While all the determinants of tropism of HIV replication are still being elucidated, the major determinants reside in the envelope of the virus and are present in both the extracellular (gp120) and the transmembrane (gp41) components. However, other genes (gag, nef, vpr, vif) and cis-acting elements (the long terminal repeat, LTR, the primer binding site, PBS, and the polypurine tract, PPT) may also modulate replication in different cell types. Our previous work has involved characterizing determinants of tropism that reside in the envelope gene. The approach involves passaging a virus with reduced replication capacity on cells in which that reduced replication ability is manifested. Because of the high propensity of HIV to mutate, the emergence of variants with increased replication capacity is frequently seen. The type and location of mutations that account for the observed phenotype have been informative in delineating those genes or regions of the HIV genome invloved in the particular function. Recently, we have applied this approach to the study of the primer binding site (PBS) and the polypurine tract (PPT) of the ROD strain HIV-2. While most HIV-1 and HIV-2 strains use the host tRNA Lys3 species as primer for the minus-strand cDNA synthesis, the ROD10 clone of HIV-2 carries a sequence change in the PBS that results in a different tRNA being incorporated. This tRNA corresponds to a tRNA species that we refer to as tRNA Lys7. The question of whether tRNA Lys7 is used for HIV-2 ROD10 replication and whether the non-canonical PBS is maintained during passage was addressed by sequencing the PBS after passaging in PBMC and T-cell lines. There was rapid conversion to the canonical PBS after even a single passage in either PBMC or cell lines. After two passages, the proportion of the PBS corresponding to the Lys7 tRNA was about 30 to 50%. Further passages are underway to determine if the non-canonical PBS reverts completely. In addition, since there are determinants other than the PBS for incorporation of the appropriate tRNA, we are constructing hybrid viruses between ROD10 and ROD2, another isolate of HIV-2 ROD that has the canonical PBS and uses tRNA Lys3 as primer, in order to identify and map these other determinants. HIV-2 ROD has another unusual feature. The PPT is not the usual all-purine canonical element of the majority of retroviruses but has a PPT that is interrupted with a pyrimidine residue. Since this virus is infectious, we sequenced the PPT from passaged virus and found that the non-canonical PPT was retained on culture. Furthermore, that there are three clones of ROD and all three have this PPT demonstrates that this PPT exists in vivo, that it is fully functional, and that it is selected for. Future studies will assess the interaction between the PPT and RT/RNaseH by replacing the ROD PPT with a canonical PPT and measuring the stability of this PPT.
