Annexin A2 is a soluble mediator of macrophage activation - CiteSeerX

Annexin A2 is a soluble mediator of macrophage activation Jennifer F. A. Swisher, Utsha Khatri, and Gerald M. Feldman1 Laboratory of Molecular and Developmental Immunology, Division of Monoclonal Antibodies, Office of Biotechnology Products, Center for Drug Evaluation and Research, Food and Drug Administration, Bethesda, Maryland, USA

Abstract: On the surface of the macrophage, annexin A2 tetramer (A2t) serves as a docking protein or recognition element for bacterial and viral pathogens. Plasma levels of free A2t have been reported to increase following infection, although the mechanistic significance of this observation is unclear. Although annexin A2 had generally been thought to play an anti-inflammatory role, soluble A2t stimulates MAP kinase activity in bone marrow stromal cells downstream of a recently cloned receptor. This raises the question of whether A2t activates human macrophages via MAP kinases and whether it might be capable of acting as an inflammatory mediator. To this end, human monocytederived macrophages were treated with soluble A2t and MAP kinase phosphorylation, p65 NF-␬B activation, and inflammatory mRNA and protein levels were measured. It was found that A2t caused rapid phosphorylation of several MAP kinases, as well as translocation of p65 NF-␬B to the nucleus. A2t stimulated the production of TNF-␣, IL-1␤, and IL-6, as well as several members of the chemokine family within 24 h, which are capable of recruitment and/or activation of a broad range of leukocyte classes. Furthermore, A2t-activated macrophages demonstrated enhanced phagocytic ability for the ingestion of GFP-expressing Escherichia coli. These data are the first to suggest the participation of an annexin in microbial clearance, as well as the establishment of inflammation and the immune response, including the recruitment and activation of immune cells to the site of infection. J. Leukoc. Biol. 82: 1174 –1184; 2007. Key Words: MAP kinase signaling 䡠 NF-␬B signaling 䡠 inflammatory cytokines 䡠 phagocytosis

INTRODUCTION Proteins that coordinate inflammation and the acute-phase response must mount a rapid, efficient defense against invading microorganisms at local and systemic levels, but the amplitude of this response can be difficult to contain. Thus, safeguards must also be built into the system to protect the host, although a perfect compromise cannot always be achieved. It is not surprising then that although many proteins 1174

Journal of Leukocyte Biology Volume 82, November 2007

involved in these events appear to be predominantly either proinflammatory or anti-inflammatory in nature, some are involved in both the initiation and the resolution of an immune response. In this regard, members of the annexin family of proteins have been reported to be largely anti-inflammatory in their actions; annexin A1 has clearly been shown to mitigate anti-inflammatory events downstream of glucocorticoid induction [1, 2], as well as inhibit arachidonic acid production via direct inhibition of phospholipase A2 (PLA2) [1, 2]. It has also been demonstrated that annexins A1 and A2 play a crucial role in the phagocytosis of apoptotic lymphocytes [3], an event at the end of an immune response, which results in the reduction of inflammation by the release of immunosuppressive cytokines. However, evidence is beginning to accumulate that the role of soluble, extracellular annexins in inflammation and the immune response may not be so simple. The annexins are calcium-dependent phospholipid binding proteins that are more divergent in function than structure [4]. This protein family comprises at least 20 members that share a highly conserved, tightly packed ␣-helical core domain, as well as calcium and phospholipid binding domains [4]. Their divergent N termini, however, appear to allow for their diverse and distinct biological functions [5]. They are largely intracellular and most play important roles in membrane trafficking, plasma membrane reorganization during signaling, and calcium regulation [4]. Additional roles, however, have been identified for particular annexins on the extracellular surface of endothelial cells and certain leukocyte populations [3, 4, 6, 7]. Although much has recently been learned about annexin A2 on the surface of macrophages, nothing is known about its role as a soluble mediator in the early events of infection and the coordination of the innate immune response. Annexin A2 is unique among the annexins in that it is known to function largely in heterotetramer form, composed of two 36-kDa subunits of annexin A2 (“p36”) complexed with two 11-kDa S100A10 protein subunits (“p11”). It is highly expressed on the surface of macrophages, where it enjoys promiscuous binding of several proteins and through these interactions, participates in many facets of macrophage biology. It is required for recognition of apoptotic T lymphocytes

1

Correspondence: Laboratory of Molecular and Developmental Immunology, Division of Monoclonal Antibodies, Office of Biotechnology Products, Center for Drug Evaluation and Research, Food and Drug Administration, Bldg. 29A, Rm. 3C22, 29 Lincoln Dr., HFD-123, Bethesda, MD 20892, USA. E-mail: [email protected] Received March 13, 2007; revised July 20, 2007; accepted July 26, 2007 doi: 10.1189/jlb.0307154

0741-5400/07/0082-1174 © Society for Leukocyte Biology

[3], serves as a platform to nucleate thrombolytic activity by binding both tPA and plasminogen [7–9], as well as functioning as a docking protein for cytomegalovirus [10 –12], respiratory syncitial virus [13], Pseudomonas aerurginosa [14], and macrophage-trophic HIV [15, 16]. Its importance as a recognition element for such pathogenic agents predicts the existence of a soluble form of this “receptor” as a decoy, and indeed, soluble annexin A2 tetramer (A2t) is found at low levels in the sera of healthy people [17] and at elevated levels in serum during an infection [18]. Recent studies on osteoclastogenesis have demonstrated that soluble A2t stimulates the proliferation and differentiation of osteoclast precursors through MAP kinase activation in bone marrow stromal cells, resulting in downstream production of GM-CSF and RANKL [19 –21]. More recently, a receptor for A2t has been cloned and its requirement for downstream signaling in stromal cells has been demonstrated [22]. This receptor was shown to be expressed in cells of monocytic lineage, as well as many other cell types. As osteoclasts represent a type of specialized macrophage, the importance of soluble, extracellular A2t in promoting osteoclast differentiation suggests that it may play an important role in macrophage function as well. MAP kinase activation is well known to play a central role in macrophage activation through production and release of inflammatory cytokines [23]. If A2t were to induce macrophage activation via this pathway, it would provide a mechanism for the body to defend itself against the use of surface-bound A2t as a viral and bacterial receptor: cells could shed A2t and gain an extra measure of protection by concomitant activation of surrounding cells. To explore these possibilities, human monocyte-derived macrophages were treated with soluble A2t and MAP kinase signaling and downstream cytokine and chemokine production, along with effector functions, were measured. In this report, we demonstrate that A2t elicits activation of the MAP kinases, in addition to the nuclear translocation of NF-␬B, resulting in inflammatory cytokine production as well as chemokine production. Furthermore, this activation was reflected in enhanced macrophage effector function, as A2t brought about an acceleration of macrophage phagocytosis of Escherichia coli bacteria.

MATERIALS AND METHODS Isolation and culture of human blood monocytes Human peripheral blood monocytes were obtained from healthy volunteers by leukapheresis. Monocytes were further purified from mononuclear cells by Ficoll-Hypaque sedimentation followed by countercurrent centrifugal elutriation. Monocytes were resuspended at a density of 3.0 ⫻ 106 cells per millilter in DMEM (Invitrogen, Carlsbad, CA, USA) containing 2 mM L-glutamine and supplemented with 10 ␮g/ml gentamicin and were plated at 6 ⫻ 106/2 ml per well in 6-well plates or 3 ⫻ 106/ml per well in 12-well plates. After 4 – 6 h, heat-inactivated fetal bovine serum (HyClone, Logan, UT, USA) was added to 10%, and adherent monocytes were cultured for 7 days to enable differentiation into mature monocyte-derived macrophages.

