Induction of nonspecific cytotoxicity by monoclonal anti-T3 antibodies

The effects of monoclonal anti-T3 antibodies on the effector phase of cytotoxic T lymphocytes (CTL) were studied with respect to antigen-specific and antigen-nonspecific lysis of different target cells. Anti-T3 antibodies inhibited the antigen-specific lysis by CTL generated in mixed lymphocyte cultures (MLC), but they concomitantly augmented the nonspecific killing of third-party cells such as the cell lines Daudi, Raji, and K562. This nonspecific cytotoxicity was induced by various anti-T3 antibodies, whereas antibodies reactive with other antigens expressed on the cytotoxic effector cells lacked any such activity. Anti-T3 antibodies induced nonspecific cytotoxicity only when activated T cells, obtained by primary MLC, by repeated restimulation, or after cloning, were used. The antibodies had no effect on unstimulated peripheral T lymphocytes or thymocytes. The inhibition of the antigen-specific lysis and the induction of nonspecific lysis by anti-T3 was dose dependent, and both effects occurred at the same concentration range of anti-T3. F(ab')2 fragments of anti-T3 inhibited the specific lysis but were not able to induce cytotoxic activity, indicating that this induction is an Fc-dependent process. When different target cells were tested, only Fc receptor-positive cells were susceptible for this nonspecific cytotoxicity. Thus, anti-T3 antibodies have a dual effect on effector CTL: they inhibit antigen-specific lysis and concomitantly induce nonspecific lysis in an Fc-dependent way.     

Augmentation by anti-T3 antibody of the lymphokine-activated killer cell-mediated cytotoxicity

This study showed that a mAb (145-2C11) against the T3 epsilon-chain of the TCR complex augmented the cytotoxic activity of the lymphokine-activated killer (LAK) effectors. The LAK cells were induced by culturing normal spleen cells with purified human rIL-2. Adding alpha T3 at the effector phase of the cytotoxic reactions augmented the LAK-mediated cytotoxicity. The alpha T3-augmented LAK killing was seen only with tumor targets, and there was no increase of killing against Con A-induced lymphoblasts. The augmentation effect was dose dependent on both the amounts of alpha T3 and the number of LAK cells added. A very low concentration of alpha T3 (1/10,000 dilution of culture supernatants) was sufficient to induce alpha T3-augmented LAK-mediated cytotoxicity. Human rIL-2 at 10 to 30 U/ml was sufficient to generate LAK cells for maximal alpha T3 augmentation, whereas 300 to 1000 U/ml of IL-2 were needed to generate maximal LAK activity when tested in the absence of alpha T3. LAK cells generated for longer periods of time showed a progressive increase of alpha T3-augmented cytotoxicity. For some targets, the alpha T3-augmented LAK killing was FcR dependent as evidenced by the ability of alpha FcR mAb 2.4G2 to inhibit, and for others it was not inhibited. The alpha T3-augmented killing did not correlate with the FcR expression on target cells as defined by 2.4G2. The LAK cells were both Lyt-2+ and Lyt-2-, but the LAK cells involved in alpha T3-augmented killing were exclusively Lyt-2+. Preincubation of LAK cells with alpha T3, but not preincubation of targets with alpha T3, resulted in augmented killing suggesting that the alpha T3 effect was unrelated to an antibody-dependent cell-mediated cytotoxicity. Our findings indicate that alpha T3 is a potent reagent to augment the cytotoxic reaction of LAK cells. These results suggested that a relationship might exist between the T3 complex and the cytotoxic activity of a subpopulation of Lyt-2+ LAK cells.     

Heterogeneity of long-term cultured activated killer cells induced by anti-T3 antibody

