In Vitro Induction of Human Regulatory T Cells Using Conditions of Low Tryptophan Plus Kynurenines.

Thymic regulatory T cells (tTregs) and induced regulatory T cells (iTregs) suppress murine acute graft-versus-host disease (GVHD). Previously, we demonstrated that the plasmacytoid dendritic cell indoleamine 2,3-dioxygenase (IDO) fosters the in vitro development of human iTregs via tryptophan depletion and kynurenine (Kyn) metabolites. We now show that stimulation of naïve CD4+ T cells in low tryptophan (low Trp) plus Kyn supports human iTreg generation. In vitro, low Trp + Kyn iTregs and tTregs potently suppress T effector cell proliferation equivalently but are phenotypically distinct. Compared with tTregs or T effector cells, bioenergetics profiling reveals that low Trp + Kyn iTregs have increased basal glycolysis and oxidative phosphorylation and use glutaminolysis as an energy source. Low Trp + Kyn iTreg viability was reliant on interleukin (IL)-2 in vitro. Although in vivo IL-2 administration increased low Trp + Kyn iTreg persistence on adoptive transfer into immunodeficient mice given peripheral blood mononuclear cells to induce GVHD, IL-2-supported iTregs did not improve recipient survival. We conclude that low Trp + Kyn create suppressive iTregs that have high metabolic needs that will need to be addressed before clinical translation.

An essential role of Ffar2 (Gpr43) in dietary fibre-mediated promotion of healthy composition of gut microbiota and suppression of intestinal carcinogenesis.

Composition of the gut microbiota has profound effects on intestinal carcinogenesis. Diet and host genetics play critical roles in shaping the composition of gut microbiota. Whether diet and host genes interact with each other to bring specific changes in gut microbiota that affect intestinal carcinogenesis is unknown. Ability of dietary fibre to specifically increase beneficial gut microbiota at the expense of pathogenic bacteria in vivo via unknown mechanism is an important process that suppresses intestinal inflammation and carcinogenesis. Free fatty acid receptor 2 (FFAR2 or GPR43) is a receptor for short-chain fatty acids (acetate, propionate and butyrate), metabolites of dietary fibre fermentation by gut microbiota. Here, we show FFAR2 is down modulated in human colon cancers than matched adjacent healthy tissue. Consistent with this, Ffar2(-/-) mice are hypersusceptible to development of intestinal carcinogenesis. Dietary fibre suppressed colon carcinogenesis in an Ffar2-dependent manner. Ffar2 played an essential role in dietary fibre-mediated promotion of beneficial gut microbiota, Bifidobacterium species (spp) and suppression of Helicobacter hepaticus and Prevotellaceae. Moreover, numbers of Bifidobacterium is reduced, whereas those of Prevotellaceae are increased in human colon cancers than matched adjacent normal tissue. Administration of Bifidobacterium mitigated intestinal inflammation and carcinogenesis in Ffar2(-/-) mice. Taken together, these findings suggest that interplay between dietary fibre and Ffar2 play a key role in promoting healthy composition of gut microbiota that stimulates intestinal health.

Indoleamine 2,3-dioxygenase, Tregs and cancer.

The IDO pathway is implicated in a number of settings which lead to acquired peripheral tolerance. One such setting may be the functional tolerance displayed by tumor-bearing hosts toward tumor-associated antigens. Foxp3(+) Tregs are now recognized as a major contributor to tumor-induced immune suppression and functional tolerance. Emerging evidence links the IDO pathway with Treg biology at several points. The first is the ability of IDO-expressing DCs to drive the differentiation of naive CD4(+) T cells toward a Foxp3+ (inducible Treg) phenotype. The second link is the ability of IDO-expressing DCs to directly activate mature, pre-existing Tregs for markedly enhanced suppression of target cells. And the third link is the ability of IDO to prevent the inflammation-induced conversion ("reprogramming") of Tregs into pro-inflammatory T-helper-like cells in vivo. Taken together, these findings suggest that IDO may represent an important regulatory checkpoint influencing Treg activity: both by stabilizing and augmenting the suppressive phenotype, and by preventing Treg reprogramming into non-suppressive helper-like cells.

A key in vivo antitumor mechanism of action of natural product-based brassinins is inhibition of indoleamine 2,3-dioxygenase.

