We consider joint optimization of transmit power, serving base station cluster (BSC) for coherent reception, beamforming and subcarrier allocation for the uplink channel of a multicell multicarrier system. The objective is to minimize the sum transmit power subject to user-specific rate constraints. Since the problem is nonconvex and NP-hard, finding the optimal solution is challenging and not practically appealing. We propose two bit loading algorithms, in which the data rate of users is iteratively increased until the rate targets are satisfied. A joint power control, BSC selection and beamforming design problem with fixed rate allocation per subcarrier is optimally solved in every iteration. A bit switching (BSW) algorithm is further proposed in order to search for better rate allocations. The performances of the proposed algorithms are compared to upper and lower bounds achieved by the capacity-achieving scheme and equal rate allocation over subcarriers (ER), respectively. Simulation results show that proposed algorithms achieve significantly better performance than the ER scheme. The performance gains become larger as the number of users is increased.