RNA and protein expression of Paternally Expressed Gene 10 (PEG10) and associated translational frameshifting in mammalian tissues

Abstract

A genetic recoding event termed −1 frameshifting is often used by retroelements (retroviruses and retrotransposons) including HIV-1 to produce two different sized proteins from a single mRNA. This mechanism appeared to be ideal for a drug target against HIV-1 because successful viral replication is absolutely dependent on this rare translational event, and yet the mechanism was not apparently used in humans or indeed any organisms in the animal kingdom. Now, two human genes, Paternally Expressed Gene 10 (PEG10) and C-C chemokine receptor 5 (CCR5) (confirmed only in 2014) have been documented to use −1 frameshifting for their expression, challenging whether the mechanism is a suitable drug target in HIV-1. To further evaluate feasibility of such a strategy, my study focused on PEG10, identified in 2001. The PEG10 frameshift mechanism was investigated and compared with that of HIV-1. The expression of PEG10 was also analysed in mouse and rat tissue samples by qPCR, immunoblotting and mass spectrometry to determine where rodent Peg10 protein was being synthesised and whether the frameshift mechanism was being used. The PEG10 frameshift element has a slippery sequence (G GGA AAC) and a pseudoknot secondary structural element that together promote frameshifting, and these are different from the sequences and structures found in HIV-1. Immediately downstream of the slippery sequence is a UCC ‘intercodon’. Substitution of the UCC with a UGA stop codon reduced −1 frameshifting 4-fold in PEG10, suggesting a decoding mechanism was responsible. Frameshifting did not decrease further however, when the stop codon decoding factor, eRF1, was overexpressed. This was in contrast to the effect found with the HIV-1 frameshift element when the intercodon was substituted with UGA. However, the alternative decoding molecule, opal (UGA) suppressor tRNA, increased −1 frameshift efficiency, but the non-cognate amber (UAG) suppressor tRNA unexpectedly also modestly affected −1 frameshifting compared with the parental serine tRNA from which the suppressor tRNAs were derived. Mass spectrometry analysis of the frameshift peptides produced as a result of ribosome translation through the PEG10 element showed the shift from the 0 frame to the −1 frame largely, but not exclusively, occurred before the intercodon, consistent with the equivocal evidence that the 0 frame intercodon was being decoded before frameshifting. Peg10 mRNA was detected in most tissues at a low level and in a small number of tissues at a significantly increased level, for example, in endocrine organs particularly placenta and hypothalamus. By contrast, Peg10 protein was expressed in only a small number of tissues (embryo and extra-embryonic) as previously documented, and it was also detected in lactating rat hypothalamus for the first time. Although there was no definitive evidence that −1 frameshifting occurred in the hypothalamus, as only Peg10 ORF1 peptides were identified by mass spectrometry in protein extracts from this tissue, it was deduced from an immunoblot analysis and mass spectrometry that a product of identical size produced in placenta was an ORF1-ORF2 fragment. Nevertheless, frameshifting has been confirmed definitively only in placenta, amniotic membrane and embryonic tissues. The outcome of −1 frameshifting in the virus HIV-1 and the human gene PEG10 are identical regarding the translation of a second open reading frame as a fusion protein (gag-pol for HIV-1 and ORF1–2 for PEG10). Despite this, the results of this study, together with concurrent studies in the laboratory, indicated that there are differences in the frameshift mechanisms between HIV-1 and PEG10. Considering these findings, −1 frameshifting in HIV-1 appears to remain a suitable target for therapeutic intervention.