In this paper, the bipartite consensus tracking problem on one class nonlinear multi-agent systems with antagonistic interactions is investigated in details. The communication among agents is depicted as a directed signed graph, which is assumed to be structurally balanced. The bipartite consensus tracking problem is furthermore formulated by integrating with the signed graph network. Based on the nearest neighborhood rule, the distributed control law is designed accordingly with the aid of Laplacian matrix, pinning matrix, signed function and the coupling gain parameter. The asymptotic stability of the closed-loop system is rigorously proved by a Lyapunov function with the aid of Barbalat lemma. The lower bound for the coupling gain is derived to guarantee the closed-loop system be stable. The effectiveness of the proposed approach is finally verified by simulation results.