Benzotript

The effects of glycine‐extended gastrin (G‐Gly) on the invasion by colon cancer cells through stromal extracellular matrix and the role of metalloproteinases (MMPs) in this invasion were investigated. We found that 10−9–10−6 M G‐Gly significantly increased the invasiveness of 2 human colon cancer cell lines, LoVo and HT‐29, both expressing the G‐Gly‐specific binding site but little gastrin/CCK‐B receptor (gastrin receptor). LoVo cells expressed MMP‐1, ‐2, ‐3 and ‐9. An amount of 10−7 M G‐Gly enhanced collagenase MMP‐1 expression. Overexpression of enhanced green fluorescent protein (EGFP)‐fused MMP‐1 in LoVo cells, by cDNA transfection, enhanced invasiveness through type I collagen gel. Immunofluorescence study revealed that G‐Gly increased the number of cytoplasmic vesicles containing MMP‐1, some vesicles being released from the cells. The MMP‐1 vesicles contained one of the ubiquitous coat proteins, Golgi‐localized, γ‐adaptin ear‐containing, ARF‐binding proteins‐2 (GGA‐2). MMP‐1 also colocalized with CD147 (EMMPRIN, an extracellular matrix metalloproteinase inducer in adjacent stromal cells). It was suggested that G‐Gly increased the number of vesicles containing MMP‐1 and that MMP‐1 interacted with CD147 to increase invasion. G‐Gly significantly enhanced the production of MMP‐3, an activator of MMP‐1 and ‐9, as well as gelatinase MMP‐9 activity. The G‐Gly‐mediated MMP‐9 increase was inhibited by treatment with anti‐MMP‐3 IgG and MMP‐3 siRNA. Furthermore, G‐Gly increased the proMMP‐2 level, although no activated MMP‐2 was found in conditioned medium in either the presence or the absence of G‐Gly. By contrast, gastrin (10−7 M) had no effect on the levels of these MMPs or the invasiveness of colon cancer cells in type I collagen gel and Matrigel. These effects of G‐Gly on the activity and expression of MMPs and the invasiveness of colon cancer cells were inhibited by treating the cells with a broad‐spectrum metalloproteinase inhibitor (CGS27023A) and nonselective gastrin/CCK receptor antagonists (proglumide and benzotript). But a gastrin/CCK‐B receptor antagonist (YM022) did not inhibit the increased invasion by G‐Gly. Together, these results demonstrate that G‐Gly renders colon cancer cells more invasive by increasing MMP‐1 and MMP‐3 expressions via the putative G‐Gly receptor and would thus be a good molecular target in a clinical setting. © 2004 Wiley‐Liss, Inc.

Gastrin17gly acts as a growth factor for the colonic mucosa. Studies of the receptor involved have generally been restricted to its binding properties, and no investigation of the structure of gastrin17gly receptors on human colorectal carcinoma cell lines has yet been reported. The aim of this study was to optimise the conditions for binding of gastrin17gly to the human colorectal carcinoma cell line DLD-1, and to investigate the structure of the receptor responsible. Binding of 125I[Met15]gastrin17gly to DLD-1 cells was measured in competition experiments with increasing concentrations of either gastrin17gly or gastrin17, or with single concentrations of gastrin receptor antagonists. The molecular weights of the gastrin17gly binding proteins were determined by gel electrophoresis and autoradiography after covalent cross-linking of 125I[Nle15]gastrin2,17gly to cells or membranes with disuccinimidyl suberate. The IC50 value for binding of gastrin17gly to DLD-1 cells was 2.1+/-0.4 microM. Binding was inhibited by the non-selective gastrin/cholecystokinin receptor antagonists proglumide and benzotript, but not by the cholecystokinin-A receptor antagonist L364,718, or the gastrin/cholecystokinin-B receptor antagonist L365,260. The molecular weight of the major gastrin binding protein on DLD-1 cells or membranes was 70,000. We conclude that the major gastrin17gly binding site on the human colorectal carcinoma cell line DLD-1 is clearly distinct from the cholecystokinin-A and gastrin/cholecystokinin-B receptors, but is similar in some respects to the gastrin/cholecystokinin-C receptor.