Phenotype and antitumor activity of ascitic fluid monocytes in patients with ovarian carcinoma.

Monocytes/macrophages (MO/MA) represent a major leukocyte population in the peritoneal cavity of patients with epithelial ovarian cancer (EOC). We examined the phenotypic characteristics and antitumor cell activity of ascitic MO in patients with EOC. MO/MA phenotype was compared with MO in peripheral blood by two- and three-color flow cytometry. Cytotoxic/cytostatic effects of different cytokines on cultured EOC cells were measured by initial labeling or uptake inhibition of [methyl-3H] thymidine. Malignant ascites had higher proportion of MO/MA with the CD14brightCD16+ phenotype than peripheral blood. Cell surface antigen expression of activation and differentiation in peripheral blood and ascites, including CD38, CD40, CD64, and CD86, was higher on CD14brightCD16- and CD14brightCD16+ than on CD14dimCD16- cells. HLA-DR expression was higher on ascitic MO/MA than peripheral blood MO. Significant cytotoxic/cytostatic activity was elicited by treating ascitic MO/MA with interferon-gamma (IFN-gamma) and interleukin-2 (IL-2), but not with interleukin-12, paclitaxel, granulocyte-monocyte colony-stimulating factor (GM-CSF), or tumor necrosis factor-alpha (TNF-alpha). Soluble CD40Lt did not enhance MO/MA cytotoxic activity, and inhibited IFN-gamma or IL-2 induced cytoxicity. We conclude that MO/MA from ascites have elevated proportions of CD14brightCD16+ cells, showing phenotypic features of activation. IFN-gamma induces the cytotoxic and cytostatic activity of MO/MA that is inhibited by CD40Lt.

Maturation of dendritic cells from ovarian cancer patients.

PURPOSE: Dendritic cells (DC) are the most potent antigen-presenting cells of the immune system. We have shown that DC, defined as LN-DR+ leukocytes from the ascites of patients with ovarian or peritoneal carcinoma, have the cell surface characteristics of immature cells. Moreover, p70 interleukin-12 has not been detected in the ascites of ovarian cancer patients in vivo. In the current study, we examined the effects of in vitro treatment of DC from peripheral blood and ascites samples of patients with ovarian cancer with either cytokines or proteolytic enzymes (polyenzyme preparation). METHODS: Mononuclear leukocytes from the ascites of 15 patients or peripheral blood from 9 patients with epithelial ovarian cancer were cultured with tissue culture medium containing either papain, trypsin and chymotrypsin for 5-7 days or recombinant human granulocyte-macrophage colony-stimulating factor, tumor necrosis factor alpha and interleukin-4 (complete medium) or tissue culture medium alone. RESULTS: Phenotypic analysis of DC obtained on days 5-7 of the culture showed higher proportions of CD83+, CD40+ and CD80+ cells when cultured with cytokines or enzymes as compared with DC cultured with medium alone. Stimulation of allogeneic T cells was detected by the mixed leukocyte reaction (MLR) and higher concentrations of IL-12 were detected after growing these cells in the presence of cytokines or enzymes as compared to tissue culture medium alone. CONCLUSION: Our studies demonstrate for the first time that DC obtained from the peritoneal cavity and peripheral blood of ovarian cancer patients after culturing in the presence of a polyenzyme preparation, will undergo maturation. Further studies are warranted to determine whether polyenzyme preparations facilitate DC maturation in vivo.

Impairment of antigen-specific cellular immune responses under simulated microgravity conditions.

