Impaired Object Recognition but Normal Social Behavior and Ultrasonic Communication in Cofilin1 Mutant Mice

Abstract

Autism spectrum disorder (ASD), schizophrenia (SCZ) and intellectual disability (ID) show a remarkable overlap in symptoms, including impairments in cognition, social behavior and communication. Human genetic studies revealed an enrichment of mutations in actin-related genes for these disorders, and some of the strongest candidate genes control actin dynamics. These findings led to the hypotheses: (i) that ASD, SCZ and ID share common disease mechanisms; and (ii) that, at least in a subgroup of affected individuals, defects in the actin cytoskeleton cause or contribute to their pathologies. Cofilin1 emerged as a key regulator of actin dynamics and we previously demonstrated its critical role for synaptic plasticity and associative learning. Notably, recent studies revealed an over-activation of cofilin1 in mutant mice displaying ASD- or SCZ-like behavioral phenotypes, suggesting that dysregulated cofilin1-dependent actin dynamics contribute to their behavioral abnormalities, such as deficits in social behavior. These findings let us hypothesize: (i) that, apart from cognitive impairments, cofilin1 mutants display additional behavioral deficits with relevance to ASD or SCZ; and (ii) that our cofilin1 mutants represent a valuable tool to study the underlying disease mechanisms. To test our hypotheses, we compared social behavior and ultrasonic communication of juvenile mutants to control littermates, and we did not obtain evidence for impaired direct reciprocal social interaction, social approach or social memory. Moreover, concomitant emission of ultrasonic vocalizations was not affected and time-locked to social activity, supporting the notion that ultrasonic vocalizations serve a pro-social communicative function as social contact calls maintaining social proximity. Finally, cofilin1 mutants did not display abnormal repetitive behaviors. Instead, they performed weaker in novel object recognition, thereby demonstrating that cofilin1 is relevant not only for associative learning, but also for "non-matching-to-sample" learning. Here we report the absence of an ASD- or a SCZ-like phenotype in cofilin1 mutants, and we conclude that cofilin1 is relevant specifically for non-social cognition.

Keywords: ADF/cofilin; USV; actin depolymerizing factor; cognition; dendritic spine; memory.

Figure 1

Impaired object recognition in cofilin1 mutant mice. (A) Similar to control (CTR), knockout (KO) mice spent equal time exploring both objects during the acquisition phase. Total time exploring the objects was increased in KO mice when compared to CTR mice. (B) Activity of KO mice was increased during the acquisition phase as they more often transited between the compartments of the three-chambered box. (C) While CTR mice showed a preference for the novel object in the recognition trial, KO mice spent equal time exploring the familiar and the novel object. *p < 0.05.

Figure 2

Normal reciprocal social interaction and interaction-induced ultrasonic vocalizations (USVs) in cofilin1 mutants. (A) Time spent in active social behavior was similar in CTR and KO mice in 1 min intervals during the 5 min test period. (B) The reciprocal character of social activity was similar in CTR and KO mice. In the vast majority of cases, mice of both groups replied with a social behavior to a previous social activity. (C) During 5 min of reciprocal social interaction, the number of USV emissions was similar in CTR and KO mice. (D) Graph showing the positive correlation of USV emitted during direct reciprocal social interaction and time spent in active social interaction for CTR mice. KO mice showed a trend for this positive correlation, which did not reach statistical significance. (E) Representative ethograms of CTR and KO pairs during a 5 min test phase (1–6, min right to the dashed line) upon 1 min of habituation (left to the dashed line). Depicted are ultrasonic vocalizations (black bars in row 1), six different social activities (i.e., facial sniffing, anogenital sniffing, following, social grooming, push past, crawling under/over, red bars in rows 2–7), three different non-social activities (i.e., rearing, digging, self-grooming, blue bars in rows 8–10), and other non-social activities (green bars in row 11). Ethograms revealed that the emissions of USV calls correlate with active social behavior both in CTR and KO mice. (F) In CTR and KO mice, the vast majority of USV calls have been emitted during an active social behavior. *p < 0.05.

Figure 3

Social approach and social recognition was normal in cofilin1 mutants. (A) Similar to CTR, KO mice spent much more time exploring the cage with a social stimulus than the empty cage (non-social). Exploration time of the social and the non-social stimuli was not different between CTR and KO mice. (B) CTR and KO mice were both able to discriminate between novel and familiar social stimuli. Both groups spent more time exploring novel than familiar social stimuli. *p < 0.05; **p < 0.001.