Interaction of the heterozygous nude gene with the asplenia trait in mammary tumorigenesis.

The BALB/c mouse strain has been shown to contain endogenous mouse mammary tumor virus (MMTV) proviral sequences. However, no exogenous MMTV particles have been detected in their tissues. Female BALB/c mice from our colonies exhibit a very low incidence of spontaneous mammary tumors (SMT); less than 1% at up to 20 mo of age. Immunodeficient BALB/c mice heterozygous for the nude gene (nu/+, +/+), for the dominant hemimelia gene associated with asplenia (+/+, Dh/+), or for both traits (nu/+, Dh/+) have been examined for SMT incidence and the presence of MMTV proviruses. Based on restriction digestion with Eco RI, Bam HI, and Pst I, the immunodeficient mice have an MMTV provirus copy number and organization identical to the BALB/cCrgl strain. This MMTV DNA pattern is distinct from the MMTV proviruses in C3H/He, C57BL/6J and CBA/CaJ mice, which were parental strains of the immunodeficient mutants. Normal female BALB/c or BALB/c heterozygous for the asplenic trait do not develop significant numbers of SMT at up to 19 mo of age. In contrast, an incidence of 23.8% and 57.7% SMT was observed in BALB/c nu/+ heterozygotes, and in BALB/c nu/+, Dh/+ heterozygotes, respectively. These results indicate that agenesis of the spleen, concomitant with the presence of the heterozygous nude gene, contribute to a high incidence of SMT in the low-SMT BALB/c mouse strain.

The altered tumoricidal capacity of macrophages isolated from tumor-bearing mice is related to reduce expression of the inducible nitric oxide synthase gene.

Nitric oxide (NO) is a major effector molecule in the destruction of tumor cells by activated macrophages. However, in many cases, developing neoplasms appear to be capable of impairing steps in the complex process leading to NO production as a means of avoiding immune destruction. After activation with lipopolysaccharide (LPS), peritoneal-elicited macrophages (PEM) from mice bearing mammary tumors display alterations in their ability to lyse tumor cells due to reduced production of NO. In contrast, when these same cells are stimulated with LPS in combination with interferon gamma (IFN-gamma), they are able to produce NO and lyse targets at normal levels. Since tumor-associated macrophages are intimately associated with the cells of the developing tumor, their ability to produce NO and lyse tumor targets is likely to be more relevant to controlling tumor growth. This population of macrophages exhibited a more profound inability to produce NO and lyse targets and, unlike the PEM, was not able to upregulate these functions even when treated with combinations of LPS and IFN-gamma. Northern and Western blots revealed that inducible nitric oxide synthase (iNOS) mRNA and protein levels correlated directly with the ability of each macrophage population to produce NO, and the levels of these macromolecules were altered sufficiently in tumor bearers' macrophages to account for the diminished NO production described. These results indicate that a spatial gradient of suppression of macrophage cytolytic activity and iNOS expression exists in mammary tumor-bearing mice, whereby macrophages from within the tumor exhibit a more pronounced suppression than the more distally located PEM. This suppression may be due to proximity of the macrophages to the developing tumor, macrophage maturational state, or both.

Isolation of a nitric oxide inhibitor from mammary tumor cells and its characterization as phosphatidyl serine.

