Nagahiro Minato, Barry R. Bloom, Burton Janis, David Rowlands, Katherine R. Spindler, Bert L. Semler, Charlotte L. Jones, John J. Holland, Frank M. Horodyski, Elizabeth A. Grabau, and Lola M. Reid
Oligonucleotide mapping of viral RNA, peptide mapping of virus proteins, and virion RNA termini sequencing all show that long term persistent infection by Vesicular Stomatitis Virus (VSV) leads to extensive and continuous genome mutation. The Dl particles present in persistently infected cells apparently help drive this rapid mutation of infectious virus because serial high multiplicity passages of VSV in acute lytic infections generates more oligonucleotide map changes in virus RNA than does alternating high and low m.o.i. passage, or low m.o.i. passage. Furthermore, virus populations in persistently - infected cultures are selected for resistance to the Dl particles present initially in the carrier cultures. The mutant virus clones recovered after many years of persistence contain more than 2% of mutated bases at the 3′ end of virion RNA and about 15% at the 5′ end. The mutant virus recovered after years of persistence does not rapidly revert, and is as stable as the original input virus clone. A number of biologically interesting mutant phenotypes have been recovered from these persistently infected cells. These include: (I) ts smaII plaque, avirulent mutants; (2) mutants resistant to Dl particles originally present in the persistently infected cells; (3) mutants which give extremely low virus yields on initial isolation and which are able to establish persistent infection without added Dl particles; (4) mutants selected in vivo which allow persistently infected cells to form tumors and metastases in nude mice (these strongly induce interferon and natural killer cells, but escape the natural killer cell surveillance due to virus mutation); (5) mutants arising spontaneously in the carrier cells in culture which are poor inducers of interferon and of natural killer cells, and which also allow carrier cells to form tumors and metastases due to viral mutations; (6) mutants which show reduced neutralization kinetics with heterologous and homologous antisera; (7) mutants exhibiting apparent smaller size M protein, or a major new virion protein band near the G protein; (8) mutants which have lost all mouse virulence (VSV and rabies virus), and mutants which acquired mouse virulence (rabies HEP Flury vaccine strain). Many of the phenotypes must involve multiple mutations and a number of them may be due to the same mutation(s). Considerable further work will be required to map and understand these phenotypes, but they should provide considerable insight into virus functions. It is clear from this work that RNA virus genomes have an unexpected degree of mutational plasticity, and that viral persistence is a highly competitive, metastable situation which shows no signs of stabilizing as viral genomes and Dl particle genomes evolve over many years.