Activity of epithelial defensin HBD-3 against a periodontal pathogen

Defensins are cationic (positive-charged) peptides with broad-spectrum antibiotic activity. In humans, there are two types of defensins, alpha (alpha) and beta (beta). Human neutrophils contain four alpha-defensins known as Human Neutrophil Peptide (HNP) 1-4. Epithelial cells produce four beta-defensins known as Human Beta Defensin (HBD) 1-4. Gram-negative anaerobic bacteria that are associated with periodontal disease are resistant to human alpha-defensins, but are killed by beta-defensins.;HBD-3 is the most active beta-defensin. HBD-3 is a longer peptide than HNP 1-4. HBD-3 has additional amino acid residues with hydrophobic side chains near the N-terminus and residues with cationic side chains at the C-terminus.;Objectives. (1) Confirm that the periodontal pathogen A.a. (Aggregatibacter actinomycetemcomitans) is resistant to HNP-1 but killed by HBD-3; (2) Determine if the N-terminal or C-terminal portion of HBD-3 can account for activity against A.a. ; (3) Determine whether HBD-3 binds to lipopolysaccharide (LPS), which covers the surface of gram-negative bacteria; (4) Determine whether binding of the hydrophobic N-terminus of HBD-3 to the hydrophobic lipid A portion of LPS accounts for activity of HBD-3 against A.a.;Methods. Non-pathogenic Escherichia coli and pathogenic A.a. Y4 bacteria were incubated with recombinant HBD-3 or HNP-1 purified from human neutrophils. Bacteria were also incubated with synthetic peptides CHRG07 and CHRG01. These peptides have sequences derived from the HBD-3 N-terminus and C-terminus, respectively. The number of viable bacteria was determined by diluting, plating on solid growth medium, and counting colonies.;Bacteria were also incubated with HBD-3 and purified LPS from E. coli or A.a. to determine whether purified LPS absorbs HBD-3 and blocks killing. Similar experiments used purified lipid A or deacylated-LPS, which lacks the hydrophobic fatty acids of the lipid A portion of LPS.;Results: HBD-3 had strong bactericidal activity against A.a. under the usual assay conditions for alpha-defensins (in dilute culture medium) and the usual assay conditions for beta-defensins (in buffer without nutrients). HBD-3 at 5 muM gave 90 to 99% killing of A.a. within 2 to 4 h. In contrast, HNP-1 had no activity against A.a. regardless of assay conditions, confirming that A.a. is resistant to HNP-1 but killed by HBD-3.;Both CHRG07 and CHRG01 killed A.a., but CHRG07 was much more active. The activity of CHRG07 was equal to that of HBD-3, indicating that the mixture of hydrophobic and cationic amino acid residues at the N-terminus can account for HBD-3 activity against A.a.;Purified LPS from E. coli or A.a. blocked the activity of HBD-3 at a 1:1 ratio of LPS to HBD-3, indicating that one molecule of HBD-3 binds to each molecule of LPS. Deacylated-LPS also blocked HBD-3 at a 1:1 ratio, but purified lipid A did not block. Although HBD-3 binds to LPS, and hydrophobic residues near the N-terminus of HBD-3 appear to be important for killing of A.a., the hydrophobic lipid A portion of LPS was not the binding site for HBD-3. Binding of HBD-3 to other hydrophobic substances such as membrane proteins or phospholipids may be important to HBD-3 activity against A.a.;Conclusions. Resistance of A.a. to leukocyte alpha-defensins is probably important to the ability of A.a. to cause disease. On the other hand, the epithelial cell beta-defensins probably help to protect healthy individuals against oral disease. Small synthetic peptides such as CHRG07 that contain the portion of HBD-3 active against the periodontal pathogen A.a. may be useful to prevent or treat gingivitis and periodontitis.