MM46 Tumor Research Articles

Abstract 3232: Pharmacokinetic analysis of anti-PD-1 and PD-L1 antibodies and evaluation of their anti-tumor effects

Abstract Objective: Recently anti-PD-1 antibodies (aPD-1 Abs), anti-PD-L1 (aPD-1) Abs have been approved. However, the difference between both Abs in pharmacokinetics and anti-tumor effects have not been fully understood. In this study, we analyzed the difference between both Abs in blood concentration, biodistribution and degradation in tumor-bearing mice by using aPD-1/PD-L1 Abs labeled with radioisotopes (In-111/I-125) and evaluated the relationship between PK and therapeutic effects. Method: Abs were labeled with In-111 via chelate agents, and labeled with I-125 through covalent bond. Tumor bearing mice were prepared by s.c. inoculation with mouse colon cancer MC38 cells or mouse breast cancer MM48 cells. The labeled Abs were intraperitoneally injected into tumor-bearing mice. Tumors and organs were harvested at several time points after the injection, and radio activities in organs were measured by a gamma counter. The accumulation of Abs were expressed as % of injected dose/g organs. Because In-111 tends to be accumulated in organs due to poor permeability and I-125 was eliminated from organs rapidly due to high permeability, the ratio of I-125 and In-111 could reflect the degradation of Abs after cellular uptake. In pharmacological studies, Abs were intraperitoneally injected into tumor-bearing mice at doses of 50 to 200 μg (2.5 to 10 mg/kg) at day 5, 8, and 12 after tumor-inoculation. Tumor volume was evaluated to evaluate tumor progression. Result and Discussion: aPD-1 Ab showed anti-tumor effect in both MC38 and MM48 tumor models in a dose-dependent manner. On the other hand, aPD-L1 Abs showed lower anti-tumor effect in MC38 models, and negligible effect in MM48 tumor bearing mice at tested doses. According to PK studies, it was observed that aPD-L1 Abs were largely accumulated in normal tissues, especially in the spleen, liver, and kidney, and degraded rapidly compared with aPD-1 Abs, resulting that the blood concentration and distribution in tumors of aPD-L1 Abs tended to be low. Moreover, aPD-L1 Abs showed more rapid degradation in both tumors than aPD-1 Ab. Conclusion: aPD-L1 Ab showed less anti-tumor effect in tested tumor models due to less distribution and faster degradation in tumors than aPD-1 Ab. Collectively, the PK of aPD-1/PD-L1 Abs which target the same axis were not equivalent and the selectivity of expression of target molecules in both normal tissues and tumors should be considered to optimize their therapeutic efficacy. Citation Format: Hiroto Hatakeyama, Taiki Kurino, Reiko Matsuda, Hiroyuki Suzuki, Ayu Terui, Tomoya Uehara, Yasushi Arano, Akihiro Hisaka. Pharmacokinetic analysis of anti-PD-1 and PD-L1 antibodies and evaluation of their anti-tumor effects [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3232.

Enhanced radiation effect on tumor growth using oxygen-rich microcapsules

The aim of this study was to determine whether oxygen-releasing microcapsules could be used to sensitize cancer cells to kill by radiation. The microcapsules were generated by spraying a mixture of 0.1% alginate and hyaluronic acid into a 0.3 mmol/l solution of CaCl2 and FeCl2. These were then subcutaneously injected around a MM48 tumor (a cell line derived from human breast cancer) in the left hind legs of C3H/HeN mice, and tumors were dosed with 60Co γ-ray radiation. We showed that the oxygen released from the microcapsules enhanced the anti-tumor effect of radiation treatment via the generation of oxygen radicals.

Two-step mechanism of macrophage-mediated tumor lysis in vitro.