The present study has characterized the short term and long term cultured murine-activated killer (AK) cells that are induced by antibody directed against the epsilon-chain of T3 complex. The conventional lymphokine AK (LAK) cells were generated by culturing normal B6 spleen cells with purified human rIL-2. The alpha T3-induced AK cells (T3AK) were induced by culturing normal B6 spleen cells with alpha T3 and were then maintained in culture medium supplemented with human rIL-2 and/or alpha T3. After initial activation with alpha T3, lymphocyte proliferation and generation of cytotoxic effectors (T3AK) were noted, and these events were related to the endogenous production of IL-2 and IL-4. Addition of alpha IL-2 and/or alpha IL-4 suppressed both the proliferative response and the cytotoxic response induced by alpha T3. In comparing the T3AK cells with the conventional LAK cells, there were many similarities as well as some distinct differences. Both cells displayed a similar cytotoxic spectrum against a variety of tumor targets. The T3AK cells usually gave much higher levels of cytotoxic activity against susceptible targets. However, the susceptibility of different tumor targets to conventional LAK cells and T3AK cells varied. The time course for the generation of lytic activity also differed between the conventional LAK and T3AK cells. One distinct difference was their ability to survive in vitro. The conventional LAK cells survived in culture for only 1 wk. The T3AK cells could survive for at least 4 to 5 wk with active growth. The serologic phenotype of the LAK precursors was asialo GM1 (AsGM1+) cells, but the T3AK precursors could be either AsGM1+ or AsGM1-, depending on the target cell. The LAK effectors were both Lyt-2+ and Lyt-2-, but the short-term T3AK effectors were exclusively Lyt-2+. The long term T3AK cells (cultured for more than 2 wk) were found to consist of Lyt-2+ and Lyt-2- cells, and these subsets of T3AK cells showed different target specificities. These findings demonstrate the heterogeneity of LAK and T3AK cells, and this heterogeneous property may contribute to their diversity in specificity against different tumor targets and thus may affect their effectiveness in the immunotherapy of cancer.     

Regulation of the alternative pathway of T cell activation by anti-T3 monoclonal antibody

T cell activation may be triggered either through the T3-Ti antigen receptor complex or via an alternative macrophage-independent pathway involving the 50KD T11 sheep erythrocyte-binding glycoprotein. Monoclonal antibodies anti-T11(2) and anti-T11(3), directed at distinct epitopes of the T11 molecule, trigger mature T cells to proliferate and express their functional programs, and induce expression of IL 2 receptors on both T3+ and T3- thymocytes. We now show that a non-mitogenic anti-T3 antibody blocks activation via the T11 pathway of not only peripheral blood T cells, but also T3+ thymocytes. Anti-T3 does not affect surface expression of T11 or the rapid augmentation of T11(3) expression after incubation of cells with anti-T11(2). However, anti-T3 inhibits generation of IL 2 receptors and production of IL 2 by T lineage cells cultured with anti-T11(2) plus anti-T11(3). In contrast, modulation of the T11 molecule by a non-mitogenic anti-T11 antibody does not inhibit activation of T cells by a mitogenic anti-T3 antibody. The ability of anti-T3 to block expression of IL 2 receptors on both thymocytes and mature T cells activated by the T11 pathway suggests that a regulatory interaction may be important during T cell ontogeny to provide a mechanism for inhibiting expansion of autoreactive clones.     

The Fc portion of intact IgG blocks stimulation of human PBMC by anti-T3

The means by which normal human serum inhibits the activation of PBMC by OKT3 has been investigated. The Fc portion of intact IgG is shown to be the major inhibitor in human serum of this OKT3-induced stimulation. Furthermore, inhibition by IgG subclasses correlated with their ability to bind to the monocyte Fc receptor, i.e., IgG1 and IgG3 produced greater inhibition than IgG2 and IgG4, and this inhibition was competitive. In contrast, hypogammaglobulinemic serum, IgA, IgM, and F(ab')2 of IgG were not inhibitory relative to intact IgG or Fc of IgG. Because of the similarities between T3 and the idiotype-defined T cell receptor for antigen, these investigations suggest that IgG might modulate the interactions between the T cell recognition complex and anti-idiotype antibody, thus regulating the idiotype network.     

Specific lysis of human tumor cells by T cells coated with anti-T3 cross-linked to anti-tumor antibody

Heteroaggregates containing anti-T3 cross-linked to anti-target cell antibodies have been shown to cause human T cells to lyse target cells that express antigens recognized by the anti-target cell antibody. In this study, we test targeted human T cells for the ability to lyse human tumor cells as a first step toward the application of this phenomenon to tumor immunotherapy. Several monoclonal anti-human tumor antibodies were assayed for binding to a number of human tumor lines and for the ability to promote specific tumor cell lysis when cross-linked with anti-T3. We found that anti-T3 cross-linked to anti-tumor monoclonal antibodies caused cloned human T cells and fresh peripheral blood T cells to lyse the tumor cells with the same specificity as predicted by the binding studies. Peripheral blood T cells were then tested in the presence of various heteroaggregates for the ability to lyse single cell suspensions prepared from fresh tumor or fresh normal tissue. These studies showed that heteroaggregates containing anti-T3 cross-linked to anti-tumor antibody cause fresh human T cells to specifically lyse fresh tumor cells, but not (with one exception) fresh normal cells.     

