Abstract
INTRODUCTION: Behavioral interventions for autism spectrum disorder show variable outcomes, highlighting the need for complementary therapies. Music-based interventions are promising, yet their molecular mechanisms remain unclear. Saliva-based RNA sequencing (RNA-seq) provides a non-invasive framework to monitor neuroimmune and metabolic dynamics, but its application in autism remains underexplored. METHODS: We explored the buccal transcriptional effects of music exposure in five individuals with autism (8-37 years; 60% female). To overcome saliva-specific limitations, we combined Poly-A selection and Human-Enriched protocols preparation methods to enhance human transcript detection and reproducibility while capturing microbial signals. RESULTS: Individually, each dataset revealed a few differentially expressed genes, but integrated analysis improved biological resolution. Consistently modulated genes included HERC6, TSPAN5, and REM2, involved in neurodevelopmental and immune functions. Enrichment analyses highlighted pathways associated with immune regulation, oxidative phosphorylation, and epithelial differentiation, hallmarks of autism, such as immune dysregulation and mitochondrial dysfunction. Co-expression network analysis identified modules correlated with music exposure. The AKNA module, previously linked to autism, was downregulated and enriched for Ras-related GTPase and immune pathways, suggesting modulation of intracellular signaling and inflammation. Conversely, upregulation of the UBE2D3 module indicated activation of endoplasmic reticulum stress responses, a contributor to autism. Exploratory metagenomics identified 15 microbial species responsive to music exposure, including Acidipropionibacterium acidipropionici and Propionibacterium freudenreichii, producers of propionic acid, a metabolite associated with autism-like behaviors and neuroinflammation. CONCLUSION: Saliva-based RNA-seq can stably capture transcriptomic and microbial responses to behavioral stimuli. Music exposure modulates neuroimmune pathways relevant to autism, supporting the biological plausibility of music therapy and demonstrating saliva-based RNA-seq as a viable, non-invasive tool for monitoring intervention outcomes.