Blood-Borne, but Not Endothelial CD40 Promotes Experimental Lung Metastasis.

During experimental lung metastasis, procoagulant tumor cells adhere to the microvascular endothelium and activate coagulation, which leads to thrombin generation and platelet activation. It is well established that hematogenous metastasis is supported by activated platelets. CD40 ligand (CD40L) expressed on, and subsequently released from, activated platelets can induce an inflammatory response, including the production of tissue factor, in endothelial cells and monocytes. Furthermore, CD40L has been shown to play an important role in platelet function and in vivo thrombus formation. These properties of CD40L may be mediated, at least in part, by CD40 signaling on platelets and vascular endothelial cells. We have previously shown that tumor cell-induced coagulation activation and experimental lung metastasis are markedly reduced in CD40L-deficient mice. In addition, we demonstrated significantly prolonged bleeding and PFA-100 closure times in mice deficient for CD40L or CD40. In the present study, we hypothesized that blood-borne (i.e. platelet-derived) CD40 is required for optimum hematogenous metastasis. Wild-type (WT) C57/BL6 mice and mice deficient for CD40 (CD40^−/−) were injected via the tail vein with 2×10⁵ B16 murine melanoma cells (n=13 per group), and tumor nodules were counted on lung surfaces 18 days after. The results revealed a significant (73%, P<0.001) reduction in lung metastasis in CD40^−/− mice. To assess the specific contribution of blood-borne CD40 to experimental lung metastasis, bone marrow cells (4×10⁶) from WT donor mice were injected (i.v.) into CD40^−/− recipient mice 24 hrs after lethal irradiation (1000 rad in 1 hr). The transplanted mice (n=14) were designated CD40-BC, as they expressed CD40 only on blood cells, but not on endothelial cells. Irradiated WT mice receiving WT bone marrow (n=17) served as transplant controls. CD40-BC mice were phenotyped 9 weeks after transplantation by flow cytometric analysis of CD40 on peripheral B lymphocytes using a PE-labeled rat anti-mouse CD40 monoclonal antibody. All of the transplanted CD40^−/− mice demonstrated the chimeric phenotype. There were no differences in platelet counts between CD40-BC and transplanted WT mice or between the non-transplanted groups (CD40^−/− vs. WT). However, bone marrow transplantation (BMT) was associated with decreased platelet counts within the recipient strains: 471±63/nl in CD40-BC vs. 644±150/nl in CD40^−/− mice (P=NS) and 472±12/nl in transplanted vs. 692±73/nl in non-transplanted WT mice (P<0.01). Compared to WT mice, the number of lung tumor nodules (median [IQR]) was significantly reduced in CD40^−/− mice (11 [10–14] vs. 41 [26–49], P<0.001). However, compared to CD40^−/− mice, lung metastasis was significantly increased in CD40-BC mice (49 [23–67] vs. 11 [10–14], P<0.001). No difference in lung seeding was observed between CD40-BC (median, 49), transplanted WT (48) and non-transplanted WT mice (41), indicating that, first, the protection against lung metastasis due to CD40 deficiency was completely reversed by selective expression of CD40 on peripheral blood cells and, second, BMT did not alter baseline susceptibility of C57/BL6 mice to B16 melanoma lung seeding. In summary, our results suggest an important contribution of CD40 to experimental lung metastasis. Furthermore, the data point to a selective role of CD40 expressed on peripheral blood cells (i.e. platelets) in this process.