Castleman's Disease and Interleukin 6

Castleman's disease is a rare condition characterized by benign hyperplastic lymph nodes. Based on the morphological features, it has been divided into hyaline-vascular, plasma cell and intermediate types. The latter two types are frequently associated with a wide variety of clinical pictures such as fever, anemia with hypotransferrinemia, hyperimmunoglobulinemia and an increase in the concentration of C-reactive protein (CRP). Although immunological disturbances have been suggested to play important roles in the pathophysiology of Castleman's disease, the precise mechanisms for the generation of its clinical pictures are still unsettled. In this respect, we have reported a pediatric case with spontaneous production of high levels of B cell differentiation factor (BCDF) activity by the hyperplastic lymph node, and we demonstrated here the strong expression of interleukin 6 (IL-6) gene in the lymph node cells. On the other hand, recent studies have revealed that IL-6 is a multifunctional cytokine; IL-6 not only induces the immunoglobulin production but also induces the acute phase reaction, and functions as an endogeneous pyrogen. In the acute phase reaction, IL-6 may induce an increase in CRP concentration and hypotransferrinemia. These studies indicate that the overproduction of IL-6 by the hyperplastic lymph node may be closely related to the pathophysiology of Castleman's disease. Therefore, it is considered that this disease is a "disorder of IL-6 production".

Regulation of T-helper cell responses to inhalant allergen during early childhood

BACKGROUND

Recent evidence suggests that preschool children manifest patterns of allergen-specific skin prick test (SPT) reactivity and in vitro T-cell cytokine production which are similar to that of either atopic or nonatopic adults. However, published studies on this age group involve small sample sizes and a restricted number of cytokines, usually in response to polyclonal stimuli.

OBJECTIVE

To elucidate the relationship between in vivo and in vitro immune responses to a major inhalant allergen house dust mite (HDM) in preschoolers.

METHODS

Peripheral blood mononuclear cells (PBMCs) from matched groups of HDM-SPT+ and SPT- 6-year-olds (n = 30 and 29, respectively) tested for PBMC responses to HDM, and cytokine production measured at both the protein and mRNA levels. Immunoglobulin (Ig) E and IgG subclass antibody titres were determined in serum. Interrelationships between in vitro and in vivo HDM responses were examined via multivariate analyses.

RESULTS

SPT reactivity to HDM was associated with in vitro production by putative T cells of interleukin (IL) -4, IL-5, IL-9, IL-10, IL-13 and low level IFNgamma, and with production in vivo of IgE and (all) IgG subclass antibodies; HDM responses in the SPT- group were restricted mainly to IL-10 and IFNgamma and very low levels of IL-4; IL-6 production from non-T-cell sources was common. The cytokine most associated with positive SPT responses was IL-9; SPT weal diameter correlated positively with IL-4, IL-5 and IL-13 and negatively with IL-10.

CONCLUSION

Detailed analysis of cytokine responses in this very young age group have the potential to uncover subtle relationships between in vivo and in vitro allergen reactivity which may be less clear in adults, in whom T-cell response patterns are modified via chronic stimulation. The present findings which suggest potentially important roles for IL-9 and IL-10 in the early phase of allergic disease, may be one such example.

TH2-polarized immunological memory to inhalant allergens in atopics is established during infancy and early childhood

BACKGROUND

There is increasing evidence that the T-cell reactivity to environmental allergens underlying expression of allergic disease in adulthood, develops initially during childhood. However, there is little information available on the kinetics of these early responses, or on the patterns of cytokine production during this period.

OBJECTIVE

The purpose of this study was twofold: to obtain further information on the reported differences between responses to food versus inhalant allergens during early childhood, and to ascertain the age-range over which T-cell responses to inhalant allergens become polarized towards the TH2 cytokine profile, in potentially atopic children.

METHODS

In vitro cytokine responses to house dust mite (HDM) and egg (OVA) were assessed by semiquantitative RT-PCR in panels of 2- and 5-year-old children and adults; lymphoproliferative responses to OVA were subjected to epitope analysis.

RESULTS

At age 2 years IL-4/IL-5 responses to HDM grouped with positive atopic family history, and by age 5 years cytokine responses correlated strongly with individual SPT reactivity to HDM. In contrast, OVA responses were restricted to weak and transient IL-5 signals in the 2-year-old family history positive group. Lymphoproliferation assays performed in parallel indicate a log-scale greater postnatal expansion of T-cell reactivity to the inhalant allergen; preliminary epitope analysis of OVA responses indicate that the number of OVA epitopes recognised decrease during early childhood.

