INTRODUCTION: Autism spectrum disorder (ASD) is a highly heritable neurodevelopmental disorder characterized by social impairments and inflexible behaviors (1), of which treatments are an unmet medical need. Transgenic animal models with homologous etiology provide a promising way to pursue the neurobiological substrates of the behavioral deficits in autism spectrum disorder (ASD). Gain-of-function mutations of MECP2 cause MECP2 duplication syndrome, a severe neurological disorder with core symptoms of ASD (2, 3). However, the developmental implications of MECP2 duplication on brain function and structure, and how the altered brain developments lead to behavioral deficits, are still underexplored.

METHODS: A human MECP2 duplication (MECP2-DP) rat model manifesting locomotor deficits and reduced social novelty preference was created by the bacterial artificial chromosome transgenic method. Both functional MRI and structural MRI were acquired from 16 male MECP2-DP rats and 15 male wildtype (WT) rats (from six litters) at postnatal 28 days, 42 days, and 56 days, respectively. Then, we performed three-way (fALFF: fractional Amplitude of Low-Frequency Fluctuations; ReHo: Regional Homogeneity; GMV: Grey Matter Volume;) MRI fusion (4-6) supervised by social novelty time and locomotor-relevant metrics to detect abnormal brain networks underlying various behavioral deficits of MECP2-DP rats (Fig.1A). A linear mixed-effect model was established to analyze the developmental implications of atypical MECP2 levels on brain function and structure spanning the three developmental milestones (Fig.1B). Correlation analysis was performed to delineate the relationship between aberrant brain development and behavioral metrics (Fig.1C).

RESULTS: Under the guidance of social novelty time and motor ability metric, brain networks underlying both social deficits (Fig.1D-F) and motor deficits (Fig.2A-C) induced by MECP2 duplication were identified. Specifically, we found that fALFF and ReHo of a network mainly composed of the dorsal medial prefrontal cortex (dmPFC) and retrosplenial cortex (RSP) was significantly decreased in MECP2-DP rats and showed negative correlations with both social novelty time and motor ability metric. Meanwhile, we found GMV of the hippocampus was significantly increased (Fig.1D-E), and GMV of the thalamus was significantly reduced (Fig.2A-B) in MECP2-DP rats. The volume of thalamus showed negative associations with motor ability metric (Fig.2C), whereas the associations between hippocampal volume and social deficit did not reach significance (Fig.1F). All identified functional-structural network did not exhibit significant changes along the three developmental milestones (Fig.1E and Fig.2B). Notably, correlations analysis (Fig.2E-F) showed that the fALFF and ReHo networks underlying social deficits negatively correlated with motor ability score, and the fALFF network underlying motor deficits negatively correlated with social novelty time. Considering the large overlap on dmPFC and RSP between the multimodal networks underlying social deficits and motor deficits (Fig.2A), these results suggest that the neurobiological bases of abnormal behavioral phenotypes induced by MECP2 duplication converge onto the fALFF decrease of a core network primarily composed of dmPFC and RSP.

CONCLUSIONS: This is the first attempt to study the relationship between brain development and behavioral deficits induced by MECP2 duplication under a multimodal fusion and longitudinal analysis framework. We suggest that gain-of-function mutations of MECP2 induce aberrant functional activities in the default-mode-like network and aberrant volumetric changes of hippocampus and thalamus, resulting in autistic-like behavioral deficits. Overall, our results provide critical insights into the biomarker of MECP2 duplication syndrome and the neurobiological underpinnings of the behavioral deficits in ASD, highlighting the power of multimodal fusion analysis.