A probable protein export sorting signal, PEP-CTERM, was described by Haft, et al. (). It is predicted to interact with a putative transpeptidase we designate exosortase. This model describes a variant with conserved motif VPLPA, rather than VPEP. It appears to be the recognition sequences for exosortase D (TIGR04152). This variant is found prominently in two members of the Rhodobacterales, namely Jannaschia sp. CCS1 and Roseobacter denitrificans OCh 114. One interesting member protein has a full-length duplication and therefore two copies of this putative sorting domain.