Open Collections

Characterization of the human gelatinase : a collagen binding domain

University of British Columbia

Matrix raetalloproteinases (MMPs) can collectively degrade most extracellular matrix components during normal tissue remodeling and have also been implicated in pathological inflammatory diseases and in tumor cell invasive growth. Exploring the MMP-ligand interactions is important in understanding the function of these enzymes. Gelatinase A, like the other MMPs, is composed of distinct functional domains. One domain unique to the gelatinases consists of three fibronectin type II-like modules. The type II modules in fibronectin provide this molecule with gelatin binding properties. Therefore, the hypothesis was formulated that the fibronectin type II-like modules also provide gelatinase A with ligand binding. The tri-modular recombinant collagen binding domain of human gelatinase A (rCBD123) expressed in E. coli bound native type I collagen as well as denatured types I, IV and V collagen, elastin, and heparin. All of the gelatinase A type I collagen binding properties were found to reside in the CBD. However, rCBD123 did not bind several substrates including native type V collagen or fibronectin. Although gelatinase A can degrade basement membrane, rCBD123 did not bind laminin, fibronectin, SPARC, or matrigel. Binding site analysis further revealed that rCBD123 can bind at least two collagen molecules simultaneously. Whereas the major binding site in native type I collagen is in the telopeptide ends, collagen denaturation exposes multiple binding sites. Lysine residues were found to be important molecular determinants for ligand interactions of the CBD. Acetylation of rCBD123 lysine residues abolished heparin binding and reduced binding to collagen. Site-specific substitution of rCBD123 lysines with alanines demonstrated that K357 in the third module is required for heparin binding. Unaltered binding to other ligands by K357A and no change from wildtype protein in secondary structural components, as assessed by circular dichroism spectral analysis, confirmed that the loss of heparin binding was not a result of structural perturbation. These results together with structure modeling of the gelatinase A CBD indicated that two or more modules are required for heparin binding. Another mutant, K263A, demonstrated reduced saturation level binding to collagen but with an unchanged Kd for the interaction pointing to the presence of more than one collagen binding sites in the domain. Studies of mechanisms for gelatinase A cell surface localization showed specific cell binding to coated rCBD123 which was inhibited by preincubation of cells with soluble rCBD123. The cellular proteins binding rCBD123 were characteristic of collagen by electrophoretic behavior and resistance to digestion by pepsin but not bacterial collagenase. In addition, cell binding to rCBD123 was abolished by treatment of the cells with collagenase, and was reduced on a collagen binding deficient mutant of rCBD123. That rCBD123 could compete progelatinase A from cultured human gingival fibroblasts and that cell binding to rCBD123 was blocked by a Bl-integrin specific antibody point to the formation of a gelatinase A/ native type I collagen/ 61-integrin cell surface attachment complex. Thus, the CBD is an essential gelatinase A ligand binding domain. CBD lysine residues are important molecular determinants of substrate specificities and module cooperativity is likely required for the.ligand interactions. Gelatinase A can be positioned via the CBD to cell surfaces where it may be stored poised for activation and proteolysis.

Thesis/Dissertation

application/pdf

2009-04-17

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/7340

Oral Biology and Medical Sciences

University of British Columbia

UBCV

DSpace