IL-4 promotes the formation of multinucleated giant cells ... - CiteSeerX

IL-4 promotes the formation of multinucleated giant cells from macrophage precursors by a STAT6-dependent, homotypic mechanism: contribution of E-cadherin Jose L. Moreno,*,† Irina Mikhailenko,† Mehrdad M. Tondravi,‡ and Achsah D. Keegan†,§,ⱍⱍ,1 *Department of Orthopedics, †Center for Vascular and Inflammatory Disease, §Program in Oncology, Marlene and Stewart Greenebaum Cancer Center, and ⱍⱍDepartment of Microbiology and Immunology, University of Maryland School of Medicine, Baltimore, Maryland, USA; and ‡Department of Hematopoiesis, Jerome H. Holland Laboratory for the Biomedical Sciences, American Red Cross, Rockville, Maryland, USA

Abstract: Multinucleated giant cells (MNG) are central players in the inflammatory response to foreign materials and in adverse responses to implants. IL-4 promotes the formation of MNG from bone marrow-derived precursors in vitro and participates in the development of the foreign body reaction in vivo. Therefore, we investigated the mechanism by which IL-4 promotes formation of MNG and engulfment of foreign bodies. We found that generation of MNG cells by IL-4 was dependent on cell density and expression of STAT6; macrophages derived from STAT6ⴚ/ⴚ mice were unable to form MNG in response to IL-4. No soluble factors including CCL2 or supernatants from IL-4-treated macrophages compensated for the lack of MNG cells in STAT6ⴚ/ⴚ cultures. We found that IL-4 must remain present during the full differentiation process and that STAT6ⴙ/ⴙ macrophage precursors retained their ability to differentiate into MNG over time. These MNG were able to internalize large particles efficiently, and the mononuclear STAT6ⴚ/ⴚ macrophages were unable to do so. Furthermore, we found that IL-4 induced expression of E-cadherin and dendritic cell-specific transmembrane protein in a STAT6dependent manner. E-cadherin expression was critical for the formation of MNG cells by IL-4; an anti-E-cadherin antibody prevented the formation of large MNG. In addition, we found that STAT6ⴚ/ⴚ progenitors failed to fuse with STAT6ⴙ/ⴙ, revealing the need for a homotypic interaction. Thus, IL-4 promotes the formation of MNG in a STAT6-dependent manner by regulating cell surface expression of E-cadherin, leading to homotypic cell fusion and the incorporation of large foreign bodies. J. Leukoc. Biol. 82: 1542–1553; 2007. Key Words: inflammation 䡠 cell differentiation 1542

䡠

phagocytosis

䡠

fusion

䡠

DC-STAMP

Journal of Leukocyte Biology Volume 82, December 2007

INTRODUCTION IL-4 is produced by TH2 lymphocytes, NKT cells, and by cells of the innate immune system, including mast cells, basophils, and eosinophils [1, 2]. Although it was identified originally as a stimulator of B cell proliferation, it is now clear that IL-4 can regulate proliferation, apoptosis, gene expression, and differentiation in many cell types [1– 6]. In particular, IL-4 can play a critical role in regulating the differentiation and functional activity of cells of the monocyte-macrophage lineage [4, 5, 7–14]. Such cells are derived from hematopoietic progenitors and can differentiate into several cell types with markedly distinct characteristics [7, 8, 15]. Some of these cell types are mononuclear, such as dendritic cells (DCs) and macrophages, and others fall into the multinucleated category. These include multinucleated giant cells (MNG), sometimes called foreign body giant cells (FBGC), and osteoclasts. MNG are closely related to osteoclasts and share not only a common origin from the hematopoietic-derived monocyte-macrophage lineage but also important morphological and functional characteristics. For example, both develop by the regulated fusion of mononuclear precursors and can be highly multinucleated [16]. However, their biologic activities are markedly different [8]. These cells perform highly specific functions: bone resorption by the osteoclast and elimination of extracellular foreign bodies by MNG. We showed previously that IL-4 directly prevents the receptor-activator of NF-␬B ligand (RANKL)-induced differentiation of myeloid precursors to osteoclasts and inhibits bone resorption by mature osteoclasts in a STAT6-dependent manner [4]. During these studies, we observed the formation of tartrate-resistant acid phosphatase (TRAP)-negative MNG in the cultures treated with IL-4. Thus, it would appear that IL-4 is a molecular switch between osteoclasts and MNG.

