Sequence analysis of the Ebola virus genome: organization, genetic elements, and comparison with the genome of Marburg virus

Abstract

Sequence analysis of the second through the sixth genes of the Ebola virus (EBO) genome indicates that it is organized similarly to rhabdoviruses and paramyxoviruses and is virtually the same as Marburg virus (MBG). In vitro translation experiments and predicted amino acid sequence comparisons showed that the order of the EBO genes is: 3'-NP-VP35-VP40-GP-VP30-VP24-L. The transcriptional start and stop (polyadenylation) signals are conserved and all contain the sequence 3'-UAAUU. Three base intergenic sequences are present between the NP and VP35 genes (3'-GAU) and VP40 and GP genes (3'-AGC), and a large intergenic sequence of 142 bases separates the VP30 and VP24 genes. Novel gene overlaps were found between the VP35 and VP40, the GP and VP30, and the VP24 and L genes. Overlaps are 20 or 18 bases in length and are limited to the conserved sequences determined for the transcriptional signals. Stem-and-loop structures were identified in the putative (+) leader RNA and at the 5' end of each mRNA. Hybridization studies showed that a small second mRNA is transcribed from the glycoprotein gene, and is produced by termination of transcription at an atypical polyadenylation signal located in the middle of the coding region. The predicted amino acid sequence of the glycoprotein contains an N-terminal signal peptide sequence, a hydrophobic anchor sequence, and 17 potential N-linked glycosylation sites. Alignment of predicted amino acid sequences showed that the structural proteins of EBO and MBG contain large regions of homology despite the absence of serologic cross-reactivity.