Functional network disorganization and cognitive decline following fractionated whole-brain radiation in mice

Cognitive dysfunction following radiotherapy (RT) is one of the most common complications associated with RT delivered to the brain, but the precise mechanisms behind this dysfunction are not well understood, and to date, there are no preventative measures or effective treatments. To improve patient outcomes, a better understanding of the effects of radiation on the brain's functional systems is required. Functional magnetic resonance imaging (fMRI) has shown promise in this regard, however, compared to neural activity, hemodynamic measures of brain function are slow and indirect. Understanding how RT acutely and chronically affects functional brain organization requires more direct examination of temporally evolving neural dynamics as they relate to cerebral hemodynamics for bridging with human studies. In order to adequately study the underlying mechanisms of RT-induced cognitive dysfunction, the development of clinically mimetic RT protocols in animal models is needed. To address these challenges, we developed a fractionated whole-brain RT protocol (3Gy/day for 10 days) and applied longitudinal wide field optical imaging (WFOI) of neural and hemodynamic brain activity at 1, 2, and 3 months post RT. At each time point, mice were subject to repeated behavioral testing across a variety of sensorimotor and cognitive domains. Disruptions in cortical neuronal and hemodynamic activity observed 1 month post RT were significantly worsened by 3 months. While broad changes were observed in functional brain organization post RT, brain regions most impacted by RT occurred within those overlapping with the mouse default mode network and other association areas similar to prior reports in human subjects. Further, significant cognitive deficits were observed following tests of novel object investigation and responses to auditory and contextual cues after fear conditioning. Our results fill a much-needed gap in understanding the effects of whole-brain RT on systems level brain organization and how RT affects neuronal versus hemodynamic signaling in the cortex. Having established a clinically-relevant injury model, future studies can examine therapeutic interventions designed to reduce neuroinflammation-based injury following RT. Given the overlap of sequelae that occur following RT with and without chemotherapy, these tools can also be easily incorporated to examine chemotherapy-related cognitive impairment.