Abstract
Clones from the bovine enteric coronavirus (F15) cDNA library were cloned in pBR322 and sequenced by the method of Sanger and Coulson. This led to the identification of a sequence of 1,300 bases which contained a single open reading frame of 690 bases yielding a protein having properties of the matrix protein (M). It was comprised of 230 amino acids with a molecular weight of 26,376 Da. It was hydrophobic and had a net charge of +8 at neutral pH. Analysis of its secondary structure could not establish a simple transmembrane arrangement of the amino acids. Comparison of its nucleotide sequence with that of BECV Mebus strain showed only a two-base change resulting in a 100% homology between the two amino acid sequences. Furthermore, a very conserved structure of M appeared on comparison with the Dayoff optimal alignment of MHV-A59, MHV-JHM, TGEV, IBV Beaudette and IBV 6/82M amino acid sequences. As the two strains of BECV, F15 and Mebus present some antigenic differences, this led us to reconsider the role of M in viral antigen specificity. A hypothesis is that, as it seems to possess the necessary information on its transmembrane region, it is an ideal candidate for the viral budding process.