Activation of autoreactive cytolytic T lymphocyte clone against human melanoma by anti-T3 monoclonal antibody and autologous accessory cells

A cytotoxic T-lymphocyte (CTL) clone Tc1.8 was derived in a limiting dilution culture from a single cell that was derived from melanoma-involved lymph node lymphocytes activated in in vitro coculture against the autologous melanoma cells (VIP). The clone Tc1.8 (T3+, T8+, T4-, and Leu7-) expressed restricted cytolytic activity against only the autologous target VIP. As it aged in continuous culture containing interleukin 2, Tc1.8 lost cytolytic activity. The cytolytic function could be restored, however, with monoclonal antibody (MoAb) against T3 (OKT3) or with F(ab')2 fractions of OKT3, and upon restimulation with irradiated accessory cells. OKT3-mediated reinduction of cytotoxicity by the aged Tc1.8 could not be achieved if the T3 molecules were modulated from the effector cell surface following overnight incubation of Tc1.8 with saturating concentrations of OKT3 MoAb. Following reactivation with OKT3 Tc1.8 gained cytolytic function against NK targets in addition to VIP. Reactivation with F(ab')2 fractions of OKT3 and with autologous accessory cells, however, maintained its restricted antigen fidelity. The NK-like activity of Tc1.8 upon reactivation with OKT3 resulted from conjugate formation between the activated Tc1.8 and NK targets via the activating ligand itself. Thus, upon stimulation with anti-T3 MoAb and with autologous accessory cells, independently, the autoreactivity could be restored in an aged and inactive CTL clone.     

Susceptibility of cytotoxic T lymphocyte (CTL) clones to inhibition by anti-T3 and anti-T4 (but not anti-LFA-1) monoclonal antibodies varies with the "avidity" of CTL-target interaction

To explore the role of the T3, T4, and LFA-1 molecules in high and low "avidity" interactions between SB2-specific cytotoxic T lymphocyte (CTL) clones and their targets, monoclonal antibody-mediated inhibition of cytotoxicity has been studied in experiments that vary the "avidity" of interaction in three different ways. 1) Previous results have been extended with respect to different CTL clones assayed on the same SB2-positive target cells. Differences between clones in susceptibility to anti-T3 inhibition paralleled variations in anti-T4 inhibition, and both correlated inversely with the "avidity" of the effector-target interaction (inferred previously from studies of conjugate dissociation). 2) A high "avidity" clone, 8.4, was identified that lysed not only SB2-positive cells but also cross-reacted on a few SB2-negative cells. Cold target inhibition studies confirmed the cross-reaction, and together with conjugate dissociation studies, indicated that cross-reaction to be of lower "avidity" than the specific recognition of SB2. Cross-reactive lysis was much more susceptible to inhibition by anti-T3 and anti-T4 than was specific lysis. 3) Anti-T3 and anti-T4 blocking was analyzed in the presence of anti-Ia antibody to reduce the amount of Ia antigen available on the target. Anti-T3 and anti-T4 antibody blocking was more efficient after the addition of anti-Ia antibody concentrations that (by themselves) produced minimal inhibition of lysis. As a control, anti-LFA-1 antibody blocking was analyzed in each of these three experimental systems that compare interactions of different "avidity"; minimal variation was observed in the efficiency of inhibition by anti-LFA-1. Thus, anti-T3 and anti-T4 inhibition correlates inversely with the "avidity" of that CTL-target interaction, but anti-LFA-1 inhibition does not.     

