To facilitate content delivery to mobile users, we propose a content caching and distribution framework for the heterogeneous OFDM networks, where a library of files available at the macro base station (MBS) can be distributively cached in multiple serving nodes (SNs). SNs are capable of both receiving unstored files from MBS and transmitting files to the associated users. For a given group of file downloading requests, the user scheduling scheme is jointly designed with the content caching scheme so that the number of served users is maximized for a given amount of spectrum and time. First, the corresponding downlink throughput maximization problem is shown to be NP-hard. Then, for the system with one MBS and one SN, the design of the joint user scheduling and caching scheme is transformed into a binary linear programming problem. For the system with one MBS and two SNs, the joint scheduling and caching (JSC) algorithm has been proposed to tap on the potential of the coded multicasting scheme between MBS and SNs. The proposed algorithm can be extended to networks with arbitrary number of SNs. Simulation results demonstrate that the JSC algorithm provides a significant sum throughput gain.