CONCLUSIONS

Inhalant allergen-specific in vitro cytokine production associated with positive skin-prick test (SPT) reactions, one of the hallmarks of adult atopy, manifests in children at or before 5 years of age; additionally, cytokine responses in SPT negative 5 year-olds are restricted to IFNgamma, as per normal adults. In contrast, T-cell responses to a typical food allergen appear to be deleted during early childhood.

Later development of Fas ligand-mediated cytotoxicity as compared with granule-mediated cytotoxicity during the maturation of natural killer cells

We classified CD56+ CD3- natural killer (NK) cells into CD2- CD56dim (CD2- NK), CD2+ CD56dim (CD2+ NK) and CD2+ CD56bright populations, and investigated mainly functional differences between the former two populations. CD2- and CD2+ NK cells were the same in their morphology and several surface molecules except for CD2. The percentages of CD2- NK cells in total NK cells were higher in the cord blood and marrow than in the peripheral blood of adults or children. Freshly isolated CD2- NK cells had CD2 in the cytoplasm, and gradually expressed it on the surface upon incubation with interleukin-2 (IL-2). These results demonstrated that CD2 is an antigen which appears on the surface during the maturation of NK cells. The granule-mediated cytotoxicities, which are mainly performed by the perforin molecule, of CD2+ NK cells against K562 and Daudi cells were higher than those of CD2- NK cells, and they were inhibited to the levels of CD2- NK cells by the addition of a blocking anti-CD2 monoclonal antibody (mAb). Fas ligand (FasL) mRNA was expressed in freshly isolated CD2+ NK cells but not in the CD2- NK cells. Neither freshly isolated NK populations showed FasL-mediated cytotoxicity, and only CD2+ NK cells lysed Fas-transfected targets after the 24-hr incubation with IL-2. Based on these results, CD2- NK cells have already developed granule-mediated cytotoxicity equal to that of CD2+ NK cells except for the CD2-associated activity, but they, unlike CD2+ NK cells, totally lack FasL-mediated cytotoxicity. These findings suggest that FasL-mediated cytotoxicity may be acquired at more mature stages of NK-cell maturation than granule-mediated cytotoxicity.

Developing patterns of T cell memory to environmental allergens in the first two years of life

Several recent studies have demonstrated cord blood mononuclear cell (CBMC) proliferation in response to food and inhalant allergens, suggesting that initial T-cell-priming may occur in utero. The findings below from an ongoing prospective study on 60 subjects provide initial information on the nature of accompanying T cell cytokine responses. We demonstrate CBMC proliferation following culture with house dust mite and ovalbumin (OVA) in 47 and 42% of subjects, respectively, compared to an overall rate of 3% for tetanus toxoid; the frequencies of these responses were comparable in neonates with and without atopic family history (FH). With the exception of IL-10, analysis of cytokine responses in allergen-stimulated cultures of CBMCs required the use of semiquantitative RT-PCR, which revealed low-level IL-4 and/or IL-5 mRNA production, in particular a 50% IL-5 response rate to OVA in FH-positive neonates. IFN-gamma responses were less frequent and required higher PCR cycle numbers for detection. Preliminary analysis of culture supernatants from a subgroup of CBMCs indicate high-level allergen-specific IL-10 responses in both FH-negative and -positive subjects, detectable by ELISA. Parallel PCR studies on MCs from 27 children (mean age 18 months) indicated a clear segregation at this age on the basis of FH, with Th0-like or mixed Th1/Th2 responses (IL-5 plus IFN-gamma) which were mainly restricted to the FH-positive group.

Allergen recognition in the origin of asthma

Allergic respiratory diseases such as bronchial asthma are believed to result directly from the repeated local expression in airway tissues of T helper (Th) 2-polarized T cell immunity to inhaled allergens. Recent evidence suggests that these T cell responses are typically primed in utero and subsequently reshaped during postnatal allergen exposure via immune deviation, leading to the eventual emergence of stable allergen-specific T cell memory which is polarized towards the Th1 (normal) or Th2 (atopic) phenotype. The underlying Th1/Th2 switching process is influenced by a number of host and environmental factors that are poorly understood. Prominent amongst these are factors that affect the kinetics of maturation of immune competence during the early postnatal period. In particular, there is mounting evidence that the immunological milieu at the materno-fetal interface is naturally skewed towards the Th2 phenotype (possibly an evolutionary adaptation to protect the placenta against the toxic effects of Th1 cytokines). Furthermore, this bias appears to be preserved for varying periods into infancy, which may account for the presence of a high risk 'window' for allergic sensitization in early postnatal life. It is hypothesized that the principal impetus for postnatal development of a normal Th1/Th2 balance (and hence closure of the high risk sensitization window) is provided via contact with Th1-stimulatory commensal and pathogenic micro-organisms at the body's major mucosal surfaces.