Microgravity has been implicated to play a role in the observed immune dysfunction of astronauts and cosmonauts after either short-term or long-term space travel. These reports, together with studies describing increased levels of microorganisms in the space cabin environment suggest potential risk for in-flight incidences of infectious diseases. In order to understand the mechanism underlying these immune defects, it is important to have a ground-based model that would reliably mimic the effects of microgravity on antigen-specific immune function. We tested the utility of the rotating wall vessel (RWV) technology developed at NASA as a model system because in the RWV the culture medium and the cells rotate synchronously with the vessel, thereby creating simulated microgravity conditions. We compared the RWV to the conventional tissue culture flask (T-flask), for culturing immune precursor cells with cytotoxic T lymphocyte (CTL) activity against synthetic viral peptides. We observed a significant loss of antigen-specific CTL activity in RWV cultures, but not in those from the T-flask, irrespective of the peptide immunogen used for inducing the primary immune response in different mouse strains. Loss of CTL activity in RWV cultures coincided with a significant reduction in CD8+ cells as well as CD4+ cells and DEC205+ dendritic cells, suggesting adverse effects of RWV culturing on both the effector and accessory cells for the loss of antigen-specific CTL function. These results provide a strong parallel to the reported defects in cell-mediated immunity during space travel and strongly support the utility of the RWV technology as an effective ground-based model for identifying key steps in immune cell dysfunction related to microgravity.

Characteristics of human dendritic cells generated in a microgravity analog culture system.

Generation of an effective immune response requires that antigens be processed and presented to T lymphocytes by antigen-presenting cells, the most efficient of which are dendritic cells (DC). Because of their influence on both the innate and the acquired arms of immunity, a defect in DC would be expected to result in a broad impairment of immune function, not unlike that observed in astronauts during or after space flight. In the study reported here, we investigated whether DC generation and function are altered in a culture environment that models microgravity, i.e., the rotary-cell culture system (RCCS). We observed that RCCS supported the generation of DC identified by morphology, phenotype (HLA-DR+ and lacking lineage-associated markers), and function (high allostimulatory activity). However, the yield of DC from RCCS was significantly lower than that from static cultures. RCCS-generated DC were less able to phagocytose Aspergillus fumigatus conidia and expressed a lower density of surface HLA-DR. The proportion of DC expressing CD80 was also significantly reduced in RCCS compared to static cultures. When exposed to fungal antigens, RCCS-generated DC produced lower levels of interleukin-12 and failed to upregulate some costimulatory/adhesion molecules involved in antigen presentation. These data suggest that DC generation, and some functions needed to mount an effective immune response to pathogens, may be disturbed in the microgravity environment of space.

Dendritic cell-mediated stimulation of the in vitro lymphocyte response to Aspergillus.

Lymphocytes play a major role in host defense against Aspergillus, but little is known about the contribution of dendritic cells (DC) to antifungal immunity in humans. We have observed that DC derived from normal volunteers phagocytose heat-killed A. fumigatus conidia. Following 24 h of exposure to the fungus, DC displayed an increase in the mean fluorescence intensity of HLA-DR, CD80, and CD86, and an increase in the percentage of CD54(+) cells. These DC also displayed increased production of IL-12. DC derived from CD34(+) progenitors or monocytes stimulated autologous lymphocytes to proliferate and produce high levels of interferon-gamma, but not interleukin-10, in response to fungal antigen. DC generated from CD34(+) progenitors collected prior to autologous or allogeneic stem cell transplantation also partially restored the in vitro antifungal proliferative response of lymphocytes obtained from patients 1 month after transplantation. These results suggest that DC are important to host-response to A. fumigatus, and that ex vivo-generated DC might be useful in restoring or enhancing the antifungal immunity after hematopoietic stem cell transplantation.

Pretransplant tumor antigen-specific immunization of allogeneic bone marrow transplant donors enhances graft-versus-tumor activity without exacerbation of graft-versus-host disease.

