Genome properties of beet virus Q, a new furo-like virus from sugarbeet, determined from unpurified virus.

Based solely on the information that beet virus Q (BVQ) contains tubular particles, the entire nucleotide sequence of its tripartite genome was determined from unpurified virus in ca. 40 ml crude sap from locally infected Chenopodium quinoa. A starting sequence for RNA 1 was generated using primers corresponding to highly conserved helicase domains in the respective RNAs of furo-, pomo-, peclu-, hordei- and tobraviruses, and was extended by a walking random-primed cDNA approach. The similarity of the 3' ends of furoviral RNAs allowed starting sequences for BVQ RNAs 2 and 3 to be obtained once the 3' end of RNA 1 was known. BVQ RNA 1 encodes a protein with a methyltransferase-like, a variable and a helicase-like region, and for a readthrough protein which, in addition, contains an RNA-dependent RNA polymerase region. RNA 2 carries the coat protein gene, a coat protein read-through protein gene and two additional ORFs which may have arisen by deletions from an originally larger readthrough domain. RNA 3 carries a triple gene block resembling that of several other rod-shaped viruses. The 5' UTRs of the three RNAs have the potential to form a series of hairpins with C-A and C-C mismatches resembling those found in tymoviral RNAs. The 3' ends can be folded into tRNA-like structures which are preceded by a long hairpin-like structure and an upstream pseudoknot domain. BVQ belongs to the recently proposed genus Pomovirus; it shows evolutionary relationships to furoviruses in sensu stricto, peclu-, hordei-, tobra-, tymo-, tobamo-, carla- and potexviruses.