A killing defect of natural killer cells as an underlying immunologic abnormality in childhood systemic lupus erythematosus

OBJECTIVE

To elucidate the nature of the natural killer (NK) cell system in children with systemic lupus erythematosus (SLE), we performed phenotypic and functional studies of circulating NK cells during the course of childhood SLE. For comparison, similar examinations were undertaken in juvenile rheumatoid arthritis (JRA).

METHODS

Twenty-five children with SLE and 27 children with JRA were studied; 5 of these children with SLE were examined 6 to 67 months before the overt progression to SLE. The number and cytolytic function of NK cells were determined, using flow cytometry, 51Cr release, and single-cell cytotoxicity assays.

RESULTS

At the diagnosis of SLE, a decrease in NK cells defined as CD16+ or CD56+ was the most prominent of the numerical changes in lymphocyte subsets. In regard to cytolytic function, NK activity in children with SLE was greatly reduced at diagnosis: at the single cell level, their NK cells were defective in killing and recycling abilities. Although the relative number of NK cells and their recycling capacity returned to normal with the improvement of active SLE, the killing defect persisted during the inactive phase; there was no persistent NK cell abnormality in JRA. Reduced NK activity due to a killing defect was demonstrable early in the course of SLE: the NK activity and killing capacity values were profoundly decreased in 5 children before the overt progression to SLE.

CONCLUSION

It would appear that NK cell functional abnormality, characterized by a killing defect, is an underlying immunological abnormality during the course of childhood SLE.

Frequently relapsing minimal change nephrotic syndrome with natural killer cell deficiency prior to the overt relapse of Hodgkin's disease

A 15-year-old boy developed minimal change nephrotic syndrome (MCNS) during remission of Hodgkin's disease. Natural killer (NK) cell activity was practically absent at the onset of MCNS, with a value of 3% compared with the normal value of 44.1% +/- 7.8% (mean +/- SD). Treatment with prednisolone resulted in transient remission of MCNS and partial improvement of NK cell activity. Extensive investigations for Hodgkin's disease were performed at 1- to 3-month intervals; a relapse finally became apparent 25 months after the diagnosis of MCNS. Successful treatment of Hodgkin's disease resulted in complete disappearance of proteinuria and normalisation of NK cell activity. Frequently relapsing MCNS with NK cells deficiency during remission of Hodgkin's disease appears to imply its subclinical relapse.

Inhalant allergen-specific T-cell reactivity is detectable in close to 100% of atopic and normal individuals: covert responses are unmasked by serum-free medium

BACKGROUND

It is widely held that in vitro T cell responses to allergens are more prominent in atopic than in normal individuals, though this conclusion is based upon culture techniques which fail to detect proliferative responses in a significant minority of atopics and many normals.

OBJECTIVES

Study allergen-specific proliferative responses of T cells cultured in serum-free medium (SFM). Examine associations between atopic status, age and T cell reactivity.

METHODS

Initially, peripheral blood mononuclear cells were stimulated with allergens or antigens in SFM, and compared with cells cultured in RPMI + 10% fetal calf serum or human AB serum. Subsequently, T cell reactivity was studied in 34 adults (20-49 years), 27 children (2-13 years), and 19 infants (< or = 10 weeks) using SFM alone.

RESULTS

Compared with serum-supplemented medium, SFM enhanced net T cell proliferation, both in bulk culture and when cloning at limiting dilution. In many subjects, SFM unmasked T cell reactivity to allergens which was not otherwise evident, and lowered the threshold allergen levels required for in vitro T cell triggering. For most allergens, T cell proliferative responses did not differ between adults who had specific IgE, and those who did not. The most vigorous responses observed were to ubiquitous inhalant allergens, which stimulated T cells from close to 100% of adults and children, and over 60% of infants. In contrast, responses to the 'vaccine' antigen tetanus toxoid were completely absent in the latter age group, but present in the majority of adults and children.

CONCLUSIONS

These findings suggest that the extent of active T cell recognition of environmental allergens has been hitherto underestimated, and further that these responses may frequently be initiated in very early life. Additionally, these findings reinforce the notion that qualitative (as opposed to quantitative) variations in specific T cell reactivity ultimately determine allergen responder phenotype.

