Non-immunogenic murine hepatocellular carcinoma Hepa1-6 cells expressing the membrane form of macrophage colony stimulating factor are rejected in vivo and lead to CD8+ T-cell immunity against the parental tumor.

Hepatocellular carcinoma is a lethal disease and methods that develop effective cellular-based immunotherapy are needed. We retrovirally transduced non-immunogenic mouse Hepa1-6 hepatoma cells with the gene encoding the membrane form of macrophage colony stimulating factor (mM-CSF). Excess recombinant M-CSF and phagocytosis-inhibiting chemicals blocked macrophage-mediated killing of cloned mM-CSF transfected Hepa1-6 hepatoma cells. Macrophages derived from Hck(-/-)Fgr(-/-) and Lyn(-/-) triple knockout mice, which are incapable of performing phagocytosis, failed to kill the mM-CSF transduced cells. The mM-CSF transfected tumor clones failed to grow when injected into C57BL/6 or C57L/J mice. Splenocytes from these vaccinated mice displayed cytotoxicity against parental Hepa1-6 cells, but not against B16 and CT-26 tumor cells in vitro. Mice that rejected the mM-CSF transfected Hepa1-6 tumor subsequently rejected parental Hepa1-6 cells but not the B16 melanoma cells when rechallenged. Elimination of the CD8+ effector cells by an anti-CD8 antibody and complement treatment prevented the adoptive transfer of anti-Hepa1-6-specific immunity into naive animals. Thus, mM-CSF provides a method of generating effective anti-tumor immune responses by macrophages and cytotoxic T cells against the parental Hepa1-6 cells. Our work suggests that mM-CSF transduced hepatoma cells could be used as a tumor vaccine to stimulate immune responses against hepatocellular carcinoma.

Gene expression and gene therapy in experimental duodenal ulceration.

Gastroduodenal ulceration is still poorly understood and changes in gene expression may provide new mechanistic insights. Previously, we demonstrated that angiogenic growth factors are potent ulcer healing agents, and the synthesis of bFGF, PDGF and VEGF is enhanced early in duodenal ulcer healing. The initial molecular event in duodenal ulceration seems to be the organ-specific early release of ET-1 in the pre-ulcerogenic stages after the administration of duodenal ulcerogen cysteamine in rats. We also briefly review here data from literature indicating a central role of ET-1 in gastroduodenal ulceration. After studying the involvement of immediate early genes (e.g. egr-1, Sp1) in ulcer development, we now investigated expression of other genes in the duodenal mucosa in the early stages of chemically induced duodenal ulceration in rats. Following a brief review of principles of gene expression and gene therapy, we review our preliminary gene expression studies, involving monitoring about 1200 genes which revealed about 160 signals and prominent changes in about 30 genes in the early stages of experimental duodenal ulceration. Cysteamine enhanced ET-B receptor gene expression in 30 min, while transcription factors (MAX, STAT 3) showed increased expression in 12 h. We recently also initiated gene therapy studies to enhance the local synthesis of PDGF and VEGF to accelerate duodenal ulcer healing, using a single dose of naked DNA (ND) or adenoviral (AV) vectors of VEGF and PDGF in rats with cysteamine-induced duodenal ulcers. Gene therapy with ND or AV of VEGF or PDGF significantly accelerated chronic duodenal ulcer healing, and increased levels of VEGF and PDGF were detected by Western blotting and ELISA in duodenal mucosa after both VEGF and PDGF gene therapy. Thus, gene expression studies provide new insights into the molecular mechanisms of duodenal ulceration and VEGF or PDGF gene therapy seems to be a new option to achieve a rapid ulcer healing.

Dexamethasone increases the expression of membrane macrophage colony stimulating factor from retrovirally transduced tumor cells expressing macrophage colony stimulating factor.

