Identification of a subpopulation of lymphocytes in human peripheral blood cytotoxic to autologous fibroblasts

Translational studies in anterior cruciate ligament repair

Translational research, which can be explained as the principle of combining advances in both basic research and clinical understanding in a bedside-to-bench-to-bedside approach, has become one of the central themes of present-day medical research. One orthopedic problem that has strongly benefited from such an approach is tissue-engineering-enhanced primary repair of the anterior cruciate ligament. Recent years have shown a clearer definition of the clinical problem and established an underlying mechanistic cause of the incapacity of the anterior cruciate ligament to heal-the premature loss of provisional scaffold in the wound site. These clinical findings were then translated into a research objective, namely, to replace the missing scaffold with a biomaterial with appropriate structural and bio-stimulatory characteristics. Subsequently, a tissue-engineering-based treatment using a collagen-platelet composite was developed and tested in vitro. After proofing the efficacy of this new treatment in the laboratory, it was translated into a potential clinical application, which showed highly successful results in structural integrity and biomechanical capacity in large animal testing. This approach of defining the scientific mechanism underlying a clinical observation and then using that information to design new therapies is but one example of how translational research in tissue engineering can help define and develop new treatments for challenging problems faced by patients.

Spontaneous cell-mediated cytotoxicity (SCMC) in various mammalian species and chickens: selective reaction pattern and different mechanisms

Cultured cell lines derived from donors of various species served as 75Se- or 51Cr-labelled targets in microcytotoxicity assays. As in human donors, considerable spontaneous cell-mediated cytotoxicity (SCMC) was exhibited by peripheral blood lymphoid effector cells from healthy chickens, mink, swine, cattle, horses and tigers. Although all target lines tested could be lysed in SCMC, there was no 'general SCMC target' for all species or all individuals of one species. The selectivity, the kinetics, and the strength of SCMC reflected the capacity of the effector cells rather than a 'susceptibility' of the target to lysis. Classical major histocompatibility complex products, determinants specific for distinct cell types or subpopulations, or antigens associated with Epstein-Barr virus, herpes virus ateles, herpes virus papio, or Marek's disease virus were not appreciably involved in SCMC. No 'malignant phenotype' on targets was required for the cytotoxic effect. Yet undefined xenogeneic, allogeneic, and individual structures seem to be involved in SCMC. Depending on their expression, a target can be classified as 'species-related' or as 'individual-related'. Intermediate forms are conceivable. SCMC might comprise different cytotoxic mechanisms. Antibodies are not required for lysis to occur, but serum factors including immunoglobulins can have considerable influence on SCMC.

Connection of atopic disease in Japanese patients with juvenile dermatomyositis based on serum IgE levels

Serum IgE levels of 22 patients with juvenile dermatomyositis (JDMS), 44 normal children, and 43 patients with adult dermatomyositis were compared. The geometric mean of serum IgE levels was significantly higher in the juvenile patients when compared with normal children (p < 0.01) and adult patients (p < 0.01). Of the 22 patients with JDMS, 11 (50%) had elevated serum IgE levels accompanied by atopic disease. Dermatomyositis was accompanied by atopic dermatitis (AD) in 9 (41%) of these 22 patients, a high prevalence when compared with reports among the general population. Following the appearance of muscular symptoms in JDMS patients with AD, the ratio of OKT4 to OKT8 cells rose due to a reduction in the percentage of OKT8-positive cells, along with further elevations in serum IgE levels and intractable cutaneous manifestations of dermatomyositis. All this may be a result of an interaction between the immuno-mechanism of this disease and that of AD. We suspect that, in general, children with impaired cell-mediated immunity, presenting with such symptoms as AD, may have a higher tendency for developing JDMS.

Natural killer cell regulation of murine embryonic pulmonary fibroblast survival in vivo

