coli DH5α and P. aeruginosa ATCC 14207, but not S. Typhimurium ATCC 23564. Both CclA and AS-48 target the cytoplasmic membrane, but differ slightly in their mode of action. AS-48 forms nonselective pores (Gálvez et al., 1991), whereas CclA generates anion-selective pores (Gong et al., 2009). It is not clear whether the differences between AS-48 and CclA toward Salmonella arise from differences in the mode of action or from differences in the strains tested. To lend a broader context to our findings with the UAL307 bacteriocins,

we also examined the activity of gallidermin and SubA. Our results show that when tested in combination with EDTA, gallidermin has comparable activity to nisin against Gram-negative bacteria. Because the receptor molecule for nisin and gallidermin (lipid II) is highly conserved across the prokaryotes, once these lantibiotics are able to Pexidartinib mouse access the cytoplasmic membrane, they are more likely to display a killing effect compared with CbnBM1 or PisA, which require a specific EIItman permease receptor for binding. Indeed, upon cotreatment with EDTA, both lantibiotics were more active than either CbnBM1 or PisA against the strains of E. coli and Salmonella that were tested. Conversely, although it had little selleck screening library effect against S.

Typhimurium ATCC 23564, CclA showed activity against E. coli DH5α and P. aeruginosa ATCC 14207 comparable to that of the lantibiotics. Our other point of comparison, SubA, is a non-LAB circular bacteriocin with unusual thioether cross-links. Reports indicate that SubA is able to directly inhibit the growth of some Gram-negative bacteria, including certain strains of E. coli and

Pseudomonas, and is able to inhibit additional Gram-negative strains when subjected to heat stress (Shelburne et al., 2007). In contrast, we found that SubA combined with EDTA did not inhibit Gram-negative bacteria significantly, leading us to speculate whether EDTA was interfering with the activity of SubA. In support of this hypothesis, we found that when EDTA was used in combination with SubA, its activity toward a sensitive Gram-positive organism was reduced. It has been reported that many anionic antimicrobial peptides exert maximal activity when complexed with also cationic species (Brogden, 2005). SubA is an anionic bacteriocin, and because EDTA chelates Mg2+ and Ca2+ ions, it may be that the experimental conditions ‘inactivated’ SubA. If an alternate OM destabilizing strategy was used, it is likely that a greater killing effect from SubA would be observed. However, SubA may also require a membrane-bound receptor: SubA can interact directly with lipid bilayers, causing pore formation, albeit at concentrations higher than those required for antimicrobial activity (Thennarasu et al., 2005).