Strategies for Assessing Autistic-Like Behaviors in Mice

Autism spectrum disorder (ASD) is a developmental brain disorder that manifests social communication or interaction impairments and restricted, repetitive patterns of behaviors or interests^1,2. In 2022, approximately 1 in 100 children were diagnosed with ASD globally³. According to the Centers for Disease Control and Prevention (CDC, USA), the prevalence of ASD has increased by 30% since 2008 and is up more than 2-fold since 2000^4,5. Individuals with ASD may also exhibit co-morbidities, such as intellectual disability (ID) (35.2%, IQ ≤ 70), attention-deficit/hyperactivity disorder (ADHD) (50%-70%), and other genetic syndromes^2,4,6.

The use of animal models in ASD research, especially rodents, has provided significant insights into the impact of various environmental factors, including diet, drugs, exercise, and enrichment^7,8,9,10, as well as genetic mutations such as Shank, Fmr1, Mecp2, Pten, and Tsc mutant^11,12,13, on ASD symptoms. Mouse models are commonly used to investigate ASD due to their social nature and shared genetic, biochemical, and electrophysiological features with humans. For instance, by deletion of a specific gene (such as Shank3, Fmr1, Cntnap2, and Pten), aberrant social and repetitive behaviors can be recapitulated, providing strong validity of the study^14,15,16. Here, we provide protocols for studying parallels between animal genetic models and human ASD symptoms¹⁷. We describe the self-grooming and three-chamber social interaction test, which reflect two core symptoms in ASD patients, namely restricted, repetitive patterns of behavior and social interaction (communication) impairments, respectively.

Based on the DSM-V (Diagnostic and Statistical Manual of Mental Disorders of the American Psychiatric Association 5^th Edition) and ICD-11 (International Classification of Diseases 11th Revision), ASD patients engage in restricted, repetitive, and stereotyped behavior patterns, in particular, non-functional body-focused repetitive behaviors (BFRBs), such as rocking, stimming, nail-biting, hair pulling, skin picking, or toe-walking^18,19. In animals, repetitive behavior is manifested by prolonged and repetitive self-grooming. Grooming is one of the most common innate activities among rodents, with approximately 40% of their wake time spent on grooming^20,21. It is instinctive for mice to lick their skin or fur to remove foreign dirt from the body surface, which serves to maintain body cleanliness, prevent injury, remove parasites, and regulate temperature. Grooming is categorized into two types: social grooming (allo-grooming), involving grooming by another mouse, and self-grooming. Self-grooming shows a stereotyped and conserved sequencing pattern consisting of four stages (mostly discrete and non-sequential)^22,23. In stage I (Elliptical stroke), mice initiate grooming by first licking both paws and then grooming around the nose with their paws. In stage II (Unilateral stroke), mice use their paws to wipe their face asymmetrically. In stage III (Bilateral stroke), mice symmetrically wipe their head and ears. In stage IV (Body licking), mice transition to body licking by moving their head backward and may extend grooming to the tail and genitals. When mice are individually placed in a clear cage, self-grooming behavior can be readily recognized and observed. Mice increase self-grooming behavior when faced with stress, pain, or social disruption, rendering the self-grooming test crucial when researching neurological disorders²². Different mouse models of ASD, including those with genetic mutations (such as Fmr1^−/y, Shank3B^−/-, NL1^−/−), pharmacological interventions (such as DO34, PolyI:C), and specific inbred strains (like BTBR and C58/J), have demonstrated excessive repetitive self-grooming behavior^24,25,26,27.

Alterations in social behavior serve as one of the criteria for assessing ASD. According to the DSM-V and the ICD-11, ASD patients display persistent social communication and social interaction impairments^18,19. These may manifest in verbal and nonverbal communication deficits (i.e., abnormal eye contact, gestures, and facial expression), lack of sharing interest and emotions with others, unawareness of social contextual cues, or difficulties developing relationships. In line with the social impairment symptoms, various behavioral tasks have been designed and optimized to assess social interactions in mice, such as the direct social interaction test, the three-chamber social approach and preference for social novelty test, and analysis of ultrasonic vocalizations (USVs)^16,28. The three-chamber social interaction test is an extensively used experiment for evaluating ASD-related behaviors^17,29,30,31. The apparatus comprises three connected chambers; the left and right chambers contain a wire cage that may be either empty or occupied by a mouse, enabling the test mouse to interact freely with both cages. Two measurements help assess different aspects of social behavior in the test mouse during the three-chamber experiment. First, the test mouse is scored for the time spent interacting with the empty cage (novel object) versus a cage that contains a novel mouse. This part of the task provides insight into the mouse's sociability. Next, an unfamiliar mouse is placed into the previously empty wire cage. The time difference in interaction of the test mouse between the unfamiliar and familiar mouse measures the preference for social novelty. In this part of the task, a control mouse prefers to interact with an unfamiliar rather than the previously encountered mouse, which was already present in the sociability part of the test. Deficits in social interaction and decreased motivation of interacting with novel mice are generally found in mouse model of ASD. The three chamber test has proven robust since its invention. It has been used to study social phenotypes in various mouse models of ASD, including Fmr1^−/−, Shank3B^−/-, Cntnap2^−/−, and the BTBR inbred strain^{32,33,34,35,36}.

The two tests utilize naturally occurring, spontaneous behavior of mice as meritorious tools for studying ASD-like behavior. Since they are considered low-stress tests, it is feasible to conduct both tests within the same group of mice to measure ASD-like behavior, with the self-grooming test being performed first and the three-chamber social interaction test on subsequent days. The protocols we provide present an essential tool for the assessment of ASD-like behavior and the development of new therapeutics^29,30,31. Ultimately, they would contribute to improving outcomes for individuals affected by ASD.