Uninfected Target Cells Research Articles

Cytotoxic lymphocytes generated in vivo with acute measles virus infection

We studied the generation of cytotoxic lymphocytes in adults during an outbreak of acute measles virus infection. Nine patients were studied determining in particular whether virus-specific cytotoxic T lymphocytes could be directly detected in peripheral blood during this acute infection. The cytotoxicity of PBL was assayed against measles virus-infected and uninfected phytohemagglutinin-induced blast cells of matched and mismatched HLA, A, B, and C types, in a standard 4-h 51Cr release assay. There was greater cytotoxicity against measles virus-infected than uninfected target cells in at least one sample from every patient. In 4 patients this preferential lysis of virus infected cells was greater (a difference of more than 10% virus-specific lysis) against HLA-matched than mismatched targets. This preference for HLA A and B matched infected target cells was also clearly seen when the effector PBL were depleted of FC receptor bearing cells. The other 5 subjects exhibited no evidence of preferential lysis of HLA-matched measles virus-infected cells. All 9 patients limited the spread of measles virus infection and recovered equally from the acute infection. These studies provide some evidence to suggest that MHC-restricted virus-specific CTL are detectable in human peripheral blood during acute measles virus infection, albeit only with low frequency, but are not necessarily associated with recovery from disease.

Gamma interferon production and cytotoxicity of spleen cells from mice infected with Semliki Forest virus.

Interferon (IFN) production by spleen cells from normal mice, mice acutely infected with Semliki Forest virus (SFV) or mice immune to SFV was measured after stimulation in vitro with either infectious or inactivated SFV. All three classes of spleen cells made IFN-alpha beta in response to infectious SFV. Spleen cells taken from mice late, but not early, after infection, or from immune mice, made IFN-gamma in response to inactivated SFV. Amounts of INF-gamma and IFN-alpha beta were similar. Normal spleen cells made no IFN (of any type) in response to inactivated SFV. The cell type producing IFN-alpha beta appeared to be the macrophage, whilst both T-lymphocytes and macrophages were necessary for IFN-gamma production. During the acute infection, the ability of spleen cells to lyse both virus-infected and uninfected target cells arose earlier than the ability to produce IFN-gamma. However, cytotoxicity towards uninfected cells fell to near background levels by day 7, whilst cytotoxicity towards infected targets remained high at that time, when IFN-gamma production was at its peak. IFN-gamma production is therefore temporally associated with cytotoxicity specifically directed against virus-infected targets, and the ability to produce IFN-gamma is a late response to SFV infection.

Antibody-dependent cellular cytotoxicity of Coxiella burnetii-infected J774 macrophage target cells.

Cell-mediated immunity is clearly the critical host defense mechanism against human Coxiella burnetii infection (Q fever); the role of specific antibody is unclear. By using a mouse macrophage tumor cell line, J774, persistently infected with C. burnetii phase I organisms, in a standard 51Cr-release cytotoxicity assay, we explored the possibility that antibody-dependent cellular cytotoxicity may be immune mechanism in Q fever. After 16 h of incubation in the presence of immune sera from Q fever hepatitis or endocarditis patients, nonimmune human peripheral blood effector cells specifically lysed infected J774 target cells; no 51Cr release was seen in the presence of nonimmune sera or uninfected target cells. An effector/target ratio of at least 5:1 was required, and monocytes were more efficient effector cells than lymphocytes. Cytotoxicity was blocked by preincubation of effector cells with purified aggregated human immunoglobulin G, indicating the role of Fc receptor-bearing effector cells. Two nonphagocytic lymphoid tumor cell targets, passively coated with C. burnetii, did not induce substantial immune-specific cytolysis, suggesting that bystander lysis does not explain the observation of specific lysis. Although antibody-dependent cellular cytotoxicity may participate in primary defense, alternatively, it may facilitate the dissemination of C. burnetii or surreptitiously participate in granuloma formation.

Sendai-virus-induced cell-mediated cytotoxicity in vitro. The role of viral glycoproteins in cell-mediated cytotoxicity.