Macrophages from mice bearing large D1-DMBA-3 mammary tumors have a decreased capacity to kill tumor targets. This effect is due to an impaired ability to produce nitric oxide (NO) in response to lipopolysaccharide (LPS) stimulation. Here we report that the DA-3 tumor cell line, derived from the in vivo adenocarcinoma D1-DMBA-3, produces a factor that inhibits both NO production/release and cytotoxicity of LPS-activated peritoneal exudate macrophages (PEM). However, other complex macrophage functions such as phagocytosis, superoxide production, mitochondrial dehydrogenase activity, and synthesis of proteins were not reduced by this factor. The NO inhibitor has been found to be lipid in nature. Lipid extracts from DA-3 cell culture supernatants were purified by repeated silica gel column chromatography. The active molecule was unambiguously characterized as phosphatidyl serine (PS) by fast atom bombardment tandem mass spectrometry. Preliminary results indicate a lack of induced NO synthase (iNOS) activity in the lysates of LPS-activated PEM pretreated with PS. The ubiquity of PS in the inner leaflet of biological membranes and its NO inhibitory property, suggest that this phospholipid may be one of the long elusive molecules responsible for regulating physiological levels of NO in the host and hence preventing cellular dysfunction and/or tissue damage. Furthermore, the possible overexpression and shedding of PS by DA-3 tumor cells may represent a novel mechanism to impair macrophage cytotoxicity, a host function that contributes to the protection against developing neoplasms.

Influence of mammary tumor progression on phenotype and function of spleen and in situ lymphocytes in mice.

Peripheral immunity to an immunogenic chemically induced mouse mammary adenocarcinoma has been demonstrated in the syngeneic host, i.e., BALB/c mice, by several in vivo and in vitro cell-mediated immune assays. Analysis of the responsiveness of the immune cells within the in situ population may prove to be more indicative of the actual interactions between host and tumors. A centrifugal elutriation procedure was used to isolate tumor-infiltrating lymphocytes. A fluorescence-activated cell sorter revealed that most in situ cells were of the T-cell lineage as determined by the absence of surface immunoglobulin (sIg) and the presence of the Thy 1.2 antigen on their surfaces, while in mice with large tumors there were 27.4% Thy 1.2+ and 47.9% sIg+ cells Further studies using fluorescein-conjugated monoclonal anti-Lyt 1 or anti-Lyt 2 antibodies revealed a decrease in the levels of Lyt 1+ cells and an increase in the percentage of Lyt 2+ lymphocytes in the spleens of mice bearing large tumors. Within the Thy 1.2+ population infiltrating mammary tumors, the proportion of lymphocytes presenting the Lyt 1 marker was decreased in comparison to that of spleens of normal mice (0.58 vs. 0.83). At the same time the relative proportions of Lyt 2+ cells within the Thy 1.2+ population was increased in the in situ population (0.50) compared to those of spleens of normal mice (0.28). Examination of the functional abilities of the in situ lymphocytes (ISL) revealed that ISL derived from small tumors responded to the T-cell mitogens phytohemagglutinin (PHA) and concanavalin A (Con A), although at lower levels than those of the splenocytes of the animals from which they were obtained. Responses to PHA were lost rapidly in ISL from large tumors, whereas Con A responses were more durable. In contrast to the spleen cells of tumor bearers, ISL failed to respond to tumor-associated antigens (TAA) at any stage of the disease. The changes in the subsets of T-cells in the spleens of tumor-bearing mice and in ISL may provide an explanation for the decreased activation of ISL by mitogens and for the lack of responsiveness to TAA observed in the splenocytes of mice with large tumors and in the ISL isolated from the mammary adenocarcinomas.

Antigenic determinants in a virus-induced mouse mammary tumor recognized by cell-mediated immune assays.