Many different tumor cell types (breast, ovarian, glioma, liver and colon) were retrovirally transduced with the human macrophage colony stimulating factor (M-CSF) gene (either the membrane associated form [mM-CSF] or the secreted form [sM-CSF]). These cells were tested for their ability to display increased amounts of mM-CSF in response to dexamethasone. M-CSF-transfected tumor cells expressed additional mM-CSF in response to 18-72 h incubations with 3-15 microg/ml dexamethasone, while non-transfected parental cells were unaffected by this treatment. Increased mM-CSF protein expression on the M-CSF transduced cells was observed by flow cytometry and Western blotting using M-CSF specific antibodies. Northern blot analysis revealed an increase in the mM-CSF specific transcripts within the dexamethasone-treated mM-CSF transduced cells, but this was not seen within the non-transfected tumor cells that were treated with dexamethasone. ICAM-1 expression was unaffected by dexamethasone treatment, indicating that this response is mM-CSF specific. All trans-retinal and 1,25-dihydroxy vitamin D3 compounds that have been reported to induce M-CSF expression failed to increase mM-CSF. When dexamethasone-treated mM-CSF transfected clones were used as target cells for macrophage-mediated cytotoxicity assays, an increased killing with the dexamethasone-treated cells was seen. The macrophage-mediated cytotoxicity of these mM-CSF expressing tumor cells was blocked with excess recombinant M-CSF by saturating M-CSF receptors on the macrophage that is required for this form of tumor cell killing. This work suggests the possibility that dexamethasone may prove useful for vaccination purposes using mM-CSF retrovirally transfected tumor cells.

Membrane macrophage colony-stimulating factor on MADB106 breast cancer cells does not activate cytotoxic macrophages but immunizes rats against breast cancer.

Weakly immunogenic, but highly malignant, rat MADB106 breast cancer cells were retrovirally transduced with the membrane form of macrophage colony-stimulating factor (mM-CSF). The cloned mM-CSF-transfected MADB106 cells physically conjugated with macrophages, but were not killed by the macrophages in 48-h cytotoxicity assays. Macrophages killed the mM-CSF-expressing tumors in the presence of noncytotoxic doses of either taxol or taxol plus cisplatin. This indicated that macrophages bind to the mM-CSF expressed on the tumor cells, but for successful macrophage cytotoxicity to occur against mM-CSF-transduced tumor cells other factors must be present. The mM-CSF-transfected tumor cells were rejected when inoculated subcutaneously into normal rats. Cloned MADB106 tumor cells which expressed high amount of mM-CSF were rejected, while tumor cells that displayed lower levels of mM-CSF grew in 60% of the inoculated rats. The mM-CSF-transfected tumors that grew were smaller and had a greater amount of necrosis, compared to the viral vector tumors. Rats that spontaneously rejected the mM-CSF-transfected MADB106 cells showed rechallenge resistance to unmodified parental MADB106 and R3230Ac breast cancers, but not to the F98 glioma. These observations suggest that breast cancer-specific immunity was induced by the inoculation of mM-CSF-expressing MADB106 tumor cells.

Review article: transcription factors and growth factors in ulcer healing.

This review is focused on recent investigations demonstrating a pharmacological and pathophysiologic role in gastroduodenal ulceration for growth factors such as basic fibroblast growth factor (bFGF), platelet-derived growth factor (PDGF) and vascular endothelial growth factor (VEGF), as well as for transcription factors. Our experiments revealed accelerated healing, without decreased gastric acid secretion, of chronic cysteamine-induced duodenal ulcers in rats treated daily for 3 weeks with intragastric administration of bFGF, PDGF or VEGF. Our recent studies also indicate a pathophysiological role of endogenous growth factors in the natural history of experimental duodenal ulcer development and healing. More recently, we investigated the genetic regulation of these growth factors in experimental duodenal ulceration. Since gene expression is most effectively controlled by transcription factors, proteins that bind to cis-acting elements of DNA and guide the binding of polymerase II to start the transcription of specific mRNA, we tested the hypothesis that the expression of IEGs and their transcription factor products, such as Egr-1 and Sp1, might precede the increased synthesis of bFGF, PDGF and VEGF in duodenal ulcer healing. Indeed, the duodenal ulcerogen cysteamine, but not its nonulcerogen and toxic analogue ethanolamine, rapidly increased duodenal (but not gastric) mucosal levels of ET-1, which was followed by enhanced expression of Egr-1 and a decrease in Sp1 in the preulcerogenic stage of duodenal ulceration. These changes in levels of ET-1 and expression of transcription factors were also accompanied by increased expression of the CDK inhibitor p21. Thus, not only growth factors such as bFGF, PDGF and VEGF, but also transcription factors such as Egr-1 and Sp1 and the cell cycle regulator p21, may play a role in the natural history of experimental duodenal ulceration.

