Role of interferon in human natural killer activity against target cells infected with HSV-1

Human natural killer (NK) cells show high cytotoxic activity against target cells infected with herpes simplex virus type 1 (HSV-1). Substantial amounts of interferon (IFN) were generated in co-cultures of NK effector cells and infected target cells; however, the cytotoxic activity seen against a specific infected cell target did not correlate with the amount of IFN induced. The production of IFN increased steadily from 4 to 18 hr of co-culture, as did NK activity; however, IFN production peaked 4 hr later than NK activity. Pretreatment of NK effector cells with exogenous IFN increased cytotoxic activity against all targets tested, but the differential pattern of reactivity against cells infected with wild type and mutant viruses was unaltered. When effector cells were treated with the RNA synthesis inhibitor actinomycin D before co-culture with virus-infected targets, IFN production was markedly reduced, without a concomitant reduction in cytotoxicity. Similarly, the addition of anti-IFN antiserum to co-cultures greatly decreased the available IFN present, but had no effect on NK activity. We conclude that the induction of cytotoxic activity in co-cultures of NK effector cells and HSV-1-infected target cells is independent of the induction of IFN.     

Lymphocyte-mediated cytotoxicity to cells infected with measles virus: I. Use of in vitro infected leukocytes as autologous target cells; absence of virus-specific cytotoxic lymphocytes

We investigated lymphocyte-mediated cytotoxicity in humans to autologous cells infected with measles virus. Mononuclear leukocytes, isolated from peripheral blood, were stimulated by phytohemagglutinin (PHA) and infected with measles virus. At 72 hr after infection, about 80% of the cells could be lysed by antibodies against measles virus and human complement, which meant that at that time the expression of virus-specific antigens on the cell surface was maximal. Such PHA-stimulated, infected leukocytes were used as target cells in an assay for lymphocyte-mediated cytotoxicity. Effector lymphocytes were obtained from the same donor who had provided the target cells, and were tested for their cytotoxicity directly after isolation.Lymphocytes obtained from adult humans, with a history of natural measles infection contracted during childhood, were not found to be cytotoxic to autologous infected cells, unless antibodies against measles virus were present during the assay. The same response, though to a lesser extent, was observed with cord blood lymphocytes obtained from healthy neonates. This indicates that the observed cytotoxicity does not reflect acquired cellular immunity but rather antibody-dependent cellular cytotoxicity (ADCC).     

Interactions between human lymphocytes and paramyxovirus-infected cells: adsorption and cytotoxicity.

The capacity of human lymphocytes to adhere to paramyxovirus-infected monolayers and their capacity to kill paramyxovirus-infected cells was investigated. A large fraction of human lymphocytes was found to adhere firmly to the paramyxovirus-infected monolayers. Predsorption of lymphocytes on mumps virus-infected cells impaired their adsorption to a second cell monolayer of the same type. The cytotoxic activity of lymphocytes against mumps virus-infected cells was also reduced after predsorption on mumps virus- or Newcastle disease virus-infected (NDV) cell monolayers. Exposure of lymphocytes to trypsin did not significantly decrease either adsorption or cytotoxicity. Pretreatment of lymphocytes with neuraminidase (NANase) partly inhibited adsorption whereas cytotoxicity was not decreased. Cell fractionation experiments after rosetting of the lymphocytes with sheep erythrocytes (E) indicated that T cells were equally or better adsorbed than "non-T" cells. Taken together with previous experiments which showed that the majority of T lymphocytes are not cytotoxic against mumps virus-infected cells these results suggest that adherence of lymphocytes to infected cells and cytotoxicity may be unrelated phenomena.     

The effect of interferon on lymphocyte-mediated effector cell functions: selective enhancement of natural killer cells.

The effect of human interferons on different types of lymphocyte-mediated killer assays was explored. Killing by T cells generated through mixed lymphocyte cultures as well as antibody-dependent lymphocyte-mediated cytotoxicity was not influenced by the addition of interferon. Enhancement of cytolysis produced by natural killer cells was observed when interferon was added during the assay, but enhancement could also be induced if the effector cells were pretreated with interferon for 2 hr prior to the lytic reaction. Killing of a cell line susceptible to natural killing was increased and a cell line which is normally relatively resistant to this type of killing became a susceptible target.     