Allogeneic bone marrow transplantation (BMT) causes a beneficial graft-versus-tumor (GVT) immune response that is often associated with graft-versus-host disease (GVHD). There is substantial interest in developing therapeutic strategies that augment GVT without GVHD. We have demonstrated recently that immunization of BMT donors with cellular tumor vaccines leads to curative GVT but induces unacceptable GVHD because of the presence of recipient minor histocompatibility antigens (mHAgs) in whole-cell tumor vaccines. This study tested the hypothesis that immunization of BMT donors against a defined tumor-specific antigen with a vaccine not containing recipient mHAgs would help to separate the two responses by enhancing GVT activity without exacerbating GVHD, even when cellular vaccines were used after BMT. Recipient strain C57BL/6 fibrosarcoma cells engineered to express the well-characterized model tumor antigen, influenza nucleoprotein (NP), were used in these studies. C3H.SW donors were immunized against NP prior to BMT, and cytolytic T cells were transferred along with bone marrow into irradiated H-2-matched, mHAg-mismatched C57BL/6 recipients with established micrometastatic 205-NP tumors. Donor immunization led to a significant increase in GVT activity, as measured by reduction in tumor growth and enhanced survival. However, deaths in recipients of tumor antigen-specific immune BMT ultimately occurred because of the growth of antigen-loss variants; such tumor growth did not occur in animals receiving BMT from donors treated with whole-cell vaccines. Donor immunization did not lead to an exacerbation of GVHD, even when BMT recipients received additional immunization after BMT with a 205-NP "whole" tumor cell vaccine (which was shown to induce fatal GVHD when used for donor immunization). In conclusion, immunization of allogeneic BMT donors against a tumor-specific antigen significantly enhances GVT activity without an associated exacerbation of GVHD.

Immunization of allogeneic bone marrow transplant recipients with tumor cell vaccines enhances graft-versus-tumor activity without exacerbating graft-versus-host disease.

Allogeneic bone marrow transplantation (BMT) induces 2 closely associated immune responses: graft-versus-tumor (GVT) activity and graft-versus-host disease (GVHD). We have previously shown that pretransplant immunization of allogeneic BMT donors with a recipient-derived tumor cell vaccine increases both GVT activity and lethal GVHD because of the priming of donor T cells against putative minor histocompatibility antigens (mHAgs) on the tumor vaccine cells. The work reported here tested the hypothesis that tumor cell vaccination after BMT would produce an increase in GVT activity without exacerbating GVHD. C3H.SW donor bone marrow and splenocytes were transplanted into major histocompatibility complex-matched, mHAg-mismatched C57BL/6 recipients. One month after BMT, recipients were immunized against either a C57BL/6 myeloid leukemia (C1498) or fibrosarcoma (205). Immunized recipients had a significant increase in survival and protection against tumor growth in both tumor models, and significant tumor protection was seen even in recipients with preexisting micrometastatic cancer before immunization. Alloreactivity appeared to contribute to the in vitro anti-tumor cytolytic activity, but in vivo immunity was tumor specific, and no exacerbation of GVHD was observed. Although the immunodominant mHAg B6(dom1) was shown to be expressed by all B6 tumors tested and was largely responsible for the alloreactivity resulting from tumor immunization of donors, the in vitro alloreactivity of immune recipients was more restricted and was not mediated by recognition of B6(dom1). In conclusion, post-transplant tumor immunization of allogeneic BMT recipients against either a leukemia or a solid tumor can increase GVT activity and survival without exacerbating GVHD.

Pathogenesis I: interactions of host cells and fungi.

The interactions of host cells and fungi during infection represent a complex interplay. Although T helper 1 (Th1)-mediated immunity is primarily responsible for acquired resistance to Paracoccidioides brasiliensis, studies have demonstrated that polymorphonuclear neutrophils play a critical role in providing an early resistance to this organism. One study has shown that the invasiveness of Candida albicans requires adherence, particularly to endothelial cells, which in turn are stimulated to express various cell-markers and pro-inflammatory cytokines as part of a proactive resistance to invasion. Somewhat in contrast to infection with C. albicans, it has been shown that the capsular glucuronoxylomannan of Cryptococcus neoformans causes the shedding of host-cell adherence molecules (L-selectins) needed for the migration of host-inflammatory cells to sites of infection and likely explains, in part, the reduced host inflammatory response to this organism. Resistance to aspergillosis is often associated with the immune status of the host. In one set of studies, it has been demonstrated that lymphocytes have little direct effect on the organism, but that antigen-presenting dendritic cells stimulate the production of Th1 cytokines, suggesting a positive role for the dendritic cell in host-response. Similarly, another study has shown that among the regulatory cytokine networks that Th2-associated cytokines (e.g., interleukin-10) likely play a detrimental role in the resistance of the host to Aspergillus fumigatus.

