The packaging determinants of feline immunodeficiency virus (FIV) consist of two discontinuous core regions, extending from R to ϳ150 bp of the 5 untranslated region and the first ϳ100 bp of gag. However, the role of sequences intervening between the core regions in packaging has not been clear. A mutational analysis was conducted to determine whether the intervening sequences played a role in FIV RNA packaging, using an in vivo packaging assay complemented with semiquantitative reverse transcriptase PCR. Our analyses reveal that the intervening sequences are dispensable not only for vector RNA packaging but also for propagation, confirming the discontinuous nature of the FIV packaging signal.A series of recent studies have suggested that the packaging determinants of feline immunodeficiency virus (FIV) are complex and multipartite, consisting of at least two discontinuous core regions, one located upstream of the major splice donor sequence from R/U5 at the 5Ј end to the first ϳ150 bp of the 5Ј untranslated region (UTR), while the other is within the first 100 bp of gag (2, 3, 6, 7). To determine whether the region intervening between the core determinants was required for maintaining the stability of the FIV packaging signal, we modified our previously described vector, MB15 (2), and constructed a series of transfer vectors, AG002 to AG004, that maintained the R/U5 and 100 bp of gag but kept only the first 90, 120, or 150 bp of the 5Ј UTR, generating incremental deletions between the core packaging determinants (Fig. 1). In a second series, AG013 to AG015, the deletions were replaced with heterologous sequences of the same lengths to further determine whether the deleted/substituted region had a role at the structural level or acted only as a spacer (Fig. 1).The two series of mutant vector RNAs were tested for their ability to be packaged by FIV proteins, using our well-established in vivo packaging assay along with TR394, which contains the entire 270-bp UTR with a 333-bp gag sequence (1-3). The amount of transfer vector RNA packaged into viral particles was analyzed by reverse transcriptase PCR (RT-PCR) (10) for variable cycle numbers, and the resulting products were Southern blotted and hybridized using an R/U5 probe. All mutant vector RNAs, either with deletions or deletion/ substitution, were packaged into the virus particles but with different efficiencies, while the control vectors, MB15 and TR394, were packaged at similar levels ( Fig. 2A). This was despite the fact that all cultures produced similar levels of viral particles and the transfection efficiencies were within twofold of each other ( Fig. 2B and C). Since the PCRs were conducted across the deleted or deleted/substituted region, the size of the PCR fragment varied with each construct, confirming that correct packaging constructs were being expressed in each transfection.To determine the packaging efficiencies of the various constructs accurately, RT-PCRs were repeated using primers within the R/U5 region that would result in same-sized fragme...