Analysis of effector cells in human antibody-dependent cellular cytotoxicity with murine monoclonal antibodies

We examined purified human large granular lymphocytes, peripheral monocytes, and T cells for their ability to mediate antibody-dependent cellular cytotoxicity (ADCC) with murine monoclonal antibodies. We also evaluated the effects of pretreatment of cells with interleukin 2 and interferon to augment ADCC activity. MB3.6, a murine monoclonal antibody directed against the GD3 ganglioside, induced high levels of ADCC. This ADCC was mediated predominantly, if not completely, by human killer cells (large granular lymphocytes) whereas other effector cell populations demonstrated no significant cytotoxic activity in 6- or 18-hr assays. The IgG2a an anti-melanoma antibody 9.2.27 generated low or no ADCC with most normal donors or melanoma patients. IL 2 was a very potent booster of ADCC activity. Interferon alpha also was effective, whereas interferon gamma did not augment but rather inhibited reactivity. We tested a large panel of antibodies of various isotype against colon carcinoma cells and found that gamma-3 isotype antibodies more frequently generated ADCC and produced higher levels of cytotoxic activity than did IgG1 or IgG2 antibodies. It appears that a variety of parameters can affect ADCC reactions, including the type of effector cell and its level of activation, the isotype of the antibody, and properties of the target cell line such as its susceptibility to lysis.     

Lysis of human T cell leukemia virus infected T and B lymphoid cells by interleukin 2-activated killer cells

The human T cell leukemia (HTLV-1) retrovirus is the etiologic agent for adult T cell leukemia. Interleukin 2 (IL-2) activated killer (AK) cells have been shown to lyse freshly explanted tumor cells in vitro and have been used as a form of adoptive immunotherapy for the treatment of cancer. In this report, the ability of AK cells to lyse HTLV-1-infected targets was examined. Normal lymphocytes, when cultured in recombinant IL-2 for periods of 3 to 7 days, killed infected T and B cell lines. The precursor for these AK cells resided in the CD-16 antigen-positive subset (i.e., natural killer (NK) cells). Resting T cells, NK cells, or unfractionated lymphocytes did not lyse the infected targets. However, when isolated NK cells were incubated for 24 hr in IL-2, suboptimal cytolysis was induced whereas activation of NK cells with a four pulse of IL-2 was insufficient to generate effector cells. The results of performing cold target inhibition studies with Epstein-Barr virus-infected B cell lines and HTLV-1-infected T and B cell lines suggest that there are discrete subsets (i.e., clonotypic) in the AK population that preferentially lyse a given virally infected cell line. Thus to consider AK cells as true polyspecific killer cells may be inaccurate. Alternately AK cells may express a number of different receptors with variable affinities for the Epstein-Barr virus- and HTLV-1-infected cell lines. In addition, it was shown that HTLV-1-infected B cells are relatively resistant to AK cell-mediated lysis. These results clearly indicate that AK cells but not resting NK cells kill HTLV-1-infected cells.     

Human natural cell-mediated cytotoxicity against fetal fibroblasts. III. Morphological and functional characterization of the effector cells

We have isolated and characterized human natural killer cells cytotoxic to human fetal fibroblasts utilizing adsorption-elution of the effector cells from target cell-coated beads. The cell associated with enriched cytotoxicity was slightly larger than small- to medium-sized lymphocytes, the cytoplasm was pale and characteristically granular. In direct surface marker analysis the cell was Fc-receptor-positive, formed E-rosettes, and displayed strong either diffuse or granular ANAE reactivity in the cytoplasm. The ANAE reactivity could not be inhibited with sodium fluoride and in mitogen and antigen stimulation experiments the cell had T-cell characteristicis. The cell type was termed large granular lymphocyte and we suggest that it is the main direct effector cell for natural killer activity against human fetal fibroblasts.     

Phorbol esters enhance spontaneous cytotoxicity of human lymphocytes, abrogate Fc receptor expression, and inhibit antibody-dependent lymphocyte-mediated cytotoxicity

Phorbol esters with tumor promoter activity enhance the spontaneous cytotoxicity of human lymphocytes against a variety of target cell lines, with an efficiency that correlates with their potency as tumor promoters or skin irritants. Analysis of surface marker expression of the lymphocytes cytotoxic after treatment with phorbol ester identified the cytotoxic cell subset as that containing natural killer cells. Although gamma-interferon (IFN gamma) is produced by T cells treated with phorbol esters, IFN gamma is probably not the mediator of enhancement of natural killer cell activity, because anti-IFN gamma antibodies failed to block this enhancement. Spontaneous cell-mediated cytotoxicity is inhibited when phorbol esters are present during the cytotoxic assay, but is enhanced when the effector cells are pretreated with these agents. On the other hand, antibody-dependent cytotoxicity mediated by lymphocytes is inhibited by phorbol ester pretreatment of the effector cells or by phorbol esters present during the cytotoxic assay. Treatment of lymphocytes with phorbol esters at 37 degrees C, but not at 4 degrees C, completely abrogates in 1 to 2 hr the expression of the receptor for the Fc fragment of IgG, as detected by rosette formation with IgG-sensitized erythrocytes and by reactivity with anti-Fc receptor antibodies. The inhibition of antibody-dependent cytotoxicity by phorbol esters is probably secondary to their effect on the Fc receptor.     