HER-2/neu peptide specificity in the recognition of HLA-A2 by natural killer cells.

Although natural killer (NK) cells have been described as non-MHC-restricted, new evidence suggests that NK activity can be either up- or down-regulated after interaction with the peptide-MHC-class-I complex expressed on target cells. However, the epitope(s) recognized by NK cells have remained ill-defined. We investigated NK cell recognition of synthetic peptides representing a portion of a self-protein encoded by the HER-2/neu (HER-2) proto-oncogene and presented by HLA-A2. HER-2 nonapeptides C85, E89, and E75 were found partially to protect T2 targets from lysis by freshly isolated and interleukin-2(IL-2)-activated NK cells (either HLA-A2(+) or A2(-)). This inhibition was not solely due to changes in the level of HLA-A2 expression or conformation of serological HLA-A2 epitopes. Using single-amino-acid variants at position 1 (P1) of two HER-2 peptides, we observed that protection of targets was dependent on the sequence and the side-chain. These results suggest similarities in the mechanism of target recognition by NK and T cells. This information may be important for understanding the mechanisms of tumor escape from immunosurveillance and could help explain the aggressiveness of HER-2-overexpressing tumor cells.

Multidimensional flow-cytometric analysis of dendritic cells in peripheral blood of normal donors and cancer patients.

We studied the potential of multidimensional flow cytometry to evaluate the frequency and maturation/activation status of dendritic cells in minimally manipulated peripheral blood mononuclear cell preparations (i.e., only separated on Ficoll-Hypaque) of normal donors and cancer patients. A rare subset of HLA-DR+ leukocytes (less than 1% mononuclear cells) was detected in blood of normal donors that displayed all the features of dendritic cells: these cells had high forward-light-scatter characteristics and coexpressed CD4, CD86 and CD54 surface antigens, but lacked the lineage-associated surface markers of T cells, B cells, monocytes, granulocytes or NK i.e. they were CD3-, CD19-, CD20-, CD14-, CD11b-, CD16-, CD56-). These physical and phenotypic properties were virtually identical to those of immunomagnetically sorted leukocytes characterized as dendritic-cells on the basis of morphology, phenotype and high stimulatory activity in allogeneic mixed-lymphocyte cultures. Using this flow-cytometric approach we observed that the frequency of dendritic cell-like cells in peripheral blood mononuclear cell specimens of cancer patients receiving chemotherapy alone or those recovering from stem cell transplantation was significantly lower than that of normal individuals (mean +/- SE: 0.36 +/- 0.05%, 0.14 +/- 0.06%, and 0.75 +/- 0.04% respectively). Multidimensional flow-cytometric analysis of dendritic cells might represent an important new tool for assessing immunocompetence, and for monitoring the effects of therapeutic regimens on the immune system.

Activated lymphocytes reduce adherence of Aspergillus fumigatus.

Lymphocytes comprise up to 30% of the cells present in human bronchoalveolar lavage fluid and thus could participate in host response to infectious Aspergillus fumigatus conidia. We have examined the possibility that lymphocytes might play a role during early infection by either damaging the fungus or interfering with adherence. When incubated with A. fumigatus conidia for 20 h, highly purified 5-day-old lymphocytes activated with either IL-2 or phytohaemagglutinin, but not untreated lymphocytes, were consistently able to reduce residual fungal biomass as estimated by a metabolic assay. T lymphocytes, but not NK cells, appeared to be responsible for this activity. Lymphocytes bound both A. fumigatus conidia and hyphae, and the antifungal activity of the lymphocytes required direct lymphocyte fungus contact. In a separate set of experiments using release of 51Cr from 51Cr-loaded fungi as an estimate of fungal damage, lymphocyte-induced loss of fungal biomass was found to be due to loss of fungal adherence rather than to direct fungal damage. The detached hyphae were also found to be metabolically intact and to have normal morphology by electron microscopy. These data demonstrate that IL-2- and phytohaemagglutinin-activated lymphocytes exhibit a contact-dependent ability to reduce adherence of germinating conidia of A. fumigatus to a surface.

