Search

Individuals with mannose-binding lectin (MBL)-deficiency are at an increased risk from infections with mannose-bearing microorganisms. We have investigated two quantitative research assays for measuring MBL protein in serum for routine diagnosis. The evaluation of 817 serum samples with a nephelometric assay revealed two deficiencies, a number far below the postulated 5-10% of the population. Reevaluation of 102 serum samples with an MBL-ELISA detected low levels in 27 cases (26.4%) and clear deficiencies in 21 samples (20.4%). In our hands, the MBL-ELISA permitted the detection of decreased levels of MBL in serum, as occurs in individuals with homozygous or heterozygous MBL gene mutations; in contrast, the nephelometric assay appeared to be unsuitable for the detection of MBL deficiencies. We support the routine measurement of MBL in serum, especially in children with frequent infections.

The expression of galectin-3 (formerly known as IgE-binding protein or Mac-2) in rat bone marrow (BM) was investigated by FACS, immunocytochemical and immunoblot analysis. The functional significance of rat recombinant galectin-3 on mouse recombinant granulocyte-macrophage colony-stimulating factor (GM-CSF)-driven proliferation of macrophage progenitors and gene transcription was further examined. Immunocytochemical analysis of in situ BM sections demonstrated galectin-3 in myelopoietic cells and surrounding stroma, whereas erythropoietic and lymphopoietic environments essentially lacked galectin-3 expression. FACS analysis demonstrated that incubation of freshly isolated BMC with lactose, a competing ligand for galectin-3 binding to glycoconjugates, decreased binding of antigalectin antibodies to cells primarily expressing the myeloid antigen recognized by mAb His-54. Similarly, lectin-mediated binding of exogenous galectin-3 to myeloid lineage cells was also demonstrated. Immunoblot analysis of BM eluates demonstrated galectin-3 both in the extracellular matrix and in a lactose elutable form, bound to the surface of BMC. [3H]Thymidine incorporation studies on BMC cultured in the presence of galectin-3 demonstrated suppression of GM-CSF-induced proliferation by galectin-3. In addition, differential display analysis of immediate early gene expression in BMC cultured in the presence of galectin-3 revealed a 76.2% inhibition of GM-CSF-induced gene transcription by galectin-3 assessed by the number of PCR-fragments generated. Our data suggest a role for galectin-3 in the organization of myelopoietic compartments in rat BM and regulation of the action of growth factors on myelopoietic precursor cells.

Mature macrophages (M phi) differ from other rat leukocytes by their ability to bind soybean agglutinin (SBA). In this study we identify the SBA-binding structure on rat bone marrow-derived M phi (BMDM phi). Precipitation of iodinated membrane proteins from rat bone marrow cells (BMC) and BMDM phi with SBA revealed a major glycoprotein of Mr 160 kDa on BMDM phi but not on BMC. In addition minor bands migrating at 70 and 26 kDa were seen. Stimulation of BMDM phi with 100 nM SBA induced a decrease in surface density of Thy1.1 (MRC OX7) and His54 and an increase in the expression of MRC OX6 (RT1.B/I-A), MRC OX17 (RT1.D/I-E), MRC OX41 (gp 110/120), MRC OX42 (CD11b/c), Macl (CD11b/CR3) and Mac2 (galectin-3/IgE binding protein) antigen. Expression of other M phi differentiation antigens recognized by mAb MRC OX43 (M phi, endothelial cells) and ED9 (M phi/CD14 like) were not significantly altered. BMDM phi derived from cultures with M phi colony-stimulating factor (M-CSF) and SBA showed increased oxidative burst and phagocytic activity compared to cells cultured with M-CSF alone. Our data suggest that binding of a 160-kDa membrane glycoprotein on M phi by N-acetylgalactosamine-specific lectins stimulates M phi differentiation and activation.

