The increasing prevalence of antibiotic-resistant strains of pathogenic bacteria is a major healthcare problem. Antibacterial lysins are enzymes that cleave the peptidoglycan of the bacterial cell wall. These proteins hold potential as a supplement or an alternative to traditional antibiotics since they are active against antibiotic resistant strains. However, antibacterial lysins are rapidly eliminated from the systemic circulation, which limits their application. Dimerization of an anti-pneumococcal lysin Cpl-1 has been demonstrated to decrease the clearance rate of this protein in mice. In the present work, we constructed a dimer of an anti-staphylococcal lysin lysostaphin by fusing it with an anti-parallel α-helical dimerization domain. Lysostaphin dimer had a more favorable pharmacokinetic profile with increased terminal half-life and area under the curve (AUC) values compared to monomeric lysostaphin. However, the staphylolytic activity of dimerized lysostaphin was decreased. This decrease in activity was likely caused by the dimerization; since the catalytic efficacy of lysostaphin dimer towards pentaglycine peptide was unaltered. Our results demonstrate that, although dimerization is indeed beneficial for the pharmacokinetics of antibacterial lysins, this approach might not be suitable for all lysins, as it can negatively affect the lysin activity.

Antibacterial lysins are promising proteins that are active against both antibiotic-susceptible and antibiotic-resistant bacterial strains. However, a major limitation of antibacterial lysins is their fast elimination from systemic circulation. PEGylation increases the plasma half-life of lysins but renders them inactive. Here we report the construction of a fusion protein of lysostaphin, a potent anti-staphylococcal lysin, and an albumin-binding domain from streptococcal protein G. The resulting fusion protein was less active than the parent enzyme lysostaphin, but it still retained significant antibacterial activity even when bound to serum albumin. The terminal half-life of the fusion protein in rats was five-fold greater than that of lysostaphin (7.4 vs. 1.5 h), and the area under the curve increased more than 115 times. Most importantly, this increase in systemic circulation time compensated for the decrease in activity. The plasma from rats that received an injection of the fusion protein retained bactericidal activity for up to 7 h, while plasma from rats that received plain lysostaphin lacked any detectable activity after 4 h. To the best of our knowledge, this is the first report of an antibacterial lysin with both improved pharmacokinetic parameters and prolonged bactericidal activity in the systemic circulation.

Prokaryotic restriction-modification (R-M) systems defend the host cell from the invasion of a foreign DNA. They comprise two enzymatic activities: specific DNA cleavage activity and DNA methylation activity preventing cleavage. Typically, these activities are provided by two separate enzymes: a DNA methyltransferase (MTase) and a restriction endonuclease (RE). In the absence of a corresponding MTase, an RE of Type II R-M system is highly toxic for the cell. Genes of the R-M system are linked in the genome in the vast majority of annotated cases. There are only a few reported cases in which the genes of MTase and RE from one R-M system are not linked. Nevertheless, a few hundreds solitary RE genes are present in the Restriction Enzyme Database (http://rebase.neb.com) annotations. Using the comparative genomic approach, we analysed 272 solitary RE genes. For 57 solitary RE genes we predicted corresponding MTase genes located distantly in a genome. Of the 272 solitary RE genes, 99 are likely to be fragments of RE genes. Various explanations for the existence of the remaining 116 solitary RE genes are also discussed.