1 Correspondence: Center for Vascular and Inflammatory Diseases, University of Maryland, Baltimore, 800 W. Baltimore St., Baltimore, MD 21201, USA. E-mail: [email protected] Received January 23, 2007; revised August 10, 2007; accepted August 13, 2007. doi: 10.1189/jlb.0107058

0741-5400/07/0082-1542 © Society for Leukocyte Biology

MNG are the central players in the chronic inflammatory response to foreign materials in vivo [17, 18]. These cells have been observed at the surface of a wide variety of implants (i.e., vascular, cardiovascular, orthopedic, breast prostheses) for extended periods of time [7, 18 –20] and are thought to be responsible for a variety of biomaterial-mediated, adverse responses, which lead to degradation and implant failure [20 – 25]. IL-4 not only promotes the formation of MNG in vitro, but it also participates in the development of the foreign body reaction at the tissue-material interface in vivo [18]. However, the mechanism by which IL-4 promotes the formation of MNG is unclear. Therefore, we investigated the mechanism by which IL-4 regulates macrophage fusion during the formation of MNG. In this study, we show that IL-4 promotes the formation of MNG in a STAT6-dependent manner, in part by regulating cell surface expression of E-cadherin, leading to homotypic cell fusion and the incorporation of large foreign bodies.

MATERIALS AND METHODS Cell culture Nonadherent bone marrow (BM) mononuclear cells were isolated from femurs and tibias of 4- to 6-week-old wild-type BALB/c STAT6⫹/⫹ (Taconic Laboratories, Germantown, NY, USA) and BALB/c STAT6⫺/⫺ female mice. (Breeding pairs of STAT6⫺/⫺ mice, crossed onto the BALB/c background for more than 10 generations, were obtained from Dr. William E. Paul, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD, USA, and bred in-house.) Cells were cultured overnight at 37°C and 5% CO2 atmosphere in ␣-10 [␣-MEM supplemented with penicillin, streptomycin, and glutamine (all from BioWhittaker, Walkersville, MD, USA) and 10% heatinactivated FBS (Invitrogen, Frederick, MD, USA)] to deplete adherent stromal cells. The nonadherent mononuclear cells were isolated over Ficoll-Hypaque density gradients as described [26]. These nonadherent BM mononuclear cells were cultured in the presence of recombinant cytokines to generate MNG, and in other experiments, they were cultured initially for 48 h in the presence of 20 ng/ml recombinant mouse (rm)M-CSF (R&D Systems Inc., Minneapolis, MN, USA) to generate M-CSF-dependent macrophages. In all of the cultures, cells were allowed to adhere for 2 h prior to the addition of cytokines. The cells were cultured for 5 days and then fixed and stained in one step with 1% Crystal violet in 95% ethanol, washed with water, and air-dried before analysis. Cells were plated at 2 ⫻ 106 cells/ml, unless otherwise stated.

Analysis of MNG formation To induce MNG formation, BM cells from BALB/c STAT6⫹/⫹ and STAT6⫺/⫺ mice were plated on glass coverslips (Nalge Nunc International, Naperville, IL, USA) at a density of 2 ⫻ 106 cells/ml with 20 ng/ml rmM-CSF and 10 ng/ml rmIL-4 (R&D Systems Inc.). After 5 days in culture, the cells were washed with PBS, fixed with 4% formaldehyde in PBS, and stained with rhodamine phalloidin and Hoechst dye (both from Molecular Probes, Eugene, OR, USA), according to the manufacturer’s instructions. Digital pictures at low power (10⫻ and 20⫻ objectives) were taken with a microscope (ECLIPSE E400, Nikon, Japan) under UV light for the quantification of the number, area, and fusion index of the MNG. Cell area was calculated over the images captured with the 10⫻ objective by delineating all the MNG cells contained on the field. All the pictures were analyzed using image analysis software (Bioquant Image Analysis, Nashville, TN, USA). The fusion index was determined as described previously [7] by dividing the total number of nuclei within MNG, defined as cells containing greater than or equal to three nuclei/cell, by the number of total nuclei counted and multiplying by 100. Three (10⫻ objective) and five (20⫻ objective) fields per well were counted in duplicate. The images from three independent experiments were evaluated. The unpaired Student’s t-test was used for statistical analysis between wild-type and STAT6⫺/⫺ groups.

