Opsonic activity of human immune serum on in vitro phagocytosis of Plasmodium falciparum infected red blood cells by monocytes.

In vitro human monocytes from normal blood donors ingest red blood cells infected with Plasmodium falciparum more efficiently than normal red blood cells (NRBC). The phagocytic activity of human monocytes for infected red blood cells (IRBC) is greatly enhanced by the addition of immune sera obtained from individuals living in areas with endemic malaria. In contrast, the addition of sera obtained from individuals recovering from a first infection, or pooled normal sera, does not result in increased phagocytosis of IRBC. The phagocytosis enhancing activity of immune sera is associated with the IgG fraction and IgG depleted sera do not stimulate phagocytosis. Enhanced immune serum mediated phagocytosis occurs as a result of opsonization of IRBC. This was demonstrated by experiments in which monocytes or IRBC were preincubated with immune serum prior to the phagocytic assay. The opsonic activity could be absorbed by IRBC but not by NRBC. The opsonization of IRBC and subsequent phagocytosis were also dependent on the stage of development of the intracellular parasite. IRBC containing schizonts and trophozoites were preferentially phagocytosed as compared with ring forms. The role of malaria induced surface alterations and/or malaria surface antigens in the opsonization of IRBC by immune sera is discussed. These experiments suggest that phagocytosis of P. falciparum IRBC by monocytes may play a role in the immune elimination of malaria infection in humans.     

Capacity of passively administered antibody to prevent establishment of Brucella abortus infection in mice.

In contrast to immunity against some other facultative intracellular parasites, protective immunity against Brucella abortus is mediated in mice by antibodies as well as by cell-mediated immune responses. It was the purpose of this study to determine whether antibody alone would prevent infection with B. abortus. The majority (82%) of CD-1 outbred mice infected with 100 CFU of virulent B. abortus 2308 preincubated with graded quantities of an O polysaccharide-specific IgG2a monoclonal antibody (MAb) were free of infection 1. 2, 4, and 6 weeks later, based on detection limits of 13 brucellae per spleen and 39 per liver. Infection was present in 95% of control animals. Similar results were obtained with a challenge dose of 500 CFU, but with a challenge dose of 5,000 CFU, infection became established even with the highest concentration of MAb used (50 micrograms of MAb per 5,000 brucellae). Pretreatment with an O polysaccharide-specific IgG1 MAb or with convalescent-phase serum diminished but did not prevent establishment of infection by 100 CFU of B. abortus. A majority of culture-negative mice tested 6 weeks after infection were serologically negative, which could have signified either the absence of previous infection or the early elimination of infection. In an in vitro test system, all of the antibody preparations were efficient in opsonizing B. abortus. Effective killing of the organism by unelicited mouse peritoneal macrophages occurred in conventional but not in endotoxin-free medium, suggesting that activated macrophages were required for killing of opsonized B. abortus. These results emphasize the potential importance of antibodies in the immunoprophylaxis of brucellosis and suggest that the design of a successful vaccine will require the induction of antibodies not only of appropriate specificity but also of the optimal isotype for mediating protective functions.     

Ex vivo desequestration of Plasmodium falciparum-infected erythrocytes from human placenta by chondroitin sulfate A

We performed ex vivo experiments with Plasmodium falciparum-infected human placentas from primi- and multigravida women from Cameroon. All women, independent of their gravida status, had anti-chondroitin sulfate A (CSA) adhesion antibodies which cross-reacted with heterologous strains, such as FCR3 and Palo-Alto(FUP)1, which were selected for CSA binding. These antibodies, directed against the surface of infected erythrocytes obtained by flushing with CSA (IRBC(CSA)), were restricted to the immunoglobulin G3 isotypes. Massive desequestration of parasites was achieved with soluble CSA but not with anti-ICAM-1 and anti-CD36 monoclonal antibodies. All of the CSA-flushed parasites were analyzed immediately by using in vitro assays of binding to Saimiri brain endothelial cells (SBEC) expressing various adhesion receptors. Parasites derived from all six placentas displayed the CSA adhesion phenotype. However, only partial inhibition of adhesion was observed in the presence of soluble CSA or when Sc1D SBEC were treated with chondroitinase ABC. These results suggest that an additional adhesive molecule of IRBC(CSA) which binds to an unidentified receptor is present in the placenta. This new phenotype was lost once the parasites adapted to in vitro culture. We observed additional differences in the CSA adhesion phenotype between placental parasites and in vitro-cultured parasites panned on endothelial cells carrying CSA. The minimum size of fractionated CSA required for a significant inhibition of placental IRBC(CSA) adhesion to Sc1D cells was 1 to 2 kDa, which contrasts with the 4-kDa size necessary to reach equivalent levels of inhibition with panned IRBC(CSA) of this phenotype. All placental IRBC(CSA) cytoadhered to Sc17 SBEC, which express only the CSA receptor. Panning of IRBC(CSA) on these cells resulted in a significant quantitative increase of IRBC cytoadhering to the CSA of Sc1D cells but did not change their capacity for adhesion to CSA on normal placenta cryosections. Our results indicate that the CSA binding phenotype is heterogeneous and that several distinct genes may encode P. falciparum-CSA ligands with distinct binding properties.     

