J Leukocyte Bio 66/6 - Semantic Scholar

Human polymorphonuclear neutrophils express a B7-1-like molecule Anja Windhagen, Susanna Maniak, Andreas Gebert,* Isabel Ferger, Ulrich Wurster, and Fedor Heidenreich Laboratory of Neuroimmunology, Department of Neurology, and *Center of Anatomy, Medical School Hannover, Germany

Abstract: Polymorphonuclear neutrophils (PMN) are part of the innate immune system and are first-line effector cells in acute inflammatory responses. On activation PMNs secrete cytokines and oxygen metabolites that might be involved in the regulation of the acquired immune response. We show here that peripheral blood PMNs constitutively express a B7-1-like molecule as detected by immunostaining with several B7-1 antibodies. Reverse transcriptase-polymerase chain reaction using three sets of primers spanning different regions of B7-1 indicate dissimilarities at the mRNA level. B7-1 mRNA is expressed in bone marrow cells and lipopolysaccharide (LPS)-stimulated but not in unstimulated PMNs. The B7-1-like molecule is localized to the cytoplasmic granules and translocated to the cell surface after stimulation with LPS or interleukin-12 in some donors. Binding of CTLA4-Ig suggests that the B7-1-like molecule can interact with functional B7 ligand and might be important in the immunobiology of PMNs. J. Leukoc. Biol. 66: 945–952; 1999. Key Words: intracellular · inflammatory responses · bone marrow cells

INTRODUCTION Polymorphonuclear neutrophils (PMNs) are first-line effector cells in microbial infections and on stimulation produce a variety of bioactive substances, e.g., enzymes, cytokines, and reactive oxygen metabolites, that mediate cytotoxicity and inflammatory reactions in the tissue [1–3]. Recently several reports suggested that PMNs may be capable of presenting antigen to T lymphocytes because they can be induced to express major histocompatibility complex II (MHC II) molecules [4]. Eosinophils from interleukin-5 (IL-5) transgenic mice express both B7-1 and B7-2 [5] and expression of B7-2 was also found on eosinophils in inflammatory skin lesions in mice [6]. The B7- molecules deliver the most potent costimulatory signals to T cells [7] and simultaneous blockade of these molecules causes complete inhibition of T cell responses [8, 9]. In humans B7-2 but not B7-1 is expressed on resting B cells, dendritic cells, monocytes, and microglia [7, 10–12]. After stimulation, these cells also express B7-1 and increase the level

of B7-2. The question of whether the two costimulatory molecules B7-1 and B7-2 have different or overlapping functions is unresolved as yet [13–16]. The interactions between PMNs and T cells might be important in human inflammatory diseases, e.g., inflammatory central nervous system (CNS) diseases like bacterial or viral meningitis, in which PMNs are the predominant cell type in early phases of the immune response. However, PMNs can also be found in inflammatory lesions of autoimmune diseases like rheumatoid arthritis or experimental autoimmune encephalomyelitis (EAE), an animal model for multiple sclerosis (MS) [17, 18]. We examined the expression of costimulatory molecules B7-1 and B7-2 in inflammatory CSF specimens from patients with bacterial meningitis. It was surprising to find that in all CSF samples staining of PMNs with the monoclonal Ab (mAb) 104 (B7-1, mAb 104 antigen) was detected. We then investigated expression and regulation of B7 molecules in PMNs.

MATERIALS AND METHODS Patients Cerebrospinal fluid (CSF) was obtained from eight patients who underwent routine diagnostic lumbar puncture for bacterial meningitis at the Department of Neurology, Medical School Hannover, during November 1996 and August 1997 (Table 1). Patients gave informed consent that samples be used for research purposes after clinical diagnostic tests were completed. None of the patients had received antibiotic treatment before the experiments. Patients 3 and 4 had tuberculous meningitis, patient 7 had Streptococcal meningitis. In the other patients CSF culture for bacteria was not revealing.

Reagents Primary antibodies used were anti-B7-1 (clone mAB104, IgG1, Immunotech, Marseille, France; clones EW3.1F1.F5, IgG2b, and EW3.4B2.C4, IgG2a, gifts from Genetics Institute, Cambridge, MA; clone L307.4, IgG1, from Becton Dickinson, and clone P1.H5.A1.A1, IgG1, from Ancell, Bayport, MN), anti-B7-2 (clone FUN-1, IgG1, PharMingen, San Diego, CA), anti-CD16 (clone 3G8, IgG1, PharMingen), and nonspecific mouse IgG1, IgGb, and IgG2a isotype controls (DAKO, Glostrup, Denmark). CTLA4-Ig was obtained from Ancell. Human B7-1/IgG-Fc fusion protein (a gift from Genetics Institute) was used as positive control for Western blot.

Correspondence: Anja Windhagen, Department of Neurology, Medical School Hannover, 30623 Hannover, Germany. E-mail: [email protected] Received March 1, 1999; revised August 2, 1999; accepted August 3, 1999.

Journal of Leukocyte Biology Volume 66, December 1999 945

TABLE 1.

