Given an n-vertex graph G=(V,E) and a set R \subseteq {{x,y} | x,y \in V} of requests, we consider to assign a set of edges to each vertex in G so that for every request {u, v} in R the union of the edge sets assigned to u and v contains a path from u to v. The Minimum Certificate Dispersal Problem (MCD) is defined as one to find an assignment that minimizes the sum of the cardinality of the edge set assigned to each vertex. This problem has been shown to be LOGAPX-complete for the most general setting, and APX-hard and 2-approximable in polynomial time for dense request sets, where R forms a clique. In this paper, we investigate the complexity of MCD with sparse (tree) structures. We first show that MCD is APX-hard when R is a tree, even a star. We then explore the problem from the viewpoint of the maximum degree Δof the tree: MCD for tree request set with constant Δis solvable in polynomial time, while that with Δ=Ω(n) is 2.56-approximable in polynomial time but hard to approximate within 1.01 unless P=NP. As for the structure of G itself, we show that the problem can be solved in polynomial time if G is a tree.