Dendritic cells induce immunity and long-lasting protection against blood-stage malaria despite an in vitro parasite-induced maturation defect

Dendritic cells (DC) suffer a maturation defect following interaction with erythrocytes infected with malaria parasites and become unable to induce protective malaria liver-stage immunity. Here we show that, by contrast, maturation-arrested DC in vitro are capable of the successful induction of antigen-specific gamma interferon (IFN-gamma) and interleukin 4 (IL-4) T-cell responses, antibody responses, and potent protection against lethal blood-stage malaria challenge in vivo. Similar results were found with DC pulsed with intact parasitized Plasmodium yoelii or Plasmodium chabaudi erythrocytes. Cross-strain protection was also induced. High levels of protection (80 to 100%) against lethal challenge were evident from 10 days after a single immunization and maintained up to 120 days. Interestingly, correlation studies versus blood-stage protection at different time points suggest that the immune effector mechanisms associated with protection could change over time. Antibody-independent, T-cell- and IL-12-associated protection was observed early after immunization, followed by antibody and IL-4-associated, IFN-gamma-independent protection in long-term studies. These results indicate that DC, even when clearly susceptible to parasite-induced maturation defect effects in vitro, can be central to the induction of protection against blood-stage malaria in vivo.     

Immunobiological behaviour of rabbit precipitating and non-precipitating (co-precipitating) antibodies.

Complement fixing capacity, antibody-dependent cell cytotoxicity and in vivo opsonic function of rabbit anti-DNP precipitating and co-precipitating antibodies were analysed. Precipitating antibody activated the complement system, but co-precipitating antibody did not. Both antibody preparations behaved similarly in antibody-dependent cytotoxicity reactions. When chicken erythrocytes were employed as target cells, both antibodies were inactive with homologous (rabbit) lymphocytes as effector cells but were active with heterologous lymphoid effector cells (human, guinea-pig). Clearance of antigen from the blood of mice was accelerated following injection of an optimal dose of precipitating antibody but co-precipitating antibody was ineffective even at different doses. Competition between precipitating and co-precipitating antibodies in complement fixation and antigen elimination following interaction with antigen successively or at the same time was investigated. In these tests, both antibodies competed with antigen by mass but not by affinity. The possibility exists that co-precipitating antibody has one high and one low affinity binding site and thus acts as an univalent blocking antibody and is therefore not able to form active complexes with antigen. The possible role of co-precipitating antibody in some humoral and cellular immune mechanisms is discussed.     

Opsonization as an effector mechanism in human protection against asexual blood stages of Plasmodium falciparum: functional role of IgG subclasses.

A phagocytic assay performed with human peripheral mononuclear cells and malaria-infected erythrocytes enabled the study of opsonizing antibodies in human serum from donors presenting different levels of protective immunity. Opsonizing activity was found in sera from individuals who could be considered immune, i.e. asymptomatic parasite carriers and subjects residing in endemic regions who presented neither symptoms nor parasites. This contrasted with subjects showing an absence of symptoms or who had experienced only a single malarial infection. All sera contained high levels of antimalarial antibodies, as shown by immunofluorescence assay (IFA). IgG of different subclasses were immunopurified from these sera. All subclasses were shown to bind to the surface of infected erythrocytes by FACS analysis, but only IgG1 and IgG3 were able to mediate opsonization. IgG2 and IgG4 did not opsonize and inhibited the opsonizing activity of IgG1 and IgG3 in competition experiments. These results suggest the existence of a correlation between immune protection, the ability of serum to mediate opsonization of infected erythrocytes and the predominance, in this serum, of IgG1 and IgG3 over IgG2 and IgG4 directed against the surface of the infected erythrocytes.     

