Inhibition of macrophage tumoricidal activity by glucocorticoids

In this study, the effect of corticosteroids on the activation of macrophages to a fully tumoricidal state was examined. Thioglycolate-elicited peritoneal exudate macrophages from C3H/HeJ mice were rendered cytolytic for P815 mastocytoma cells in a two-signal tumoricidal assay that used recombinant interferon-gamma (rIFN-gamma; 1 to 10 U/ml) as a "priming" signal and butanol-extracted lipopolysaccharide (But-LPS; 0.1 to 5 micrograms/ml) as a "trigger" signal. Treatment of macrophages with either rIFN-gamma alone or But-LPS alone failed to result in significant cytolytic ability. Tumoricidal activity was markedly inhibited in a dose-dependent fashion when glucocorticoids were added simultaneously to the cultures with rIFN-gamma and But-LPS at concentrations ranging from 1 X 10(-10) to 1 X 10(-5) M. Nonglucocorticoid sex hormones failed to inhibit tumoricidal activity in this system under identical culture conditions. Inhibition was most effective if the glucocorticoids were added simultaneously with the priming and triggering signals (rIFN-gamma and But-LPS); however, if the glucocorticoids were added 24 hr after the signals were provided to the cultures, suboptimal inhibition was observed. Experiments that dissociated the priming phase of activation from the triggering phase showed that glucocorticoids inhibited both the rIFN-gamma-induced priming stage as well as the But-LPS-induced triggering stage of activation. These observations provide evidence that glucocorticoids, but not other steroid hormones, inhibit the activation of macrophages to a fully tumoricidal state by interfering with either the priming or triggering signals in this two-signal model of macrophage activation.     

In vitro modulation of macrophage tumoricidal activity

Summary NaIO₄ treatment of mouse adherent peritoneal cells or lymphocyte-free cloned macrophages enhances their cytotoxic and tumoricidal activity. 5×10^–3 M NaIO₄ treatment of nontumoricidal BCG-activated macrophages renders them completely tumoricidal, whereas the same treatment of stimulated (peptone-normal) macrophages renders them weakly tumoricidal. Addition of LPS in nanogram quantities too low to enhance tumor cell killing by untreated peptone-normal macrophages causes NaIO₄-treated peptone-normal macrophages to be maximally tumoricidal. The activating action of NaIO₄, MAF, or LPS can be potently, but inconsistently, blocked or reversed by the reducing agent NaBH₄ or the aldehyde-reacting agent dimedone. NaIO₄ treatment of lymphocyte-free macrophage colonies does not make them cytotoxic, but NaIO₄-treated colony macrophages are cytotoxic for tumor cells when cultured in 10 ng/ml LPS (an amount of LPS inadequate to render untreated colony macrophages cytotoxic). Supernatants of NaIO₄-treated adherent peritoneal cells contain MAF activity. Thus, the NaIO₄-induced enhancement of peritoneal cell tumoricidal activity may result from both direct NaIO₄ activating effects on macrophages and indirect NaIO₄ effects through NaIO₄-induced MAF production.     

Relationships between suppressor macrophages and macrophage precursors in the spleens from tumor-bearing mice

Suppressor macrophages (Mφ) which can inhibit mitogen-induced lymphocyte proliferation appeared in the spleens of mice bearing transplanted MC-A fibrosarcoma cells. An analysis of the ontogeny of such Mφ revealed additional suppressor activity directed against macrophage stem cells. Treatment of spleen cell suspensions with carbonyl iron followed by centrifugation removed suppressor Mφ but did not deplete Mφ-colony forming cells (M-CFC) which could be demonstrated in soft agar culture in L-cell conditioned medium (LCM). Untreated spleen cells had normal numbers of M-CFC; phagocyte-depleted mononuclear cells showed a threefold increase in M-CFC 14 days after subcutaneous inoculation of 10⁶ MC-A cells per mouse. Further increases in M-CFC were also evident in similar preparations on Days 21 and 28 when the M-CFC concentration reached a maximum of eight times the normal level. The Mφ which developed from the M-CFC grown in the presence of LCM were later shown to have indomethacin-sensitive suppressor activity suggesting the mediation of this phenomenon by prostaglandins. These observations suggest that locally produced phagocytic suppressor Mφ from the spleens of tumorbearing mice play important roles not only as inhibitors of lymphocyte proliferation as reported earlier, but also as regulators of monocyte-Mφ production.     