Western blot analysis and NF-␬B transcription factor ELISA For whole cell lysates, 7- to 14-day-old monocyte-derived macrophages were treated with Annexin A2: S100A10 tetramer (isolated from bovine lung;

Biodesign International, Saco, ME, USA or United States Biologicals, Swampscott, MA, USA) at 250 ng/ml (2.6 nM) or the indicated concentration for the indicated times. Plates were then placed on ice, rinsed with ice-cold PBS, lysed directly with 500 ␮l per well of a 1:1 mixture of RIPA buffer (1⫻ PBS with 1% NP-40, 0.5% sodium deoxycholate, 0.1% SDS with 2 mM activated sodium orthovanadate and 2 mM AEBSF) and NuPAGE 4⫻ LDS sample buffer (Invitrogen). Each lysed sample was transferred to a 1.6 ml Eppendorf tube, placed in a 95°C heat block for 10 s, and genomic DNA sheared by forcing the sample through a 26-gauge needle three times. For nuclear and cytoplasmic lysates, cells were treated with A2t for the indicated times, with or without 30’ pretreatment with p38 MAP kinase inhibitor SB203580 (5 ␮M; Sigma Chemical Company, St. Louis, MO, USA) or proteasome inhibitor MG-132 (10 ␮M; Sigma). Plates were washed with ice-cold PBS, and cells collected by scraping into 800-␮l ice-cold PBS. Cells were pelleted at 1000 g for 2 min., resuspended in 300 ␮l of solution A (50 mM HEPES, pH 7.9, 10 mM KCl, 1 mM EGTA, 1 mM EDTA, 2 mM sodium orthovanadate and 2 mM AEBSF) and left on ice for 15 min. Triton X-100 was added to 1%, cells were briefly vortexed, and nuclei were spun down at 5000 g for 3 min. Supernatant (cytoplasmic extract) was collected and placed at – 80°C until use. Pelleted nuclei were briefly washed in ice-cold solution A, resuspended in 50 ␮l solution C (50 mM HEPES, pH 7.9, 400 mM NaCl, 1 mM EGTA, 1 mM EDTA, 2 mM sodium orthovanadate and 2 mM AEBSF) and placed on a rocking platform at 4°C for 15 min. Debris was pelleted at 15,000 g at 4°C for 10 min., and nuclear lysates were stored at – 80°C until use. Protein concentrations of both cytoplasmic and nuclear lysates were determined by BCA assay (Pierce, Rockford, IL, USA). Ten micrograms of each nuclear extract was used per well in the TransAM p65 NF-␬B transcription factor ELISA per manufacturer’s protocol (Active Motif, Carlsbad, CA, USA). Briefly, nuclear extracts were incubated in wells coated with oligos comprising the NF-␬B consensus site. Plates were washed, incubated with anti-p65 primary antibody for 1 h, followed by 1 h incubation with HRP-labeled secondary antibody, and colorimetric detection. OD450 values presented were subtracted for the absorbance of a blank well, to which no nuclear lysate was added but which was processed otherwise identically to the other samples. For electrophoretic analysis, samples were supplemented with 10% ␤-mercaptoethanol, heated again for 10 min at 95°C, and loaded as indicated onto 4 –12% NuPAGE bis-Tris gradient gels with accompanying molecular weight markers (RPN800; GE Healthcare Life Sciences, Piscataway, NJ, USA). Samples were run at 125 V for 1 h using 1⫻ MES NuPAGE running buffer (Invitrogen). Proteins were then transferred to a nitrocellulose membrane using 1⫻ NuPAGE transfer buffer (125 V for 1.5 h) and membranes were blocked for 15 min with 5% BSA in TBST (50 mM Tris, 150 mM NaCl, 0.05% Tween 20). Primary antibodies were added at the following concentrations for overnight incubation: anti-phospho-ERK1/2 (CST, #9101), 1:1000; anti-ERK (CST, #9102), 1:1000; anti-phospho-p38 (Cell Signaling Technology (CST), Danvers, MA, USA, #9211), 1:500; anti-p38 (CST, #9212), 1:1000; anti-phospho-JNK/ SAPK (CST, #9251), 1:500; anti-JNK/SAPK (CST, #9252), 1:1000; anti-p65 NF-␬B (Santa Cruz Biotechnology, Santa Cruz, CA; #sc-109), 1:200. Blots were then washed, incubated with 1:2000 HRP-conjugated secondary antibodies (Amersham) in blocking buffer, developed using ECL Plus (Amersham), and detected using the Fuji LAS-3000 (Fujifilm, Tokyo, Japan).

Gel purification of p11, p36, and A2t A2t, p11, and p36 were isolated by electrophoresis. 40 ␮g native A2t was heated for 10 min at 70°C in 1⫻ LDS loading buffer (Invitrogen) and 10 ␮g loaded per lane in each of four lanes. Excision of bands was carried out using Coomassie stained neighbor lanes as guides, and proteins were eluted overnight in 10 mM Tris-HCl, pH 7.5, with 0.1% SDS at 37°C with shaking. Proteins were concentrated and exchanged into buffer without SDS using Centriprep YM10 and Microcon YM10 concentrators (Millipore, Billerica, MA, USA) before employing a final G25 spin column to remove particulates. The “mock elution” negative control sample was obtained by performing the identical process on a gel slice from an unloaded lane. Finally, the protein concentrations of each eluted fraction were measured using the BCA assay (Pierce), and the mock eluted sample was adjusted to a volume similar to the adjusted samples.

Swisher et al. Annexin A2 induces macrophage activation

1175

Genomic DNA isolation Genomic DNA was isolated from 6 ⫻ 106 macrophages cultured in one well of a six-well plate using the Illustra Tissue and Cells Genomic Prep Mini-Spin Kit (GE Healthcare, New York, NY, USA).