Normal peritoneal macrophages from C3H/He mice could not lyse syngeneic MM46 tumor cells in co-operation with syngeneic antitumor antibody. Thus, normal macrophages could not effectively participate in the antibody-dependent macrophage-mediated tumor lysis in vitro. However, after activation in vivo by stimuli, such as lipopolysaccharide, BCG, or glycogen, macrophages could co-operate with antitumor antibody in cytolysis of target cells. In the cytolysis nonspecific activation of normal macrophages was an essential first step, followed by specific tumor lysis in the presence of an antitumor antibody (second step). Immune macrophages from resistant mice were apparently equal in functional state to activated macrophages. A two-step mechanism of tumor lysis in vitro in a syngeneic mammary tumor system is proposed.

Scanning electron microscopic study of the three-dimensional structure of the collagen sheath surrounding cancer cells after single high-dose irradiation

The effect of single high-dose irradiation on collagen networks in tumor cells was examined by scanning electron microscopy. MM46 tumor cells were transplanted into C3H/He mice and locally irradiated with a 30 Gy dose (6 MeV) 7-8 days posttransplantation. Mice were killed at various time points postirradiation, and tumor tissues were excised and fixed. After 2 N NaOH maceration, tissues were washed and stained with 1% tannic acid. They were then postfixed with 1% OsO4, dehydrated, and dried. The tissues were coated with gold, and the three-dimensional structure of the collagen fibers was observed using scanning electron microscopy. In the control group, the chambers formed by collagen fibers were larger on days 7 and 14 than those on days 1 and 3. In the irradiated group, the chambers were largest and most irregular in shape on day 7. The chamber size had decreased approximately 50% by day 14, and there was evidence of tumor cell regeneration. In addition, several fenestrations in the chamber wall developed in the irradiated group. Fenestrations in the collagen chamber walls surrounding tumor cells might allow transfer of metabolites that provide nutrition to the tumor cells, leading to their regeneration.

The change of proliferating cell nuclear antigen and apoptosis of the MM46 mammary cancer cells of the mouse after single high-dose irradiation

The effects of intraoperative radiotherapy on tumor cells were elucidated by immunohistochemical examination of changes in the level of proliferating cell nuclear antigen. In addition, the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling method was used to examine the level of apoptosis in mouse MM46 tumor cells after a single high dose of irradiation (30 Gy, 6 MeV). A significant decrease in the number of tumor cells compared to controls was observed on the 3rd, 7th, and 14th days following irradiation, but not on the 1st day. Consistent with these results, the proliferating cell nuclear antigen labeling index of irradiated cells decreased significantly on the 3rd, 7th, and 14th days following irradiation, but not on the 1st day. By comparison, the regrowth area on day 14 only showed no difference compared to the control group. The apoptotic index increased on the 7th and 14th day after irradiation, but at a lower rate than the observed decrease in the level of proliferating cell nuclear antigen. We speculate that the major mechanism of single high-dose radiation effect is inhibition of DNA synthesis.

Complete Regression of Mouse Mammary Carcinoma with a Size Greater than 15mm by Frequent Repeated Hyperthermia Using Magnetite Nanoparticles

Magnetite cationic liposomes (MCLs) have a positive surface charge and are used as a heating mediator for hyperthermia, because they generate heat in an alternating magnetic field (AMF) due to hysteresis loss. In our previous paper, hyperthermia using MCLs was applied to animals having several types of tumors in mice, rats, hamsters, and rabbits, and a strong anticancer effect was observed. For mice, complete tumor regression was observed when the tumor size was smaller than 5 mm. However, a protocol for large tumors is necessary for the clinical application. In the present paper, a protocol for tumors with a size greater than 15 mm in mice was investigated. MCLs were injected into an MM46 tumor (size, 15 mm) in C3H/HeN mice, which were subjected to AMF for 30 min. The temperature at the surface of the tumor reached 45°C and was maintained by controlling the magnetic field intensity. Hyperthermia treatment was repeated twice with 24-h intervals (repeated hyperthermia; RH), and RH was carried out until complete tumor regression was observed. Complete tumor regression was achieved in all mice treated once, twice or six times with RH, and mice acquired antitumor immunity. This protocol, which is termed frequent RH, is a potent approach for cancer therapy.