Frequent natural killer cell abnormality in children in an area highly contaminated by the Chernobyl accident

Since January 1991, we have been performing thyroid surveys and hematologic and immunologic screening on children in Chechersk, Belarus, a city situated in one of the areas most seriously contaminated with high levels of radionuclides after the Chernobyl accident. Ten children selected from 713 children because of goiter did not show a decrease in humoral immunity or in the number and function of T cells. By contrast, natural killer (NK) cell activity against K562 cells was depressed in 4 of these 10 children. The clinical and laboratory findings indicated that previously reported diseases with NK cell dysfunction could be excluded. A comparative analysis of NK cell activity in children from areas with and without high 137Cs levels revealed a high frequency of abnormal NK cell activity only in children from the area contaminated by radioactive fallout. In addition, there was no correlation between NK cell activity and NK cell number as percentage in the children from the area with high 137Cs levels. Neither activity nor number of NK cells was correlated with the body content of 137Cs. Thus, the frequent abnormality of NK cell function may not have been due to actual internal exposure to the long-lived radionuclide.

[Functional abnormalities of natural killer cells in childhood hemophagocytic syndrome]

To analyze the abnormality of natural killer (NK) cells in childhood hemophagocytic syndrome (HPS), we investigated the numbers, cytolytic functions of circulating NK cells, and responsiveness to cytokines by flow cytometry, 51Cr-releasing and single-cell assay in 13 patients. In the active phase, the relative number of CD 16+ cells was decreased in 3 patients. NK activity was depressed in 6 patients but normal in the remaining 7 in the active phase. Lymphokine-activated killer (LAK) activity showed the similar tendency. The responsiveness of NK cells to IFN-gamma was absent, but that to IL-2 was normal in the active phase. Single-cell assay showed depressed maximal recycling capacity (MRC) and killing capacity values in the active phase and in remission phase, respectively. From these results, NK cells have such functional abnormalities as the absence of responsiveness to IFN-gamma and depressed recycling capacity in the active phase of childhood HPS. Further, abnormality of killing capacity of NK cells may exist even in the remission phase.

Absence of responsiveness to interferon-gamma but not to interleukin 2 and depressed recycling as natural killer cell abnormalities in childhood lymphohistiocytic syndrome

To characterize the abnormalities of natural killer (NK) cells in childhood lymphohistiocytic syndrome (LHS), we investigated the number and cytolytic functions of circulating NK cells in 10 LHS children using flow cytometry, 51Cr-release and single-cell assays. In the active phase, the numbers of CD16+ or CD56+ cells and NK activity were normal in more than half of the patients or otherwise decreased. Despite the treatment with interferon-gamma (IFN-gamma) there was no significant increase in NK activity in the children with LHS: the values (mean +/- S.D.) of 32.2% +/- 14.2% became 35.0% +/- 13.3% (P > 0.05) when the control values changed from 45.5% +/- 8.5% to 54.2% +/- 10.1% after the stimulation. However, the NK cells normally responded to interleukin 2 (IL-2). In contrast, NK cells from 9 patients with infectious mononucleosis (IM) responded well to both IFN-gamma and IL-1 (P < 0.01). At the single-cell level, their NK cells had defective recycling capacity with normal killing capacity. The maximal recycling capacity (MRC) values (mean +/- S.D.) were 3.6 +/- 0.8 as compared to the control value of 5.5 +/- 0.9 (P < 0.01). Two of the patients studied had extremely high levels of serum IFN-gamma (9.8 U/ml and 158.0 U/ml) as compared to the control value of < 0.4 U/ml. NK cells may have been strongly stimulated by IFN-gamma in vivo, which probably yields superficially normal NK cell activity in the face of the absence of responsiveness to IFN-gamma but not to IL-2. The defective recycling may be another abnormality of NK cells in LHS.

[Hemophagocytic syndrome with high level of interferon-gamma in the advanced stage]

Serum cytokines, lymphocyte subsets of peripheral blood and natural killer cell activity were serially assayed in a 3-year-old girl with hemophagocytic syndrome. Laboratory findings showed pancytopenia, increased levels of transaminases and hyperferritinemia and proliferation of histiocytes in bone marrow and pleural effusion. The administration of prednisolone resulted in a temporary improvement followed by a further exacerbation. The patient died in spite of the treatment with VP-16 and THP-adriamycin. Serum interferon-gamma (IFN-gamma) markedly increased in active phases. TNF-alpha, IL-1 beta and GM-CSF concentrations were normal or slightly elevated. The CD4/8 ratio of peripheral lymphocytes and natural killer cell activity also fluctuated according to clinical stage. These results suggested that IFN-gamma was the most important cytokine to activate histiocytes in this case.