Purified tumor cell membrane (PTCM) fractions from spontaneous BALB/cfC3H mammary adenocarcinomas elicit blastogenic responses in spleen cells of tumor-bearing mice. Previous studies have demonstrated that splenic T-cells are the major responding cell population but have not clarified whether the antigens responsible for the reactions are exclusively viral in origin or may involve nonviral tumor-associated antigens (TAA). Pretreatment of PTCM preparations with polyvalent anti-murine mammary tumor virus (MuMTV) completely obliterated the ability of PTCM to stimulate an innocent bystander cytotoxicity reaction; however, it reduced the blastogenic response by only 60%, suggesting that the two assays may measure different antigenic reactivities. The specificity of the anti-MuMTV blocking for viral antigens was demonstrated by the complete absorption of the serum-blocking reaction with purified MuMTV particles in the cytotoxicity assay; however, absorption was only partial in the blastogenesis assay. Incubation with purified Rauscher murine leukemia virus particles failed to absorb the neutralizing effect of the sera. These data suggest that putative TAA can be detected in the blastogenesis assay but not in the cytotoxicity reaction. Ammonium sulfate-precipitated immunoglobulin fractions of sera from tumor-bearing BALB/cfC3H mice (TBMS) completely blocked the stimulatory potential of PTCM in both the blastogenesis and the cytotoxicity assays. In the reciprocal experiment to that described above, preabsorption of TBMS immunoglobulins with purified MuMTV completely removed the inhibition of the cytotoxicity but again only partially restored the blastogenic response. The reaction could be completely restored by preabsorption of TBMS with PTCM. These results support the contention of nonvirion antigen involvement in the blastogenesis reaction. In conclusion, these two assay systems detect different antigenic determinants on the MuMTV-expressing tumor cell membranes. Both viral and other antigens appear to be relevant in this model system, and serum factors present in the immunoglobulin fractions of tumor-bearing mice can inhibit T-cell responses directed at either kind of antigenic moieties.

Kinetics of responses to tumor antigens and mouse mammary tumor virus in BALB/cCrgl and BALB/cfC3H mice.

Spleen cells from BALB/cCrgl mice responded to murine mammary tumor virus (MuMTV) in cell-mediated immune assays at higher levels than did the spleen cells from the syngeneic BALB/cfC3H mice. Implantation in BALB/cCrgl with a chemically induced mammary tumor and in BALB/cfCrgl mice with spontaneous mammary tumors (SMT) arising in the same strain resulted in sensitization of these animals to the antigens of their tumors. Reactivities peaked 3 weeks after transplantation, whereas no positive reactions could be detected when tumors reached maximum size. A kinetic study with the use of MuMTV antigen(s) showed that the responses of lymphocytes from BALB/cfC3H with SMT followed the same pattern as that obtained with tumor antigens, which indicated that this might be a de novo sensitization. In sharp contrast, a steady type of response to MuMTV was observed with the spleen cells from BALB/cCrgl mice; i.e., the levels of responsiveness to MuMTV did not significantly vary at any stage of tumor development. In vivo studies explored the possible relevance of the in vitro cell-mediated immunity to the host defenses. MuMTV-expressing mammary tumor cells were implanted in BALB/cCrgl and BALB/cfC3H mice. The total incidence of tumors was significantly reduced and a delay occurred in their time of appearance in BALB/cCrgl mice in relation to BALB/cfC3H animals. Thus the in vitro reactivity to MuMTV antigen(s) in BALB/cCrgl mice was found to be coincidental with a degree of protection against the development of MuMTV-expressing mammary tumors.

Dependence of initiation of an innocent bystander cytotoxicity reaction on the association of viral antigens and tumor cell membranes in mice.

In chemically induced mouse mammary tumors in BALB/c mice, no murine mammary tumor virus (MuMTV) antigens could be detected in the tumor cell membranes. In contrast, in mammary tumors of spontaneous appearance in BALB/cfC3H mice neonatally infected with MuMTV, viral antigens could be detected by immunofluorescence. Specific activation of immune spleen lymphocytes in vitro by homologous tumor cell membrane preparations resulted in an innocent bystander cytotoxicity reaction in BALB/c mice bearing either the chemical- or virus-induced mammary tumors. Non-tumor bearers did not respond in this reaction. The cytotoxicity effector cell was a nylon-adherent Thy 1.2+, Lyt 1+, Lyt 2+ lymphocyte. Induction of the reaction in the chemically induced mammary tumor bearers was related to tumor-associated antigens, but in the mice with MuMTV-induced tumors, it was possible that all responses were solely due to viral antigens. Biochemical analyses using immunoprecipitation techniques indicated that the major external glycoprotein of MuMTV (gp52) was present in the tumor cell membrane preparations. Preincubation of the cell membranes of virus-induced tumors with anti-MuMTV completely blocked the capacity of this preparation to induce the cytotoxicity. Preabsorption of the anti-MuMTV with purified MuMTV removed all the blocking activity of these sera, indicating that the antigenic determinants recognized were of viral origin. However, purified MuMTV failed by itself to elicit the innocent bystander cytotoxicity. This observation indicates that an association with membranes was necessary for the viral antigens to initiate and/or effect this cell-mediated immune reaction.

