Computer Analysis of the Conservation and Uniqueness of Ribozyme-Targeted HIV Sequences

Ribozymes are currently being developed as therapeutic agents against AIDS (
1
,
2
). Successful cleavage of the mRNAs produced by HIV or the HIV genome itself could produce improvement by significantly reducing the patients’ viral burden. However, choosing the appropriate target sites in HIV RNA is complicated by its high mutability. It is estimated that nine sequence errors occur with each reverse transcription of the genome (
3
). Since ribozyme action is highly sequence-dependent, the virus may thus escape ribozyme cleavage with time. However, some mutations will inactivate critical viral functions and cannot be propagated in the host. The goal of ribozyme design is to locate and target the sequences in HIV RNA whose importance results in conservation in a majority of patients. An additional concern is that these highly conserved HIV sequences may not be virus-specific. If an important functional motif is also present in normal human mRNA, it too might be cleaved by the HIV therapeutic ribozyme, possibly resulting in harmful side effects. For these reasons, a procedure has been developed for searching the available sequence databases to determine (
1
) an approximate percentage of conservation among HIV strains for a potential ribozyme target site and (
2
) which known human mRNAs have a similar sequence. The search process can be performed relatively easily using any computer system with access to the Internet.