Investigation of Interactions between Rev and Microtubules: Purification of Wild-type and Mutant Rev Protein and Optimization of Microtubule Depolymerization Assays

As a logical pharmaceutical target for antiviral drugs, HIV-1 Rev is a regulatory protein essential for viral infection (Hope, 1999). The development of antiviral drugs that target Rev has been hindered by the lack of high-resolution structural information due to the protein\u27s tendency to aggregate in solution. While searching for solution conditions rendering Rev amenable to crystallographic analyses, Watts et al., (2000) discovered a novel in vitro interaction between Rev and microtubules (MTs) whereby addition of equimolar Rev and tubulin forms bilayered rings called Rev-tubulin toroidal complexes (RTTs). RTTs are similar to those seen when MTs are mixed with certain anti-MT drugs and KinI kinesins (Watts et al., 2000). Coupled with the sequence homology that exists between KinI kinesins and Rev, I hypothesize that Rev and KinI\u27s are depolymerizing MTs by a shared mechanism. This mechanism may include the binding of the Rev/KinI to the end of the MT inducing a curved confirmation in the MT thereby destabilizing it to promote depolymerization. I propose to test this hypothesis through measuring Rev-MT interactions by adapting biochemical and microscopy-based assays used to measure MT depolymerization by KinI proteins. These assays require microgram amounts of highly purified wild-type and mutant Rev proteins as well as purified tubulin from which MTs can be polymerized in vitro. To this end, I have purified wild type Rev with no visible contaminants on coomasie stained gels. Rev mutants R42A and E57A can also be purified with limited visible contaminants. However, the appropriate controls can be generated from non-expressing cells to address this issue. Rev mutants A37D and R39A can also be partially purified. Using purified Rev proteins, I then applied sedimentation assays to measure Rev-stimulated MT depolymerization. There was a statistically significant time dependence for wild type Rev to depolymerize MTs although there was no evidence for concentration dependence. Visual assays demonstrate no significant difference in the length of MTs treated with Rev although Rev decreased the number of MTs on the coverslip over time. This could contribute to the finding that MT depolymerization by Rev is time dependent. These results demonstrate that it is possible to measure Rev-MT interactions in vitro although it is clear that these assays are deficient in certain ways, including the ability of MTs to depolymerize on their own and RTTs potentially pelleting during the depolymerization assay. However, it is likely that cycling tubulin and/or using taxol to further stabilize the MTs can remedy the deficiency of MT depolymerization on their own. EM could also be used to determine if RTTs are pelleting during the depolymerization assay. After using mutant Rev proteins in the depolymerization and visual assays, it is the long-term goal that mechanistic information about Rev will lend valuable evidence to the study of KinI kinesins. Generating structural Rev information may also be helpful in the drug design of anti-mitotic peptides