We examined the role of the natural killer (NK) cell in controlling the survival of embryonic pulmonary fibroblasts in vivo. In vitro, both primary embryonic fibroblasts and an embryonic fibroblast line (10T1/2) were lysed by syngeneic C3H/HeN splenocytes threefold more efficiently than primary adult fibroblasts. The membrane phenotype of the effector cells was typical of NK cells. It was asialo GM1+, Lyt2.1-, Lyt 1.1-, Thy 1.2-. The cytotoxicity of the effector cell could be enhanced by IFN-alpha/beta but was deficient in the C3H/HeJ bg/bg mutant. Iododeoxyuridine (131I-dUrd)-labeled embryonic fibroblasts were injected intravenously into syngeneic mice with either enhanced or deficient NK function and their survival in the lung was quantitated. Enhanced fibroblast survival was detected in the NK deficient C3H/HeJ beige (bg/bg) mutant strain compared to its normal littermate C3H/HeJ (bg/+). A second method of NK depletion by pretreatment with rabbit anti-asialo GM1 antiserum also produced a striking increase in fibroblast survival. Poly(I:C) significantly enhanced the elimination of pulmonary fibroblasts from the lung between 4 and 24 hr after injection. Poly(I:C) did not enhance clearance of pulmonary fibroblasts in the C3H/HeJ (bg/bg) mutant, but did so in the normal littermate C3H/HeJ (bg/+). In conclusion, we have shown that the survival of embryonic pulmonary fibroblasts was inversely correlated with in vivo NK activity suggesting a possible role for this cytotoxic cell in the control of fibroblast growth in vivo.

Experimental immunotherapy with NK-like cells. A preliminary report

Natural killer (NK) cell activity was generated in the spleen of C3H/HeN mice by i.p. administration of poly I:C, while i.p. injection of BCG primarily promoted the generation of NK-like cells in peritoneal exudates (PE). A single injection of 10 mg of BCG 9 days before s.c. challenge with the MBT-2 murine bladder cancer was found to induce a 45% protection against tumor take. However, a single injection of 100 micrograms poly I:C 16 h before tumor cell challenge did not protect the animals against tumor take. Intratumoral injection of either PE cells from BCG-immunized or spleen cells from poly I:C-treated mice into mice developing tumor, was capable of suppressing tumor growth in vivo. The mean tumor diameters of these two experimental groups of animals on day 40 were significantly smaller (P less than 0.005) than in the controls, and they survived approximately 10 days longer than the controls. Since this in vivo tumor suppressive effect by the lymphoid cell population correlated with the increase in NK-like cell activity assayed in vitro, and most of the adherent cells had been removed before injection, it is suggested that the antitumor function of the lymphoid cell population may be mostly due to the presence of "activated" NK or NK-like cells. These results support the concept of NK therapy for cancer.

Generation and regulation of autocytotoxicity in mixed lymphocyte cultures: evidence for active suppression of autocytotoxic cells

Using limiting dilution analysis, we have detected both the generation and suppression of autocytotoxic cells following autologous or allogeneic stimulation in vitro. Assay conditions were established in which the cytotoxic response toward an allogeneic sensitizing cell was consistent with a traditional single-hit kinetic model. Under identical conditions, cytolytic activity toward autologous phytohemaglutinin-activated lymphoblasts exhibited a distinct biphasic response. At low responder cell doses, a clear autocytotoxic response was observed. However, at higher responder cell numbers, this autocytotoxic reaction disappeared. This biphasic pattern of autocytotoxicity developed after stimulation with allogeneic or autologous peripheral blood mononuclear leukocytes (PBL) or Epstein-Barr virus-transformed B cells. This pattern of response is consistent with the counterpoised actions of two distinct cell populations, an autoaggressive population and a lower frequency autosuppressor population. Autocytotoxicity was not the result of mitogenic or xenogeneic antigenic stimulation, as it was observed after stimulation with autologous PBL in autologous serum and an autologous interleukin 2 preparation. Thus, cells capable of autocytotoxicity are present in peripheral blood but at a lower frequency than allocytotoxic T lymphocytes. Furthermore, autoaggressive cells are down-regulated by an autologous suppressor population. These findings suggest that immunologic self-tolerance is, at least in part, an actively maintained condition. Disturbances in this autoregulatory network may have relevance to the pathogenesis of some autoimmune diseases and graft-versus-host disease.

Cytodestructive mechanisms provoked by food antigens. I. Direct, allergic autocytotoxicity

In vitro challenge of peripheral white blood cells (WBC) from 62 food sensitive patients and 22 asymptomatic control individuals was performed using standardized extract of cow milk, corn and wheat. 81% of the patients reacted at least toward one food antigen, showing disintegration of WBC, as measured by trypan blue exclusion technique. The presented experimental condition of the direct binding between food antigen and cell wall membrane suggests that this phenomenon is multicellular in its origin. The interpretation of these studies is of distinct value in directing human food hypersensitivity research in the future.