Patients with Bacterial Meningitis

Disease mAb 104 CSF PMNs, Monocytes, Lymphocytes, duration, intramAb 104 Patient cells/µL % % % days cellular surface, %

1 2 3 4 5 6 7 8

120 2120 171 213 330 48 1336 1213

91 94 54 47 63 51 12 98

2 1 4 9 11 9 22 0

7 5 42 44 26 40 66 2

1 1 24 10 9 14 19 1

⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

— — 2, 5 — — — 79 —

Shown are the total CSF cell count, the percentages of different cell types, and the percentages of PMNs that had surface staining with mAb 104 (B7-1). In all samples intracellular staining of PMNs with mAb 104 was detected.

Cytokines used for stimulation were IL-8 at 5 µg/mL, interferon-␥ (IFN-␥) at 1000 U/mL, and IL-12 at 50 ng/mL all from R & D Systems (Minneapolis, MN). Other cytokines used were recombinant IL-1 and IL-6 from Tebu (Frankfurt, Germany) at 100 ng/mL, granulocyte colony-stimulating factor (G-CSF) from Amgen (Thousand Oaks, CA) at 100 ng/mL, GM-CSF from PeproTech, Inc. (Rocky Hill, NY) at 30 ng/mL, and tumor necrosis factor ␣ (TNF-␣) from Alexis Corp. (Grunberg, Germany) at 1000 U/mL. Other reagents used were PMA at 10 ng/mL, N-formyl-methionyl-leucylphenylalanine (fMLP) at 10-7 M, and lipopolysaccharide (LPS) at 10 µg/mL, all from Sigma (Deisenhofen, Germany).

Cell purification and culture PMNs were isolated from the peripheral blood of healthy donors as previously described [19]. In brief, blood was layered over Ficoll and peripheral blood mononuclear cells (PBMCs) were removed. PMNs were separated from red blood cells (RBCs) by sedimentation in 3% hydroxyethyl starch HES 450/0.7 (Fresenius, Bad Homburg, Germany). Remaining RBCs were lysed with H2O. The resulting cell population was stained with the naphthol-AS-D-chloroacetate (NCAE) kit from Sigma to determine monocyte contamination and percentage of eosinophils in the preparation. Here neutrophils are stained red, whereas eosinophils are not stained. PMNs were suspended in RPMI-1640 (Sigma) supplemented with 10% FCS (GIBCO, Eggenstein, Germany) and 1% HEPES (Sigma) and cultured at 37°C for up to 44 h in 5-mL polypropylene tubes (Greiner, Nurtingen, Germany) at 107/mL in the presence of cytokines or LPS. Stimulation with PMA or fMLP was performed for 25 min. To determine cell viability the Trypan blue exclusion test was used. To inhibit shedding of B7 from the cell surface by protease activity a protease inhibitor cocktail (Roche Molecular Biochemicals) 1 tablet/10 mL was added to the culture medium. Expression of B7-1 and B7-2 was examined after 15, 30, 60, and 120 min and after 8, 16, 20, and 44 h on the surface and in the cytoplasm by FACS analysis and immunohistochemistry. For RT-PCR, RNA was extracted after 6, 12, and 20 h of stimulation. Bone marrow was obtained from two healthy subjects who donated bone marrow for transplantation purposes at the Department of Hematology, Medical School Hannover, after informed consent. Bone marrow cells were isolated by Ficoll separation to remove RBCs and granulocytes.

Immunocytochemistry For APAAP staining, cytospins were fixed in acetone for 2 min. Primary antibody was added in phosphate-buffered saline (PBS)/1% bovine serum albumin (BSA) for 30 min, and fixed for 2 min in 0.04% glutaraldehyde (Merck, Darmstadt, Germany). Rabbit anti-mouse IgG was added for 30 min followed by incubation with APAAP mouse monoclonal antibody, both from DAKO. Bridging antibody and APAAP complex were each repeated for 15 min. Substrate solution (Fast-red tablets, KEM EN TEC, Copenhagen, Denmark) was added in 0.1 M Tris buffer for 15 min. Counterstaining was done with Hemalum solution (Merck) for 2 min. Slides were mounted with Aquamount Improved (BDH Laboratory Supplies, Poole, UK) and examined by light microscopy. The fixation with acetone allows for intracytoplasmic staining. For immunofluorescence cells were transferred to Bio-Rad adhesion slides

946

Journal of Leukocyte Biology Volume 66, December 1999

(Bio-Rad, Munich, Germany) and fixed in 4% PFA (Sigma) for 30 min. After blocking with 2% fetal calf serum (FCS) washing was done in EBSS. Cells were incubated with primary antibody in EBSS/10% human AB⫺ serum (Sigma) for 30 min at room temperature followed by secondary antibody [fluorescein isothiocyanate (FITC)-or phycoerythrin (PE)-conjugated goat-anti-mouse IgG1 or IgG, Southern Biotechnology, Birmingham, AL] for another 30 min. For cytoplasmic staining 0.1% saponin (Sigma) was added to the washing and staining buffers during the whole process. Slides were covered with AquaPolymount (Polysciences, Warrington, PA) and viewed under a confocal microscope.

Flow cytometry For flow cytometry the same primary and secondary antibodies as above were used after fixation of cells with 4% PFA for 15 min. Cells were stained with or without 0.1% saponin buffer that was supplemented with 10% human AB⫺ serum to block unspecific antibody binding for 15 min for each primary and secondary antibody at room temperature and analyzed on a FACScan (Becton Dickinson). Mean fluorescence intensities were determined from data histograms with the use of CellQuest software. As positive control for B7 staining, the Epstein-Barre virus (EBV)-transformed B cell line LE509 (a gift from R. Schmidt, Hannover), which expresses both B7-1 and B7-2 on the cell surface, was used.

