In vivo generation of hybrids between different species of RNA phages.

Hybrids between different species or genera of the single-stranded RNA coliphages have not been found in nature. Here, it has been shown that viable hybrids between different phage species can easily be generated in the laboratory by in vivo recombination. cDNA of species I phage MS2 located on a plasmid and lacking part of its 5' untranslated leader (5' UTR) was complemented with another plasmid carrying the 5' half of the genome of fr, a species I phage, or of KU1, a species II representative with low sequence similarity. When the two plasmids were present in the same cell there was spontaneous production of hybrid phages. Interestingly, these hybrids did not arise by a double or single crossover that would replace the missing MS2 sequences with those of fr or KU1. Rather, hybrids arose by attaching the complete 5' UTR of fr or KU1 to the 5' terminus of the defective MS2 phage. Several elements of the 5' UTR then occurred twice, one from KU1 (or fr) and the other from MS2. These redundant elements are in most cases deleted upon evolution of the hybrids. As a result, the 5' UTR of KU1 (or fr) then replaced that of MS2. It was earlier shown that this 5' UTR could assume two alternating structures that facilitated transient translation of the proximal maturation gene. Apparently, this timer function of the 5' UTR was exchangeable and could function independently of the rest of the genome. When hybrids were competed against wild-type, they were quickly outgrown, probably explaining their absence from natural isolates.

Peptide display on live MS2 phage: restrictions at the RNA genome level.

The potential of the RNA phage MS2 to accommodate extra amino acids in its major coat protein has been examined. Accordingly, a pentapeptide was encoded in the genome as an N-terminal extension. In the MS2 crystal structure, this part of the coat protein forms a loop that extends from the outer surface of the icosahedral virion. At the RNA level, the insert forms a large loop at the top of an existing hairpin. This study shows that it is possible to maintain inserts in the coat protein of live phages. However, not all inserts were genetically stable. Some suffer deletions, while others underwent adaptation by base substitutions. Whether or not an insert is stable appears to be determined by the choice of the nucleic acid sequence used to encode the extra peptide. This effect was not caused by differential translation, because coat-protein synthesis was equal in wild-type and mutants. We conclude that the stability of the insert depends on the structure of the large RNA hairpin loop, as demonstrated by the fact that a single substitution can convert an unstable loop into a stable one.