Regulation of major histocompatibility complex (MHC) class II antigen expression by cytokines has been suggested to play a major role in the initiation and propagation of immune and autoimmune processes. The analysis of class II gene regulation benefits greatly from the existence of mutants with defects in regulatory factors. We report the establishment of a subclone of the human monocytic cell line U937, termed C119/9, with unusual cytokine regulation of MHC class II expression. In contrast to the parental U937 cell line, only tumor necrosis factor (TNF)-alpha, and not interferon (IFN)-gamma induces the expression of MHC class II antigens on C119/9 cells, and paradoxically, this induction was inhibited almost completely by IFN-gamma. The HLA-DR induction is controlled at the transcriptional level by the first 150 bp of the class II promoter which contains all the class II consensus elements. Both HLA-DR and -DQ mRNA are induced by TNF-alpha treatment, and both are diminished upon co-treatment with TNF-alpha and IFN-gamma. This antagonism between TNF-alpha and IFN-gamma seem to be restricted to MHC class II genes. This subline of U937 cells may be useful in further studies of MHC class II regulation.

Previously we have reported the differentiation-dependent expression of a soybean agglutinin (SBA)-binding structure on rat bone marrow cells (BMCs) during their differentiation into macrophages (møs). In the present study we tried to analyze the functional role of the SBA-binding structure in BMC proliferation and differentiation. Addition of SBA to BMC cultures driven into mø differentiation by recombinant murine granulocyte-macrophage colony-stimulating factor (rmGM-CSF), resulted in a two- to threefold increased proliferation rate compared with rmGM-CSF alone. However, the number of colonies in methyl cellulose was not increased by SBA. The effect of SBA was dose dependent (from 4 to 83 pM SBA), with a maximum effect at 83 pM. Experiments to detect a possible synergistic effect of additional cytokines produced by BMC after SBA treatment were inconclusive. The enhancing effect of SBA was also seen when high-density cells, which did not proliferate in response to rmGM-CSF (mainly granulocytes), were removed. Therefore, SBA may increase the CSF reactivity of re-sponsive mø progenitor cells directly by binding to N-acetylgalactosamine residues on their surface. J. Leukoc. Biol. 55: 127–132; 1994.

T-T cell interactions are known to be of importance in the generation of immune responses and have been postulated to play a role in autoimmunity. Antiergotypic T lymphocytes are cells, which react with other activated T cells, which they may consecutively neutralize. Antiergotypic T cells have been described to play a role in animal models of autoimmunity and have been found in patients with rheumatoid arthritis. It was the aim of the present study to analyse, whether antiergotypic T cells were also present in the blood of patients with endocrine autoimmunity and to characterize them in vitro. We found that peripheral blood mononuclear cells from patients with Graves' disease, Hashimoto's thyroiditis and Diabetes mellitus show a pronounced proliferative response, when stimulated with autologous T cell lines. This response is dependent on the state of activation of the stimulator cell population, but is independent of its phenotype and is not MHC restricted. In contrast to normal control cells PBMC from patients with endocrine autoimmune disorders respond better to autologous T lymphocytes, than to allogeneic T cells and to autologous E- monocytic cells. T cells responsive to activated autologous T cell lines can also be expanded from PBMC. They may be CD4+ or CD8+ and recognize autologous T cells in the presence of autologous PBMC as feeder cells. These results suggest the presence of a T-T cell network in patients with endocrine autoimmune disease, which may have an important regulatory role in the disease process.

Alveolar macrophages (Amphi;) represent an immunologically distinct sub-population within the reticuloendothelial system. Phagocytosis and possibly antigen presentation by Amphi; are essential components of specific and innate primary immune defence processes against inhaled material. The mphi;-restricted sheep erythrocyte receptor sialoadhesin (Sn) is a member of the immunglobulin superfamily and binds specifically to sialic acid-containing structures such as selectins and was originally identified as the sheep erythrocyte receptor (SER) responsible for sialic acid-dependent binding of native sheep erythrocytes (SE) to resident murine bone marrow macrophages in rosetting assays. Sn expression has been demonstrated on murine and rat mphi; in lymphatic organs and is recognised by the monoclonal antibody (mAb) ED3 in the rat. In addition, sialic acid-dependent receptor (SAR) activities that mediate rosette formation of alveolar, peritoneal, splenic and bone marrow-resident rat mphi; with SE pretreated with gangliosides and SER-like activities between native SE and trypsinised Amphi;, have been described. The binding activities of both SAR and Sn show similar characteristics suggesting that these molecules are closely structurally related or identical. To clarify the relationship between Sn, SAR and SER-like activities, the binding of mAb ED3 to isolated rat Amphi; was investigated by flow cytometry and rosetting assays. It is demonstrated that rat Amphi; express Sn and evidence is provided that SAR and SER-like activities are mediated by Sn.

