Drug delivery is the major hurdle to success of therapeutic treatments. A large number of potential drugs with high intrinsic activities fall by the wayside during the development because they have unfavourable physicochemical properties.The aim of this proposal is to develop a general strategy for the encapsulation of guest molecules inside container molecules. Current encapsulation strategies make use of- strong covalent bonds to form the molecular container under kinetic control, which results in essentially irreversible imprisonment of the guest; or- weak non-covalent interactions to assemble the capsules under thermodynamic control, leading to relatively labile products.Neither of these strategies is suitable for drug delivery. We will develop a new strategy towards encapsulation using thermodynamically controlled synthesis to assemble capsules that are held together by covalent bonds that are stable under ordinary conditions but that will release their guest under the influence of a physiologically relevant trigger.Capsules will be assembled by linking relatively simple building blocks together through disulfide bonds. We will prepare equilibrium mixtures (dynamic combinatorial libraries) of potential capsules under conditions of reversible disulfide bond formation. Introduction of a guest molecule into these mixtures should lead to the stabilisation of those capsules that can accommodate this guest.The concentration of the best capsules will increase, which should enable their identification. Once the structure of the selected capsules has been elucidated the same reversible chemistry can be used for thermodynamically-controlled capsule synthesis.Then, the conditions required for the release of the guest from the resulting capsules will be established in model systems that mimic a cellular environment. This approach may be developed for future use in drug targeting or other applications requiring controlled release of chemical substance.