Reverse transcriptase-polymerase chain reaction (RT-PCR) Total cellular RNA was extracted from 2 ⫻ 106 cells and cDNA was prepared as described previously [20, 21]. Each 25-µL PCR contained 5 µL cDNA, 0.5 µg of forward and 0.5 µg reverse primers, 1.2 U of Taq polymerase, and 10 µL of a mix of dNTPs and Taq buffer that was prepared as a master mix for each set of samples. Thermal cycling was done at 94°C for 1 min, 55 or 60°C for 1 min, and 72°C for 1.5 min for 28–35 cycles. Primer sequences for ␤-actin and IL-12 were as described previously [20]: forward primer B7-1, 58 AAC TCg CAT CTA CTg gCA AAA ggA gAA 38; reverse primer B7-1, 58 ggg AAA CTg TTg TgT TgA Tgg CAT TTA 38; forward primer B7-2, 58 gTA TTT Tgg CAg gAC CAg gA 38; reverse primer B7-2, 58 gCC gCT TCT TCT TCT TCC AT 38. To determine B7-1 mRNA expression, two additional primer pairs (B7-1a and B7-1b) were used that amplify regions from the amino- or carboxy-terminal end of the B7-1 cDNA (accession number M27533). B7-1a forward 58 ATg ggC CAC ACA Cgg Agg-38, B7-1a reverse 58 Tgg gCg CAg AgC CAg gAT CA-38, B7-1b forward 58 ACT ggC AAA Agg AgA AAg AAA-38, B7-1b reverse 58 ATA CAg ggC gTA CAC TTT CC-38. PCR products were separated on 1.5% agarose gels, stained with ethidium bromide, and analyzed on a video imaging system (Eagle-Eye, Stratagene).

Immunoblotting Whole-cell lysates were prepared from 5 ⫻ 106 untreated and PMA- or fMLP-stimulated PMNs by homogenizing in 0.3 mL PBS/0.1% Tween containing a protease inhibitor cocktail (Roche Molecular Biochemicals) 1 tablet/10 mL. Sixty micrograms of protein from cell lysates or four micrograms protein from cell supernatant was loaded in each lane on 7.5–18.5% T gradient gels and run overnight under non-reducing conditions. Recombinant B7-1/IgG-Fc fusion protein (0.5 ng) was used as positive control. The separated proteins were transferred electrophoretically to 0.2-µm nitrocellulose membranes in a tank blot apparatus. The membranes were blocked with 5% skimmed milk and incubated overnight with 0.4 µg/mL mAb 104 (Immunotech). After three washes with TBS-Tween a biotinylated rabbit anti-mouse antibody (Dako, E0354 at 1:2000) was added and developed for 45 min using streptavidin/ alkaline phosphatase complex and BCIP/NBT as color substrates.

RESULTS Neutrophils in the CSF from bacterial meningitis To investigate expression of costimulatory molecules on CSF cells from patients with bacterial meningitis, samples were http://www.jleukbio.org

Expression of B7 molecules in peripheral blood PMNs

Fig. 1. Immunocytochemistry of CSF cells from a patient with bacterial meningitis (acute phase, PMN percentage 98%). (A) mAb 104, (B) B7-2 mAb. All PMNs show positive staining with mAb 104, but not B7-2 mAb. B7-2 staining, however, is detectable on CSF monocytes. The same results were obtained also from the other CSF specimen tested.

To determine whether expression of B7-1 is a common feature of PMNs or is present only in PMNs in inflammatory reactions we next examined PMNs from the peripheral blood of healthy donors. We performed surface and intracytoplasmic staining for B7-1 and B7-2 mAbs. The PMN preparation obtained consisted of 96–99% neutrophils as determined by CD16 expression and NCAE staining, the contaminating cells being eosinophils. Monocytes and lymphocytes were not detectable. PMNs from the peripheral blood expressed the mAb 104 antigen (B7-1) in the cytoplasm, whereas B7-2 expression was not detectable (Fig. 3). Staining was also obtained by binding of two other antibodies to B7-1 (EW3.1 and P1.H5.A1.A1) as well as CTLA4-Ig, the functional ligand for B7-1 and B7-2. However, the B7-1 antibodies L307.4 and EW3.4 did not stain PMNs (Fig. 3). This staining pattern was defined as B7-1-like. Eosinophils also showed positive staining with mAb 104 (not shown), however, further study of eosinophils with immunofluo-

stained with the B7-1 (mAb 104) and B7-2 mAbs. In all samples both intracellular and surface staining of CSF cells was performed. PMNs from all CSF samples showed intracellular staining with mAb104 (B7-1; Fig. 1A) but not B7-2. Lymphocytes and monocytes present in some CSF samples (Table 1) did not express B7-1. In addition, PMNs from CSF of two patients (one with tuberculous meningitis and one with streptococcal meningitis) that both were in the subacute phase of the disease showed surface expression of B7-1 as detected by staining with mAb 104 (Fig. 2). B7-2 was detected on CSF monocytes only (Fig. 1B).

Fig. 2. Surface staining with mAb 104 of CSF neutrophils from two patients with bacterial meningitis. Neutrophils were gated by forward/sideward scatter characteristics. (A) patient 7, (B) patient 3 (see Table 1). In (A) there were 79% of CSF neutrophils expressing B7-1 on the cell surface; in (B) the percentage of positively stained neutrophils was 2.5%.

