Selenium-Dependent Read Through of the Conserved 3’-Terminal UGA Stop Codon of HIV-1 nef

Abstract

Objectives: The HIV-1 nef gene terminates in a 3’-UGA stop codon, which is highly conserved in the main group of HIV-1 subtypes, along with a downstream potential coding region that could extend the nef protein by 33 amino acids, if readthrough of the stop codon occurs. It has been proposed that antisense tethering interactions (ATIs) between a viral mRNA and a host selenoprotein mRNA are a potential viral strategy for the capture of a host selenocysteine insertion sequence (SECIS) element. This mRNA hijacking mechanism could enable the expression of virally encoded selenoprotein modules, through translation of in-frame UGA stop codons as selenocysteine (Sec). Here, our aim was to assess whether readthrough of the 3’-terminal UGA codon of nef occurs during translation of HIV-1 nef expression constructs in transfected cells, and whether selenium-based mechanisms might be involved. Material and Methods: To assess UGA codon readthrough, we used fluorescence microscopy image analysis and flow cytometry of HEK 293 cells transfected with full length HIV-1 nef gene expression constructs including the 3’-UGA stop codon and a predicted thioredoxin reductase 1 (TXNRD1) antisense region spanning the UGA codon, engineered with a downstream in-frame green fluorescent protein (GFP) reporter gene. These were designed so that GFP can only be expressed by translational recoding of the UGA codon, that is, if the UGA codon is translated as an amino acid or bypassed by ribosomal hopping. To assess readthrough efficiency, appropriate mutant control constructs were used for 100% and 0% readthrough. We used anti-TXNRD1 siRNA to assess the possible role of the proposed antisense interaction in this event, by knockdown of TXNRD1 mRNA levels. Results: UGA stop codon readthrough efficiency for the wild-type nef construct was estimated by flow cytometry to be about 19% (P < 0.0001). siRNA knockdown of TXNRD1 mRNA resulted in a 67% decrease in GFP expression in this system relative to control cells (P < 0.0001), presumably due to reduced availability of the components involved in selenocysteine incorporation for the stop codon readthrough (i.e. the TXNRD1 SECIS element). Addition of 20 nM sodium selenite to the media enhanced stop codon readthrough in the pNefATI1 plasmid construct by >100% (P < 0.0001), that is, more than doubled the amount of readthrough product, supporting the hypothesis that selenium is involved in the UGA readthrough mechanism. Conclusion: Our results show that readthrough of the 3’-terminal UGA codon of nef occurs during translation of HIV-1 nef expression constructs in transfected cells, that this is dependent on selenium concentration, and the presence of TXNRD1 mRNA, supporting the proposed antisense tethering interaction.

SARS-CoV-2 Main Protease Targets Host Selenoproteins and Glutathione Biosynthesis for Knockdown via Proteolysis, Potentially Disrupting the Thioredoxin and Glutaredoxin Redox Cycles

“…Any benefit for the virus of SelenoF knockdown is likely to be related to its roles as an ER-resident protein (and thus a potential disruptor of viral protein assembly), so there are clearly other viral strategies at play in addition to inhibition of DNA synthesis [ 26 ]. Taken together, these previous findings support the possibility that, like a number of other RNA viruses including HIV-1, Ebolavirus, and Zika virus [ 67 , 68 , 69 ], SARS-CoV-2 may be targeting certain selenoprotein mRNAs by antisense [ 26 , 33 ].…”

“…Significantly, based on a combination of computational and experimental lines of evidence, our group has demonstrated the possibility of virus-host RNA-RNA antisense interactions targeting host selenoprotein mRNAs, particularly isoforms of thioredoxin reductase, as a possible factor in previous observations linking Se status to viral pathogenicity [ 67 , 68 , 69 ].…”

The relevance of selenium to viral disease with special reference to SARS-CoV-2 and COVID-19

“…An example of this type of potential antisense interaction is shown as RNA secondary structures for SARS-CoV-2 v. human TNXRD3 in Taylor et al (68) . The possibility of antisense interactions between viral RNAs and host selenoprotein mRNA was first proposed for HIV-1 and Ebola virus based on computational analysis (68) and has been supported experimentally for HIV-1 and Zika virus (69,70) .…”

Selenium, Stroke, and Infection: A Threefold Relationship; Where Do We Stand and Where Do We Go?

“…The procedures via which this is achieved are unique host–virus interactions such as antisense tethering (ATI) [ 97 ]. ATIs between viral and selenoproteome mRNAs reflect a gain-of-function procedure during the viral lifecycle, using the captured host selenocysteine insertion sequence (SECIS) element to bypass a UGA stop codon and read through it as selenocysteine [ 97 , 98 ]. Virus–host ATIs utilize an established readthrough mechanism that governs the incorporation of selenocysteine in selenoproteins albeit with contextual UGA to selenocysteine efficiency [ 99 ], potentially intended as a limiting step in their translational rhythms [ 100 ].…”