Recognized as an efficient approach to reduce fossil fuel consumption and alleviate environment crisis, the adoption of electric vehicles (EVs) in urban transportation system is receiving more and more attention. EVs will tightly couple the operations of urban transportation network (UTN) and power distribution network (PDN), necessitating the interdependent traffic-power modeling to optimize the on-road EV charging decisions. In this article, the optimal charging pricing problem of public electric vehicle charging stations (EVCSs) considering the interdependent operations of UTN and PDN is modeled from the view of the charging network operator (CNO), which is formulated as a bi-level optimization problem. In order to match the spatial size of UTN and power capacity of PDN, multiple PDNs are incorporated in the modeling to support EVCSs at different parts of UTN, permitting charging load migration across PDNs. Fixed point theory is used to analyze properties of the CNO's optimal charging pricing problem. An iterative method is designed to obtain the optimal charging pricing result, which solves the CNO's pricing subproblem and PDN market clearing subproblem in turn. Numerical case study results validate the effectiveness of proposed models and methods, and demonstrate the impacts of the CNO pricing on the operations of UTN and PDN.