Dissociable roles of central striatum and anterior lateral motor area in initiating and sustaining naturalistic behavior
Although much is known about how corticostriatal circuits mediate behavioral selection, most previous work has been conducted in highly trained animals engaged in instrumental tasks. Understanding how corticostriatal circuits mediate behavioral selection and initiation in a naturalistic setting is critical to understanding how the brain chooses and executes behavior in unconstrained situations. Central striatum (CS), an understudied region that lies in the middle of the motor-limbic topography, is well-poised to play an important role in these processes since its main cortical inputs (Corbit et al., 2019) have been implicated in behavioral flexibility (lateral orbitofrontal cortex (Kim and Ragozzino, 2005)) and response preparation (anterior lateral motor area, ALM) (Li et al., 2015), However, although CS activity has been associated with conditioned grooming behavior in transgenic mice (Burguiere et al., 2013), the role of CS and its cortical inputs in the selection of spontaneous behaviors has not been explored. We therefore studied the role of CS corticostriatal circuits in behavioral selection in an open field context. Surprisingly, using fiber photometry in this unconstrained environment, we found that population calcium activity in CS was specifically increased at onset of grooming, and not at onset of other spontaneous behaviors such as rearing or locomotion. Supporting a potential selective role for CS in the initiation of grooming, bilateral optogenetic stimulation of CS evoked immediate onset grooming-related movements. However, these movements resembled subcomponents of grooming behavior and not full-fledged grooming bouts, suggesting that additional input(s) are required to appropriately sequence and sustain this complex motor behavior once initiated. Consistent with this idea, optogenetic stimulation of CS inputs from ALM generated sustained grooming responses that evolved on a time-course paralleling CS activation monitored using single-cell calcium imaging. Furthermore, fiber photometry in ALM demonstrated a gradual ramp in calcium activity that peaked at time of grooming termination, supporting a potential role for ALM in encoding length of this spontaneous sequenced behavior. Finally, dual color dual region fiber photometry indicated that CS activation precedes ALM during naturalistic grooming sequences. Taken together, these data support a novel model in which CS activity is sufficient to initiate grooming behavior, but ALM activity is necessary to sustain and encode the length of grooming bouts. Thus, the use of an unconstrained behavioral paradigm has allowed us to uncover surprising roles for CS and ALM in the initiation and maintenance of spontaneous sequenced behaviors.