Impaired vasopressin neuromodulation of the lateral septum leads to social behavior deficits in Shank3B+/- male mice

Bortolozzo-Gleich, Maria Helena; Bouisset, Guillaume; Geng, Lan; Pino, Antonia Ruiz; Nomura, Yuki; Han, Shuting; Li, Yulong; Leroy, Félix

doi:10.1038/s41467-025-61994-6

Download PDF

Article
Open access
Published: 23 July 2025

Impaired vasopressin neuromodulation of the lateral septum leads to social behavior deficits in Shank3B^+/- male mice

Nature Communications volume 16, Article number: 6783 (2025) Cite this article

4230 Accesses
73 Altmetric
Metrics details

Subjects

Abstract

The neuropeptide arginine-vasopressin (AVP) has been repeatedly associated with the autism spectrum disorder (ASD) but the underlying mechanisms remain unclear. As Shank3B^+/- male mice, a model of ASD, exhibit deficits in sociability and social aggression, we focused on the lateral septum (LS), a brain region involved in the regulation of motivated behaviors and observed reduced AVP inputs from the bed nucleus of the stria terminalis (BNST) to LS. Manipulating AVP release from the BNST to LS of wild-type male mice, we found that AVP promotes both sociability and social aggression. Blocking the vasopressin receptor 1a (AVPR1a) in LS impaired sociability, while blocking the receptor 1b (AVPR1b) disrupted social aggression. Consequently, selective activation of AVPR1a or AVPR1b rescued the respective behavioral deficits in Shank3B^+/- male mice. These findings reveal that AVP release in LS modulates two distinct social behaviors via different receptors and highlight a possible strategy to rescue sociability during ASD.

Leveraging a translational research approach to drive diagnostic and treatment advances for autism

Article 01 April 2022

Specification of oxytocinergic and vasopressinergic circuits in the developing mouse brain

Article Open access 14 May 2021

Vasopressin regulates social play behavior in sex-specific ways through glutamate modulation in the lateral septum

Article 20 September 2024

Introduction

Impairment in social interactions is a hallmark of the autism spectrum disorder (ASD). Typical social deficits include initiating interactions, responding to the initiations of others, maintaining eye contact, sharing enjoyment, reading the non-verbal cues of others, and taking another person’s perspective. Social interactions are tightly regulated by several neuropeptides, including arginine-vasopressin (AVP), and studies have linked changes in vasopressinergic neuromodulation to ASD^1,2,3,4. Thus, variants in the adjacent AVP gene regions have been associated with ASD diagnosis and endophenotypes^3,5,6. In addition, magnetic resonant imaging (MRI) revealed changes in structure and functional connectivity in brain regions containing AVP⁷. Moreover, AVP levels in the plasma of children with ASD or their mothers has been associated with autistic symptoms^8,9,10,11,12. AVP can bind to four distinct receptors: AVP receptor 1a (AVPR1A), AVP receptor 1b (AVPR1B), AVP receptor 2 (AVPR2), and the oxytocin receptor (OXTR). Among these, AVPR1A, AVPR1B, and OXTR are primarily located in the central nervous system and show a high degree of polymorphism in patients with autism spectrum disorder (ASD)^{13,14,15,16,17,18,19}. In contrast, AVPR2 is predominantly expressed in the kidney²⁰.

In rodent, subcutaneous administration of AVP in male rats extended the duration of social memory²¹. Further investigations showed that ventricular injections of AVP in male rats immediately after initial encounters with stimulus animals enhanced social memory²², similar to peripheral AVP administration. Stimulation of the supraoptic nucleus and paraventricular nucleus in male rats led to AVP release in the hypothalamus and facilitated social recognition²³, similar to direct infusion of AVP in LS^24,25. Conversely, peripheral AVP antagonist injections impaired social recognition in intact male rats but had no effect on castrated ones²⁶, suggesting that sex hormones may indirectly influence social recognition through modulation of the AVP system. Central administration of anti-AVP serum also resulted in impaired social recognition in male rats²⁷. Knocking out AVPR1a impaired social recognition in male²⁸ but not in female²⁹ mice, although this effect was not seen in all AVPR1a-KO lines³⁰. Knocking out AVPR1b also leads to a deficit in social recognition and social memory (including the Bruce effect) as well as decreased social aggression and social motivation³¹. Finally, intracerebral administration of AVP in male oxytocin receptor knockout mice lowers aggression and fully reverts social and learning defects by acting on AVPR1a receptors³². Despite the extensive evidence that AVP facilitates social interactions, the mechanisms by which defects in vasopressinergic neuromodulation may lead to the social interaction deficits exhibited by ASD patients and mouse models remain elusive.

The lateral septum (LS) is a key region of the limbic system that supports many social and affiliative behaviors, including sociability, social discrimination, and social aggression³³. Autopsy and MRI studies of ASD patients have identified developmental changes in the septum^3,34,35. In rodents, LS is richly innervated by vasopressinergic fibers and expresses AVP receptors³³. Consequently, male Brattleboro rats lacking AVP showed impaired social recognition, which can be restored by micro-dialysis of AVP into LS³⁶. While the source of septal AVP remained unclear for a long time^37,38, recent evidence suggests AVP inputs to LS in mice originate from the bed nucleus of the stria terminals (BNST) and promote social interactions³⁹. In addition, our previous results demonstrate that AVPR1b activation in LS facilitates social aggression in male mice⁴⁰.

Here, we hypothesize that impairment of AVP release from the BNST to LS leads to the social behavior deficits exhibited by Shank3B^+/- male mice, a mouse model of ASD⁴¹. We first show that sociability and intermale social aggression are decreased in Shank3B^+/- male mice. Then, we confirmed that AVP⁺ fibers to LS originate from the BNST before investigating vasopressinergic modulation of LS in Shank3B^+/- mice. We found a decrease in AVP⁺ fiber density in LS and AVP⁺ cell bodies in BNST, as well as a decrease in Avpr1a mRNA expression in LS and BNST of mutant mice. This decrease correlated with fewer excitatory terminals impinging on BNST^AVP neurons. Using calcium recording of BNST^AVP neurons terminals in LS and a biosensor for AVP in male WT mice, we probed the function of septal AVP release and correlated the fibers activation and AVP release with sociability and social aggression. We then used chemogenetic, terminal-specific silencing and gene expression knockdown to demonstrate that release of AVP from BNST to LS promotes sociability and social aggression in male WT mice. Consistent with the lower density of BNST^AVP neurons observed previously, AVP biosensor recordings in male Shank3B^+/- mice indicated a decrease in AVP release during sociability and aggression. We then infused AVP receptor-specific antagonists in LS of WT mice and demonstrated that activation of AVPR1a in LS promotes sociability, while activation of AVPR1b is known to support social aggression⁴⁰. Finally, we leveraged this functional segregation and activated one or the other septal AVP receptor in Shank3B^+/- mice in order to rescue sociability or social aggression.

Results

Sociability and social aggression are impaired in Shank3B ^+/- mice

Prior studies have established that Shank3B^+/- mice exhibit both genetic and behavioral parallels to symptoms observed in autism spectrum disorder (ASD), where sociability/social investigation deficits are predominant^41,42. However, the exploration of other potential social behavior deficits was never particularly investigated, prompting us to investigate whether Shank3B^+/- mice also exhibit changes in antagonistic behaviors, such as intermale social aggression. First, we confirmed the sociability deficits in Shank3B^+/- mice. Male Shank3B^+/- mice and WT littermates were introduced for 5 min in an open arena with two pencil cups in opposite corners. One pencil cup was empty while the other one had a C57Bl6/J mouse of the same age and sex (Fig. 1a). We measured the time the test mouse spent exploring each pencil cup and found that while WT littermates exhibited a preference for the social cup, Shank3B^+/- male mice displayed sociability deficits (Fig. 1b–d) as previously reported⁴².

Then, we tested the mutant mice for social aggression using the resident-intruder test, a behavioral assay designed to assess territorial intermale aggression⁴³. Contrary to the sociability test, where the test mouse exhibits a preference but cannot fight, this test allows for direct interaction between two mice, where a wider range of social behaviors can be displayed (exploration, dominance, and aggression). Male Shank3B^+/- mice and WT littermates were single-housed for a week before a male Balb/cBy mouse of the same age was introduced into their home-cage. We chose to use Balb/cBy mouse intruders for their low aggressivity⁴⁴ in order to minimize the likelihood of attacks being initiated by the intruder (Fig. 1e). Typically, introducing an unfamiliar mouse into the home cage of a singly-housed resident prompted social exploration from the resident mouse, usually followed by social dominance behavior (see “Methods”). In some cases, social dominance eventually escalated to social aggression, defined as one or several biting bouts (Fig. 1f). These male-to-male interactions can be viewed as territorial interactions. Shank3B^+/- mice showed a decrease in social exploration (similar to the result of the sociability test) and a tendency for a decrease in social dominance (Fig. 1I, j). Regarding the aggressive behavior, while all WT littermates escalated from social interaction to aggression within the 10 min duration of the test, only two-third of the Shank3B^+/- mice displayed social aggression (Fig. 1g–j). This difference led to an increase latency to attack and decreased attack duration (Fig. 1k, I). Overall, these results indicate that Shank3B^+/- mice display deficits in intermale social aggression in addition to sociability deficits.

To further analyze the differences in social exploratory behavior between WT littermates and mutant mice, we applied a supervised annotation method on movies acquired during a 2 min period in an open arena containing a male Balb/cBy mouse of the same age (Supplementary Fig. 1a). We measured the time mice spent in nose-to-nose and nose-to-tail interactions, as well as following a same-sex adult conspecific (Supplementary Fig. S1b, c). The results indicated no significant differences in nose-to-nose interactions, which can also be initiated by the stimulus mouse. However, they revealed a significant reduction in exploratory social behavior initiated by the test mouse in the mutant mice group (Supplementary Fig. 1d–f). There were no differences in the distance traveled during the test that could explain these findings (Supplementary Fig. 1g).

Vasopressinergic neuromodulation from the BNST to LS is dysregulated in Shank3B ^+/- mice

We decided to investigate AVP inputs to the septum and injected the BNST of Avp-Cre mice with a Cre-dependent adeno-associated virus (AVV) expressing GCaMP6f (Fig. 2a). Three weeks later, we perfused the mice and observed GCaMP6f⁺ cells at the injection site and GCaMP6f⁺ fibers in LS (Fig. 2b). As the injection site was unilateral, we observed mostly ipsilateral projections. To confirm this BNST-LS AVP projection, we injected a Cre-dependent monosynaptic retrograde herpes simplex virus (HSV) expressing GFP in LS of Avp-Cre mice (Fig. 2c) and observed retrogradely labeled GFP⁺ cell bodies in posterior BNST as well as GFP⁺ axon fibers in LS (Fig. 2d). Since AVP is also expressed in the hypothalamus and central amygdala³⁸, we carefully looked for GFP⁺ cell bodies in these regions but could not find any. Overall, these results indicate that the BNST provides the major source of vasopressin to LS.

