Neurexin-II deletion results in autistic-like social behaviours

R Cohen1,2, E Hawthornthwaite1, I Edwards2, J Dachtler1. 1University of Leeds, Leeds, UK, 2University College London, London, UK

The neurexin family of genes are presynaptic cell adhesion molecules that form a trans-synaptic link to the postsynaptic neuroligins (Fatemi, 2015). Recent genetic screens have found that deletions within the neurexin-II (NRXN2) gene are associated with autism (Fatemi, 2015; Gauthier et al., 2011). Within this study, we have used a mouse model with a homozygous deletion of Nrxn2 (Missler et al., 2003) to further understand how these deletions result in autistic-like behaviours, particularly in relation to reduced sociability – a core symptom of autism.

Methods

The generation and confirmation of depletion of neurexin-II has been previously described (Missler et al., 2003). A cohort of adult (>p56) neurexin-II knockout (KO; n=18) and wild-type littermates (WT; n=31) on a C57BL/6J background were tested for social investigation and social memory in a three-chambered arena. To confirm a lack of social interest, a novel cohort (WT n=11, KO n=13) was tested for social discrimination of soiled bedding (obtained from a cage of 4 adult male C57BL/6J mice) vs. clean bedding. We also performed the buried food test (as described in (Yang & Crawley, 2009)) on the second cohort. In order to determine which brain areas could be underlying these social deficits, fixed brains of WTs (n=4) and KOs (n=2) were reacted for c-Fos immunoreactivity (Santa Cruz) and visualised by a DAB peroxidase stain 60 minutes following social exposure to two novel conspecifics (10 minutes each). Finally, to investigate whether the social response could be recovered, we delivered oxytocin (1 mg/kg) via IP injection 45 minutes prior to testing for social investigation and memory in the three-chambered arena (KO n=4).

Results

A repeated measures two-way ANOVA found that KO mice failed to show a preference for the cage containing a novel mouse (WT:141±7 vs KO:90±9s, p=0.001) and displayed no ability to discriminate between a previously explored mouse and a second novel conspecific (WT:86±6 vs KO:55±7s, p=0.039). Similarly, KO mice spent significantly less time exploring the cage containing the soiled bedding (WT:143±6 vs KO:119±9s, p=0.036). The absence of social response was unlikely to be due to poor olfaction, as as upon performance of a t-test there was no significant difference in latency to find the buried food (WT:53±11 vs KO:202±81s, p=0.09). Tests for c-Fos immunoreactivity following social exposure to novel conspecifics and statistical testing with two-way ANOVA revealed that the basolateral amygdala had the strongest reduction in number of c-Fos positive cells (WT:69±21 vs KO:50±7), suggesting that reduced activation within the amygdala may be responsible for a lack of sociability in the KOs. Although we hypothesised that KO sociability could be improved using the pro-social drug oxytocin, preference for an empty cage over social contact with a novel mouse was observed (125±17s vs 51±26, respectively). However, there was a minor improvement in the social recognition of a second novel conspecific compared to the first (60±19 vs 34±3s, respectively).

In conclusion, we suggest that Nrxn2 KO mice are a useful tool for exploring the mechanisms underlying altered social behaviours, while further work is required to develop drug treatments to improve sociability.

(1) Fatemi, S (2015). The Molecular Basis of Autism. Springer: Switzerland, Chapter 17.

(2) Gauthier, J. et al., 2011. Truncating mutations in NRXN2 and NRXN1 in autism spectrum disorders and schizophrenia. Human genetics, 130(4), pp.563–573.

(3) Missler, M. et al. 2003. Alpha-neurexins couple Ca2+ channels to synaptic vesicle exocytosis. Nature, 423(6943), pp.939-948.

(4) Yang, M & Crawley, J.N. (2009). Simple behavioural assessment of mouse olfaction. Current Protocols in Neuroscience, Chapter p.Unit–8.24.