RNA isolation and analysis 6 ⫻ 106 macrophages cultured in six-well plates were treated with 50 –5000 ng/ml A2t, AV, p11 (S100A10), or p36 for the times indicated. Total RNA was extracted using RNeasy Mini Kit spin columns (Qiagen, Valencia, CA, USA) per the manufacturer’s instructions. For RT-PCR, 1 ␮g RNA from each sample was placed in a 20 ␮l reaction containing 1⫻ reaction buffer (50 mM Tris-HCl, pH 8.3, 50 mM KCl, 6 mM MgCl2, and 2% glycerol supplemented with 1 mM DTT) with 0.4 ␮M each dNTP (Promega, Madison, WI, USA), 250 ng random hexamer primers, 1 ␮l RNaseout (Invitrogen) and 1 ␮l SuperScript II (Invitrogen). 25 ng cDNA was then used in each reaction for PCR amplification of the following genes using the primers below in a series of gradient PCR reactions (94°C for 4 min, followed by 30 cycles at 94°C for 30 s, 55– 62°C for 30 s, 72°C for 30 s, and finally extension at 72°C for 10 min) to ensure that each set yields a single amplicon per primer set, as well as to ascertain the optimum temperature conditions for real-time PCR. For real-time PCR, cDNA corresponding to 25 ng RNA was used for each reaction in duplicate. PCR reactions were conducted using SYBR Green and detection with the real-time PCR System 7900 (Applied Biosystems, Foster City, CA, USA). Reaction conditions were as follows: 15 s at 95°C for melting, 30 s at 62°C for annealing, 30 s at 72°C for extending, for a total of 40 cycles. Primer sequences were as follows: ␤-actin forward: TCGTCGACAACGGCTCCGGCATGTGC; ␤-actin reverse: TTCTGCAGGGAGGAGCTGGAAGCAGC; IL-6 forward: CACAGACAGCCACTCACCTC; IL-6 reverse: TTTTCTGCCAGTGCCTCTTT; IL-1␤ forward: GGACAAGCTGAGGAAGATGC; IL-1␤ reverse: TCGTTATCCCATGTGTCGAA; TNF-␣ forward: ATGAGCACTGAAAGCATGATCC; TNF-␣ reverse: GAGGGCTGATTAGAGAGAGGTC; TGF-␤ forward: GGCCAGATCCTGTCCAAGC; TGF-␤ reverse: GTGGGTTTCCACCATTAGCAC; COX-2 forward: CTGGCGCTCAGCCATACAG; COX-2 reverse: ACACTCATACATACACCTCGGT; IP-10 forward: GGAACCTCCAGTCTCAGCACC; IP-10 reverse: CAGCCTCTGTGTGGTCCATCC; Gro-␣ fwd: ATGGCCCGCGCTGCTCTCTCC; Gro-␣ rev: GTTGGATTTGTCACTGTTCAG; ICAM-1 fwd: TCTGTGTCCCCCTCAAAAGTC; ICAM-1 rev: GGGGTCTCTATGCCCAACAA; IL-8 fwd: AGGTGCAGTTTTGCCAAGGA; IL-8 rev: TTTCTGTGTTGGCGCAGTGT. Gene expression was determined using the relative quantification formula: ⌬⌬CT ⫽ (CT Target ⫺ CT actin)Test ⫺ (CT Target ⫺ CT actin)Control where CT is the fractional cycle number that reaches a fixed threshold, CTest is the test of interest, and CControl is the reference control (RNA from untreated cells). ⌬CT is the difference between gene expression in treated cells and reference control cells. The fold increase was calculated using 2⫺⌬⌬CT. For certain reactions, genomic DNA (1 ng [(⬃286 copies of a single copy gene, if the genome of a single cell is 3.5 pg) to 50 ng (⬃14,300 copies)] was titrated in the same plate in order to obtain a standard curve. Copies of transcript/cell number were calculated using the cell number and RNA yield from each sample, the fraction of that sample used in the reverse transcription reaction, combined with the fraction of the RT reaction that was used as the real-time substrate.

Cytokine array 1.2 ⫻ 107 macrophages (from each of two separate donors) were either treated with 250 ng/ml A2t or an equal volume of PBS in 1.6 ml DMEM and incubated for 24 h. Supernatants were then removed and clarified with a 20,000 g spin (10 s at 4°C), and applied to each of two separate cytokine array membranes (R&D Systems, Minneapolis, MN, USA). Membranes were processed per manufacturer’s instructions. Chemiluminescence was detected using the Fuji LAS-3000 (Fujifilm) and quantified using Image Gauge software, version 4.22 (Fujifilm). Manual background subtraction was performed prior to quantitation, and “fold change” was calculated as the difference between the average values for both cytokine spots on A2t-treated vs. untreated arrays.

Cytokine measurement by Luminex assay 6.0 ⫻ 106 macrophages (from each of three separate donors) were either treated with 250 ng/ml A2t for 3.5, 7, or 24 h, or an equal volume of PBS for

1176


24 h (untreated control). Supernatants were then removed and clarified with a 20,000 g spin (10 min at 4°C). Cytokine assays were performed in duplicate using human cytokine 10-plex antibody bead kit (Biosource International, Camarillo, C, USA) and a Luminex 100 analyzer (Luminex Corporation, Austin, TX, USA) for IL-1␤, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, gamma interferon (IFN-␥), tumor necrosis factor-␣ (TNF-␣), and GM-CSF. Cytokine data analysis was performed using the MasterPlex QT quantitation software (MiraiBio, Alameda, CA, USA).

Phagocytosis and Fc gamma receptor staining For phagocytosis of latex beads, carboxylate-modified, fluorescent yellow-green latex beads (Sigma) were incubated with 2 mg/ml goat IgG (Sigma) overnight in 50 mM MES buffer, pH 5.0, at 4°C. Beads were then washed in DMEM with 10% FCS. Cells were treated for 3 h with 2.6 nM A2t, and the equivalent of 5 ␮l of a 2.5% solid solution was added to each well of a six-well plate in 800 ␮l total volume/well. Cells were allowed to incubate with the beads for the indicated times at 37°C; then beads were removed, wells were washed with 0.25% trypsin, and fresh trypsin (800 ␮l/well) was added for 30 min at RT. Macrophages were washed off of the plates with DMEM with 10% FCS, spun for 7 min at 1600 rpm (4°C), and resuspended in ice-cold PBS with 1.0% paraformaldehyde for analysis on a FACSCalibur flow cytometer. For bacterial phagocytosis, E. coli strain DH5␣ transfected with a GFP-mut3 encoding plasmid (pCM18 [24]) was a generous gift from Dr. Robert Palmer (NIDCR, National Institutes of Health, Bethesda, MD, USA). Bacteria were cultured in Luria-Bertani broth with 150 ␮g/ml erythromycin to an OD600 of 0.4 then washed with PBS and brought to a concentration of 6 ⫻ 108/ml in RPMI with 5% serum, and without antibiotics. Macrophages were washed with this same media, and bacteria seeded at a ratio of 100:1 by placing 0.5 ml of the bacteria in each well of a 12-well plate (containing 3⫻106 cells/well). Incubations were carried out for the times indicated at 37°C. Extracellular bacteria were then removed by three vigorous washings and a 10 min trypsin treatment (macrophage adherence confirmed by fluorescent microscopy), and cells were removed from the plate and resuspended in ice-cold PBS with 1% paraformaldehyde for analysis. For surface staining, adherent macrophages were washed twice with ice-cold PBS, detached using a rubber policeman into PBS with 2% FBS and 0.01% azide, and stained with mouse anti-human CD16, anti-human CD32 (BD) or goat anti-human CD64 (Santa Cruz Biotechnology). Samples were washed with 2 ml staining buffer, then counterstained with a 1:200 dilution of Alexa 647-donkey-anti-mouse (for anti-CD16 and 32) or Alexa 488- donkey-anti-goat antibody (Invitrogen) in the dark for 30 min at 4°C. Samples were washed once more, resuspended in PBS containing 1.0% paraformaldehyde, and analyzed on a FACSCalibur flow cytometer, using a gate set for whole, intact macrophages.

RESULTS A2t activates ERK, JNK, and p38 kinases in human monocyte-derived-macrophages To determine whether MAP kinase pathways are activated upon treatment of macrophages with soluble annexin A2 tetramer (A2t), their phosphorylation levels were determined using Western blot analysis. As shown in Fig. 1, macrophages treated with 2.6 nM (250 ng/ml) A2t demonstrated a rapid induction in phosphorylation of MAP kinases ERK1/2 (Fig. 1A), p38 (Fig. 1B), and JNK (Fig. 1C). The kinetics of activation of these three pathways were similar: phosphorylation could be detected within 5 min, continued to increase for 15 min, but returned to baseline levels within 30 min. Comparing this behavior with that elicited by a strong, canonical activator of MAP kinase pathways, stimulation of macrophages with LPS resulted in a more prolonged signal than that achieved by A2t (Fig. 1D); for example, activation of ERK1/2 continued to increase at a rapid rate for the first 30 http://www.jleukbio.org

Fig. 1. Annexin A2 tetramer induces phosphorylation of MAP kinases. Human monocyte-derived macrophages (6⫻106) were cultured in DMEM with FCS for 7–10 days. Cells were then treated with 2.6 nM (250 ng/ml) A2t, 2.6 nM (85 ng/ml) annexin AV, or 10 ng/ml LPS for the indicated times. Whole cell lysates were then analyzed for phospho-p42/44 MAPK (ERK1/2) (A, D), phospho-p38 MAPK (B), or phoshpo-p46 JNK (C) with phospho-specific antibodies by Western blot analysis. Antibodies against total protein were used as internal loading controls for each. Data shown are representative of eight donors.

min and reached higher overall levels, consistent with its well-documented behavior in macrophages [23]. In contrast, another member of the annexin family, annexin AV, failed to induce any activation of MAP kinases during this time frame (Fig. 1A). As the overall protein levels of these kinases did not change over time (Fig. 1A–D), these results clearly demonstrate that treatment of human primary macrophages with soluble A2t induces a rapid but transient stimulation of all three classes of MAP kinases.