Treatment of peripheral blood lymphocytes from normal donors with small amounts of purified Sendai virions results in enhanced cellular cytotoxicity in vitro to uninfected tissue culture target cells (virus-dependent cellular cytotoxicity (VDCC)), without any obvious correlation to the natural cytotoxicity (NK) displayed by the lymphocytes in the absence of virus. Removal from the virions of the two surface components present in the viral envelope, the HN glycoprotein (gp 71), carrying haemagglutinating and neuraminidase activity, and the F glycoprotein (gp 49), carrying fusion activity, by treatment with pronase abrogated their capacity to induce VDCC. Similar results were obtained when virions lacking the HN glycoprotein after treatment with chymotrypsin were added to the lymphocytes. In contrast, treatment of the virus particles with trypsin, which removed the F glycoprotein, did not affect their capacity to induce VDCC. When the solubilized and separated peplomers were used for lymphocyte treatment, either alone or in combination, the purified HN glycoprotein had full capacity to induce VDCC, whereas the F glycoprotein was inactive. These results suggest that the HM peplomer is solely or primarily responsible for the cytolytic activity arising in non-sensitized lymphocytes when confronted with certain viruses.

Effects of Interferon on Natural Killing and Antibody-Dependent Cellular Cytotoxicity Against Varicella-Zoster Virus-Infected and Uninfected Target Cells

Pretreatment of normal adult peripheral blood mononuclear cells with interferon-α (IFN-α) resulted in enhanced natural killing (NK) of varicella-zoster virus (VZV)-infected and uninfected target cells. In contrast antibody-dependent cell-mediated cytotoxicity (ADCC) against VZV-infected targets was not augmented by preincubation of effector cells with IFN-α. Preincubation of uninfected target cells with IFN-α significantly reduced their susceptibility to NK. In contrast, VZV-infected target cells preincubated with IFN-α remained fully susceptible to lysis by NK and also by ADCC. Since IFN has been detected in vesicular fluid and plasma of individuals with VZV infections, the in vitro mechanisms detected in this study may have biological relevance.

The role of interferon in the NK cell killing of virus-infected target cells

Spleen cells from uninfected CBA mice are more cytotoxic for Sendai virus-infected L929 cells than for uninfected cells and the lymphocytes responsible have the properties of NK cells. Preincubation of spleen cells with culture supernatants from Sendai virus-infected L929 cells increases the cytotoxicity for uninfected target cells. This increase in cytotoxicity can also be produced by pretreatment with purified mouse interferon. The enhancing effect of both the infected culture supernatants and purified interferon can be neutralized with anti-interferon serum. It is concluded that the preferential killing of Sendai virus-infected L929 cells by NK cells is dependent on the induction of interferon and that interferon will increase NK cell cytotoxicity for uninfected target cells.

Measles- or mumps virus-infected cells forming rosettes with lymphocytes from patients with multiple sclerosis.

Peripheral blood lymphocytes (PBLs) from patients with multiple sclerosis (MS) were studied to determine the frequency at which they formed rosettes with target cells persistently infected with measles or mumps virus. Results were compared with (1) the rosette-forming capability of lymphocytes from age- and sex-matched normal control subjects and (2) the rosette-forming capability of lymphocytes with uninfected target cells from patients with MS. Comparison of mean measles antibody titers in patients with MS was significantly higher than in control subjects. A similar comparison for mumps antibodies showed a significant differences. There was no significant difference between patients and control subjects in the frequency of lymphocytes that formed rosettes, no matter which target cell was used. When data obtained using target cells infected with measles were analyzed according to sex or clinical classification, no significant difference was observed. Lymphocytes from patients or control subjects formed significantly more rosettes when reacted with virus-infected rather than uninfected target cells. These data suggest that PBL rosette formation with measles- or mumps-infected cells may represent nonspecific adherence rather than specific adherence.

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.

Natural cell-mediated cytotoxicity of bovine mononuclear cells against virus-infected cells.

