Identification of Host Factors that Mediate the Nuclear Retention of Unspliced and Partially Spliced HIV-1 Transcripts

About This Grant

Project Summary Transcription of a HIV-1 provirus generates a 9-kilobase full-length transcript, that either remains unspliced or undergoes extensive alternative splicing resulting in generation of over 100 transcripts. While the completely spliced viral mRNAs can easily access the host mRNA nuclear export pathway, the unspliced genomic RNA and the partially spliced mRNAs are retained in the nucleus by unknown mechanisms. Nuclear export of intron- containing HIV-1 RNAs is mediated by the viral Rev protein (translated from fully spliced mRNAs) which binds to the Rev-response element (RRE) on these RNA subsets and tethers them to the host CRM1 export protein. How intron containing HIV-1 transcripts are retained in the nucleus is a long-standing question. The majority of cellular pre-mRNAs are alternatively spliced, and recruitment of mRNA export factors is closely coupled with splicing. Accordingly, the first widely accepted model proposes that lack of splicing leads to a block in deposition of RNA export factors on the unspliced and partially spliced HIV-1 transcripts. On the other hand, it is noteworthy that ~5% of protein-coding genes in humans do not contain introns. Hence, splicing is not a strict pre-requisite for RNA export. The second model proposes that intron-containing HIV-1 transcripts are actively retained in the nucleus due to distinct features present within these RNAs. This model is supported by the finding that codon- optimization of unspliced and partially spliced viral mRNAs overcomes nuclear retention independent of splicing modulation. On the other hand, the minimal sequence features necessary and sufficient for nuclear retention and whether trans-acting host factors are involved remain unknown. A distinguishing feature of the HIV-1 genome is its unusually biased nucleotide composition, rich in adenosines (~36%) and poor in cytosines (~18%). Preliminary studies provided herein raise the possibility that this property may underlie the nuclear retention of unspliced and partially spliced HIV-1 transcripts. The central hypothesis of this application is that unspliced and partially spliced HIV-1 mRNAs are actively retained in the nucleus by host proteins that recognize adenosine- rich sequences on these transcripts. In preliminary studies, we found that altering the codon usage of multiple reporter genes (e.g. GFP, mCherry, firefly luciferase) to resemble HIV-1 codon usage, hence making them adenosine-rich, results in their nuclear retention and dependence of reporter gene expression on Rev/RRE. In Aim 1, we propose to identify the minimal features of these RNAs that result in nuclear retention and determine whether this property is conserved in other mammalian species. In Aim 2, we propose to leverage these minimal HIV-1-like reporter RNAs in genome-wide CRISPR and siRNA screens to identify putative host factors that mediate nuclear retention of HIV-1 unspliced and partially spliced transcripts. The proposed studies have the potential to significantly advance our understanding of a fundamental aspect of HIV-1 replication. Knowledge gained herein will also more broadly impact viral and eukaryotic gene regulation fields.