Fetal liver T cell receptor gamma/delta+ T cells as cytotoxic T lymphocytes specific for maternal alloantigens

We have established fetal liver-derived T cell receptor (TCR) gamma/delta+, CD3+ T cell lines that are cytotoxic for maternal T cells. Fetal liver-derived lymphoid progenitors yielded predominantly TCR-gamma/delta+ cell clusters when cultured on fetal bone marrow-derived stromal cells in the presence of a cytokine cocktail under magnetic force. These tightly adherent clusters were cloned by limiting dilution and the resulting cell lines analyzed for phenotype and function. Six of eight TCR-gamma/delta lines from 8-9.5-wk gestation fetuses were V delta 2+ as compared with zero of eight lines from later stages of gestation (10 and 15 wk), where all the lines were V delta 1+. In cytotoxicity assays, these TCR-gamma/delta+, CD3+, CD4-, and CD8+ or CD8- long-term cultured lymphoid cells (LLC) were killer cells active against the class I antigens on maternal T cells. Of the cell lines, the CD8+ TCR-gamma/delta+ LLC had the highest levels of killer activity. Thus fetal liver TCR-gamma/delta+ T cells may play a crucial role in protection against invading maternal T cells generated in the feto-maternal interaction.

A recycling defect as a characteristic of natural killer cells in childhood acute lymphoblastic leukemia

The cytolytic function of natural killer (NK) cells and their responsiveness to interferon-alpha and IL-2 were investigated in children with acute lymphoblastic leukemia (ALL) using 51Cr-release and single-cell assays. For comparison, such NK cell functions were similarly assayed in neuroblastoma. NK activity in ALL children was extremely low at onset, but it increased gradually during remission and finally reached normal levels. At the single-cell level, their NK cells at onset were defective in the binding, lytic, and recycling abilities. Although the binding and lytic defects improved to normal levels during remission, the recycling, which increased gradually during remission, was still low even after the long-term remission in ALL: the maximal recycling capacity values were 1.9 +/- 0.4 (p less than 0.001) at onset and 4.6 +/- 0.6 (p less than 0.05) after 5 y of complete remission, as compared to the value in control children of 5.4 +/- 0.7. On the other hand, children with neuroblastoma had no recycling defect after completing the therapy: their maximal recycling capacity value was 5.6 +/- 0.7. Bone marrow cells in ALL were also depressed in their recycling ability at all stages. Interferon-alpha and IL-2 could enhance NK activity and IL-2 could generate lymphokine-activated killer activity at all stages of ALL; however, the recycling defect hardly improved with these treatments. Thus, NK cells in childhood ALL have a recycling defect as a functional characteristic.

Development of natural killer cytotoxicity during childhood: marked increases in number of natural killer cells with adequate cytotoxic abilities during infancy to early childhood

The cytotoxicity of natural killer (NK) cells against K562 cells and their responsiveness to interferon-alpha and interleukin 2 (IL-2) were studied throughout childhood using 51Cr-release and single-cell assays. Although NK activity was extremely low in the neonatal period, it almost reached the adult level during 1 to 5 mo of age and remained at that level thereafter. At the single-cell level, the binding, lytic, and recycling abilities were also depressed in the neonatal period, but these abilities improved conspicuously after this period; in particular, the lysis and recycling were at higher levels during 6 mo to 4 y of age. The absolute numbers of circulating cytotoxic NK cells were high during infancy to early childhood: they were 54 +/- 24 (mean +/- SD/mm3) in neonates, 115 +/- 48 in 1- to 5-mo-old infants, 121 +/- 42 in 6- to 12-mo-old infants, 93 +/- 26 in 1- to 4-y-old children, and 42 +/- 16 in adults. Interferon-alpha and IL-2 could enhance NK activity throughout childhood. The IL-2 enhancement was prominent especially in the neonatal period; IL-2 yielded a 2.5-fold increase in the number of cytotoxic cells and improved the recycling to the adult level. At older ages, interferon-alpha and IL-2 yielded 1.4- and 1.9-fold increases in the number of cytotoxic cells, respectively, but did not enhance the recycling. The increased number of NK cells with adequate cytotoxic abilities during infancy to early childhood indicates the predominance of NK immunity during these periods. IL-2 is a cytokine that induces high levels of NK cytotoxicity even in neonates.