Hematopoiesis is viewed as a differentiating system emanating from a pluripotent hematopoietic stem cell capable of both self-renewal and differentiation. By identifying and characterizing a novel and highly specific in vitro mitogenic response to the N-acetyl glucosamyl/sialic acid specific, stem cell-binding lectin wheat germ agglutinin (WGA), we demonstrate the existance of a rare (0.1%), plastic adherent precursor in rat bone marrow capable of proliferation (two to seven divisions) in response to WGA. Stimulated cells possess a lineage (lin)low/- immunophenotype and immature blastoid morphology (WGA blasts). A subsequent proliferative response to stem cell factor (SCF), the ligand for the proto-oncogene receptor tyrosine kinase c-kit, is characterized by an initial maturation in immunophenotype and subsequent self-renewal of cells (SCF blasts) without differentiation for at least 50 generations. Although granulocyte colony-stimulating factor (G-CSF), interleukin (IL) -6, IL-7, and IL-11 synergize with SCF to increase blast colony formation, cytokines such as granulocyte-macrophage CSF or IL-3 are without significant effect. At all time points in culture, however, cells rapidly differentiate to mature neutrophils with dexamethasone or to mainly monocytes/macrophages in the presence of 1alpha,25-dihydroxyvitamin D3, characterized by cell morphology and cytochemistry. Removal of SCF during blast maturation, self-renewal, or induction of differentiation phases results in apoptotic cell death. Data indicate a pivotal role for SCF/c-kit interaction during antigenic maturation, self-renewal, and apoptotic protection of these lineage-restricted progenitors during non-CSF-mediated induction of differentiation. This approach provides a source of many normal, proliferating myelomonocytic precursor cells, and introduces possible clinical applications of ex vivo expanded myeloid stem cells.

In a previous report, we demonstrated intense lymphocytic infiltration of all benign prostatic hypertrophy (BPH) tissues analyzed in conjunction with HLA-DR expression on normally MHC-class-II-negative prostate epithelial cells. The composition of these infiltrates (70 to 80% CD3+ T-cells, but no granulocytes) resembles the situation seen in immune responses against altered self or self rather than against foreign antigens (infection). In the present study, phenotypic and functional immunoassays were used in order to investigate whether T-cells in BPH are indeed activated, and whether this activation is systemic or restricted locally to the prostate. Analysis of T-cell activation marker expression and proliferation requirements provided substantial evidence that these infiltrating lymphocytes, in contrast to their peripheral counterparts, are chronically activated. Since local accumulation of activated lymphocytes can cause tissue destruction, high concentrations of cytokines, and consequently tissue rebuilding, this process might contribute to the pathogenesis of BPH.

Suramin, a polyanionic compound, which has been in clinical use for the treatment of African trypanosomiasis for several decades, has recently been introduced in clinical oncology. Its effects on the immune system seemed therefore of interest. In the present study we tried to elucidate in vitro how suramin affected different functions of human peripheral blood mononuclear cells (PBMC). Suramin suppressed the proliferation of PBMC in response to various stimuli, including OKT3, phytohemagglutinin (PHA), phorbolmyristate-acetate (PMA) and ionomycin, purified protein derivate of Mycobacterium tuberculosis (PPD) and antibodies against CD2. It also inhibited the binding of monoclonal antibodies to T cell surface antigens. This effect was not dependent on the isotype of the antibody, but seemed to be highly epitope-specific. Among a panel of antibodies against one antigen, only a few were affected by the compound. Whereas the binding of Leu3a and OKT3 was for instance fully suppressed by suramin, OKT4 and Leu4 binding was only slightly affected. Suramin also decreased the expression of T cell surface molecules such as CD2, CD25 and CD4 in preactivated PBMC and had pronounced effects on cytokine production. Interestingly the compound had adverse regulatory effects on different cytokines. Whereas the secretion of interferon-gamma was completely suppressed by suramin, interleukin-2 (IL-2) and IL-4 production was stimulated. These results demonstrate that suramin affects T cell function in multiple different ways. This will have to be considered, when suramin is used in the treatment of cancer patients.

