Abstract Virtually every step of HIV-1 replication as well as numerous cellular antiviral defense mechanisms are regulated by viral and cellular RNA-binding proteins (RBPs) that recognize distinct sequence or structural features on viral RNAs. One such interaction takes place between the HIV-1 major structural protein, Gag, and the viral genomic RNA. Gag packages two copies of an unspliced positive strand viral genome in particles, which are selected from a pool of cellular and spliced viral mRNAs in excess. How viral genomes are selected for packaging and why only two copies are packaged in a single virus particle remain poorly understood. By extension of findings from simple retroviruses, such as murine leukemia virus (MLV), it has long been thought that HIV-1 selective genome packaging is similarly regulated by a cis-acting packaging signal, Psi (Ψ), located within the 5’ untranslated region of the genome. However, unlike MLV, disruption of regions in Ψ only modestly impacts packaging and regions outside of the Ψ sequence might contribute to genome encapsidation. Thus the precise rules that govern HIV-1 selective genome packaging still remain poorly understood. HIV-1 genome has an unusually biased nucleotide composition, rich in adenosines (~36%) and poor in cytosines (~18%). We have previously discovered that Gag binding to the HIV-1 genome is highly dynamic and undergoes several changes coincident with its membrane binding, multimerization, and proteolytic maturation. In particular, we found that while cytosolic Gag bound to guanosine-rich sequences, its binding preference shifted towards sequences with adenosine-rich nucleotide composition at the plasma membrane concomitant with its multimerization. In this application, we propose to test the novel idea that the overall nucleotide content of the HIV-1 genome contributes to its selective packaging. We will conduct a series of genetic approaches in which the nucleocapsid (NC) domain of Gag is replaced by heterologous RNA-binding domains and by determining the efficiency with which minimal genomes with A-rich vs. A-poor nucleotide content are packaged into virions (Aim 1). Through replacement of NC by heterologous RNA-binding domains and increasing NC copy number per Gag molecule, we propose to determine whether dimeric genome packaging is driven by specificity, affinity and avidity of Gag towards viral RNAs (Aim 2). Overall, this project will provide novel insight into mechanisms of selective genome packaging. Understanding this process is not only significant from a basic molecular biology standpoint but will also impact gene therapy and CRISPR-based engineering tools which depend on retroviral systems for efficient delivery into host cells.