Fig. 3. Expression of B7-1-like molecule in PMNs. (A–F) Cytoplasmic staining of PMNs with different B7-1 Abs (mAb 104, EW3.1, P1.H5.A1.A1, L307.4) as well as B7-2 and CTLA4-Ig. Positive staining of PMNs is seen with Abs 104, P1.H5.A1.A1, EW3.1, and CTLA4-Ig but not with B7-2 mAb or L307.4. B7-1 Abs EW3.4 also did not stain PMNs (not shown). (G, H) Surface staining of EBV cells with mAb 104 and EW3.4. Compared with mAb104 and the other B7-1 antibodies used, EW3.4 only showed weak staining of EBV cells.

Windhagen et al. B7-1 expression in human neutrophils

947

rescence was not successful due to high nonspecific staining and autofluorescence. EBV cells were used as positive control because these cells constitutively express high levels of B7-1 and B7-2. All B7-1 antibodies used in this study showed positive staining of EBV cells. Compared to the other B7-1 antibodies the mAb EW3.4 at optimal concentrations showed only weak staining (Fig. 3). Staining with mAb 104 in PMNs showed a speckled pattern and was morphologically localized to the intracytoplasmic granules. This was determined by the typical size and localization of the granules on confocal microscopy (Fig. 4). To further characterize the B7-1-like antigen in PMNs, Western blot was performed with mAb 104. Several distinct bands were revealed that showed a molecular mass between 30 and 65 kDa (Fig. 5) as well as diffuse staining in this region. The molecular mass of the observed bands is similar to those described for the B7-1 protein in B cells.

Surface expression of B7-1-like antigen in PMNs after activation Because the B7-1-like molecule is localized to the cytoplasmic granules we examined whether it is released from PMNs after degranulation. PMNs were stimulated with PMA or fMLP, which are complete secretagogues. The cell lysates and supernatants were collected and examined for the presence of the B7-1-like molecule by Western blot. Staining with mAb 104 revealed the same pattern of bands in the cell lysates after PMA stimulation, whereas in the supernatant there was only weak staining of a band with a molecular mass of 30 kDa (Fig. 5). The same results were obtained with fMLP stimulation (data not shown). These results indicate that the B7-1-like molecule is not released by degranulation.

To further study the regulation of the B7-1 like molecule in PMNs we examined whether surface expression of the B7-1like molecule could be induced on PMNs from the peripheral blood of healthy donors. Because most intense staining was seen with mAb 104 this antibody was used for these experiments. PMNs were stimulated with a panel of cytokines. LPS, IFN-␥, and IL-12 could induce surface expression of the B7-1-like molecule. This finding was reproduced in six out of ten experiments. Figure 6 shows an example of surface staining with mAb 104 on PMNs after stimulation with LPS, IL-12, and IFN-␥ for 24 h. With the other cytokines tested (IL-1, IL-6, G-CSF, GM-CSF, TNF-␣, IL-8) no surface expression of the B7-1-like molecule was observed.

Expression of B7 mRNA in neutrophils To test whether B7-1 and B7-2 were present at the transcriptional level we performed RT-PCR for B7-1 and B7-2 mRNA from unstimulated neutrophils and after 6, 10, and 20 h of stimulation with different stimuli. Resting PMNs did not express detectable levels of B7-1 or B7-2 mRNA. mRNA expression of B7-1 but not B7-2 was induced in PMNs stimulated with LPS after 20 h (Fig. 7), whereas after 6 and 10 h of stimulation B7-1 mRNA was not yet detectable. Stimulation with IL-8, IL-12, or IFN-␥ did not induce B7-1 mRNA. Other stimuli that did not induce B7-1 mRNA expression were G-CSF, TNF-␣, and GM-CSF (not shown). Simultaneously with B7-1 mRNA expression we observed expression of IL-12 mRNA in PMNs. Because B7-1 mRNA was absent from PMNs we investigated whether there was expression of B7-1 in granulopoietic cells in the bone marrow. Bone marrow cells were isolated from bone marrow aspirate by Ficoll separation and examined by flow

Fig. 4. Confocal laser scanning microscopy shows that mAb 104 antigen (a and b) but not B7-2 (c) is expressed in PMNs. High-power optical sections running across the cytoplasm of the PMNs (b) reveal that the epitopes detected are associated with the characteristic small cytoplasmic granules of the PMNs (arrowheads in b), but are absent from the cell surface. Only negligible fluorescence is seen in controls (d) using a nonspecific control mouse IgG1 antibody. a, c, d ⫻330; b ⫻1550.

948


http://www.jleukbio.org

Fig. 6. Surface expression of B7-1-like molecule can be induced by stimulation of PMNs in vitro. (A) unstimulated, (B) LPS stimulated, (C) IFN-␥stimulated, and (E) IL-12-stimulated PMNs; line histogram isotype control, dotted histogram mAb 104 staining. (D) unstimulated PMNs, surface staining with anti-CD16 antibody (dotted histogram) as marker for neutrophils and positive control. (F) PMNs stimulated with IFN-␥ surface staining with B7-2 antibody (dotted histogram). Stimulation with the different cytokines was performed for 24 h. Fig. 5. Detection of B7-1-like molecule in PMNs by Western blot. Whole-cell lysate of unstimulated and PMA-stimulated PMNs as well as supernatant of PMA-stimulated PMNs was examined for presence of B7-1-like antigen using the mAb104. The molecular mass of the observed bands from cell lysate is in the range of 30 to 65 kDa. After PMA stimulation the B7-1 molecule is still detected in the cell lysate and only a weak band with a molecular mass of 30 kDa is detected in the supernatant.

