Abstract 1154: Age of type I collagen is critical for regulation of HT-1080 cell proliferation in 3D matrices

Abstract

Abstract The tumor microenvironnement consists of cells, soluble factors, signaling molecules, extracellular matrix (EM), and mechanical cues that can promote neoplastic transformation, support tumor growth and invasion, and foster therapeutic resistance. In addition to provide a physical barrier against invasion, type I collagen, the most abundant protein in EM, is also known to play a role in tumor cell proliferation (1). DeClerck et al. (2) has reported a direct inhibitory effect of fibrillar type I collagen on tumor cell proliferation when compared to the monomeric one. Type I collagen is a life-long protein (half-life 15 yrs) susceptible to undergo non enzymatic post-translational modifications such as production of Advanced Glycation End Products (AGEs). These modifications are able to influence the behavior of tumor cells. Here, we investigated whether regulation of HT-1080 cell proliferation is a consequence of type I collagen aging in terms of fibrillar state of the protein (2,3), or of the level of AGEs, also known to stimulate cell proliferation (4). Rat tail type I collagen were prepared from adult (2 months) and old (2 years) animals. A significant increase in AGEs has been observed in collagen extracted from old rats, compared to the adult. Under 3D growth condition, HT-1080 cells proliferate rapidly in old type I collagen relative to the adult one. This effect was not observed in 2D coating culture. The low rate of proliferation in adult collagen is accompanied by a downregulation of ERK1/2 activation, and an upregulation of p21, the inhibitor of cell cycle progression. This age-dependent cell proliferation regulatory effect does not involve α2β1 integrins. Neither Pi3K/Akt nor JNK pathways were involved in this process. Accumulating evidence suggest that Discoidin Domain Receptor 2 (DDR2) is a Receptor Tyrosine Kinase (RTK) with the unique ability among RTKs to respond to fibrillar collagen (5). To determine whether a similar response is elicited by DDR2 in both collagens, DDR2 was immunoprecipitated from cell lysates, and phosphorylated DDR2 was detected by immunoblotting. High level of DDR2 phosphorylation was observed after 3 days culture in adult collagen compared to the old one, whereas DDR2 expression remains unchanged in the presence of both collagens. Moreover, the DDR2 kinase function inhibitor, nilotinib, restored cell proliferation in adult collagen to a level similar to that observed in the old one. Taken together, these data establish a role for DDR2 in critical events during restriction of tumor cell proliferation induced by adult type I collagen.