Natural killer cell immunodeficiency in siblings: defective killing in the absence of natural killer cytotoxic factor activity in natural killer and lymphokine-activated killer cytotoxicities

A immunodeficiency of natural killer cells as effectors for natural killer and lymphokine-activated killer cytotoxicities was first demonstrated in siblings. Two of three male siblings persistently lacked natural killer activity against K562 target cells as assayed by a 51Cr-release assay: percent lysis values were less than 1.0% as compared to the normal lymphocyte values of 43.5% +/- 6.2% (mean +/- SD). Their lymphocytes did not develop natural killer cell activity by changing effector to target ratios, prolonging the incubation time, or stimulating them with interferon-alpha or interleukin 2. Numbers of lymphocytes bearing Leu-7, CD16, or NKH-1 were normal but those of Leu-7-, CD16+ cells were decreased as estimated by flow cytometry. Single cell-in-agarose assays showed normal numbers of natural killer cells capable of binding to a target cell but incapable of killing it. They had depressed levels of lymphokine-activated killer activity, which was totally eliminated by the treatment with OKT3 and complement. This result indicates that the patients' natural killer cells are also defective in the capacity to work as effectors for lymphokine-activated killer activity. The patients' natural killer cells did not produce natural killer cytotoxic factor activity. Antibody-dependent cellular cytotoxicity and cytotoxic T lymphocyte cytotoxicity were normal. These results demonstrate a selective natural killer cell deficiency as effectors for natural killer and lymphokine-activated killer cytotoxicities with a familial tendency, in which there is defective killing with the absence of natural killer cytotoxic factor activity.

Giant lymph node hyperplasia (Castleman's disease) with spontaneous production of high levels of B-cell differentiation factor activity

A 13-year-old girl presented with general fatigue, back pain, anemia, hyperimmunoglobulinemia, and a mediastinal mass on chest radiograph. A mass was surgically removed, and its histologic examination determined the diagnosis of giant lymph node hyperplasia (Castleman's disease). With removal of the hyperplastic lymph node, the clinical symptoms soon disappeared and the abnormal laboratory findings were markedly improved within 1 month: serum IgG levels decreased from 4350 mg/dl to 1829 mg/dl. Immunostaining on the lymph node sections revealed polyclonal B-lymphocyte and T-lymphocyte populations. The patient's lymph node cells were cultured without any mitogenic stimulation, and the culture supernatants were assayed for their B-cell differentiation factor (BCDF) activity to induce IgG production by our Epstein-Barr virus-transformed cell line. The patient's lymph node cells produced high levels of BCDF activity: the supernatants could increase the IgG production from 140 ng/ml to 410 ng/ml when the values became from 140 ng/ml to 142 ng/ml or 148 ng/ml with those of the control lymph node cells. These results suggest that the hyperimmunoglobulinemia and its prompt improvement with removal of the hyperplastic lymph node may have been related to the spontaneous production of high levels of BCDF activity by the lymph node cells in the patient.

Induction of lymphokine-activated killer and natural killer cell activities from cryopreserved lymphocytes

Lymphokine-activated killer (LAK) and natural killer (NK) cells were studied for their capacity to retain cytotoxicity after cryopreservation. LAK cells were generated by a 4-day culture of lymphocytes with recombinant interleukin-2 (rIL-2). Cytotoxicity was measured by 51Cr-release assay at effector:target ratios of 10:1 to 80:1. Cryopreserved LAK cells retained 58.8 to 87.4 percent of cytotoxicity, as compared with that in fresh control cells. Cryopreserved NK cell activity against K562 and Molt-4 targets was 45.7 to 67.9 percent of the respective values of the fresh control cells. The responsiveness of NK cells to polyinosinic-polycytidilic acid (poly I:C), interferon-alpha (IFN-alpha), or rIL-2 remained intact. Activated NK cell activity after poly I:C or IFN-alpha stimulation and that after rIL-2 were, respectively, comparable to and higher than the endogenous NK cell activity of the fresh cells. The composition of lymphocyte subsets as determined by flow cytometry using monoclonal antibodies did not change after cryopreservation, indicating that cellular loss of the given subsets did not occur during the procedure. The retention of substantial levels of cytotoxicity in cryopreserved LAK and NK cells may make them promising candidates as cytotoxic effector cells.