In PMNs from the CSF of patients with bacterial meningitis as well as from peripheral blood of healthy donors, expression of a B7-1-like molecule was detected in the cytoplasmic granules. This is a novel finding because expression of costimulatory

cytometry. The resulting cell population contained about 30% CD16-positive cells but no mature PMNs. Cells were gated on the CD16⫹ population and intracytoplasmic B7-1 was detected while there was no surface expression of B7-1, similar to the results obtained with PMNs (Fig. 8). Bone marrow cells also expressed B7-1 mRNA as determined by RT-PCR using three sets of B7-1 PCR primers that amplify different regions of the B7-1 cDNA (Fig. 9). It is interesting that with the primers B7-1 and B7-1b PCR product of expected length was obtained, whereas no distinct PCR product could be generated with the primers B7-1a. The B7-1a forward primer is complementary to a sequence encoding the signal peptide, the reverse primer to a sequence in the IgG-like V-type domain. The same pattern was obtained when cDNA from LPS-stimulated PMNs was used (Fig. 9).

DISCUSSION We investigated the expression of B7 costimulatory molecules in PMNs from inflammatory CSF samples and peripheral blood.

Fig. 7. Expression of B7-1 mRNA is induced after stimulation of PMNs with LPS (but not IL-8, IL-12, or IFN-␥) for 20 h measured by RT-PCR. Simultaneously, mRNA of IL-12 p40 is increased, whereas B7-2 mRNA is not detectable. ␤-Actin, 30 PCR cycles; B7-1, B7-2, and IL-12p40, 35 PCR cycles.


949

Fig. 8. Expression of B7-1-like molecule in bone marrow cells. (A) Surface staining with CD16 mAb, (B) surface staining with mAb 104, (C) intracytoplasmic staining with mAb 104. Line histogram isotype control, dotted histograms CD16 or mAb 104. Cells were gated on the CD16expressing population.

molecules in human PMNs has not been reported so far, although expanding research has been directed toward the interaction of the innate and acquired immune systems. Because the B7-1-like molecule in PMNs might represent a novel link between PMNs and T cells, which express two B7 ligands, we further characterized this molecule in relation to other B7 molecules. The initial observation was made using the mAb 104 that specifically stains B7-1. Four additional B7-1 antibodies were tested and two also recognized B7-1 in PMNs, whereas two did not. There are several possible explanations for this B7-1-like staining pattern in PMNs. One hypothesis would be the existence of different glycosylation forms of B7-1 in PMNs similar to findings that were recently described with human B7-2. It was shown that human T cells express a hypoglycosylated form of B7-2 that has different binding affinities to CD28 and CTLA4 compared with the B7-2 expressed on EBV cells [22]. B7-1 has eight potential N-linked glycosylation sites in the extracellular region and the molecular mass of the glycosylated protein has been found to be in the range of 44–60 kDa, whereas the predicted size of nonglycosylated B7-1 is 30 kDa [23]. The B7-1 like molecule in PMNs shows bands with a molecular mass of between 30 and 65 kDa on Western blot. From this finding it can be concluded that both molecules are similar in size; the bands at 30 kDa might represent hypo- or non-glycosylated protein. The Ab L307.4, which did not stain the B7-1-like molecule in PMNs, recognizes an epitope in the IgG-like V-type domain of B7-1 [24], indicating that there are differences between both molecules in this epitope. The epitope recognized by mAb EW3.4 is not known. Because the mAb EW3.4 showed only weak staining on EBV cells that express high levels of B7-1 the

Fig. 9. Expression of B7-1 mRNA in bone marrow cells and PMNs using three sets of B7-1 primers spanning different parts of the B7-1 cDNA. B7-1, cDNA position 499–888, PCR product 389 bp; B7-1a, cDNA position 318–641, PCR product 323 bp; B7-1b, cDNA position 512–1182, PCR product 671 bp. In bone marrow cells and LPS-stimulated PMNs a PCR product is generated with primers B7-1 and B7-1b, but not B7-1a.

