Abstract

RNA-protein interactions underlie many key processes in viral replication and antiviral immune responses. In this thesis, I reported our investigations on the roles of two RNA-binding proteins, Gag and IFIT1, in mediating HIV-1 genome packaging and regulating innate immune gene expression, respectively. The HIV-1 genome comprises a dimer of unspliced, positive sense mRNA (gRNA) that is packaged with high efficiency, with more than 90% of viral particles containing two copies of gRNA. Gag is the major HIV-1 structural protein that coordinates all steps of virion assembly and possesses a nucleocapsid domain (NC) required for gRNA binding and genome packaging. In infected cells, viral RNA is surrounded by a vast excess of cellular RNAs. Thus, it was previously thought that HIV-1 achieves efficient genome packaging through the selective binding of gRNA by Gag, mediated by specific high-affinity interaction between NC and the cis-acting packaging signal Ψ located within the HIV-1 5’-UTR. However, deletion of regions in Ψ only modestly reduces genome packaging, which led us to hypothesize that there are additional accessory factors important for this process. RNA plays a structural role in HIV-1 virion assembly and Gag-Gag multimerization was recently found to be important for genome packaging. Thus, an emerging model posits that HIV-1 gRNA is packaged selectively because it is more efficient at driving Gag multimerization than cellular RNAs. However, the feature(s) of viral RNA that make it a suitable catalyst for Gag multimerization are not well understood. Gag was previously reported to change its RNA binding specificity throughout HIV-1 virion morphogenesis, switching from G/U-rich binding in the cytoplasm to purine-rich (A/G) binding at the PM and in immature virions. This switch was proposed to play a role in genome packaging because HIV-1 gRNA has an unusually biased nucleotide composition (~40% adenosine). I investigated whether the binding preference of Gag toward purine-rich motifs plays a role in HIV-1 genome packaging and, consequently, whether the adenosine (A) richness of HIV-1 gRNA plays an accessory role in genome packaging. To this end, we replaced NC with heterologous RNA binding domains (RBDs) derived from cellular RNA binding proteins of the hnRNP and SRSF families to generate Gag chimeras with altered RNA binding specificity. We identified one chimera, Gag-SRSF5, that packaged gRNA nearly as efficiently as WT. Unexpectedly, while Gag-SRSF5 is the sole exception, all Gag chimeras were able to find and retain gRNA at the PM at near WT level. However, past this stage of gRNA recruitment to the PM, the majority of Gag chimeras exhibited assembly defects and failed to form buds. Gag-SRSF5 variants with reduced RNA binding affinity towards purine-rich motifs and altered specificity for C/G- or G-rich motifs displayed reduction in genome packaging efficiency. Altogether these results suggest that the biased nucleotide composition of the HIV-1 genome serves as a molecular signature that facilitate efficient genome packaging by promoting rapid Gag multimerization on viral RNA. These results also provide further supporting evidence for the model where HIV-1 gRNA is packaged efficiently based on the kinetics of Gag multimerization, and not based on specific interaction between NC and Ψ because diverse RBDs could find and retain Ψ-containing gRNA at the PM. Another long-standing question in the field is how HIV-1 ensures the packaging of a gRNA dimer. We hypothesized that dimeric gRNA packaging requires a tight balancing of Gag binding affinity and avidity towards gRNA, where the late arriving pool of Gag molecules need to dissociate from the bound cellular or viral RNA before arriving at virion assembly sites to drive the growth of immature Gag lattices. Therefore we replaced NC with up to seven tandem Zn2+ fingers from cellular nucleic acid binding protein (Gag-CNBPs) and duplicated/triplicated NC to generate chimeras with enhanced RNA binding affinity/avidity. Despite binding to more than WT Gag in cells with increasing number of Zn2+ fingers as intended, none of the Gag-NCx2/3 and Gag-CNBP chimeras packaged gRNA more efficiently than WT. Nevertheless, increasing the number of Zn2+ fingers progressively enhanced gRNA localization at the PM and genome packaging efficiency. These data suggest that the RNA binding affinity of Gag needs to be above a certain threshold to ensure efficient packaging and there is a clear selection process at the plasma membrane that limits the amount of gRNA packaged into virions. All Gag chimeras are non-infectious. Thus, we next asked whether they could block infectious when co-assembled with WT Gag in virions. Co-transfection of Gag chimeras with WT Gag in virus-producing cells exerted dominant negative effects, severely blocking infection while having negligible effects on particle release and genome packaging. Given their potent antiviral effects, Gag chimeras, which can co-assemble with WT Gag in virions, could be developed into anti-HIV-1 therapeutics that pose high genetic barriers to virus evolving resistance. Lastly, I described our investigation on the cellular protein interferon-induced protein with tetratricopeptide repeats-1 (IFIT1). IFIT1 is an interferon stimulated gene (ISG) that blocks viral translation by binding to viral RNA lacking 2′-O methylation and interacting with cellular proteins involved in translation. Emerging evidence suggests that IFIT1 also plays a role in regulating immune gene expression at the transcriptional level during lipopolysaccharide (LPS) stimulation and at the translational level during IFN response. Thus I monitored global transcriptional and translational changes in WT versus IFIT1-knockout THP-1 cells treated with and without type I interferon (IFN) stimulation using RNA-seq and ribosome profiling, respectively. For both IFN-treated THP-1 cells, IFIT1 deletion led to the transcriptional upregulation of immune genes, including ISGs and cytokines. At the translational level, the expression of immune genes was also upregulated in the absence of IFIT1. Interestingly, a number of pro-inflammatory cytokine genes are translated less efficiently in THP-1 cells upon IFN stimulation in an IFIT1 independent manner. Together, these data suggest that IFIT1 negatively regulates the expression of certain immune genes during IFN stimulation and certain cytokine genes are subjected to translational control during IFN response.

Committee Chair

Sebla Kutluay

Committee Members

Dave Wang; Liang Shan; Michael Diamond; Michele Leroux

Degree

Doctor of Philosophy (PhD)

Author's Department

Biology & Biomedical Sciences (Molecular Microbiology & Microbial Pathogenesis)

Author's School

Graduate School of Arts and Sciences

Document Type

Dissertation

Date of Award

7-24-2025

Language

English (en)

Author's ORCID

https://orcid.org/0000-0003-0359-0408

Download

Included in

Biology Commons

Share

COinS