A2t stimulates p65 NF-␬B activation It is also well documented that NF-␬B plays a crucial role in macrophage activation via the transcription of inflammatory mediators downstream of multiple types of stimulatory events. As MAP kinases (which are known to play an important role in macrophage activation) had been demonstrated to be activated by A2t, we next examined whether NF-␬B is activated in order to fully stimulate canonical macrophage activation. To this end, subcellular fractionation was performed and nuclear and cytoplasmic extracts from macrophages treated with A2t were probed to determine whether p65 NF-␬B had migrated from the cytosol upon A2t treatment (Fig. 2A). The blot was then reprobed for ␤-actin to demonstrate the separation of these two cellular compartments, as nuclear extracts should be free of this cytoplasmic marker. Whereas MAP kinase activation by A2t is immediate and transient, p65 NF-␬B demonstrably relocates to the nucleus upon macrophage exposure to A2t but does so more slowly and over a longer period of time, continuing to leave the cytoplasm and accumulate in the nucleus over the 2-h duration of this experiment (Fig. 2A). p65 NF-␬B translocation to the nucleus does not necessarily guarantee its activation. To address its ability to bind its target sequence directly, as well as to investigate whether NF-␬B activation was secondary to effects elicited by the more rapid MAP kinase activation, nuclear extracts from cells treated with A2t, in the presence or absence of NF-␬B or MAP kinase inhibitors, were examined in an ELISA format that measures NF-␬B consensus site binding activity for individual NF-␬B family members. Indeed, in the two donors tested, it was confirmed that p65 translocates to the nucleus, but it was also shown that it becomes activated and is available to bind its consensus site (Fig. 2B).

The delayed and prolonged timing of p65 NF-␬B mobilization to the nucleus led us to inquire whether it might be secondary to the effects elicited by the more rapid MAP kinase activation or whether A2t directly activates both pathways in a parallel but temporally distinct manner. MG132 is a proteasomal inhibitor that prevents the degradation of ⌱␬〉, thereby inhibiting the release of p65 that would normally allow it to translocate to the nucleus. Pretreatment with MG132 reduced p65 activity in A2t-treated cells to baseline levels in the first donor, although only partial p65 inhibition was achieved in macrophages from the second donor (Fig. 2B). Interestingly, SB203580, a MAP kinase inhibitor, reduced p65 activity in both donors, but this reduction did not equal that achieved with MG132 (Fig. 2B), suggesting only partial involvement of MAP kinase-

Fig. 2. A2t induces p65 NF-␬B to redistribute to the nucleus. Macrophages were treated with A2t, with or without 30-min pretreatment with the indicated inhibitors, for the indicated times (min). Cells were then fractionated to produce nuclear and cytoplasmic extracts. (A) 30 ␮g of the indicated extracts were probed for p65 NF-kB by Western blot analysis. Blot was also probed for ␤-actin, a cytoplasmic marker. (B) Nuclear extracts were tested for p65 activity by binding of plated DNA oligos bearing the NF-kB consensus site, followed by colorimetric detection of p65 by HRP-linked antibody.


1177

dependent events in NF-␬B activation. Nonetheless, these data clearly demonstrate that A2t engages both of the canonical signal transduction pathways involved in macrophage activation—the MAP kinase and NF-␬B pathways.

A2t-activated macrophages undergo a rapid but transient increase in inflammatory cytokine transcription As both MAP kinase and NF-␬B pathways are involved in inflammatory cytokine production, mRNA levels of several mediators of inflammation were measured using quantitative real-time PCR (Fig. 3). Treatment of macrophages with A2t resulted in an increase in mRNA for TNF-␣, IL-1␤, and IL-6 in all donors tested, inducing a dramatic increase in three of the four donors shown (Fig. 3A). In all four donors, IL-6 levels were found to be increased more than 50-fold above basal transcription levels at 3 h, and TNF-␣ and IL-1␤ message levels were similarly increased in 3 out of 4 donors shown (Fig. 3A). Transcript number per cell for each of these important cytokines was also calculated using genomic DNA to generate standard curves. Expressed in this manner, transcript levels ranged from 20 to 85 transcripts per cell in resting macrophages: 30 to 85 transcripts per cell for TNF-␣ and 20 to 80 per cell for IL-6 and IL-1␤, in agreement with previously published data [25]. After 1 h of exposure to 2.6 nM A2t, TNF-␣ transcripts were increased to 2.7 ⫻ 102 to 1.6 ⫻ 105 transcripts/cell, after 3 h, IL-6 transcript levels rose to 3.6 ⫻ 103 to 1.7 ⫻ 104 transcripts/cell. COX2 mRNA levels were also consistently increased, although to a lesser degree (data not shown). Transcript levels for TGF-␤1, which can play a mixed role in inflammation but is generally regarded as antiinflammatory, were either unaffected or reduced, depending on the donor (data not shown). By 6 h after treatment, however, transcript levels for all three cytokines had begun to fall precipitously and were reduced nearly to baseline levels for most cytokines and donors by 24 h (Fig. 3A). As a negative control, annexin AV was added at the same molar concentration to macrophages, and RNA was collected after 1 h to examine whether it had any effect on inflammatory gene induction (Fig. 3B). Consistent with its inability to activate MAP kinases (Fig. 1A) and p65 NF-␬B translocation (data not shown), annexin AV failed to induce a meaningful increase in mRNA for any of the genes examined. Biologically relevant concentrations of A2t may well be highly variable depending on the donor, as well as the immune status of the individual [18]. Therefore, the effect of A2t concentration on TNF-␣ mRNA levels in treated macrophages was examined to determine their range of responsiveness to A2t, as well as to examine whether higher doses might result in inhibition by some mechanism of interference. As might be expected from Fig. 3A, the level of TNF-␣ induction varied widely between donors. In addition, the dose responsiveness (or EC50 for TNF-␣ induction) exhibits donor variability as well. As shown in Fig. 3C, a semilogarithmic plot of dose vs. mRNA levels reveals a roughly linear relationship between the concentration of A2t used and the response observed between 250 ng/ml and 2.5 ␮g/ml (2.6 and 26 nM, respectively) in donors 6 and 8, as evidenced by TNF-␣ mRNA levels. (Data are shown in two plots in order to accommodate the differences 1178


between the magnitude of response between donors.) Macrophages from donor 5 appear to reach saturation at a slightly higher A2t concentration (⬃5 ␮g/ml or 52 nM), whereas donor 8 has not reached saturation by 10 ␮g/ml. Interestingly, A2t appears to interfere with its own activity at higher concentrations in the three donors for which A2t reached saturation in this experiment. Most importantly, the concentration of A2t used in this study (250 ng/ml or 2.6 nM) is within the linear range of demonstrated biological responsiveness in all donors. Thus, the effects elicited by this concentration would be expected to be representative of a broad range of A2t concentrations in vivo. To examine whether these effects are only elicited by A2t or if they can be induced by Annexin A2 (p36) or S100A10 (p11) on its own, the tetramer was subjected to denaturing LDSPAGE electrophoresis, followed by excision and elution, in order to purify the different components and test them for their ability to activate inflammatory gene transcription (Fig. 3D). A2t was also purified as a positive control, and material derived from a “mock” elution performed on an empty piece of acrylamide gel was used as a negative control. The data demonstrate that while purified tetramer retained the ability to induce TNF-␣ transcription (Fig. 3D), neither p36 nor p11 could do so on its own (Fig. 3D), suggesting that soluble annexin A2 (or S100A10) activates macrophages only in the context of the whole annexin A2 tetramer.

