Joint employment of multi-hop data forwarding and mobile data-collector is a popular technique for efficient data collection in energy constraint and delay sensitive wireless sensor networks (WSNs). Existing tree-based data forwarding methods take the joint advantages of clustering and multi-hop data forwarding. However, the performances of all these approaches hardly meet the desired level of efficiency and thus, finding an efficient tree-clustering method to save network energy and extending the network lifetime is still a relevant issue in WSN. In this work, we study the problem of multi-hop data forwarding and propose a novel tree-clustering scheme named energy efficient tree clustering (EETC) which minimizes network energy consumption and extend the network lifetime while maintaining a pre-bound tour delay of the mobile sink. EETC uses a heuristic clustering algorithm named Optimal Generation of Clusters (OGENCL) in the clustering phase. In the proposed method, the number of relay hops between a cluster member node and the CH has been restricted to balance the network load. For further balancing the network load, we use an upper bound on the cluster size. The OGENCL problem is formulated as a Mixed Integer Linear Programming (MILP) Problem. Extensive simulations have been performed and compared with existing works to show the effectiveness of the proposed scheme for network load balance, energy consumption and network lifetime.