Autism spectrum disorder (ASD) is a neurodevelopmental disorder, characterized by impaired social communication, abnormal repetitive behaviors and restricted interests and/or sensory behaviors. It has been widely accepted that ASD involves a complex interplay of both genetic and environmental risk factors. Existing medications are only symptomatic treatments, there are no effective treatments that can improve these core social behavior deficits. Recent studies indicated that synaptic development and abnormal myelination are linked to the pathogenesis of ASD. The stable tubule only polypeptide (STOP) protein, also known as microtubule-associated protein 6, plays an important role in neuronal development and synaptic plasticity. Our previous studies showed that STOP protein was significantly reduced in the plasma of autistic subjects and in the cortex of BTBR T+ Itpr3tf (BTBR) mouse model of ASD. Furthermore, studies have shown that Epothilone D, a taxol-like microtubule-stabilizing agent, could alleviate behavioral and synaptic deficits in STOP-null mice. Here, we further evaluate whether Epothilone D treatment is sufficient to modulate the autism-like behaviors in the BTBR mice, and explore the underlying mechanism. BTBR mice were treated either with Epothilone D dissolved in 99% dimethyl sulfoxide (DMSO) or with 99% DMSO vehicle. Our studies demonstrated that the restricted and repetitive behaviors of BTBR mice were improved after Epothilone D treatment, which could be achieved by improving microtubule stability and further regulating the expression of excitatory synapse-related and myelin-related proteins. These results indicate that microtubule stability may be a new and promising therapeutic target for treating patients with ASD.