(165h) Characterization of LL37 Binding to Collagen through Collagen-Binding Domains (CBDs)

Antimicrobial peptides (AMPs) exhibit a broad spectrum of antimicrobial activity against pathogenic microorganisms and have been evaluated as alternatives to conventional antibiotics for treating wound infections. However, the major challenges in the application of AMP-based therapies are the poor understanding of their mechanism of action, narrow therapeutic ratio, and toxicity to mammalian cells. Recent work from our lab demonstrated that binding AMPs to wound dressing materials with collagen binding domains (CBDs) has potential in delivering novel, non-cytotoxic, antimicrobial, and pro-healing therapies. Our lab developed a chimeric peptide, cCBD-LL37, comprised of LL37, (a human AMP derived from cathelicidin) with a C-terminal CBD derived from collagenase (cCBD; collagenase CBD; peptide sequence: TKKTLRT) and an intervening FLAG linker domain. We found that cCBD-LL37 and unmodified LL37 peptides retained their antimicrobial activity, but were not cytotoxic to human fibroblasts, when adsorbed onto type I collagen scaffolds. The overall goal of the present study is to evaluate the mechanisms of LL37 and cCBD-LL37 binding to collagen and collagen-tethered AMP activity. To achieve this goal, we first measured the concentration-dependent (0.01-0.5 mg/ml) viscoelastic deposition of type I collagen onto hydrophilic SiO₂ substrates by measuring frequency and dissipation in quartz crystal microbalance with dissipation (QCM-D). We found that the collagen layer thickness did not increase with the collagen concentrations greater than 0.1 mg/ml. We then compared 0.1mg/ml type I collagen deposition on hydrophilic SiO₂ substrates with hydrophobic polystyrene (PS) substrates. QCM-D revealed a thick (up to 200nm) collagen layer adsorbed on PS and showed a larger mass adsorption on PS compared to SiO₂. To observe the morphology of the deposited collagen layer, we used immunohistochemistry (IHC) to detect collagen on the coated sensor crystals. We also performed atomic force microscopy (AFM) to further analyze the physical properties of deposited type I collagen on freshly cleaved mica. IHC and AFM revealed the presence of a mat of collagen fibrils. Through a combination of QCM-D, IHC, and AFM, the conditions for preparing an adsorbed layer of collagen fibrils were thus established as the basis for measuring peptide-collagen binding, as the basis for building detailed models of peptide-collagen mechanisms of interaction. To quantify the binding capacities between AMPs and collagen, we adsorbed 0.1 mg/ml type I collagen onto 96 well plates and designed an anti-LL37 enzyme-linked immunosorbent assay (ELISA) to measure LL37 and cCBD-LL37 binding to collagen as a function of peptide concentration. LL37 and cCBD-LL37 both bound to type I collagen, although the results suggested that a higher quantity of LL37 binds to collagen compared to cCBD-LL37 at the same peptide concentration. The binding signal of cCBD-LL37 increased with increasing peptide concentration and then reached a saturation binding point, while the binding signal for LL37 did not positively correlate with increasing peptide concentration. The impacts of these studies will help to develop a novel antimicrobial peptide delivery system and elucidate a general mechanism of CBD-mediated binding of AMPs.