Mutations that occur in response to the HIV-1 protease inhibitors are responsible for the development of multidrug cross-resistance to these antiproteases in AIDS treatment. One alternative to inhibiting the active site of HIV-1 protease is to target the dimer interface of the homodimeric enzyme at the antiparallel beta-sheet formed by the interdigitation of the C- and N-ends of each monomer. This region is highly conserved and is responsible for approx. 75% of the dimer-stabilization energy. The strategies that have been used to design small molecules to target the interface antiparallel beta-sheet have produced lipopeptides, guanidinium derivatives and peptides (or peptidomimetics) cross-linked with spacers. The mechanism of inhibition was determined using a combination of kinetic and biophysical methods. These dimerization inhibitors proved equally active in vitro against both wild-type and mutated proteases. They are therefore promising alternatives to active-site-directed inhibitors in AIDS therapy. Disruption of protein-protein interactions by small molecules is a new way to obtain potentially therapeutic molecules.