Transmembrane TNF-α—a Tumor Marker For Leukemia And A Target For Antibody Therapy Study Of Unknown Ligand(s) For Tim3and Their Function

Herein we reported a case of follicular lymphoma with50.26%clonal maligment lymphocytes and50%IGH/BCL-2positive tumor cells. To determine whether endothelial cells (ECs) within the tumor share the malignant feature, we isolated and purified endothelial cells from the tumor with immunomagnetic beads using a monoclonal antibody against CD31, a surface marker of endothelial cell. Thereafter, we identified ECs according to their morphology and found that ECs were consistently flat and elongate with a lot of Weibel-Palade bodies in the cytoplasm. Results of flow cytometry confirmed that ECs isolated from the follicular lymphoma expressed high level of both vWF and CD34and the purity of the ECs fraction was>90%. Next we used FISH to check chromosomal aberration in the purified ECs and found that some of ECs had only one fusion signal for the green IGH probe and red BCL2probe in contrast to typical t(14;18)(q32;q21) translocation with two fusion signals. This phenomenon was also observed in the tumor cells. It may be a different breakpoint of chromosome14containing IGH gene in this case, inducing a loss of fusion signal which also indicates t(14;18)(q32;q21) translocation. The positive cells amounted to18%of isolated endothelial cells from the tumor, indicating that a part of ECs from follicular lymphoma has the same chromosome aberration as the neoplastic cells. TNF-a is a double-edged sword to tumor: it can kill the tumor cells without injury of the normal cells, on the otherhand, it promotes tumor growth. TNFR-/-mice had unexpectedly delay growth of tumor in contrast to wild type mice. TNF-a expressed by tumor facilitates their survival and prevents apoptosis. In this study, we detected the expression of transmembrane TNF-a (TM-TNF-a) in leukemia and clarified the role of TM-TNF-a in leukemia. The killing effect of antibody specific to TM-TNF-a on leukemia cells in vivo and in vitro was observed. The main results are as follows:1. The expression of TM-TNF-a in acute leukemia. We used flow cytometry to determine the expression of TM-TNF-a on mononuclear cells from bone marrow of10healthy donors,33AML and22ALL patients. TM-TNF-a strong positive-primary leukemia cells were observed in19out of22(86.36%) ALL patients,24out of33(75%) AML patients including13out of15(86.67%) M5patients, indicating TM-TNF-a expression on leukemia cells at high level in part (about two-thirds) of patients.2. Comparing the expression of TM-TNF-a between normal stem cells and leukemia stem cells. TM-TNF-a strong positive-stem cells (CD34+CD38) were observed in88.89%of AML patients and85.71%ALL patients, but not in normal donors. The expression of TM-TNF-a in leukemia stem cells is positively correlated with that in leukemia cells.3. The relationship of TM-TNF-α expression in ALL with extramedullary infiltration. The expression level of TM-TNF-a on leukemia cells in ALL with extramedullary infiltration is significantly higher than that in ALL without extramedullary infiltration (P<0.05), pointing out that the more highly TM-TNF-α is expressed by leukemia cells, the more easily leukemia cells infiltrate extramedullarily.4. Inhibition of proliferation and promotion of apoptosis by knockdown TM-TNF-a expression. Silence of the TNF-a gene expression resulted in decreased proliferation and more apoptosis in THP-1cells.5. The CDC and ADCC induced by a monoclonal antibody against TM-TNF-a. In contrast to negative control (treated with Rituximab), the antibody of TM-TNF-a had CDC effect about30-35%and ADCC effect about40-45%, similar to Remicade, a monoclonal antibody against TNF-a.6. The killing effect of antibody of TM-TNF-a in vivo. The mice treated with antibody to TM-TNF-a had lower incidence of tumor and smaller tumor size in comparison with the mice treated with Rituximab. The effect of antibody to TM-TNF-a had no significant difference from that of Remicade.Conclusion:TM-TNF-a is expressed at high level in part of leukemia patients, and closely correlated with proliferation and extramedullary infiltration of leukemia cells. Treatment with antibody to TM-TNF-α decreases the incidence of tumor and inhibits the tumor growth. The results of this study provide experimental evidences and a new clue to TM-TNF-a targeting therapy for leukemia. So far, Galectin-9is only one known ligand for Tim-3. The earlier work in our lab shows that Tim3expressed by endothelial of lymphoma inhibits the activation and proliferation of T cells which leads to tumor immune escape. In order to understand the signal pathway of Tim-3, it's prokaryotic and eukaryotic protein were constructed and expressed in this study. We found that the other ligand(s) of Tim-3except for galectin-9is present in a mononuclear tumor cell line THP-1. We used co-Immunoprecipitation (Co-IP) and Liquid Chromatography-Mass Spectrometry (LC-MS) to find the unknown membrane molecules co-precipitated with Tim-3, and studied the effects of Tim-3on this cell line. The main results are as follows:1Construction and prokaryotic expression of sTim-3. The gene sequence of extracellular domain of human Tim-3was modified according to E.coli favourite codons, which was inserted into prokaryotic expression vector pet28+and expressed successfully by E.coli. This sTim-3was able to bind to the surface of two kinds of lymphoma cell lines, Su and Hut78.2Construction and eukaryotic expression of human sTim-3-IgGlCⅡ fusion protein. The signal peptide was combined with the gene sequence of extracellular domain of Tim-3(sTim-3), which was then linked to the sequence of constant region if IgGl heavy chain (sTim3-IgG1CH) by multiplex PCR. This fused gene was inserted into eukaryotic expression vector pIRES2-EGFP and then, stably transfected into CHO cell line. It has been found that this transfectant expressed and secreted the fusion protein sTim3-IgG1CH detected by flow cytometry and Western blot.3Screen the cell lines that express the unknown ligands of Tim-3with the fusion protein sTim3-IgG1CH. Among8kinds of leukemia cell lines, HL-60-. Kassumi, RS4-11Raji and Su showed the consistency of the expression rate of galectin-9with the binding rate of sTim3-IgG1CH to cells. However, the expression rate of galectin-9in THP-1was55%, less than the binding rate of sTim3-IgG1CH to cells (99%), indicating a possibility that unknown ligands of Tim-3may be present on THP-1.4Membrane molecules coprecipitated with Tim-3. Using Co-IP and LC-MS, we found that the membrane molecules in THP1cells co-precipitated with Tim-3were galectin-1and galectin-8except for galectin-9, as well as "inhibitor1of T cell activation", a signal molecule in Tim-3signal pathway.5The effects of sTim3-IgG1CH on THP-1. Assaying the effect of sTim3-IgG1CH on the release of cytokines by THP-1using CBA kits, it was found that sTIM-3significantly inhibited TNF-a release and slightly promoted IL-10secretion.Conclusion:In this study, we found that the unknown ligands for Tim-3may be galectin family members. sTIM-3stimulation significantly inhibits TNF-a release from THP-1, indicating a possibility that TIM-3might promote M2macrophage polarization. Although this conclusion should be confirmed by further studying, the results provide a new clue to explore the immune suppressive mechanisms of TIM-3.