Opsonization as an effector mechanism in human protection against asexual blood stages of Plasmodium falciparum: Functional role of IgG subclasses

A phagocytic assay performed with human peripheral mononuclear cells and malaria-infected erythrocytes enabled the study of opsonizing antibodies in human serum from donors presenting different levels of protective immunity. Opsonizing activity was found in sera from individuals who could be considered immune, i.e. asymptomatic parasite carriers and subjects residing in endemic regions who presented neither symptoms nor parasites. This contrasted with subjects showing an absence of symptoms or who had experienced only a single malarial infection. All sera contained high levels of antimalarial antibodies, as shown by immunofluorescence assay (IFA). IgG of different subclasses were immunopurified from these sera. All subclasses were shown to bind to the surface of infected erythrocytes by FACS analysis, but only IgG1 and IgG3 were able to mediate opsonization. IgG2 and IgG4 did not opsonize and inhibited the opsonizing activity of IgG1 and IgG3 in competition experiments. These results suggest the existence of a correlation between immune protection, the ability of serum to mediate opsonization of infected erythrocytes and the predominance, in this serum, of IgG1 and IgG3 over IgG2 and IgG4 directed against the surface of the infected erythrocytes.     

Demonstration of the role of cytophilic antibody in resistance to malaria parasites (Plasmodium berghei) in rats.

This paper reports the results of a study of the nature of the immune response against Plasmodium berghei parasites by inbred rats. A macrophage-cytophilic antibody specific for malarial antigens was identified and characterized. Detection of the antibody on the macrophage surface was accomplished by the parasite adherence tests and by the indirect fluorescent antibody technique. Isolation and purification of the macrophage-cytophilic and opsonic antibodies from hyperimmune rat serum was accomplished by QAE-Sephadez A-50 elution chromatography, and of the macrophage-cytophilic antibody by adsorption with and elution from syngeneic macrophages as well. Characterization of the cytophilic antibody as immunoglobulin G1 was done by immunoelectrophoresis and by Ouchterlony-type double diffusion in gel. Passive protection tests in weanling inbred rats have demonstrated that the opsonizing antibody conferred some protection against P. berghei. The macrophage-cytophilic antibody, on the other hand, was not protective alone but acted synergistically with the opsonizing antibody.     

Babesia bovis: In vitro phagocytosis promoted by immune serum and by antibodies produced against protective antigens

The in vitro phagocytosis of bothBabesia bovis-infected red cells and of parasites exposed by lysis of infected red blood cells is demonstrated in a phagocytic mouse model. Twenty-fourB. bovis immune sera were tested alone or as a pool as were antibodies (DS antibodies) raised against aB. bovis protective fraction, prepared by dextran sulfate precipitation. All the immune sera failed to promote significant levels of phagocytosis, whereas the other antibodies (DS antibodies) consistently induced phagocytosis of infected cells in all the experiments carried out. This study shows that antibody specificity is critical to the opsonization of infected red cells and parasites during in vitro phagocytosis and suggests that phagocytosis is one of the mechanisms in the in vivo immune response againstBabesia species.     

Non-Determinant Specificity of Feedback Immunosuppression by IgG Antibodies Injected after the Antigen

The determinant specificity of the IgG-mediated suppression of the humoral immune response in mice was studied. One hour before or 2.5, 6, 12, or 24 h after the injection of sheep erythrocytes (SRBC) or SRBC-TNP, CBA/Ca mice received SRBC-specific monoclonal IgG antibodies. The antibodies did not cross-react with TNP or goat erythrocytes, the latter an antigen which shows 30% cross-reactivity with SRBC. Five days later the determinant-specific plaque-forming cell response against SRBC and the non-determinant-specific response against goat erythrocytes and TNP were determined. Regardless of whether the antibodies were injected before or after the antigen, they suppressed not only the response to the antigenic determinant they bound to, but also the response to other epitopes on the same antigen. This shows that Fc parts of the IgG molecules play a crucial part in suppression of the in vivo antibody response even when, as in a natural situation, the antigen is presented to the immune system before the antibody.     

Intrasplenic immunization with infected hepatocytes: a mouse model for studying protective immunity against malaria pre-erythrocytic stage.