In situ quantification of T-lymphocyte subsets and Langerhans cells in the inflammatory infiltrate of atopic eczema

Tissue from the acute, non-infected eruption of sixteen atopic eczema subjects was subjected to an indirect immunoperoxidase technique using monoclonal antibodies recognizing T-lymphocyte subsets, Langerhans cells and natural killer cells. Over half the cells infiltrating the dermis were T lymphocytes, including a large majority of helper T cells and relatively few suppressor T cells. Langerhans cells were present in significant proportions in the dermis and probably reflected increased antigen presentation within the affected skin. There was no evidence of increased natural killer cell presence. This study suggests that type IV hypersensitivity may be implicated in the aetiology of atopic eczema.

Natural cell-mediated cytotoxic activity against isolated chondrocytes in the mouse

The ability of lymphoid cells from normal mice to exert a natural cytotoxic activity against isolated syngeneic and allogeneic epiphyseal chondrocytes was studied by means of 51Cr release assay. We found, that both spleen and peritoneal cells, but not thymocytes, exerted an anti-chondrocyte cytotoxic effect. Addition of unlabelled chondrocytes markedly reduced experimental 51Cr release and the inhibitory effect was proportional to the number of 'cold' cells added. This indicate, that chondrocyte lysis was due to specific effector-target interaction. As no cytotoxicity was observed against isolated fibroblasts, our results could not be explained by lysis of fibroblasts contaminating chondrocyte cultures.

Human and murine cutaneous graft-versus-host diseases. Potential models for the study of immunologically mediated skin diseases

Cutaneous graft-versus-host disease in humans can be regarded as a potential model for the study of lymphocytotoxic reactions directed against the epidermis, and immunologically mediated cutaneous sclerosis. Toxic epidermal necrolysis and lichen planus are closely reproduced in graft-versus-host disease; sclerosis of chronic graft-versus-host disease cannot be regarded as strictly similar to any type of scleroderma. Sclerosis is the most consistently reproducible feature of cutaneous graft-versus-host disease in mouse chimeras; it offers the opportunity to study a dermal sclerosing reaction which may be considered as T cell dependent. Murine graft-versus-host disease presently appears debatable for use as a model of T cell dependent injury of the epidermis since stable established epidermal lesions are difficult to reproduce.

Histiocytosis X

To clarify salient issues pertaining to histiocytosis X--a syndrome that includes Letterer-Siwe disease, Hand-Schuller-Christian disease, and eosinophilic granuloma--the authors review the epidemiologic data and the histologic, morphologic, and clinical bases for diagnosis and prognosis. Histiocytes are defined and their possible histogenesis outlined, and Langerhans cells, which may be a leading element in active lesions, are characterized. The authors outline hypothetic pathogenetic schema, which they recommend be tested by recently developed immunologic and genetic means, since histiocytosis X, at least in its disseminated form, remains an unpredictable disease for which there is no proven effective therapy.

Interferon-induced autologous reactivity of human natural killer cells

The reactivity of human natural killer (NK) cells against normal autologous target cells was studied. The following major observations were made: (i) non-stimulated NK cells did not lyse autologous target cells, although allogeneic cells were killed to some extent. However, treatment of effector cells with interferon (IFN) resulted in induction of autologous cytotoxicity, and both autologous and allogeneic fibroblast targets were efficiently killed without preference. This was demonstrated both with unfractionated lymphocytes and with effector cells highly enriched in large granular lymphocytes (LGL). (ii) NK cells bound to autologous cells as efficiently as to allogeneic cells. The cells binding to autologous cells were able to lyse K562 cells, demonstrating their NK nature. After stimulation with IFN autologous cells were also killed. (iii) The exemption from autologous cytotoxicity by the non-stimulated NK cells seemed to result from a block at a post-binding stage, and there did not appear to be any need for accessory cells in the induction of autologous reactivity by IFN treatment of the LGLs. The results thus point to important post-binding-stage regulatory mechanisms saving normal cells from the cytotoxic action of circulating NK cells and may help to understand the suggested role of NK cells in certain autoimmune diseases.

