A Perturbational Approach to Understand the Neural Basis of Resting-State Functional Connectivity

Spatiotemporal correlations in the resting-state fMRI (rsfMRI) signal are commonly interpreted as an index of reciprocal interareal communication and functional connectivity. However, structurally based models of rsfMRI connectivity have not been comprehensively validated through experimental manipulations in the living brain. Perturbational approaches are critically required for a deeper understanding of brain-wide rsfMRI coupling for two main reasons. First, the correlative nature of rsfMRI makes measures of functional connectivity fickle indicators of direct interareal interaction, as synchronous signals could reflect the contribution of global fluctuations, or could be driven by spatially remote functional interactions with a third region. Second, a deeper understanding of the fundamental elements and dynamic rules underlying the establishment of rsfMRI coupling is required to reliably back-translate clinical alterations in rsfMRI connectivity into testable neurophysiological events and models that can help understand, diagnose or treat brain disorders. By allowing to link specific neural events to brain-wide patterns of brain activity, the combined use of rsfMRI and neuro-perturbational techniques offer the opportunity to bridge this knowledge gap, enabling a disambiguation of the correlative nature of rsfMRI coupling. Using this approach, here we show that chronic silencing of the prefrontal cortex (PFC) via overexpression of a potassium channel results in paradoxical rsfMRI overconnectivity of thalamo-cortical components of the mouse default mode network (DMN). Acute chemogenetics silencing of the PFC produced analogous patterns of rsfMRI oversynchronization, an effect that was functionally relayed to wider cortical areas by polymodal thalamic regions. Importantly, in vivo electrophysiological recordings showed that PFC silencing is associated with reduced gamma power and prominently increased slow oscillatory activity. The observed breakdown between cortical output and rsfMRI coupling challenges prevailing structurally based models of rsfMRI connectivity and suggests that long-range functional coupling is critically orchestrated by locally and remotely generated slow- oscillatory rhythms.