950


failure of this antibody to stain PMNs might be due to low sensitivity or might indicate differences in the epitope recognized. Alternatively the B7-1-like molecule may be a distinct member of the B7 family of costimulatory molecules. A novel costimulatory molecule distinct from B7-1 and B7-2 was recently described on activated human myoblasts and this molecule could be detected by the BB-1 antibody and CTLA4-Ig but not other B7-1 mAbs [25]. This possibility is supported by our finding that one set of the PCR primers used (B7-1a primers) does not amplify, whereas the other B7-1 primers amplified a PCR product of expected length. The B7-1a primers are complementary to regions in the signal sequence and IgG-like V-type domain of the B7-1 cDNA. These findings suggest that the B7-1 like molecule in PMNs is different from B7-1 at the mRNA level, possibly at the amino-terminal region containing the IgG-like V-type domain or the signal sequence. Alternatively, this result could be explained by the existence of different B7-1 splicing forms in PMNs compared with B cells or monocytes. For the PCR analysis it was important to exclude contamination of the PMN preparation with monocytes because these are potent producers of B7-1 after stimulation with LPS. For this purpose the PMN preparation was stained with naphthol-AS-Dchloroacetate, where no mononuclear cells were detectable. Moreover, monocytes have different kinetics of B7 mRNA expression and produce high levels of both B7-1 and B7-2 mRNA on LPS stimulation with a maximum at about 6 h [26], whereas PMNs started B7-1 mRNA expression only after 20 h. It cannot be ruled out that eosinophils express B7-1 mRNA, although the amount of B7-1 mRNA detected was not dependent on the eosinophil percentage of the PMN preparation. It is interesting that in PMNs stimulated with LPS there was expression of both B7-1 and IL-12 with the same kinetics. Simultaneous expression of B7-1 and IL-12 was previously found in inflammatory lesions from multiple sclerosis brains, indicating that this might be of importance in inflammatory conditions in vivo [20]. Recent findings showed that IL-12 mRNA in neutrophils is expressed after LPS stimulation for 20 h [27]. Other mRNA transcripts are expressed constitutively in peripheral blood PMNs like actin and MHC class I [28]. The absence of B7-1 mRNA in unstimulated PMNs is unusual because in most cell types protein expression in the absence of mRNA expression is not known. However, in PMNs it was previously reported that constitutive expression of IL-6 is present in the cytoplasm in the absence of IL-6 mRNA. In immature cells in the bone marrow, however, IL-6 mRNA was detectable and correlated to the intracytoplasmic level of circulating neutrophils [29]. Similarly, our results show B7-1 http://www.jleukbio.org

mRNA expression in bone marrow cells and expression of B7-1-like protein in the CD16⫹ bone marrow population. This suggests that intracytoplasmic stores of B7-1-like molecule accumulate during granulopoiesis in the bone marrow. After stimulation with LPS mRNA for B7-1 is transcribed, indicating that mature PMNs can produce new B7-1-like molecule if their survival is prolonged as in inflammatory conditions. B7-1 is known as a surface molecule of APCs that is important for the costimulation of T cells. PMNs have so far not been associated with T cell costimulation, although they do secrete cytokines that can influence T cell function [3]. The expression of B7-1 on APCs requires external stimuli and is tightly regulated. B7-1 appears on monocytes and B cells about 2 days after stimulation and is therefore thought to influence the later stages of an immune response [30, 31]. Because the B7-1-like molecule was localized to the intracytoplasmic granules we used degranulating stimuli to induce release of B7-1-like molecule from the cells. However, after PMA or fMLP stimulation B7-1-like molecule could still be detected in the cell lysates but not in the cell supernatant in Western blots, indicating that it is not released by degranulation. The faint band with a molecular mass of 30 kDa that was detectable in the supernatant might represent non-glycosylated protein released from small numbers of dying cells. In PMNs intracellular proteins may translocate to the cell surface, e.g., Cd11b and CD45 translocate from granules to the cell surface after activation [32] and rapid surface expression from an intracellular pool was described for molecules of the CD18 complex [33]. Expression of B7-1-like molecule on PMNs was observed after considerably longer stimulation. PMNs as terminally differentiated cells are relatively shortlived, undergoing death by apoptosis. Apoptosis in the mature neutrophil is a constitutive process that results in a rapid turnover of the circulating neutrophil population with a t1/2 of 5–6 h in vivo [34, 35]. However, survival of neutrophils can be greatly extended after exposure to microenvironmental signals involved in infection and immunity such as LPS, streptococci, or several cytokines like IL-6, G-CSF, or GM-CSF [3]. To determine viability of the neutrophils we performed trypan blue exclusion and CD16 staining [36] and consistently ⬎90% of cells were viable in our experiments. Therefore surface expression of a B7-1-like molecule on PMNs may be of biological significance when under optimal in vivo conditions survival is prolonged, particularly at sites of inflammation. Similar to our findings with the B7-1-like molecule it was reported that PMNs may express MHC II on the cell surface after prolonged in vitro culturing [4]. These authors also reported that MHC II expression on PMNs can only be induced in some donors, indicating individual patterns of responsiveness. Similarly, we observed surface expression of B7-1-like molecule in six out of ten individuals, although we did not perform longitudinal experiments. Surface expression of B7-1-like molecule on PMNs was demonstrated in CSF from two out of eight patients with meningitis. These two patients had a longer disease duration of 24 and 19 days and a percentage of PMNs in the CSF cell population of 54 and 12%, the remaining cells being monocytes and lymphocytes. This finding indicates that translocation of the B7-1-like molecule to the cell surface of PMNs may be

