In this paper, we investigate the resource allocation issues in spectrum-sharing-based two-tier networks, which are composed of several underlaid femtocells and a macrocell. The macrocell base station (MBS) partially shares its sub-carriers with femtocell users (FUs) while being protected by per-sub-carrier interference temperature constraint. We formulate this multi-user-multi-channel problem as a non-cooperative game, in which each FU acts individually to maximize its own profits, regardless of the impact on other players. To avoid serious interference to MBS, we introduce a price-based utility function of FUs with a per-channel interference constraint instead of the spectrum mask constraint. We prove the existence and uniqueness of Nash Equilibrium (NE) and introduce fully distributed algorithm based on simultaneous iterative-water-filling (IWF) like algorithm, which adjusts the sub-carrier price iteratively and the maximum transmit power. This approach enables FUs to select suitable sub-carriers and allocate the power in a distributed manner. Meanwhile, it gives sufficient protection to MBS in each sub-carrier. Experimental results have shown that the proposed approach converges fast to the NE while providing sufficient protection to the macrocell users.