In this paper, output-feedback adaptive stabilization is investigated for a class of nonlinear systems with unknown control directions. First, through a linear state transformation, the unknown control coefficients are lumped together and the original system is transformed to a new system for which control design becomes feasible. Then, after the introduction of an observer and an estimator for state and parameter estimates, respectively, a constructive design procedure is given for the output-feedback adaptive stabilizing controller using integrator backstepping and tuning function techniques. It is shown that the controller designed ensures the original system state converges to the origin whereas all the other closed-loop system states are bounded. Simulation results are illustrated to show the effectiveness of the proposed approach.