Malaria liver forms are the target of antibody or T-cell-mediated immune mechanisms induced by previous or subsequent developmental stages of the parasite. The potential for vaccine development of antigens expressed exclusively in the liver stages has not been fully explored partly because of the lack of an experimental animal model. Here we show that protective immunity against sporozoite-induced infection with Plasmodium yoelii and P. berghei can be obtained by intrasplenic injection of a small number of liver stages of the parasites. The serum of the protected animals did not contain antibodies against sporozoites, liver or blood stage malaria parasites. Protective immunity was abolished by depletion of either CD4+ or CD8+ T cells from the vaccinated mice before challenge.     

Immune depression and macroglobulinemia in experimental subchronic trypanosomiasis.

The effects of subchronic trypanosomiasis upon immune responses were examined in Trypanosoma gambiense infection and in subcurative treatment of T. brucei- and T- equiperdum-infected mice. About 60% of the mice infected with T. gambiense developed a subchronic infection similar to human trypanosomiasis, characterized by the absence of circulating trypanosomes. The animals died between 1 and 12 months after infection with elevated serum immunoglobulin M (IgM) levels (16 times the normal level). After 1 month of infection, the mice showed a normal primary antibody response against sheep erythrocytes, as tested by hemagglutination, despite their high serum IgM levels. After more than 1 month of infection, about 20% of the mice showed depressed hemagglutination titers (25% of control), whereas all relapsed mice that contained circulating parasites showed a pronounced suppression. Elimination of the blood parasites with Berenil treatment restored immune competence, which persisted until the relapse of the animals. Identical results were obtained in T. brucei-infected mice. Berenil treatment abolished the immune depression against sheep erythrocytes, but did not cure the animals, which relapsed with the development of a new state of immune depression. T. gambiense and T. brucei infections were always followed by a marked increase of serum IgM levels. Hypergammaglobulinemia was also induced in relapsing T. equiperdum-infected mice treated with Berenil. No immune depression against sheep erythrocytes could be detected. It appeared that immune depression was not the result of clonal exhaustion (measured by the serum IgM level) but seemed to be closely associated with the presence of living trypanosomes.     

Low incidence and high titers of antibodies to hepatitis B virus X-protein in sera of Chinese patients with hepatocellular carcinoma

Sera of patients from China with hepatocellular carcinoma (HCC) were tested for the presence of HBc/HBe- and HBx antibodies by immunoblotting using recombinant MS2 or beta gal fusion proteins as substrate. Antibodies against HBx were detected in four out of 68 HBsAg positive and in one out of three HBsAg negative sera, antibodies against HBc/HBe in 52 and two serum samples, respectively. Competition experiments in which sera were preincubated with individual viral proteins synthesized in E. coli were carried out to demonstrate the specificity of signals obtained in immunoblot analyses. In the five anti-HBx positive sera, the antibody titer against X fusion protein was higher than against core fusion protein and in one of these sera anti-x activity could be demonstrated even at a serum dilution of 1:50,000. These data indicate that X antibodies occur rarely in Chinese patients and are not serodiagnostic for HCC. The high titer of X antibodies in some patients shows that the X protein can be highly immunogenic in vivo. Induction of antibody formation may be triggered by X protein expressed from integrated viral DNA.     

Anti-idiotypic antibody with potential use as an Eimeria tenella sporozoite antigen surrogate for vaccination of chickens against coccidiosis.

Anti-idiotypic antibodies were raised in rabbits against four monoclonal antibodies with specificity for the surface antigenic determinants of Eimeria tenella sporozoites, the infective stage of the coccidial parasite. Two of the monoclonal antibodies (1073 and 15-1) transferred passive protection in chickens against E. tenella infection. The polyclonal anti-idiotype antibody preparations against protective monoclonal antibodies contained specificities for the paratope-associated idiotypes of these monoclonal antibodies, as assessed by the competitive inhibition of binding of the homologous idiotype-anti-idiotype by the sporozoite antigen. Competitive inhibition of binding of homologous idiotype-anti-idiotype by the parasite antigen was not observed when the anti-idiotype antibody preparations against monoclonal antibodies 1546 and 1096 were tested. The anti-idiotype 1073 and 15-1 antibodies functioned as surrogate antigens in vivo when used for vaccination of young chickens, as evidenced by the induction of partial protective immunity against subsequent challenge infection with virulent parasites and induction of antisporozoite antibodies. These data clearly support the view that anti-idiotypic antibodies raised against the paratope-associated idiotypes can mimic pathogen antigens and therefore can provide a possible alternative approach for the vaccination of chickens against coccidiosis.     