Direct demonstration of the human suppressor inducer subset by anti-T cell antibodies

Prior studies indicated that sera of patients with active juvenile rheumatoid arthritis (JRA) contain anti-T cell antibodies reactive with the T4+ inducer population. More important, depletion of this T cell subset with JRA anti-T cell antibodies (JRA+ T cells) and C abrogated T5/T8+ suppressor T cell function. In the present study, we utilized Ig-coated plate techniques and JRA anti-T cell antibodies to fractionate the T4+ population into T4+JRA+ and T4+JRA- subsets and characterize the individual T4+ inducer subset. It was shown that whereas only the T4+JRA- population responded maximally to the soluble antigens, TT and mumps, both T4+JRA+ and T4+JRA- subsets proliferated equally well to mitogens and alloantigens. Furthermore, B cell immunoglobulin production induced by T4+JRA- T cells was approximately twice that induced by the reciprocal T4+JRA+ subset. In contrast, the T4+JRA+ subset alone activated T8+ T cells to become suppressor effector cells. These results suggest that the T4+JRA+ subset is the inducer of suppressor subpopulation whereas the T4+JRA- subset functions maximally as the inducer of B cells. It is believed that the suppressor inducer population may have a central role in the immunoregulatory network in man.     

Analysis of the monocyte Fc receptors and antibody-mediated cellular interactions required for the induction of T cell proliferation by anti-T3 antibodies

The induction of human T cell proliferation by antibodies that cross-link T3 antigens is dependent on functional interactions of anti-T3 antibodies with monocyte Fc receptors. In this report, we used a panel of anti-T3 antibodies of differing heavy chain isotype and a variety of other monoclonal antibodies to analyze several features of the antibody-mediated interactions between T cells and monocytes that are required for mitogenesis. Whereas three IgG2a anti-T3 antibodies were mitogenic for cells from all individuals, IgM and IgG2b anti-T3 antibodies did not induce T cell proliferation in any donor and could block the proliferative responses induced by other mitogenic anti-T3 antibodies. Dose-response analyses with four IgG1 anti-T3 antibodies demonstrated donor heterogeneity as reported by other investigators. However, in contrast to these previous reports, high concentrations of IgG1 anti-T3 antibodies were found to be mitogenic for all donors, indicating that this heterogeneity is based on relative rather than absolute defects in low responder monocytes. Cell mixing experiments in which monocytes from two different low responder donors were co-cultured with T cells and IgG1 anti-T3 antibodies did not identify any complementary defects, suggesting that the low responder phenotype results from a relatively restricted polymorphism. To assess the nature of the signals required for inducing T cell proliferation, nonmitogenic anti-T3 antibodies were co-cultured with other pan-T cell antibodies having the IgG2a isotype. The combination of signals from T3 antigen cross-linkage and those independently generated by other IgG2a antibodies bound to monocyte Fc receptors did not induce T cell proliferation. Hence, it appears that the T3 antigen or closely associated structures must be clustered at the monocyte membrane for mitogenesis. Finally, in competitive inhibition experiments, the isotype specificity of monocyte Fc receptors involved in the induction of T cell proliferation was examined. Two distinct Fc receptor sites, one that binds murine IgG2a and IgG3 antibodies and a second that binds murine IgG1 antibodies, were identified. Murine IgM or IgG2b did not appear to bind either of these receptor sites. Taken together, these data indicate that human monocytes have two distinct Fc receptor sites, which must specifically and directly interact with T cell-bound anti-T3 antibodies for mitogenesis.     

Regulation by anti-CD2 monoclonal antibody of the activation of a human T cell clone induced by anti-CD3 or anti-T cell receptor antibodies

In this study the effect of anti-cluster designation (CD) 2 monoclonal antibodies (mAb) on the activation of a cloned human T cell line, HY837, after triggering the CD3/T cell receptor (TcR) complex by anti-CD3 or anti-TcR mAb is described. HY837, which reacts with a series of mAb directed at different epitopes on the TcR, could be induced to proliferation and interleukin 2 (IL-2) production by soluble mAb directed at the CD3/TcR complex in the absence of accessory cells. mAb directed at the CD2 epitope T11-1 were shown to block the IL-2 production by HY837, as well as the expression of the IL-2 receptor, induced by anti-CD3 mAb, resulting in the inhibition of the proliferative response. The effect of anti-CD2 mAb on the proliferative response of HY837, induced by anti-CD3 mAb, was not due to a competition for Fc binding sites. In contrast, the proliferative responses and IL-2 production of HY837, induced by mAb directed at the TcR, were shown to be enhanced by the action of the anti-CD2 mAb. These results indicate that effects mediated by anti-CD3/TcR mAb cannot always be extrapolated to antigen-mediated effects and show that anti-CD2 mAb may regulate the T cell response, induced by mAb directed at the CD3/TcR complex, depending on which part of this complex is triggered during activation.     