Suppression of natural killer activity in human blood and bone marrow cultures by bone marrow-adherent OKM1-positive cells

Maintenance and regulation of natural killer (NK) cell activity in human bone marrow cultures were studied using K562 leukemia cells as targets. Culture of bone marrow cells in medium supporting long-term generation of myeloid cells resulted in a rapid loss of NK activity in 1-3 days. In contrast, antibody-dependent cytotoxicity to an NK-resistant tumor was maintained for more than 7 weeks. Horse serum, a component of the myelopoietic culture medium, was found to diminish NK cytotoxicity of blood and bone marrow cultures whereas hydrocortisone supplement did not. In addition, an adherent cell is present in bone marrow which greatly inhibits NK activity. Nonadherent bone marrow cells exhibited higher cytotoxicity than unfractionated cells at all days of culture; adherent cells were not cytotoxic to K562. Purified adherent marrow cells inhibited the cytotoxic capacity of nonadherent blood or marrow mononuclear cells during coculture. Indomethacin, an inhibitor of protaglandin synthesis, augmented levels of NK activity in cultures of bone marrow cells, indicating that macrophages may be suppressing this effector function via prostaglandins. Further identification of the adherent suppressor cells came from experiments in which suppression was prevented by treatment of the adherent cells with monoclonal OKM1 antibody plus complement. This study shows that bone marrow-adherent OKM1-positive cells, presumably macrophages, negatively regulate NK activity, and it defines conditions for analysis of the generation and/or positive regulation of NK cells in human bone marrow.

Quantitative analysis of T-lymphocyte subsets in atopic eczema, using monoclonal antibodies and flow cytofluorimetry

An analysis of T-lymphocyte subpopulations was performed on thirty-eight consecutive patients with atopic eczema, using monoclonal antibodies and flow cytofluorimetry. The total T-lymphocyte and helper T-cell numbers did not differ significantly from those of a matched control group, but there was a highly significant reduction in suppressor T-cell numbers among the atopic group. Both suppressor T-cell deficiency and circulating IgE elevation were found to correlate positively with disease activity. Food specific IgE levels to milk and egg were ascertained by radioallergosorbent (RAST) studies. These proved positive in seven out of thirty-five atopics (20%).

Interferon-induced resistance to the killing by NK cells: a preferential effect of IFN-gamma

HeLa cells show a decrease of susceptibility to the killing by natural killer (NK) cells when treated with IFN-alpha, beta, or gamma. The concentrations at which preparations of IFN-alpha or beta induce the resistance to killing are those which also induce resistance of HeLa cells to infection by vesicular stomatitis virus (VSV). Stimulation of the killing activity of NK cells is also induced at that same range of concentrations of IFN-alpha and beta. In contrast with preparations of IFN-gamma, induction of the resistance to killing occurs at IFN concentrations which have only marginal stimulatory effect on the activity of NK cells and have no antiviral effect against VSV. IFN-gamma, produced with cloned IFN-gamma cDNA, is as effective as lymphocyte-produced IFN in inducing the resistance to natural killing. The potent effect of IFN-gamma on the target cells is, therefore, not due to the function of lymphokines which might contaminate lymphocyte-produced preparations of IFN-gamma, but a genuine property of the IFN itself.

Human spontaneous lymphocyte-mediated cytotoxicity (SLMC) against malignant and normal tissue-derived target cell lines tested in autologous and allogeneic combinations by the microcytotoxicity assay

Effector cell types and effector mechanisms of human spontaneous lymphocyte-mediated cytotoxicity (SLMC) were studied in a 44-h microcytotoxicity titration assay. Peripheral blood lymphocytes from cancer patients and controls were used as effector cells either unfractionated or after fractionation by rosetting techniques or affinity chromatography. The possible immunoglobulin dependency of the reactions was studied by incorporation of specific Fab fragments of rabbit anti-human IgG antibodies in the incubation mixtures. Twelve different target cell lines of either high or low sensitivity to SLMC and with or without easily detectable HLA antigens were used. Most of the target cells were cell lines derived from transitional cell carcinoma of the urinary bladder (TCC). Both allogeneic and autologous lymphocyte target cell combinations were tested. Although high- and low-sensitivity target cells differed significantly in susceptibility to lysis, the predominating SLMC was displayed by Fc-receptor-positive lymphocytes in both allogeneic and autologous combinations. Addition of the Fab anti-immunoglobulin reagent to the incubation mixtures resulted in strong inhibition of cytotoxicity regardless of the type of target cells used and in allogeneic as well as in autologous lymphocyte target cell mixtures. However, in some combinations no inhibition was seen and inhibition was usually not complete, suggesting that both immunoglobulin-dependent (i.e., ADCC-like) and immunoglobulin-independent mechanisms were involved in the cytotoxicity reactions. The results of the microcytotoxicity assay were compared with those obtained with aliquots of the same lymphocytes and target cells in an 18-h 51Cr-release assay. While similar results were obtained with high-sensitivity target cells, with low-sensitivity targets and in some autologous combinations the two assay systems registered lymphocyte/target cell interactions which differed with regard to specificity, effector cell type, and immunoglobulin dependency.