Human lymphocyte cytotoxicity against mumps virus-infected target cells. Requirement for non-T cells.

Subpopulations of human lymphocytes were tested for their capacity to kill mumps virus-infected target cells in a 51-chromium release asaay. Using two different cell fractionation techniques, lymphocytes were fractionated into T cell-enriched (primarily T cells) and T cell-depleted (primarily B cells) subpopulations. Filtration of lymphocytes through columns coated with human immunoglobulin and rabbit anti-human-immunoglobulin (Ig-anti-Ig) rendered the resulting T-cell preparation inactive as effector cells against target cells carrying mumps virus antigens. In the second technique, lymphocytes were fractionated by centrifugation into two fractions according to their ability to form spontaneous rosettes with sheep erythrocytes (E). The E-rosette-forming population (primarily T cells) was shown to lack cytotoxic activity against mumps virus-infected target cells. This activity was present in the nonrosetting population. The results suggest that the effector cells involved in this cytotoxic system are of a non-T variety.     

Human natural cell-mediated cytotoxicity against fetal fibroblasts. II. Isolation and target cell specificity of the effector cells.

Human peripheral Wood lymphocytes were depleted of natural killer cells cytotoxic against human fetal fibroblasts by allowing them to attack the fibroblast targets grown on plastic beads followed by gravity sedimentation under conditions in which single cells floated but the attacker cells sedimented with the carrier beads. The attacker cells could be released from the bead-grown targets and shown to be greatly enriched in natural cytotoxic activity. The effector cells depleted by fibroblast adsorption were also depleted of cytotoxic activity against other monolayer targets whereas suspension grown lymphoma and leukemia cells (MOLT-4, RAJI, and K-562) were killed as effectively as by non-depleted effector cells. In competition assays other monolayer cells inhibited the natural cytotoxicity against fetal fibroblasts but the suspension-grown cells were unable to compete. The results suggested that different effector cell populations were probably involved when monolayer vs suspension targets were used in assays for human natural cell-mediated cytotoxicity. The separation was not, however, functionally complete since in competition assays with suspension-grown target cells also monolayer cells were able to compete. Preliminary morphological characterization of the natural killer cells against fetal fibroblasts is also presented.     

Lymphocyte-mediated cytotoxicity to autologous cells infected with measles virus. II. Specificity of the cytotoxic reaction and characterization of the effector cells involved.

The mechanism of lymphocyte-mediated cytotoxicity to cells infected with measles virus was investigated. Cytotoxicity was measured in a direct assay, immediately after the isolation of lymphocytes from human peripheral blood; mononuclear leukocytes, infected with measles virus in vitro, served as autologous target cells. Virus-specific cytotoxicity required the presence of both IgG antibodies against measles virus and of effector lymphocytes. The effector lymphocytes had Fc receptors and were mainly present in a fraction of non-T lymphocytes. Monocytes were not cytotoxic but rather inhibitory. These results indicate that lysis of virus-infected cells in this direct assay is due to antibody-dependent cellular cytotoxicity (ADCC), caused by K cells. Control experiments showed that the virus-infected target cells were sensitive to incubation with human serum or IgG, resulting in a nonspecific increase of ⁵¹Cr release; however, this did not affect the results of K-cell cytotoxicity. Maximal virus-specific lysis by ADCC did not reach the level obtained by complement-dependent cytotoxicity. Possible explanations for this difference are discussed.     

Virus specificity of human influenza virus-immune cytotoxic T cells.