Agents that interfere with tumoral immune tolerance may be useful to prevent or treat cancer. Brassinin is a phytoalexin, a class of natural products derived from plants that includes the widely known compound resveratrol. Brassinin has been demonstrated to have chemopreventive activity in preclinical models but the mechanisms underlying its anticancer properties are unknown. Here, we show that brassinin and a synthetic derivative 5-bromo-brassinin (5-Br-brassinin) are bioavailable inhibitors of indoleamine 2,3-dioxygenase (IDO), a pro-toleragenic enzyme that drives immune escape in cancer. Like other known IDO inhibitors, both of these compounds combined with chemotherapy to elicit regression of autochthonous mammary gland tumors in MMTV-Neu mice. Furthermore, growth of highly aggressive melanoma isograft tumors was suppressed by single agent treatment with 5-Br-brassinin. This response to treatment was lost in athymic mice, indicating a requirement for active host T-cell immunity, and in IDO-null knockout mice, providing direct genetic evidence that IDO inhibition is essential to the antitumor mechanism of action of 5-Br-brassinin. The natural product brassinin thus provides the structural basis for a new class of compounds with in vivo anticancer activity that is mediated through the inhibition of IDO.

Tryptophan catabolism prevents maternal T cells from activating lethal anti-fetal immune responses.

The murine conceptus is protected from maternal immunity by cells expressing indoleamine dioxygenase (IDO), which catabolizes tryptophan. Induction of lethal maternal anti-fetal immunity requires effective pharmacologic inhibition of IDO enzyme activity and the presence of maternal T cells, but not B cells and also depends on the degree of maternal-fetal tissue incompatibility. Based on these findings, we propose a model to explain the role of IDO in suppressing maternal immunity and the mechanism of fetal allograft rejection, when IDO activity is inhibited during gestation. This model incorporates observations that fetal allograft rejection is T cell dependent, antibody-independent and is accompanied by a novel type of inflammation involving extensive complement deposition at the maternal-fetal interface, when IDO activity is blocked during murine pregnancy.

Extinguishing maternal immune responses during pregnancy: implications for immunosuppression.

Mammals owe their existence to immunosuppressive processes that prevent fetal rejection in utero. Blocking tryptophan catabolism during murine pregnancy allows maternal T cells to provoke fetal allograft rejection. Cells expressing indoleamine 2,3-dioxygenase (IDO), which catabolizes tryptophan, prevent T cell cycle progression and enhance activation induced T cell death. Here, we discuss the role of cells expressing IDO in regulating maternal T cell immunity during pregnancy and consider whether this mechanism might contribute to immunological discrimination by promoting T cell tolerance in other circumstances.

Prevention of T cell-driven complement activation and inflammation by tryptophan catabolism during pregnancy.

Indoleamine 2,3 dioxygenase (IDO) activity during pregnancy protects developing fetuses from maternal immune responses in CBA mice. We show here that fetal allografts were rejected only in mating combinations where paternally inherited tissue antigens elicited potent maternal T cell responses after exposure to IDO inhibitor. IDO inhibitor treatment triggered extensive inflammation at the maternal-fetal interface in susceptible mating combinations, which was characterized by complement deposition and hemorrhagic necrosis. Identical inflammatory responses occurred in B cell-deficient (RAG-I-/-) mothers that carried a monoclonal cohort of CD8+ T cells specific for a single paternally inherited fetal major histocompatibility complex antigen. Thus, fetal allograft rejection was accompanied by a unique form of inflammation that was characterized by T cell-dependent, antibody-independent activation of complement. In contrast, no inflammation, complement deposition or T cell infiltration was elicited when mice carrying syngeneic fetuses were exposed to IDO inhibitor. These data show that IDO activity protects the fetus by suppressing T cell-driven local inflammatory responses to fetal alloantigens.

Maturation of antigen-presenting cells is compromised in HLA-G transgenic mice.

The human MHC class Ib antigen HLA-G is thought to regulate maternal immune responses during pregnancy. Here we show that expression of HLA-G in transgenic mice diminished cellular immunity by inhibiting maturation of myelomonocytic cells into functional antigen-presenting cells (APC). Skin allografts applied to HLA-G transgenic mice survived longer and resultant T cell responses were less potent compared to control mice. T cells from HLA-G mice responded normally to allogeneic APC and immunohistological analyses of spleen revealed no marked abnormalities. However, spontaneous outgrowths of myeloid cells were observed when bone marrow or splenocytes from HLA-G mice were cultured in vitro, but functionally competent APC did not develop spontaneously in bone marrow cultures supplemented with granulocyte macrophage colony stimulating factor (GM-CSF). Addition of lipopolysaccharide (LPS) to GM-CSF-derived bone marrow cultures rescued APC maturation. Studies using HLA-G tetrameric reagents revealed that HLA-G-specific binding activity was associated with CD11c(+) myelomonocytic cells, while binding to lymphoid and NK cell subsets was undetectable. These data show that spontaneous maturation of functionally competent dendritic cells (DC) is compromised in HLA-G mice. We hypothesize that HLA-G inhibits maturation of DC via receptor-mediated interactions with myelomonocytic precursors, which render immature DC precursors unable to receive signals from activated T cells.