Complete regression of mouse mammary carcinoma with a size greater than 15 mm by frequent repeated hyperthermia using magnetite nanoparticles

Magnetite cationic liposomes (MCLs) have a positive surface charge and are used as a heating mediator for hyperthermia, because they generate heat in an alternating magnetic field (AMF) due to hysteresis loss. In our previous paper, hyperthermia using MCLs was applied to animals having several types of tumors in mice, rats, hamsters, and rabbits, and a strong anticancer effect was observed. For mice, complete tumor regression was observed when the tumor size was smaller than 5 mm. However, a protocol for large tumors is necessary for the clinical application. In the present paper, a protocol for tumors with a size greater than 15 mm in mice was investigated. MCLs were injected into an MM46 tumor (size, 15 mm) in C3H/HeN mice, which were subjected to AMF for 30 min. The temperature at the surface of the tumor reached 45°C and was maintained by controlling the magnetic field intensity. Hyperthermia treatment was repeated twice with 24-h intervals (repeated hyperthermia; RH), and RH was carried out until complete tumor regression was observed. Complete tumor regression was achieved in all mice treated once, twice or six times with RH, and mice acquired antitumor immunity. This protocol, which is termed frequent RH, is a potent approach for cancer therapy.

Biological response modifier activity of Lactococcus lactis 332

The biological response modifier (BRM) activity of a preparation of heat killed cells of Lactococcus lactis 332 which can produce large amount of lactate in culture, was investigated in C3H/He and ICR mice. Intraperitoneal injection of Lactococcus lactis 332 caused an accumulation of neutrophils and macrophages in the peritoneal cavity of the mice, similarly to BRMs used clinically. As a parameter of the activation of macrophages, the effect of the Lactococcus lactis 332 preparation on the production of tumor necrosis factor (TNF) was examined. The Lactococcus lactis 332 preparation was revealed to have both priming and triggering activities for the production of TNF. The TNF level in the sera reached about 1000 IU/ml in mice 2 h after the triggering injection. This preparation also stimulated peritoneal macrophages to produce TNF in vitro. Intratumoral injection of the Lactococcus lactis 332 preparation regressed MM46 tumor cells in C3H/He mice. These results suggest that the Lactococcus lactis 332 preparation is a biological response modifier (BRM) with various activities on phagocytes similarly to a streptococcal antitumor agent, OK432, used clinically.

Morphological study of cytotoxicity produced by PSK-induced polymorphonuclear leukocytes (PMNs) and Nocardia rubra cell wall skeleton.

The morphologic changes in PMNs induced by an i.p. injection of PSK, a polysaccharide from the mycelia of Coriolus versicolor, and tumor cells undergoing cell death, were evaluated by immunohistochemical staining and electron microscopy. Male C3H/He mice, 8-10 -weeks old, received an i.p. injection of 125 mg/kg of PSK. Their PMNs were obtained 6 h after the PSK injection by peritoneal lavage. N-CWS (Nocardia rubra cell wall skeleton) was added at the start of the chromium release assay using the MM46 mammary carcinoma cell line, which is syngeneic to C3H/He mice, as target cells. During the cytotoxic assay, the cells were fixed at various time points. The MM46 cells expressed ICAM-1 while the PMNs expressed both ICAM-1 and LFA-1 as determined by immunohistochemical staining and immunoelectron microscopy using anti-ICAM-1 and anti-LFA-1 antibodies. PMNs with ruffle-like microvilli adhered to the MM46 tumor cells 30 min after the addition of N-CWS. Immunoelectron microscopic findings suggested that the adhesion molecules were LFA-1 on the PMNs and ICAM-1 on the MM46 tumor cells, but cell fusion between the PMNs and tumor cells was not observed. The MM46 tumor cells gradually lost their microvilli, which showed cell damage, and died 6-7 h after the addition of the N-CWS. This time course of tumor cell death is compatible with the results of the cytotoxic assay. Pretreatment of PMNs by anti-LFA-1 antibody suppressed 1% lysis of MM46 tumor cells from 90% to 10% (p < 0.01). These data suggest that adhesion molecule on the surface of PMNs such as LFA-1 might play an important role on signal transduction of these PMNs cytotoxic function in this experimental system.