**Fig. 2: Vasopressinergic system from BNST (bed nucleus of stria terminalis) to LS (lateral septum) is decreased in *Shank3B*^+/- mice.**

In order to investigate AVP release in LS in Shank3B^+/- mice, and since antibodies against AVP are known to poorly label AVP⁺ cell bodies in the BNST^37,38, we relied on in situ hybridizations to label Avp and Avpr1a mRNAs in the BNST of Shank3B^+/- mice and WT littermates. We found a decrease in the number of Avp⁺ and Avpr1a⁺ cell bodies in the BNST of the mutant mice (Fig. 2e–h). Then, we used anti-AVP antibodies to label fibers and found a reduction in AVP⁺ pixels in anterior but not posterior LS (Fig. 2I, j and Supplementary Fig. 2a, b). Since AVP⁺ fibers in LS originate from the BNST, we can assume that the decrease in AVP⁺ in LS is due to a lower number of AVP⁺ cells in the BNST. Finally, we also observed a decrease in the number of Avpr1a⁺ cell bodies in LS (Fig. 2k, l). All together, these results indicate that the vasopressinergic neuromodulation from BNST to LS is impaired in Shank3B^+/- mice. Importantly, we did not find any difference in AVP⁺ cell number in the paraventricular nucleus of the hypothalamus, the region containing the highest density of AVP⁺ cells³⁸, suggesting that the dysregulation of AVP release is not generalized (Supplementary Fig. 2c, d).

SHANK proteins are multidomain scaffold proteins of the postsynaptic density of glutamatergic synapses and promote normal synaptic functioning⁴⁵. As mutation of SHANK has been shown to lead to a decrease in excitatory synapses in the anterior cingulate cortex⁴⁶ and the hippocampal region CA1⁴⁷, we hypothesized that fewer glutamatergic synapses impinging onto BNST^AVP cells in Shank3B^+/- mice may lead to a decrease in their density. We injected the BNST of Avp-Cre x Shank3B^+/- mice and WT controls with a Cre-dependent virus expressing channelrhodopsin coupled to YFP to label AVP⁺ cells in this region (Fig. 2m). After two weeks of viral expression, animals were perfused, and slices containing the BNST were labeled for the pre-synaptic vesicular glutamatergic transport VGLUT1 to quantify glutamatergic inputs (Fig. 2n).

We observed a strong reduction in the number, density, and size of VGLUT1⁺ clusters apposed to the BNST^AVP cells of Shank3B^+/- mice (Fig. 2o–q), suggesting a reduced excitatory input for these cells. We made sure that although the ratio of VGLUT1⁺ pixels differed (Supplementary Fig. 3a), the mean area of the surface used to count VGLUT1⁺ clusters was unchanged (Supplementary Fig. S3). In addition, VLGUT1 cluster density in the entire BNST was not different between Shank3B^+/- and WT littermates, indicating that the decrease in glutamatergic inputs is not global (Supplementary Fig. 3c, d).

To address whether the BSNT is involved in the social behavioral deficits observed in Shank3B^+/- mice, we looked for differences in the expression of the immediate early gene c-fos following the resident-intruder test. We performed the resident-intruder test in Shank3B^+/- mice and WT littermates. One hour later, we perfused the mice that displayed at least one aggressive episode and labeled the LS and BNST for c-fos (Supplementary Fig. 4a, b). We found a decrease in c-fos⁺ cells in the posterior but not in the anterior BNST nor in anterior or posterior LS (Supplementary Fig. 4c–f). In addition, in situ hybridization showed that the posterior BNST contains a higher density of Avp⁺ cells (22.9 ± 3.4 cells per mm²) compared to the anterior BNST (8.8 ± 1.0 cells per mm², Supplementary Fig. 5a, b).

Silencing BNST^AVP neurons impairs sociability and social aggression

What are the behavioral functions supported by BNST^AVP cells, and could a defect in AVP release from the BNST to LS induce the social behavioral deficits observed in Shank3B^+/- mice? To address these questions, we used chemogenetic to silence BNST^AVP cells. Avp-Cre mice and WT littermates were injected in the BNST with a Cre-dependent AAV expressing an inhibitory DREADD tagged with mCherry. After 2 weeks, mice were injected intra-peritoneally with the DREADD agonist CNO 30 min prior to the sociability test (Fig. 3a–d). Unlike control littermates, Avp-Cre mice expressing iDREADD failed to prefer the social cup (Fig. 3e, f). A week later, we injected the same mice with saline and performed the same test. Here, both groups exhibited a preference for the social cup (Fig. 3g, h). Overall, silencing BNST^AVP cells impaired sociability/social exploration between male mice. Mice were then singled housed for a week and subjected to the resident-intruder test following CNO injection (Fig. 3i). Only a small proportion of Avp-Cre mice demonstrated aggression as opposed to the majority of WT littermates (Fig. 3j). In addition, the time they spent attacking was decreased (Fig. 3k). A week later, the same mice were injected with saline and subjected to the same test. Both groups exhibited social aggression levels comparable to the WT littermates injected with CNO (Fig. 3l–m).

**Fig. 3: Silencing AVP+ neurons from the bed nucleus of the stria terminalis (BNST) impairs sociability and social aggression.**

To better delineate the role of BNST^AVP cells, we also silenced these cells during additional tests on similar cohorts of mice. First, mice explored an open arena for 10 min in order to assess locomotor activity and anxiety (Supplementary Fig. 6a). Avp-Cre mice and WT littermates traveled the same distance and spent a similar amount of time in the center vs. surrounds (Supplementary Fig. 6b–d), suggesting that BNST^AVP cells do not regulate locomotion or anxiety. Then, we tested the mice for social novelty preference. Mice were introduced to an open arena containing 2 wire cups positioned diagonally, each of which housed an unfamiliar, same-age mouse. Animals could freely explore for 5 min and were removed from the open arena and individually placed in a cage for 30 min. We then replaced one of the now familiar animals under the wire cup with an unfamiliar mouse. The test animals were reintroduced in the arena and had the possibility to socially interact freely for 5 min with the familiar or novel stimulus mouse (Supplementary Fig. 6e)⁴⁸. Both groups exhibited social novelty preference, and total interaction times during learning or recall were similar (Supplementary Fig. 6e–i). Overall, BNST^AVP cells promote exploration and aggression of other males but do not support locomotion, anxiety, or social novelty preference.

BNST^AVP cell axonal projections to LS are activated during sociability and social aggression

We then used fiber photometry to image the AVP⁺ terminals in LS. We bilaterally injected the BNST of Avp-Cre mice with a Cre-dependent AAV expressing the calcium sensor GCaMP6f before implanting an optical ferrule above the LS. (Fig. 4a–c). Two weeks after viral injection, animals were subjected to a direct social interaction test, which consisted in 2 min of baseline recording in the home-cage followed by 2 min of presentation of a novel mouse (Fig. 4d). We calculated the frequency and amplitude of the GCaMP6f fluorescent signal during baseline and social presentation. No significant difference was observed in the frequency of the signal, but the amplitude of the signal was higher in the social condition, indicating that BNST^AVP fibers to LS are activated during social encounters (Fig. 4e, f).

**Fig. 4: AVP⁺ axons from the bed nucleus of stria terminalis (BNST) to the lateral septum (LS) are activated during sociability and social aggression.**

Three days after, we performed the sociability test while recording terminal activity (Fig. 4g). We calculated the peri-stimulus time histogram using the start of interaction either with the social or the empty cup to synchronize traces and found that the area under the curve of the calcium peak was larger when interacting with the social cup (Fig. 4h–i). Similarly, the peak amplitude showed a tendency to be larger when approaching the social cup (Fig. 4j). These results confirmed the previous results of the direct social interaction test.

After the sociability test, animals were isolated for a week before recording the calcium activity during the resident-intruder test (Fig. 4k). Here, we calculated the peri-stimulus time histogram using the start of interaction or biting to synchronize traces (Fig. 4l). First, we compared the peak amplitude and area under the curve during biting to 3 s baseline before the biting. Both measures were significantly higher during biting, indicating that the vasopressinergic fibers originating from the BNST in the LS are also active during an aggressive encounter (Fig. 4m, n). We then asked if there was a difference between the magnitude of AVP⁺ fiber responses to social interactions or social aggression. As mice usually explore the intruder during the initial phase of the resident-intruder test, we calculated the peri-stimulus time histogram aligned to the start of social interaction and compared both responses. Peak amplitude of the response was higher during biting than social interaction, but not the area under the curve (Fig. 4o, p). We then employed a linear support vector machine (SVM) decoder to classify social from non-social transients (see “Methods”). Decoder performance was compared to a shuffled baseline where transient labels were randomly permuted. The decoder performed better on actual recordings compared to the shuffled distribution, indicating that social and non-social transients contain distinct activity patterns (Fig. 4q). Overall, BNST^AVP fibers projecting to LS are active during social interactions and aggression, with social aggression recruiting the fibers more efficiently.

Silencing BNST^AVP cell projection to LS impairs sociability and social aggression

To prove that this vasopressinergic projection from BNST to LS is necessary to promote sociability and social aggression, we used an optogenetic silencing approach. We injected the BNST of Avp-Cre mice and WT littermates with a Cre-dependent AAV expressing an enhanced mosquito homolog of the vertebrate encephalopsin (eOPN3) that can suppress synaptic output through a brief illumination of the presynaptic terminals⁴⁹. We then implanted an optical ferrule above LS (Fig. 5a, b). After 2 weeks of viral expression, we performed the sociability test twice. Each test was performed 3 days apart with the light on and off conditions counterbalanced for each mouse (Fig. 5c, d). With the light on, WT littermates displayed a preference for the social cup while Avp-Cre mice did not (Fig. 5e, f). Without light, however, both groups exhibited social preference (Fig. 5g, h). When comparing light conditions for the same mice, Avp-Cre mice displayed reduced interaction with the social cup when the terminals were inhibited (Fig. 5i). Animals were then single-housed for a week and habituated to the patch cord in their home-cage before performing the resident-intruder test with counterbalanced light conditions (Fig. 5c, j). In line with results from the sociability test, turning on the light decreased social exploration for Avp-Cre mice (Fig. 5k, l). We then analyzed the social aggression exhibited by each group. Avp-Cre mice displayed a decrease in the number of attack bout only when with the light on (Fig. 5m, n). Finally, when comparing the same mice across light conditions, Avp-Cre mice but not WT littermates displayed a tendency toward a decrease in total attack time (Fig. 5o). Overall, these results indicate that silencing BNST^AVP pre-synaptic terminals to LS impairs male to male sociability and intermale social aggression, therefore recapitulating the chemogenetic silencing of BNST^AVP cells and demonstrating the importance of this projection in regulating these social behaviors.