Effect of L-arginine on the retention of macrophage tumoricidal activity

It has been reported that the tumoricidal activity of macrophages (M phi) depends on L-arginine and that L-arginine metabolites such as reactive nitrogen intermediates alter M phi physical capacities. The aim of this report is to investigate the dose-related effect of L-arginine on the expression and retention of M phi tumoricidal activity. Cytotoxicity of M phi activated by IFN-gamma plus LPS was detected in the presence of about 0.1 mM or more of L-arginine. This paralleled the NO2- production in the presence, but not in the absence, of L-arginine. On the other hand, activated M phi were destined to die and lost their tumoricidal activity with time in the presence of 0.3 mM or more L-arginine. They retained, however, considerable activity in the absence or presence of 0.15 mM L-arginine. This retention of M phi cytotoxicity was longer when M phi were preactivated by 100 ng/ml than 10 ng/ml of LPS in combination with IFN-gamma. Addition of indomethacin, an inhibitor of prostaglandin production, did not prevent the decay of M phi cytotoxicity but rather facilitated it even in the absence of L-arginine. Regardless of indomethacin, consecutive stimulation with LPS or LPS plus IFN-gamma during culture was effective in maintaining the tumoricidal activity at a high level. In addition, we found that M phi which had lost tumoricidal activity during culture in L-arginine deficient medium could be reactivated by LPS to attack tumor target cells.     

Depressed T-cell reactivity and suppressor activity of lymphoid cells from cyclophosphamide-treated mice

A study was made of the in vitro proliferative activity of thymus-derived lymphoid cells from cyclophosphamide-treated mice (Cy-mice) and the relationship between this and some in vivo immunological responses. The proliferative response to phytohaemagglutinin (PHA) and allogeneic cells was depressed for up to 3 weeks after drug treatment in spleen and lymph node cells, responsiveness recovering more rapidly in lymph node cells. Cell concentration in culture was shown to be important in such measurements as cells from some Cy-mice were able to inhibit their own proliferation and that of normal lymph node cells. No stable soluble factor responsible for this effect could be isolated. It was shown that in vitro proliferative activity is not a good indicator of in vivo T-cell capability as indicated by the very rapid recovery of ability to reject skin grafts and the fairly rapid recovery of ability to produce cytotoxic cells compared to the slower recovery of in vitro T-cell activities.     

Lipopolysaccharide-mediated macrophage activation: the role of calcium in the generation of tumoricidal activity

As we have reported, calcium ionophore A23187 activates macrophages for tumor cell killing, and the activated macrophages produced a soluble cytotoxic factor (M phi-CF) that is similar, if not identical, to tumor necrosis factor. Based on these observations, we have investigated whether calcium is involved in the activation mediated by another potent macrophage activator, namely lipopolysaccharide (LPS). We first showed that A23187 caused uptake of extracellular calcium-45 by macrophage monolayers, whereas LPS did not. Because in this system rapid changes would not have been detected, several other approaches also have been used. We have examined the effect of depleting extracellular calcium by using medium containing no added calcium, supplemented with 1 mM EGTA. In no case did depletion result in decreased M phi-CF production by LPS-treated macrophages. Measurements using the fluorescent intracellular calcium indicator Quin 2 have also been performed. The calcium ionophore ionomycin caused a rapid change in the intracellular Quin 2 signal. LPS, even at a concentration in vast excess of that required to activate the macrophages, caused no change in the signal during a 2-hr period. If the macrophages were loaded with high doses of Quin 2 or another intracellular chelator, TMB-8, M phi-CF production decreased and cytotoxic activity was impaired. These data indicate that one or more of the processes involved in M phi-CF production does require calcium, but that activation mediated by LPS occurs without the influx of extracellular calcium or redistribution of intracellular calcium.     

