Computer-aided search for effective antisense RNA target sequences of the human immunodeficiency virus type 1.

For the biological and therapeutic application of antisense nucleic acids, there is a need to identify effective local target regions of given cellular target mRNAs or viral single-stranded nucleic acids. One critical parameter for the effectiveness of antisense nucleic acids could be the potential of intramolecular folding of a given sequence element of the target strand and the antisense strand, respectively. The folding potential of such subsequences was calculated by using an established secondary structure prediction algorithm. For the genomic RNA and the complementary RNA strand of the human immunodeficiency virus type 1 (HIV-1), an energy profile was calculated that monitors the local folding potential of each sequence position surrounded by a window of given length ranging from 50 to 400 nucleotides. The resulting energy profile was compared to the effectiveness of HIV-1-directed antisense RNAs. It was found that significant minima of the local folding potential (high delta G values) correlated with antisense RNA target regions involved in strong inhibition of HIV-1 replication that had been measured independently in two earlier studies by using different experimental approaches. Conversely, antisense RNAs directed against local subregions with a high folding potential (low delta G values) showed weak or no antiviral effect in human cells. The results indicate that analyses of the local folding potential of a given target RNA can support the selection of effective target sequences for antisense RNA.