Immunology at the maternal-fetal interface: lessons for T cell tolerance and suppression.

Mammalian reproduction poses an immunological paradox because fetal alloantigens encoded by genes inherited from the father should provoke responses by maternal T cells leading to fetal loss. Current understanding of T cell immunobiology and the critical role of inflammatory processes during pregnancy is reviewed and discussed. Lessons derived from studies on the regulation of T cell responsiveness during mammalian gestation are considered in the wider context of T cell tolerance toward some microbial infections and tumors, avoidance of autoimmunity, and tissue allograft rejection.

Tryptophan catabolism and T-cell tolerance: immunosuppression by starvation?

Some macrophages inhibit microbial infections by producing indoleamine 2,3 dioxygenase (IDO), which catabolizes tryptophan. Here, Andrew Mellor and David Munn discuss evidence that cells that synthesize IDO protect the mammalian fetus from maternal T-cell attack and argue that this mechanism might have wider implications for the control of T-cell responses.

Inhibition of T cell proliferation by macrophage tryptophan catabolism.

We have recently shown that expression of the enzyme indoleamine 2, 3-dioxygenase (IDO) during murine pregnancy is required to prevent rejection of the allogeneic fetus by maternal T cells. In addition to their role in pregnancy, IDO-expressing cells are widely distributed in primary and secondary lymphoid organs. Here we show that monocytes that have differentiated under the influence of macrophage colony-stimulating factor acquire the ability to suppress T cell proliferation in vitro via rapid and selective degradation of tryptophan by IDO. IDO was induced in macrophages by a synergistic combination of the T cell-derived signals IFN-gamma and CD40-ligand. Inhibition of IDO with the 1-methyl analogue of tryptophan prevented macrophage-mediated suppression. Purified T cells activated under tryptophan-deficient conditions were able to synthesize protein, enter the cell cycle, and progress normally through the initial stages of G1, including upregulation of IL-2 receptor and synthesis of IL-2. However, in the absence of tryptophan, cell cycle progression halted at a mid-G1 arrest point. Restoration of tryptophan to arrested cells was not sufficient to allow further cell cycle progression nor was costimulation via CD28. T cells could exit the arrested state only if a second round of T cell receptor signaling was provided in the presence of tryptophan. These data reveal a novel mechanism by which antigen-presenting cells can regulate T cell activation via tryptophan catabolism. We speculate that expression of IDO by certain antigen presenting cells in vivo allows them to suppress unwanted T cell responses.

Macrophage suppression of T cell activation: a potential mechanism of peripheral tolerance.

The mechanisms of induction and maintenance of tolerance in self-reactive T cells in the periphery are poorly understood. Current models assume that successful T cell activation only occurs if ligation of the T cell receptor (signal 1) by antigen presenting cells (APCs) is accompanied by a costimulatory signal (signal 2), and that signal 1 in the absence of signal 2 is either ignored or is tolerizing. However, there is also evidence for the existence of macrophages (M phi) capable of suppressing T cell activation both in vitro and in vivo. The possibility of a more actively induced tolerance exists, in which the M phi itself responds to T cell-mediated signals in a tolerogenic fashion. This would help to resolve the paradox that tissue M phi, which act as scavengers of self-antigen, can also act as professional APCs. The ability of tissue macrophages to actively suppress T cells would further underscore the importance of the innate immune system in regulating adaptive immune responses.

Prevention of allogeneic fetal rejection by tryptophan catabolism.

In 1953 Medawar pointed out that survival of the genetically disparate (allogeneic) mammalian conceptus contradicts the laws of tissue transplantation. Rapid T cell-induced rejection of all allogeneic concepti occurred when pregnant mice were treated with a pharmacologic inhibitor of indoleamine 2,3-dioxygenase (IDO), a tryptophan-catabolizing enzyme expressed by trophoblasts and macrophages. Thus, by catabolizing tryptophan, the mammalian conceptus suppresses T cell activity and defends itself against rejection.