The inhibitory effect of preoperative immunochemotherapy on the lymph node metastasis of murine MM48 tumor

A murine tumor model showing metastases to lymph nodes (LN) was established by intradermally implanting highly metastatic MM48 tumor cells (2 × 106) in C3H/HeN mice. We were searching for the most effective immunochemothera-peutic modality to treat this metastatic tumor. The combination therapy with chemotherapeutics, granulocyte colony-stimulating factor (G-CSF) and OK-432 on Days 8–11 was found to be remarkably effective, making the solid tumor disappear in more than half of the treated mice, though all of them eventually died of LN metastasis, as all the control mice did. Then an attempt was made to cure the mice from such fatal metastatic tumors with combined immunochemotherapy prior to surgical resection on Day 14. The combination therapy with chemotherapeutics, G-CSF and OK-432 more strongly inhibited the metastases then, and more than 85% of the mice survived. When the survivors were rechallenged with MM48 tumor cells, all of them rejected and survived without recurrences and metastases, indicating the acquirement of specific immunity. It is expected that this preoperative immunochemotherapy may be clinically useful for the treatment of malignant neoplasms.

Utilization of leucine methyl ester for the generation of hybridomas producing monoclonal antibodies specific to tumor-associated antigens.

In an attempt to obtain monoclonal antibodies specific to tumor-associated antigens, C3H/He mice were immunized with syngeneic MM2 tumor cells, and the primed spleen cells were fused with P3-X63-Ag8.653 myeloma cells. The outgrowth of hybridomas, however, was extremely low and monoclonal antibodies were not obtained. The reason for the low hybridoma growth was studied. It was found that MM2 cells used as the immunogen, the fusion partner myeloma cells and the resulting hybridomas shared at least one tumor-associated antigen, namely Q5 antigen. Because of this common antigen, cytotoxic cells, presumably cytotoxic T lymphocytes, which were lytic to the hybridomas, were induced during the culture for generation of the hybridomas. Removal of lysosome-rich cells, including cytotoxic T lymphocytes, from the primed spleen cells before the fusion by treatment with leucine methyl ester, a lysosomotropic agent, drastically improved the outgrowth of hybridomas. By this method, seven stable hybridoma clones producing monoclonal antibodies specific to tumor-associated antigens were obtained. Two of the seven clones were found to secrete monoclonal IgM species, which reacted with the extra-cellular region of the Q5 antigen. This procedure will be an option when production of monoclonal antibodies specific to cell-surface antigens is intended and outgrowth of hybridomas is unexpectedly low.

Prevention of lymph node metastases by adoptive transfer of CD4+ T lymphocytes admixed with irradiated tumor cells.

CD4+8- T lymphocytes with potent antitumor activity in vivo were obtained in peritoneal exudate cells by immunizing mice with irradiated MM48 tumor cells admixed with OK-432. These immune CD4+ T cells were used in adoptive immunotherapy for prevention of lymph node metastases after removal of the primary tumor. Complete cure of metastases was obtained by adoptive transfer of CD4+ T cells admixed with irradiated MM48 tumor cells, but not by CD4+ T cells alone. To analyze the curative effect of admixing tumor cells on the prevention of metastases, a model of 1-day tumor inoculated with macrophages was used. Administration of immune CD4+ T cells alone resulted in the regression of local tumor in more than half of the mice, although all of them eventually died of lymph node metastases. On the other hand, adoptive transfer of immune CD4+ T cells plus irradiated tumor cells resulted in the complete regression of local tumors in all the mice, which survived without any sign of metastasis. The curative effect of the immune CD4+ T cells obtained by admixing irradiated tumor cells was tumor-specific. Macrophages induced by OK-432 (tumoricidal), implanted together with tumor, assisted tumor regression more than did macrophages elicited by proteose peptone (non-tumoricidal) in the same adoptive transfer system. Administration of recombinant interleukin-2 instead of stimulant tumor cells did not enhance, but rather eliminated the constitutive antitumor activity of CD4+ T cells. On the other hand, exogenous recombinant interleukin-1 was more effective in the enhancement of antitumor activity of the CD4+ T cells as compared with stimulant tumor cell administration. In this case, the activating states of macrophages at the implanted tumor site had no influence on the therapeutic efficacy. A possible role of macrophages for induction of tumor-specific cytotoxic T cells that were mediated by tumor-specific CD4+ T cells is discussed.