Inhibition of macrophage tumoricidal activity by immune complexes and altered erythrocytes

Engagement of the macrophage membrane by biologic particles including insoluble immune complexes inhibited the development of lymphokine-mediated nonspecific tumoricidal activity by murine macrophages. The degree of inhibition was dependent on the dose of particles and the lymphokine concentration. Inhibition was not due to macrophage cell death or to diminution of cell adherence after ingestion of the immune complexes. Soluble immune complexes were not inhibitory, although approximately 10% of the complexes became cell-associated. Monomeric or heat-aggregated IgG was also not inhibitory. IgG-opsonized erythrocytes (EA) were inhibitory and inhibition was dependent on the degree of opsonization. In contrast, nonopsonized erythrocytes (E), which did not bind to macrophages, were not inhibitory. Phagocytosis of glutaraldehyde-treated E or E carrying IgM antibody and complement (EAC) also led to a reduction of tumorilytic activity. Insoluble immune complexes were inhibitory when added either before or after lymphokine. Phagocytosis was neither sufficient nor necessary to cause inhibition because 1) ingestion of polystyrene latex beads did not diminish tumoricidal activity, and 2) macrophages plated on IgG-coated surfaces were inhibited with respect to the tumoricidal function. Inhibition was not affected when indomethacin (10(-6) M) was included in the assay, which indicated that prostaglandins were not involved in the process. Thus, macrophage tumoricidal responsiveness may be compromised by interaction of biologic substances with macrophage plasma membranes. This process may thereby inactivate an important host defense mechanism against neoplastic cells.     

Dissociation between macrophage tumoricidal capacity and suppressive activity: analysis with macrophage-defective mouse strains

Macrophages (M phi diameter) from three mouse strains with genetically distinct M phi diameter deficits (C3H/HeJ, A/J, and P/J) were unable to develop high cytolytic and cytotoxic activity against tumor cells in vitro when exposed to agents (MAF and IFN-beta) that strongly increased the tumoricidal capacity of M phi diameter from nondefective C3H/HeN mice. Nevertheless, the tumoricidal deficits of M phi diameter from the defective strains did not affect their suppressive capacity on Con A-induced lymphoproliferation, nor their ability to react to IFN-beta by decreasing suppressive activity. In fact, natural suppressive activity and IFN-beta-induced changes in the suppression of M phi diameter from C3H/HeJ, A/J, and P/J mice were highly comparable to those of C3H/HeN M phi diameter, thus stressing the dissociation between the mechanisms governing M phi diameter suppression and M phi diameter tumoricidal activity. Analysis of the modulation by MAF and IFN-beta of M phi diameter ability to release the oxygen metabolites O2- and H2O2, molecules possibly involved in the effector mechanism of both M phi diameter cytotoxicity and suppression, revealed a close correlation with the patterns of suppressive activity in both nondefective and defective strains. In contrast, no correlation between the production of oxygen-reactive species and M phi diameter tumoricidal activity was observed. The ability of MAF- and IFN-beta-treated M phi diameter to produce PGE, a molecule of major importance in M phi diameter-mediated suppression and possibly involved also in the regulation of M phi diameter tumoricidal activity, again paralleled M phi diameter suppressive capacity. Thus, the mechanisms controlling M phi diameter antitumor activity appeared to be clearly distinct from those involved in M phi diameter suppression.     

Functional heterogeneity of macrophage precursor cells from spleen of Leishmania donovani-infected and untreated mice