relevant in inflammatory conditions in vivo. Because surface expression of the B7-1-like molecule was only detected in the subacute phase of bacterial meningitis it can be hypothesized that other cells possibly in combination with bacterial factors like LPS are important to induce the translocation of B7-1-like molecule to the cell surface. Alternatively, surface expression of B7-1-like molecule might reflect age or maturation of PMNs in the course of chronic inflammation. The possibility that the B7-1-like molecule is shed or released from other cells and secondarily bound to PMNs cannot be ruled out, however, this would be more likely if all PMNs in the CSF had been positive. After in vitro stimulation of PM a low level of surface expression of B7-1-like molecule could be induced as detected by a shift of fluorescence intensity in the whole-cell population. From this it cannot be determined whether all PMNs translocate a small amount of B7-1-like molecule to the surface or whether this phenomenon is a result of B7-1-like molecule released from a few cells. Because the stimuli used did not induce cell death as determined by trypan blue exclusion or loss of B7-1-like molecule in a subpopulation as determined by immunocytochemistry it seems more likely that the whole PMN population translocates B7-1-like molecule from the intracytoplasmic pool to the cell surface. It is possible that the stimuli used here are not optimal and that other unknown factors, possibly contact with other cells like mononuclear cells at an inflammatory site, are necessary to induce a stronger surface expression. In the course of an inflammatory reaction PMNs are among the first cells to enter the tissue. Here they can perform a variety of biological functions, including phagocytosis, cytokine secretion, degranulation, or translocation of stored molecules to the cell surface. We have presented evidence for a B7-1-like molecule as a new member of the B7 family in PMNs from inflammatory CSF specimen and peripheral blood. This molecule may have functional activity because surface expression can be induced in vitro and was also observed in vivo in an inflammatory CSF specimen. Moreover, the B7-1-like molecule binds the functional B7 ligand CTLA4-Ig and may therefore have a role in the interaction between PMNs and T cells.

ACKNOWLEDGMENTS This work was supported by an internal grant of the Medical School Hannover (A. W.), and by the Deutsche Forschungsgemeinschaft (Ge 647/2-2 to A. G.). We thank Reinhold Schmidt, Klaus Resch, Per Ho¨llsberg, and David Hafler for helpful discussions and Fransziska Bode for excellent technical assistance.

REFERENCES 1. Yeaman, G., Collins, J., Currie, J., Guyne, P., Wira, C., Fangers, M. (1998) IFN-␥ is produced by polymorphonuclear neutrophils in human uterine endometrium and by cultured peripheral blood polymorphonuclear neutrophils. J. Immunol. 160, 5145–5153. 2. Marcinkiewicz, J. S. (1997) Neutrophil chloramines: missing links between innate and acquired immunity. Immunol. Today 18, 577–580.


951

3. Cassatella, M. S. (1995) The production of cytokines by polymorphonuclear neutrophils. Immunol. Today 16, 21–26. 4. Gosselin, E., Wardwell, K., Rigby, W., Guyres, P. (1993) Induction of MHC Class II on human polymorphonuclear neutrophils by granulocyte/ macrophage colony-stimulating factor, IFN-␥ and IL-3. J. Immunol. 151, 1482–1490. 5. Tamura, N., Ishii, N., Nakazawa, M., Nagoya, M., Yoshinari, M., Amano, T., Nakazima, H., Minamis, M. (1996) Requirement of CD80 and CD86 molecules for antigen presentation by eosinophils. Scand. J. Immunol. 44, 229–238. 6. Reiser, H., Schneebergers, E. (1996) Expression and function of B7-1 and B7-2 in hapten-induced contact sensitivity. Eur. J. Immunol. 26, 880–885. 7. Reiser, H., Stadeckers, M. J. (1996) Costimulatory B7 molecules in the pathogenesis of infectious and autoimmune disease. N. Engl. J. Med. 335, 1369–1377. 8. Linsley, P. S., Wallace, P. M., Johnson, J., Gibson, M. G., Greene, J. L., Ledbetter, J. A., Singh, C., Teppers, M. A. (1992) Immunosuppression in vivo by a soluble form of the CTLA-4 T cell activation molecule. Science 257, 792–795. 9. Lenschow, D. J., Zeng, Y., Thistlethwaite, J. R., Montag, A., Brady, W., Gibson, M. G., Linsley, P. S., Bluestones, J. A. (1992) Long-term survival of xenogeneic pancreatic islet grafts induced by CTLA4-Ig. Science 257, 789–792. 10. Caux, C., Vanbervliet, B., Masscrier, C. (1994) B70/B7-2 is identical to CD86 and is the major functional ligand for CD28 expressed on human dendritic cells. J. Exp. Med. 180, 1841–1847. 11. deSimone, R., Giampaolo, A., Giometto, B. (1995) The costimulatory molecule B7 is expressed on human microglia in culture and in multiple sclerosis acute lesions. J. Neuropathol. Exp. Neurol. 54, 175–187. 12. Dangond, F., Windhagen, A., Groves, C. J., Haflers, D. A. (1997) Constitutive expression of costimulatory molecules by human microglia and its relevance to CNS autoimmunity. J. Neuroimmunol. 76, 132–138. 13. Guerder, S., Meyerhoff, J., Flavells, R. (1994) The role of the T cell costimulator B7-1 in autoimmunity and the induction and maintenance of tolerance to peripheral antigen. Immunity 1, 155–166. 14. Guerder, S., Eynon, E., Flavells, R. (1998) Autoimmunity without diabetes in transgenic mice expressing beta-cell specific CD86, but not CD80: parameters that trigger progression to diabetes. J. Immunol. 161, 2128– 2140. 15. Kuchroo, V. K., Prabhu Das, M., Brown, J. A., Ranger, A. M., Zamvil, S. S., Sobel, R. A., Weiner, H. L., Nabavi, N., Glimchers, L. H. (1995) B7-1 and B7-2 costimulatory molecules activate differentially the Th1/Th2 developmental pathways: application to autoimmune disease therapy. Cell 80, 707–718. 16. Freeman, G. J., Boussiotis, V. A., Anumanthan, A. (1995) B7-1 and B7-2 do not deliver identical costimulatory signals, since B7-2 but not B7-1 preferentially stimulates the initial production of IL-4. Immunity 2, 523–532. 17. Edwards, S., Halletts, M. (1997) Seeing the wood for the trees: the forgotten role of neutrophils in rheumatoid arthritis. Immunol. Today 18, 320–328. 18. Lassmann, H. S. (1983) Chronic relapsing experimental allergic encephalomyelitis: its value as an animal model for multiple sclerosis. J. Neurol. 229, 207–220. 19. Bo¨yum, A. S. (1968) Isolation of mononuclear cells and granulocytes from human blood. Isolation of mononuclear cells by one centrifugation, and of granulocytes by combining centrifugation and sedimentation at 1 g. Scand. J. Clin. Lab. Invest. 97 (Suppl.), 77. 20. Windhagen, A., Newcombe, J., Dangond, F., Strand, C., Woodroofe, N., Cuzner, M. L., Haflers, D. A. (1995) Expression of costimulatory molecules