**Fig. 5: Silencing AVP⁺ terminals from the bed nucleus of the stria terminalis (BNST) to the lateral septum (LS) impairs sociability and social aggression.**

Genetic silencing of Avp expression in the BNST impairs sociability and social aggression

We have demonstrated that BNST^AVP neurons projecting to LS promote sociability and social aggression, but it is unclear whether this is mediated by AVP release. To address this question, we used a short-hairpin RNA approach to silence Avp mRNA expression. We injected AAVs expressing a shRNA against Avp or a scrambled shRNA as well as GFP in the BNST of C57BL6/J WT mice and observed cell bodies labeled in the BNST and fibers in LS (Fig. 6a–c). Confirming the efficacity of our approach, AVP labeling in LS following anti-Avp shRNA expression in the BNST was decreased compared to control shRNA expression (Supplementary Fig. 7), similar to previous results⁵⁰. After 1 week of viral expression (shRNAs under U6 promoter readily express within 1-2 days), mice performed the sociability test (Fig. 6d). While mice that expressed the scrambled shRNA demonstrated normal social preference, mice expressing the shRNA against Avp failed to prefer the social cup, suggesting that AVP release from BNST^AVP cells is necessary to support sociability (Fig. 6e, f). We subsequently single-housed animals for a week and subjected them to the resident-intruder test (Fig. 6g). In line with the results of the sociability test, mice expressing the shRNA against Avp exhibited a higher latency to interact and less interaction time during the resident-intruder test (Fig. 6h–j). Regarding aggressive behaviors, 5/10 control mice bit the intruder at least once during the test while only 1/10 mice expressing the shRNA against Avp did so (Fig. 6h). The attack duration was also decreased and there was a tendency for the number of tail rattles to decrease as well (Fig. 6k, l). Together, these results indicate that the release of AVP from BNST^AVP cells promotes social aggression in addition to sociability.

**Fig. 6: Knocking-down *Avp* expression in the bed nucleus of the stria terminalis (BNST) impairs sociability and social aggression.**

Septal AVP release is reduced in Shank3B ^+/- mice during social exploration and aggression

Given the behavioral deficits exhibited by Shank3B^+/- mice and reduced density AVP⁺ fibers in their anterior LS, we sought to directly test whether septal AVP release during social interaction and social aggression was decreased in mutants compared to WTs. To achieve this, we leveraged a novel AVP biosensor called AVP0.3. Shank3B^+/- and WT littermate mice were injected with AAV9.hSyn.AVP0.3 into vLS (ventral lateral septum) to turn every neuron near the injection site into an AVP detector. Then, we implanted an optical ferrule above the injection site (Fig. 7a–c). After recovery, mice were subjected to the resident-intruder test while monitoring vasopressinergic events using fiber-photometry (Fig. 7d). Aligning the fluorescent signals from WT mice to the onset of social interactions revealed an increase in both the area under the curve and peak amplitude compared to baseline (Fig. 7f, g). Mutant mice, in contrast, failed to exhibit increased activity during social interactions (Fig. 7h, i). Then, we directly compared the area under the curve or peak amplitude between WT and Shank3B^+/- mice and observed that vasopressinergic events were significantly larger in WT mice (Fig. 7j, k). We then applied the same linear SVM decoder method as described previously to classify between social and non-social transients in WT and mutant mice. In WT, decoder performance on our recordings was significantly higher than the shuffled distribution, indicating distinct activity patterns between social and non-social transients, consistent with the calcium recordings results (Fig. 4q). In contrast, the decoder’s performance was similar between the biosensor recordings from mutant mice and the shuffled distribution. Furthermore, decoder performance was lower in mutants compared to WT (Fig. 7l).

We also aligned the fluorescent signals to the biting onset and found that WT mice exhibited an increase in peak amplitude and a higher area under the curve during biting compared to baseline (Fig. 7m–o). In contrast, no significant changes were observed in Shank3B^+/- mice (Fig. 7p, q). As for social exploration, direct comparisons of the area under the curve or peak amplitude during biting revealed that WT exhibited higher activity compared to mutants (Fig. 7r, s). Overall, these findings indicate that AVP release is reduced in Shank3B^+/- mice during social exploration or aggression of another male mouse.

Activation of AVPR1a and 1b rescues sociability and social aggression respectively in Shank3B ^+/- mice

Since the release of AVP and subsequent activation of AVPR1a in LS are decreased in Shank3B^+/- mice, we sought to compensate for it by infusing receptor-specific agonizts in LS in order to rescue the social behavioral deficits exhibited by the mouse model of ASD. Prior studies have shown that septal AVPR1a and AVPR1b supports different social behaviors⁵¹. Indeed, knocking out AVPR1a impairs social discrimination, which can be rescued by in-trans expression of AVPR1a in LS⁵². On the other hand, AVPR1b is expressed at the presynaptic terminals of CA2 pyramidal neurons, where its activation facilitates the flow of information from CA2 to LS to promote social aggression⁴⁰. In addition, infusion of the AVPR1b antagonist SSR149415 in dorsal LS abolishes social aggression⁴⁰. First, to further characterize the effect of blocking AVPR1a or 1b on sociability, we implanted C57BL6/J WT mice with a cannula guide in LS and infused them with saline, SSR149415 (AVPR1b antagonist) or SR49059 (AVPR1a antagonist, Supplementary Fig. 8a, b). Each mouse was infused with one of the drugs in randomized order and subjected to 3 sessions of the sociability test (Supplementary Fig. 8c). Blocking of AVPR1a abolished sociability while blocking AVPR1b had no effect on it (Supplementary Fig. 8d, e). In addition, drug infusion had no effect on total interaction time with both cups and total distance traveled (Supplementary Fig. 8f, g). Overall, these results suggest a complete dichotomy of action of AVP in LS whereby activation of AVRP1a promotes sociability with no effect on social aggression, while activation of AVPR1b promotes social aggression with no effect on sociability. This is likely due to the organization of LS in parallel pathways, allowing separate regulation of different motivated behaviors⁵³.

Then, we implanted Shank3B^+/- mice with a cannula guide in LS and infused them with saline, the AVPR1b agonist DDAVP, or a combination of AVP and SSR149415 in order to activate AVPR1a while blocking AVPR1b since there is no valid AVPR1a agonist (Fig. 8a, b). First, each mouse was infused with one of the drugs in randomized order and subjected to 3 sessions of the sociability test (Fig. 8c). Activating AVPR1a in Shank3B^+/- mice rescued sociability, unlike activation of AVPR1b (Fig. 8d–f). In addition, drug infusion had no effect on total interaction time with both cups and total distance traveled (Supplementary Fig. 8a–c). Then, we isolated the mice for 1 week, and each mouse was infused with one of the drugs in randomized order and subjected to 3 sessions of the resident-intruder test (Fig. 8c, g). Here, activating AVPR1b in Shank3B^+/- mice increased social aggression unlike activation of AVPR1a as indicated by the number of mice showing attack (Fig. 8h) and the number of attack bouts (Fig. 8i). Activation of AVPR1b also induced a tendency toward an increase in tail rattling duration (Supplementary Fig. 9e), while activation of either receptor induced a tendency toward an increase in dominance behavior (Supplementary Fig. 9f). Overall, these results show that activation of AVRP1a in Shank3B^+/- mice rescues sociability with no effect on social aggression while activation of AVRP1b rescues social aggression with no effect on sociability.

**Fig. 8: Activation of AVPR1a or AVPR1b in LS of *Shank3B*^+/- mice rescues sociability or social aggression respectively.**

Discussion

First, our results demonstrate the importance of AVP release in LS to facilitate social exploration between male mice and therefore promote sociability. The increase in social exploration led to an increase in sociability when mice had the choice to explore an empty cup versus a cup containing another male mouse. Our results are similar to a recent study which utilized optogenetics to show that inhibiting BNST^AVP cells reduces social investigation in male, but not in female mice, whereas stimulating the same cells increased social investigation in both sexes³⁹. Further, they found that optogenetic stimulation of BNST^AVP cell inputs to LS increased social investigation and some anxiety-like behavior in males but not in females³⁹. Contrary to this study, we did not observe a change in anxiety when manipulating the activity of BNST^AVP cells. This difference may stem from experimental reasons since we used the open arena test while they favored the elevated zero maze.

In addition to promoting social exploration, our results also demonstrate a function for AVP in promoting intermale social aggression. Indeed, male to male social exploration usually precedes social aggression, and both social interactions can be viewed as part of a continuum of territorial interaction behaviors facilitated by AVP release from the BSNT. Correlation between septal AVP release and social aggression was previously documented in male rats^54,55,56, but the source of AVP release in LS, as well as a direct causal link, remained missing. Despite a lack of evidence suggesting that vasopressin may originate from other sources that the BNST, we acknowledge that the AVP biosensor approach would respond to the release of AVP from any source, not just the BNST. In addition, we used virgin male mice with limited aggressive experience throughout this study since these two factors are known to increase aggression^57,58. Overall, our findings are consistent with these and other studies involving the BNST in social aggression^59,60. In addition, our previous study highlighted the role of presynaptic AVPR1b at the CA2 to LS synapse to facilitate social aggression without further investigating the source of AVP⁴⁰. Now, we show that AVP is released from the BNST to LS to facilitate social aggression through AVPR1b-mediated potentiation of the CA2 to LS synapse. Interestingly, examination of the intruder-evoked immediate early gene activation in Avpr1b-KO male mice revealed a decrease in EGR-1 expression in the BNST, suggesting a lesser recruitment of the BNST in these mice during aggression⁶¹. This observation remained puzzling since AVPR1b is absent from the BNST⁶². Here, we propose that, because the BNST receives inputs from LS⁶³, hypo-activation of the BNST in Avpr1b-KO male mice likely stems from the decreased transmission of CA2 inputs to LS in the absence of pre-synaptic AVPR1b. A decrease in BNST activation may, in turn, decrease AVP release to LS. Disruption of this AVPR1b-dependent positive feedback loop between LS and the BNST would contribute to Avpr1b-KO male mice exhibiting less aggression⁶⁴. In normal conditions, this loop might be responsible for the facilitation of aggression during repetitive intruder presentation⁴³, similar to the LS-ventral tegmental area loop involving dopamine release in LS^58,65.

