Interleukin 8 (IL-8) selectively inhibits immunoglobulin E production induced by IL-4 in human B cells.

The effect of interleukin 8 (IL-8) on IL-4-induced immunoglobulin E (IgE) production was studied. IL-4 induced IgE and IgG4 production by tonsillar mononuclear cells (MNC) without affecting IgM, IgG1, IgA, IgG2, or IgG3 production. IL-8 inhibited IL-4-induced IgE and IgG4 production, whereas it had no effect on IgM, IgG1, IgA, IgG2, and IgG3 production. The inhibitory effect by IL-8 was specific, since it was blocked by anti-IL-8 mAb, but not by control IgG1. Although interferon gamma (IFN-gamma) also inhibited IgE and IgG4 production by MNC stimulated with IL-4, the inhibitory effect of IL-8 was not mediated by IFN-gamma, since the IL-8-induced inhibition could not be blocked by anti-IFN-gamma. Furthermore, anti-IL-8 mAb had no effect on IFN-gamma-induced inhibition. Moreover, addition of IL-5 or IL-6 did not reverse IL-8-induced inhibition of IgE production. In contrast to these observations with MNC, IL-4 failed to induce IgE and IgG4 production by purified B cells. However, combined treatment of purified B cells cells with IL-4 and anti-CD40 antibody resulted in IgE but not IgG4 production. IL-8 inhibited this IgE production without affecting IgM, IgG1, IgG2, IgG3, IgG4, or IgA production, whereas IFN-gamma, IFN-alpha, or prostaglandin E2 (PGE2) failed to do so. These results indicate that IL-8 antagonizes IL-4-induced IgE production by directly affecting B cells through a specific mechanism that is different from IFN-gamma, IFN-alpha, or PGE2.

In vivo radiolocalization of antiosteogenic sarcoma monoclonal antibodies in osteogenic sarcoma xenografts.

Monoclonal antibodies Ost6 and Ost7 (mouse Immunoglobulin G1) to human osteogenic sarcoma were isolated from ascitic fluid and labeled with radioiodine. After injection into athymic nu/nu mice with s.c. xenografts of human osteogenic sarcoma, the uptake of radioactivity in tumors, visceral organs, and blood was determined. Five days after injection, Ost6 and Ost7 showed preferential accumulation in tumors (tumor:blood ratio, 4.3). Furthermore, with testicular and bladder tumors, both unreactive with Ost7, there was no localization of radiolabeled Ost7 in xenograft growths. When Ost7 was labeled with 131I, its accumulation into human osteogenic sarcoma could be clearly visualized by whole-body gamma-scintigraphy without computer-assisted data processing.

Detection of human osteosarcoma-associated antigen(s) by monoclonal antibodies.

Hybrid cell lines have been derived from a fusion between mouse myeloma cells, NS1, and spleen cells from mice immunized with freshly resected osteosarcoma cells from an untreated patient. Of the 276 hybrids obtained, five secreted antibodies which bound to osteosarcoma tissues but not to autologous skin fibroblasts. The antibodies from three of these five hybrids, OST6, OST7, and OST15, reacted with all of five osteosarcoma tissues and with one chondrosarcoma tissue but not with other malignant or benign tumors. Tests of various normal tissues were negative, except for weak binding to a subpopulation of chondrocytes in articular cartilage. The reciprocal binding inhibition test showed that OST6, OST7, and OST15 antibodies were directed against different antigenic determinants.

Adenosine inhibits divalent cation influx across human neutrophil plasma membrane via surface adenosine A2 receptors.

Adenosine and its analogues inhibited increases in divalent cation influx stimulated by platelet-activating factor (PAF) and formyl-methionyl-leucyl-phenylalanine (FMLP) in a dose-dependent fashion. This effect was antagonized by theophylline, an adenosine receptor antagonist. When extracellular adenosine was removed by adenosine deaminase, the effect of adenosine was completely abolished. Two adenosine analogues with different affinities for adenosine receptor subtypes, 5'-N-ethylcarboxamideadenosine (NECA) and L-N6-phenylisopropyladenosine (PIA), also inhibited divalent cation influx, NECA being more potent than PIA. These results suggest that adenosine and its analogues inhibit divalent cation influx across neutrophil plasma membranes via surface adenosine A2 receptors. Adenosine had little effect on the initial peaks of intracellular free calcium rises induced by chemoattractants, but it inhibited the subsequent rise in free calcium. Since calcium influx through the divalent cation channels or neutrophil plasma membranes is responsible for maintaining free calcium concentration following the initial peaks, we suggest that adenosine modulates neutrophil function by interfering with this calcium influx.

