We demonstrate a novel and efficient bioprocess for production of the cephalosporin intermediates, 7-aminocephalosporanic acid (7-ACA) or 7-amino deacetoxycephalosporanic acid (7-ADCA). The Streptomyces clavuligerus expandase gene or the Cephalosporium acremonium expandase-hydroxylase gene, with and without the acetyltransferase gene, were expressed in a penicillin production strain of Penicillium chrysogenum. Growth of these transformants in media containing adipic acid as the side chain precursor resulted in efficient production of cephalosporins having an adipyl side chain, proving that adipyl-6-APA is a substrate for either enzyme in vivo. Strains expressing expandase produced adipyl-7-ADCA, whereas strains expressing expandase-hydroxylase produced both adipyl-7-ADCA and adipyl-7-ADAC (aminodeacetylcephalosporanic acid). Strains expressing expandase-hydroxylase and acetyltransferase produced adipyl-7-ADCA, adipyl-7-ADAC and adipyl-7-ACA. The adipyl side chain of these cephalosporins was easily removed with a Pseudomonas-derived amidase to yield the cephalosporin intermediates.

SummaryColicins are a diverse family of large antibacterial protein toxins, secreted by and active against Escherichia coli and must cross their target cell's outer membrane barrier to kill. To achieve this, most colicins require an abundant porin (e.g. OmpF) plus a low‐copy‐number, high‐affinity, outer membrane protein receptor (e.g. BtuB). Recently, genetic screens have suggested that colicin N (ColN), which has no high‐affinity receptor, targets highly abundant lipopolysaccharide (LPS) instead. Here we reveal the details of this interaction and demonstrate that the ColN receptor‐binding domain (ColN‐R) binds to a specific region of LPS close to the membrane surface. Data from in vitro studies using calorimetry and both liquid‐ and solid‐state NMR reveal the interactions behind the in vivo requirement for a defined oligosaccharide region of LPS. Delipidated LPS (LPSΔLIPID) shows weaker binding; and thus full affinity requires the lipid component. The site of LPS binding means that ColN will preferably bind at the interface and thus position itself close to the surface of its translocon component, OmpF. ColN is, currently, unique among colicins in requiring LPS and, combined with previous data, this implies that the ColN translocon is distinct from those of other known colicins.

No abstract