This paper focuses on two-cell multiple-input multiple-output (MIMO) Gaussian interfering broadcast channels (MIMO-IFBC) with K cooperating users on the cellboundary of each BS. It corresponds to a downlink scenario for cellular networks with two base stations (BSs), and K users equipped with Wi-Fi interfaces enabling to cooperate among users on a peer-to-peer basis. In this scenario, we propose a novel interference alignment (IA) technique exploiting user cooperation. Our proposed algorithm obtains the achievable degrees of freedom (DoF) of 2K when each BS and user have M = K + 1 transmit antennas and N = K receive antennas, respectively. Furthermore, the algorithm requires only a small amount of channel feedback information with the aid of the user cooperation channels. The simulations demonstrate that not only are the analytical results valid, but the achievable DoF of our proposed algorithm also outperforms those of conventional techniques.