The ability of mononuclear cells (MC) from peripheral blood of normal cattle to lyse a variety of cells was tested in a 51Cr-release microcytotoxicity assay. Several types of bovine cells infected with parainfluenza 3 virus (PI3V) were susceptible to natural cytotoxicity. Bovine cells infected with infectious bovine rhinotracheitis virus or noncytopathogenic bovine viral diarrhea virus, uninfected bovine cells, and human cell lines MOLT-3, HSB-2, K562, and U-937 were not susceptible. The period of time that target cells need to be infected with PI3V to achieve maximal cytotoxicity was determined. Target cells were infected with PI3V and MC, added 1 h later. After the addition of effector cells, significant levels of cytotoxicity were recorded at 17 h. Maximal cytotoxicity occurred 22 to 24 h postinfection. To define the optimal time that MC must be present, cells were infected with PI3V for a total of 24 h, and MC were left in contact with target cells for various time intervals. Maximal cytotoxicity was recorded when effector cells were present for 20 h, suggesting that a period of activation was needed to stimulate effector cell function. Removal of adherent mononuclear cells on Sephadex G-10 columns did not reduce the low level of cytotoxicity against uninfected target cells, but markedly reduced the level of cytotoxicity against PI3V-infected cells. The effector cell was nonphagocytic and nonadherent. These characteristics and the fact that target cell lysis was independent of genetic restriction indicate that effector cells are similar to natural killer cells described in other species.

Cytotoxic T cells are induced in mice infected with lymphocytic choriomeningitis virus strains of markedly different pathogenicities.

The ability of two lymphocytic choriomeningitis virus substrains to induce cytotoxic T-lymphocyte (CTL) responses in intracerebrally infected mice was examined. One strain, designated A (aggressive), provoked a convulsive type of death in 100% of the mice within 8 to 9 days, whereas the other strain, designated D (docile), killed less than 10% of the mice during 28-day observation periods. CTL activity was assessed by the capacity of partially purified splenocytes to lyse 51Cr-labeled L-cell targets infected with either type of lymphocytic choriomeningitis substrain. The CTL population was identified by its sensitivity to anti-Thy-1 serum and its inability to lyse uninfected target cells or infected target cells with which it differed at the level of antigens controlled by the major histocompatibility gene complex. A strong CTL response developed in mice infected with either lymphocytic choriomeningitis substrain, although the activity provoked by substrain D was somewhat less than that seen after substrain A infection. Peak CTL activities induced by both strains occurred at about the same time. Even though docile virus replicated more extensively in the brain than did aggressive virus and fluorescent antibody staining revealed similar distributions of viral antigen, no inflammatory response was noted in the brains of mice infected with docile virus. These results are discussed with regard to the role of CTLs in mediating classic central nervous system pathology.

Hormone-dependent terminal differentiation in vitro of chicken erythroleukemia cells transformed by ts mutants of avian erythroblastosis virus

Chicken erythroblast cell strains and a cell line transformed by ts mutants of avian erythroblastosis virus (AEV) terminally differentiate when shifted to the nonpermissive temperature (42°C). The differentiated cells resemble mature erythrocytes with respect to morphology and ultrastructure, expression of differentiation-specific cell-surface antigens, pattern of protein synthesis and hemoglobin content. Terminal differentiation is dependent on conditions favoring the differentiation of normal erythroid progenitor cells, including an erythropoietin-like factor. Colonies of ts AEV cells grown at 42°C in semisolid medium resemble erythrocyte colonies derived from normal erythroid progenitor cells. The colonies obtained were comparable in size or slightly larger than the late erythroid precursor (CFU-E) colonies. These results suggest that AEV-transformed cells are blocked at a stage of differentiation that is more advanced than that of the uninfected target cells. ts AEV cells are irreversibly committed to terminal differentiation within 20 to 30 hr after shift to 42°C.

Cytotoxic cells induced after Chlamydia psittaci infection in mice.