Mitogen-induced blastogenesis and receptor mobility inhibition by breast cancer serum with elevated orosomucoid (alpha 1-acid glycoprotein) levels.

Sera from cancer patients have been shown to suppress normal lymphoid cell responsiveness in vitro. In the present study, sera from breast cancer patients were demonstrated to be inhibitory to the concanavalin A (Con A)-, Proteus vulgaris-derived phytohemagglutinin-, and pokeweed mitogen-induced blastogenic responses of normal lymphoid cells. Orosomucoid (OR) (alpha 1-acid glycoprotein), an acute-phase reactant, was elevated in these sera, and a positive correlation existed between the OR level in the sera and its immunosuppressive capacity. When normal lymphoid cells were reached in fluorescein isothiocyanate-labeled Con A, cells that had been preincubated in breast cancer sera known to contain an elevated level of OR showed a significant decrease in Con A receptor mobility as compared to the receptor mobility of the same lymphoid cells preincubated in normal sera. Thus a component(s) from the sera of the breast cancer patients had the capacity to inhibit lymphocyte activation. This inhibition might have resulted from an interaction of OR with the membrane.

Lymphoproliferative responses to mouse mammary tumor virus in lymphocyte subsets of breast cancer patients.

Altered immunologic reactions were observed in breast cancer patients as compared to those in normal subjects. Lymphoproliferative responses to murine mammary tumor virus (MuMTV) were significantly enhanced in peripheral blood mononuclear cells from patients with metastatic disease. These reactivities occurred with mammary tumor virus purified from either mouse milk or infected feline kidney cells but not with Rauscher murine leukemia virus. For the assessment of the role of peripheral blood mononuclear leukocyte subpopulations in the responsiveness to MuMTV, the cell preparations were fractionated according to their ability to form spontaneous rosettes with sheep red blood cells (E-rosettes). The effectiveness of the separation was ascertained by means of cell surface markers, i.e., presence of surface immunoglobulins or a T-cell marker. Leu-1 antigen, and mitogen-induced blastogenesis. The responsiveness to the MuMTV antigen(s) was associated with the T-cell subset, identified as the E-rosetting. Leu-1-positive, and surface immunoglobulin-negative population. Although some subjects with the normal population gave positive reactions, the results reveal an apparent association between high levels of responsiveness to MuMTV within the T-lymphocyte subset and breast cancer disease.

Maximizing differences in the concanavalin A-induced blastogenic responses of lymphocytes from breast cancer patients and controls by the use of alpha-methyl-D-mannoside.

In an attempt to magnify differences in the immune responses of potentially immunosuppressed cancer patients and normal controls, an assessment was made on the effects of the competitive inhibitor alpha-methyl-D-mannoside on the concanavalin A (Con A)-induced blastogenic responses of lymphocytes from each of these populations. Lymphocytes from breast cancer patients with metastatic disease were significantly deficient in their capability to undergo blast transformation regardless of whether the monosaccharide inhibitor was added to the assay cultures. In contrast, lymphocytes from breast cancer patients who did not display metastatic disease were capable of normal blastogenic responses to Con A. The addition of alpha-methyl-D-mannoside to lymphocyte cultures caused a significantly greater inhibition of the blastogenic responses of these patients' cells as compared to cells of normal controls. Thus the monosaccharide seems to serve as a useful reagent for optimizing differences between lymphocyte blastogenic responses of normal donors and those of immunodepressed donors. The results suggest that lymphocytes from breast cancer patients without clinically evident metastases possess some modification of their cell membrane. One possibility discussed was that the number or distribution of receptors for Con A on the membrane of lymphocytes of these patients is deficient.