Detection of allogeneic Qa/TL and Ly specificities on murine tumor cells with IgD in tumor-regressor serum.

Serum from C3H/He mice, which show regression of MM2 tumor cells after transplantation and removal (regressor serum, RS) contains non-gammaglobulin components that cross-react with various tumor cells of mice [22, 23]. In addition to tumor cells, various allogeneic lymphocytes are also susceptible to an RS-dependent lymphocyte-mediated cytotoxic reaction. To identify tumor cell surface antigens that cause the cross-reactive host response, the serum components were analyzed by absorption of RS with allogeneic lymphocytes. RS components were found to recognize allogeneic lymphocyte antigens including Qa-2 and Ly6.2. Specificity for the Qa-2 antigen was further tested using Qa-2-congenic mice. The expression of Qa-2 antigen was detected on the surfaces of MM2 and other tumor cells derived from H-2k mice (seven among nine cell lines tested) by a membrane immunofluorescence method using a Qa-2-specific mAb. Physical characteristics of the Qa-2-specific component in RS were determined and found to differ from those of regular IgGs but to be similar to those of IgDs. Using an enzyme-linked immunosorbent assay with an IgD-specific mAb and Qa-2-lacZ fusion protein, the existence of IgD in RS with specificity for Qa-2 was confirmed. These results suggest that the RS component with Qa-2 specificity is an IgD, the specificity and physiological role of which are unknown.

Inhibitory effects of plant secondary metabolites on cytotoxic activity of polymorphonuclear leucocytes.

The inhibitory effects of 151 natural products, representing most of the frequently occurring types, on the cytotoxicity to MM2 tumor cells of polymorphonuclear leucocytes (PMN) induced by TAK, a polysaccharide immunomodulator, were examined. Forty-two compounds inhibited the TAK-induced activation of PMN. Among them, some naturally occurring quinones and various alkaloids (nicotine, Cinchona alkaloids, isoquinoline alkaloids such as cepharanthine, and indole alkaloids such as ajmaline) exhibited potent inhibitory effects. Using the inhibition assay for monitoring, the extracts of Hydrangea Dulcis folium, Scopoliae rhizoma, Cinchona cortex, Magnoliae cortex, Stephania tuber, and Rauwolfia radix were analysed to characterize the active constituents.

Antibody Production Useful for Cytotoxicity against Mouse MM2 Tumor Cells by Peritoneal Macrophages

We applied an antibody-dependent macrophage-mediated cytotoxicity (ADMC) test in order to analyze the effector mechanism of the host-mediated antitumor effects induced by OK-432. Adherent peritoneal exudate (PE) cells were obtained from each of high (C3H/He) and low (B10) responder mice treated with OK-432, and 51Cr-labeled MM2 tumor cells were used as target cells. The cytotoxic activity in vitro coincided well with the results obtained in in vivo antitumor experiments. When adherent PE cells from C3H/He mice reacted with anti-MM2 serum from B10 mice, the degree of ADMC was significantly lower than that obtained with the anti-MM2 serum from C3H/He mice. The removal of IgG1 from the anti-MM2 serum induced in B10 mice resulted in the enhancement of ADMC activity. Then mean level of IgG2 in each of anti-MM2 sera from C3H/He and B10 mice was higher than in normal serum, and the IgG1 level in the antiserum from B10 mice was also higher than that in the serum from normal B10 mice. The present work suggested that the active component(s) in anti-MM2 serum participating in ADMC was a specific antibody of the IgG2 subclass, and that the inhibiting factor(s) was the IgG1 subclass.

