Beet necrotic yellow vein virus (BNYVV) is an economically important pathogen of sugar beet and has been found worldwide, probably as the result of recent worldwide spread. The BNYVV genome consists of four or five RNA components. Here, we report analysis of sequence variation in the RNA3-p25, RNA4-p31, RNA2-CP, and RNA5-p26 genes of 73 worldwide isolates. The RNA3-p25 gene encodes virulence and avirulence factors. These four sets of gene sequences each fell into two to four groups, of which the three groups of p25 formed eight subgroups with different geographical distributions. Each of these subgroup isolates (strains) could have arisen from four original BNYVV population and their mixed infections. The genetic diversity for BNYVV was relatively small. Selection pressure varied greatly depending on the BNYVV gene and geographical location. Isolates of the Italy strain, in which p25 was subject to the strongest positive selection, were able to overcome the Rz1-host resistance gene to differing degrees, whereas other geographically limited strains could not. Resistance-breaking variants were generated by p25 amino acid changes at positions 67 and 68. Our studies suggest that BNYVV originally evolved in East Asia and has recently become a pathogen of cultivated sugar beet followed by the emergence of new resistance-breaking variants.

Ten Japanese field isolates of beet necrotic yellow vein virus (BNYVV) were transmitted to Tetragonia expansa by inoculation with sap from rootlets of sugar-beet seedlings, to which the virus had been transmitted by the fungus Polymyxa betae. RNA extracted from BNYVV particles obtained from the T. expansa leaves was analysed by agarose gel electrophoresis. Some isolates contained RNA-1 (7.1 kb), RNA-2 (4.8 kb), RNA-3 (1.85 kb) and RNA-4 (1.5 kb) and the others contained, in addition, RNA-5 (1.4 kb). Further isolates, derived from single lesions produced by these isolates, had a variety of RNA compositions. Some contained only RNA-I and RNA-2. Others contained, in addition, RNA-3, RNA-4, RNA-5 or RNA-6 (1.0 kb), or combinations of two or three of these components. Such isolates generally maintained their RNA composition on further subculture, and their particles had length distributions corresponding to their RNA components. Isolates containing RNA-I + 2 + 3 caused yellow or strongly chlorotic local lesions in T. expansa, Beta vulgaris, B. macrocarpa and Chenopodium quinoa, and caused systemic stunting and yellow mosaic in B. macrocarpa and, occasionally, in B. vulgaris. In contrast, isolates containing RNA-1 + 2 + 4 or 1 + 2 + 5 induced chlorotic lesions, those containing RNA-1 + 2 + 6 or 1 + 2 induced faint chlorotic lesions, and none of these isolates easily infected B. macrocarpa systemically. Isolates containing different combinations of RNA-3,-4 and -5 induced more severe symptoms than those containing a single RNA. Such synergistic effects occurred between RNA-3 and RNA-4 or RNA-5, or between RNA-4 and RNA-5 or RNA-6, but not between RNA-3 and RNA-6, or between RNA-5 and RNA-6. These small RNA species therefore contain the genetic determinant(s) for lesion type and for ability to infect B. vulgaris and B. macrocarpa systemically. RNA-1 and RNA-2 are viral genome components. The other RNA components have some characteristics of viral satellite nucleic acids but they may not all be dispensable if the BNYVV isolates are to survive in nature.

In plants, RNA silencing is part of a defense mechanism against virus infection but there is little information as to whether RNA silencing-mediated resistance functions similarly in roots and leaves. We have obtained transgenic Nicotiana benthamiana plants encoding the coat protein readthrough domain open reading frame (54 kDa) of Beet necrotic yellow vein virus (BNYVV), which either showed a highly resistant or a recovery phenotype following foliar rub-inoculation with BNYVV. These phenotypes were associated with an RNA silencing mechanism. Roots of the resistant plants that were immune to foliar rub-inoculation with BNYVV could be infected by viruliferous zoospores of the vector fungus Polymyxa betae, although virus multiplication was greatly limited. In addition, virus titer was reduced in symptomless leaves of the plants showing the recovery phenotype, but it was high in roots of the same plants. Compared with leaves of silenced plants, higher levels of transgene mRNAs and lower levels of transgene-derived small interfering RNAs (siRNAs) accumulated in roots. Similarly, in nontransgenic plants inoculated with BNYVV, accumulation level of viral RNA-derived siRNAs in roots was lower than in leaves. These results indicate that the RNA silencing-mediated resistance to BNYVV is less effective in roots than in leaves.

About half of Japanese isolates of beet necrotic yellow vein virus (BNYVV) were found to contain RNA 5 molecules, which were also detected in virus isolates from China and France. Sequence comparisons of RNA 5 (nucleotides 327 to 1171) in 25 isolates showed that there are up to 8% sequence differences, and that RNA 5 variants fall into three groups: group I contains most of the Japanese and Chinese isolates, group II two Japanese isolates, and group III four French isolates. The group I isolates fall into three small clusters. In the 26 kDa coding region of RNA 5, there was a maximum of 1.5% nucleotide sequence differences (6 amino acid changes) within the group and 8.4% nucleotide sequence differences (17 amino acid changes) between the groups. Comparisons of the coat protein gene of RNA 2 revealed that most of the Japanese and Chinese isolates belonged to the A type strain, but some isolates were of the B type. The French isolates (P type) were closely related to those of the A type. Mixed infections of the two types of virus and the two groups of RNA 5 were detected in a small area of Hokkaido. BNYVV might have been introduced into Japan and China by a similar route from at least two origins. These results, together with other evidence, suggest that the three groups of RNA 5 variants separated from an original population a long time ago and, thereafter, the group I population diverged further into three clusters, which may have been associated with the A type strain rather than the B type.