Role of protein kinase C in interleukin 1, anti-T3, and mitogenic lectin-induced interleukin 2 secretion

Activation of T-lymphocytes by antigen, mitogenic lectins, or antibodies against the T-cell receptor complex, particularly in the presence of IL1, induces the secretion of the T-cell growth factor IL2. IL2 then has a major role in regulating the duration and magnitude of the immune response. Interaction of antigen, antibodies against the T-cell receptor complex, or mitogenic lectins with T-lymphocytes also induces hydrolysis of membrane phospholipids, leading to the production of diacylglycerol, an activator of the Ca2+- and phospholipid-dependent phosphotransferase, protein kinase C (PKC). Phorbol esters, potent activators of PKC, augment secretion of the T-cell growth factor, interleukin 2 (IL2). Activation of PKC may therefore serve as an important early event in the production and secretion of IL2. We have determined whether IL2 secretion can be induced in the murine cell T-lymphocyte line LBRM 331A5, where PKC is inhibited by staurosporine or sphingosine or in cells where PKC is depleted by prolonged incubation with high concentrations of phorbol esters. In cells in which PKC was either inhibited or depleted, antibodies against the T3 portion of the T-cell receptor complex and the mitogenic lectin phytohemagglutinin (PHA) still triggered IL2 secretion. In addition, the monokine IL1 augmented this IL2 secretion irrespective of whether PKC was inhibited or depleted. These data indicate that activation of PKC is not an obligatory step for IL2 secretion in LBRM 331A5 murine T-lymphocytes.     

The events of primary T cell activation can be staged by use of Sepharose-bound anti-T3 (64.1) monoclonal antibody and purified interleukin 1

A mitogenic anti-CD3 ("T3") monoclonal antibody (64.1), that stimulates polyclonal T cell activation by a mechanism believed to be similar to antigen via binding to the T cell receptor complex, was utilized in soluble (SOL) and Sepharose-bound (SEPH) forms to dissect the role of accessory cells (AC) and interleukin 1 (IL 1) in supporting T cell activation. The T cell activation pathway was dissected into "early" events including expression of interleukin 2 receptors (IL 2R), increased RNA content, IL 2 release, and "late" (DNA synthesis) events. Unseparated peripheral blood mononuclear cells progressed through all stages of activation when stimulated by either form of 64.1. Stringent AC depletion by plastic adherence, nylon wool adherence, and L-leucine methyl ester (selectively lyses AC) prevented early and late T cell responses to either form of 64.1. The addition of highly purified IL 1 replenished both early and late T cell responses to SEPH-64.1 but not to SOL-64.1. Although SOL-64.1 stimulation of purified T cells induced modulation of the CD3 complex, only SEPH-64.1 induced IL 1 responsiveness, and exogenous IL 1 was then able to support synthesis of RNA, secretion of IL 2, expression of IL 2R, and ultimately, DNA synthesis. Therefore, the stages of early T cell activation owing to stimulation of the CD3-T cell receptor complex and IL 1 responsiveness have been dissected.     

Separation of bivalent anti-T cell immunotoxin from Pichia pastoris glycoproteins by borate anion exchange

A major problem encountered in the large-scale purification of the bivalent anti-T cell immunotoxin, A-dmDT390-bisFv(G4S), from Pichia pastoris supernatants was the presence of host glycoproteins exhibiting similar charge, size, and hydrophobicity characteristics. We overcame this problem by employing borate anion exchange chromatography. The borate anion has an affinity for carbohydrates and imparts negative charges to these structures. We found that at a concentration of sodium borate between 50 and 100 mM, the nonglycosylated immunotoxin did not bind to Poros 50 HQ anion exchanger resin, but glycoproteins, including aggregates related to the immunotoxin, did. By using this property of the immunotoxin in the presence of sodium borate, we successfully developed a 3-step purification procedure: (i) Butyl-650M hydrophobic interaction chromatography, (ii) Poros 50 HQ anion exchange chromatography in the presence of borate, and (iii) HiTrap Q anion exchange chromatography. The final preparation exhibited a purity of greater than 98% and a yield of greater than 50% from the supernatant. Previously, boronic acid resins have been used to separate glycoproteins from proteins. However, combining borate anion with conventional anion exchange resins accomplishes the separation of the immunotoxin from glycoproteins and eliminates the need to evaluate nonstandard resins with respect to good manufacturing practice guidelines.     