Acquirement of protective immunity in mice through infection with an attenuated isolate and its failure in parent virulentPlasmodium berghei

In virulentPlasmodium berghei infection, mice showed suppressive responses to sheep red blood cells SRBC (PFC) as well as the parasite antigen (DTH) and developed autoantibodies against homologous lymphocytes. On the other hand, mice infected with an attenuated variant derived fromP. berghei did not show these responses but developed solid protective immunity against parent parasite infection accompanied by high antibody titre. When such an immune serum was transferred into mice, attenuated parasite infection was completely eliminated. These results show that an attenuated variant stimulates antibody production, which contributes to protection against the parasites. In contrast, in virulentP. berghei infections harmful immunopathological responses against the host are more prominent than protective immune responses against the parasites.     

Rhesus monkeys protected against Plasmodium knowlesi malaria produce antibodies against a 65,000-MrP. knowlesi glycoprotein at the surface of infected erythrocytes.

Sera from 27 rhesus monkeys immunized in various ways against the H strain of Plasmodium knowlesi were analyzed by quantitative crossed immunoelectrophoresis. The reaction of the sera was compared with a reference immune serum only reactive with P. knowlesi-specific 65,000-Mr glycoprotein-immune component 13 (gp65/ic13) in membranes of infected rhesus monkey erythrocytes. Triton X-100-solubilized, 125I-labeled membranes of schizont-infected erythrocytes were used as an antigen. Sera from 9 or 10 monkeys immunized by repeated infections with P. knowlesi reacted with gp65/ic13. In 6 of 10 sera, anti-gp65/ic13 was the only antibody reacting with host cell membrane proteins. In contrast, vaccination of 15 monkeys with predominantly sexual stages or trophozoites of P. knowlesi in Freund complete adjuvant resulted in protection against blood challenges in 7 monkeys, only 2 of which contained precipitating antibody against gp65/ic13. None of the sera from monkeys not protected by infections or vaccinations contained detectable levels of precipitating antibodies against gp65/ic13. Our data indicate that gp65/ic13 acts as a prominent immunogen in vivo during natural p. knowlesi infections of rhesus monkeys. There is a positive correlation suggested between anti-gp65/ic13 antibody and protection in the monkeys analyzed. This correlation does not apply to monkeys protected against P. knowlesi malaria by vaccination, pointing to other effective immune defense mechanisms.     

Antiparasite adherence activity in Thai individuals living in a P. falciparum endemic area

Two types of antimalaria antibodies in the serum of 54 villagers living in a malaria endemic area of Thailand were determined by indirect immunofluorescence assay in order to define the status of malaria immunity within the group. Antibodies to parasite-derived antigens in the membrane of ring stage-infected erythrocytes were very high (> or = 1:1,250) in 44%, moderate to low (< or = 1:250) in 37% of the sera, and the rest did not have the antibody. However, all the sera had antibodies to antigens of the intraerythrocytic mature parasites, showing a very high level in 65% and moderate to low levels in 37% of the sera. Sera with high antibody titers to either type of antigen significantly inhibited cytoadherence of P. falciparum-infected erythrocytes. All the sera variably inhibited rosette formation of the parasites but showed no association with the antibody titers. These results suggest that the antibodies to cytoadherence and rosette formation can be elicited and sustained in the malaria experienced host while living in the endemic area. This may be a natural preventive mechanism against the severity of P. falciparum infection in the infected host. How long the antiparasite adherence activity will last remains to be investigated.     

Dominance of Immunoglobulin G2c in the Antiphosphorylcholine Response of Rats Infected with Trichinella spiralis