In this study, we focused on male mice given the sexual dimorphism of the vasopressinergic projection from the BNST to LS^66,67 and the numerous factors regulating the differentiation of these cells. Thus, testosterone levels around P7 in rats determine the sexual differentiation of AVP projections to LS, and without testosterone, females and castrated males exhibit few BNST^AVP inputs to LS⁶⁶. Consequently, peripheral AVP antagonist injection impairs social recognition in intact male rats but has no effect on castrated ones⁶⁸. The fact that females and castrated males are able to perform social discrimination suggests that other mechanisms can also support this behavior. In addition, AVP⁺ cells in the BNST and medial amygdala also express progesterone receptors, acting to suppress AVP expression in these regions⁶⁹. As a result, progesterone injection in male rats impairs social recognition⁷⁰, which can be rescued by direct AVP infusion into LS²⁵. Finally, ablation of the pineal gland⁷¹ or depletion of norepinephrine from the olfactory bulb⁷² also blocks AVP-mediated social recognition. Overall, these results suggest that AVP expression in the BNST is highly regulated in male rodents, particularly by reproductive hormones.

Female mice exhibit about half AVP⁺ in BNST and less AVP⁺ fibers in LS compared to males⁶⁷. In addition, optogenetic inhibition in female mice did not affect social investigation, and optogenetic stimulation increased female investigation of males but not of females³⁹, suggesting that septal AVP release in female mice does not facilitate social interactions to the same extent, confirming similar results in female rats^26,32. Septal AVP release might even have an opposite role in female rodents since enhanced oxytocin release within the ventral LS, combined with reduced AVP release within the dorsal LS, is required for aggression in virgin female rats⁷³. Similarly, blocking AVPR1a in LS increased social play behavior in males but decreased it in females^74,75, demonstrating opposite regulation of the same behavior in the same study. Further studies on the mechanisms of AVP release in female rodents are warranted to fully understand its function. Finally, AVPR1a antagonist infusion in the amygdala of rats impair impairs maternal memory but not the memory of other adults⁷⁶, which suggests that AVP release in different regions supports distinct types of social memory/preferences. Overall, whether AVP expression in the BNST of female rodents is subject to the same control remains to be explored.

Extending our findings to a pathological context, our results in Shank3B^+/- mice point to an impairment of AVP neuromodulation in LS due to a decrease in Avp⁺ cells in the BNST and Avpr1a⁺ cells in LS. This correlates with a decrease of VGLUT1⁺ excitatory terminals impinging on BNST^AVP neurons. As SHANK3b is a presynaptic protein located on excitatory terminals, we propose that the mutation carried by the mutant mice led to a decrease in excitatory inputs received by BNST^AVP neurons and consequently to their death. Interestingly, the decrease was specific to BNST^AVP neurons and not general to the entire BNST, suggesting that the SHANK3b mutation does not affect every excitatory terminal to the same extent. We also found a decrease in Avpr1a⁺ cells in the BNST of the mutant mice. Given the 50 % overlap between Avp⁺ and Avpr1a⁺ cells in the BNST (Supplementary Fig. 10a, b), we suppose that autocrine activation of these cells provides a negative feedback loop for AVP release. This may explain the negative correlation between AVP release and social aggression observed in the BNST⁵⁶. AVP release from PVN to BNST might also dampen BNST^AVP cell activity⁷⁷. A previous study of Magel2-knockout male mice modeling the Prader-Willi and Schaaf-Yang neurodevelopmental syndromes associated to ASD also showed a decrease in AVP⁺ fibers in LS⁷⁸. Magel2-KO mice exhibit impaired social novelty preference⁷⁹, which can be rescued using AVP infusion in LS or optogenetic stimulation⁷⁸. In our study, however, social novelty preference was not affected by chemogenetic silencing of BNST^AVP cells (Supplementary Fig. S6e–i). Whether dysregulation of AVP inputs to LS is a common feature of ASD models and whether social deficits exhibited by other ASD models can be rescued by manipulating AVP signaling in LS remains to be explored.

Building on these findings, the discovery that AVPR1a promotes sociability while AVPR1b facilitates social aggression demonstrates how a single peptide release can facilitate separate behaviors via specific receptor activation. This segregation of action enabled us to selectivity rescue sociability in our ASD model without affecting social aggression. Previous studies in humans have demonstrated the potential for using AVP agonizts or AVPR1a antagonists to improve social deficits in children⁸⁰ or rescue social communications in adults^81,82,83 respectively, but without mechanism, these studies have generated limited interest. Our results open the way to novel AVP-based therapeutic approaches⁸⁴, aiming at rescuing sociability deficits in ASD patients without facilitating other undesirable behaviors such as social aggression.

Methods

Further information and requests for reagents may be directed to Felix Leroy (felxfel@aol.com).

Ethical approval

All the experimental procedures were in conformity with the directive 2010/63/EU of the European Parliament and of the Council, and the RD 53/2013 Spanish regulation on the protection of animals use for scientific purposes, approved by the government of the Autonomous Community of Valencia, under the supervision of the Consejo Superior de Investigaciones Científicas Committee for Animal use in Laboratory.

Animals

8- to 16-week-old male C57BL6/J (Jackson Laboratories, #000664), Avp-Cre^+/- mice (Jackson Laboratories, #023530) and Shank3B^+/- mice (Jackson Laboratories, #017688) were used as experimental subjects. For experiments with mutant mice, the same age wild-type (WT) littermates were used as the control group. Male Balb/cBy mice (Jackson Laboratories, #001026) were used as intruders during the resident intruder test, and C57BL6/J male mice (Jackson Laboratories, #000664) as stimulus mice during the social interaction tests.

Stereotaxic surgeries

Surgical procedure

Animals were anesthetized with isoflurane and placed in the stereotaxic apparatus. An intramuscular injection of Carprofen was used as an anti-inflammatory drug. Eyes were covered with ophthalmic gel (Viscotears 2 mg/g) to prevent corneal desiccation. The hair of the head was removed, and the tissue was sterilized. Following the opening of the head’s tissue, a small craniotomy was performed above the target region, and a thin glass pipette was placed on the desired depth to deliver the viral content. Injections were performed using the nano-inject II (Drummond Scientific) with a rate of 23 nl every 11 s with a 10 s delay between each. The skin was then closed with surgical glue, and animals rested in a recovery chamber until restoration of normal locomotor activity.

Viral vectors

For anterograde tracing, 100 nL of AAV2/1 syn.FLEX.GCaMP6f.WPRE.SV40 (Addgene, #100833-AAV1) was injected unilaterally in the posterior BNST of Avp-Cre mice (injection coordinates accordingly to the Allen Brain Atlas: AP: +0.02 mm, ML: ±0.75, DV: −4.75 from the cranium). These coordinates were used for all subsequent BNST injections.

For retrograde tracing, 100 nl of the herpes simplex virus hEF1a.LSIL.GFP (Rachael Neve, Massachusetts General Hospital, #RN406) was injected into the LS of Avp-Cre mice (coordinates: AP: +0.24 mm, ML: ±0.50, DV: −3.30 from the cranium). These coordinates were used for all subsequent LS injections.
For AVP⁺ cell labeling in VGLUT1 experiments, AAV2/9EF1a.DIO.hChR2 (E123T/T159C).eYFP8 (Addgene, #35509) was injected in the posterior BNST of Avp-Cre x Shank3B^+/- and WT controls mice.
For chemogenetic experiment, 200 nl of AAV2/8 hSyn.DIO.hM4D(Gi)-mCherry (Addgene, #44362-AAV8) was bilaterally injected in the posterior BNST of Avp-Cre mice and WT littermates.
For fiber-photometry calcium recording, 100 nl of AAV2/1 syn.FLEX.GCaMP6f.WPRE.SV40 (Addgene, #100833-AAV1) was bilaterally injected in the posterior BNST of Avp-Cre mice.
For AVP biosensor recordings, 200 nl of AAV9-hSyn-AVP0.3 was unilaterally injected in LS of C57BL6/J.
For optogenetic silencing experiments, 200 nl of AAV2/1 hSyn1-DIO-eOPN3-mScarlet-WPRE (Addgene, #125713-AAV1) was injected bilaterally in the posterior BNST of Avp-Cre and WT littermates.
For the Avp knock-down experiment, 200 nl of AAV2/9 GFP.U6.anti-Avp-shRNA or AAV2/9 GFP.U6.scramble-shRNA (Vector Biolabs, #253437-shAAV) were injected bilaterally in the BNST of C57BL6/J mice as experimental or control condition respectively. All viruses expressed for a minimum of 2 weeks before the beginning of experiments, except for the shRNA-expressing AAVs, that expressed for 1 week only.

Fiber optic implants

Animals were anesthetized with isoflurane and placed in the stereotaxic apparatus. An intramuscular injection of Carprofen was used as an anti-inflammatory drug. The scalp was sterilized and completely removed. VetbondTM (3 M™ #7000002814) was applied on the peripheries of the cut to hold the tissue. Two screws were placed on the back of the cranium to provide a sturdy foundation for the implant. A craniotomy was performed above the target region, and an optical ferrule (200 µm core, black ceramic ferrule, Neurophotometrics) was lowered until the desired depth. Superglue was applied to hold the lens in position, and dental cement (GC FujiCEM 2) was applied to cover the exposed skull and to maintain the optical ferrule in place. For fiber-photometry calcium and vasopressin recordings, the optical ferrule implant was placed in LS at the following coordinates AP: + 0.24, ML: 0.5, DV: − 2.80 (from the cranium). For optogenetics silencing of BNST terminals, the optical ferrule implant was placed in LS at the following coordinates AP: + 0.24, ML: 0, DV: − 2.80 (from the cranium). Animals remained undisturbed for a week after the procedure.