Changes in sensitivity to anticancer drugs during TPA-induced cellular differentiation in a human T-lymphoblastoid cell line (MOLT-4).

After four days of treatment with 10(-8) M TPA, differentiation of the human T-lymphoblastoid cell line MOLT-4 was induced along the T cell lineage, confirmed by a fall in adenosine deaminase and 5'-ectonucleotidase and a rise in purine nucleoside phosphorylase activity. TPA-treated cells became resistant to the cytotoxic effects of 1-beta-D-arabinofuranosylcytosine (Ara-C), 9-beta-D-arabinofuranosyladenine (Ara-A), and 2-chlorodeoxyadenosine. This was, in part, due to the altered cell cycle distribution (accumulation of cells in the G1 phase), since the toxicity of Ara-C and Ara-A is S phase specific. The diminished rate of Ara-C transport concomitant with Ara-CTP formation after TPA treatment is considered to be the biochemical basis for this acquired resistance.

Effect of high-dose methotrexate on plasma hypoxanthine and uridine levels in patients with acute leukemia or non-Hodgkin lymphoma in childhood.

We investigated the effect of high dose methotrexate (HDMTX) therapy on plasma hypoxanthine (Hx) and uridine (UR) concentrations in 12 children with acute lymphoblastic leukemia (ALL) or non-Hodgkin lymphoma (NHL). The initial plasma Hx level before the first administration of HDMTX (1 g/m2) was significantly higher in patients (25.5 +/- 17.5 microM) than that in healthy adult controls (4.0 +/- 1.4 microM). By 48 or 72 hours after the beginning of MTX infusion, the Hx concentration had decreased to 7.9 +/- 7.7 microM and 4.7 +/- 4.1 microM, respectively. This decrease of plasma Hx concentration after MTX infusion was also observed with the second course of HDMTX (3 g/m2) therapy. On the other hand, the plasma UR level did not change significantly. The in vitro treatment with 2 microM MTX of hypoxanthine-guanine phosphoribosyltransferase (HGPRT)-deficient mutant cells selected from HL-60 lowered the excretion of Hx into the culture medium. These data suggest a possible new explanation of the synergism of HDMTX and 6-thiopurines, for example 6-mercaptopurine and 6-thioguanine, since plasma Hx is considered to counteract 6-thiopurine toxicity through competition at the level of HGPRT.

Biochemical characterization of U937 cells resistant to L-asparaginase: the role of asparagine synthetase.

A human histiocytic lymphoma cell line, U937, is highly sensitive to L-asparaginase with an ID50 of about 0.0001 U/ml after 72 hr of culture. When U937 cells were made resistant to either L-asparaginase (1 U/ml) or asparagine deprivation, the activity of asparagine synthetase increased to 80- or 7-fold of the wild type, respectively. The phenotype of the resistance to L-asparaginase turned out to be stable under nonselective conditions for over several months. The hybrids between L-asparaginase sensitive (Molt4) and resistant (HL-60) cell lines revealed the latter phenotype in terms of L-asparaginase sensitivity and the activity of asparagine synthetase. Furthermore, U937 cells resistant to L-asparaginase could survive in glutamine-free media with 1.5-fold elevation of glutamine synthetase activity. These results altogether clarify the role of asparagine synthetase in L-asparaginase toxicity and have a good implication for the clinical use of L-asparaginase.

Involvement of CD11b/CD18 in enhanced neutrophil adhesion by Fc gamma receptor stimulation.

Neutrophils showed a rapid and transient adhesion to immunoglobulin G (IgG)-coated plates compared with their adhesion to bovine serum albumin (BSA)-coated plates: the adhesion reached a peak after 15 min of incubation and then gradually returned to almost the basal state in 60 min. The addition of monomeric IgG or anti-Fc gamma RII monoclonal antibody (mAb) (IV.3) suppressed the increase in adhesion, whereas anti-Fc gamma RIII mAb (3G8) was hardly effective, indicating that the interaction of Fc gamma R, especially Fc gamma RII, with coated IgG is involved in the process. Adhesion was also blocked by cytochalasin B, suggesting that functional actin filament structures are crucial. Protein kinase inhibitors, erbstatin and genistein, inhibited the adhesion in a dose-dependent manner. The adhesion was inhibited by anti-CD11b (M1/70) and anti-CD18 (MHM23, TS1/18) mAbs. Moreover, neutrophils from a patient with complete leukocyte adhesion deficiency syndrome did not show increased adhesion to IgG-coated plates. The adhesion of neutrophils to fibrinogen- and BSA-coated plates was also increased when Fc gamma R was stimulated in the fluid phase with soluble aggregated IgG, which was also inhibited by anti-CD11b mAb. Stimulation of neutrophil Fc gamma R with soluble aggregated IgG enhanced the expression of CD11b in concert with the enhanced adhesion. These data collectively suggest that stimulation via Fc gamma R evokes a tyrosine kinase-dependent and actin filament-dependent intracellular signal that enhances the specific and nonspecific adhesive activity of neutrophils, presumably through the activation of CD11b/CD18.