Development of systemic immunity to glioblastoma multiforme using tumor cells genetically engineered to express the membrane-associated isoform of macrophage colony-stimulating factor.

We investigated the ability of Fischer rat T9 glioblastoma cells transduced with cDNA genes for the secreted (s) or membrane-associated (m) isoform of M-CSF to elicit an antitumor response when implanted into syngeneic animals. Intracranial (i.c.) implantation of 1 x 10(5) T9 cells expressing mM-CSF (T9/mM-CSF) resulted in 80% tumor rejection. Electron microscopy of the T9/mM-CSF tumor site, 2-4 days postimplantation, showed marked infiltration by macrophages, many of which were in physical contact with the T9/mM-CSF cells. Animals that rejected T9/mM-CSF cells were resistant to i.c. rechallenge with T9 cells, but not syngeneic MadB106 breast adenocarcinoma cells, suggesting that T9-specific immunity can be generated within the brain via the endogenous APCs. Intracranial injection of parental T9, vector control (T9/LXSN), or T9 cells secreting M-CSF (T9/sM-CSF) was 100% fatal. Subcutaneous injection of 1 x 10(7) T9/sM-CSF, T9/LXSN, or parental T9 cells resulted in progressive tumors. In contrast, T9/mM-CSF cells injected s.c. were destroyed in 7-10 days and animals developed systemic immunity to parental T9 cells. Passive transfer of CD3+ T cells from the spleens of immune rats into naive recipients transferred T9 glioma-specific immunity. In vitro, splenocytes from T9/mM-CSF-immunized rats specifically proliferated in response to various syngeneic glioma stimulator cells. However, only marginal T cell-mediated cytotoxicity was observed by these splenocytes in a CTL assay against T9 target cells, regardless of restimulation with T9 cells. Subcutaneous immunization with viable T9/mM-CSF cells was effective in eradicating i.c. T9 tumors.

Macrophages that kill glioma cells expressing the membrane form of macrophage colony stimulating factor are resistant to prostaglandin E2 and interleukin-10.

Malignant rat T9 glioma cells retrovirally transduced with the membrane form of macrophage colony stimulating factor (mM-CSF) were killed by bone marrow derived macrophages in 24 h cytotoxicity assays. Prostaglandin E2 (PGE) and interleukin-10 (IL10) were tested for their ability to block this tumoricidal reaction. Only at very high nonphysiological concentrations of PGE (10(-5) and 10(-6) M) was this cytotoxicity inhibited. Use of high doses of theophylline, a phosphodiesterase inhibitor, also prevented macrophages from killing the mM-CSF transduced target cells. IL10 did not alter the killing potential of the mM-CSF tumoricidal macrophages, even though IL10 reduced the production of nitric oxide by macrophages in response to tumor necrosis factor and lipopolysaccharide. IL10 enhanced the growth of bone marrow macrophages suggesting that IL10 has a complex role in the regulation of tumoricidal macrophages. Thus, the mM-CSF may be an ideal agent to treat tumors that utilize either of these two immunosuppressive defense mechanisms that may block other forms of treatment.

Macrophages kill T9 glioma tumor cells bearing the membrane isoform of macrophage colony stimulating factor through a phagocytosis-dependent pathway.

Rat T9 glioma cells transfected with the gene for the membrane isoform of macrophage-CSF (mM-CSF) but not for the secreted isoform of M-CSF were directly killed by bone marrow-derived macrophages. Macrophage-mediated cytolysis of the mM-CSF-transfected clone was blocked by using chemical inhibitors of phagocytosis such as iodoacetate, 2-deoxyglucose, gadolinium chloride, and cytochalasin B. In contrast, macrophage-mediated killing of mM-CSF-expressing tumor cells was augmented by the microtubule inhibitor, colchicine. Use of nitric oxide and reactive oxygen intermediate inhibitors failed to alter the macrophage-mediated killing of the mM-CSF-transfected tumor cells. Photomicroscopy, using immunohistochemical staining with the anti-Hck Ab to distinguish macrophages from tumor cells, revealed that phagocytosis began within 2 h after addition of the mM-CSF-bearing tumor cells. Photocinematography confirmed that macrophages first phagocytosized and then lysed the internalized mM-CSF transfectant cells. Using annexin V and acridine orange staining techniques, macrophages phagocytosized living mM-CSF-transfected tumor cells.