A2t increases macrophage chemokine secretion As A2t stimulates inflammatory cytokine mRNA production, presumably by way of MAP kinase and NF-␬B activation, it was of interest to investigate the downstream effects of these initial events in terms of cytokine secretion, as this represents an important mechanism by which macrophages play their crucial role in innate and adaptive immunity. To investigate multiple targets simultaneously, supernatants from A2t treated or untreated macrophages were pooled from two donors at 24 h and applied to nitrocellulose arrays containing antibodies to 36 different cytokines, chemokines, and immunomodulatory proteins (Fig. 4A and Table 1). Untreated human monocyte-derived macrophages were found to secrete detectable levels of interferon-gamma inducible protein 10 (IP10; also known as CXCL10), the IL-1 receptor agonist (IL-1RA), macrophage inhibitory factor (MIF), MCP-1, growth-regulated oncogene ␣ (Gro-␣ or CXCL1), soluble intercellular adhesion molecule-1 (sICAM-1), interleukin 8 (IL-8), and serpin E1. Most of these have previously been documented as proteins constitutively expressed by human monocyte-derived macrophages. For the most part, macrophages treated for 24 h with A2t exhibited increased secretion of many of these, with the most marked increases seen with three chemokines: Gro-␣ (CXCL1), IP10 (CXCL10), and IL-8, as well as the soluble receptor sICAM-1 (Fig. 4). In contrast with untreated macrophages, A2t-treated macrophages secreted IL-6, RANTES (CCL5), and complement protein C5a at detectable levels at 24 h (Fig. 4A and Table 1). In addition, a small but measurable reduction in secretion of MCP-1 was observed (Fig. 4A and Table 1). Although A2t induced the production of several chemokines, A2t itself was not found to http://www.jleukbio.org

Fig. 3. A2t induces the rapid and efficient transcription of inflammatory cytokines. (A) Macrophages from four individual donors were treated with 2.6 nM (250 ng/ml) A2t or (B) 2.6 nM AV for the indicated times. Total RNA was isolated, and quantitative RT-PCR was used to assess transcript levels of the indicated cytokines. (C) Increase in TNF-␣ transcripts after 1 h of treatment with the indicated concentrations of A2t in each of four individual donors. (D) A2t was separated into its individual components, p11 (S100A10), and p36 (annexin A2) by electrophoresis, recovered by elution, and tested for their ability to induce transcription of inflammatory cytokines. The intact tetramer was also eluted and used as the positive control.


1179

Fig. 4. Increased expression and induction of chemokines and cytokines in A2t-treated macrophages. (A) Clarified supernatants from untreated human monocyte-derived macrophages (6⫻106 from each of two donors) or from cells treated with 2.6 nM (250 ng/ml) A2t for 24 h were incubated with cytokine arrays (R& D Systems) and detected with HRP-conjugated secondary antibodies. The names of protein targets of interest are inserted for reference, and fold changes between treated and untreated samples are listed in Table 1. (B) Induction of genes identified in the array are confirmed by real-time PCR in four individual donors, treated with 2.6 nM A2t for 1 h. Comparison of IL-1␤ (C), TNF-␣ (D), and IL-6 (E) protein levels induced by 2.6 nM A2t is shown for indicated times (Luminex). Data are presented as the mean ⫾ SE of duplicate wells from three pooled donors.

be chemotactic when tested in standard chamber assays at the concentrations investigated in this work (data not shown). Macrophages from all donors tested thus far have responded to A2t with the induction of the same genes, although there is a certain degree of variability in the magnitude of these responses (Fig. 3). The cytokine array was used to explore the A2t-dependent secretion of chemokines and cytokines in a pooled sample, and thus, it was of interest to examine the 1180


donor-to-donor variability in some of the targets identified by this array. To this end, primers were designed for four of these targets and RNA from the four donors measured for inflammatory cytokines TNF-␣, IL-6, and IL-1␤ in Fig. 3 were examined for the transcription of some of these newly identified genes: IP10, IL-8, ICAM-1, and Gro-␣ (Fig. 4B). In agreement with the results of the cytokine array, it was found that all four of these targets were indeed induced by A2t in all four donors, http://www.jleukbio.org

TABLE 1. Fold Changes for Proteins Present in A2t-induced or Uninduced Supernatants as Indicated by Cytokine Array

Protein control 1 control 2 RANTES C5a IP10 IL-1RA MCP-1 MIF Gro-␣ (CXCL-1) CCL-1 IL-6 sICAM-1 serpin E1 IL-8 control 3

Fold change (A2t/control) 0.8 0.9 2.8 1.6 3.5 0.8 0.72 0.78 4.5 1.5 20.3 2.2 0.9 3.4 1.1

Image Gauge software (Fujifilm) was used to quantitate the (background subtracted) intensity of signal on positive positions of the R&D Systems cytokine array shown in Fig. 4. MIF, macrophage inhibitory factor; Gro-␣, growth regulated oncogene ␣; sICAM, soluble intercellular adhesion molecule-1.

with similar kinetics to those seen for IL-6 and IL-1␤, peaking (in most cases) at 3 h, then dropping down to basal transcription levels by 24 h (Fig. 4B). Early transcriptional events downstream of A2t treatment indicate vigorous synthesis of TNF-␣ and IL-1␤, although mRNA levels begin to drop significantly by 6 h (Fig. 3A). IL-6 transcription appeared to be more universally elevated, and message levels were maintained longer than was observed for these other two, and so it was perhaps not surprising that although IL-6 was picked up by the cytokine array, IL-1␤ was barely detectable and TNF-␣ was undetectable in A2t treated supernatants collected at 24 h (Fig. 4A). Presumably, uptake, degradation, and/or the presence of soluble receptor forms could account for these apparent discrepancies. To elucidate this point, Luminex bead assays were used to examine earlier time points and to confirm that the initial burst in the transcription of these cytokine genes does indeed result in the production of these inflammatory mediators. Analysis of 3-, 7-,

and 24-h supernatants from A2t-treated macrophages revealed that TNF-␣ and IL-1␤ proteins are actively synthesized, processed, and secreted within 3 h (Fig. 4C–E), and thus, other factors (such as uptake, degradation, soluble receptors, or array sensitivity) must be responsible for the lack of detectable TNF-␣ and IL-1␤ in the 24-h supernatants (Fig. 4A).

Increase in phagocytosis of GFP-labeled E. coli in A2t-treated macrophages As macrophages also play a crucial role in the direct removal of pathogens from the site of infection, it was of interest to determine whether A2t could invoke an improvement in the ability of macrophages to engulf foreign objects. To test this important effector function, A2t-treated macrophages were examined for their relative efficiency in ingesting fluorescent latex beads or GFP-expressing E. coli DH5␣. It was found that phagocytosis of latex beads, whether opsonized or untreated, was not consistently augmented (Fig. 5A) in A2t-treated macrophages. In fact, in some cases, phagocytosis of opsonized latex beads was modestly decreased in these cells, depending on the length of A2t treatment (Fig. 5A and data not shown). In contrast, enhanced bacterial phagocytosis of GFP-expressing E. coli was demonstrated by A2t-treated macrophages at all time points and was consistent across all donors tested (Fig. 5B.) As it has been found that TNF-␣ can temporarily decrease Fc receptor expression in macrophages [26], expression of CD16, CD32, and CD64 was examined in A2t-treated and untreated macrophages used for these phagocytosis experiments. It was found that the relative efficiency of latex bead phagocytosis corresponded with the expression of Fc receptors, especially CD16 and CD32 (Fig. 5C).