The antibody response to the L1 stage of Trichinella spiralis has been described as biphasic. Worms resident in the intestine during the first week of infection stimulate an antibody response against a subset of larval proteins. L1 larvae in the muscle at the end stage of infection stimulate a second antibody response against tyvelose-bearing glycoproteins. Antityvelose antibodies protect rats against challenge infection with larvae. The aim of this study was to characterize the rat B-cell response against larval antigens during the intestinal phase of T. spiralis infection and to test the antiparasitic effects of such antibodies. Strain PVG rats were infected orally with 500 larvae. Antibodies specific for phosphorylcholine-bearing proteins of L1 larvae first appeared in serum 9 days postinfection. Absorption experiments showed that the majority of antilarval antibodies produced in rats 16 days after infection with T. spiralis were specific for phosphorylcholine-bearing proteins. A fraction of these antibodies bound to free phosphorylcholine. Immunoglobulin G2c (IgG2c) producing cells in the mesenteric lymph node dominated this early antibody response. IgG2c is associated with T-independent immune responses in the rat; however, a comparison of athymic rats with euthymic controls suggested that only a small fraction of the phosphorylcholine-related antibody response against T. spiralis was T independent. Phosphorylcholine is a common epitope in antigens of bacteria and nematode parasites and has been shown to be a target of protective immunity in certain bacteria. A monoclonal IgG2c antibody was prepared from infected rats and shown to be specific for phosphorylcholine. Monoclonal phosphorylcholine-specific IgG2c failed to protect rats against intestinal infection with T. spiralis. Therefore, our findings do not support a role for phosphorylcholine-bearing antigens in immune defense against T. spiralis; however, the potency of the immune response induced suggests an immunomodulatory role for the lymphocytes involved.     

Antibody recognition of rodent malaria parasite antigens exposed at the infected erythrocyte surface: specificity of immunity generated in hyperimmune mice

In regions where malaria is endemic, inhabitants remain susceptible to repeated reinfection as they develop and maintain clinical immunity. This immunity includes responses to surface-exposed antigens on Plasmodium sp.-infected erythrocytes. Some of these parasite-encoded antigens may be diverse and phenotypically variable, and the ability to respond to this diversity and variability is an important component of acquired immunity. Characterizing the relative specificities of antibody responses during the acquisition of immunity and in hyperimmune individuals is thus an important adjunct to vaccine research. This is logistically difficult to do in the field but is relatively easily carried out in animal models. Infections in inbred mice with rodent malaria parasite Plasmodium chabaudi chabaudi AS represent a good model for Plasmodium falciparum in humans. This model has been used in the present study in a comparative analysis of cross-reactive and specific immune responses in rodent malaria. CBA/Ca mice were rendered hyperimmune to P. chabaudi chabaudi (AS or CB lines) or Plasmodium berghei (KSP-11 line) by repeated infection with homologous parasites. Serum from P. chabaudi chabaudi AS hyperimmune mice reacted with antigens released from disrupted P. chabaudi chabaudi AS-infected erythrocytes, but P. chabaudi chabaudi CB and P. berghei KSP-11 hyperimmune serum also contained cross-reactive antibodies to these antigens. However, antibody activity directed against antigens exposed at the surfaces of intact P. chabaudi chabaudi-infected erythrocytes was mainly parasite species specific and, to a lesser extent, parasite line specific. Importantly, this response included opsonizing antibodies, which bound to infected erythrocytes, leading to their phagocytosis and destruction by macrophages. The results are discussed in the context of the role that antibodies to both variable and invariant antigens may play in protective immunity in the face of continuous susceptibility to reinfection.     

Cytomegalovirus-specific cytotoxicity mediated by non-T lymphocytes from peripheral blood of normal volunteers.

The cytotoxicity of circulating mononuclear cells from normal volunteers was determined using human leukocyte antigen (HLA)-typed, low-passaged human skin fibroblasts infected with cytomegalovirus as target cells. Peripheral blood lymphocytes from both seropositive and seronegative individuals possessed virus specific cytotoxicity. Although all target cells used were susceptible to virus specific lysis, lymphocytes from some individuals were more active against some target cells than others. This differential cytotoxicity did not follow a consistent pattern of HLA restriction. Some variations in cytotoxic activity were noted on sequential studies of individual volunteers. Studies of fractionated lymphocytes from selected immune and nonimmune individuals demonstrated that cytotoxicity of lymphocytes from both groups was mediated by nonadherent, Fc receptor bearing cells which did not form rosettes with sheep erythrocytes. Repeated washing sometimes decreased cytotoxicity of lymphocytes from immune individuals, and addition of serum containing antibody to cytomegalovirus enhanced cytotoxicity, suggesting antibody dependence. It is concluded that cytotoxic lymphocytes from nonimmune volunteers possessed characteristics of natural killer cells, whereas those from immune volunteers probably consisted of both natural killer cells and antibody-dependent killer cells.