA, other Nlgs, synaptic proteins and neuroreceptors have been shown to influence proper social spacing. As mentioned earlier, nlg2 and nlg4 [68], acetylcholine [64] (all of those only studied in males) and GABA [62] are intriguing candidates. Indeed, nlg4, which is important for social spacing, also modulates GABA neurotransmission, although this modulation was studied in the context of sleep regulation. However, GABA is crucial for normal social spacing, as demonstrated by the consequence of its mislocalization in mitochondria, in CYFIP haploinsufficient flies [62].

    Of note, sexual dimorphism in social spacing in flies has also been reported in mutation of another candidate gene for neuropsychiatric disorders. Indeed, the effect of downregulating the expression of dABCA (ATP-binding cassette protein A) was stronger in males than in females [54]. A link with sexual dimorphism in the underlying neurocircuitry itself has been found in C. elegans, where Neuroligin post-synaptic binding partner, Neurexin, is localized in and affects the function of a sexually dimorphic neuron (which happens to be GABAergic [98]). And the DA-signalling circuitry is known to be sexually dimorphic, at least in its response to stress [99]. Although, unlike in humans, those examples are unrelated to sex chromosomes, they might inform the molecular basis and neural circuitry of sex differences observed in the severity and occurrences of disorders such as autism spectrum disorders [100].

    Future research in the field will need to combine these various data to offer an integrated and more comprehensive understanding of how the players underlying social spacing behaviour work together. For example, there might be a structural association between CYF1P and Nlgs, as both affect the actin cytoskeleton [56,101104].

    In conclusion, for the first time to our knowledge, we have demonstrated that social space and sociability are recoverable after isolation in a sex-specific manner. We have also shown that DA is required for the recovery from social isolation in a sex-specific manner and that nlg3 is required for DA levels to respond to the social environment. If those gene by environment by sex interactions are evolutionarily conserved, it might help us better understand the molecular basis of the social difficulties encountered by humans with neurodevelopmental disorders in response to changing social environments.

    Ethics

    This work did not require ethical approval from a human subject or animal welfare committee.

    Data accessibility

    The datasets analysed for this study can be found in the Dryad Digital Repository [105].

    Supplementary material is also available online [106].

    Declaration of AI use

    We have not used AI-assisted technologies in creating this article.

    Authors’ contributions

    R.T.Y.: conceptualization, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, validation, visualization, writing—original draft, writing—review and editing; A.M.S.: formal analysis, investigation, methodology, writing—original draft; J.M.K.: formal analysis, investigation, writing—original draft; B.W.-R.: investigation, methodology, supervision; R.D.: funding acquisition, resources, supervision, writing—original draft; A.F.S.: conceptualization, formal analysis, funding acquisition, investigation, methodology, project administration, resources, supervision, visualization, writing—original draft, writing—review and editing.

    All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

    Conflict of interest declaration

    The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

    Funding

    This project was funded by an internal and provincial grant to R.T.Y., NSERC Fellowship to R.T.Y. and A.M.S., NSERC Discovery grants 05420-2019 to R.D., and 05054-2022 to A.F.S.

    Acknowledgements

    We would like to thank Dr Mark Bernards, Dr Repon Saha, and Karina Kaberi for their knowledge and expertise obtaining the dopamine level quantification.

    Footnotes

    Electronic supplementary material is available online at https://doi.org/10.6084/m9.figshare.c.7358208.

    Author was undergraduate at the time of data collection

    Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original author and source are credited.

    References

    Comments