The rapid proliferation of advanced mobile devices has created a growing demand for data content. Existing approaches (e.g. relying on cellular infrastructures) cannot keep up with the large volume of content generated and requested, without the deployment of new expensive infrastructure. Exchanging content of interest opportunistically, when two nodes are in range, presents a low cost and high bandwidth alternative for popular, bulky content. Yet, efficiently collecting, storing, and sharing the content while preventing selfish users from impairing collaborative ones, poses major challenges. In this paper, we present MobiTrade, a collaborative content dissemination system on top of a delay tolerant network. It allows users to head out in the real world, express locally their interests, and wait to get notified whenever an encountered device has content(s) matching these interests. Even though interactions are done between neighboring wireless devices (locally), MobiTrade implements a trading scheme that motivates mobile devices to act as merchants and carry content across the network to satisfy each other's interests. Users continuously profile the type of content requested and the collaboration level of encountered devices. Based on this knowledge, an appropriate utility function is used to rank these requests and collect an optimal inventory of data that maximizes the expected value of stored content for future encounters. Using NS3 simulations based on synthetic and real mobility traces, we show that MobiTrade achieves up to 2 times higher query success rates compared to other content dissemination schemes. Furthermore, we show that MobiTrade successfully isolates selfish, non-collaborative devices. Finally, using a simple game theoretic framework we show that turning on our MobiTrade mechanism is an efficient Nash Equilibrium.