Macrophages can recognize and kill tumor cells bearing the membrane isoform of macrophage colony-stimulating factor.

NBXFO hybridoma cells produced both the membrane and secreted isoforms of macrophage colony-stimulating factor (M-CSF). Murine bone marrow cells stimulated by the secreted form of M-CSF (sM-CSF) became Mac1+, Mac2+, Mac3+, and F4/80+ macrophages that inhibited the growth of NBXFO cells, but not L1210 or P815 tumor cells. In cytotoxicity studies, M-CSF activated macrophages and freshly isolated macrophages killed NBXFO cells in the presence of polymyxin B, eliminating the possibility that contaminating lipopolysaccharide (LPS) was responsible for the delivery of the cytotoxic signal. Retroviral-mediated transfection of T9 glioma cells with the gene for the membrane isoform of M-CSF (mM-CSF), but not for the secreted isoform of M-CSF, transferred the ability of macrophages to kill these transfected T9 cells in a mM-CSF dose-dependent manner. Macrophage-mediated killing of the mM-CSF transfected clone was blocked by using a 100-fold excess of recombinant M-CSF. Catalase, superoxide dismutase, and the nitric oxide inhibitor, N-omega-nitro-arginine methyl ester (NAME), did not effect macrophage cytotoxicity against the mM-CSF transfectant T9 clones. T9 parental cells when cultured in the presence of an equal number of the mM-CSF transfectant cells were not killed, indicating specific target cell cytotoxicity by the macrophages. Electron microscopy showed that macrophages were capable of phagocytosizing mM-CSF bearing T9 tumor cells and NBXFO hybridoma cells; this suggested a possible mechanism of this cytotoxicity. This study indicates that mM-CSF provides the necessary binding and triggering molecules through which macrophages can initiate direct tumor cell cytotoxicity.

Macrophage colony stimulatory factor-activated bone marrow macrophages suppress lymphocytic responses through phagocytosis: a tentative in vitro model of Rosai-Dorfman disease.

NBXFO hybridoma cells produced macrophage colony-stimulating factor (M-CSF), which stimulated the growth of murine bone marrow-derived macrophages with potent suppressor activity. These macrophages suppressed lymphocyte responses to mitogens and antigens in a dose-dependent manner. Using a transwell chamber, we demonstrated that macrophages needed physical contact with the lymphocytes to suppress lymphocyte proliferation on day 1 in the concanavalin A mitogen reaction. In addition, no soluble suppressor factor was detected at that time. The number of lymphocytes disappeared with time when they were cocultured with the macrophages. Electron microscopy revealed that the macrophage phagocytosized the lymphocytes after 7 1/2 h. Dextran sulfate, heparan, and fucoidan prevented the macrophages from suppressing the lymphocytes. This phenomenon resembles the human disease sinus histiocytosis, also called Rosai-Dorfman disease, in which macrophages (histiocytes) phagocytosize autologous lymphocytes; occasionally, this disease is associated with immunological abnormalities. Thus we believed that macrophage-activating cytokines, such as M-CSF, may stimulate macrophages to phagocytose lymphocytes in vivo.

The identification of the neonatal NBXFO hybridoma cell and its mediator.

The neonatal spleen:myeloma hybridoma cell, NBXFO, with immunosuppressive properties supported rodent hematopoietic colony formation. We identified this hybridoma to be an undifferentiated fibroblast that produced macrophage colony-stimulating factor (M-CSF). The bone marrow cells that grew in the presence of the NBXFO supernate were macrophages and were immunosuppressive towards lymphocytes. Neutralizing anti-M-CSF antibody partially inhibited the actions of the neonatal splenic suppressor cells. Neonatal splenocytes, but not the other parental cell line, FO, induced macrophage colony formation, possessed surface-associated M-CSF, and possessed M-CSF-specific transcripts. Therefore, we believe that the M-CSF-producing phenotype was contributed by a fibroblastic stromal cell and that these stromal cells could be responsible for the in situ generation of neonatal splenic suppressor cells.

Prostaglandin E2 induces up-regulation of murine macrophage beta-endorphin receptors.

