Dr Stéphane Baudouin

Social skills are necessary to form and maintain social interactions and to define relationship among individuals. Dysfunction of social skills can lead to severe psychiatric disorders such as autism. The heterogeneity of social skills, social behaviour and social disabilities impedes research on their molecular and cellular underpinnings. To break down this complexity we develop a two-step approach where we first define the molecular and cellular mechanisms underlying specific social skills and second analyse the role of these skills in complex social behaviour. To achieve this goal we use wild-type and genetic mouse models with defects in specific social skills.

With this approach we seek to understand two fundamental aspects of social abilities:

• What is the specificity (i.e. molecular identity, localisation and connectivity) of the neuronal circuits dedicated to particular social skills?
• How the development of social skills influences complex social behaviour and what is the role of social enrichment in sociability?

Modeling social disabilities in mice

Figure1: The three-chamber sociability and social-novelty paradigm test the ability to recognize and discriminate new mice. Mice lacking Neuroligin-3 or-4 show defects in social discrimination the three-chamber test.

Psychiatric disorders like autism and schiozophrenia are in part caused by genetic factors. The genetic of these disorders is highly complex. For example to date about 500 genes have been associated to various forms of autism (for a complete list see the Simons Foundation database: https://gene.sfari.org), each of them accounting for only a small fraction of the cases of the disorder. Among the genes linked with these pathologies, many code for proteins involved in neuronal development and synaptic functions, leading to the hypothesis of de-regulation of pathways specifically associated with neuronal functions. Consequently, investigating psychiatric disorders and by extension social disabilities in model systems such as mice can be achieved by introducing human mutations associated with social dysfunctions. Some of these mouse models recapitulate of symptoms of the disease and in some cases to alterations of social skills. For example, the lack of Neuroligin-3 or 4, two synaptic adhesion molecules associated with autism, show in behavioural tests reduced in interest for social novelty (as assessed by discrimination, Figure 1). The high genetic heterogeneity of these disorders and the low penetrance of each single mutation imply that the study of mutations one-by-one gives a low probability to gain insight on the pathophysiology of the disorders. So conceptually, our approach is divided in two steps. We first identify molecular and neuronal pathophysiology associated with specific social disabilities in these models. Secondly we identify the general mechanisms underlying the common social disabilities by comparative study in different mouse models with similar social disabilities.

Control of social skills

Many studies have shown the negative impact of social isolation on the subsequent development of social behaviour. In addition olfactory sensory deprivation, anosmia induced by nasal injection of zinc-sulfate, strongly influence dominance behaviour and formation of social hierarchy in mice. On the contrary social enrichment is a promising behavioural therapy improving social skills in autistic patients. However it remains unknown how positive social experiences, encounter of one or several unfamiliar mice, can influence complex social behaviour. To address this question we will use a genetic approach to control memory traces generated during specific social experiences. We will subsequently use the control we obtained on the corresponding neuronal populations to analyse the role of prior social experience on complex social interactions. To this purpose we will use synthetic designer receptor exclusively activated by designer drugs (DREADD), the G coupled receptors hM3Dq and hM4Di (Figure 2). The ligands of DREADDs can either excite or inhibit neuronal populations, when binding to the HM3Dq and hM4Di respectively. The knowledge we gain is then transferred in our different animal models, to see if and how these neuronal circuits are associated with pathological conditions.

Figure 2: Example of an experimental design for the study of neuronal networks involved in the control of social skills

An essential stage in the development of new treatments is the characterisation of cellular and molecular mechanisms underlying specific aspect of the symptoms, in particular social behaviour dysfunctions. Our work seeks to help: (i) advancing understanding of the biological basis of social skills and social behaviour, (ii) providing precise biological rational as to neuronal identity and molecular targets in which to screen novel pharmacological interventions.