Three-dimensional Structure and Evolution of HIV-1 Protease

Abstract

HIV-1 proteinase processes its virally encoded polyproteins into mature structural proteins and enzymes that are essential for viral propagation. As a consequence the proteinase is an attractive target for prospective antiviral agents for the treatment of AIDS, and knowledge of its tertiary structure an important step in drug design. Following the observation (Toh et al. 1985) that retroviral proteinases shared a highly conserved sequence Asp-Thr/Ser-Gly with the pepsins, it has been hypothesised (Pearl and Taylor, 1987; Blundell et al. 1988) on the basis of sequence analysis and modelling studies that these enzymes exist as dimers closely similar in three-dimensional structure to the ancestral dimeric proteinase suggested for the aspartic proteinases (Tang et al. 1978). This has now been confirmed, first by X-ray analysis of a synthetic HIV-1 proteinase in the laboratory of Wlodawer (Weber et al. 1989) and then for a recombinant enzyme in our own laboratories (Lapatto et al. 1989). These X-ray structure analyses indicated that the overall fold of the HIV-1 proteinase closely resembled that of the RSV-proteinase (Miller et al., 1989). The Asp-Thr-Gly sequences adopt a conformation closely similar to that of the pepsin-like aspartic proteinases but organised symmetrically in the dimer about the crystallographic 2-fold axis. However, the N- and C-termini together form an intermolecular four-stranded sheet, which is central to the stability of the dimer, in contrast to the inter-subunit sheet of the pepsins, which has six antiparallel strands arranged around the pseudo dyad.