BASAL FOREBRAIN CONTROL OF LARGE-SCALE FMRI NETWORKS

Resting-state fMRI (rsfMRI) is widely used to map brain regions exhibiting synchronous spontaneous BOLD signal fluctuations. These patterns show highly stereotyped spatial organization, delineating so-called resting-state networks (RSNs). Accumulating evidence suggests that RSN organization is not static, but highly flexible, and shaped by both external input and internal brain state. However, the neural mechanisms orchestrating the dynamic organization of RSNs remain unclear.
Neuromodulatory nuclei like the basal forebrain (BF), a major source of cortical acetylcholine and key regulator of arousal, are strong candidates for causally controlling RSNs dynamics. Yet, whether and how BF output causally organizes large-scale RSNs is unknown. Here, we addressed this question using targeted manipulations of the BF in mice. Specifically, we reduced global BF output via pan-neuronal genetic lesioning. We then used chemogenetics to dissect the relative contribution of parvalbumin-positive (Pv+) and cholinergic (Chat+) BF neurons in controlling RSNs activity.
Chronic BF ablation reconfigured fMRI connectivity extensively, strengthening default mode network (DMN) prefrontal connectivity, while broadly reducing cortico-cortical coupling. Conversely, activation of Pv+ and Chat+ neurons increased integration between the DMN and somatomotor networks. Notably, these effects were state-dependent: lightly sedated animals showed weaker and qualitatively distinct RSN reconfigurations, suggesting that arousal gates BF-dependent control of RSNs dynamics. Together, our findings show that the BF causally control RSNs dynamics, with distinct BF neuronal pathways converging to modulate DMN activity and cortico-cortical integration. These results provide a mechanistic framework for interpreting state-dependent network reconfigurations observed with fMRI, and for linking neuromodulatory control to large-scale functional connectivity.