Autism spectrum disorder (ASD) is associated with atypical brain network organization. Functional connectivity has been extensively studied using fMRI, which often reports reduced long-range connectivity alongside local hyperconnectivity. However, BOLD-based connectivity reflects neurovascular coupling and therefore provides an indirect estimate of neuronal activity. Complementary information may be obtained from 18FFDG microPET, which quantifies regional glucose uptake—an index of synaptic energy demand integrated over minutes—although FDG-derived network organization has been scarcely examined in ASD. Here, “metabolic connectivity” refers to between-subject covariance in regional 18FFDG uptake (group-level metabolic covariance networks), not within-subject temporal coupling as in BOLD-fMRI functional connectivity. We tested whether the C58/J mouse strain—an ASD-relevant model with social deficits and repetitive behaviors—recapitulates ASD-relevant metabolic network-level alterations. Using 18FFDG microPET, we constructed ROI-wise metabolic covariance networks by correlating uptake values across animals and compared C58/J mice with C57BL/6 controls. The C58/J network exhibited higher clustering and a more locally cohesive organization than the C57BL/6 network. Nodal degree/edge density was reduced in the olfactory bulb, hippocampus, and hypothalamus. In contrast, the motor-related regions—including the striatum, brainstem, and superior and inferior colliculi—showed a higher degree/denser covariance. These findings suggest that C58/J mice display FDG-derived metabolic covariance network features qualitatively consistent with those reported in ASD FDG-PET studies, supporting this strain as a tool to investigate ASD-relevant pathophysiological mechanisms and to evaluate candidate interventions.
Verdugo-Díaz et al. (Wed,) studied this question.