952


21.

22.

23.

24.

25.

26.

27.

28. 29.

30.

31.

32.

33.

34.

35.

36.

B7-1 (CD80), B7-2 (CD86), and Interleukin 12 cytokine in multiple sclerosis lesions. J. Exp. Med. 182, 1985–1996. Wucherpfennig, K., Newcombe, J., Li, H., Keddy, C., Cuzner, M., Haflers, D. (1992) T cell receptor V alpha-V beta repertoire and cytokine gene expression in active multiple sclerosis lesions. J. Exp. Med. 175, 993–1002. Ho¨llsberg, P., Scholz, C., Anderson, D., Greenfield, E., Kuchroo, V., Freeman, G., Haflers, D. (1997) Expression of a hypoglycosylated form of CD86 (B7-2) on human T cells with altered binding properties to CD28 and CTLA-4. J. Immunol. 159, 4799–4805. Freeman, G., Freedman, A., Segil, J., Lee, G., Whitman, J., Nadlers, L. (1989) B7, a new member of the Ig superfamily with unique expression on activated and neoplastic B cells. J. Immunol. 143, 2714–2722. Engel, P., Wagner, N., Tedders, T. (1995) CD80 workshop report, leucocyte typing V, white cell differentiation antigens. In Proceedings of the Fifth International Workshop and Conference, Boston 1993. Oxford, UK: Oxford University Press, 682–683. Behrens, L., Kerschensteiner, M., Misgeld, T., Goebels, N., Wekerle, H., Hohlfelds, R. (1998) Human muscle cells express a functional costimulatory molecule distinct from B7-1 and B7-2 in vitro and in inflammatory lesions. J. Immunol. 161, 5943–5951. Schmittel, A., Scheibenbogen, C., Keilholzs, U. (1995) Lipopolysaccharide effectively upregulates B7-1 (CD80) expression and costimulatory function of human monocytes. Scand. J. Immunol. 42, 701–704. Cassatella, M., Meda, L., Gasperini, S., D’Andrea, A., Ma, X., Trinchieris, G. (1995) Interleukin-12 production by human polymorphonuclear leukocytes. Eur. J. Immunol. 25, 1–5. Jack, R., Fearons, D. (1988) Selective synthesis of mRNA and proteins by human peripheral blood neutrophils. J. Immunol. 140, 4286–4293. Terebuth, P., Otterness, I., Strieters, R. (1992) Biologic and immunohistochemical analysis of interleukin-6 expression in vivo: constitutive and induced expression in murine polymorphonuclear and mononuclear phagocytes. Am. J. Pathol. 140, 649–657. Boussiotis, V., Freeman, G., Gribben, J., Daley, J., Gray, G., Nadlers, L. (1993) Activated human B lymphocytes express three CTLA-4 binding counter-receptors which costimulate T-cell activation. Proc. Natl. Acad. Sci. USA 90, 11059–11063. Lenschow, D., Sperling, A., Cooke, M., Freeman, G., Rhee, L., Decker, D., Gray, G., Nadler, L., Goodnow, C., Bluestones, J. (1994) Differential upregulation of the B7-1 and B7-2 costimulatory molecules following immunoglobulin receptor engagement by antigen. J. Immunol. 153, 1990–1997. Lacal, P., Pulido, R., Sanchez-Madrid, F., Mollinedos, F. (1988) Intracellular location of T200 and Mo1 glycoproteins in human neutrophils. J. Biol. Chem. 263, 9946–9951. Kishimoto, T., Jutila, M., Lakey Berg, E., Butchers, E. (1989) Neutrophil Mac-1 and MEL-14 adhesion proteins inversely regulated by chemotactic factors. Science 245, 1238–1241. Lee, A., Whyte, M., Hasletts, C. (1993) Inhibition of apoptosis and prolongation of neutrophil functional longevity by inflammatory mediators. J. Leukoc. Biol. 54, 283–288. Watson, R., Redmond, H., Wang, J., Condron, C., Bouchier-Hayss, D. (1993) Neutrophils undergo apoptosis following ingestion of Escherichia coli. J. Immunol. 156, 3986–3992. Dransfield, I., Buckle, A., Savill, J., McDowall, A., Haslett, C., Hoggs, N. (1994) Neutrophil apoptosis is associated with a reduction in CD16 (Fc␥RIII) expression. J. Immunol. 153, 1254.

http://www.jleukbio.org