Functionally active Epstein-Barr virus-transformed follicular dendritic cell-like cell lines.

Follicular dendritic cells (FDC) are unique nonlymphoid cells found only in germinal centers. FDC can be distinguished from other accessory cells based on a characteristic set of cell surface markers. It is known that FDC are able to rescue germinal center B cells from apoptosis. To investigate the role of FDC in the process of selection and maturation of B cells during germinal center reactions, we tried to establish factor-independent immortalized FDC-like cell lines. Because freshly isolated FDC express the Epstein-Barr Virus (EBV) receptor CD21, we attempted EBV transformation on isolated FDC. After incubation of FDC-enriched cell populations with EBV, cell lines were obtained consisting of slowly duplicating very large cells. These cell lines have a fibroblast-like morphology but could be clearly distinguished from several human fibroblast cell lines by displaying a different phenotype including intercellular adhesion molecule 1, CD40, and CD75 expression. Detection of the EBV-encoded proteins latent membrane protein 1 and Epstein-Barr virus nuclear antigen 2 in our FDC-like cell lines implicated successful EBV transformation. FDC-like cells are able to bind nonautologous B cells and preserve the latter from apoptosis. The binding of B cells to FDC-like cells is dependent on adhesion via lymphocyte function-associated antigen 1/intercellular adhesion molecule 1 and closely resembles the pattern of emperipolesis as described by others. These data demonstrate that FDC can be successfully infected by EBV, and that the cell lines obtained share phenotypic and functional characteristics with freshly isolated FDC.

Diarrhoea in HIV-infected patients: no evidence of cytokine-mediated inflammation in jejunal mucosa.

OBJECTIVE: To determine whether a mucosal cytokine-mediated inflammatory response is involved in cryptosporidial or microsporidial diarrhoea, as well as in diarrhoea of unknown origin in HIV-infected patients. DESIGN: Prospective study. METHODS: Jejunal biopsies were obtained from HIV-infected patients with diarrhoea. Controls were HIV-infected and HIV-seronegative patients without diarrhoea. Two biopsies were homogenized immediately and two other biopsies were first cultured for 20 h. Cytokines [tumour necrosis factor (TNF), interleukin (IL)-1 beta, IL-6, IL-8, IL-10], soluble TNF receptors (sTNFR) p55 and p75, and soluble IL-2 receptor (sIL-2R) were assessed in the homogenates and in the supernatants by sandwich enzyme-linked immunosorbent or enzyme-linked binding assays. The cytokine receptors were also measured in serum. RESULTS: Six HIV-infected patients with cryptosporidiosis, six with microsporidiosis, seven with diarrhoea of unknown origin, seven without diarrhoea, and seven HIV-seronegative patients were eligible. Four patients were excluded because of the presence of other pathogens. No cytokines were detected in immediately homogenized jejunal tissue. Following culture, IL-6 and IL-8 levels were higher in HIV-infected patients with diarrhoea of unknown origin than in HIV-seronegative controls without diarrhoea, although this was not statistically significant. No differences in serum or post-culture supernatant sTNFR p55 and p75 levels existed between the HIV-infected patients with or without diarrhoea. sTNFR, IL-1 beta, IL-10 and the sIL-2R were only detected in low amounts or not at all, and were equally distributed among all patient groups. CONCLUSIONS: This study indicates that mucosal cytokine-mediated inflammatory responses do not play an important role in the pathogenesis of different types of diarrhoea in HIV-infected patients. These results do not support the use of immunomodulatory therapy in these patients.

An efficient procedure for the generation of human monoclonal antibodies based on activation of human B lymphocytes by a murine thymoma cell line.

A new, efficient procedure for the generation of human monoclonal antibodies has been developed. The procedure is based on the activation of human B cells in microwells by murine thymoma EL4B5 cells. This mode of B cell stimulation leads to proliferation of at least one per eight of human B cells and to a high rate of antibody production. Subsequently, supernatants of the microwells are screened by ELISA for the presence of antibody of the desired specificity and B cells from selected wells are hybridized by electroporation. To optimize the procedure, the kinetics of the B cell expansion induced by EL4B5 cells were analysed. Counting and phenotyping of cultured cells at different time points indicated that the peak of B cell expansion occurred at day 5 for tonsil B cells (16-fold increase) and at day 7 for peripheral blood B cells (20-fold increase). The B cells did not merely proliferate but also differentiated, as indicated by loss of CD20 expression and increase of CD38 expression. At the peak of B cell expansion, B cells could be hybridized efficiently with myeloma cells. The majority of the resultant hybridomas secreted human immunoglobulin. The efficiency of the procedure is exemplified by the generation of hybridomas secreting human IgG against Haemophilus influenzae from limited numbers of either human tonsil B lymphocytes or peripheral blood B lymphocytes.