Aortic endothelial cells regulate proliferation of human monocytes in vitro via a mechanism synergistic with macrophage colony-stimulating factor. Convergence at the cyclin E/p27(Kip1) regulatory checkpoint.

Monocyte-derived macrophages (Mphis) are pivotal participants in the pathogenesis of atherosclerosis. Evidence from both animal and human plaques indicates that local proliferation may contribute to accumulation of lesion Mphis, and the major Mphi growth factor, macrophage colony stimulating factor (MCSF), is present in atherosclerotic plaques. However, most in vitro studies have failed to demonstrate that human monocytes/Mphis possess significant proliferative capacity. We now report that, although human monocytes cultured in isolation showed only limited MCSF-induced proliferation, monocytes cocultured with aortic endothelial cells at identical MCSF concentrations underwent enhanced (up to 40-fold) and prolonged (21 d) proliferation. In contrast with monocytes in isolation, this was optimal at low seeding densities, required endothelial cell contact, and could not be reproduced by coculture with smooth muscle cells. Intimal Mphi isolated from human aortas likewise showed endothelial cell contact-dependent, MCSF-induced proliferation. Consistent with a two-signal mechanism governing Mphi proliferation, the cell cycle regulatory protein, cyclin E, was rapidly upregulated by endothelial cell contact in an MCSFindependent fashion, but MCSF was required for successful downregulation of the cell cycle inhibitory protein p27(Kip1) before cell cycling. Thus endothelial cells and MCSF differentially and synergistically regulate two Mphi genes critical for progression through the cell cycle.

Monocyte-induced downregulation of nitric oxide synthase in cultured aortic endothelial cells.

Since endothelium-dependent vasodilation is altered in atherosclerosis and enhanced monocyte/endothelial interactions are implicated in early atherosclerosis, we evaluated the effects of monocytes on the endothelial nitric oxide (NO) pathway by estimating release of biologically active NO from cultured endothelial cells and levels of constitutive NO synthase (ecNOS). NO release was estimated in a short-term bioassay using endothelial cell-induced cGMP accumulation in vascular smooth muscle (SM) cells. Exposure of SM cells to porcine aortic endothelial cells (PAECs) and human aortic endothelial cells (HAECs) produced large increases in SM cGMP content; this increase was prevented by NG-nitro-L-arginine methyl ester, the inhibitor of endothelial NOS. Confluent monolayers of PAECs and HAECs cocultured with monocytes also stimulated SM cGMP formation; however, NO release from these cultures was attenuated in a coculture time (2 to 48 hours)- and monocyte concentration (20 to 200 x 10(3) per well)-dependent manner. This effect of monocyte adhesion appeared to be selective for NO release since other biochemical pathways, such as atriopeptin-and isoproterenol-induced cyclic nucleotide accumulation within the endothelial cells, were not altered by monocytes. The effects of adherent monocytes on NO release were mimicked by monocyte-derived cytokines tumor necrosis factor (TNF)-alpha and interleukin (IL)-1 alpha. Furthermore, the conditioned medium of monocytes contained significant quantities of these cytokines. Conditioned medium, as well as monocytes physically separated from the endothelial cells, attenuated NO release, suggesting that soluble factors may mediate the effects of monocytes. An IL-1 beta neutralizing antibody fully prevented the NO dysfunction in response to directly adherent monocytes. Superoxide dismutase, catalase, 4,5-dihydroxy-1,3-benzene disulfonic acid (Tiron), and exogenous L-arginine failed to improve NO release, suggesting that oxidant stress-induced inactivation of NO or limited substrate availability were not primarily responsible for the inhibiting effects of monocytes. Western blot analysis revealed reduced quantities of ecNOS in monocyte/endothelium cocultures, as well as in HAECs treated with monocyte-conditioned medium or TNF-alpha. Thus, adhesion of monocytes to endothelial cells and monocyte-derived secretory products downregulate steady state levels of ecNOS, an event associated with attenuated release of biologically active NO. This mechanism may potentially contribute to diminished endothelium-dependent and NO-mediated vasodilation in early atherosclerosis.

Monocyte- and cytokine-induced downregulation of angiotensin-converting enzyme in cultured human and porcine endothelial cells.