Natural killer cell activity in atopic dermatitis

Natural killer (NK) cell activity against K562 cells was studied in 12 male adults with atopic dermatitis (AD). In a 4-h chromium release microcytotoxicity assay normal NK cell function and its augmentation by interferon (IFN-alpha) were observed in AD patients. Slightly higher NK cell activity was observed in patients with allergic respiratory symptoms. The role of NK cells in the pathogenesis of atopic dermatitis is discussed.

Autocytotoxic cells against normal leucocytes identified in mouse mesenteric lymphnodes

Plaques were formed in monolayers of leucocytes prepared from mesenteric lymphnodes (LN) of normal mice. The plaques were first detected after 2 h incubation, followed by a sudden increase in number at 7 h, and reached a peak between 11 and 21 h. Large cells were observed in almost all plaques under the microscope. The large cells attached surrounding leucocytes via cytoplasmic extensions, and held aggregated leucocytes on their cell surface. Aggregated leucocytes were lyzed and/or phagocytized by the large cells. Results suggest that plaque formation was due to the direct, indiscriminate killing of normal leucocytes by the large cells. The large cells increased with age. The large cells were contained in the adherent cell fraction but not the non-adherent fraction, and were removed from leucocytes by the carbonyl iron method. Thus, the large effector cells may be related to macrophages.

Effect of human fibroblast interferon on natural killer cell activity: stimulation in vitro and inhibition in vivo

This paper describes the influence of human fibroblast interferon (IFN-beta) on the cytotoxic activity of natural killer cells (NK) in vitro and in vivo using the blood of healthy donors and myeloma patients. IFN-beta stimulates NK activity against all target cells tested in vitro in a dose-dependent way up to 250% of pretreatment values. At higher IFN concentrations, stimulation returned to baseline values. Stimulation was most pronounced in the lowest lymphocyte to target cell ratio. 1- to 2-h preincubation of effector cells with IFN was enough to achieve maximal stimulation. The effector cells of IFN-treated myeloma-patients, or patients with herpes zoster, showed a clear reduction of toxicity against all cells tested during the first infusion, as compared to the pretreatment values.

Umbilical-cord-blood-derived suppressor cells of the human natural killer cell activity are inhibited by interferon

The natural killer (NK) activity of umbilical-cord-derived lymphocytes was studied. The general level of activity was lower than with adult lymphocytes against K-562 cells and fetal fibroblasts. The activity could be boosted by interferon pretreatment of effect or cells, fractionated with Percoll density gradient centrifugation, and the suppressive activity of different fractions was tested on highly enriched adult buffy-coat-derived NK cells. Allogeneic adult NK cell activity could be inhibited in 9 of 20 cases tested with small and medium-sized T lymphocytes (Percoll fractions 4-5) from the umbilical cord. The suppressive capacity was further enriched in fractions forming rosettes (RFC) with antibody-coated human erythrocytes (EA). Such EA-RFC of percoll fraction 4-5 from umbilical cord exerted a strong suppressive activity in each case tested. Pretreatment of EA-RFC with interferon regularly abolished the suppressive effect. We conclude that there are Fc-receptor-positive small/medium-sized T lymphocytes in the umbilical cord blood which can efficiency suppress the cytotoxic activity of NK cells and that the suppressive activity can be abolished by interferon pretreatment of the suppressor cells.

Lymphocytes infiltrating human breast cancers lack K-cell activity and show low levels of NK-cell activity

Lymphocytes infiltrating human primary mammary carcinomas lack ADCC and show low levels of natural cytotoxicity. The peripheral blood lymphocytes, however, show a variable but prominent level of cytotoxicity. Lymphocyte preparations from breast tumours, when mixed with autologous blood lymphocytes, significantly suppress their prominent killer- (K- and NK-) cell activities.

