9629440 Smith Virus Molecular Evolution: Fitness, Competition, and Recombination In a natural infection of an animal by an RNA virus, the virus consists of a 'swarm' of virus with a distribution of related, but not identical genetic material. A major cause of diversity in these viruses is that the mechanism for copying their genetic material is prone to copying errors. In the special case of retroviruses, which contain two copies of their genetic material, a high rate of recombination during the copying of the RNA genome into DNA allows it to have some advantages proposed to be important for sexually reproducing organisms: removing damage to the genetic material; increasing the rate of incorporation of new mutation into the population; and increasing variability by allowing new combinations of mutations. The goal of this proposal is to measure precisely several aspects of virus growth; to use these results to predict the ability of mutants to compete with each other in the presence or absence of a novel environment (specific chemical selection); and to examine recombination between viral genomes at these mutation sites. A member of this group, HIV-1, possesses unique qualifications as a model system to study the molecular evolution of a RNA virus: it is available as a clone for which the entire nucleotide sequence has been determined, and chemically-resistant mutants are known for which the exact mutation has been identified. This allows the mutant and its isogenic wild-type strain, with exactly the same sequence except for the mutation in question, to be compared. Likewise, competition of completely isogenic strains in the presence or absence of specific selection can be carried out. The rate of evolution of this virus in nature is approximately 100,000 to one million time higher than that estimated for genes of higher organisms. In addition, the half-life of this virus in nature is on the order of 2 days, such that large numbers of generations may be studied in a rela tively short time. This model system therefore provides a unique opportunity to study questions in population biology that heretofore have been modeled theoretically, or in which sufficient data are difficult to obtain due to long generation times.