Rat bone marrow cells (BMC) cultured in the presence of murine colony-stimulating factor type 1 (CSF-1) differentiate within 7 days into a cell population containing 96–100% macrophages (Mφ). In this study, binding of 10 different fluorescein isothiocyanate (FITC)-conjugated lectins to cultured BMC at various stages of differentiation into Mφ was investigated. Only soybean agglutinin (SBA) showed a binding pattern that was significantly correlated to Mφ differentiation. Nearly all adherent (A) cells bound SBA. Binding of SBA to nonadherent (NA) cells increased from 20% on day 0 to 80% on day 8. NA cells were separated by means of a fluorescence-activated cell sorter into SBA-, SBA±, and SBA+ fractions. On day 0 and day 2, Mφ, blast-like cells, eosinophils, and some neutrophils were found in the SBA+ population. From day 4 onwards, the SBA+ fraction contained almost exclusively Mφ. Neutrophils and some blasts were found in the SBA-population on days 0 and 2. In the SBA± fraction, mainly blasts and lymphocytes were identified. With increasing time of culture, Mφ or Mφ precursors prevailed also in the SBA-/± cell population. Cells forming colonies in soft agar in the presence of CSF-1 were highly enriched in the SBA± fraction. A large number of cells in S-G2/M phases but few colony-forming cells were found in the SBA+ population. Our data suggest that SBA is a useful additional tool to define late stages of Mφ differentiation.

Sheep erythrocytes were coated with bencylpenicilloyl-(BPO)groups. Different incubation periods resulted in erythrocyte preparations with different hapten density. Complement dependent lysis induced by IgM or IgG antibodies was studied with the cell preparations. The calculation of hapten density on the erythrocyte surface was not possible by direct measurement of coupled radioactive BPO, since more than 90% of radioactive material was found in the soluble supernatant after osmotic cell lysis and less than 10% was fixed to the cellular membrane. Measurement of membrane bound immunologically relevant BPO-groups was achieved, therefore, by comparison of the inhibitory capacity of the test cells with that of a standard cell preparation. The latter consisted of tannic acid treated erythrocytes coated with protein complexed radioactive BPO. Surface hapten density of the different target cell preparations varied between 1.9 × 10 5 and 4.8 × 10 5 BPO-groups per cell depending on the time of incubation. Complement dependent antibody mediated cell lysis was significantly reduced by reduction of haptenic sites per target cell, IgG induced lysis being much more affected than hemolysis induced by IgM antibodies. Statistical calculations led to the conclusion that 18,000 protein islets per cell bearing 4 or more BPO-groups are not sufficient for hemolysis induced by IgG antibodies. 48,000 protein islets with this hapten density are necessary for «optimal« sensitization. IgG antibodies must be apparently bound to the cell surface in bivalent form.