Cultured murine bone marrow macrophages specifically bound 125I-labeled beta-endorphin. Binding was displaceable by 100 times molar excess of full-length beta-endorphin but was insensitive to the opioid receptor antagonist, naloxone. Binding was inhibited by beta-endorphin's C-terminal tetrapeptide, lys-lys-gly-glu, but not by the truncated N-terminal 27 amino acid fragment, indicating that binding of beta-endorphin to this receptor is dependent on its C-terminus. Macrophages incubated for 24 h with 10(-8)-10(-5) M prostaglandin E2 showed a dose-dependent increase in beta-endorphin binding, implying receptor up-regulation. This was also observed in response to the phosphodiesterase inhibitor, isobutylmethylxanthine, indicating that regulation of these receptors may be mediated through a cAMP-dependent process. This is the first demonstration that beta-endorphin receptor expression can be positively regulated.

Hematopoietic colony stimulatory factor formation by murine mesangial cells: gene expression and biological activity.

This study examined the ability of mesangial cells to synthesize colony-stimulating factors (CSF), cytoregulatory peptides associated with the differentiation and proliferation of hematopoietic cells. Conditioned media obtained from SV-40 transformed murine mesangial cells stimulated the growth of murine bone marrow progenitor cells of the myeloid series. Differential analysis of these cells showed the presence of both macrophages and granulocytes. Cellular identification of bone marrow colonies stimulated in response to mesangial cell conditioned media was examined by flow cytometric analysis and revealed the presence of F4/80 antigen positive macrophages (67%) and Gran-1 antigen positive granulocytes (21%). Neutralizing antibodies to macrophage colony-stimulating factor (M-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF) but not antibody to interleukin-3 (IL-3), or stem cell factor (SCF) significantly inhibited the growth of the progenitor cells induced by mesangial cell conditioned media. Utilizing Northern blot analysis, murine mesangial cells expressed mRNA transcripts for M-CSF, GM-CSF, and granulocyte colony-stimulating factor (G-CSF). Further studies were performed to determine optimal incubation conditions for mesangial cell CSF gene expression. These studies revealed that both GM-CSF and G-CSF mRNA were maximally expressed at early time points (4 and 8 h of incubation), while M-CSF mRNA expression remained unchanged during the incubation of mesangial cells from 4-48 h. Incubation of mesangial cells with various concentrations of fetal bovine serum (FBS, 0.5-15%) markedly increased the mRNA expression of M-CSF, GM-CSF and G-CSF in a dose-dependent manner. These studies indicated that transformed murine mesangial cells are able to synthesize and secrete biologically active CSF that are associated with the migration and proliferation of circulating mononuclear cells in the glomerulus. Furthermore, observations regarding the role of duration of incubation and the media concentration of FBS on mesangial cell CSF mRNA expression may provide useful data to understand the optimal conditions for studies that examine the gene expression of basal or inducible CSF in mesangial cells.

Detection of interleukin 2 receptors on murine lymphocytes using fluorescent interleukin 2.

Murine interleukin 2 receptors found on freshly isolated and on in vitro activated lymphocytes were identified using a fluorescent interleukin 2 (IL2F). Three percent of freshly isolated small thymocytes bound the IL2F; these cells appeared to be dual CD4 and CD8 positive cells. Ten percent of the larger thymocytes also bound the IL2F; phenotypically, these cells were more heterogenous in their CD4/CD8 composition than the small IL2F+ thymocytes. Freshly isolated splenocytes bound more IL2F than did the thymocytes. Twenty-four percent of the small splenocytes were IL2F+ and they were mostly B220+ cells. Half of the larger splenocytes were IL2 receptor positive and these cells consisted of B and T cells. Using mitogen stimulated splenocytes, three times as many LPS stimulated B220+ blasts bound the fluorescent IL2 than freshly isolated large B220+ cells; this level of IL2F binding was maintained for four days. Of the Con A blasts, more CD8+ cells (30%) bound IL2F than did CD4+ blasts (19%); these cells maintained this level of IL2F binding for only three days. The IL2F binding could be completely inhibited by excess unlabeled IL2 and could be inhibited by 92% using a monoclonal antibody directed against the IL2 binding region of the IL2 alpha receptor, indicating that IL2F can bind to both IL2 alpha and IL2 beta receptors.

The role of cytokines in graft-versus-host reactions and disease.

