The Expression Of LSECtin In Melanoma Cells Inhibits Antitumor T Cell Immune Responses And Promotes The Growth Of Mouse Melanoma

Although melanoma is substantially more immunogenic than other tumors, itusually remains refractory to immunologic manipulation, despite the fact that largenumbers of tumor-infiltrating lymphocytes (TIL) are often found at melanoma sites.Most melanoma patients, even those with advanced-stage disease, have circulatingmelanoma antigen-specific CD8+T cells. However, these T cells are likely ineffectiveat inducing disease regression or preventing progression. Much research is going intothe efforts to elucidate the underlying mechanisms by which the tumor cells exploit todestroy the effector function of tumor-specific CD8+T cells. Despite tumor-specificCTLs and helper T cells generated, many of these effector cells are finally “turnedoff” at the tumor site by a number of immunosuppressive mechanisms, includingtumor-induced antigen presentation impairment, down-regulation of HLA molecules,the elaboration of immunosuppressive factors, as well as the influence of regulatorycell populations that may contribute to this immunosuppressive network, includingregulatory T cells (Treg) and myeloid-derived suppressor cells (MDSC). Recently,another immune resistance mechanism has gained much attention, specifically, thesuppressive actions of co-inhibitory molecules.Liver Sinusoidal Endothelial Cell Lectin (LSECtin), which belongs to the C-typelectin receptor superfamily, is a type II transmembrane protein that is highlyexpressed in the liver and lymph node. Our previous studies revealed that LSECtin,when expressed in the liver, acted as a co-inhibitory molecule and limited the abilityof T-cell immunity to promote HBV tolerance. Immunologically, tumors are quitesimilar to chronic viral infections. Because the potential role of LSECtin in antitumorimmunity remains unknown, we hypothesized that tumor cells might hamperantitumor T cell immune responses through LSECtin and that this mechanism mighthelp to shift the balance toward an immunosuppressive environment at the tumor site.First, we found that melanoma had a higher expression of LSECtin and almost30%of melanoma patients had increased expression of LSECtin compared withnormal people by examining the expression of LSECtin from a RNA cancer surveychip which consists of7types of cancer. Further, in a melanoma specific array, wefound almost half of the melanoma patients had increased expression of LSECtincompared with normal people. Next, immunohistochemistry (IHC) analysis of themelanoma tissue chip and flow cytometry analysis of melanoma patient’s fresh tumor cells had demonstrated the expression of LSECtin in melanoma cells at protein level.In all, we found that LSECtin was expressed by melanoma cells.Previous studies showed that the B16murine melanoma cell line on the C57BL/6mouse strain background is a well-studied model of human melanoma. We nextdetermined whether the B16cells also expressed LSECtin. We found that theexpression of LSECtin in B16cells were feeble at both mRNA and protein level,while tumor cells extracted from B16-inoculated mice had a significantly higherinduction of LSECtin expression at both mRNAand protein level compared with B16cells. This suggested that LSECtin expression in B16cells was not constitutive andcould possibly be induced in vivo. To test this, a tumor extract solution from freshisolated melanoma was obtained and used to stimulate B16cells. We found that thetumor extract solution could upregulate the expression of LSECtin in B16cellsmediated by intratumoral cytokines IL-6and IL-10.Because LSECtin was not constitutively expressed by B16cells, we obtained astable transfectants B16-LSECtin to study the role of tumor-derived LSECtin. First,all stable transfectants exhibited similar growth rates, indicating that B16-LSECtincells had no growth advantage over B16-mock cells in vitro. Remarkably,B16-LSECtin cell administration resulted in a more rapid melanoma growth in bothwild-type and LSECtin KO mice. This indicated that tumor-derived LSECtin couldpromote B16melanoma progression in mice. Furthermore, we explored the impact ofLSECtin blockade on B16melanoma progression by injecting C57BL/6mice withLSECtin antibody to investigate whether tumor cells could overwhelm antitumor cellresponses in vivo through LSECtin-dependent mechanisms. Wild type and LSECtinKO C57BL/6mice were inoculated with B16cells and treated with anti-LSECtin orcontrol antibody. We found that anti-LSECtin significantly prevented B16tumorprogression. Collectively, our results demonstrated that tumor-derived LSECtinactually promotes B16tumor growth in vivo.Because LSECtin was previously shown to dampen T cell-mediated antiviralimmune responses, we first hypothesized that tumor-derived LSECtin could promoteB16melanoma growth by inhibiting T cell-mediated antitumor immunity. Todetermine which mechanisms are involved in LSECtin-mediated tumor promotion,tumor tissue histology was assessed. Ki67was used to detect tumor cell proliferation.There was no difference in cell proliferation between the two groups, indicating that,in this model, tumor cell growth arrest was not a main component of tumor control. To determine whether apoptosis could contribute to tumor promotion in our model, aterminal-deoxynucleoitidyl transferase mediated nick end labeling (TUNEL) assaywas performed. Only marginal apoptosis and no difference were observed betweenthe two groups. Further, we depleted CD4+and CD8+T cells in mice prior totreatment. As expected, the LSECtin-mediated melanoma promotion was abrogated inresponse to the depletion of both CD4+and CD8+T cells. Taken together, these dataimply that LSECtin-mediated melanoma promotion was mainly due to weakenedantitumor T cell immunity.To assess the impact of LSECtin on antitumor T cell immune responses, whetherthe number or/and effector function of T cells within tumor are regulated by LSECtinneeds to be addressed. We first isolated tumors and assayed them for T cell infiltrationusing flow cytometry. The absolute number of CD3+and CD8+T cells/gram tumordecreased significantly by approximately three-fold in B16-LSECtin mice, comparedwith B16-mock mice. Further, we found that LSECtin expressed by B16cells couldinhibit the proliferation of B16-specific CD8+T cells through coculture assays. Next,wild type C57BL/6mice were inoculated with B16-LSECtin or B16-mock cells, andtumor-draining lymph nodes (TDLN) or splenocytes were isolated to determine todetermine the effector function by intracellular cytokine staining (ICS) and ELISPOT.We found that LSECtin-expressing B16cells could inhibit the effector functions ofB16-specific T cells, both in the TDLN and the periphery. In the other hand, we haddemonstrated that LSECtin expressed by B16cells could inhibit the secretion ofIFN-γ from CD8+T cells in vitro. Collectively, LSECtin expressed by B16cells notonly reduces the number of T cells within tumor, but also impairs the effector functionof T cells to promote melanoma progression.Together, our studies had revealed that: as a coinhibitory molecule, LSECtin wasexpressed by B16cells to dampen anti-tumor T cell immune responses for melanomaprogression. Importantly, we indicated that LSECtin interacted with LAG-3to inhibitthe secretion of IFN-γ of T cells. LSECtin expressed on melanoma cells appears to bea novel possible mechanism of immune escape of melanoma cells and provides afoundation for potential combinatorial immunotherapy strategies.