We recently described the bone marrow-derived macrophage precursor, which is able to spontaneously and extracellularly kill protozoa of the genus Leishmania. These nonadherent, nonphagocytic macrophage precursor cells are present in the spleen of healthy mice only in a small quantity. However, high numbers of proliferating macrophage precursors are isolated from the spleen of Leishmania donovani-infected mice. Macrophage precursors from spleens of diseased animals are able to kill spontaneously the promastigote as well as the amastigote form of L. donovani. The mechanism of the spontaneous leishmanicidal activity of macrophage precursor cells derived from spleens of L. donovani-infected mice was investigated. This effector function could be defined in part as an antibody-dependent cellular cytotoxicity. In addition we assessed the role of CSF-1-containing L cell-conditioned supernatant at the leishmanicidal activity of these immature cells of the macrophage lineage. For that purpose, nonadherent spleen cells from healthy mice were cocultivated with this CSF-1-containing medium for 4 days. These in vitro proliferated macrophage precursor cells from untreated mice showed an increased leishmanicidal activity. Thereby we established a further activation mechanism for proliferating splenic macrophage precursor cells responsible for the observed killing of L. donovani pro- and amastigotes. The spontaneous cytotoxicity of macrophage precursors from spleens of L. donovani-diseased animals is thus defined as a cooperative effect of antibody-dependent cellular cytotoxicity and Macrophage-CSF activation.     

Binding of activated macrophages to tumor cells through a macrophage lectin and its role in macrophage tumoricidal activity

A 45-60 kDa Gal/GalNAc-specific macrophage lectin was found to participate in the interaction between tumor cells and tumoricidal macrophages activated by an antitumor streptococcal preparation, OK-432, and in the tumoricidal activity of the activated macrophages. The binding between OK-432-elicited activated macrophages and murine mastocytoma P-815 cells was inhibited on preincubation of the macrophages with a neoglycoprotein (Gal-BSA) or a complex-type glycopeptide (unit B) which was a specific inhibitor of the macrophage lectin. This binding of the macrophages to P-815 cells was also inhibited on the addition of anti-macrophage lectin antiserum. Contrary to the case of OK-432-elicited macrophages, the binding of thioglycolate-elicited (responsive) macrophages to P-815 cells was inhibited only a little by Gal-BSA and unit B, and not inhibited by the antiserum. Furthermore, the tumoricidal activity of the activated macrophages was inhibited by the addition of the anti-macrophage lectin antiserum. These results suggest that the binding of activated macrophages to tumor cells through the Gal/GalNAc-specific macrophage lectin is an important part of the tumor cell killing mechanism.     

Effects of various inhibitors of arachidonic acid oxygenation on macrophage superoxide release and tumoricidal activity

Macrophages release a variety of arachidonic acid metabolites after treatment with various membrane triggers or particulate stimuli. We examined the role of phospholipase and lipoxygenase inhibitors in the modulation of superoxide production and tumor cytolysis by murine macrophages. Superoxide was induced by the soluble stimulus, phorbol myristate acetate (PMA), and the particulate stimulus, opsonized zymosan, and was measured by the reduction of ferricytochrome c with the use of a micro ELISA reader. Macrophage-mediated tumor cytolysis was induced by hybridoma-derived, macrophage-activating factor (MAF) and was quantitated by 51Cr release from P815 target cells. In both assays, 72-hr peptone-elicited macrophages were used. Dexamethasone, and to a lesser degree hydrocortisone, inhibited superoxide release and MAF-induced tumor cytolysis. Inhibition in the superoxide assay required pretreatment with corticosteroid. Only the gold compound, auranofin, inhibited superoxide when given simultaneously with stimulant. Other phospholipase inhibitors, including mepacrine and 4-bromophenacyl bromide, and several lipoxygenase inhibitors, including BW755c, nordihydroguaiaretic acid (NDGA), and 5,8,11,14-eicosatetraynoic acid (ETYA), failed to modulate either macrophage response at nontoxic concentrations. At the concentrations tested in the tumoricidal and superoxide assays, mepacrine and 4-bromophenacyl bromide inhibited the release of 14C-arachidonic acid from macrophages stimulated with opsonized zymosan. Our data strongly suggest that corticosteroids suppress macrophage superoxide production and tumoricidal function by a nonphospholipase-dependent mechanism.     

