Bovine viral diarrhea virus is an economically important animal pathogen that may cause a fatal gastrointestinal disease in beef and dairy herds. Infection of a fetus with this virus during the first trimester leads to the birth of animals that are persistently infected for life. Some animals remain healthy, while others develop severe mucosal disease. The lethal outcome is a consequence of RNA recombination that produces a cytopathic virus.
Pathogenicity of bovine viral diarrhea virus is associated with the synthesis of a the viral protein NS3. This protein is not produced by the noncytopathic virus that persistently infects cows for life. Absence of the protein is due to failure to cleave the precursor of NS3, called NS2-3. In cells infected with the cytopathic, disease-causing virus, NS3 is produced because the virus has acquired an extra cleavage site. This difference is illustrated in the diagram (click for a larger view).
The extra cleavage site in the viral protein is acquired when the viral RNA of the noncytopathic virus recombines with cellular RNA. This exchange of sequence probably occurs when the enzyme copying the viral RNA briefly switches to a cellular RNA, and then back to the viral RNA. The result is a copy of the viral RNA into which a cellular sequence has been inserted.
The cleavage site for NS3 can be created in several ways. One of the most frequent is the insertion of a cellular RNA sequence coding for ubiquitin (UCH in the diagram). This small protein can be cleaved by members of a family of cellular proteases (proteases are enzymes that cut proteins). The insertion of ubiquitin leads to cleavage of NS2-3 and the production of NS3. The recombinant viruses replicate faster than noncytopathic viruses and cause disease in cattle. Why pathogenicity is associated with release of the NS3 protein, which is involved in viral RNA synthesis, is not known.
The production of pathogenic pestiviruses by recombination with cellular RNA is another illustration of the many unexpected pathways of viral evolution.
Meyers, G., Tautz, N., Dubovi, E., & Thiel, H. (1991). Viral cytopathogenicity correlated with integration of ubiquitin-coding sequences. Virology, 180 (2), 602-616 DOI: 10.1016/0042-6822(91)90074-L
I thought ubiquitination is very specific. Doesn't the ubiquitinated protein have to be adenylated etc.? The structure of the viral protein would have to be highly specific so that it wouldn't be chopped to pieces by a proteosome.