Developmental changes of GM-CSF gene inducibility in embryonal carcinoma cells.

Murine embryonal carcinoma (EC) P19 cells, a tissue culture model of early embryonic development, failed to produce cytokines, such as interleukin-3 (IL-3), IL-4, granulocytemacrophage colony stimulating factor (GM-CSF) and interferon-beta (IFN-beta) at the mRNA level. Differentiation induced by retinoic acid (RA) released this repression to produce some cytokines. GM-CSF and IFN-beta genes were expressed in response to PMA/A23187, poly(I):poly(C), IL-1 alpha, forskolin, or LPS stimulation in differentiated P19 cells, whereas IL-3 and IL-4 genes were not expressed. To elucidate the mechanism of the GM-CSF gene induction after differentiation, we transfected a series of 5' deletion mutants of the mouse GM-CSF promoter fused to the bacterial CAT gene. The 740-bp fragment of the 5'-flanking region mediated the positive response. Deletion analysis revealed that the 5' boundary region of the DNA element required for activation lies between positions -95 and -84 and the region upstream of position -95 appears inhibitory. These results indicate that the maturation of the transcriptional machinery after differentiation results in the activation of the GM-CSF gene.

Interferon and (2'-5')oligoadenylate enhance the expression of low affinity receptors for IgE (Fc epsilon R2/CD23) on the human monoblast cell line U937.

The regulation of low affinity IgE receptor (Fc epsilon R2/CD23) expression on the human monoblast cell line U937 was examined by an anti-Fc epsilon R2/CD23 monoclonal antibody (H107) and the cDNA probe for Fc epsilon R2/CD23. Alpha interferon (IFN-alpha) and its intracellular mediator, (2'-5')oligoadenylate (2, 5-A), induced Fc epsilon R2/CD23 expression on U937 with no significant increase of the Fc epsilon R2/CD23 mRNA. PMA and IFN-gamma increased both surface Fc epsilon R2/CD23 expression and the Fc epsilon R2/CD23 mRNA levels. IFN-alpha effectively induced 2, 5-A synthetase activity in U937 cells, whereas IFN-gamma induced little. The results suggest that the mechanisms of enhancement of Fc epsilon R2/CD23 expression on U937 cells by IFN-alpha and IFN-gamma are different and that 2, 5-A may play an important role in the Fc epsilon R2/CD23 expression on U937 cells induced by IFN-alpha.

Low burst-promoting activity (BPA) production by cord mononuclear cells is due to functional immaturity of monocytes.

Cord blood mononuclear cells produced lower burst-promoting activity (BPA) than adult peripheral blood mononuclear cells when stimulated with phytohemagglutinin (PHA). In order to examine the cellular basis of the low production of BPA by PHA-stimulated cord blood mononuclear cells in the context of the functional immaturity of T cells or monocytes, we studied BPA production by T cells or monocytes from cord blood and adult peripheral blood. Cord T cells produced as much BPA as adult T cells. Monocytes themselves did not produce significant BPA at the concentration used in this experiment (1 x 10(5)/ml). BPA production by adult T cells was significantly enhanced by the presence of autologous monocytes. BPA production by cord T cells was also enhanced by the presence of adult monocytes but not by that of cord monocytes. Cord monocytes did not enhance BPA production by adult T cells either. These results indicate that cord monocytes are primarily responsible for the low BPA production by PHA-stimulated cord mononuclear cells.

Mechanism of enhancement of neutrophil survival by granulocyte colony-stimulating factor and adenine.

An in vitro study was performed on the effect of recombinant human granulocyte colony-stimulating factor (rhG-CSF) and adenine on the survival of purified human neutrophils. The addition of rhG-CSF (1 to 100 ng/mL) or adenine (100 microM) enhanced the survival of neutrophils. The maintenance of O2- production in response to N-formylmethionyl-leucyl-phenyl-alanine (FMLP) suggested that these neutrophils were functionally alive. Neutrophils in cultures had shown two distinct biochemical changes during cell death: DNA fragmentation and depletion of cellular adenosine triphosphate (ATP) pools. Treatment with rhG-CSF (10 ng/mL) significantly delayed the appearance of DNA fragmentation as measured quantitatively by diphenylamine or by agarose gel electrophoresis. On the other hand, adenine had no effect on the generation of DNA fragmentation. The decrease of ATP during incubation for 12 hours was similar in control and rhG-CSF-treated neutrophils, while rhG-CSF prevented the further decline of ATP seen in control cultures. In contrast, adenine (100 microM) preserved ATP at levels significantly higher than in controls at both 12 hours and 24 hours of incubation. Our results suggest that rhG-CSF and adenine promote the survival of neutrophils in vitro by different mechanisms.