Effects of whole body X-irradiation and cyclophosphamide treatment on induction of macrophage tumoricidal function in mice

The influence of whole body X-irradiation (200–800 R) and subcutaneous cyclophosphamide (CY) treatment (150–500 mg/kg) was studied on the ability of adjuvants to induce cytotoxic macrophages in vivo. Surprisingly, radiation or CY therapy alone produced growth inhibitory macrophages whose function peaked within 2 days after treatment. When adjuvants such as Bacillus Calmette Guérin (BCG), pyran copolymer, or glucan were administered ip within 2 hr after sublethal (600 R) X-irradiation, adjuvant-induced cytotoxic function was depressed but not ablated. In addition, when noninduced peritoneal macrophages were obtained 6 days after lethal (800 R) X-irradiation, their ability to be activated in vitro by lymphokine or fibroblast-derived interferon preparations was only slightly depressed at all concentrations of inducer tested. When BCG, pyran, or glucan was administered ip concurrently with sc CY treatment, only the ability of BCG to activate macrophages was markedly reduced, indicating separate mechanisms for the induction of tumoricidal macrophages. A better understanding of the interaction of chemotherapeutic and/or radiation regimens with adjuvants which affect macrophage function may be instrumental to rationalized immunotherapy protocols.     

Prevention of macrophage tumoricidal activity by agents known to increase cellular cyclic AMP.

The concept of T-T cell interaction which was first suggested during cell-mediated immune response to alloantigens was evaluated in a syngeneic tumor system. The combination of lymph node and thymus cells from BALB/c mice immune against syngeneic tumor cells, mKSA, was shown to exhibit collaboration with respect to in vitro generation of effector cells capable of preventing growth of corresponding tumor cells in the tumor cell neutralization assay. While each cell population of either anatomical site did not prevent tumor growth when tested alone, combinations of both did. The antigen specificity of effector cells generated by synergizing cultures was similar to that of effectors derived from cultures containing optimal numbers of responding lymph node cells. The lymph node and thymus cell populations participating in synergy were found to be thymus dependent. These results suggest that we may be dealing with the same or similar T₁- and T₂-cell populations described before as displaying synergy in response to alloantigens in the graft versus host, mixed lymphocyte, and cell-mediated cytotoxicity reactions.     

Heterogeneity of the effectors of natural killer activity in CBA mice

The natural cytotoxicity of Percoll fractions of CBA mice splenocytes was tested against cells of human erythroblastosis suspension culture K-562 and murine C3HA hepatoma solid culture MGXXIIa. The 1.076 g/ml dense fraction was shown to have the highest activity in 3H-uridine cytotoxic test against the cell targets. Immunosera prepared by the Golub method against CBA mouse brains and exhausted supplementary by syngeneic thymocytes decreased the natural cytotoxicity of CBA mouse splenocytes against MGXXIIa tumor cells and, on the contrary, increased it against K-562 tumor cells.     

B cell subsets in spleens of BALB/c mice: identification and isolation of MMTV-expressing and MMTV-responding subpopulations

The DNA of BALB/c mice contains two genomic- and one subgenomic-size MMTV proviruses that appear to be preferentially expressed in their spleen cells, although intact MMTV virions cannot be detected in the tissues of these mice. This retrovirus antigen expression is restricted to a subpopulation of B lymphocytes, as determined by double label immunofluorescence studies. Nylon-adherent, SIg-positive spleen lymphocytes from BALB/c mice are capable of being stimulated by purified MMTV in lymphocyte transformation assays. Two possibilities were explored: the MMTV-positive cells are the responders to MMTV in the blastogenesis assay, or there exists two B lymphocyte subsets, one expressing the MMTV antigen(s) and the other responding to the virus. Depletion of MMTV-positive, nylon-adherent cells with anti-MMTV and complement resulted in no significant change in the blastogenesis to MMTV, indicating that the MMTV-negative lymphocytes are the responders in this reaction. These results were confirmed by positive selection experiments by using a fluorescence-activated cell sorter. Two populations of nylon-adherent cells, on the basis of the presence or absence of MMTV antigen in their surfaces, were obtained by a two-way sorting procedure and were used in lymphocyte transformation assays. MMTV-expressing lymphocytes were found to be nonresponsive to MMTV antigens, although high levels of [3H]thymidine incorporation were observed in the MMTV-negative, nylon-adherent cell cultures exposed to MMTV. These data indicate that in normal BALB/c mice, expression of endogeneous retrovirus genetic information is restricted to a nylon-adherent spleen cell subset that is different from the one responding in blastogenesis to the viral antigens.     