Structural Characterization and Antitumor Activity of a Pectic Polysaccharide from the Roots ofAngelica acutiloba1

The polysaccharide fraction from the root of Angelica acutiloba Kitagawa showed a potent antitumor activity against ascitic form of Sarcoma-180, IMC carcinoma, and Meth A fibrosarcoma as well as the solid form of MM-46 tumor. An active polysaccharide, AR-4E-2, was purified by precipitation with cetyl-trimethylammonium bromide, anion-exchange chromatography on DEAE-Sepharose, and gel filtration on Sepharose CL-4B from the polysaccharide fraction. An active polysaccharide fraction showed a weak anti-complementary activity. AR-4E-2 was composed of arabinose, galactose, and rhamnose in the molar ratios of 3.3:1.0:0.7, and also contained 14.5% galacturonic acid and 3.2% protein. Methylation analysis and base-catalysed beta-elimination studies suggested that AR-4E-2 contained a rhamnogalacturonan moiety in which 2,4-di-substituted rhamnose residues were attached to 4-substituted galacturonic acid through position 2 of rhamnose. AR-4E-2 also contained highly branched 3,5-arabinan and (1----4)-galactan.

Cyclophosphamide‐dependent Lymph Node Modification in Lymph Node Metastasis of MM48 Tumor Cells in Syngeneic Mice

We investigated the role of immunosuppressive activity induced in the regional lymph nodes (RLN, popliteal lymph nodes) in the establishment of lymph node metastasis by cyclophosphamide (CY) administration. The CY treatment led to the elimination of suppressive activity with the appearance of positive immune responses, and the inhibition of lymph node metastasis of MM48 tumor cells. In CY‐treated mice, the removal of RLN together with the primary tumor lowered the survival rate compared with the mice in which the RLN remained intact. During 4 days after primary tumor resection, the proliferation of tumor cells in the RLN was significantly decreased in CY‐treated mice. These results suggested that the induction of suppressive activity in the lymph node is closely associated with the establishment of lymph node metastasis.

Modulation of Immune Response against Tumor Cells by the in vivo Administration of an Autoreactive Th Clone

The immunization of C3H mice with irradiated syngeneic MM48 tumor cells induced specific tumor‐neutralizing cells (TNC). The TNC activity was mediated by the L3T4+, Ly‐2− T cell population, and the generation of TNC coincided with the appearance of delayed‐type hypersensitivity response against MM48 antigen. The administration of an auto‐I‐Ak reactive T helper cell clone MS202 to normal C3H mice resulted in the facilitation of growth of MM48 tumor due to the induction of T suppressor (Ts) cells. Splenic T cells from animals given this T cell clone inhibited the TNC activity of immunized mice resulting in the escape of MM48 from the TNC effect. The surface phenotype of the Ts cells was L3T4+, Ly‐2−. The Ts cells induced by the clone MS202 were totally antigen‐nonspecific, and were able to suppress both the effector and inductive processes of TNC. The results suggest the presence of a physiological MHC‐restricted T cell circuit that regulates immune responses against the growing tumors.

Induction of tumoricidal macrophages from bone marrow cells of normal mice or mice bearing a colony-stimulating-factor-producing tumor.