Induction of "innocent bystander" cytotoxicity in nonimmune mice by adoptive transfer of L3T4+ Lyt-1+2- mammary tumor immune T-cells.

Spleen cells from BALB/c mice, bearing a syngeneic mammary adenocarcinoma, nonspecifically lyse xenogeneic target cells following in vitro reexposure to mammary tumor-associated antigens. The effectors of this reaction have been shown to be Thy-1+ Lyt-1+2+ nylon-adherent cells. Tumor-immune spleen cells are also able to induce nonimmune spleen cells to express nonspecific cytotoxicity. Studies were undertaken to determine whether this inducer activity is mediated by the effector population or by a functionally distinct subset. Cell separation studies demonstrated that the inducers and effectors of innocent bystander cytotoxicity can be separated based upon the property of adherence to nylon wool. Further examination of the nylon-nonadherent inducer population indicated that the phenotype is L3T4 + Lyt-1+2-. By flow cytometry the inducer subset was shown to express a higher density of Thy 1 antigen than that expressed by the cytotoxic effectors. When adult thymectomized mice were implanted with mammary tumors, nonspecific effectors were not generated. In contrast, spleen cells from the same experimental animals were able to induce nonspecific cytotoxicity in normal mice following adoptive transfer of their spleen cells. Thus, these data support the conclusion that during the course of mammary tumor growth, at least two functionally and phenotypically distinct cell populations interact for the expression of "innocent bystander" cytotoxicity.

Adenocarcinoma R-3327 of the Copenhagen rat as a suitable model for immunological studies of prostate cancer.

An experimental animal model for the study of host-prostatic tumor cell interactions has been described. R-3327, a line of prostatic adenocarcinoma of the Copenhagen rat, has been proven to be immunogenic to its syngeneic host as evidenced by two different in vitro cell-mediated immune assays. Specificity of the responses has been ascertained on the basis of absence of response: (a) of nonimmune lymphocytes to the R-3327 tumor antigen(s); (b) of R-3327 immune lymphocytes to several normal tissues including normal prostate; (c) of immune lymphocytes to unrelated squamous cell prostatic carcinoma of the Copenhagen rat. Furthermore, the presence of tumor has an effect in several nonspecific aspects of host response, inducing splenomegaly, heightened responses to nonspecific mitogens in lymphocyte transformation assay, and increased levels of killer cell action. Since there are many histological, biochemical, and functional analogies between this tumor line and human prostate carcinomas, this system appears to be suitable for immunological and possible immunotherapeutic studies of this type of neoplasia.

Characterization of the effector cells mediating an "innocent bystander" cytotoxicity reaction induced by a syngeneic mammary adenocarcinoma in mice.

Spleen cells from BALB/c mice bearing a syngeneic mammary adenocarcinoma nonspecifically destroy xenogeneic targets following in vitro induction with mammary tumor-associated antigens. Studies were undertaken to characterize the effector cell population(s) mediating this "innocent bystander" cytotoxicity reaction. Fractionation experiments using phagocyte-depleted spleen cells revealed that the effector population was adherent to nylon wool columns. Flow cytometric analysis of the nylon-adherent cells revealed the presence of a minor population of Thy 1.2+ cells. Following treatment of the nylon-adherent cells with anti-Thy 1.2 and complement, the cytotoxic activity was abolished. Furthermore, when those cells expressing the Thy 1.2 antigen were positively selected by cell sorting, they were able to mediate the cytotoxic reaction. In contrast, nylon-adherent Thy 1.2-negative cells were unable to mediate the reaction following selection by cell sorting. Depletion studies with anti-Lyt 1 or anti-Lyt 2 and complement also abolished this cytotoxic response. Additional studies demonstrated that nylon-adherent spleen cells from mammary tumor-bearing mice were not able to lyse natural killer cell-sensitive targets. These data suggest that the cells which effect tumor antigen-induced "innocent bystander" cytotoxicity, or their activated precursors, are nylon-adherent Thy 1.2+, Lyt 1+2+ T-cells.