The macrophage response to bacteria: flow of L-arginine through the nitric oxide and urea pathways and induction of tumoricidal activity

The consequences of the interaction of heat-killed bacteria and lipopolysaccharide (LPS) with a pure population of bone marrow-derived mononuclear phagocytes (BMM?) were investigated, utilizing changes in the flow of L-arginine and expression of tumoricidal activity as parameters of macrophage (m?) function. Gram negative bacteria and LPS proved potent in inducing the flow of L-arginine through the nitric oxide and the urea pathways but were mostly poor in eliciting tumoricidal activity. Gram positive bacteria affected the metabolism of L-arginine only little but were often efficient in triggering tumoricidal activity. The findings show that the m? response to bacteria, which may determine the outcome of their interaction with the host, may differ considerably depending on the type of bacteria.     

Colony-stimulating factors and regulation of macrophage tumoricidal and microbicidal activities

Conditioned medium from antigen- or mitogen-stimulated spleen cells, lymphokines, contained factors that induced formation of granulocyte and macrophage colonies in cultures of bone marrow cells (CSF). Lymphokines also contained factors that induced macrophage non-specific tumoricidal activity against fibrosarcoma 1023, antibody-dependent tumoricidal activity against lymphoma 18-8, and antimicrobial activities against amastigotes of the protozoan parasite, Leishmania tropica. The factors that regulated macrophage effector functions, however, were different from those that induced colony formation, and could be distinguished from CSF by Sephadex gel chromatography or heat sensitivity. To further analyze a role for CSF in induction of macrophage effector activities, conditioned medium from several nonlymphoid cell sources (L-929, WEHI-3, and endotoxin-treated lung cells) were assayed for CSF activities and capacity to induce tumoricidal and microbicidal activities. Conditioned medium that contained either macrophages CSF (CSF-1) or the factor that induced formation of both macrophage and granulocyte colonies failed to activate macrophages for effector activities against fibrosarcoma 1023, lymphoma 18-8, and L. tropica amastigotes (either resistance to infection or intracellular destruction). These data suggest that CSF has no direct role in activation of macrophages for tumoricidal and microbicidal activities against these targets.     

Poly(A)-containing RNA from the spleens of mice with Chagas' disease triggers in vitro macrophage resistance to Trypanosoma cruzi

Mouse peritoneal macrophages exposed to the RNA from the spleens of mice infected with Trypanosoma cruzi (iRNA) exhibit enhanced resistance to this parasite. The poly(A)-containing iRNA was found to be the active fraction. No such activity was observed in macrophages incubated with RNA from normal mice (nRNA) or with synthetic poly A.     

Mutations in Salmonella typhimurium recovered from livers and spleens of mice

Balb/c mice were inoculated intraperitoneally with TA2662, a smooth derivative of the Salmonella typhimurium Ames tester strain (TA102) which carries the mutable hisG locus on a multicopy plasmid, or TA103, which carries the same hisG gene on the chromosome. The bacteria were recovered at various times from the livers and spleens of the infected mice. Total numbers of bacteria were determined and the mutant frequency was estimated. The frequency of occurrence of histidine prototrophs in experiments using TA2662 was substantially above the frequency found with this strain grown in vitro. The mutant frequencies in experiments using TA103 recovered from mice were also highly significantly increased above background. We did not identify factors which might suggest selection in vivo for histidine prototrophs. There is sufficient histidine in body fluids of the host for the growth of His- bacteria. The His- and His+ derivatives were found to grow equally well in vitro in the presence of amounts of histidine approximating concentrations known to exist in vivo. It is probable that mutations in TA2662 are greatly underestimated, since the hisG-containing plasmid is lost at relatively high frequency during incubation in a variety of conditions.