Guide cannula implants

The mouse scalp was removed and scored before a hole was drilled (AP: + 0.24, ML: ± 0.00). A cannula guide extending 2.4 mm below the pedestal (Plastics One, #C315G 2-G11-SPC) was lowered slowly and kept in place using superglue. The skull was then covered with dental cement (GC FujiCEM 2), and dummy cannulas (Plastics One, #C315DC-SPC) were inserted into the guides. The mice were returned to their home cages and left to recover for at least 3 days. Mice were immobilized by the experimenter, and the dummy cannula was removed. A cannula (Plastics One, #C315I-SPC) projecting 1.7 mm from the tip of the cannula guide was mounted.

Histology and immunohistochemistry

Animals were anesthetized with isoflurane and intracardially perfused with 10 mL of saline. The brains were quickly extracted and incubated in 4% PFA overnight. The brains were washed for 1 h in PBS, and 50 µm slices of the regions of interest were sliced using a Leica VT1000S vibratome (Leica Biosystems).

Immunohistochemistry

The slices were permeabilized for 2 h in PBS with 0.5% Triton-X100 (T9284, Sigma-Aldrich) before being incubated overnight at 4 °C with primary antibodies diluted in PBS with 0.5% Triton-X in PBS. The slices were washed in PBS for 1 h, then incubated for 2 h or overnight at 4 °C with secondary antibodies from Thermo-Fisher Scientific at a concentration of 1:500 diluted in PBS with 0.1% Triton-X. Hoechst counterstain was applied (Hoechst 33342 at 1:1000 for 30 min in PBS at RT) prior to mounting the slice using fluoromount (Sigma-Aldrich).

c-fos labeling

For c-fos labeling, primary incubation was performed overnight at 4 °C with anti-c-fos antibody (1:1000, Abcam, #ab190289). Secondary incubation was performed with an anti-rabbit antibody conjugated to Alexa 488 (#A11039) at a concentration of 1:500 for 2 h at room temperature.

Anti-vasopressin labeling

For vasopressin labeling, primary incubation was performed overnight at 4 °C with anti-vasopressin antibody (1:1000, Merck, #PC234L). Secondary incubation was performed with an anti-rabbit antibody conjugated to Alexa 488 (#A11039) (1:500) overnight.

Anti-VGLUT1 labeling

For anti-VGLUT1 labeling, primary incubation was performed overnight at 4 °C with anti-VGLUT1 antibody (1:2000, Millipore, #ab5905). Secondary incubation was performed with an anti-guinea pig antibody conjugated to Alexa 594 (#A11076) (1:500) for 2 h at room temperature.

RFP labeling

For RFP labeling intensification of the signal, primary incubation was performed overnight at 4 °C with rabbit anti-RFP (1:500, Rockland Antibody, #600-401-379), and secondary incubation was performed with anti-rabbit antibody conjugated to Alexa 568 (#A11011) (1:500) for 2 h at room temperature.

GFP labeling

For intensification of the GFP labeling, primary incubation was performed overnight at 4 °C with and chicken anti-GFP (1:1000, Aves, #GFP-1020) antibodies. Secondary incubation was performed with anti-chicken antibody conjugated to 488 (#A11039) (1:500) overnight.

In situ hybridization

Animals were anesthetized using isoflurane and decapitated. The brains were quickly extracted and immersed in cooled 2-methylbutane for 6 s before being stored at 80 °C. 20 µm slices of the region of interest were prepared using a Leica cryostat (CM3050 S, Leica Biosystems) and mounted on Superfrost Plus microscope slides (12-550-15, FisherBrand). The slices were then processed following the RNAscope® Multiplex Fluorescent Detection Reagents v2 (CN:232110, ACD Bio) with the probes for Avp in C1 (#401391) and Avpr1a in C3 (#418061-C3). Protease IV was applied for 2 min, and TSA Vivid Dyes 520 was used to visualize Avp and 650 for Avpr1a DAPI was applied for 30 s prior to mounting using fluoromount.

Drugs

For hM4D activation, i.p. administration of clozapine-N-oxide dihydrochloride (CNO, Hello Bio, HB6149) dissolved in physiological saline (0.9% NaCl) at a dose of 3 mg kg−1 in a volume of 10 ml kg−1, was used 30 min before the behavioral experiments. I.p. injections of saline (0.9% NaCl) were used 30 min before the behavioral experiments as a control condition. All animals (control and ID) received i.p. CNO injections and saline injections. Saline experiments were performed with 3 days of interval for the sociability test and 1 week of interval for the resident-intruder test to prevent heightened aggression resulting from close repetitive exposure to an intruder in the home-cage.

For the guide cannula experiments on WT mice, C57BL/6 mice were infused in LS with 1 μl of saline, 1 μl of the AVPR1b antagonist SSR149415 at 2 μM, or 1 μl of the AVPR1a antagonist SR49059. For the guide cannula experiments on mutant mice, Shank3B^+/- mice were infused in LS with 1 μl of saline, 1 μl of the AVPR1b agonist DDAVP at 100 μM solution, or 1 μl of AVP at 1 mM combined with together with the antagonist of AVPR1b, SSR149415 at 2 μM to selectively activate AVPR1a only. All drugs came from Tocris (Table 1). Drugs were infused at a rate of 0.2 μl per minute using a programmable syringe pump (Chemyix Inc.) mounted with a 2 µl syringe (Hamilton #88511). After infusion, animals were undisturbed for 5 min before the start of the experiment. Drugs were randomized per animal during the sessions, and each session was performed with 3 days interval for the sociability test, and 1 week of interval for the resident-intruder test. For checking the guide cannula infusion localization in the brain, mini-Ruby (Invitrogen, D1868) was used. A total of 1 μl was infused at a rate of 0.2 μl per minute. 60 min after infusion, animals were perfused.

Table 1 Reagents and resources information

Full size table

Behavioral tests

All animals were housed with littermates (maximum of 5 per cage) at a temperature of 24 °C with 12 h daily illumination and food and water ad libitum. The light cycle was not inverted, and experiments were conducted under dim light conditions.

Open arena test

Animals were placed in a white open arena (OA) (60 cm x 60 cm) and allowed to explore freely the space for 10 min. Automatic tracking software (Any-Maze 7, Stoelting) was used to quantify the time spent in the center and surroundings, as well as the total distance traveled by the animal.

Social interaction test

Animals were insolated in their home-cage, and 2 min calcium recording sessions were conducted in the empty home-cage. Subsequently, a stimulus mouse (Balb/cByJ) was introduced for another 2 min. Animals were allowed to freely explore, and the frequency and amplitude of the calcium signal was quantified for both conditions.

Sociability test

Wire cup cages were placed diagonally in opposite corners of the open arena. One of the cups was empty, and the other had a male, same-aged, unfamiliar mouse under it. Test subjects were allowed to freely explore the arena for 5 min. The same automatic tracking software described above was used to generate regions of interest around the cups and to provide an output of the total time spent interacting with both cups. A discrimination index was used to calculate the interaction time with both cups: (time with social cup – time with empty cup) / total interaction time.

Social novelty preference test

For the learning trial, mice were placed in the open arena containing two wire cups with one male same-age unfamiliar mouse under each cup. The cups were placed diagonally at opposite sides of the arena, and the test mouse could freely explore for 5 min. The test mouse was then single-housed in a chamber for 30 min. For the recall trial, one of the animals under the wire cup was replaced with a novel, unfamiliar mouse. The test mouse was then placed again in the open arena and could freely explore. The automated software for tracking was used for quantifying the interaction time with both animals under the cup. A discrimination index was used to calculate the interaction time with familiar and novel mouse for the recall condition: (time with novel – time with familiar) / total interaction time.

Resident-intruder test

Animals were single-housed in a cage with food and water ad libitum for a week prior to the testing day. During this week, animals were undisturbed, and the bedding of their cage was not changed in order to preserve odors that can favor the development of territoriality. Test mice were virgin and did not receive any sexual or aggressive experience during isolation. An intruder male mouse (Balb/cBy) was placed inside the cage, and the social interactions were recorded for 10 min. The software Any-Maze (Stoelting) was used for manual scoring of the social behaviors displayed. Similar to our previous study⁴⁰, the following social behaviors were quantified: social exploration (facial and ano-genital sniffing initiated by the resident mouse), social dominance (resident mouse rising onto its hindlimbs to scruff the intruder’s head, as well as excessive allo-grooming, chasing or mounting) and attack (biting attack followed by fighting and tail rattling). We show example pictures (Fig. 1f), and example videos can be found in our previous publication⁴⁰. For the c-fos experiments, the same protocol was followed, but following completion of the test, animals were undisturbed for 60 min for the expression of the immediate-early gene and then perfused as previously described.

Development of the GRAB_AVP0.3 sensor

The development, optimization, and characterization strategy of the GRAB_AVP0.3 followed the methodology of our previously developed GRAB sensors⁸⁵. The GRAB_AVP0.3 was developed based on human AVPR2 and optimized by mutagenesis screening of fluorescent protein, linker region, and GPCR. Then GRAB_AVP0.3 was cloned into a pAAV vector and packaged into an AAV virus for in vivo expression.

Fiber-photometry calcium recording data acquisition

Animals were habituated to the optical fiber patch cord for 1 day before the test. The optical fiber patch cord was placed on top of the mouse’s implant, and the sociability test and resident-intruder test were performed as previously described while recording calcium activity from vasopressinergic BNST terminals in LS. Data acquisition was performed using the FP3002 system from Neurophotometrics controlled via Bonsai (Open-Ephys). LEDs delivering two excitation wavelengths (470 nm for detection of GCaMP6f signal and 415 nm for a calcium-independent control) intercalated at 40 Hz throughout recording sessions. Fluorescence emission was focused onto a CMOS sensor for detection, with a region of interest drawn around the end of the plug of the patch cord. A key press trigger was used in Bonsai to save the timestamps of precise social interactions aligned to the calcium signal (social exploration and biting).

Fiber-photometry vasopressin recording data acquisition

AVP biosensor data acquisition was conducted using a DORIC system (Basic FMC). Two LEDs (405 nm and 465 nm) were coupled to a fluorescence mini-cube (FMC) to deliver light into optical ferrules permanently implanted above the dLS. Light was delivered at a final intensity of 12 μW (465 nm) and 249.5 μW (405 nm) at the tip of the patch-cord. Emitted light between 420 and 450 nm (with 405 nm excitation) and 500 and 540 nm (with 465 nm excitation) were collected through the FMC on separate fiber-coupled Newport 2151 photo-receiver modules. The fluorescent signals were collected in AC-low mode and converted to voltage via the formula V = PRG, where V is the collected voltage, P is the optical input power in watts, R is the photodetector responsivity in amps/watts (0.2 – 0.4), and G is the trans-impedance gain of the amplifier. Raw signals and 405 nm excitation (isosbestic signal) were recorded using Doric Neuroscience Studio software.