The ability of spleen cells from Chlamydia psittaci-infected mice to lyse C. psittaci-infected and uninfected target cell monolayers was studied. The cytotoxicity assay used was a terminal label method in which the number of adherent target cells surviving the interaction with effector cells was determined by measuring the uptake of [3H]uridine by such cells. It was observed that in the first few days postinfection (3 to 5), spleens contained cells that lysed infected and uninfected targets with equal efficiency. Subsequently, infected targets were killed primarily. The activity of effector spleen cells for infected targets continued, although at a reduced level, beyond 21 days postinfection. Intact effector cells were required since a disruption by sonication resulted in a loss of cytotoxicity. The enhanced killing observed with infected targets was also observed when target cells were sensitized with heat- or UV-inactivated C. psittaci. This study suggests that the induction of cytotoxic cells after C. psittaci infection may contribute to the ability of the host to control multiplication of the microorganism.

Natural cytotoxicity of murine cytomegalovirus-infected cells mediated by mouse lymphoid cells: role of interferon in the endogenous natural cytotoxicity reaction.

Lymphoid cells from unstimulated normal C57BL/6J mice were shown to lyse murine cytomegalovirus (MCMV)-infected syngeneic mouse embryo fibroblasts but not uninfected mouse embryo fibroblasts. This cytotoxicity by mouse effector cells was not restricted to MCMV-infected syngeneic cells since MCMV-infected xenogeneic rat heart fibroblasts were also lysed. Characterization of the effector cells mediating this cytotoxicity against MCMV-infected cells indicated that the effector cells are similar to described natural killer (NK) cells mediating lysis of tumor cells and virus-infected cells. Because of the described augmentation of NK activity by interferon, we examined the role of interferon in the NK reaction. Although low levels of virus-induced interferon were detectable in supernatants of MCMV-infected mouse embryo fibroblasts, no interferon was detectable in supernatants of MCMV-infected rat heart fibroblasts, a target significantly more sensitive to NK cytolysis than infected mouse embryo fibroblasts. We were able to augment the NK reaction against MCMV-infected cells by in vitro treatments with interferon. However, the amounts of interferon required for augmentation were significantly greater than the amounts generated by infected target cells. In vitro interferon-stimulated NK cells retained selective cytotoxic activity since they continued to remain incapable of lysing uninfected target cells. MCMV-infected rat heart fibroblasts induced more interferon and were also more susceptible to NK activity than MCMV-infected mouse embryo fibroblasts. In spite of this difference in interferon-inducing capacity, there was no augmentation of cytotoxicity of MCMV-infected mouse embryo fibroblasts when mouse splenocytes were cocultivated with both target cells. Finally, when production of interferon in the NK reaction was inhibited by the addition of actinomycin D, no reduction of NK activity was seen. Our findings suggest that native mouse NK cells can discriminate between MCMV-infected cells and uninfected cells, this ability leading to the selective lysis of the virus-infected cells. Furthermore, although we could demonstrate augmentation of NK activity by interferon, interferon activation of NK cells may not be a necessary precondition for the development of endogenous NK activity.

Induction of natural killer cells during murine influenza virus infection

Cells which are cytotoxic for both virus-infected and uninfected target cells can be recovered from the spleens of mice injected with either infectious or non-infectious influenza A virus. Peak activity occurs at 1-2 days and decreases to low levels by day 6. The effector cells are insensitive to anti-Thy 1 antibody and complement treatment, are not H-2 restricted, do not adhere to plastic and are unaffected by silica or carrageenan. In this sense and in the pattern of susceptibility to lysis of a variety of cultured cell lines, these effector cells have the properties of natural killer (NK) cells and are referred to as such. They are present to an increased level of activity in nude (nu+/nu+) mice and to a low level of activity in beige (bg+/bg+) mice, but upon injection of virus there is a significant increase in activity in both hosts. Such cells were also recovered from the lungs of mice infected intranasally with a lethal or sublethal dose of virus. In the former case, maximum activity was reached 2 days post infection and the activity remained high until death; in the latter case, peak activity was reached 4 days after virus inoculation and by day 11 the activity had decreased to pre-infection levels. After intranasal inoculation of influenza virus, both beige mice and their heterozygous littermates contained similar levels of infectious virus in their lungs. However, this result does not eliminate the possibility that these cells may help to limit virus infection.

Spontaneous cytotoxicity against virus-infected cells: relationship to NK against uninfected cell lines and to ADCC.

