Intracellular trafficking of retroviral Gag and RNA during late replication
Open Access
- Author:
- Bann, Darrin V
- Graduate Program:
- Cell and Molecular Biology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 09, 2013
- Committee Members:
- Leslie Joan Parent, Dissertation Advisor/Co-Advisor
Leslie Joan Parent, Committee Chair/Co-Chair
Sarah Bronson, Committee Member
Edward Joseph Gunther, Committee Member
Jeffery Thomas Sample, Committee Member
David Joseph Spector, Committee Member - Keywords:
- Retrovirus assembly
RNA trafficking
Live-cell imaging
Mouse mammary tumor virus
Rous sarcoma virus
Virus-cell interactions - Abstract:
- Retroelements, including retroviruses and retrotransposons, are ubiquitous throughout eukaryotes and replicate by reverse-transcribing their RNA genome into DNA and stably integrating into the chromosome of the host cell. Retroviruses encode a very limited genome and the most simple retroviruses carry out all of the functions needed to replicate using only three proteins: Gag, the viral structural protein that orchestrates the assembly of nascent virus particles using the genomic RNA as a scaffold; Pol, the reverse transcriptase and integrase; and Env, the envelope glycoprotein that studs the outside of virus particles. Because retroviruses encode such a limited genome, they are dependent on host factors to mediate or facilitate many aspects of the virus life cycle including transcription of viral RNAs, assembly of nascent particles, and targeting viral proteins to the plasma membrane where budding occurs. However, the host factors involved in many aspects of retrovirus replication are poorly characterized. Our goal was to understand the intracellular trafficking pathways of retroviral Gag and RNA during the late (post-integration) phase of replication to identify cellular factors implicated in genome encapsidation, particle assembly, and plasma membrane localization of Gag. After integration, full-length retroviral RNA is transcribed by cellular machinery and exported from the nucleus. This full-length RNA serves two roles in retroviral replication, acting as an mRNA to direct the translation of the Gag, Gag-Pro, and Gag-Pro-Pol structural proteins, and as a genomic RNA (gRNA), which is encapsidated into nascet virus particles. The requirement for retroviruses to efficiently assemble virus particles suggests that a mechanism exists to separate full-length RNA to be used for translation from identical RNA molecules destined for encapsidation. However, the mechanisms that retroviruses use to separate RNA molecules destined for these two fates have remained an enigma. We hypothesized that cellular RNA-binding proteins interact with the viral RNA and/or Gag to target a population of viral RNA for encapsidation. To test this hypothesis, we chose mouse mammary tumor virus (MMTV) as a model system. MMTV assembles complete, immature capsids in the cytoplasm, so we reasoned that sites of capsid assembly might also be the location where Gag first interacts with the viral genome. Therefore, by studying cellular factors associated with these sites of assembly, we could gain insight into cellular proteins that target viral RNA for encapsidation. As part of an unbiased screen to identify cellular proteins that interact with MMTV Gag, we identified ribosomal protein, large 9 (RPL9) as a Gag-interacting partner. A single RPL9 protein is present in each large ribosomal subunit and, strikingly, overexpression of RPL9 induced the accumulation of Gag and RPL9 in nucleoli where ribosome subunit assembly occurs. Furthermore, Gag interacted with RPL9 in nucleoli and under steady state conditions we identified a population of Gag within nucleoli. Depletion of extraribosomal RPL9 using siRNA resulted in decreased MMTV virus production, suggesting that Gag interacts with RPL9 in the nucleolus to facilitate virus production. Although the mechanism by which RPL9 influences virus production is not known, because RPL9 is an RNA-binding protein one possibility is that RPL9 interacts with Gag to help Gag identify viral genomes for encapsidation. Several other cellular proteins have also been reported to facilitate encapsidation of viral RNAs and assembly of virus particles. For example, the Gag protein and genomic RNA of the yeast Ty3 retrotransposon, which follows an assembly pathway similar to MMTV, interact with a conserved set of cellular mRNA-binding proteins associated with stress granules (SGs) and processing bodies (PBs) to facilitate Ty3 replication. These observations led us to conduct parallel experiments in MMTV to test whether SG and PB proteins also influenced MMTV particle assembly. We found that specific SG- and PB-associated proteins colocalized with MMTV Gag in the cytoplasm in translationally-repressed complexes. Furthermore, a subset of these proteins interacted with MMTV Gag through an RNA-dependent mechanism and colocalized with a subviral RNA reporter. To our surprise, the viral RNA appeared to recruit SG-associated proteins and Gag into complexes, suggesting that SG and/or PB proteins may bind the viral RNA and serve as a “landmark” used by Gag to identify viral RNAs for encapsidation. Indeed, reducing the expression of specific SG or PB-associated proteins using siRNAs resulted in decreased virus production, indicating that SG- and PB-associated proteins play an essential role in efficient virus production. Together, these data led to a model where SG- and/or PB-associated proteins repress the translation of a portion of viral RNA, thereby targeting this RNA for encapsidation. After selecting a genome for encapsidation all retroviral Gag proteins must traffic to the plasma membrane to bud from the host cell as a virus particle. Gag proteins of many retroviruses recognize the plasma membrane by the presence of the plasma membrane-specific phospholipid phosphatidylinositol-(4,5)-bisphosphate [PI(4,5)P2]. The Gag protein of Rous sarcoma virus (RSV) transiently traffics through the nucleus and then, in contrast to MMTV, localizes predominantly to the plasma membrane, which serves as the site of virus particle assembly. However, previous reports indicated that RSV Gag localized to the plasma membrane through a PI(4,5)P2-independent mechanism. By contrast, we found that enzymatic depletion of PI(4,5)P2 caused the mis-localization of RSV Gag from the plasma membrane to an intracellular compartment. Strikingly, we found that mutants of RSV Gag deficient in nuclear trafficking were also less sensitive to the depletion of PI(4,5)P2, suggesting a previously unknown role of nuclear membrane-free PI(4,5)P2 in RSV replication. Together, these data shed light on novel ways in which retroviruses interact with cellular proteins to locate the viral genome, assemble virus particles, and achieve plasma membrane localization required for budding.