Background

High-field mouse fMRI is an important translational tool to bridge the gap between invasive research in mouse models of neuropsychiatric diseases and clinical research in patients. However, compared to human fMRI, mouse fMRI studies are typically limited due to anesthesia, which is necessary to suppress body motion and stress during fMRI acquisition. Performing mouse fMRI studies under light anesthesia can ameliorate body motion, but the extent to which functional brain networks and connectivity are influenced by light anesthesia remains unclear. In general, small doses of dexmedetomidine, medetomidine, isoflurane or a mixture of these anesthetics are commonly used to achieve light anesthesia to investigate brain function in rodents (Grandjean et al., 2014; Bukhari et al., 2017; Tsurugizawa et al., 2019 and 2020a). However, these anesthetics induce not only suppression of consciousness (Munglani et al., 1993) but also vasoactive modulation of neurovascular coupling (Tsurugizawa et al., 2010 and 2016), which potentially impacts the relationship between neuronal activation and vascular response. In particular, light sedation alters the BOLD response to physiological stimuli and thus affects the validity of task-based fMRI (Tsurugizawa et al., 2013a). Anesthesia clearly depresses consciousness, even if residual brain function is comparable to an awake state. In addition, it is impossible to perform cognitive tasks under anesthesia. In summary, fMRI under anesthesia, even with light anesthesia, is not comparable to fMRI in a conscious state, and thus rodent fMRI performed in an anesthetized state considerably narrows opportunities to comprehensively characterize brain function and connectivity. Hence, we developed an awake fMRI system for mice that does not necessitate anesthesia, thus enabling mouse fMRI experiments that are analogous to human fMRI.

Previous studies, including our own, have developed fMRI systems and protocols to acquire fMRI in awake mice. These have been used to investigate responses to fear conditioned stimulation (Harris et al., 2015) and optogenetics (Desai et al., 2011) and to investigate resting state functional connectivity (Bergmann et al., 2016; Yoshida et al., 2016; Madularu et al., 2017). Key limitations of the protocols used in these studies include the use of restraint tubes, absence of earplugs to reduce scanner noise (Mowery et al., 2019; Kurioka et al., 2020) and unclear animal handling and acclimation methods. We first developed an awake-mouse fMRI system to investigate the intragastric stimulation of capsaicin (Tsurugizawa et al., 2013b). More recently, we enhanced the system with the addition of a head fixation apparatus and acclimation training. The new system was used to investigate brain function in mice with a chromosome duplication (15q dup) resulting in abnormal behavior resembling ASD symptoms (Nakatani et al., 2009; Tsurugizawa et al., 2020b).

In this report, we describe our fMRI system and protocol for acquiring fMRI in awake mice, including cranioplastic surgery, acclimation training and fMRI acquisition. The protocol enables routine resting-state and task-fMRI mouse studies, where anesthetization of animals introduces a major confound.