Search

Mouse monoclonal antibody FU-MK-1, raised against a human gastric adenocarcinoma, recognizes a glycoprotein antigen (termed MK-1 antigen) present on most carcinomas and seems to be valuable in immunodiagnosis and immunotherapy of various cancers. In a recent study, we constructed a mouse/human chimeric antibody, designated Ch FU-MK-1, by fusing the FU-MK-1 V(H) and Vkappa genes to the human Cgamma1 and Ckappa genes, respectively. In the present study, we tested combination immunotherapy of Ch FU-MK-1 with human lymphokine-activated killer (LAK) cells in vitro and in mice with severe combined immunodeficiency (SCID) bearing human MK-1-expressing tumors. In in vitro experiments, Ch FU-MK-1 effectively mediated antibody-dependent cell-mediated cytotoxicity (ADCC) against MK-1-expressing MKN-74 cells, which was completely blocked by an anti-FcR antibody. Since the apoptotic pathway as well as the necrotic pathway have been shown to be utilized in various cytotoxic effector mechanisms, we investigated the role of apoptosis in ADCC mediated by LAK cells and Ch FU-MK-1 against MKN-74 cells. The implication of the apoptosis during ADCC was demonstrated by means of both a terminal-deoxynucleotidyltransferase-mediated dUTP-biotin nick-end-labeling assay and a propidium iodide staining method. In vivo antitumor activity of combination treatment with LAK cells and Ch FU-MK-1 was estimated using SCID mice inoculated s.c. with MKN-74 cells. The i.v. administration of LAK cells and i.p. administration of Ch FU-MK-1 and interleukin-2 (IL-2) produced a marked growth inhibition of MKN-74 tumors in SCID mice. When the actual tumor weights were measured 16 days after initiation of treatment, more than 70% reduction was observed in the group receiving LAK cells plus Ch FU-MK-1 plus IL-2 as compared to the control untreated group. Together these results suggest that Ch FU-MK-1 may serve as a potentially useful immunotherapeutic reagent for human MK-1-expressing tumors.

Mouse monoclonal antibodies against CD3 on human T lymphocytes have been used for therapy in organ-transplant patients as a potent immunosuppressive agent or for treatment of cancer as a potent T cell activating agent. However, an inherent problem in their in vivo application is the human anti-mouse antibody response. In this study, we cloned and sequenced the variable region genes of the heavy and light chains (VH and V kappa) of a mouse anti-human CD3 monoclonal antibody (OKT3) using the reverse transcription-polymerase chain reaction method. Then, we constructed a mouse/human chimeric antibody, designated as Ch OKT3, by fusing the OKT3 VH and V kappa genes to the human heavy and light chain constant region genes (C gamma 1 and C kappa) derived from a human plasma cell leukemia line (ARH77), respectively. The chimeric gene constructs were sequentially co-transfected into mouse non-Ig-producing hybridoma cells (Sp2/0) by electroporation. The Ch OKT3 antibody thus prepared bound to human peripheral blood mononuclear cells and competitively inhibited the binding of the parental MAb OKT3 to the blood mononuclear cells, indicating that this chimeric antibody seems to be suitable for in vivo therapeutic approaches.

Whole-body autoradiography (WBAR) was used to study the biodistribution of 125I-labeled mouse-human chimeric antibody (Ch F11-39) to carcinoembryonic antigen (CEA) in athymic nude mice bearing the CEA-producing MKN-45 human gastric carcinoma xenografts. Significantly high uptake of 125I-Ch F11-39 in the tumors obtained by tissue-counting technique was confirmed by WBAR of mice of 12, 24, 48, and 96 h postinjection of 125I-Ch F11-39. When compared with histochemical or immunohistochemical staining results of the tumor tissue sections, imaging profiles of 125I-Ch F11-39 obtained by WBARs were topographically correlated with histopathological findings of tissues and immunohistochemical localization of CEA in the tumor tissues, indicating that the accumulation of 125I-Ch F11-39 at the tumor site is based on its specificity for CEA. These results demonstrate that this chimeric antibody may serve as a potential useful diagnostic and/or therapeutic reagent for human CEA-producing cancers.

Binding reactivities of 62 anti-CEA MAbs from 10 different research groups with cell membrane-bound CEA and with free CEA in solution were compared by inhibition of MAb binding to CEA-expressing tumor cells by free CEA. Bindings of 30 MAbs to the cell membrane-bound CEA (280 ng CEA/2 x 10(5) cells) were inhibited by approximately equal amounts of free CEA, indicating that binding affinities of about half the MAbs for cell membrane-bound CEA are similar to those for free CEA, respectively. Bindings of 15 MAbs to the cell membrane-bound CEA were easily inhibited by free CEA of less than half the amount of the cell membrane-bound CEA, while inhibition of bindings of the remaining 17 MAbs required twice more free CEA than the amount of cell membrane-bound CEA, showing that about one-fourth of the MAbs have higher affinities for free CEA and the remaining about one-fourth of the MAbs possess higher affinities for cell membrane-bound CEA. These results help form the basis for selecting the anti-CEA MAbs for use in clinical applications, such as serum CEA assay, tumor imaging and immunotherapy.

To reduce heterophilic antibody interference in a two-site immunoassay for carcinoembryonic antigen (CEA), we utilized a human/mouse chimeric antibody to CEA as the tracer. One mouse monoclonal antibody (MAb), F82-61, which reacts with an epitope present on the domain N of CEA, was immobilized on 96-well polystyrene microtiter plates. A human/mouse chimeric antibody (Ch F11-39), which recognizes an epitope present on the domain B3 of CEA, was biotinylated for the tracer (Ch F11-39 system). Another MAb F11-39, the parental MAb of Ch F11-39, was also biotinylated and used as the control tracer (F11-39 system). For a fair comparison, the same 503 serum samples from healthy individuals were simultaneously assayed in the present study. When a tentative common reference limit of 5 ng/ml was used, the false positive rate with the Ch F11-39 system was only 2.8% (14/503) and that with the F11-39 system was 29.0% (146/503). Adding normal mouse serum (NMS; 1%) or a mixture of purified mouse IgG subclasses (heterophilic blocking reagent (HBR, 15 micrograms/test)) to the F11-39 system reduced the false positive rate from 29.0% to 6.2% (31/503) or 4.8% (24/503), respectively, suggesting that heterophilic antibodies reactive with mouse IgG gave rise to the high positive rate in normal populations with the F11-39 system. On the other hand, the false positive rate with the Ch F11-39 system was only slightly reduced from 2.8% to 2.6% (13/503) or to 2.0% (10/503) by adding NMS or HBR to the Ch F11-39 system. The false positive rates with two commercially available assay systems, CEA Roche EIA.DM or Abbott IMx CEA, were 5.4% (27/503) and 5.8% (29/503), respectively, which both corresponded roughly to that with the F11-39 system including NMS or HBR. These results indicate that the application of human/mouse chimeric antibodies in two-site immunoassays is more effective for reducing interference from heterophilic antibodies than the adding of NMS or purified mouse IgG in the assay using conventional MAbs.