Bone marrow cells (BMC) flushed from femora of Lewis rats were cultured in Dulbecco's modification of Eagle's medium supplemented with mouse L929 cell supernatant as a source of colony-stimulating factor (CSF). Differentiation of macrophage progenitor cells into macrophages (Mø) and expression of various markers were kinetically assessed. The proportion of Mø increases from approximately 4% in freshly isolated BMC to 100% after 7-8 days of cell culture. These cells, termed bone marrow cell-derived macrophages (BMDMø), adhere to and spread on plastic surface; exhibit Mø morphology; stain intensely for nonspecific esterase; are able to phagocytose latex particles, IgG-sensitized erythrocytes, and C3-coated red cells; and express receptors for IgG and C3. A subpopulation of BMDMø expresses MHC class II antigens as demonstrated by immunofluorescence using MRC OX6 and MRC OX17 monoclonal antibodies which recognize antigens coded in the I-A or I-E subregion of the MHC, respectively. Collectively, our results show that supernatant from mouse L929 cells supports and is continuously required for proliferation and differentiation of rat BMC into typical Mø, and suggest that mouse CSF cross-reacts with the putative receptor on rat Mø.

Two hybridoma antibodies (VEP8 and VEP9) raised against the promyelomonocytic leukemia cell line HL60 have previously been shown to distinguish human granulocytes and monocytes from other cells of the peripheral blood. We report here that both antibodies recognize the carbohydrate structure 3-fucosyl-N-acetyllactosamine with the following sequence: (formula; see text) This structure is the same as that recognized by a hybridoma antibody against mouse teratocarcinoma cells (anti-SSEA-1) which recognizes an early embryonic antigen in the mouse. Until recently this carbohydrate structure was considered to be rare among glycoproteins and glycosphingolipids. However, there is a growing list of human and animal glycoproteins in which this sequence has been detected by chemical and immunochemical methods. In this article we survey this information and discuss how this and other carbohydrate structures behave as differentiation- or tumor-associated antigens.

Human alveolar macrophages (AM) from thirteen patients, who were suffering from various lung diseases were harvested by bronchoalveolar lavage. Peripheral blood monocytes from eight healthy donors were isolated by Ficoll-Hypaque gradient centrifugation and adherence to plastic surface. To detect the VEP13 antigen (CD16) on these cells, a rosette assay employing ox erythrocytes coated by the CrCl 3 method with purified VEP13 monoclonal antibody (E o -VEP13) was used. A mean of 31.3 % of freshly isolated AM and 3.9 % of blood monocytes formed E o -VEP13 rosettes. Monocytes cultured for 3 or 6 days in the presence of a supernatant from mouse L929 cells, which had been shown previously to improve long-term viability of human monocytes in culture, showed 12.5 % and 25.3 % E o -VEP13 rosettes, respectively. No significant increase in VEP13 antigen expression was noted by culturing monocytes without L929 cell supernatant. The factor in L929 supernatant that induces VEP13 antigen expression has not been identified. Tunicamycin at 10 υg/ml inhibited significantly VEP13 antigen expression on monocytes. In contrast, IgG rosette formation was not reduced by tunicamycin. Our data show that subpopulations of native human AM and peripheral blood monocytes cultured in presence of a supernatant of L929 fibroblasts containing mainly murine CSF may express the CD16 antigen, which is normally found on large granular lymphocytes (LGL). Suppression by tunicamycin indicates that Fc receptor glycosylation takes place during a later differentiation step of mononuclear phagocytes.

The hapten density on the surface of sheep red cells, coated with benzylpenicilloic acid (BPO) was determined. This was achieved by comparison of the inhibitory capacity of the test cells with that of a standard cell preparation of known antigen density. IgG-induced hemolysis was much more affected by the recution of haptenic sites/cell than IgM-induced hemolysis. It was concluded that more than 18,000 protein islets on the sheep red cell surface bearing at least four haptenic sites were necessary for IgG-induced lysis. Target cells with 190,000 BPO groups/cell could be protected by anti-BPO-IgG against the lytic effect of simultaneously added IgM. Using target cells with 240,000 haptenic groups/cell or more a synergistic hemolytic action of IgG- and IgM-anti-BPO antibodies could be observed.