Optogenetic silencing

Animals were habituated to the optical fiber patch cord for 1 day before the test. The optical fiber patch cord was placed on top of the mouse’s implant, and the sociability test and resident-intruder test were performed as previously described. In the experimental condition of silencing, light stimulation was supplied using the Cobolt-Jive 561 nm, adjusted at 5 mW, and applied during all the testing time. In control conditions, mice were subjected to the test with the patch cord connected but with no light stimulation. The experimental and control conditions were randomized between batches of animals and performed with an interval of 3 days between the sociability test and 1 week between the resident-intruder test, to prevent heightened aggression resulting from close repetitive exposure to an intruder in the home-cage.

Data analysis

Quantification of the density of cell bodies

In situ hybridization was used to label AVP and Avpr1a mRNA in the BNST and LS of Shank3B^+/- and WT littermates, as previously described. Equivalent-sized images of the anterior and posterior LS and BNST were acquired with an inverted confocal microscope (SPEII, Leica) with identical settings. The pictures were loaded in the software Fiji (Schneider et al., 2012) and subjected to manual quantification using the cell counting function.

Quantification of the density of axonal fibers

Immunohistochemistry for labeling vasopressin was performed accordingly to the descriptions above. Equivalent-sized images of the anterior and posterior LS of Shank3b^+/- and WT littermates were acquired with an epifluorescent microscope (Thunder, Leica). The settings for image acquisition was the same for all samples. The pictures were loaded in Fiji software and transformed in to an 8-bit image. A threshold was set to cover all the white pixels of the image, which corresponded to the GFP signal. A selection of the white pixels was then generated, and the area corresponding to the GFP labeling was measured. Data was plotted as the total area of GFP pixels labeled⁸⁶.

Quantification of density of VGLUT1 clusters

Immunohistochemistry for labeling VGLUT1 was performed accordingly to the descriptions above. For both Shank3b^+/- and WT controls, equivalent-sized images of the BNST were acquired (x63 magnification) with identical settings on a confocal microscope (SPEII, Leica). The images were then converted to 8-bit format using ImageJ. A threshold was uniformly applied to all images to enhance the detection and count the different VGLUT1 clusters present on the soma and fibers of AVP cells. Finally, the ROIs used to outline the AVP cells are associated with the number of clusters to generate the VGLUT1 synapse density for each quantified cell.

Quantification of c-fos + cells

The automated program SimpylCellCounter⁸⁷ was used for the quantification of c-fos⁺ cells. Equivalent-sized images from anterior and posterior BNST and LS of Shank3B^+/- and WT littermates were acquired with identical settings in the Thunder microscope (Leica) were loaded in the program. The software utilized binary thresholding and morphological functions from the open-source computer vision library OpenComputerVision (OpenCV) to identify and quantify the cells based on shape and size. The size was defined as 5, and the circularity threshold as 0.8. An output of the images with the quantified cells was generated, and the cells on the total area were plotted.

Fiber photometry and AVP biosensor data analysis

The Guppy software⁸⁸ was used for fiber photometry and AVP biosensor data analysis. The raw data and the behavioral timestamps were loaded, and the first seconds of recording were removed for artifact correction. A subtraction of the background fluorescence was calculated using a time-fitted running average of the 470 nm channel relative to the 415 nm control channel using a least squares polynomial fit of degree 1. The ∆F/F was calculated by subtracting the fitted control channel from the signal channel and dividing by the fitted control channel using the formula: (470–415 nm)/415 nm. A peak enveloping Fourier transform was applied to the ∆F/F signal across the entire trace to identify peaks in activity. The data was presented as the deviation of the ∆F/F from its mean (z-score) using the formula: (∆F/F – mean of ∆F/F/ Standard deviation of ∆F/F). The z-score of each interaction of interest were analyzed during the sessions based on a time window of − 3 to 3 s for the PSTH analysis. For the sociability test, the interaction with the object only (empty cup) was used as a control condition. For the resident-intruder test, the pre-biting or pre-social interaction transients were used as a control condition from − 3 to − 1 seconds. The area under the curve and the peak amplitude were plotted accordingly to these time windows.

Decoder analysis

Fiber photometry recordings of calcium or AVP biosensor signal were used to discriminate between non-social and social transients. In each session, we detected the transients and computed the peak amplitude and the area under the curve of each transient. To discriminate whether the recording came from a non-social or social condition, we trained linear support vector machine (SVM) decoders, using both peak amplitude and transient strength of all detected transients. We trained decoders by randomly selecting 70% data points as a training set and evaluated the decoder performance using the rest 30% as a testing set. To correct for class imbalance, as a standard procedure, we implemented a cost for misclassification that is inversely proportional to the number of samples in each class. This random sampling procedure was repeated 20 times to avoid sampling bias, and the results were averaged across repeats for each session. To generate a baseline distribution, we shuffled the labels (non-social vs social condition) of the data points and trained decoders as above. This shuffling procedure was repeated 1000 times. Decoder performance was evaluated as the area under the curve (AUC) of the receiver operating characteristic (ROC) curve. A statistical test (one-sided Wilcoxon signed-rank test) was performed to compare the AUC between the real data and the top 95% quantile of the shuffled distribution. For the biosensor data, we performed the decoder analysis to discriminate between social and non-social transients in WT and Shank3B^+/- mice, using the same procedure as described above.

Supervised annotation in DeepOF analysis

Videos of WT or Shank3B^+/- mice freely interacting in an open arena with a stimulus mouse were captured using a DMK 27BUR0135 (The IMAGING SOURCE) camera, positioned 130 cm above the arena for a top-down perspective. The recordings were acquired with Bonsai-RX software. Initially, the test mouse was allowed to explore the arena freely for 10 min. Following this, a stimulus mouse (BALB/c, males, 7–11 weeks old) was introduced into the arena, and a 2 min interaction period was recorded and saved for subsequent analysis. The interactions between the test and stimulus mice were analyzed using DeepOF software (version 0.7.2)^89,90. To facilitate this, the body parts of both mice were tracked with DeepLabCut (version 2.2.3)⁹¹ using a custom-trained model. DeepOF analysis was then conducted based on the calculated trajectories for each body part^89,90. Finally, each frame was classified into specific behaviors using DeepOF’s pre-trained network.

Statistics and reproducibility

Statistical analyses and figure plotting were performed using Prism version 9 (GraphPad). No statistical methods were used to predetermine sample sizes, but sample size was selected based on previous experience and on estimation from related studies. Unless specified otherwise, bar graphs represent mean ± s.e.m. Sample sizes and statistical tests are reported in the figure legends. * indicates p < 0.05, ** indicates p < 0.01, *** indicates p < 0.001 and **** indicates p < 0.0001. When multiple observations were done in the same mouse, nested statistical tests were used to take into account the lower degree of freedom. Experiments using viral injections for anterograde and retrograde tracing were independently repeated a minimum of three times, producing similar outcomes. For experiments involving viral expression or optical ferrule implant, only animals exhibiting proper viral expression and accurate optical ferrule localization were included in the analysis.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The data generated in this study are provided in the Source Data file. Request for raw data files should be directed to Felix Leroy (felxfel@aol.com). Source data are provided in this paper.

Code availability

The code used to create the classifiers analysis of the fiber-photometry data is available at https://github.com/hanshuting/AVP_decoder_manuscript. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