We investigated the effects of monocytes on endothelial cell (EC) ectoenzyme activity. Coculture of human aortic ECs with human monocytes (2 x 10(5) monocytes per 2-cm2 well) led to a decrease in EC angiotensin-converting enzyme (ACE) activity (64.5 +/- 3.5% of control) but not aminopeptidase N, aminopeptidase P, and 5'-nucleotidase activities. Similar results were obtained using human umbilical vein EC-human monocyte and porcine aortic EC-porcine monocyte cocultures. The decrease in ACE activity was monocyte concentration and coculture time dependent, reaching a maximum of 65% decrease in activity at 120 hours. Monocyte-mediated reduction in ACE activity did not require cell to cell contact, since exposure of ECs to conditioned medium from cocultures (CCCM) or from monocyte cultures (MCM) produced a decrease in ACE activity similar to that observed in EC-monocyte cocultures. Exogenously added tumor necrosis factor (TNF)-alpha and interleukin (IL)-1 alpha, two known secretory products of monocytes, simulated the effects of monocytes on ACE activity. Western blot analysis revealed a decrease in the amount of ACE protein in TNF-alpha-treated and CCCM-treated ECs compared with control ECs. Both TNF-alpha and IL-1 alpha were present in CCCM and MCM but not EC-conditioned medium. Incubation of the cocultures with a mixture of neutralizing antibodies against TNF-alpha and IL-1 totally abolished the monocyte-induced decrease in ACE activity. In conclusion, monocytes decrease ACE activity in cultured ECs through the release of cytokines such as TNF-alpha and IL-1.

Recombinant human macrophage colony-stimulating factor in nonhuman primates: selective expansion of a CD16+ monocyte subset with phenotypic similarity to primate natural killer cells.

The CD16 receptor (Fc gamma R-III) is found on many tissue macrophages (M phi s), but its expression on circulating monocytes is restricted to a small, phenotypically distinct subset. The number of these CD16+ monocytes may be markedly increased in response to sepsis, human immunodeficiency virus infection, or metastatic malignancy. We have recently shown that the CD16+ monocyte population is selectively expanded by administration of recombinant human macrophage colony-stimulating factor (rhM-CSF). In the current study, we used the highly rhM-CSF-responsive cynomolgus primate model to further characterize this novel monocyte population. Animals treated with rhM-CSF underwent a progressive and essentially complete conversion to the CD16+ monocyte phenotype, with up to a 50-fold increase in the number of CD16+ cells. This increase was paralleled by the emergence of a population of circulating cells that morphologically resembled large granular lymphocytes (LGLs). However, quantitatively, this population corresponded closely to the number of CD16+ monocytes, and fluorescence-activated cell sorting (FACS) confirmed that they were the same. In addition to their LGL-like morphology, many rhM-CSF-induced CD16+ monocytes showed a pattern of size, granularity, and quantitative cell surface marker expression that closely resembled the pretreatment LGL/natural killer (NK) cell population but that did not resemble the pretreatment monocyte population. However, rhM-CSF-induced CD16+ monocytes could be distinguished from LGL/ NK cells by fact that they all expressed cell surface receptors for rhM-CSF, and many of them showed reduced but detectable phagocytic and respiratory burst activity. Studies of human subjects treated with rhM-CSF also showed an analogous population of "LGL-appearing" CD16+ mononuclear cells. Thus, our studies reveal a previously unsuspected ability of cells in the monocyte lineage to adopt a phenotype similar to that of LGL/NK cells. The extent of this phenotypic convergence suggests that the two lineages retain access to elements of a similar developmental pathway.

Selective activation-induced apoptosis of peripheral T cells imposed by macrophages. A potential mechanism of antigen-specific peripheral lymphocyte deletion.

The self-reactive T cells that escape clonal deletion in the thymus must be suppressed by the less well characterized process of peripheral tolerance. In this study, we show that monocyte-derived macrophages (M phi) undergoing terminal differentiation in the presence of macrophage CSF (M-CSF) acquire the ability to selectively induce apoptosis of T cells in an activation-specific fashion. Lymphocytes were stimulated via the TCR using anti-CD3 cross-linking, staphylococcal superantigen, or allogeneic mixed-leukocyte cultures. T cells activated while in contact with M-CSF-derived M phi exited the resting G0 state and re-entered the cell cycle, but experienced a sustained arrest near the first G1/S transition, followed by progressive apoptosis. In contrast, lymphocytes that were not stimulated remained viable, and could later activate normally when removed from contact with M phi. Functionally, exposure of T cells to alloantigens presented by M-CSF-derived M phi resulted in a selective depletion of the alloresponsive T cell population, while preserving reactivity to other mitogens and to alloantigens from different donors. The ability of M phi to impose activation-induced apoptosis on lymphocytes was regulated developmentally, being absent in fresh monocytes, progressively acquired during differentiation in M-CSF, and abrogated if monocytes were exposed to IFN-gamma before differentiation. We speculate that this novel interaction may help to selectively delete autoreactive T cells that respond to self Ags presented by noninflammatory tissue M phi.