Identification of structural differences between different forms of interleukin-2 (IL-2) using anti-(human recombinant) IL-2 monoclonal antibodies.

We have generated a panel of murine monoclonal antibodies (mAbs) directed against recombinant human interleukin-2 (rh IL-2). All mAbs have similar affinities (Kaff approximately equal to 10(-8) M) and are of the IgG1 isotype. Specificity of the mAbs has been established using an ELISA method and immunoprecipitation of native human interleukin-2 (nh IL-2) present in supernatants from PHA-stimulated human mononuclear cells (PBMNC). Four of the nine mAbs inhibit the rh IL-2-dependent proliferation of a murine cytotoxic T-lymphocyte line (CTLL-2). The estimated ID50, in the presence of 0.75 IU rh IL-2/ml, ranges from 2.0 micrograms/ml to 30 micrograms/ml final concentrations of the antibodies. Using different forms of IL-2 we found that the mAbs give different patterns of inhibition of the IL-2-dependent proliferation. One mAb (AMCIB 27) is able to discriminate between rh and nh IL-2. Findings with another mAb (AMCIB 24) indicate that possible functional differences between human IL-2 and recombinant murine (rm) IL-2 are caused by differences in the active sites. The results of this investigation show that it is possible to obtain mAbs, after immunization with rh IL-2, that differ in their ability to inhibit the biological action of different forms of IL-2.

Improvement of EBV transformation and cloning efficiency of human B cells using culture supernatants from lymphoblastoid cell lines.

Exposure of human B cells to Epstein-Barr virus (EBV) usually results in low frequencies of transformed cells. The transformed cells can be cloned poorly by limiting dilution, even when feeder cells are used. In recent years it has become clear that growth and antibody production of EBV-transformed cells are influenced by auto- and paracrine growth factors. Therefore, supernatants from the lymphoblastoid B cell lines JY and Raji were used as a source of growth factors to investigate their effect during EBV transformation of human B cells and consequently on cloning by limiting dilution of these transformed cells. Initial experiments to clone three established EBV-transformed B cell lines showed a strong increase in outgrowth of the number of cells in the presence of the supernatant (range: 1 per 2-8 of the originally plated cells) as compared to cells cultured without the supernatant (range: 1 per 17-100 of the originally plated cells). Transformation efficiencies of freshly isolated tonsil B cells were not influenced by the supernatant and were generally less than 1%. In contrast, transformation efficiency was increased up to 9.4% if B cells were both transformed and cultured in the presence of the supernatant. Cloning efficiencies increased if the cells used were transformed in the presence of the supernatant. Best results were seen when the supernatant was present during transformation and cloning of the cells. The presence of the supernatant during transformation and/or cloning of the B cells dramatically enhanced the number of B cells secreting IgG. Cloning of two established tonsil B cell lines resulted in a large number of B cell clones.(ABSTRACT TRUNCATED AT 250 WORDS)

Direct evidence that human follicular dendritic cells (FDC) rescue germinal centre B cells from death by apoptosis.

It is supposed that FDC have a pivotal role in the rescue of germinal centre (GC) B lymphocytes from apoptosis. However, formal proof for this hypothesis has not as yet been presented. In the present study FDC and GC B cells were isolated from human tonsils and cultured. When brought into culture FDC and B cells rapidly formed spherical clusters. T cells were not observed inside these clusters. At different time points cultures of FDC and B cells were supravitally stained with Hoechst 33258 or acridine orange and examined by direct observation using fluorescence microscopy. Viable B cells appeared to be profoundly restricted to clusters, whereas cells not taking part in clusters all had an apoptotic appearance. The formation of clusters could be prevented by addition of MoAbs against CD11a (LFA-1 alpha) or CD49d (VLA-4 alpha), resulting in an apoptotic appearance of virtually all B lymphocytes. The present data demonstrate that a physical interaction between FDC and germinal centre B lymphocytes is able to rescue the latter from apoptotic cell death.

Contrasting effects of interleukin 4 (IL-4) on the in vitro proliferation of human B cells is due to the presence of B cell subsets in different activation states.