Role of protein kinase C in neutrophil survival enhanced by granulocyte colony-stimulating factor.

Recombinant human granulocyte colony-stimulating factor (rhG-CSF) (10 ng/mL) prolonged human neutrophil survival in culture by at least 36 hours. The addition of H-series compounds at concentrations that are considered to inhibit both protein kinase C (PKC) and cyclic adenylate monophosphate (cAMP)-dependent protein kinase (PKA) counteracted the effect of rhG-CSF. Concomitantly, the inhibition of nucleosomal DNA fragmentation by rhG-CSF was canceled. At lower concentrations, presumably capable of inhibiting only PKA, however, the compounds exhibited marginal effects on rhG-CSF-mediated increase of cell survival. These PKC inhibitors did not influence the priming effect of rhG-CSF significantly, as determined by O2- production stimulated by N-formyl-L-methionyl-L-leucyl phenylalanine (fMLP). Our results suggest that PKC plays an important role in the mechanism by which rhG-CSF promotes neutrophil survival, in striking contrast with the priming effect elicited by rhG-CSF.

Characterization of an eosinophilic leukemia cell differentiation factor (ELDF) produced by a human T cell leukemia cell line, HIL-3.

An adult T cell leukemia cell line, HIL-3, constitutively secretes a factor which induces the phenotypical and functional eosinophilic differentiation of a human eosinophilic leukemia cell line, EoL-1. Biochemical characteristics of the factor, termed eosinophilic leukemia cell differentiation factor (ELDF), were examined. ELDF was precipitated by 35 to 65% saturated ammonium sulfate from the culture supernatants of HIL-3 cells (HIL-3 sup). ELDF was eluted in a peak corresponding to a molecular weight of 30 to 40 kd by gel filtration. Isoelectric focusing in the Rotofor showed that ELDF had isoelectric points of 5 to 6. ELDF was trypsin-sensitive and stable to heat treatment at 65 degrees C for 30 minutes but labile at 80 degrees C or pH lower than 3. Half of the activity adhered to lentil-lectin but not to Con-A, indicating that a part of ELDF is glycoprotein with an N-linked carbohydrate moiety, which did not seem to be essential for ELDF activity. The biochemical characteristics of ELDF and blocking experiments using cytokine-specific neutralizing antibodies suggest that ELDF is different from gamma-interferon (IFN-gamma), tumor necrosis factor-alpha (TNF-alpha), granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-5 (IL-5) and interleukin-2 (IL-2), which may exist in HIL-3 sup, and that ELDF may be a previously unrecognized leukemia differentiation factor.

Dibutyryl cyclic AMP induces formyl peptide receptor expression and chemotactic responses in a human eosinophilic cell line, EoL-1.

We investigated actin polymerization and the increase of cytosolic free calcium concentration ([Ca2+]i) in a human eosinophilic leukemia cell line, EoL-1, in response to stimulation with chemotactic factors; we also investigated the effect of dibutyryl cyclic AMP (dbcAMP) on the responses. EoL-1 cells under normal culture conditions did not show either actin polymerization or an increase in [Ca2+]i when stimulated with formyl-methionyl-leucyl-phenylalanine (fMLP). Expression of formyl peptide receptors was not detectable on untreated EoL-1 cells, either. Dibutyryl cAMP induced the expression of formyl peptide receptors and the responsiveness to fMLP. The responsiveness of EoL-1 cells to the complement fragment C5a and platelet-activating factor (PAF) was also induced or enhanced by dbcAMP. The growth of EoL-1 cells was decreased and the proportion of cells in the G0/G1 phase of the cell cycle was increased by the treatment of EoL-1 cells with dbcAMP. The proportion of eosinophilic granule-containing cells and the content of eosinophil cationic protein (ECP) in EoL-1 cells was also increased when they were stimulated with dbcAMP for a longer period. The responsiveness of EoL-1 cells to fMLP, C5a, and PAF was shown to be regulated independently. EoL-1 cells and dbcAMP seem to be useful for examining chemotactic receptor expression and its signal transduction mechanisms.

Specific localization of tissue-type transglutaminase in adrenocorticotropin-producing cells of the human pituitary gland as demonstrated by immunohistochemistry.