Phase II trial of murine monoclonal antibody D612 combined with recombinant human monocyte colony-stimulating factor (rhM-CSF) in patients with metastatic gastrointestinal cancer.

In a Phase II study, 14 patients with metastatic gastrointestinal cancer received the mAb D612 (40 mg/m2, days 4, 7, and 11) in combination with recombinant human monocyte colony-stimulating factor [(rhM-CSF) 80 micrograms/kg/days 1-14]. The combined treatment was well tolerated and resulted in characteristic biological activity associated with each of the agents. Thus, 10 of 14 patients experienced D612-associated secretory diarrhea, which responded to the prostaglandin inhibitor Indomethacin in 5 of 7 patients. rhM-CSF therapy was associated with peripheral monocytosis (peak absolute monocyte count, 1444 +/- 394/mm3) and thrombocytopenia (nadir count, 78 +/- 10/mm3). Monocyte surface marker analysis revealed a high baseline expression of CD16+ cells in our patient population with an additional increase with rhM-CSF therapy. We observed a correlation between the degree of thrombocytopenia and the pretreatment CD16+ monocyte count. Of the plasma cytokines assayed, serum Neopterin demonstrated the most consistent increase during rhM-CSF therapy. There was a significant difference in the half-life of the first and last dose of D612 (35.8 +/- 2 versus 27 +/- 2.9 h; P < 0.05). Eleven of fourteen patients developed low-moderate levels of anti-D612 antibody. Despite the observed biological activity of both rhM-CSF and D612 and the previously described in vitro synergy, no clinical antitumor responses were observed in this Phase II study.

Antibody-independent phagocytosis of tumor cells by human monocyte-derived macrophages cultured in recombinant macrophage colony-stimulating factor.

Human monocytes exposed in vitro to recombinant macrophage-colony-stimulating factor (rhMCSF) differentiate into monocyte-derived macrophages (MDM), which mediate efficient antibody-dependent cytotoxicity (ADCC) against tumor cells. We and others have shown that this form of ADCC is unusual in that phagocytosis, rather than extracellular lysis, appears to play the major role in target cell killing. In this study, we asked whether the phagocytic form of cytotoxicity seen with ADCC could occur in the absence of an opsonizing antibody. We now report that, whereas cell lines derived from solid tumors are often resistant to antibody-independent cytotoxicity, malignant cells of lymphoid origin appear particularly susceptible to such antibody-independent killing. We found that all of nine lymphocytic leukemia and lymphoma cell lines tested in a total of 35 experiments, plus all four samples of fresh leukemic blasts, were consistently susceptible to antibody-independent MDM cytotoxicity. Antibody-independent cytotoxicity against these cells was efficient (40%-63% killing) at effector: target (E:T) ratios as low as 2:1. Like ADCC, antibody-independent cytotoxicity involved phagocytosis of target cells, as demonstrated by ingestion of fluorescently labeled targets and analysis by flow cytometry. At the time of phagocytosis, the majority of target cells retained membrane integrity, as indicated by the direct transfer of intracellular [51Cr]chromate from radiolabeled targets to phagocytosing MDM, without release of the label into the medium. However, in contrast to ADCC, we found that the degree of antibody-independent cytotoxicity was not a function of the E:T ratio. Instead, a constant proportion of the available target cells were killed regardless of the E:T ratio, suggesting that target cell recognition, rather than effector cell potency, might be the limiting factor in determining cytotoxicity. In additional experiments, we have also identified a second tumor cell type, nueroblastoma, as being susceptible to antibody-independent phagocytosis (all of five cell lines tested, cytotoxicity 40%-93%, E:T = 3:1). Our data thus indicate that the cytotoxicity induced by rhMCSF is not confined to antibody-mediated killing, and that phagocytosis can play a significant role in target cell destruction even in the absence of opsonizing antibody.