Sequence and binding activity of the autolysin-adhesin Ami from epidemic Listeria monocytogenes 4b

Ami is an autolytic amidase from Listeria monocytogenes that is targeted to the bacterial surface via its C-terminal cell wall anchoring (CWA) domain. We recently showed that the CWA domain from Ami of L. monocytogenes EGD (serovar 1/2a) (Ami 1/2a) mediated bacterial binding to mammalian cells. Here we studied the sequence and binding properties of Ami from CHUT 82337 (serovar 4b) (Ami 4b). The Ami 4b polypeptide is predicted to be 770 amino acids long (compared with the 917 amino acids of Ami 1/2a from EGD). Ami 1/2a and Ami 4b are almost identical in the N-terminal enzymatic domain (approximately 98% amino acid identity), but the sequence is poorly conserved in the C-terminal CWA domain, with only approximately 54% amino acid identity and eight GW modules in Ami 1/2a compared with six GW modules in Ami 4b. The purified Ami 4b CWA domain efficiently bound serovar 4b bacterial cells and only poorly bound serovar 1/2a bacterial cells. The Ami 4b CWA domain was also significantly less able to bind Hep-G2 human hepatocytic cells than the Ami 1/2a CWA domain. We sequenced the ami regions encoding CWA domains of reference strains belonging to the 12 L. monocytogenes serovars. The phylogenic tree constructed from the sequences yielded a binary division into group I (serovars 1/2a, 1/2b, 1/2c, 3a, 3b, 3c, and 7) and group II (serovars 4a, 4b, 4c, 4d, and 4e). This is the first direct evidence of divergence between serovars 1/2a and 4b in a gene involved in the adhesion of L. monocytogenes to mammalian cells, as well as the first demonstration of allelic polymorphism correlated with the somatic antigen in this species.