The virus specificity of human in vitro cytotoxic T cell responses to influenza virus was studied with the use of peripheral blood mononuclear leukocytes from normal adult volunteers. Previous natural exposure of these donors to a variety of type A influenza viruses was documented by HI antibody titers. Cells sensitized in vitro with A/HK or A/PR8 were cytotoxic for autologous target cells infected with A/HK, A/PR8, or A/JAP 305 type A influenza viruses, but not for B/HK-infected or uninfected cells. B/HK-sensitized effector cells lysed target cells infected with B/HK but not targets infected with type A viruses. A/HK- and A/PR8-immune effector populations were shown to recognize cross-reactive antigens on A/HK- and A/PR8-infected target cells by cold target competition. Influenza-immune effector cells were cytotoxic for virus-infected autologous targets but much less so for virus-infected allogeneic targets. This self-restriction suggested that the cytotoxicity was largely T cell-mediated and was confirmed by cell separation analysis. Thus, the human secondary cytotoxic T cell response in vitro to influenza viruses is predominantly directed against cross-reactive determinants on cells infected with serologically distinct type A influenza viruses.     

Human cytotoxic T lymphocytes (CTL) induced by phytohemagglutinin: conditions for CTL generation and effect of interferon

The present study demonstrates that human peripheral blood mononuclear cells (PMC) can be stimulated in vitro to become cytotoxic T lymphocytes (CTL) by PHA. A significant cytotoxic activity of PMC was detected 48 hr after the culture initiation in the presence of 5 micrograms/ml of PHA and the peak level of the activity was obtained by culturing PMC for 72 hr. The cytotoxic cells require the presence of PHA as a cell agglutinin for the expression of their cytotoxic activity. The effector cells mediating the activity were identified as T lymphocytes by E-rosette fractionation of PMC. In this system, removal of carbonyl iron phagocytosed or attached cells from PMC did not abrogate CTL generation of PMC. In addition, human alpha-interferon did not augment CTL generation or expression of their activity. Although the target cells employed were sensitive to natural killer (NK) cells, the effector cells induced by PHA did not seem to have any relation to the NK cells. The present study may provide a useful tool to analyze for precursors of killer T cells.     

Leu-Leu-OMe sensitivity of human activated killer cells: delineation of a distinct class of cytotoxic T lymphocytes capable of lysing tumor targets

Sensitivity to L-leucyl-L-leucine methyl ester (Leu-Leu-OMe) was used to characterize the phenotype of human activated killer cells. Natural killer cells (NK) and the precursors of both the alloantigen-specific cytotoxic T lymphocytes (CTL) and the NK-like activated killer cells generated after stimulation with allogeneic cells were deleted from human peripheral blood lymphocytes by preincubation with Leu-Leu-OMe. It was noted, however, that cytotoxic lymphocytes could be generated from Leu-Leu-OMe-treated lymphocyte precursors after 2 to 6 days of culture with the nonspecific mitogen, phytohemagglutinin (PHA). The characteristics of these killer cells indicated that they were a unique population that could be distinguished from other cytotoxic cells. Killing by these cells exhibited slow kinetics in that 18 hr cytotoxicity assays were required to detect full cytotoxic potential. When 18 hr assays were used, PHA-stimulated cytotoxic cells generated from Leu-Leu-OMe-treated lymphocytes were able to kill both NK-sensitive K562 cells and the relatively NK-resistant renal cell carcinoma cell line, Cur. These cytotoxic lymphocytes were HNK-1, Leu-11b (CD16), and OKM1 (CR3)-negative at both the precursor and effector stage of activation. Furthermore, these cells were derived from a CD3-positive precursor. Finally, killing by activated effectors was inhibited by OKT3. Unlike activation of Leu-Leu-OMe-sensitive large granular lymphocytes, generation of these cytotoxic T cells was totally prevented by treatment with mitomycin c before stimulation. Thus, a unique class of tumoricidal T cells can be characterized by resistance of lymphocyte precursors to a concentration of Leu-Leu-OMe, which has been shown to ablate NK, mixed lymphocyte culture-activated NK-like cytotoxic precursors, and the precursors of alloantigen-specific CTL.     