CD4+ and CD8+ T lymphocytes are responsible for the initiation of graft-versus-host disease (GVHD) which limits the efficacy of human allogeneic bone marrow transplantation. Activated T cells are a rich source of cytokines in vitro, and many of these same molecules are also found in graft-versus-host reactions (GVHR) and GVHD. However, T cells are not the only source of these soluble mediators. Cytokines control the actions of many cell types both in vitro and in vivo, so these soluble factors can easily influence the actions of both the engrafted hematopoietic cells and the host's cells. The complex manner in which cells and cytokines interact in GVHR and GVHD is reviewed here. We speculate that cytokine cascades play major roles in many aspects of GVHR and GVHD.

Effects of interleukin 4 upon human tumoricidal cells obtained from patients bearing solid tumors.

The use of human interleukin 4 (IL4) in the generation of tumoricidal effector cells derived from cancer patients undergoing either interleukin 2/lymphokine activated killer cells (IL2/LAK) therapy or tumor derived activated cell (TDAC) therapy was examined in the present study. Human IL4 alone did not generate LAK cells from patients undergoing IL2/LAK therapy when cultured in media containing 2% AB serum (five out of six patients). However, when the serum free medium, Aim V, was used, IL4 did generate LAK cells (eight out of nine patients), although not as cytotoxic as those activated with IL2. When cells were cultured in both IL2 and IL4 during the generation of LAK cells (3-7 days), better cell recoveries were frequently observed. The cytolytic activity of these cells against Daudi target cells was slightly reduced when compared to that response induced by IL2. This inhibition of LAK activity by IL4 was dose dependent with 1,000 U/ml IL4 producing maximal effects. This form of inhibition did not correlate with any phenotypic differences between those cells cultured in IL2 and those cells cultured in IL2 plus IL4. The IL4 mediated inhibition was also observed when the cells were cultured in Aim V medium which contains indomethacin. This inhibition induced by IL4 could not be overcome by using supra-optimal IL2. In addition to its effect during the generation of IL2 induced LAK cells, IL4 also appeared to reduce the cytolytic activity of pre-activated mature LAK cells. These results suggest that IL4 has a complex role in regulating the actions of LAK cells induced by lymphokines. When IL4 was used with IL2 during the first 5 weeks of growth of TDAC, an enhanced growth was observed when compared to the growth of TDAC when only one lymphokine was used. Besides the growth enhancement of TDAC, a better cytolytic response was observed when both lymphokines were used together. Thus, for the best growth of TDAC both IL2 and IL4 are required.

The generation of human lymphokine-activated killer cells in various serum-free media.

Recent data generated in our laboratory indicates that human lymphokine-activated killer (LAK) cells can be induced in several commercially available serum-free medium formulations. Interestingly, LAK cells were also generated in RPMI 1640 medium plus 1000 U/ml IL-2 without addition of human serum. However, higher levels of cytolytic activity were generally obtained in the commercial serum-free formulations. Cytotoxic activity was observed both at 3 and 7 days against a LAK-sensitive cell line (Daudi). Maximal cytolytic activity was usually produced by 7 days of in vitro culture in 1000 U/ml of interleukin-2 (IL-2).

Morphology and lytic mechanisms of interleukin 3-dependent natural cytotoxic cells: tumor necrosis factor as a possible mediator.

Populations of interleukin 3 (IL 3)-dependent cells can be derived from mouse bone marrow that display natural cytotoxicity (NC) against Wehi-164 target cells but do not display natural killing against YAC-1 cells. These bone marrow-derived NC cells cultured up to 2 mo in IL 3 do not contain rearranged T cell receptor beta-chain genes. They appear to be mast-like cells by electron microscopy and contain heterogeneous type granules. The molecules that mediate NC appear to be contained in these granules and are preformed because protein synthesis inhibitors have no effect on the capacity of IL 3-dependent NC cells to lyse Wehi-164 target cells. In addition to the IL 3-dependent bone marrow-derived cells, the basophilic leukemia cells, RBL-1, but not P815 mastocytoma cells were found to mediate NC against Wehi-164 cells. Both bone marrow-derived NC and RBL-1 cells can lyse L929 cells in 18 hr, suggesting that the putative NC mediator may be related to lymphotoxin/tumor necrosis factor (TNF). Recombinant human TNF displayed identical properties as NC cells; both entities possessed the same target cell specificity and had similar kinetics of target cell killing. The use of polyclonal rabbit antimouse TNF antibody blocked the actions of NC cells. Thus we believe that the mediation of NC is through the actions of a TNF-like molecule.