Nonadherent cells of the bone marrow of C3H/HeN mice were incubated for 3 days with the culture supernatant of an L-929 cell line containing macrophage-colony-stimulating factor. Approximately, 70% of the cells became phagocytic, adherent to plastic dishes and positive for nonspecific esterase staining. The adherent cells exhibited a weak tumoricidal activity against syngeneic mammary carcinoma cells, and the cytotoxicity was strongly augmented by the addition of bacterial lipopolysaccharide to the cytotoxicity assay. The cytotoxicity induced by lipopolysaccharide was also shown to be mediated by Thy1.2- and asialo-GM1+ cells, and was abrogated by the addition of carrageenan. Macrophage-colony-stimulating-factor-producing (D66) and nonproducing (A23) variants were separated from the MM48 tumor line in in vitro culture following limiting dilution. There was no difference between these two variants in either the in vitro growth rate or the susceptibility to macrophage-mediated cytotoxicity. C3H/HeN mice inoculated i.p. with D66 survived longer than did those inoculated i.p. with A23. C3H/HeN mice bearing D66 or A23 as an ascitic form were given i.p. injections of Nocardia rubra cell wall skeleton (N-CWS). N-CWS significantly prolonged the survival period of mice bearing D66, whereas it exhibited no apparent antitumor effect on mice bearing A23. The increase in the cell number of D66 in the peritoneal cavity was significantly retarded, compared with that of A23. In contrast, the number of peritoneal macrophages increased more in D66-bearing mice than in A23-bearing mice. The increase in the peritoneal macrophage number was further augmented by an i.p. injection of N-CWS. Peritoneal macrophages of D66-bearing mice exhibited apparent tumoricidal activity against MM48 tumor cells in the presence of lipopolysaccharide, and the cytotoxicity was significantly augmented by i.p. injection of N-CWS. On the other hand, the responsiveness of peritoneal macrophages to lipopolysaccharide was found to be poor in A23-bearing mice and the tumoricidal activity was only weakly augmented by N-CWS. These results strongly suggest that M-CSF plays an important role not only in the maturation of macrophage progenitors but also in the induction and the accumulation of activated macrophages.

Production of colony-stimulating factor by tumor cells and the factor-mediated induction of suppressor cells.

Spleen mononuclear cells of C3H/HeN mice were cultivated with mitomycin C-treated tumor cells, X5563, MH134, MM48, MM46, and FM3A/R, all of which were of syngeneic origin, in a medium containing normal syngeneic mouse serum but not FCS. There was a proliferative response to X5563, MH134, and MM48, but not to the two other tumor cells, MM46 and FM3A/R. The responder spleen cells were found to be nonadherent cells with a phenotype of Thy-1-L3T4-Lyt2-Ig-Macl-, which were neither mature T and B cells nor mature macrophage/granulocytes. It was also found that the proliferation of these nonadherent no-marker cells was mediated by tumor cell-derived soluble factors but not by direct stimulation with tumor cells. The responsible factor was a molecule(s) with a Mr of 23 to 25 kDa, which had a CSF activity inducing granulocyte (G)-, macrophage (M)- and G + M-colonies in the bone marrow cells. Neutralization tests of this factor-induced proliferation of spleen cells revealed that a major part of the factor may be GM-CSF or a molecule closely related to it. Incubation of spleen mononuclear cells with these GM-CSF-like tumor cell factors resulted in induction of myeloblastic/promyelocytic cells with a phenotype of Mac-1+2+Ia+ Thy-1-L3T4-Lyt2-Ig- in the spleen cell cultures, which could suppress mitogenic responses of the spleen cells to T and B cell mitogens. GM-CSF-like activity could also be detected in the serum of mice bearing X5563, MH134, and MM48, but not in those bearing MM46 and FM3A/R. Subcutaneous inoculation of C3H/HeN mice with these X5563, MH134, and MM48 tumor cells generated massive metastasis in the lung and lymph nodes, whereas MM46 and FM3A/R produced no macroscopic tumor cell metastasis. These results strongly suggest the possibility that in some tumor cell-host systems, a GM-CSF-like factor(s) produced constitutively by the tumor cells may play an important role in the development of tumor metastasis, mediating through suppression of lymphoid tissues of the host.