As the number of schema repositories grows rapidly and several web-based platforms exist to support publishing schemas, \emph{schema reuse} becomes a new trend. Schema reuse is a methodology that allows users to create new schemas by copying and adapting existing ones. This methodology supports to reduce not only the effort of designing new schemas but also the heterogeneity between them. One of the biggest barriers of schema reuse is privacy concerns that discourage the participants from contributing their schemas. Addressing this problem, we develop a framework that enables privacy-preserving schema reuse. To this end, our framework supports users to define their own protection policies in the form of \emph{privacy constraints}. Instead of showing original schemas, the framework returns an \emph{anonymized schema} with maximal \emph{utility} while satisfying these privacy constraints. To validate our approach, we empirically show the efficiency of different heuristics, the correctness of the proposed utility function, the computation time, as well as the trade-off between utility and privacy.