Cellular immunoadsorption and cold cell competition assays were used to examine the specificity of NK. Lymphocytes nonadherent to monolayers of cells persistently infected with Sendai virus (SV-HEp2) were almost totally depleted of NK against the uninfected tumor cell lines K562, MOLT-4, and HSB, as well as against SV-HEp2, when assayed immediately after separation. Lymphocytes adherent to SV-HEp2 monolayers were also partially depleted of NK against these targets. After overnight incubation, the nonadherent cells remained relatively depleted of NK against all targets, compared to controls. The adherent cells displayed a variable amount of activity, depending on the donor; in all cases, however, substantial NK activity was observed in this fraction against all targets, infected and uninfected. Exogenous interferon failed to augment the activity of the nonadherent lymphocyte fraction. Cellular immunoadsorption on monolayers of K562 or HSB was similarly found to partially deplete NK against SV-HEp2, as well as against the homologous cells. Cold cell inhibition studies showed that SV-HEp2 and, to a lesser extent, HEp2, could inhibit the lysis of K562; similarly, K562 could partially inhibit the lysis of SV-HEp2. We concluded that infected and uninfected target cells are lysed by overlapping populations of NK effector cells. In addition, the ADCC activity of SV-HEp2-adherent and nonadherent lymphocyte fractions was examined immediately after separation. Whereas the nonadherent lymphocytes were depleted of both ADCC and NK, the adherent lymphocytes were partially depleted of NK yet contained high levels of ADCC activity.

Interferon induction of natural killer cytotoxicity in human neonates

Low natural killer cytotoxicity has been associated with serious herpes simplex virus infections. Neonates' mononuclear cells had significantly lower NKC to HSV infected (P less than 0.001) and uninfected (P less than 0.005) target cells than did adults' MC. Under appropriate conditions, both neonates' MC to NKC was increased by exogenous human leukocyte interferon. Although neonates' and adults' MC had a similar sensitivity to interferon, there was a significant (P = 0.05) lack of consistency in neonates' MC response. Twenty-three percent (4/17) of neonates' MC samples did not respond with increased NKC in the presence of interferon, unlike adults' MC, all of which demonstrated increased NKC. Neonates' MC did not suppress adults' MC-NKC. These data demonstrate the efficacy of interferon as an NKC stimulator in the neonate, but indicate a heterogeneous response of neonatal cells. We are encouraged with the prospect of clinical trials of interferon to treat severe viral infections in neonates. We urge the inclusion of immune surveillance in these trials.

A terminal labelling assay using 3H-uridine to measure cell-mediated cytotoxic responses against parainfluenza type 3 infected target cells

The standardisation of a terminal labelling cytotoxicity assay which used the incorporation of 3H-uridine to measure target cell survival of the effector phase is described. This assay was applied to a preliminary study of the cell-mediated responses of sheep following intravenous inoculation of Pi-3 virus. The assay provided a simple, accurate, and reproducible method to measure cytotoxic responses against Pi-3 infected target cells. The experimental data did not require mathematical corrections for non-specific effects as in the 51 Cr-release assay. When cytotoxicity could be detected in uninfected target cell cultures it was low and statistically insignificant

Genetically restricted cell-mediated cytotoxicity in cattle immune to Theileria parva.

The protozoan parasite Theileria parva, which is transmitted by the tick Rhipicephalus appendiculatus, produces an acute fatal infection in the lymphoid system of susceptible cattle1. This disease is a serious constraint to livestock improvement and production in large areas of East Africa1. The parasite invades host lymphocytes, inducing rapid proliferation followed by widespread lymphocytolysis2. Cattle which recover from theileriosis (East Coast fever) spontaneously, or which are immunized by infection and treatment with tetracycline3, are resistant to reinfection with the same isolate of T. parva for at least 3 yr4. Immunity against infection with the parasite cannot be ascribed to the production of specific antibodies5, but can be transferred adoptively between twins with thoracic duct leukocytes from the immunized partner6. These observations suggest that protective immunity is associated with cell-mediated mechanisms. We have now examined the capacity of leukocytes from immune cattle to lyse parasitized lymphoblastoid and non-parasitized tumour cell lines either directly or after stimulation in an autologous mixed lymphocyte reaction (MLR). In contrast to the nonspecific lytic activity of leukocytes from immune cattle reported by Pearson et al.7, we describe the sequential appearance in the lymph and blood of immune cattle, of cytotoxic leukocytes with activity restricted to target cells carrying the autologous genotype. These observations suggest that a major component of protective immunity to T. parva is mediated by cytotoxic cells which lyse parasitized cells in a genetically restricted fashion. We have also found that during a primary infection with T. parva cytotoxicity was manifested against allogeneic parasitized cells and xenogeneic uninfected target cells, but not against autologous infected cells. The features of cell-mediated immunity to T. parva during primary infection and immunization are discussed more fully elsewhere8.

