COMPUTER-ASSISTED RATIONAL DESIGN OF IMMUNOSUPPRESSIVE COMPOUNDS

We describe the rational design of immunosuppressive peptides without relying on information regarding their receptors or mechanisms of acti on. The design strategy uses a variety of topological and shape descri ptors in combination with an analysis of molecular dynamics trajectori es for the identification of potential drug candidates. This strategy was applied to the development of immunosuppressive peptides with enha nced potency. The lead compounds were peptides, derived from the heavy chain of HLA class I, that modulate immune responses in vitro and in vivo. In particular, a peptide derived from HLA-B2702, amino acids 75- 84 (2702.75-84) prolonged skin and heart allograft survival in mice. T he biological activity of the rationally designed peptides was tested in a heterotopic mouse heart allograft model. The molecule predicted t o be most potent displayed an immunosuppressive activity approximately 100 times higher than the lead compound.