References

Harony, H. & Wagner, S. The contribution of oxytocin and vasopressin to mammalian social behavior: Potential role in Autism Spectrum Disorder. Neurosignals 18, 82–97 (2010).
Article CAS PubMed Google Scholar
Lukas, M. & Neumann, I. D. Oxytocin and vasopressin in rodent behaviors related to social dysfunctions in autism spectrum disorders. Behav. Brain Res. 251, 85–94 (2013).
Article CAS PubMed Google Scholar
Zhang, R., Zhang, H. F., Han, J. S. & Han, S. P. Genes related to oxytocin and arginine-vasopressin pathways: Associations with Autism Spectrum Disorders. Neurosci. Bull. 33, 238–246 (2017).
Article CAS PubMed PubMed Central Google Scholar
Cataldo, I., Azhari, A. & Esposito, G. A review of oxytocin and arginine-vasopressin receptors and their modulation of Autism Spectrum Disorder. Front. Mol. Neurosci. 11, https://doi.org/10.3389/fnmol.2018.00027 (2018).
Francis, S. M. et al. Variants in adjacent oxytocin/vasopressin gene region and associations with ASD diagnosis and other Autism autism-related endophenotypes. Front. Neurosci. 10, https://doi.org/10.3389/fnins.2016.00195 (2016).
Dempster, E. L. et al. Further genetic evidence implicates the vasopressin system in childhood-onset mood disorders. Eur. J. Neurosci. 30, 1615–1619 (2009).
Article PubMed Google Scholar
Shou, X. J. et al. A volumetric and functional connectivity MRI study of brain arginine-vasopressin pathways in autistic children. Neurosci. Bull. 33, 130–142 (2017).
Article CAS PubMed PubMed Central Google Scholar
Xu, X. J. et al. Mothers of autistic children: Lower plasma levels of oxytocin and arg-vasopressin and a higher level of testosterone. PLoS ONE 8, https://doi.org/10.1371/journal.pone.0074849 (2013).
Zhang, H. F. et al. Plasma oxytocin and arginine-vasopressin levels in children with Autism Spectrum Disorder in China: associations with symptoms. Neurosci. Bull. 32, 423–432 (2016).
Article CAS PubMed PubMed Central Google Scholar
Oztan, O. et al. Cerebrospinal fluid vasopressin and symptom severity in children with autism. Ann. Neurol. 84, 611–615 (2018).
Article CAS PubMed PubMed Central Google Scholar
Oztan, O., Garner, J. P., Constantino, J. N. & Parker, K. J. Neonatal CSF vasopressin concentration predicts later medical record diagnoses of autism spectrum disorder. Proc. Natl. Acad. Sci. USA 117, 10609–10613 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Carson, D. S. et al. Arginine Vasopressin Is a Blood-Based Biomarker of Social Functioning in Children with Autism. PLoS ONE 10, e0132224 (2015).
Article PubMed PubMed Central Google Scholar
Yirmiya, N. et al. Association between the arginine vasopressin 1a receptor (AVPR1a) gene and autism in a family-based study: mediation by socialization skills. Mol. Psychiatry 11, 488–94 (2006).
Article CAS PubMed Google Scholar
Kim, S. J. et al. Transmission disequilibrium testing of arginine vasopressin receptor 1A (AVPR1A) polymorphisms in autism. Mol. Psychiatry 7, 503–507 (2002).
Article CAS PubMed Google Scholar
Wassink, T. H. et al. Examination of AVPR1a as an autism susceptibility gene. Mol. Psychiatry 9, 968–72 (2004).
Article CAS PubMed Google Scholar
Francis, S. M. et al. ASD and Genetic associations with receptors for oxytocin and vasopressin—AVPR1A, AVPR1B, and OXTR. Front. Neurosci. 10, 516 (2016).
Article PubMed PubMed Central Google Scholar
Dempster, E. L. et al. Evidence of an association between the vasopressin V1b receptor gene (AVPR1B) and childhood-onset mood disorders. Arch. Gen. Psychiatry 64, 1189–1195 (2007).
Article CAS PubMed Google Scholar
Tansey, K. E. et al. Functionality of promoter microsatellites of arginine vasopressin receptor 1A (AVPR1A): implications for autism. Mol. Autism 2, 3 (2011).
Article CAS PubMed PubMed Central Google Scholar
Yang, S. Y. et al. Family-based association study of microsatellites in the 5’ flanking region of AVPR1A with autism spectrum disorder in the Korean population. Psychiatry Res. 178, 199–201 (2010).
Article CAS PubMed Google Scholar
van den-Ouweland, A. M. W. et al. Mutations in the vasopressin type 2 receptor gene (AVPR2) associated with nephrogenic diabetes insipidus. Nat. Genet. 2, 99–102 (1992).
Article Google Scholar
Dantzer, R., Bluthe, R. M., Koob, G. F. & Le Moal, M. Modulation of social memory in male rats by neurohypophyseal peptides. Psychopharmacology 91, 363–368 (1987).
Article CAS PubMed Google Scholar
Le Moal, M., Dantzer, R., Michaud, B. & Koob, G. F. Centrally injected arginine vasopressin (AVP) facilitates social memory in rats. Neurosci. Lett. 77, 353–359 (1987).
Article PubMed Google Scholar
Engelmann, M., Ludwig, M. & Landgraf, R. Simultaneous monitoring of intracerebral release and behavior: endogenous vasopressin improves social recognition. J. Neuroendocrinol. 6, 391–395 (1994).
Article CAS PubMed Google Scholar
Dantzer, R., Koob, G. F., Bluthé, R. M. & Le Moal, M. Septal vasopressin modulates social memory in male rats. Brain Res. 457, 143–147 (1988).
Article CAS PubMed Google Scholar
Bychowski, M. E., Mena, J. D. & Auger, C. J. Vasopressin infusion into the lateral septum of adult male rats rescues progesterone-induced impairment in social recognition. Neuroscience 246, 52–58 (2013).
Article CAS PubMed Google Scholar
Bluthe, R.-M. & Dantzer, R. Social recognition does not involve vasopressinergic neurotransmission in female rats. Brain Res. 535, 301–304 (1990).
Article CAS PubMed Google Scholar
van Wimersma Greidanus, T.jB. & Maigret, C. The role of limbic vasopressin and oxytocin in social recognition. Brain Res. 713, 153–159 (1996).
Article PubMed Google Scholar
Bielsky, I. F., Hu, S. B., Szegda, K. L., Westphal, H. & Young, L. J. Profound impairment in social recognition and reduction in anxiety-like behavior in vasopressin V1a receptor knockout mice. Neuropsychopharmacology 29, 483–493 (2004).
Article CAS PubMed Google Scholar
Bielsky, I. F., Hu, S. B. & Young, L. J. Sexual dimorphism in the vasopressin system: lack of an altered behavioral phenotype in female V1a receptor knockout mice. Behav. Brain Res. 164, 132–136 (2005).
Article CAS PubMed Google Scholar
Wersinger, S. R. et al. Vasopressin 1a receptor knockout mice have a subtle olfactory deficit but normal aggression. Genes Brain Behav. 6, 540–551 (2007).
Article CAS PubMed Google Scholar
Caldwell, H., Wersinger, S. & Young, W. in Advances in Vasopressin and Oxytocin; From Genes to Behaviour to Disease. (Elsevier, 2008).
Sala, M. et al. Pharmacologic rescue of impaired cognitive flexibility, social deficits, increased aggression, and seizure susceptibility in oxytocin receptor null mice: A neurobehavioral model of Autism. Biol. Psychiatry 69, 875–882 (2011).
Article CAS PubMed Google Scholar
Menon, R. et al. Neurobiology of the lateral septum: Regulation of social behavior. Trends Neurosci. 45, 27–40 (2022).
Courchesne, E. Brainstem, cerebellar and limbic neuroanatomical abnormalities in autism. Curr. Opin. Neurobiol. 7, 269–278 (1997).
Article CAS PubMed Google Scholar
Kotagiri, P., Chance, S. A., Szele, F. G. & Esiri, M. M. Subventricular zone cytoarchitecture changes in Autism. Dev. Neurobiol. 74, 25–41 (2014).
Article CAS PubMed Google Scholar
Engelmann, M. & Landgraf, R. Microdialysis administration of vasopressin into the septum improves social recognition in Brattleboro rats. Physiol. Behav. 55, 145–149 (1994).
Article CAS PubMed Google Scholar
De Vries, G. J., Buijs, R. M. & Van Leeuwen, F. W. Sex differences in vasopressin and other neurotransmitter systems in the brain. Prog. Brain Res. 61, 185–203 (1984).
Article PubMed Google Scholar
Rood, B. D. & De Vries, G. J. Vasopressin innervation of the mouse (Mus musculus) brain and spinal cord. J. Comp. Neurol. 519, 2434–2474 (2011).
Article CAS PubMed PubMed Central Google Scholar
Rigney, N. et al. A vasopressin circuit that modulates mouse social investigation and anxiety-like behavior in a sex-specific manner. Proc. Natl. Acad. Sci. USA 121, https://doi.org/10.1073/pnas.2319641121 (2024).
Leroy, F. et al. A circuit from hippocampal CA2 to lateral septum disinhibits social aggression. Nature 564, 213–218 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Peça, J. et al. Shank3 mutant mice display autistic-like behaviours and striatal dysfunction. Nature 472, 437–442 (2011).
Article ADS PubMed PubMed Central Google Scholar
Sgritta, M. et al. Mechanisms underlying microbial-mediated changes in social behavior in mouse models of Autism Spectrum Disorder. Neuron 101, 246–259 (2019).
Article CAS PubMed Google Scholar
Koolhaas, J. M. et al. The Resident-Intruder Paradigm: A standardized test for aggression, violence and social stress. J. Vis. Exp.77, e4367 (2013).
Velez, L., Sokoloff, G., Miczek, K. A., Palmer, A. A. & Dulawa, S. C. Differences in aggressive behavior and DNA copy number variants between BALB/cJ and BALB/cByJ substrains. Behav. Genet. 40, 201 (2010).
Article PubMed Google Scholar
Sheng, M. & Kim, E. The Shank family of scaffold proteins. J. Cell Sci. 113, 1851–1856 (2000).
Article CAS PubMed Google Scholar
Wang, M. et al. Decrease of GSK-3β activity in the anterior cingulate cortex of Shank3b−/− mice contributes to synaptic and social deficiency. Front. Cell Neurosci. 13, https://doi.org/10.3389/fncel.2019.00447 (2019).
Bozdagi, O. et al. Haploinsufficiency of the autism-associated Shank3 gene leads to deficits in synaptic function, social interaction, and social communication. Mol. Autism 1, (2010).
Leroy, F. et al. Enkephalin release from VIP interneurons in the hippocampal CA2/3a region mediates heterosynaptic plasticity and social memory. Mol. Psychiatry 27, 2879 (2022).
Article CAS PubMed Google Scholar
Mahn, M. et al. Efficient optogenetic silencing of neurotransmitter release with a mosquito rhodopsin. Neuron 109, 1621–1635 (2021).
Article CAS PubMed PubMed Central Google Scholar
Rigney, N., Zbib, A., de Vries, G. J. & Petrulis, A. Knockdown of sexually differentiated vasopressin expression in the bed nucleus of the stria terminalis reduces social and sexual behaviour in male, but not female, mice. J. Neuroendocrinol. 34, https://doi.org/10.1111/jne.13083 (2022).
Zai, C. C. et al. Possible genetic association between vasopressin receptor 1B and child aggression. Psychiatry Res. 200, 784–788 (2012).
Article CAS PubMed Google Scholar
Bielsky, I. F., Hu, S.-B., Ren, X., Terwilliger, E. F. & Young, L. J. The V1a vasopressin receptor is necessary and sufficient for normal social recognition: A gene replacement study. Neuron 47, 503–513 (2005).
Article CAS PubMed Google Scholar
Besnard, A. & Leroy, F. Top-down regulation of motivated behaviors via lateral septum sub-circuits. Mol Psychiatry 8, 3119–3128 (2022).
Trainor, B. C., Greiwe, K. M. & Nelson, R. J. Individual differences in estrogen receptor alpha in select brain nuclei are associated with individual differences in aggression. Horm. Behav. 50, 338–345 (2006).
Article CAS PubMed PubMed Central Google Scholar
Beiderbeck, D. I., Neumann, I. D. & Veenema, A. H. Differences in intermale aggression are accompanied by opposite vasopressin release patterns within the septum in rats bred for low and high anxiety. Eur. J. Neurosci. 26, 3597–3605 (2007).
Article PubMed Google Scholar
Veenema, A. H., Beiderbeck, D. I., Lukas, M. & Neumann, I. D. Distinct correlations of vasopressin release within the lateral septum and the bed nucleus of the stria terminalis with the display of intermale aggression. Horm. Behav. 58, 273–81 (2010).
Article CAS PubMed Google Scholar
Oliveira, V. E. de M. in Handbook of Anger, Aggression, and Violence. (2023).
Dai, B. et al. Experience-dependent dopamine modulation of male aggression. Nature 639, 430–437 (2025).
Takahashi, A. & Miczek, K. A. Neurogenetics of aggressive behavior: Studies in rodents. Curr. Top. Behav. Neurosci. 17, 3–44 (2013).
Article Google Scholar
Trainor, B. C. & Nelson, R. J. in Handbook of Neuroendocrinology. (2012).
Witchey, S. K., Stevenson, E. L. & Caldwell, H. K. Genotypic differences in intruder-evoked immediate early gene activation in male, but not female, vasopressin 1b receptor knockout mice. BMC Neurosci. 17, https://doi.org/10.1186/s12868-016-0310-7 (2016).
Young, W. S., Li, J., Wersinger, S. R. & Palkovits, M. The vasopressin 1b receptor is prominent in the hippocampal area CA2 where it is unaffected by restraint stress or adrenalectomy. Neuroscience 143, 1031–1039 (2006).
Article CAS PubMed Google Scholar
Sun, Y., Zweifel, L. S., Holmes, T. C. & Xu, X. Whole-brain input mapping of the lateral versus medial anterodorsal bed nucleus of the stria terminalis in the mouse. Neurobiol. Stress 23, https://doi.org/10.1016/j.ynstr.2023.100527 (2023).
Wersinger, S. R., Ginns, E. I., O’Carroll, A.-M., Lolait, S. J. & Young, W. S. III Vasopressin V1b receptor knockout reduces aggressive behavior in male mice. Mol. Psychiatry 7, 975–984 (2002).
Article CAS PubMed Google Scholar
Luo, A. H., Tahsili-Fahadan, P., Wise, R. A., Lupica, C. R. & Aston-Jones, G. Linking Context with reward: A functional circuit from hippocampal CA3 to ventral tegmental area. Science 333, 353–357 (2011).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, Z., Bullock, N. A. & De Vries, G. J. Sexual differentiation of vasopressin projections of the bed nucleus of the stria terminals and medial amygdaloid nucleus in rats. Endocrinology 132, 2299–2306 (1993).
Article CAS PubMed Google Scholar
Rigney, N., Whylings, J., Mieda, M., De Vries, G. J. & Petrulis, A. Sexually dimorphic vasopressin cells modulate social investigation and communication in sex-specific ways. eNeuro 6, https://doi.org/10.1523/eneuro.0415-18.2019 (2019).
Bluthe, R. M., Schoenen, J. & Dantzer, R. Androgen-dependent vasopressinergic neurons are involved in social recognition in rats. Brain Res. 519, 150–157 (1990).
Article CAS PubMed Google Scholar
Auger, C. J. & Vanzo, R. J. Progesterone treatment of adult male rats suppresses arginine vasopressin expression in the bed nucleus of the stria terminalis and the centromedial amygdala. J. Neuroendocrinol. 18, 187–194 (2006).
Article CAS PubMed Google Scholar
Bychowski, M. E. & Auger, C. J. Progesterone impairs social recognition in male rats. Horm. Behav. 61, 598–604 (2012).
Article CAS PubMed PubMed Central Google Scholar
Appenrodt, E., Juszczak, M. & Schwarzberg, H. Septal vasopressin induced preservation of social recognition in rats was abolished by pinealectomy. Behav. Brain Res. 134, 67–73 (2002).
Article CAS PubMed Google Scholar
Dluzen, D. E., Muraoka, S. & Landgraf, R. Olfactory bulb norepinephrine depletion abolishes vasopressin and oxytocin preservation of social recognition responses in rats. Neurosci. Lett. 254, 161–164 (1998).
Article CAS PubMed Google Scholar
Oliveira, V. E. de M. et al. Oxytocin and vasopressin within the ventral and dorsal lateral septum modulate aggression in female rats. Nat. Commun. 12, 2900 (2021).
Veenema, A. H., Bredewold, R. & De Vries, G. J. Sex-specific modulation of juvenile social play by vasopressin. Psychoneuroendocrinology 38, 2554–2561 (2013).
Article CAS PubMed Google Scholar
Bredewold, R., Smith, C. J. W., Dumais, K. M. & Veenema, A. H. Sex-specific modulation of juvenile social play behavior by vasopressin and oxytocin depends on social context. Front. Behav. Neurosci. 8, https://doi.org/10.3389/fnbeh.2014.00216 (2014).
Nephew, B. C. & Bridges, R. S. Arginine vasopressin V1a receptor antagonist impairs maternal memory in rats. Physiol. Behav. 95, 182–186 (2008).
Article CAS PubMed PubMed Central Google Scholar
Rigney, N., de Vries, G. J. & Petrulis, A. Modulation of social behavior by distinct vasopressin sources. Front. Endocrinol. 14, https://doi.org/10.3389/fendo.2023.1127792 (2023).
Borie, A. M. et al. Correction of vasopressin deficit in the lateral septum ameliorates social deficits of the mouse autism model. J. Clin. Invest. 131, https://doi.org/10.1172/jci144450 (2021).
Fountain, M. D., Tao, H., Chen, C. A., Yin, J. & Schaaf, C. P. Magel2 knockout mice manifest altered social phenotypes and a deficit in preference for social novelty. Genes Brain Behav. 16, 592–600 (2017).
Article CAS PubMed Google Scholar
Parker, K. J. et al. A randomized placebo-controlled pilot trial shows that intranasal vasopressin improves social deficits in children with autism. Sci. Transl. Med 11, https://doi.org/10.1126/scitranslmed.aau7356 (2019).
Hollander, E. et al. Balovaptan vs placebo for social communication in childhood Autism spectrum disorder: A randomized clinical trial. JAMA Psychiatry 79, 760–769 (2022).
Article PubMed PubMed Central Google Scholar
Schnider, P. et al. Discovery of balovaptan, a vasopressin 1a receptor antagonist for the treatment of Autism Spectrum Disorder. J. Med. Chem. 63, 1511–1525 (2020).
Article CAS PubMed Google Scholar
Bolognani, F. et al. A phase 2 clinical trial of a vasopressin V1a receptor antagonist shows improved adaptive behaviors in men with autism spectrum disorder. Sci. Transl. Med. 11, https://doi.org/10.1126/scitranslmed.aat7838 (2019).
László, K. et al. Vasopressin as possible treatment option in Autism Spectrum Disorder. Biomedicines 11, https://doi.org/10.3390/biomedicines11102603 (2023).
Wang, H. et al. A tool kit of highly selective and sensitive genetically encoded neuropeptide sensors. Science 382, https://doi.org/10.1126/science.abq8173 (2023).
Montanari, R. et al. Callosal inputs generate side-invariant receptive fields in the barrel cortex. Sci. Adv. 9, https://doi.org/10.1126/sciadv.adi3728 (2023).
Bal, A., Maureira, F. & Arguello, A. A. SimpylCellCounter: an automated solution for quantifying cells in brain tissue. Sci. Rep. 10, 12570 (2020).
Sherathiya, V. N., Schaid, M. D., Seiler, J. L., Lopez, G. C. & Lerner, T. N. GuPPy, a Python toolbox for the analysis of fiber photometry data. Sci. Rep. 11, 1–9 (2021).
Article ADS Google Scholar
Miranda, L., Bordes, J., Pütz, B., Schmidt, M. V. & Müller-Myhsok, B. DeepOF: a Python package for supervised and unsupervised pattern recognition in mice motion tracking data. J. Open Source Softw. 8, 5394 (2023).
Article ADS Google Scholar
Bordes, J. et al. Automatically annotated motion tracking identifies a distinct social behavioral profile following chronic social defeat stress. Nat. Commun. 14, 4319 (2023).
Mathis, A. et al. DeepLabCut: markerless pose estimation of user-defined body parts with deep learning. Nat. Neurosci. 21, 1281–1289 (2018).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We thank Antoine Besnard for comments on the manuscript. This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research, innovation program (grant agreement No 949652 to F.L.), NIH grant (U01NS120824 to Y.L.), National Natural Science Foundation of China (323B2033 to L.G.), and an Ambizione Grant from the Swiss National Science Foundation (PZ00P3_216312 to S.H.). S.H. also received funding from the University Research Priority Program (URPP) “Adaptive Brain Circuits in Development and Learning” (AdaBD). F.L. also acknowledges support from the CIDEGENT grant from the Valencian Community, the Proyecto de Generación de Conocimiento (PID2022-141262NA-I00) from the Agencia Estatal de Investigacion and the Severo Ochoa Foundation. The project that gave rise to these results received the support of a fellowship from “la Caixa” Foundation (ID 100010434) to M.H.B-G. The fellowship code is LCF/BQ/DI20/11780018.

