1Department of Plant Pathology & Physiology, Clemson University, Clemson, South Carolina 29634-0377, USA; 2Department of Biological Sciences, University of South Carolina, Columbia, South Carolina 29208, USA

Background and objectives
The genus Ilarvirus, despite containing by far the greatest number of individual viruses and causing economic problems in many crops, remains the least extensively studied of the four viral genera with tripartite genomes that comprise the family Bromoviridae [1]. Members of the genera Alfamovirus, Bromovirus and Cucumovirus have all been the subject of detailed studies at the molecular level. However, until 1993 only a single complete sequence for the RNA 3 of tobacco streak virus (the type member of the ilarviruses) had been reported [2]. Alfalfa mosaic virus (AMV, the sole alfamovirus) shares properties in common with members of the ilarviruses. Both AMV and the ilarviruses require the presence of coat protein (CP) to activate the infection process. Moreover, the CP of one ilarvirus will activate the genome of another ilarvirus, and the CP of AMV has been shown to activate the genomes of ilarviruses and vice versa [3]. As a result, there has long been debate as to whether AMV should be a separate entity or should be included with the ilarviruses. Over the past 5 years this laboratory has sequenced a number of ilarviruses in an attempt to determine the relationship between AMV and the ilarviruses. Our initial work with viruses in subgroups 1 and 2 of the genus Ilarvirus indicated that the ilarviruses and AMV could be considered as distinct, as the ilarviruses possessed a 2b ORF whereas AMV did not [4]. The work reported here indicates that some ilarviruses lack a 2b ORF and, as a result, AMV can be included in the genus Ilarvirus.

Materials and methods
Prune dwarf virus (PDV) and Humulus japonicus virus (HJV) were grown, purified, cloned and sequenced as described previously [5-7]. Sequences were assembled using GeneJockey 2 software. Phylogenetic analyses were completed using CLUSTAL W [8]. Dendrograms were drawn using NJPLOT (Manolo Gouy, University of Lyon, France).

Results and conclusions
The RNA 2 of PDV and HJV are both of a size similar to that reported for other ilarviruses. However, there was no indication of the presence of a 2b ORF, similar to that found in cucumoviruses and some ilarviruses, in the sequence of either virus. Total RNA preparations of ilarviruses in subgroups 1 and 2 displayed a band of RNA (4a) smaller than subgenomic RNA 4, but this band was not detected in other ilarviruses. Phylogenetic analysis of the polymerase signatures [9] of sequenced ilarviruses indicated that AMV could be included in a cluster with HJV and PDV, and that AMV was more closely related to the remaining ilarviruses than was either HJV or PDV. Thus it is proposed that AMV be considered a member of the genus Ilarvirus rather than existing in a separate genus.

1. Murphy FA, Fauquet CM, Bishop DHL et al., 1995. Archives of Virology, Supplement 10, 450-457.
2. Cornelissen BJC, Janssen HK, Zuidema D, Bol JF, 1984. Nucleic Acids Research 12, 2427-2437.
3. van Vloten-Doting L, 1975. Virology 65, 215-225.
4.Ge X, Scott SW, Zimmerman MT, 1997. Archives of Virology 142, 1213-1226.
5. Bachman EJ, Scott SW, Ge X, Vance VB, 1994. Virology 201, 127-131.
6. Ge X, Scott SW, 1994. Journal of General Virology 75, 2841-2846.
7. Adams AN, Clark MF, Barbara DJ, 1989. Annals of Applied Biology 114, 497-508.
8. Thompson JD, Higgins DG, Gibson TJ, 1994. Nucleic Acids Research 22, 4673-4680.
9. Candresse T, Morch MD, Dunez J, 1990. Research in Virology 141, 315-329.