With the popularity of storing large data graph in cloud, the emergence of subgraph patt matching on a remote cloud has been inspired. Typically, subgraph patt matching is defined in terms of subgraph isomorphism, which is an NP-complete problem and sometimes too strict to find useful matches in certain applications. And how to protect the privacy of data graphs in subgraph patt matching without undermining matching results is an important conc. Thus, we propose a novel framework to achieve the privacy-preserving subgraph patt matching in cloud. In order to protect the structural privacy in data graphs, we firstly develop a k-automorphism model based method. Additionally, we use a cost-model based label generalization method to protect label privacy in both data graphs and patt graphs. During the generation of the k-automorphic graph, a large number of noise edges or vertices might be introduced to the original data graph. Thus, we use the outsourced graph, which is only a subset of a k-automorphic graph, to answer the subgraph patt matching. The efficiency of the patt matching process can be greatly improved in this way. Extensive experiments on real-world datasets demonstrate the high efficiency of our framework.