DISCUSSION It has been thought for some time that annexin A2 is predominantly an anti-inflammatory agent largely because of its similarities to annexin A1, which has been amply demonstrated to play such a role [4]. The frequent coregulation of these two proteins also lends credence to this hypothesis. On the other hand, data exist to support a role for A2t in the proinflammatory process: as it is a receptor for several viruses and bacteria

Fig. 5. Effect of soluble A2t treatment on macrophage phagocytosis. (A and B) Human monocyte-derived macrophages were treated with 2.6 nM A2t (dotted line) for 3 h and tested against untreated macrophages (solid line) for their ability to ingest opsonized fluorescent beads (A) or GFP-expressing E. coli DH5a (B). (C) 3 h A2t treated (black) and untreated (gray) macrophages were stained with mouse IgG1 against CD16, CD32, or goat polyclonal against CD64, followed by Alexa 488-donkey-anti-mouse or Alexa 488-donkey-anti-goat secondary antibody. All data were adjusted for isotope controls.


1181

[4, 16], the cell may have evolved a mechanism to shed it as a soluble decoy. The observation that plasma levels of A2t are elevated in individuals fighting certain infections supports this hypothesis [18]. More importantly, A2t has recently been shown to activate the ERK MAP kinases in stromal cells [19]. As MAP kinases are key regulators of macrophage activation [27], it is reasonable to assume that A2t might stimulate this pathway, and thus, cells shedding A2t could capitalize on this additional function as an inflammatory mediator. Indeed, we have shown for the first time that A2t activates macrophages, stimulating MAP kinase and NF-␬B signaling pathways and inducing the production of three key cytokines involved in the establishment of inflammation: TNF-␣, IL-1␤, and IL-6. These important but potentially damaging inflammatory mediators taper off quickly. Likewise, soluble A2t induces a transient wave of chemokine synthesis as well, presumably capable of inducing chemotaxis and activating multiple leukocyte subpopulations. In addition, A2t-induced macrophage activation results in increased bacterial phagocytosis. The data suggest that A2t induces an inflammatory burst in macrophages but that it quickly begins to resolve, acting perhaps more as an initial beacon toward the site of infection than an unchecked mediator of massive inflammation. Increased phagocytic efficiency could then, in turn, translate into an increased antigenpresenting capacity induced by soluble A2t, affecting the adaptive immune response as well. The panel of chemokines that was induced as a result of A2t treatment suggests that it plays a role in the recruitment of several key leukocyte populations to the site of insult, as well as in their activation upon arrival. IL-8 is a chemoattractant for neutrophils, as well as a neutrophil activating protein [28] and induces the adherence of monocytes to the endothelium [29]. IP10 attracts monocytes [30], NK cells [31], and activated T lymphocytes [32], which is especially intriguing given the very recent finding that annexin A1, arguably the closest member of the family to annexin A2 in structure and function, plays an unprecedented role in heightening the amplitude of activated T cell signaling through the TCR [33]. Moreover, s-ICAM-1 represents the soluble form of a multifunctional protein that can act as a docking protein for viruses [34] that also activates the immune system by recruiting neutrophils and inducing their degranulation [35], as well as activating macrophages [36]. These activated macrophages can, in turn, become a source of s-ICAM-1 by shedding it from their surface [37]. Likewise, as A2t has been found to be shed or secreted from a growing number of cell types [18, 21, 38 – 40] and as the recently cloned receptor for A2t appears to be expressed in many leukocyte subsets [22] and MAP kinases and NF-␬B play a central role in the activation of many or all of them [27], A2t may represent a paracrine or even autocrine factor in the establishment of an immune response. Phagocytosis represents an important component in the first line of defense against invading microorganisms. Consistent with a role in the early, nonspecific phase of an immune response, macrophages that are exposed to A2t prior to bacterial exposure show a heightened capacity to engulf E. coli bacteria. Presumably, this is mediated by scavenger and/or other pattern recognition receptors rather than through recognition of opsonizing antibodies, as no consistent increase in Fc 1182


surface expression or phagocytosis of opsonized beads was observed in A2t-treated macrophages (Fig. 5A and C). This is consistent with the fact that the initial engulfment of bacteria, both for the purpose of elimination. as well as for antigen presentation, takes place before the humoral response would have time to contribute in earnest. Indeed, it has been found that TNF-␣ can actually provoke a reduction of Fc receptors on the macrophage surface in the early hours of activation [26]. Additionally, it has been shown that annexin A2 binds to lipid A [41], the core of the LPS glycolipid that is found on the outer surface of gram-negative bacteria. As the annexins were named for their ability to “annex” membranes [4], it stands to reason that A2t might also facilitate a bridge between the bacterium and the macrophage surface, thereby increasing the efficiency of phagocytosis. Presumably, if A2t secretion is an element of host defense brought about by the presence of a microbial invader, it would be expected to play a role early in the time course of infection. It is understandable then that A2t treatment of naı¨ve macrophages would induce such nonspecific defenses of innate immunity. It remains to be seen what effects A2t will play downstream of these events. Our findings regarding the proinflammatory nature of A2t do not exclude the possibility of A2t participation in anti-inflammatory events. Indeed, it is entirely plausible that the effect of A2t on macrophages is context-dependent. The results reported here reflect A2t treatment of resting macrophages— cells that have not encountered an infectious agent prior to A2t exposure. Likewise, they are naı¨ve to any inflammatory mediators that might have been released in their environment due to such an encounter. Thus, it is entirely possible that A2t could provide an early warning signal upon macrophage encounter of a pathogen, but that its rapid and brief cytokine burst or activation might then give way to a dampening of cellular responsiveness to A2t or other, more damaging inflammatory mediators. Indeed, it stands to reason that host proteins involved in the establishment of inflammation would also play a role in its control and resolution, as has been observed with proteins such as TGF-␤, which is involved in the early events of leukocyte recruitment and activation [42], as well as the resolution of the inflammatory response [43]. Interestingly, like TGF-␤, A2t appears to dampen its own activity at higher concentrations in some donors (Fig. 3). If A2t plays roles in both induction and resolution of inflammation, one way to shut itself off as the reaction progresses is to start to inhibit its own action as concentrations accumulate over time. Another possibility is that soluble A2t can adopt alternative conformations that are not attainable for surface-bound A2t, providing a basis for role differentiation. The fact that A2t enjoys more than one mode of binding—protein- or lipid-dependent, and calciumdependent or -independent— underscores this possibility [44 – 47]. This seems unlikely, however, as a structural basis for discrimination between proinflammatory and anti-inflammatory roles, especially in light of recent reports that very similar signaling and transcription events have been recently observed downstream of surface A2t ligation by tPA or plasminogen [48, 49]. In addition, the role of surface-bound A2t in phagocytosis of apoptotic lymphocytes [3] highlights the fact that cell-cell interactions never occur by way of a single ligand-receptor contact, but instead, involve an interface of interacting proteins http://www.jleukbio.org