Histiocytosis-X

Twelve of 17 patients with histiocytosis-X were immunologically abnormal, as shown by the presence of circulating lymphocytes spontaneously cytotoxic to cultured human fibroblasts or of antibody to autologous erythrocytes. The patients also had a notable lack of histamine H2 surface receptors on their T lymphocytes, suggesting a suppressor-cell deficiency. The lymphocyte abnormalities were reversed in vitro after incubation in a crude extract of calf thymus gland, and therefore all 17 patients were treated with daily intramuscular injections of this extract. With this therapy, 10 patients entered complete remission -- a response at least as good as that observed in historical controls treated with chemotherapy. A positive clinical response was associated with an increase in the number of T-cell histamine H2 receptors to normal levels and with correction of the other immunologic abnormalities. The results of this preliminary study justify a larger prospective clinical trial of thymic extract and further investigation of the immunoregulatory mechanisms in histiocytosis-X.

Development of a nonthrombogenic collagenous blood-prosthetic interface

Investigations to develop an implantable assist pump for prolonged circulatory support have been impeded by accumulation of friable thrombus on the prosthetic interface, with subsequent embolization. To circumvent this problem, the textured, fibril surface of a polyurethane pump chamber (mat thickness 430 microns) was inoculated with cultured bovine fetal fibroblasts (labelled with thymidine-14C) prior to animal implantation. The pneumatically actuated device (stroke volume 75 ml), maintained a pulsatile blood flow throughout each study. In 20 calf experiments, extending up to 335 days, 30 X 10(6) fibroblasts (in 50 ml media) derived from a single Holstein fetus were distributed on the urethane surface (360 +/- 50 cells/mm2) by rotation of a sealed device for three hours (12 revolutions/hour). Following connection to the circulation, cell washout was minimal. Resultant biologic linings, examined after animal sacrifice, were densely adherent to the underlying polymer matrix, and varied in thickness from 250 micron-1.5 mm. Microscopically, fibroblasts were identified from the surface to base, accompanied by numerous collagen bundles and abundant ground substance. Amino acid analysis in 10/20 pumps implanted for 31--335 days, revealed 50 +/- 5 Hydroxyproline residues/1000 residues (50% collagen) and scant elastin. Donor fibroblasts were identified by radioautography and karyotyping. Lack of immunologic response in 12 Hereford pump recipients as confirmed by serial fibroblast cytotoxicity assays. In conclusion, an induced collagenous-blood interface permitted prolonged mechanical circulatory support in animals without thromboembolic complications.

Human graft versus host disease

Human graft versus host disease is composed of 2 distinct clinical entities, acute graft versus host disease and chronic graft versus host disease, which have different pathogenesis. Acute graft versus host disease is produced by the attack of donor immunocompetent T or null lymphocytes against recipient histocompatibility antigens. The null lymphocytes may attack antigens shared by the donor and recipient and are autocytotoxic lymphocytes which can produce acute graft versus host disease in recipients of identical twin transplants. The cessation of acute graft versus host disease occurs when suppressor lymphocytes appear in the recipient's peripheral circulation. Chronic graft versus host disease is produced by immunocompetent lymphocytes that differentiate in the recipient. Its control is unknown. Some patients with chronic graft versus host disease have in vivo activated suppressor lymphocytes which produce a secondary immunoincompetence and an increased susceptibility to bacterial sepsis and death.

Induction of autologous reactivity of human NK cells

'Mature' human natural killer cell activity could be enriched by adsorption-elution with fetal fibroblast as adsorbents against normal allogeneic skin fibroblasts, but not against autologous targets. However, when the natural killer activity was augmented by contact with tumour cells (HeLa), autologous and allogeneic skin fibroblast targets were killed without preference. Adsorption-elution with augmenting target cells as adsorbents resulted in an efficient enrichment of NK activity showing non-discriminate cytotoxicity. Morphologically, this was associated with an enrichment of large granular lymphocytes previously shown to be responsible for human NK activity. We conclude that the NK activity of human 'mature' NK cells shows a relative autologous exemption, whereas 'pre-NK' cells augmented to full activity in vitro are non-discriminative.