In this study the effect of alveolar (AC) or peritoneal (PC) lavage cells from Lewis rats on the mitogen induced proliferation of syngeneic spleen cells was investigated. When total spleen cells were stimulated with PHA (5 μg/ml) or ConA (6 μg/ml), the addition of AC or PC and of their adherent (A) or non-adherent (NA) fractions had dose dependent suppressive effects. After passing the spleen cells through a Sephadex G-10 column, the eluted cells responded poorly to the mitogens. Addition of low numbers (2x10 3 = 10%) AC or NA-AC improved this response, whereas higher concentrations (5 or 10 %) of these cells or any concentration of A-AC, PC, A-PC or NA-PC showed either none or a suppressive effect. The cells which strongly suppress mitogen induced lymphocyte proliferation have macrophage morphology, are adherent, esterase positive, bear Fc receptor for IgG and bind ganglioside coated erythrocytes. NA-AC were further separated into Ia-depleted (Ia - ) and Ia-enriched (Ia + ) population. The Ia + fraction contained 80 to 90% Ia bearing cells, supported the mitogen induced stimulation of Sephadex G-10 depleted spleen cells and was ineffective in suppression. Interestingly, these cells did not bind ganglioside treated sheep erythrocytes. On the other hand, the Ia-negative NA-AC fraction and particularly the A-AC population strongly expressed the «ganglioside receptor» as well as typical macrophage markers like Fc receptors, latex-phagocytosis and non-specific esterase. Therefore, the «ganglioside receptor» may be a useful marker to define suppressor macrophages.

Peripheral blood lymphocytes from 260 “apparently healthy” males and females, aged 20–97, have been investigated for age-related changes in a number of immunological parameters (percent and number of lymphocytes, OKT 4 + and OKT 8 + cells; OKT 4 8 ratio; intensity of fluorescence of OKT 4 and OKT 8 stained cells; membrane fluidity). Data were then reassessed after exclusion of people who did not conform to the SENIEUR PROTOCOL admission criteria of EURAGE (European Economic Community's Concerted Action Program on Aging) in order to investigate whether the differences observed were attributable to underlying disease. A slight decrease in the number and percentage of lymphocytes, OKT 4 + and especially OKT 8 + cells was found. Intensity of fluorescence of OKT 4 and OKT 8 stained cells from the elderly was reduced and may explain the lower percentages. Membrane fluidity was decreased in old persons (over the age of 75); the free cholesterol/phospholipid molar ratio in the serum increased up to the age of 75 and then declined, and it did not correlate with decreased membrane fluidity. Exclusion of the 20–60% of the study groups who were not eligible for admission according to the SENIEUR PROTOCOL criteria did not affect these results. The feasability and applicability of the PROTOCOL is discussed.

Rat alveolar and peritoneal macrophages were tested in a rosetting assay for their capacity to bind rabbit IgG antibody sensitized sheep erythrocytes (EA) before and after various times of incubation with proteolytic enzymes. With most enzymes, a biphasic effect on EA rosette formation was observed: an initial enhancement was followed by a reduction of the number of rosette-forming cells (RFC). The extent and rate of these changes depended on the type and concentration of the enzyme and on the amount of IgG antibodies used to sensitize the sheep erythrocytes. At low enzyme concentrations and with lowly sensitized EA, the phase of RFC enhancement was more prominent and prolonged. With an increase of enzyme concentration, the rate of reduction of RFC was increased. With regard to different enzymes, the highest rate of receptor degradation was found with pronase, followed by trypsin, whereas α-chymotrypsin had little receptor-degrading activity. A slight enhancement not followed by a loss of EA-rosetting activity was induced by granulocyte elastase at the concentrations studied. The results may give an explanation to contradicting reports in the literature, in which a single dose or just a few doses of proteolytic enzymes were employed and no kinetic studies were performed. In addition, EA-rosette-inhibiting material was found in supernatants of macrophage cultures, suggesting that receptor-like material was shed during incubation in serum-free medium. This material lost its EA-rosette-inhibiting capacity after proteinase treatment.

