Persistent myelin abnormalities in a third trimester-equivalent mouse model of fetal alcohol spectrum disorder.

Background: Abnormal diffusion within white matter (WM) tracts has been linked to cognitive impairment in children with fetal alcohol spectrum disorder. Whether changes to myelin organization and structure underlie the observed abnormal diffusion patterns remains unknown. Using a third trimester-equivalent mouse model of alcohol exposure, we previously demonstrated acute loss of oligodendrocyte lineage cells with persistent loss of myelin basic protein and lower fractional anisotropy (FA) in the corpus callosum (CC). Here, we tested whether these WM deficits are accompanied by changes in: (i) axial diffusion (AD) and radial diffusion (RD), (ii) myelin ultrastructure, or (iii) structural components of the node of Ranvier.

Methods: Mouse pups were exposed to alcohol or air vapor for 4 h daily from postnatal day (P)3 to P15 (BEC: 160.4 ± 12.0 mg/dl; range = 128.2 to 185.6 mg/dl). Diffusion tensor imaging (DTI) and histological analyses were performed on brain tissue isolated at P50. Diffusion parameters were measured with Paravision™ 5.1 software (Bruker) following ex vivo scanning in a 7.0 T MRI. Nodes of Ranvier were identified using high-resolution confocal imaging of immunofluorescence for Na 1.6 (nodes) and Caspr (paranodes) and measured using Imaris™ imaging software (Bitplane). Myelin ultrastructure was evaluated by calculating the G-ratio (axonal diameter/myelinated fiber diameter) on images acquired using transmission electron microscopy.

Results: Consistent with our previous study, high resolution DTI at P50 showed lower FA in the CC of alcohol-exposed mice (p = 0.0014). Here, we show that while AD (diffusion parallel to CC axons) was similar between treatment groups (p = 0.30), RD (diffusion perpendicular to CC axons) in alcohol-exposed subjects was significantly higher than in controls (p = 0.0087). In the posterior CC, where we identified the highest degree of abnormal diffusion, node of Ranvier length did not differ between treatment groups (p = 0.41); however, the G-ratio of myelinated axons was significantly higher in alcohol-exposed animals than controls (p = 0.023).

Conclusions: High resolution DTI revealed higher RD at P50 in the CC of alcohol-exposed animals, suggesting less myelination of axons, particularly in the posterior regions. In agreement with these findings, ultrastructural analysis of myelinated axons in the posterior CC showed reduced myelin thickness in alcohol-exposed animals, evidenced by a higher G-ratio.