Connected and automated vehicles (CAVs) and infrastructure-to-vehicle (I2V) communication have great potential to improve traffic safety and mobility at intersection and road segment levels. However, since transportation is an interconnected network, local transportation state change could lead to broader impacts by rerouting. In this study, we focus on investigating the impacts of en-route locational information updates on the system-level safety and mobility considering adaptive routing of CAVs. We propose a novel two-stage stochastic traffic equilibrium model to characterize the equilibrium traffic patterns considering adaptive routing behavior when en-route traffic information is provided. The proposed methodology is tested using the Orlando transportation network with real traffic and accident data. Different I2V information sharing strategies are analyzed to better understand the impacts of information sharing locations of I2V on transportation network safety and mobility, through which we generate insights on the optimal design of traffic information sharing systems.