The distribution of transglutaminases, the Ca2+-dependent protein cross-linking enzymes, in the human pituitary gland was investigated by immunohistological methods using specific antibodies. Tissue-type transglutaminase was specifically localized in ACTH-producing cells, and the cells producing GH, PRL, TSH, FSH, and LH contained no appreciable amount of the enzyme. No detectable plasma-type transglutaminase (coagulation factor XIII) was found in pituitary tissue. In a previous study we demonstrated that ACTH-producing cells contain very little Ca2+-dependent proteinases (calpain), but a remarkable amount of their inhibitor, calpastatin. Pituitary gland cells producing hormones other than ACTH contained calpains, but no detectable calpastatin. These results collectively suggest that intracellular substrate proteins in ACTH-producing cells are protected from Ca2+-dependent degradation and are substrates for Ca2+-dependent cross-linking catalyzed by the tissue-type transglutaminase. In other pituitary gland cells, conversely, the intracellular substrate proteins are more likely to undergo Ca2+-dependent degradation than cross-linking.

Hypothalamic-pituitary function in growth hormone-deficient patients with pituitary stalk transection.

We compared 1.5 T magnetic resonance (MR) image findings with hypothalamic-pituitary function in 11 patients with idiopathic pituitary dwarfism, each of whom had a history of perinatal abnormalities, and 1 patient with posttraumatic pituitary dwarfism. MR imaging revealed transection of the pituitary stalk in all patients and the formation of an ectopic posterior lobe at the proximal stump in 9 patients, none of whom had polydipsia or polyuria. Three patients without an ectopic posterior lobe had diabetes insipidus. The 5 patients who had small pituitary glands (less than 2 mm in height) had hypothyroidism with low serum TSH concentrations and low serum cortisol responses to insulin-induced hypoglycemia; however, 7 patients with normal-sized pituitary glands had normal thyroid and adrenal function. The serum GH response to GHRH did not correlate with the size of the pituitary gland. The patients with small pituitary glands had delayed or prolonged serum TSH responses to TRH and impaired serum LH and FSH responses to GnRH; 4 of the patients with normal-sized pituitary glands had normal serum TSH, LH, and FSH responses. Only 2 patients had high basal serum PRL concentrations. The endocrinological data suggest that reestablishment of the hypothalamo-hypophyseal portal circulation, which cannot be seen by MR imaging, may occur. We suggest that the primary cause of idiopathic pituitary dwarfism in many patients is injury to the pituitary stalk at birth.

The effect of cell synchronization on the efficiency of stable gene transfer by electroporation.

We synchronized thymidine kinase deficient mouse Ltk-cells by two different methods, hydroxyurea double-block treatment or aphidicolin single-block treatment and transformed them with the cloned herpes simplex virus thymidine kinase gene at various time intervals by the electroporation technique. Marked enhancement of stable transformation efficiency was observed at the time corresponding to the peak of G2/M phase. These results suggest that the G2/M phase is the most efficient period for stable gene transfer by electroporation.

The effect of adenosine on the fluorescence responses of chlorotetracycline-loaded human polymorphonuclear leukocytes.

Chlorotetracycline has been used in human polymorphonuclear leukocytes as a probe to investigate the state of membrane-bound calcium. We examined the effect of adenosine on the fluorescence responses of CTC-loaded PMNs stimulated with the synthetic chemotactic peptide, formyl-methionyl-leucyl- phenylalanine. Adenosine inhibited the decrease in CTC fluorescence in a dose-dependent fashion and its effect was reversed by theophylline, an adenosine receptor antagonist. Removal of extracellular adenosine by incubating PMNs with adenosine deaminase abolished the effect of adenosine. These data suggest that adenosine inhibits the release of membrane-bound calcium in PMNs that normally occurs in response to chemotactic stimuli, acting via PMN surface adenosine receptors.

Cell cycle related change of Ara-C transport in HL-60 cells after differentiation induction.

Using a promyelocytic leukemia cell line, HL-60, we studied the membrane transport of Ara-C before and after differentiation induced by retinoic acid (RA). In RA-treated cells, Ara-C transport was reduced and there was a concomitant increase of the ID50 values of Ara-C in comparison with the controls. By three different procedures to synchronize untreated cells, i.e. density arrest G1 phase enrichment, aphidicolin-induced S phase accumulation and the double isoleucine block method, we found that Ara-C transport was 30-50% higher in the S phase than in the G1 phase. Therefore, the observed decrease in Ara-C transport is, in part, due to the retarded growth accompanied by an accumulation of cells in the G1 phase after differentiation induction.