Baseline CBV-normalized BOLD fMRI of visual stimulation in awake vs. anesthetized mice

Abstract

Anesthesia profoundly affects baseline physiology, including CBV and evoked BOLD fMRI responses. To disentangle evoked BOLD responses into CBV and pure BOLD components, we conducted mouse fMRI experiments at 15.2 T using 5 Hz visual stimulation and a 4-second 10% O₂ hypoxic stimulus under awake and 0.5% isoflurane-anesthetized conditions. With a newly developed mouse cradle that allowed free movement of the paws, no obviously unusual behaviors were observed, and head motion was acceptable even without habituation. Anesthesia increased baseline CBV globally, impacting evoked BOLD responses. Visual stimulation elicited significant BOLD responses in visual networks, including the lateral geniculate nucleus, superior colliculus, lateral posterior thalamus, and visual cortical areas under both conditions. In the awake state, additional activation was observed in regions such as the anterior cingulate cortex, retrosplenial cortex, and prelimbic cortex. Normalizing BOLD fMRI responses using baseline CBV-weighted hypoxic response reduced the contribution of large vessels, minimizing CBV-induced variations in fMRI responses across active voxels and regions. This normalization enhanced the differences in fMRI responses between awake and anesthetized conditions, demonstrating the importance of BOLD normalization for accurate comparison between fMRI responses across different CBV levels.