Aspergillus fumigatus conidia induce a Th1-type cytokine response.

The response of human peripheral blood mononuclear cells (MNC) to Aspergillus fumigatus in vitro was evaluated. In studies of the proliferative response of MNC from 18 healthy donors to heat-killed A. fumigatus conidia, 15 displayed a significant response, with a stimulation index (SI) between 4 and 193. In contrast, all donors displayed a positive response to Candida albicans blastoconidia (SI ranged from 10 to 224). Despite the variability in reactivity to A. fumigatus conidia, the response of a particular individual was stable when retested over periods of 1-2 weeks. Supernatant from cocultures of A. fumigatus conidia with MNC contained increased levels of interferon-gamma, granulocyte-macrophage colony-stimulating factor, tumor necrosis factor-alpha, and interleukin (IL)-2, compared with unstimulated cells, but not IL-10 or IL-4. In addition, A. fumigatus induced lymphocyte surface expression of adhesion/activation-associated molecules. These results suggest that lymphocytes may contribute to host defense against Aspergillus by generating a Th1-type response.

Effects of in vitro and in vivo cytokine treatment, leucapheresis and irradiation on the function of human neutrophils: implications for white blood cell transfusion therapy.

Treatment of human polymorphonuclear leucocytes (PMNL) separated by density sedimentation (DS) from normal donors (PMNL-NL-DS) with interferon-gamma (IFN-gamma) + granulocyte colony-stimulating factor (G-CSF) lessens the damage caused by isolation and irradiation. We have studied granulocyte-macrophage colony-stimulating factor (GM-CSF) in this system, as well as the behaviour of PMNL collected by continuous flow leucapheresis (CFL) from donors treated with G-CSF (PMNL-GCSF-CFL). After isolation, PMNLs were treated with IFN-gamma + G-CSF, GM-CSF or IFN-gamma + G-CSF + GM-CSF, irradiated with 0 or 30 Gy and studied after 0 and 20 h in cell culture. All regimens reduced apoptosis of PMNL-NL-DS. Killing of Candida albicans by 20-h-old PMNL-NL-DS was best preserved by IFN-gamma + G-CSF treatment. A similar pattern of results was obtained for assays of PMNL-NL-DS chemotaxis and superoxide production. There was a consistent trend toward reduced function after irradiation in all assays. PMNL-GCSF-CFL less often demonstrated the morphological features of apoptosis, and this was further reduced by cytokine regimens containing IFN-gamma + G-CSF. In assays of C. albicans killing and chemotaxis, 20-h-old untreated PMNL-GCSF-CFL performed as well as freshly isolated PMNL-GCSF-CFL. PMNL-GCSF-CFL showed decay in CD11b (CR3), CD16 (Fc gamma III) and CD64 (Fc gamma R1) expression after 20 h in cell culture, but treatment with IFN-gamma + G-CSF preserved expression. There was a trend toward reduced function after radiation. Comparison of PMNL-GCSF separated by CFL and DS demonstrated that CFL itself is a strong inducer of the morphological features of apoptosis. This study shows that while separation by CFL, and irradiation are damaging to PMNLs, damage may be reduced by use of cytokines.

Indirect stimulatory effects of murine interleukin-12 on in vitro production of nitric oxide by mouse peritoneal cells.

The effect of murine interleukin-12 (IL-12) on L-arginine-dependent biosynthesis of nitric oxide (NO) by mouse peritoneal cells was evaluated. Interleukin-12 was found to trigger considerably enhanced production of NO in a dose-dependent manner. Antibody neutralization studies indicated that the effect of IL-12 was mediated by IFN-gamma without apparent participation of TNF-alpha. Synergistic effects of IL-12 plus lipopolysaccharide (LPS) were also observed. Our data thus provide evidence that IL-12 is a powerful but indirect modulator of NO formation. These findings may contribute to the better understanding of various biologic effects of IL-12.

Analysis of interleukin-2-activated killer cells of rhesus monkeys: striking resemblance to the human system.