Role of interferon in natural kill of HSV-1-infected fibroblasts

The production of interferon during natural killer (NK) assays against HSV-1-infected fibroblasts (NK(HSV-1)) was studied to determine whether this interferon was responsible for inducing the preferential lysis of herpes-virus-infected target cells over uninfected target cells. The interferon produced during NK(HSV-1) assays was analyzed and found to have the properties of HU-IFN-alpha. Little or no IFN was produced during NK assays against uninfected fibroblasts (NK(FS)) or K562 (NK(K562)) cells. Although the appearance of interferon in the culture supernatants seemed to parallel the development of cytotoxicity during NK(HSV-1) assays, the levels of cytotoxicity and IFN generated did not correlate, arguing against a strict quantitative dependence of cytotoxicity upon IFN production. NK(K562) and NK(FS) cytotoxicity developed with little or no production of IFN. When IFN-pretreated effector cells were used, there was still a preferential lysis of infected over uninfected target cells. This preferential lysis by IFN-treated effector cells of infected over uninfected targets was seen as early as 2 hr into the assay. Anti-IFN antibodies added to the NK assays, although neutralizing all the IFN produced during the assays, had no effect on NK(FS) or NK(K562) cytotoxic activity and caused a slightly reduction of NK(HSV-1) activity only in one of three experiments. We conclude that although IFN is generated during NK(HSV-1) assays, this IFN cannot solely account for the increased lysis of infected over uninfected cells and that NK(HSV-1) activity is in some other way dependent on the virus infection.     

Monoclonal antibody-defined surface markers of effector cells involved in human monocyte cytotoxicity

Human adherent peripheral blood mononuclear cells were cytotoxic in vitro against the murine TU5 line in a 48-hr [3H]thymidine-release assay. Monocyte-enriched adherent cell preparations contain a small and variable (usually less than 5%) contamination with large granular lymphocytes as assessed by morphology and staining with monoclonal antibody markers B73.1 and HNK1. To assess whether killing was in fact mediated by monocytes, mononuclear cells or monocyte-enriched preparations were separated using monoclonal antibodies directed against mononuclear phagocytes (Mo2, UCHM1, B44.1) or natural killer (NK) cells (B73.1 and HNK1), and a fluorescence-activated cell sorter. Cells positive for monocyte markers were highly cytotoxic against TU5, whereas negative cells were not. B73.1+ or HNK1+ cells had little or no activity. Cytotoxicity of cells positive for monocyte markers (Mo2, UCHM1, B44.1) was augmented by in vitro exposure to lymphokines or less frequently to interferon (IFN). However, cells negative for these monocytes markers were also stimulated to kill TU5 by lymphokine or IFN to an extent similar or greater than that of positive ones. IFN or lymphokines induced killing of TU5 by monocyte-depleted, B73.1-positive, lymphoid cells. These observations demonstrate that human monocytes do kill tumor cells, either in the absence of deliberate stimulation or after exposure to agents such as lymphokines. However, the possible contribution to "monocyte" cytotoxicity of minor NK cell contaminants must be taken into account particularly when agents such as IFN and lymphokines are applied, even when a relatively NK-cell-resistant target such as TU5 is used.     

Enhancement by interferon of natural killer cell activity in mice.

Injection of mice with several interferon inducers, Newcastle Disease virus, polyinosinic-polycytidylic acid and tilorone resulted in an increase in spleen cell cytotoxicity for ⁵¹chromium-labeled mouse YAC tumor target cells in 4-hr in vitro assays. This increase in spleen cell cytotoxicity was abrogated by injection of mice with potent anti-mouse interferon globulin. Inoculation of mice with mouse interferon (but not human leucocyte or mock interferon preparations) also resulted in a marked enhancement of spleen cell cytotoxicity. The extent of enhancement of spleen cell cytotoxicity was directly proportional to the amount of interferon injected and a significant increase was observed after inoculation of as little as 10³ to 10⁴ units of interferon. An effect could be detected as soon as 1 hr after injection of interferon. The increase of spleen cell cytotoxicity after inoculation of an interferon inducer was not due to a localization and accumulation of cytotoxic cells in the spleen but reflected a general increase in cytotoxic cell activity in various lymphoid tissues (except the thymus). The splenic cytotoxic cells from interferon or interferon-inducer-injected mice had the characteristics of natural killer (NK) cells since (i) interferon enhanced spleen cell cytotoxicity in athymic (nu/nu) nude mice, (ii) classical spleen cell fractionation procedures by nylon wool columns, anti-Thy 1.2 serum plus complement, anti-Ig columns, and depletion of FcR+ rosette-forming cells, failed to remove the effector cells generated in vivo or in vitro. Therefore like NK cells, interferon-induced cytotoxic cells lack the surface markers of mature T and B lymphocytes, are not adherent, and are devoid of avid Fc receptors. Furthermore like NK cells, the spleen cells from interferon-treated mice lysed various target cells (known for their sensitivity to NK cells) without H-2 or species restriction. Incubation in vitro of normal spleen cells with interferon also resulted in an increase in cytotoxicity for YAC tumor cells. We conclude that interferon acts directly on NK cells and enhances the inherent cytotoxic activity of these cells.     