Human tonsillar B lymphocytes proliferate in vitro after activation with immobilized anti-IgM antibodies in combination with interleukin 2 (IL-2) or interleukin 4 (IL-4). In addition to its proliferative effects, IL-4 is also able to inhibit B cell proliferation. An explanation for these different actions of the same protein is not as yet available. The susceptibility to the inhibitory action of IL-4 emerges at late time points after in vitro activation of the B cells, whereas IL-4 shows its growth promoting action during early stages of culture, indicating that activated IL-2 responsive B cells are the targets for IL-4 dependent inhibition. Freshly isolated tonsillar B lymphocytes also appear to be susceptible to IL-4 dependent suppression. Isotonic Percoll gradient centrifugation, that has been used by several groups as a standard procedure to isolate high density, resting used by several groups as a standard procedure to isolate high density, resting cells from tonsil B lymphocytes, does not result in separation of B cells with different in vitro proliferative response to IL-2 and IL-4. However, partial separation of cells with "resting" and "activated" phenotypes can be achieved by centrifugation in slightly hypotonic Percoll gradients. High-density fractions contain cells with a "resting" phenotype which after in vitro activation proliferate with IL-4 but not with IL-2. "Activated" B cells accumulate in the low-density fractions as concluded from the increased expression of transferrin receptors and HLA-DR molecules on these cells. However, these cells no longer proliferate in vitro with immobilized anti-IgM antibodies, possibly due to damage caused by the hypotonic treatment during centrifugation. Our data indicate that resting B cells when activated in vitro with immobilized anti-IgM antibodies proliferate primarily with IL-4 but not with IL-2. In contrast, about 15 hours after activation the cells become responsive to IL-2. At that stage, IL-4 becomes an inhibitory factor.

Release of tumor necrosis factor alpha and interleukin 6 during antibiotic killing of Escherichia coli in whole blood: influence of antibiotic class, antibiotic concentration, and presence of septic serum.

The concentration and accessibility of endotoxin can increase following antibiotic killing of gram-negative bacteria. There are indications that antibiotics may differ in this respect. We measured endotoxin levels in RPMI 1640 and tumor necrosis factor alpha (TNF-alpha) and interleukin-6 production in whole blood ex vivo after exposure of log-phase Escherichia coli to antibiotics belonging to different classes, in a final concentration of 0.5, 5, or 50 times the MIC. After 4 h of incubation at 50 times the MIC, ceftazidime and ciprofloxacin treatment resulted in levels of endotoxin, TNF-alpha, and interleukin-6 significantly higher than those of imipenem and gentamicin (P < 0.001). Similar differences in cytokine induction were measured after 8 h of incubation. At 0.5 times the MIC, the differences between the antibiotics in measured endotoxin and cytokine levels were small, with levels comparable to the levels in untreated cultures. Polymyxin B and, to a lesser degree, recombinant bactericidal/permeability-increasing protein 21 (rBPI-21) were found to be potent inhibitors of TNF-alpha release, supporting the concept that the differences between the antibiotics in cytokine production were indeed due to differences in amounts of biologically active endotoxin. The presence of serum from patients suffering from untreated sepsis decreased TNF-alpha production significantly, in a concentration-dependent manner.

Inhibition of endotoxin-induced cytokine release and neutrophil activation in humans by use of recombinant bactericidal/permeability-increasing protein.

To investigate the effects of a recombinant endotoxin-binding protein, bactericidal/permeability-increasing protein (rBPI23), on cytokine release and neutrophil activation in endotoxemia in humans, 8 volunteers were challenged twice with endotoxin and concurrently received either rBPI23 or placebo in a randomized, placebo controlled, double-blind crossover study, rBPI23 treatment significantly lowered circulating endotoxin levels (P = .02) and resulted in a significant reduction in the release of tumor necrosis factor (TNF), soluble TNF receptors p55 and p75, interleukin (IL)-6, IL-8 (P < .01 for each), and IL-10 levels (P = .02) but did not prevent the endotoxin-induced rise in body temperature. The early endotoxin-induced leukopenia was blunted (P = .08), and neutrophil degranulation, as measured by circulating levels of elastase/alpha 1-antitrypsin complexes (P = .03) and lactoferrin (P < .01), was largely prevented by rBPI23. The results of this study indicate that rBPI23 is capable of neutralizing many of the biologic effects of endotoxin in humans.