We have found numerous and exquisite homologies between the interleukin-2 (IL-2)-activated killing systems of rhesus monkeys and humans. Lymphocytes with high oncolytic and proliferative activity were generated from peripheral blood, spleen, and bone marrow of monkeys after culture with IL-2. The distribution of lymphocyte subsets in IL-2 cultures closely paralleled that seen in humans, including a decrease in CD4+ and increase in CD8+, CD38+, and CD25+ lymphocytes and an increase in density of CD2 molecules. We also describe three distinct subsets of monkey lymphocytes, CD16+,56-, CD16+,56+"dim", and CD16-,56+"bright", and show that the CD56+"bright" subset is substantially increased (to as high as 79%) after IL-2 activation. Furthermore, as in humans, the cells with oncolytic activity were characterized as CD56+, CD16+/-, and CD8+. This strong homology with humans indicates that the rhesus monkey may be a valuable preclinical model for evaluation of therapeutically relevant biological response modifiers.

Human natural killer cell development from bone marrow progenitors: analysis of phenotype, cytotoxicity and growth.

We have developed a culture system allowing for generation of NK cells from human CD34+ bone marrow progenitors. The appearance of NK cells expressing CD56+, CD3- phenotype and large granular lymphocyte morphology was observed after 2-3 weeks of culture with IL-2. The NK cell appearance coincided with development of lytic activity. NK cells generated in bone marrow cultures proliferated actively (expansion index ranged from 2- to 200-fold). The phenotype of NK cells generated from CD34+ bone marrow deviated from peripheral blood NK cells in that a lower percentage of the former cells expressed CD16, CD2, CD7, and CD8 antigens. NK cells were also generated from CD34+ populations depleted of the CD34+, CD33+ subset indicating that myeloid-committed progenitors are not required for NK cell development. The dose of IL-2 was not important for generation of NK cells; however, only high doses of IL-2 supported development of optimal NK cell cytotoxic potential. Addition of TNF-alpha facilitated IL-2-dependent NK cell generation. These data showed that NK cells can develop from early bone marrow progenitors and that this system may be instrumental in studies on NK cell lineage and differentiation.

Genesis of human oncolytic natural killer cells from primitive CD34+CD33- bone marrow progenitors.

We have found that fully functional CD56+CD3- NK cells can be generated from highly enriched populations of CD34+ human bone marrow progenitors in long term bone marrow culture (LTBMC) with IL-2. Strong lytic activity against K-562 cells was observed after 2 to 3 weeks of culture and coincided with the appearance of CD56+CD3- large granular lymphocytes, which comprised 55 to 84% of cells. In contrast, lymphocytes generated from CD34- bone marrow cells were predominantly CD56-CD3+ T cells. The phenotypic profile of NK cells generated in LTBMC differed from that of IL-2-cultured peripheral blood NK cells. Specifically, a lower percentage of LTBMC-derived NK cells coexpressed CD16, CD2, CD7, and CD8 Ag. NK cells generated in LTMBC proliferated actively, with an expansion index ranging from two- to 200-fold. Furthermore, NK cells were generated from the CD34+CD33- hematopoietic subset, which had been depleted of the myeloid-committed progenitors (CD34+CD33+). Generation of cytotoxic NK cells in LTBMC was dependent on the dose of IL-2; although CD56+CD3- NK cells were generated in LTBMC supplemented with a broad range of doses (10 to 10(3) U/ml) of IL-2, the acquisition of full cytotoxic function required a high concentration of IL-2 (10(3) U/ml). These observations demonstrate the IL-2-driven differentiation of NK cells from early bone marrow progenitors and indicate that LTBMC may provide an excellent model for studies of NK cell lineage and differentiation in normal and pathologic conditions.

Induction of cytotoxic lymphocyte subsets against leukemia by stimulation with AML blasts.