T cell-mediated cytotoxicity induced by Listeria monocytogenes. I. Activation of Listeria antigen-responsive T cells

Soon after rats are infected with Listeria monocytogenes (LM), Listeria antigen- (LMA) responsive lymphocytes are delivered to the animal's thoracic duct. These LM-responsive lymphocytes can be restimulated in vitro by LMA to generate cells that have a potent cytolytic capability. The activation of LMA responsive lymphocytes is immunologically specific and dependent upon the activity of histocompatible accessory cells. Neither cell-free LMA nor LMA-pulsed allogeneic accessory cells can promote a significant cytotoxic response by negatively selected responder lymphocytes. LM-dependent cytolytic lymphocytes differ from natural killer (NK) cells inasmuch as their activation is not facilitated by interferon (IF). Likewise, supernatants from cultures containing specifically sensitized thymus-derived (T) lymphocytes and histocompatible LMA-pulsed accessory cells fail to augment (day 2 and day 3 supernatants actually inhibit) the activation process. The results imply that the successful activation of LM dependent cytolytic lymphocytes requires the cooperative interplay of responder T cells and specific-antigen-pulsed accessory cells.     

Comparison of murine lymphokine-activated killer cells, natural killer cells, and cytotoxic T lymphocytes

Precursors and effectors of murine lymphokine-activated killer cells, natural killer cells, and cytotoxic T lymphocytes are compared. Natural killer cells are resistant to gamma-irradiation (1000 R) whereas precursors of lymphokine-activated killer cells and cytotoxic T lymphocytes are sensitive. Lower doses of gamma-irradiation (500 R) remove precursors for cytotoxic T lymphocytes but not lymphokine-activated killer cells. In addition, lymphokine-activated killer cells are regenerated before classical CTL after sublethal doses of gamma-irradiation. Natural killer cells are resistant to anti-Thy 1 and C' and anti-thymocyte serum, but sensitive to anti-asialo GM1 and complement. Precursors of cytotoxic T lymphocytes are sensitive to anti-Thy 1 and complement and anti-thymocyte serum, but are resistant to anti-asialo GM1 and complement. Precursors of lymphokine-activated killer cells are partially sensitive to anti-Thy 1 and complement and anti-thymocyte serum, but are resistant to anti-asialo GM1 and complement. Effector cells of cytotoxic T lymphocytes are sensitive to anti-Thy 1 and complement and resistant to anti-asialo GM1 and complement. Lymphokine-activated killer cell effectors are sensitive to anti-asialo GM1 and complement at 24 hr after activation. These effectors are more closely aligned with classical natural killer effectors. Lymphokine-activated killer effectors, 7 days after activation, are resistant to anti-asialo GM1 and complement and sensitive to anti-Thy 1 and complement. Relationships and differences among these cytotoxic subsets are discussed.     

Studies on the mechanism of natural killer cytotoxicity. II. coculture of human PBL with NK-sensitive or resistant cell lines stimulates release of natural killer cytotoxic factors (NKCF) selectively cytotoxic to NK-sensitive target cells

This investigation has employed the "innocent bystander" type of experimental design to determine whether soluble cytotoxic factor(s) are released during interactions between human peripheral blood lymphocytes (PBL) and NK-sensitive target cells. PBL cocultured with NK-sensitive Molt-4 or K562 target cells in the lower well of a miniaturized Marbrook culture released natural killer cytotoxic factors (NKCF), which diffused across a 0.2-mu Nucleopore membrane and lysed Molt-4 or K562 target cells cultured in the upper chamber. Coculture of PBL with the NK-resistant Raji or WI-L2 cell lines also induced release of NKCF. These factors were selectively cytotoxic to NK-sensitive targets and lysed Molt-4 and, to a lesser extent, K562 cells. However, Raji, WI-L2, and RPMI 1788 cells were all resistant to lysis. In addition, low density fractions from Percoll density gradients that were enriched for NK effector cells also released increased levels of NKCF during coculture with Molt-4 cells. Lysis of Molt-4 and K562 targets was observed after exposure to NKCF for 48 hr and 60 to 70 hr, respectively. Cellfree supernatants containing NKCF were obtained after a short time of incubation (i.e., within 5 hr of coculture of PBL with NK target cells). The factors were nondialyzable, stable at 56 degrees C for 3 hr, and showed partial loss of activity on storage at 4 degrees C or -20 degrees C for 7 days. These data suggest that NKCF may be involved in the lytic mechanism of human NK cell-mediated cytotoxicity.