Tumour-related changes in natural killer cell activity in melanoma patients. Influence of stage of disease, tumour thickness and age of patients

The influence of tumour growth on the natural killer (NK) activity of melanoma patients was examined by analysis of NK activity in 51Cr release assays before and at intervals after surgical removal of localized melanoma. In patients with Stage I and II melanoma, removal of the tumour was associated with changes in the level of NK activity which were not detected in patients who had surgery for skin graft after previous removal of the primary melanoma. The changes in NK activity after removal of melanoma were however different in patients with Stage I compared to Stage II melanoma. In the former, NK activity which appeared specifically directed toward melanoma cells was maximal 2-4 weeks after removal of the tumour and then decreased to normal levels. The NK activity after surgery was directly related to the thickness of the tumour. In patients with Stage II melanoma, NK activity did not increase but fell to low levels after removal of the tumour. The level of NK activity in these patients was not related to the thickness of their primary melanoma. In patients with Stage I and II melanoma the NK activity against melanoma cells showed a significant increase with age. The reason for the different sequence of changes in NK activity after removal of melanoma in the two patient groups is unknown. It is suggested however that the differences observed in patients with Stage II melanoma may reflect differences in the tumour or host response which contributed to the spread of the tumour to regional lymph nodes in these patients.

Autoreactive cells as mediators of the natural defense against malignant tumors

The present paper suggests that the natural defense mechanisms against tumors are part and a logical consequence of the natural mechanisms that maintain tissue uniformity. We further suggest that biologic errors steadily are accumulated in cells with increasing age, and that the number of cells with insufficient functional capacity (aberrant cells) increases with age. Since this increasing number of aberrant cells threatens the uniformity and function of the cell population, aberrant cells must be removed in order to maintain tissue uniformity. The aberrant cells may be removed by autonomous destruction or by cytotoxic attack by autoreactive cells. Such autoreactive cells are present in the normal organism throughout life, may be suppressed by thymus-derived cells, and may be activated by immune adjuvants. We suggest from theoretic and experimental reasons that the major function of autoreactive cells is the maintenance of tissue uniformity and tissue homeostasis within a cell population.

Aberrations of suppressor T cells in human graft-versus-host disease

To determine whether imbalances in immunoregulatory T-cell subsets exist in patients with graft-versus-host disease, we analyzed T cells in three patients with acute and in six patients with chronic graft-versus-host disease after bone-marrow transplantation. The normal human peripheral-blood T-cell compartment is composed of 80 per cent TH2-and 20 per cent TH2+ T cells, and defined by reactivity with subset-specific heteroantiserums. Human suppressor cells are TH2+, whereas helper cells are TH2-. Patients with acute and chronic graft-versus-host disease had abnormalities in these populations, and their T cells frequently bore la-like antigens. Patients with acute disease lacked TH2+ cells, and the reappearance of this subset preceded the cessation of disease activity. Chronic disease, in contrast, was more heterogeneous. Suppressor cells were lacking in two patients but increased in the other four. Two of these four patients had TH2+, la+ T cells, suggesting in vivo activation of suppressor cells. Studies showing that these TH2+, la+ cells actively suppressed the in vitro immune response support this hypothesis and suggest that the immunoregulatory cells may profoundly affect the overall immune response.

Do natural killer cells engage in regulated reactions against self to ensure homeostasis?

Host reactivities not requiring immunization in the mouse, especially natural resistance of irradiated animals to accept grafts of normal or malignant hemopoietic cells, were compared with NK activity against the YAC-1 lymphoma. The effects of several independent variables known to influence natural resistance in vivo had a similar effect on the NK system. Figure 12 lists an impressive array of shared properties and positive correlations. In contrast, the distinctions were few and minor. Many of the positive correlations were of particular significance since the experimental variables either have opposing or no effects on conventional induced immunity. The multiplicity and pervasiveness of these correlations suggest that the cellular mechanisms underlying natural reactivities are similar or common. Cytotoxic effectors mediating natural resistance to normal cells, tumors, and cells infected with intracellular pathogens may be distinct in terms of target selectivity, yet belong to a single cell lineage subject to common regulatory influences for differentiation and function. Regulation of reactivity via suppressor cells was studied in the NK system only. The spleens of mice selected for low levels of NK activity (resulting from young age, irradiation, and treatment with the macrophage-active agents l-carrageenan or hydrocortisone acetate) contained cells capable of inhibiting the lytic function of NK effectors taken from untreated adult donors. All the suppressor cells studied were thymus-independent, as judged by their occurrence in spleens of genetically athymic mice; the suppressive function was resistant to 2000 rads of gamma-rays administered in vitro and was not restricted by the major histocompatibility complex, without exception. However, two major classes of suppressors were identified: (a) macrophagelike cells inducible by l-carrageenan or hydrocortisone acetate, and (b) nonadherent cells found in spleens of untreated infants and of irradiated adult mice. It is proposed that the suppression of NK cytolysis demonstrated in vitro was a manifestation of regulatory mechanisms modulating the level of NK activity in vivo. Macrophagelike cells that are induced, activated, or inactivated by bacteria, viruses, hormones, and other agents may act as regulators of differentiation, maturation, and function of cells belonging to the NK lineage. Nonadherent cells could be either a distinct class of suppressors or immature NK cells capable of binding but not lysing target cells. In the latter case, regulation would be achieved via competitive binding of targets by pre-NK cells presumably in dynamic equilibrium with functional (i.e. matured) NK effectors.