Although the application of interleukin-2 (IL-2) activated lymphocytes in immunotherapy of acute myelogenous leukemia (AML) is of therapeutic interest, the high resistance of AML blasts to lymphocyte lysis may represent an obstacle to this type of therapy. However, our data shows that the leukemia resistance can be conquered by concomitant culture of lymphocytes with IL-2 and AML blasts. This approach induces not only leukemia-directed cytotoxic cells, but also promotes their growth. Additionally, multiple cytotoxic lymphocyte populations with leukemia lytic activity are induced in AML/IL-2 cultures. These include natural killer (NK) cells and subsets of T cells with both the major histocompatibility complex (MHC)-restricted and MHC-nonrestricted cytotoxic function. Thus, this protocol, which is conducive to general stimulation of cellular immune responses against leukemia, may enhance the benefits of lymphocyte therapy.

Adhesion molecules on MHC-nonrestricted lymphocytes: high density expression and role in oncolysis.

We have shown that interleukin-2 (IL-2) -activated adherent lymphocytes (A-LAK) display superior oncolytic activity against acute myelogenous leukemia (AML) blasts when compared to conventionally prepared lymphocytes with lymphokine-activated killing (LAK) activity. The A-LAK activity was generated promptly and from donors whose lymphocytes did not display any LAK activity. In comparison to LAK, a higher percentage of A-LAK expressed the CD25 and HLA class II (HLA-DR) activation-associated structures and high density of HLA class I antigens. Most striking, however, was the observation that lymphocytes from A-LAK cultures consistently contained a high density of CD2, CD11a, and CD18 adhesion molecules, as indicated cytometrically by their "bright" fluorescence intensity. Three color flow cytometric analysis indicated that virtually all CD56+,CD3- natural killer (NK) and CD56+,CD3+ T cells in unstimulated, LAK and A-LAK populations displayed this "bright" phenotype, while most CD56-,CD3+ T cells (with the exception of the small proportion found in A-LAK) were of the "dim" phenotype. The A-LAK cultures also contained a higher percentage of lymphocytes expressing CD11b (CR3 receptor) and CD54 (ICAM-1) antigens. The CD11a, CD18, and partially CD2 molecules were important in the A-LAK cytolytic mechanism against AML, since blocking of these structures with monoclonal antibodies (MAb) significantly decreased the antileukemia effect. Additionally, the ability of A-LAK to adhere to plastic was most strongly inhibited by anti-CD11a MAb and less, but significantly, by MAb against CD2, CD18, and CD56 molecules.(ABSTRACT TRUNCATED AT 250 WORDS)

Generation of MHC-nonrestricted and restricted oncolytic subsets from human bone marrow.

We analyzed the cytotoxicity and characterized the phenotype of oncolytic bone marrow (BM) lymphocyte subsets generated in vitro by interleukin-2 (IL-2) and stimulator cells (SC). Two irradiated B-lymphoblastoid cell lines (Daudi and EBV-transformed BSM) and fresh human acute myelogenous leukemia (AML) were used as SC. Stimulation with Daudi and IL-2 resulted in a substantial increase in cytotoxic activity (100- to 1000-fold) against a broad range of tumor targets, and total cellular expansion was higher compared to stimulation with IL-2 alone. The most prominent increase was observed in the CD16+ and CD56+/CD3- natural killer (NK) cell subset; however, a significant increase was also observed in CD56+/CD3+ T cells. Functional analysis of Daudi- and IL-2-generated subsets using fluorescence-activated cell sorting (FACS) revealed that most of the lytic activity was mediated by NK cells. Significant potentiation of oncolytic activity and cell growth was also seen in the cultures stimulated with BSM or fresh AML and IL-2. The highest oncolytic activity in the latter cultures was mediated primarily by CD8+, CD3+, and CD56- T cells, although NK cells also participated in cytotoxic activity. The T cell-mediated cytotoxicity was restricted by the major histocompatibility complex (MHC), since most cytotoxicity could be blocked by HLA I antibodies. Additionally, we observed that optimum stimulation of cytotoxicity required effector cell-stimulator cell contact. These data indicate that depending on the tumor used for stimulation, different lymphocyte subsets may be generated in IL-2 cultures. These different approaches may be useful in both specific and nonspecific immunotherapy.