Subset of spleen lymphocytes from BALB/cCrgl mice stimulated by mouse mammary tumor virus.

Lymphocytes from BALB/cCrgl mice react to mouse mammary tumor virus-associated antigen(s) when tested in in vitro blastogenic transformation assays. These mice have a low incidence (less than 1%) of spontaneous mammary tumors and are free from complete mammary tumor virions. We have studied the nature of the lymphoid cells mediating the lymphocyte transformation reaction to purified mammary tumor virus. With the use of nylon wool columns, the responder cells were found to belong to the nylon-adherent population. The T-lymphocytes were not stimulated by mammary tumor virus even in the presence of added macrophages. These results were reconfirmed with treatments of spleen cells with either anti-surface immunoglobulin and complement or anti-Thy 1 antigen and complement. Thus, B-cells seem to be the lymphoid population responsive to mammary tumor virus-associated antigen(s) in the spleen of BALB/cCrgl. The cause of this reactivity may be a result of any of the following: (a) horizontal transmission; (b) activation of spleen cells by viral host cell contaminants in mammary tumor virus preparations; (c) a nonspecific mitogenic reaction exerted by the virus in the system; (d) sensitization to mammary tumor virus-associated antigen(s) due to the expression of an endogenous virus. We present here data arguing against the first three possibilities. In recent work, we found evidence supporting the expression of mammary tumor virus-related antigen(s) on lymphoid cell surfaces of BALB/cCrgl. From these studies, we propose that the responses seen in our in vitro assays may represent a sensitization event resulting from exposure to an endogenous mammary tumor virus gene product.

In vitro immune responses to viral and tumor antigens in murine breast cancer.

Inhibition of migration of peritoneal exudate cells proved to be a useful measurement of cell-mediated immunity which correlated in several respects with blastogenic transformation reactions. Lectins (phytohemagglutinin and concanavalin A) inhibited the migration of peritoneal exudate cells from normal and tumor-bearing mice, whereas tumor antigen caused inhibition of migration of cells from tumor-bearing animals only. The disparity in immunogenic capacity previously observed with lymphocyte transformation studies was also manifested in migration inhibition, i.e., D1-DMBA-3 tumor being immunogenic and D1-DMBA-2 being nonimmunogenic. Using the migration inhibition and blastogenic transformation reactions, responses were obtained to mammary tumor virus (MTV) antigen(s) in cells from BALB/cCrgl mice, which are free of MTV. In contrast, cells from MTV-positive BALB/cfC3H mice failed to respond to this antigen(s) in both reactions, suggesting a form of tolerance. However, the reactions became positive after implantation with MTV-containing spontaneous mammary tumors. Two possible explanations of the origin of reactive lymphocytes, horizontal transmission, or activation of a gene coding for an MTV antigen(s), are discussed.

High incidence of mammary tumors in mice with inherited asplenia carriers for the nude gene.

A colony of mice suffering from dominant hemimelia associated with agenesis of the spleen has been developed and characterized during the past 7 years. The hereditarily asplenic (Dh/+) mice have a very low incidence (9%) of spontaneous mammary tumors (SMT). Asplenic (Dh/+) females were mated with mice homozygous (nu/nu) for hereditary athymia (nude) having a BALB/c background. BALB/c females heterozygous for the nu gene and with spleen (nu/+,+/+) have a moderate incidence (12%) of SMT, whereas nu/+,Dh/+ breeders have a drastic increase in the incidence of SMT to 46% when bred under identical conditions. Since all parent strains have a very low incidence of SMT, it appears that the spleen agenesis is a major factor accounting for an earlier and higher incidence of SMT in hereditarily asplenic (nu/+,Dh/+) mice than in normal (nu/+,+/+) siblings. The SMT express mammary tumor virus antigen(s) and possess estrogen, progesterone, and glucocorticoid receptors. The SMT rapidly metastasize and kill the host within 30 to 45 days. The BALB/c asplenic mice with SMT represent a unique model relevant to human breast cancer and for study of the function of the spleen in the development of solid tumors in general and of SMT in particular.