Lymphocyte-mediated cytotoxicity in humans during revaccination with vaccinia virus

Fifteen healthy human volunteers were revaccinated with vaccinia virus. Blood samples (4 to 7) were obtained during the 3 weeks after revaccination. Peripheral blood lymphocytes were washed extensively and tested for cytotoxicity against vaccinia-infected autologous and/or homologous skin fibroblasts. Without addition of antibodies, peak levels of killing were observed on days 7 to 9. The killing did not depend on common HLA markers. On days with peak activity, extensively washed lymphocytes showed higher levels of killing than normally washed lymphocytes. By cell separation experiments, the cell most active in killing proved to be a nonadherent, non-phagocytizing lymphocyte with Fc receptors. Serum antibodies tested in two sensitive serological assays peaked on days 14 to 17. The question of whether the killing observed is dependent on or independent of antibodies is not clarified in the present study.

Characterization of the human natural killer (NK) cell in blood and lymphoid organs

The human natural killer (NK) cell in peripheral blood is a non-thymus-dependent, (Fc+ C3-), Ig-bearing lymphocyte and is probably identical to the human K cell, although not requiring the Fc receptor, in contrast to the K cell, for the lytic process. The human NK cell in tonsil and lymph node, on the other hand, is a non-Ig- bearing T lymphocyte, lacking both Fc and C3 receptors.

Human natural cytotoxicity in the blood and lymphoid organs of healthy donors and patients with malignant disease

Natural cytotoxicity against CLA-4 and D6 target cells was determined in the peripheral blood of healthy donors and women with mammary carcinoma (localized to the breast and axilla), and in human normal and regional tumour lymph nodes. The NK cell activity in the blood of women with clinically localized mammary carcinoma showed a wide range of reactivities but was similar to the pattern detected in healthy controls. The NK cell activity in human normal lymph nodes also showed a wide range and was the same irrespective of the anatomical site of origin of the lymph node. Lymph nodes draining a variety of solid tumours had a similar spectrum of NK cell activity, there being no obvious correlation between the level of NK cell activity and such parameters as size and histological type of tumour, hyperplasia or metastatic spread in the regional node. Trypsin treatment of lymphocyte suspensions from the various lymphoid compartments suggested that the cell responsible for natural cytotoxicity in blood was probably different from the NK cell in tonsil and lymph nodes.

Immune response in humans after vaccination with vaccinia virus: generation of a virus-specific cytotoxic activity by human peripheral lymphocytes

After vaccinia virus vaccination of human volunteers, local indurations developed within 10 days, and regional adenopathy was detected in half of the individuals. Their peripheral blood lymphocytes (PBL) harvested at different days after vaccination showed specific activity against target cells infected with vaccinia virus with a peak activity at day 7. The specificity of the cytotoxic activity was not related to HLA markers, since autologous, homologous, and heterologous infected target cells were lysed with the same efficiency. The cytotoxic activity was caused by PBL that did not rosette with sheep erythrocytes and could be depleted by more than 90 percent by removing Fc receptor-bearing cells. T-cell- depleted PBL showed a one-half to two times greater cytotoxicity than intact PBL. The cytotoxic activity could also be abrogated by more than 95 percent by rabbit Fab(2) anti-human IgG. On the other hand, nonimmune PBL lysed vaccinia-infected target cells in the presence of specific antibodies against vaccinia virus, thus demonstrating that ADCC could be efficient in lysing vaccinia-infected target cells. We conclude that after vaccination, antibody-forming cells arise and provide specific anti-viral antibody and that the cytotoxic cells detected in this reaction are K cells. These experiments suggest that antibody-dependent cell cytotoxicity may be of major importance in the recovery of man to virus infections.