that provide a complex set of simultaneous signals. Thus, the context in which soluble A2t is perceived by the cell may allow it to elicit a variety of effects. It is reasonable to assume that soluble A2t and annexin A1 might diverge further in their functions than had been predicted due to the fact that they may bind different types of receptors altogether [22, 50]. In the case of the one receptor for annexin A2 that has been cloned, it appears to be active only as a tetramer, with its p11 binding partner responsible for ligation of the A2t receptor [22], whereas annexin A1 itself, or only a small peptide mimetic representing its N terminus, is sufficient for engagement of the formylated peptide receptors (FPRs) to which it binds [51]. However, the recent finding that annexin A1 has been found to participate in, rather than inhibit, T cell activation events downstream of MAP kinase and NF-␬B activation [33] suggests that not even the role of this canonically anti-inflammatory protein is so simple. It is also important to note that blocking experiments performed in conjunction with macrophage surface staining suggest that the receptor cloned from bone marrow stromal cells is not responsible for A2t-dependent macrophage activation reported in this work (data not shown.) Therefore, other receptors for A2t are likely to exist. However, as the cloned A2t receptor is expressed on T lymphocytes [19, 20, 22], its demonstrated coupling to the MAP kinase cascade (and possible downstream activation of NF-␬B) predicts that A2t may play a similar role to annexin A1 in this case. Therefore, the current studies may represent the beginning of a new understanding of annexin A2 that reveals another side to its actions through which it can augment certain immune functions while helping to maintain balance. The demonstration of macrophage activation by A2t reveals an important antimicrobial role for this soluble mediator, suggesting a complex role for A2t in innate and adaptive immunity. Investigations are under way to further elucidate the classes of pathogens or signals that may result in increased A2t secretion, as well as the consequences of that secretion during different phases of the immune response.

ACKNOWLEDGMENTS We would like to thank Manuel Osorio for his generous help with the Luminex bead assay, Nick Jakubovics and Robert Palmer for the donation of the GFP-E. coli, G. D. Roodman for providing the antibody to the A2t receptor, Sean Fitzsimmons for helpful discussions concerning real-time PCR absolute quantification, and David Frucht and Mate Tolnay for critical reading of this manuscript.

REFERENCES 1. Wu, C. C., Croxtall, J. D., Perretti, M., Bryant, C. E., Thiemermann, C., Flower, R. J., Vane, J. R. (1995) Lipocortin 1 mediates the inhibition by dexamethasone of the induction by endotoxin of nitric oxide synthase in the rat. Proc. Natl. Acad. Sci. USA 92, 3473–3477. 2. Perretti, M., Flower, R. J. (2004) Annexin 1 and the biology of the neutrophil. J. Leukoc. Biol. 76, 25–29. 3. Fan, X., Krahling, S., Smith, D., Williamson, P., Schlegel, R. A. (2004) Macrophage surface expression of annexins I and II in the phagocytosis of apoptotic lymphocytes. Mol. Biol. Cell 15, 2863–2872.

4. Gerke, V., Moss, S. E. (2002) Annexins: from structure to function. Physiol. Rev. 82, 331–371. 5. Raynal, P., Pollard, H. B. (1994) Annexins: the problem of assessing the biological role for a gene family of multifunctional calcium- and phospholipid-binding proteins. Biochim. Biophys. Acta 1197, 63–93. 6. Falcone, D. J., Borth, W., Khan, K. M., Hajjar, K. A. (2001) Plasminogenmediated matrix invasion and degradation by macrophages is dependent on surface expression of annexin II. Blood 97, 777–784. 7. Hajjar, K. A., Guevara, C. A., Lev, E., Dowling, K., Chacko, J. (1996) Interaction of the fibrinolytic receptor, annexin II, with the endothelial cell surface. Essential role of endonexin repeat 2. J. Biol. Chem. 271, 21652–21659. 8. Hajjar, K. A., Jacovina, A. T., Chacko, J. (1994) An endothelial cell receptor for plasminogen/tissue plasminogen activator. I. Identity with annexin II. J. Biol. Chem. 269, 21191–21197. 9. Cesarman, G. M., Guevara, C. A., Hajjar, K. A. (1994) An endothelial cell receptor for plasminogen/tissue plasminogen activator (t-PA). II. Annexin II-mediated enhancement of t-PA-dependent plasminogen activation. J. Biol. Chem. 269, 21198 –21203. 10. Raynor, C. M., Wright, J. F., Waisman, D. M., Pryzdial, E. L. (1999) Annexin II enhances cytomegalovirus binding and fusion to phospholipid membranes. Biochemistry 38, 5089 –5095. 11. Wright, J. F., Kurosky, A., Pryzdial, E. L., Wasi, S. (1995) Host cellular annexin II is associated with cytomegalovirus particles isolated from cultured human fibroblasts. J. Virol. 69, 4784 – 4791. 12. Wright, J. F., Kurosky, A., Wasi, S. (1994) An endothelial cell-surface form of annexin II binds human cytomegalovirus. Biochem. Biophys. Res. Commun. 198, 983–989. 13. Malhotra, R., Ward, M., Bright, H., Priest, R., Foster, M. R., Hurle, M., Blair, E., Bird, M. (2003) Isolation and characterisation of potential respiratory syncytial virus receptor(s) on epithelial cells. Microbes Infect. 5, 123–133. 14. Kirschnek, S., Adams, C., Gulbins, E. (2005) Annexin II is a novel receptor for Pseudomonas aeruginosa. Biochem. Biophys. Res. Commun. 327, 900 –906. 15. Ma, G., Greenwell-Wild, T., Lei, K., Jin, W., Swisher, J., Hardegen, N., Wild, C. T., Wahl, S. M. (2004) Secretory leukocyte protease inhibitor binds to annexin II, a cofactor for macrophage HIV-1 infection. J. Exp. Med. 200, 1337–1346. 16. Ryzhova, E. V., Vos, R. M., Albright, A. V., Harrist, A. V., Harvey, T., Gonzalez-Scarano, F. (2006) Annexin 2: a novel human immunodeficiency virus type 1 Gag binding protein involved in replication in monocytederived macrophages. J. Virol. 80, 2694 –2704. 17. Romisch, J., Schuler, E., Bastian, B., Burger, T., Dunkel, F. G., Schwinn, A., Hartmann, A. A., Paques, E. P. (1992) Annexins I to VI: quantitative determination in different human cell types and in plasma after myocardial infarction. Blood Coagul. Fibrinolysis 3, 11–17. 18. Ulvestad, E., Kristoffersen, E. K., Jensen, T. S., Matre, R. (1994) Identification of a soluble Fc gamma-binding molecule (annexin II) in human serum using a competitive ELISA. APMIS 102, 667– 673. 19. Li, F., Chung, H., Reddy, S. V., Lu, G., Kurihara, N., Zhao, A. Z., Roodman, G. D. (2005) Annexin II stimulates RANKL expression through MAPK. J. Bone Miner. Res. 20, 1161–1167. 20. Menaa, C., Devlin, R. D., Reddy, S. V., Gazitt, Y., Choi, S. J., Roodman, G. D. (1999) Annexin II increases osteoclast formation by stimulating the proliferation of osteoclast precursors in human marrow cultures. J. Clin. Invest. 103, 1605–1613. 21. Takahashi, S., Reddy, S. V., Chirgwin, J. M., Devlin, R., Haipek, C., Anderson, J., Roodman, G. D. (1994) Cloning and identification of annexin II as an autocrine/paracrine factor that increases osteoclast formation and bone resorption. J. Biol. Chem. 269, 28696 –28701. 22. Lu, G., Maeda, H., Reddy, S. V., Kurihara, N., Leach, R., Anderson, J. L., Roodman, G. D. (2006) Cloning and characterization of the annexin II receptor on human marrow stromal cells. J. Biol. Chem. 281, 30542– 30550. 23. Rao, K. M. (2001) MAP kinase activation in macrophages. J. Leukoc. Biol. 69, 3–10. 24. Hansen, M. C., Palmer, R. J., Jr., Udsen, C., White, D. C., Molin, S. (2001) Assessment of GFP fluorescence in cells of Streptococcus gordonii under conditions of low pH and low oxygen concentration. Microbiology 147, 1383–1391. 25. Cvetanovic, M., Ucker, D. S. (2004) Innate immune discrimination of apoptotic cells: repression of proinflammatory macrophage transcription is coupled directly to specific recognition. J. Immunol. 172, 880 – 889. 26. Liao, G., Simon, S. R. (1994) Temporal down-regulation of Fc gamma RIII expression and Fc gamma receptor-mediated phagocytosis in human monocyte-derived macrophages induced by TNF-alpha and IL-1 beta. J. Leukoc. Biol. 55, 702–710.