Expansion of immunoregulatory macrophages by granulocyte-macrophage colony-stimulating factor derived from a murine mammary tumor.

Using an immunogenic nonmetastatic murine mammary adenocarcinoma (D1-DMBA-3) induced in BALB/c mice by dimethylbenzanthracene, we have previously shown that splenocytes from tumor bearers have depressed lymphocyte responses to mitogens and antigens, including tumor-associated antigens. In addition, they display decreased natural killer and T-cell cytotoxic activities. Macrophages from tumor-bearing mice appear to be responsible for the suppression of T- and B-cell responses to concanavalin A, lipopolysaccharide, and tumor-associated antigens observed in tumor bearers. The appearance of these macrophages in the spleen tightly parallels the progressive growth of the tumor and the concomitant immunosuppression. Simultaneously high levels of macrophage progenitors were observed in blood, bone marrow, lung, and liver. A significant increase of colony-stimulating activity of the granulocyte-macrophage lineage was detected in the sera from tumor-bearing mice. Higher levels of this colony-stimulating activity (CSA) were detected in tumor cystic fluid as compared with the levels in serum. A tumor cell line established in vitro from the D1-DMBA-3 in vivo tumor produces high levels of a factor with CSA in culture supernatant fluids. Partial purification of the CSA from the tumor cell line supernatants was achieved using CentriCell ultrafiltration and SephacrylS-300 chromatography. These studies revealed that the molecular weight of the colony-stimulating-like factor is 32,000 to 35,000. The morphology of the colonies obtained in cultures using this factor is similar to that of the colonies that develop in the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF) but not with macrophage colony-stimulating factor (M-CSF). CSA from tumor cell supernatants was neutralized by antiserum to GM-CSF but not with anti-M-CSF or anti-granulocyte colony-stimulating factor (G-CSF). Macrophages from bone marrow or peritoneal exudates from normal mice cultured with tumor supernatant for 2 to 3 days strongly inhibit normal splenocyte responses to concanavalin A and lipopolysaccharide. The data suggest that the tumor releases a GM-CSF that alters the hemopoietic system and induces or expands macrophages, which exert a suppressive function on the immune system of tumor-bearing mice.

Splenic macrophages from tumor-bearing mice co-expressing MAC-1 and MAC-2 antigens exert immunoregulatory functions via two distinct mechanisms.

Tumor burden has been shown to induce a variety of phenotypic and functional changes in the cellular constituents of the host's immune system. These changes have been implicated as mechanisms by which tumors avoid rejection. Studies of BALB/c mice bearing a D1-DMBA-3 mammary adenocarcinoma showed alterations of the splenocyte populations. There was a five-fold increase of macrophages (M phi) that were phenotypically and functionally analyzed to establish their role in tumor-induced modifications of the host's immune response. Monoclonal antibody staining defined a Mac-1+2+ population which comprised up to 20% of the splenocytes in tumor-bearers (TB), but is negligible in spleens from normal mice. These Mac-1+2+ M phi were found to mediate down-regulation of both polyclonal and antigen-specific T and B cell responses in vitro and in vivo. Although B cell responses were suppressed via prostaglandin E2 (PGE2) production by the TB M phi, T cell responses were relatively refractory to PGE2-mediated down-regulation. Instead, they were suppressed by a contact-dependent T cell-M phi interaction. Furthermore, tumor-derived factors such as granulocyte-M phi colony-stimulating factor (GM-CSF) seem to play an important role in the induction and expansion of the Mac-1+2+ M phi. These cells appear to mediate down-regulation of the host immune responses by at least two distinct mechanisms: 1) PGE2 production and 2) a cell contact-dependent, but non-major-histocompatibility-complex-specific, interaction.

