Human immunodeficiency virus type 1 proteinase : regulation of function and catalytic activity

HIV-l proteinase activity is a prerequisite for viral replication, and one of the main targets for anti-HlV therapy. Although potent proteinase inhibitors are currently being approved for treatment of AIDS patients, these drugs are unlikely to provide treatment which can give patients a life of normal quality and duration. The development of improved drugs against HIV/AIDS requires further understanding of the function and activity of their targets. The work of this thesis has focused on the identification and characterisation of different factors that regulate HIV-l proteinase function and activity.

Methods for expression and purification of HIV-l proteinase from bacteria were developed, and proteinase activity in vitro was measured using synthetic peptides representing sequences of the naturally occurring cleavage sites. HIV-l and AMV proteinases were found to inhibit microtubule assembly in vitro, and this inhibition was associated with cleavage of microtubule- associated proteins 1 and 2. These results showed that the substrate specificity of HIV-l proteinase in vitro was not restricted to viral proteins, and indicated a potential cytotoxicity of the proteinase during viral infection. The substrate specificity of HIV-l proteinase is not only determined by the amino acid sequence of the cleavage site and by its secondary structure, but also by its accessibility. It is the tertiary or quaternary structure of the substrate protein which will ultimately determine whether a site will be cleaved or not. Such a structural determinant regulating the limited cleavage of p66 RT to p5 1 RT was identified, and consisted of a salt bridge between D488 and L465 located just above the proteinase cleavage site.

HIV-l RT was found to enhance proteinase activity in a dose-dependent manner both in vitro and in cell culture. The degree of enhancement was substrate-dependent, with the lowest effect on the proteinase autocleavage site and the site between the proteinase and RT, and with the highest effect on the cleavage site between RT and the integrase, and between p5 1 RT and RNase H. RT enhancement of proteinase activity was independent of pH and ionic strength, and was not due to an increase in proteinase dimerization. RT could enhance the activity of a tethered proteinase dimer and of the wild type proteinase to the same degree, by increasing their kcat's and decreasing their KM'S.

Cross-linking experiments of RT and proteinase showed that the proteinase cross-linked mainly to the monomer forms of RT. Studies of deletion mutants indicated that sequences in the polymerase and RNase H domain of RT are needed for maximal effect on proteinase activity. Three putative interacting sites in the RNase H domain of RT were identified using overlapping peptides coding for the entire RT sequence. The most active peptide corresponded to the region of RNase H that had to be unravelled in order for processing of RT to occur. These results suggest that in vivo, RT may regulate its own cleavage from the polyprotein precursor.

At low ionic strength, and after equilibration at 37¡, proteinase activity in vitro was found to deviate from normal Michaelis-Menten kinetics even though activity was stable under the conditions used. This could be overcome by increasing the ionic strength of the solution, or by inhibiting proteinase dissociation by avoiding preincubation or using a tethered proteinase dimer. These results indicate that deviation from normal kinetics is due to changes in the degree of proteinase dimerization upon addition of substrate, and that at least in vitro, HIV-l proteinase is activated by its substrate.