Treatment of murine lupus with monoclonal anti-T cell antibody

Three strains of autoimmune mice (MRL/lpr, NZB/NZW, and BXSB) were treated with repeated injections of rat monoclonal anti-T cell antibody (anti-Thy-1.2) in order to determine 1) the extent and duration of target cell depletion, 2) the effect of T cell depletion on the course of autoimmunity, and 3) the magnitude and consequences of the host immune response to the monoclonal antibody. Mice were treated with 6 mg of anti-Thy-1.2 every 2 wk beginning early in their disease. Treatment produced a substantial reduction in circulating T cells in all three strains. Therapy was beneficial in MRL/lpr mice. It reduced lymphadenopathy, lowered autoantibody concentrations, retarded renal disease, and prolonged life. In contrast, treatment did not improve autoimmunity in NZB/NZW mice, and it caused fatal anaphylaxis in BXSB mice. These findings demonstrate that monoclonal antilymphocyte antibodies can serve as specific probes to examine the cells that contribute to autoimmunity. Moreover, they illustrate the potential therapeutic value of monoclonal antilymphocyte antibodies when a pathogeneic cell subset can be identified. However, the same antibody may have a broad range of effects, from efficacy to severe toxicity, even in diseases that share clinical features.     

Reaction of T lymphocytes with anti-T3 induces translocation of C-kinase activity to the membrane and specific substrate phosphorylation

Reaction of the T cell membrane with monoclonal antibodies to T3 can initiate cellular activation, and this is associated with increased intracellular Ca2+ and inositol-trisphosphate (IP3) release. We therefore studied the possible involvement of Ca2+/phospholipid-dependent kinase (C-kinase) in these phenomena. Quantitative assays of exogenous substrate phosphorylation in unstimulated cells showed Ca2+/phospholipid-dependent kinase activity in the cytosol, but no comparable activity in the particulate fractions corresponding to membrane and cytoskeleton material. At concentrations of soluble anti-T3 that partially activate T cells in the absence of macrophages, there was a 50 to 60% decrease in C-kinase activity in the cytosol, with a comparable increase in activity in the membrane fraction. A similar transfer of activity was also induced with the known C-kinase activator, 12-O-tetradecanoyl-phorbol-13-acetate, although redistribution was more rapid in onset, more complete, and more sustained. Redistribution of enzyme activity was additionally confirmed by qualitative assays of endogenous substrate phosphorylation. Labeling of intact cells followed by immunoprecipitation analysis with anti-T3 indicated signal-dependent phosphorylation of two components of the T3 complex and an unidentified 94,000 substrate that was resistant to reduction and alkylation. These findings are consistent with an important role for C-kinase in transduction of membrane events by the T3-Ti complex.     

Human T cells targeted with anti-T3 cross-linked to antitumor antibody prevent tumor growth in nude mice

Human peripheral blood T cells were tested for the ability to prevent tumor growth in nude mice when targeted with anti-T3 cross-linked to antitumor antibodies. LS174T human colon adenocarcinoma cells were mixed with human PBL coated either with anti-T3 (Fab) cross-linked to 315F6 (Fab) (an antitumor monoclonal antibody) or with no antibody, and were injected subcutaneously into nude mice. Tumor growth was totally inhibited at effector to target (E:T) ratios of 7.0:1 and 2.1:1, and was partially inhibited at 0.7:1 with antibody-coated PBL, but was not inhibited by uncoated PBL. T cell-mediated protection against tumor growth occurred when an antitumor was physically cross-linked to anti-T3. Neither a mixture of unlinked anti-T3 and antitumor antibodies nor anti-human MHC class I cross-linked to antitumor antibody prevented tumor growth. Whereas in vitro cytotoxicity was mediated exclusively by T8+ cells and was augmented by brief exposure of effector cells to IL 2, tumor neutralization in vivo was mediated by both T4+ and T8+ cells and was not significantly stimulated by prior exposure of the cells to IL 2. We conclude that human T cells, when targeted with appropriate antibody heteroaggregates, can specifically inhibit tumor growth at low E:T ratios, and that cells mediating tumor neutralization in vivo may differ from those mediating cytotoxicity in vitro.