The 2A region of the aphthovirus foot-and-mouth disease virus (FMDV) polyprotein is only 18 aa long. A ' primary ' intramolecular polyprotein processing event mediated by 2A occurs at its own C terminus. FMDV 2A activity was studied in artificial polyproteins in which sequences encoding reporter proteins flanked the 2A sequence such that a single, long, open reading frame was created. The self-processing properties of these artificial polyproteins were investigated and the co-translational ' cleavage ' products quantified. The processing products from our artificial polyprotein systems showed a molar excess of ' cleavage ' product N-terminal of 2A over the product C-terminal of 2A. A series of experiments was performed to characterize our in vitro translation systems. These experiments eliminated the translational or transcriptional properties of the in vitro systems as an explanation for this imbalance. In addition, the processing products derived from a control construct encoding the P1P2 region of the human rhinovirus polyprotein, known to be proteolytically processed, were quantified and found to be equimolar. Translation of a construct encoding green fluorescent protein (GFP), FMDV 2A and β-glucuronidase, also in a single open reading frame, in the presence of puromycin, showed this antibiotic to be preferentially incorporated into the [GFP2A] translation product. We conclude that the discrete translation products from our artificial polyproteins are not produced by proteolysis. We propose that the FMDV 2A sequence, rather than representing a proteolytic element, modifies the activity of the ribosome to promote hydrolysis of the peptidyl(2A)-tRNA Gly ester linkage, thereby releasing the polypeptide from the translational complex, in a manner that allows the synthesis of a discrete downstream translation product to proceed. This process produces a ribosomal ' skip ' from one codon to the next without the formation of a peptide bond.

No abstract

In addition, we have demonstrated that the cardiovirus Theiler's murine encephalomyelitis virus (TME) 2A protein, when linked to GUS in a single ORF, is able to cleave at its own C terminus with high

No abstract

The herpes simplex virus type 1 gene UL47 encodes the tegument proteins referred to collectively as VP13/14, which are believed to be differentially modified forms of the same protein. Here we show that the major product of the UL47 gene during transient expression is VP14, suggesting that some feature of virus infection is required to produce VP13. We have tagged VP13/14 with green fluorescent protein and have demonstrated that the protein is targeted efficiently to the nucleus, where it often localizes in numerous punctate domains. Furthermore, we show that removal of the N-terminal 127 residues of the protein abrogates nuclear accumulation, and we have identified a 14-amino-acid peptide from this region that is sufficient to function as a nuclear targeting signal and transport a heterologous protein to the nucleus. This short peptide contains two runs of four arginine residues, suggesting that the VP13/14 nuclear localization signal may behave in a manner similar to that of the arginine-rich nuclear localization signals of the retrovirus transactivator proteins Tat, Rev, and Rex. In addition, by using heterokaryon assays, we show that VP13/14 is capable of shuttling between the nucleus and cytoplasm of the cell, a property that may be attributed to three leucine-rich stretches in the C-terminal half of the protein that again bear similarity to the nuclear export signals of Rev and Rex. This is the first demonstration of a tegument protein that is specifically targeted to the nucleus, a feature which may be relevant both during virus entry, when VP13/14 enters the cell as a component of the tegument, and at later times, when large amounts of newly synthesized VP13/14 are present within the cell.