Submitted to: Annual Review of Phytopathology
Publication Type: Review Article
Publication Acceptance Date: June 15, 2003
Publication Date: November 1, 2003
Citation: FRENCH, R.C., STENGER, D.C. EVOLUTION OF WHEAT STREAK MOSAIC VIRUS: DYNAMICS OF POPULATION GROWTH WITHIN PLANTS MAY EXPLAIN LIMITED VARIATION. Annual Review Of Phytopathology. 2003. Ann. Rev. Phytopathology 41:199-214. Interpretive Summary: Literature describing genetic variation in Wheat streak mosaic virus (WSMV) is reviewed. A hypothesis describing fundamental differences between population growth of plant RNA viruses versus animal or bacterial RNA viruses is presented to explain limited genetic variation in WSMV. The hypothesis states that plant virus population growth is mainly linear rather than exponential, and that systemic movement of plant viruses is subject to severe bottlenecks. As a consequence, the proportion of viral genomes that actually produce progeny and contribute to the continuation of a viral lineage is miniscule compared to the overall size of the viral population. This small effective population size is conducive to genetic drift, a random process that appears to have contributed significantly to the evolution of WSMV.
Technical Abstract: Genotypic variation of plant viruses is less than expected based on the error rate of viral RNA polymerases and the extremely large number of virions produced within a single infected plant. Generally, this has been taken as evidence of extreme negative selection, although other explanations are consistent with the data. Previous studies have defined genotypic variation among and within Wheat streak mosaic virus (WSMV), and suggest that in addition to negative selection, stochastic processes such as genetic drift are significant factors in the evolution of WSMV. A hypothesis is presented which may explain limited variation seen within plant virus populations. This hypothesis states that plant virus replication within a cell is primarily linear, rather than exponential, such that the effective population size is much smaller than the actual population size. A statistical argument is presented in which linear population growth results in a much lower accumulation of mutants in the population (compared to exponential growth) such that the extremely large population sizes of plant viruses may be realized without catastrophic mutational meltdown. Whereas lytic viruses of animals and bacteria release essentially all progeny virions produced within a cell upon lysis, the vast majority of plant virus progeny produced within a cell remain in that same cell and never produce progeny of their own. Therefore, the `extracellular¿(systemic movement) exponential phase of plant virus population growth is much less than that of a lytic virus. A statistical argument is presented in which the pattern of strain distribution within leaves and tillers of wheat plants coinfected with two strains of WSMV may be explained by cross-protection at the cellular level. This analysis indicates that the number of genomes that successfully become established within a leaf or a tiller is a surprising small number (as low as 3 or 4, but very likely no higher than 12). Therefore, systemic movement within a plant represents a severe bottleneck. Because the movement bottleneck is not blind to selection, those genotypes that become successfully established following systemic movement are chosen, at random, from a pool of more or less equally fit genotypes. Thus, neutral mutations would be expected to dominate polymorphism in the virus population. An analysis of variation among ~50 WSMV isolates from temperate North America revealed that silent mutations constitute the bulk of polymorphism observed within the population, suggesting that neutral mutations do, in fact, preferentially accumulate.