Emergence of a B lymphocyte population with ADCC effector function in mammary tumor bearing mice.

Differential expression of antibody dependent cellular cytotoxicity (ADCC) effectors was studied in normal Balb/cCrgI mice and those bearing a chemically induced 7, 12 dimethylbenzanthracene mammary adenocarcinoma. Depletion of macrophages from normal mouse splenocytes by Sephadex G-10 columns resulted in elimination of ADCC. Further separation of the normal G-10 nonadherent splenocytes on nylon wool columns did not result in any population with significant cytotoxicity. However, Balb/c mice bearing mammary tumors showed enhanced levels of ADCC which were not eliminated by macrophage removal. Lymphocytes from tumor bearers further separated on nylon wool yielded nonadherent and adherent populations both capable of effecting significant ADCC. Treatment of the nylon nonadherent cells of both normal and tumor bearing mice with anti-asialo GM1 (AGM1) and complement decreased the ADCC responses. The same treatment only marginally affected cytotoxic levels of nylon adherent cells from tumor bearers, indicating that these effectors are primarily of non-NK lineage. In addition, G-10 nonadherent, nylon adherent cells from tumor bearers separated on a fluorescence activated cell sorter based on the presence of surface immunoglobulins (slg) revealed that both the slg- and slg+ (98% pure) sorted cells were capable of functioning in ADCC. To determine whether in the tumor mice the 2% of slg- cells present in the slg+ sorted population were the ADCC effectors, mixing experiments were done in which up to 10% of slg- cells from tumor bearers were added to nylon adherent cells from normal mice. No significant increases in ADCC levels were found over that of normal mice. These experiments indicate that the 2% slg- cells were not the ADCC effectors nor were they inducing normal B cells to exert this type of cytotoxic reaction in vitro. To further substantiate the B cell lineage of the slg+ ADCC effectors, surface immunoglobulins were removed with protease treatment. After a 36 hr incubation, 92% of the cells had regenerated their slg. The results presented in this paper demonstrate that various splenic lymphoreticular populations from tumor bearers possess an enhanced cytolytic activity against antibody coated target cells. Among these is a unique nylon adherent slg+ cell that is capable of functioning as an ADCC effector.

Noninvasive imaging of c-myc oncogene messenger RNA with indium-111-antisense probes in a mammary tumor-bearing mouse model.

UNLABELLED: The c-myc oncogene is amplified in leukemia and solid tumors, thus making the c-myc messenger RNA (mRNA) a suitable target for following the progression of malignancy by noninvasive imaging with radiolabeled antisense pharmaceuticals or radiolabeled antisense oligodeoxynucleotide (RASON) probes. Considering the higher stability of phosphorothioate over phosphodiester, the probe stability and tumor localization was compared with both derivatives. METHODS: The 15-mer oligonucleotide sequence was synthesized, aminolinked [sense and antisense phosphodiester (O) and monothioester (S)] and coupled with diethylenetriamine pentaacetate (DTPA)-isothiocyanate and aliquots were lyophilized to make a DTPAAHON kit. The radionuclide 111In was chelated to DTPAAHON derivatives, and free 111In was separated by gel filtration. The radiolabeled antisense and sense probes were injected intravenously in mammary tumor-bearing BALB/c mice (1 x 10(6) cells, 8 days postinoculation). RESULTS: The highest uptake was observed at 2 hr with both thio and oxo derivatives of RASON probes, and small tumors could be imaged noninvasively. Tumor uptake and tumor/blood and tumor/muscle ratios for the sense probe (control) were significantly lower (p < 0.001) than those of the antisense probe. CONCLUSION: The radiolabeled antisense probe may provide a new sensitive tool for noninvasive imaging of c-myc oncogene mRNA for a variety of malignant tumors at an earlier stage.