1183

27. Ashwell, J. D. (2006) The many paths to p38 mitogen-activated protein kinase activation in the immune system. Nat. Rev. Immunol. 6, 532–540. 28. Baggiolini, M., Dewald, B., Moser, B. (1994) Interleukin-8 and related chemotactic cytokines–CXC and CC chemokines. Adv. Immunol. 55, 97–179. 29. Gerszten, R. E., Garcia-Zepeda, E. A., Lim, Y. C., Yoshida, M., Ding, H. A., Gimbrone, M. A., Jr., Luster, A. D., Luscinskas, F. W., Rosenzweig, A. (1999) MCP-1 and IL-8 trigger firm adhesion of monocytes to vascular endothelium under flow conditions. Nature 398, 718 –723. 30. Taub, D. D., Longo, D. L., Murphy, W. J. (1996) Human interferoninducible protein-10 induces mononuclear cell infiltration in mice and promotes the migration of human T lymphocytes into the peripheral tissues and human peripheral blood lymphocytes-SCID mice. Blood 87, 1423– 1431. 31. Taub, D. D., Sayers, T. J., Carter, C. R., Ortaldo, J. R. (1995) Alpha and beta chemokines induce NK cell migration and enhance NK-mediated cytolysis. J. Immunol. 155, 3877–3888. 32. Loetscher, M., Gerber, B., Loetscher, P., Jones, S. A., Piali, L., ClarkLewis, I., Baggiolini, M., Moser, B. (1996) Chemokine receptor specific for IP10 and mig: structure, function, and expression in activated T-lymphocytes. J. Exp. Med. 184, 963–969. 33. D'Acquisto, F., Merghani, A., Lecona, E., Rosignoli, G., Raza, K., Buckley, C. D., Flower, R. J., Perretti, M. (2007) Annexin-1 modulates T-cell activation and differentiation. Blood 109, 1095–1102. 34. Bella, J., Rossmann, M. G. (1999) Review: rhinoviruses and their ICAM receptors. J. Struct. Biol. 128, 69 –74. 35. Barnett, C. C., Moore, E. E., Moore, F. A., Biffl, W. L., Partrick, D. A. (1996) Soluble intercellular adhesion molecule-1 provokes polymorphonuclear leukocyte elastase release by CD18. Surgery 120, 395– 401 (discussion 401-2). 36. Schmal, H., Czermak, B. J., Lentsch, A. B., Bless, N. M., Beck-Schimmer, B., Friedl, H. P., Ward, P. A. (1998) Soluble ICAM-1 activates lung macrophages and enhances lung injury. J. Immunol. 161, 3685–3693. 37. Melis, M., Pace, E., Siena, L., Spatafora, M., Tipa, A., Profita, M., Bonanno, A., Vignola, A. M., Bonsignore, G., Mody, C. H., et al. (2003) Biologically active intercellular adhesion molecule-1 is shed as dimers by a regulated mechanism in the inflamed pleural space. Am. J. Respir. Crit. Care Med. 167, 1131–1138. 38. Rich, S. A. (1995) De novo synthesis and secretion of a 36-kD protein by cells that form lupus inclusions in response to alpha-interferon. J. Clin. Invest. 95, 219 –226. 39. Ma, A. S., Bell, D. J., Mittal, A. A., Harrison, H. H. (1994) Immunocytochemical detection of extracellular annexin II in cultured human skin keratinocytes and isolation of annexin II isoforms enriched in the extracellular pool. J. Cell Sci. 107, 1973–1984.

1184


40. Davis, R. G., Vishwanatha, J. K. (1995) Detection of secreted and intracellular annexin II by a radioimmunoassay. J. Immunol. Methods 188, 91–95. 41. Eberhard, D. A., Vandenberg, S. R. (1998) Annexins I and II bind to lipid A: a possible role in the inhibition of endotoxins. Biochem. J. 330, 67–72. 42. Wahl, S. M., Hunt, D. A., Wakefield, L. M., McCartney-Francis, N., Wahl, L. M., Roberts, A. B., Sporn, M. B. (1987) Transforming growth factor type beta induces monocyte chemotaxis and growth factor production. Proc. Natl. Acad. Sci. USA 84, 5788 –5792. 43. Wahl, S. M., Swisher, J., McCartney-Francis, N., Chen, W. (2004) TGFbeta: the perpetrator of immune suppression by regulatory T cells and suicidal T cells. J. Leukoc. Biol. 76, 15–24. 44. Brooks, N. D., Grundy, J. E., Lavigne, N., Derry, M. C., Restall, C. M., MacKenzie, C. R., Waisman, D. M., Pryzdial, E. L. (2002) Ca2⫹-dependent and phospholipid-independent binding of annexin 2 and annexin 5. Biochem. J. 367, 895–900. 45. Jost, M., Zeuschner, D., Seemann, J., Weber, K., Gerke, V. (1997) Identification and characterization of a novel type of annexin-membrane interaction: Ca2⫹ is not required for the association of annexin II with early endosomes. J. Cell Sci. 110, 221–228. 46. Gokhale, N. A., Abraham, A., Digman, M. A., Gratton, E., Cho, W. (2005) Phosphoinositide specificity of and mechanism of lipid domain formation by annexin A2–p11 heterotetramer. J. Biol. Chem. 280, 42831– 42840. 47. Konig, J., Gerke, V. (2000) Modes of annexin-membrane interactions analyzed by employing chimeric annexin proteins. Biochim. Biophys. Acta 1498, 174 –180. 48. Li, Q., Laumonnier, Y., Syrovets, T., Simmet, T. (2007) Plasmin triggers cytokine induction in human monocyte-derived macrophages. Arterioscler. Thromb. Vasc. Biol. 27, 1383–1389. 49. Ortiz-Zapater, E., Peiro, S., Roda, O., Corominas, J. M., Aguilar, S., Ampurdanes, C., Real, F. X., Navarro, P. (2007) Tissue plasminogen activator induces pancreatic cancer cell proliferation by a non-catalytic mechanism that requires extracellular signal-regulated kinase 1/2 activation through epidermal growth factor receptor and annexin A2. Am. J. Pathol. 170, 1573–1584. 50. Ernst, S., Lange, C., Wilbers, A., Goebeler, V., Gerke, V., Rescher, U. (2004) An annexin 1 N-terminal peptide activates leukocytes by triggering different members of the formyl peptide receptor family. J. Immunol. 172, 7669 –7676. 51. John, C. D., Sahni, V., Mehet, D., Morris, J. F., Christian, H. C., Perretti, M., Flower, R. J., Solito, E., Buckingham, J. C. (2007) Formyl peptide receptors and the regulation of ACTH secretion: targets for annexin A1, lipoxins, and bacterial peptides. FASEB J. 21, 1037–1046.

http://www.jleukbio.org