Anti-macrophage sera (AMS) produced in rabbits against rat alveolar (AM) and peritonealmacrophages (PM) were tested for their selective capacity to kill 51 Cr-labelled AM, PM and thymus lymphocytes (TL) of rats. All non-absorbed AMS were able to kill both types of macrophages as well as TL in thepresence of complement. However, the cytotoxic titre (CT) and the percent of specific 51 Cr-release was always higher on homologous macrophages (as used for immunization) than on heterologous macrophages (of the other population). On both types of macrophages it was stronger than on TL. After absorptions of AMS with insolubilized rat plasma and fetal calf serum (FCS), with raterythrocytes and with non-adherent cells from rat kidney, thymus, spleen and bone marrow (PEKL-absorptions) all AMS reacted only with macrophages and not with TL. When these antisera were absorbed repeatedly with limited amounts of AM or PM afunctional specificity was observed for the cell type which was not used for absorption. Further absorptions of these AMS with the same cells removed completely all cytotoxic activity for both macrophage populations. It is supposed that AM and PM have at least two different antigenic determinants present ontheir surface but their density is different in the two cell populations. Antigens specific for one cell type could not be detected in complement dependent cytotoxicity.

The binding specificity of rat alveolar macrophages (AM phi) for sheep erythrocytes (E) coated with gangliosides GM1, GM2, GM3, GD1a, GD1b or GT1b was analyzed in a rosette assay by studying the inhibitory effect of gangliosides, various carbohydrates, IgG, C3b-like C3, and fibronectin in this assay. The uptake of gangliosides by E was calculated from radioactivity measurements using 3H-labeled gangliosides. The different gangliosides were taken up by E at 37 degrees C to a similar extent. Uptake of 3H-labeled GM2 correlated linearly to its concn in the incubation medium. Erythrocytes pretreated with the same molar concn of GM2, GD1a, GD1b or GT1b were bound to AM phi to the same degree reaching a maximum of about 90% rosette forming cells. A mean of 17.8% AM phi-bound GM3-coated E. Treatment of E with asialo-GM2 (GA2) or GM1 did not induce significant rosette formation. A dose-dependent inhibition of rosette formation was observed when AM phi were preincubated at 0 degree C with GM2, GM3, GD1a, GD1b or GT1b, but not with GM1 or GA2 Of the tested carbohydrates, sialyl-lactose had a strong inhibitory effect, while lactose was completely ineffective. N-acetyl-neuraminic acid, N-glycolyl-neuraminic acid and N-acetyl-galactosamine were slightly inhibitory. A series of other carbohydrates including highly negatively charged compounds, as well as fibronectin, IgG or C3b-like C3 did not show significant inhibition. Our data indicate the expression of a receptor on rat AM phi recognizing carbohydrates containing sialic acid at or near the non-reducing terminus.

Flow cytometric analysis employing MRC OX 6 and MRC OX17 monoclonal antibodies recognizing determinants on RT1.B or RT1.D molecules, equivalent to murine I-A and I-E, respectively, was used to detect rat MHC class II antigen (Ag) expression. Approximately 5% of freshly isolated rat bone marrow cells (BMC) expressed RT1.B and over 30% displayed RT1.D molecules. The RT1.D+ cells were W3/13+, OX 7+, OX 19- and OX 22-. After one week culture of BMC with murine recombinant granulocyte/macrophage colony-stimulating factor (GM-CSF), regardless of concentrations, 90 to 95% of the cells were scored as bone marrow-derived macrophages (BMDM phi), and over 30% expressed both RT1.B and RT1.D Ag. GM-CSF increases the percentage of BMDM phi bearing MHC class II Ag in a concentration-dependent manner. This effect seems to be specific because antibodies to interferon-gamma, tumor necrosis factor-alpha or interleukin-4 did not reduce the number of cells expressing RT1.B and RT1.D Ag. Furthermore, GM-CSF was able to trigger expression of class II molecules on rat peritoneal macrophages (M phi) and BMDM phi resulted from cultures of BMC with mouse M phi-CSF (M-CSF), and the RT1.B and RT1.D inducing effect of GM-CSF was opposed by M-CSF, and by anti-GM-CSF antibodies. The induction of MHC class II Ag synthesis by GM-CSF on rat BMDM phi was confirmed at the mRNA level by Northern blot analysis employing cDNA probes encoding the RT1.B alpha.