Conservation of coding potential and terminal sequences in four different isolates of Borna disease virus

Abstract

We determined the complete nucleotide sequences of two poorly characterized strains of Borna disease virus (BDV) and compared them to reference strains V and He/80. Strain H1766 was almost 98% and 95% identical to strains V and He/80, respectively, whereas strain No/98 was only about 81% identical to both reference strains. In contrast to earlier reports, we found an additional A residue at the extreme 3′-end of the single-stranded RNA genome in all four BDV strains. The exact numbers of nucleotides in the four BDV genomes could not be determined due to a micro-heterogeneity at the 5′-end. If our longest sequence is a correct copy of the viral RNA, the two ends of the BDV genome would show almost perfect complementarity. All three transcription start sites, all four termination sites, both splice donor sites and both major splice acceptor sites are highly conserved, whereas a minor alternative splice acceptor site is not. The L protein of No/98 differs at 7% of its amino acid positions from the polymerase in the other strains, with most differences mapping to the C-terminal moiety of the molecule. Re-evaluation of L protein sequences of strains V and He/80 revealed differences at several positions compared to published information, indicating that variant forms of the viral polymerase have previously been characterized. These results are important because correct structures of genome ends and of the polymerase gene are the most critical parameters for the future development of techniques that will permit the genetic manipulation of BDV.