THE EXPRESSION OF VIRAL POLYCISTRONIC GENOMES

The inability of mammalian ribosomes to translate efficiently all but the 5 P3 oximal cistrons of a polycistronic mENA raises problems in the expression of viral genomes. With many viruses these problems are solved at the stages of transcription, processing and splicing of the mENA. This is obviously applicable to viruses 

A somewhat different type of mechanism is employed by Tobacco Mosaic Virus for the expression of the coat protein gene which is situated towards the 3 end of the virion ENA and is not expressed when intact virion RNA is translated in vitro. In infected plants, a low molecular weight ENA coding for the coat protein is produced, either by selective transcription or by selective processing of the viral genome (87). 

Besides these various devices exploited by viruses in the expression of their polycistronic genomes, there is in principle the additional possibility of splicing of the ENA to delete termination codons, or to delete 5 -proximal cistrons and thereby construct an RNA in which the 5 capped end is located near another cistron. Whether this possibility deserves serious consideration in respect of viruses which replicate in the cytoplasm depends on whether RNA splicing occurs uniquely in the nucleus, or whether the appropriate enzyme activities are also present in the cytoplasm. 

This question has not yet been satisfactorily answered, but seems likely to be resolved in the near future. 

I wish to thank Tim Hunt and High Pelham for numerous helpful discussions, Cathy Land and Liz Campbell for technical assistance in the unpublished work discussed here, which was supported by grants from the Medical Research Council and the Cancer Research Campaign. 

---