Author information

Authors and Affiliations

Instituto de Neurociencias CSIC-UMH, Avenida Santiago Ramon y Cajal, San Juan, de Alicante, Spain
Maria Helena Bortolozzo-Gleich, Guillaume Bouisset, Antonia Ruiz Pino, Yuki Nomura & Félix Leroy
State Key Laboratory of Membrane Biology, Peking University School of Life Sciences, Beijing, China
Lan Geng & Yulong Li
PKU-IDG/McGovern Institute for Brain Research, Beijing, China
Lan Geng & Yulong Li
Brain Research Institute, University of Zurich, Zurich, Switzerland
Shuting Han
Neuroscience Center Zurich (ZNZ), University of Zurich, Zurich, Switzerland
Shuting Han
University Research Priority Program (URPP), Adaptive Brain Circuits in Development and Learning, University of Zurich, Zurich, Switzerland
Shuting Han

Authors

Maria Helena Bortolozzo-Gleich
View author publications
Search author on:PubMed Google Scholar
Guillaume Bouisset
View author publications
Search author on:PubMed Google Scholar
Lan Geng
View author publications
Search author on:PubMed Google Scholar
Antonia Ruiz Pino
View author publications
Search author on:PubMed Google Scholar
Yuki Nomura
View author publications
Search author on:PubMed Google Scholar
Shuting Han
View author publications
Search author on:PubMed Google Scholar
Yulong Li
View author publications
Search author on:PubMed Google Scholar
Félix Leroy
View author publications
Search author on:PubMed Google Scholar

Contributions

Conceptualization: M.H.B-G. and F.L. Investigation: M.H.B-G., G.B., L.G., A.R.P. Y.N., and S.H. Writing – original draft: M.H.B-G. Writing – review & editing: M.H.B-G. and F.L. Visualization: M.H.B-G. Supervision: F.L. Funding acquisition: M.H.B-G., Y.L., F.L.

Corresponding author

Correspondence to Félix Leroy.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Nicole Rigney, Toru Takumi, and the other anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Reporting Summary

Transparent Peer Review file

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Bortolozzo-Gleich, M.H., Bouisset, G., Geng, L. et al. Impaired vasopressin neuromodulation of the lateral septum leads to social behavior deficits in Shank3B^+/- male mice. Nat Commun 16, 6783 (2025). https://doi.org/10.1038/s41467-025-61994-6

Download citation

Received: 29 August 2024
Accepted: 04 July 2025
Published: 23 July 2025
DOI: https://doi.org/10.1038/s41467-025-61994-6