Autism Spectrum Disorder — Animal Models & Behavioral Testing

There is no single best model — each captures different aspects of ASD. BTBR is the most widely used for pharmacological screening due to robust social deficits and high repetitive grooming without requiring genetic engineering. Shank3 KO targets a specific high-confidence ASD gene. VPA models environmental risk from prenatal anticonvulsant exposure. MIA with poly(I:C) models the epidemiological link between maternal infection and ASD. Most comprehensive studies use at least two models to validate findings across etiologies.

The three-chamber social approach test is the gold standard. A mouse is placed in a center chamber with access to side chambers containing either a stranger mouse under a wire cup or an empty cup. Sociability is measured as time spent with the social stimulus vs. the empty cup. Social novelty preference is then tested by introducing a second stranger mouse. Additional measures include reciprocal social interaction in a neutral arena, social dominance via the tube test, and ultrasonic vocalization recording during social encounters.

Self-grooming analysis is the primary measure — automated video scoring quantifies bout frequency, duration, and total grooming time. Marble burying counts the number of marbles buried in a set period, reflecting perseverative digging. Other measures include spontaneous alternation in the Y-Maze (reduced alternation indicates cognitive inflexibility), nose-poke pattern analysis in operant chambers, and direct observation of circling, jumping, or other stereotypies in the home cage or open field.

BTBR T+Itpr3tf/J is an inbred mouse strain that naturally exhibits low sociability, high repetitive self-grooming, altered ultrasonic vocalizations, and cognitive inflexibility — all paralleling core ASD features. The strain has agenesis of the corpus callosum and shows aberrant immune activation. Because these phenotypes are constitutive and highly reproducible, BTBR is widely used as a face-valid model for pharmacological screening without requiring genetic manipulation or developmental interventions.

Valproic acid (VPA) is administered as a single injection (500–600 mg/kg, i.p.) to pregnant dams at embryonic day 12.5 in mice. This timing targets the critical window of neural tube closure and early brain development. VPA inhibits histone deacetylases (HDACs), causing widespread epigenetic changes that disrupt neuronal proliferation, migration, and synaptogenesis. Offspring are born normally but develop social deficits, repetitive behaviors, and communication impairments by juvenile age, mimicking features seen in children exposed to VPA in utero.

MIA is modeled by injecting poly(I:C), a synthetic viral mimic, to pregnant dams at E12.5. This activates the maternal immune system via TLR3, triggering a cytokine response dominated by IL-6 and IL-17a. These cytokines cross the placental barrier and disrupt fetal brain development, particularly cortical organization and interneuron migration. The model is based on epidemiological studies showing that maternal viral infection during pregnancy increases ASD risk in offspring. Blocking IL-6 in the dam prevents the behavioral phenotype in offspring.

Yes. Shank3 KO mice are increasingly used for testing compounds that target glutamatergic signaling, mGluR5 modulation, and synaptic plasticity pathways. The excessive grooming phenotype provides a robust, quantifiable endpoint that responds to pharmacological intervention. However, colony management is more demanding than BTBR since Shank3 heterozygous breeding is required and genotyping adds cost and time. Homozygotes show the strongest phenotype but heterozygotes may better model the haploinsufficiency seen in Phelan-McDermid syndrome.

PPI is the normal reduction in startle response when a weak sensory stimulus (prepulse) precedes a strong startling stimulus by 30–500 milliseconds. PPI reflects sensorimotor gating — the brain's ability to filter incoming sensory information. PPI deficits are found in ASD patients and are robust in several ASD mouse models including VPA, MIA, and Shank3 KO. PPI is measured in an acoustic startle chamber and provides an objective, translational biomarker of sensory processing dysfunction.