Effect of herpes simplex virus infection on murine antibody-dependent cellular cytotoxicity and natural killer cytotoxicity.

Mice intraperitoneally inoculated with a sublethal dose of herpes simplex virus (HSV) produced immunoglobulin G antibody-dependent cellular cytotoxicity (ADCC) and radioimmunoassay (RIA) antibody as early as 3 days after infection. There was a rise in natural killer cytotoxicity (NKC) to infected and uninfected target cells 1 to 3 days postinfection mediated by nonadherent peritoneal cells (PC) in mice inoculated with HSV, but also with other substances commonly used in tissue culture media. HSV caused the highest and most consistent increase in NKC. PC-NKC, as ADCC, was inhibited by latex and silica, both macrophage inhibitors. PC-ADCC markedly declined 3 to 8 days after HSV inoculation. This was not due to a soluble or cellular suppressor factor, was not reversed by incubation or trypsin treatment of PC, was not associated with a change in PC Fc receptors, adherence, or acridine orange staining characteristics, and could not be induced by inactivated HSV. In vitro inoculation of PC with HSV similarly caused a reduction in the ability of PC to mediate ADCC to HSV-infected target cells. These data demonstrate the complex stimulatory and inhibitory interactions between virus and host defense mechanisms.

Generation on Infected Fibroblasts of Human T and Non-T Lymphocytes with Specific Cytotoxicity, Influenced by Histocompatibility, against Measles Virus-Infected Cells

Abstract Human peripheral blood lymphocytes (PBL) were incubated (sensitized) for 6 days on autologous or allogeneic fibroblast monolayers either uninfected or infected with measles virus. The sensitized PBL were then tested, in a 6-hr 51Cr-release assay, against autologous and allogeneic measles virus-infected and uninfected fibroblast target cells. When the PBL had been sensitized on autologous infected monolayers, they were cytotoxic to both autologous and allogeneic measles-infected targets but not to uninfected targets or to targets infected with paramyxovirus I (Sendai virus). Likewise, PBL sensitized on uninfected autologous (or allogeneic) monolayers had no cytotoxic activity against any of the infected or uninfected target cells. When the PBL had been sensitized on allogeneic infected monolayers, they were cytotoxic to measles-infected targets that were autologous either to the PBL themselves or to the sensitizing monolayer. These sensitized PBL were not, however, cytotoxic against third-party infected targets that were totally allogeneic to both the PBL and the sensitizing monolayer. Subpopulation depletion studies revealed at least two different effector lymphocytes in the sensitized PBL. One such effector was a T cell whose activity would be generated on autologous measles-infected monolayers but not on allogeneic infected monolayers. The cytotoxic activity of this cell was measles virus specific, and it required that the effector recognize as self antigens of the MHC on the target cell. The second effector cell also required in vitro generation but could be so generated on both allogeneic and autologous measles-infected, but not uninfected, monolayers. It was a non-T cell that had a receptor for the Fc fragment of IgG but lacked nonspecific esterase activity. This effector was measles virus specific, in that it failed to lyse either uninfected targets or targets infected with Sendai virus (the latter of which could be lysed by fresh PBL). When this non-T effector was generated on allogeneic infected monolayers, its cytotoxic activity showed an unusual form of MHC-related specificity, in that it lysed infected target cells autologous to either the PBL or the sensitizing monolayer but not infected third-party target cells allogeneic to both. Evidence was discussed which argued that our non-T effector was neither a conventional K nor NK cell.