Analysis of TRAIL-receptor-deficient mice in a two-stage skin tumorigenesis model reveals a role for TRAIL-receptor in suppression of metastasis formation

2377

TRAIL is a member of the TNF super family which can induce apoptosis via its two death domain-containing receptors TRAIL-R1 (DR4) and TRAIL-R2 (DR5) in human cells. In mouse cells only one apoptosis-inducing receptor for TRAIL, TRAIL-R (also called mDR5), is present. A critical role for TRAIL and TRAIL-mediated apoptosis has been shown in immune surveillance against tumor development. Interestingly, TRAIL sensitivity is significantly lower in primary and non-transformed cells than in tumor cells and increases during adenoma to carcinoma transition of colorectal carcinogenesis, suggesting a role for TRAIL in later stages of carcinogenesis. To define at which step of the tumor development process the presence or absence of TRAIL-R on transforming cells may be decisive, we generated mice deficient for TRAIL-R (TRAIL-R^-/- mice) and challenged these mice in the DMBA/TPA two-stage skin tumorigenesis model. We found that TRAIL-R^-/- mice are viable, fertile and do not show any obvious phenotype through-out the entire observation period of more than 12 months. Upon single treatment with DMBA and biweekly TPA treatments all groups of mice developed benign keratinized papillomas after 8 weeks, and some of these papillomas converted to carcinomas beginning at 20 weeks. We did not find a significant difference in number (initiation) and growth (promotion) of papillomas between TRAIL-R^-/-, TRAIL-R and wild type mice. The papilloma to carcinoma conversion rate was also similar between the genotypes. However, analysis of lymph nodes of all mice with carcinomas revealed that 44% of all TRAIL-R^-/- mice with carcinomas had developed metastases in the draining lymph nodes, compared to 22% of TRAIL-R mice and 0% of wild type mice. Finally, we wanted to test whether the difference of the appearance of metastases was due to a difference of proliferative capacity of tumors in the absence or presence of TRAIL-R. Therefore we generated several tumor cell lines from either TRAIL-R^-/- or control mice. However, we did not detect a difference in the proliferative capacity of the different tumor cell lines neither in vitro nor in vivo by orthotopic transplantation into mice, indicating that the difference in formation of metastases might possibly be due to incapacity of the tumor surveillance mechanism to delete TRAIL-R deficient cells. Together, our data suggests that the apoptosis-inducing TRAIL-R is not involved in the regulation of initiation, promotion, conversion or growth of skin tumors, but may suppress the formation of metastatic lesions.