tRNA Binding and Displacement: Implications for Specific Membrane Binding of HIV-1 Gag

Abstract

Specific trafficking of the HIV-1 Gag protein, synthesized as a 55-kDa precursor polyprotein, to the plasma membrane plays a central role in viral particle assembly. Past work strongly suggests that tRNA binding to the highly basic region (HBR) within the matrix (MA) domain regulates the plasma membrane specificity, through preventing Gag association with membranes containing prevalent acidic phospholipids (such as PS) but allowing Gag binding to membranes containing the plasma membrane-specific PI(4,5)P2. Additionally, in vitro and cellular data show that Gag selectively binds certain tRNA species; however, the MA-tRNA interface driving this selective interaction and hence membrane binding inhibition by the subset of tRNA is poorly understood. Additionally, multimerization of Gag, mediated primarily by the capsid (CA) and nucleocapsid (NC) domains, has been shown to enhance its membrane binding. Previous work has suggested that this effect is driven by increased avidity of multimerized Gag for membrane and exposure of N-terminal myristate induced upon multimerization; however, the effect of multimerization on RNA-mediated inhibition of Gag membrane binding has not yet been studied. In this thesis, I examined the contribution of individual elements of the tRNA structure to its inhibition of Gag membrane binding. Full-length Gag synthesized in vitro using reticulocyte lysates is strongly inhibited by tRNAs containing the anticodon arm of tRNAPro; however, in vitro transcribed tRNA are significantly fragmented in these experiments. In contrast, in assays using purified MA, full-length tRNALys3 showed greater inhibition of membrane binding than full-length tRNAPro. Our results demonstrate that beyond just the D loop sequence (recently reported to bind MA HBR), the entire D arm of tRNALys3 is a major determinant of strong inhibition of MA membrane binding. In this thesis, I have also explored the contribution of Gag multimerization to RNA-mediated regulation of membrane binding. Here, we show that multimerization of Gag reduces its sensitivity to tRNA-mediated inhibition. Additionally, CA-CA interactions driving the formation of higher order oligomer are essential for this resistance effect and are sufficient for inducing resistance of the MA domain to tRNA-mediated inhibition in the context of a minimal Gag construct, MACA. Finally, we found that MACA binds less tRNA than MA, which retains the same RNA binding site at the MA-HBR but lack multimerization through CA. These data suggest a new mechanism of multimerization-mediated enhancement of membrane binding, where Gag multimerization through the CA and NC domains prevent tRNA association with the MA-HBR. Together, our results provide additional detail on the mechanisms contributing to tRNA inhibition of Gag membrane binding. We demonstrate that the D-arm is a major contributor to tRNA-mediated inhibition of MA membrane binding and that multimerization of the Gag protein reduces tRNA association with MA-HBR, priming Gag for efficient membrane binding. This information could contribute to the development of RNA-based therapeutic strategies targeting the assembly stage of HIV-1 replication.