Flow cytometry M-CSF-dependent macrophages from wild-type STAT6⫹/⫹ and STAT6⫺/⫺ mice were cultured for 48 h in the presence of 20 ng/ml rmM-CSF, in the presence or absence of 10 ng/ml rmIL-4. Cells were exposed to Fluoresbrite AE YG™ microspheres with a diameter of 1 ␮m for 30 min according to the manufacturer’s instructions (1⫻108 microspheres/ml, Polysciences Inc., Warrington, PA, USA). The cells were washed extensively with PBS. Bead uptake was quantified using FACScan and analyzed by CellQuest software, Version 3.3 (Becton Dickinson Immunocytometry Systems, Bedford, MA, USA).

Antibodies mAb against E-cadherin, recognizing the extracellular domain near the transmembrane spanning region (MAB36), was purchased from BD Biosciences (San Diego, CA, USA), the rabbit polyclonal antibody against the C-terminus of E-cadherin and the goat anti-biotin-HRP were purchased from Cell Signaling Technology (Danvers, MA, USA), and the rabbit polyclonal anti-DCspecific transmembrane protein (STAMP) antibody was purchased from Cosmo Bio Co., Ltd. (Tokyo, Japan). The sheep anti-mouse IgG-HRP conjugate was purchased from Amersham (Buckinghamshire, UK), and the goat anti-rabbit IgG-HRP conjugate was obtained from Bio-Rad Laboratories (Hercules, CA, USA). As control for protein loading, rabbit anti-heat shock protein 86 (HSP86) or mouse mAb against tubulin-␣ (DM1A) was purchased from Lab Vision (Fremont, CA, USA). The mAb against the amino-terminal region of Ecadherin known to block homotypic interaction (ECCD-1) [27] was purchased from Takara Bio Inc. (Otsu, Japan).

Western blot, cell surface biotinylation, and immunoprecipitation M-CSF-dependent macrophages from STAT6⫹/⫹ and STAT6⫺/⫺ mice were cultured in the presence of 20 ng/ml rmM-CSF, in the presence or absence of 10 ng/ml rmIL-4 for various times. Subsequently, the cells were lysed in lysis buffer (50 mM Tris, pH 8.0, 150 mM NaCl, 1% Nonidet P-40, 5 mM EDTA, 10 mM Na pyrophosphate, 50 mM NaF, 0.25% Na deoxycholate, 1 mM Na orthovanadate, 1 mM PMSF, pepstatin, leupeptin, and aprotinin) and clarified by centrifugation. Protein concentration was determined in the samples by bicinchoninic acid assay (Pierce, Rockford, IL, USA), and equal amounts of protein were separated on SDS-PAGE before transfer to polyvinylidene difluoride membrane (Immobilon-P, Millipore, Bedford, MA, USA). Membranes were then blocked overnight and probed with antibodies as indicated. The membranes were washed with TBST extensively and developed using Supersignal West Pico chemiluminescent substrate (Pierce). The blots were stripped and reprobed with anti-HSP-86 or anti-tubulin-␣ (DM1A) as control. Biotinylation of macrophages was performed with Pierce EZ-Link sulfoNHS-biotin reagents according to the manufacturer’s instructions. Briefly, M-CSF-dependent macrophages from BALB/c STAT6⫹/⫹ mice were cultured for 48 h in the presence of 20 ng/ml rmM-CSF, in the presence or absence of 10 ng/ml rmIL-4. Cells (⬎95% viable by trypan blue dye exclusion) were washed three times with ice-cold PBS supplemented with 1 mM MgCl2 and 0.1 mM CaCl2. Sulfo-NHS-LC-biotin reagent diluted in PBS (1 mg/mL) was added to cells and incubated for 30 min at 4°C with occasional swirling. The reaction was quenched by treating the cells with PBS containing 100 mM glycine for 15 min at 4°C followed by extensive washing. The quenching reaction was repeated two times. Subsequently, the cells were lysed in lysis buffer and analyzed by immunoprecipitation and Western blotting. For immunoprecipitation, equal amounts of protein were incubated with the primary antibody or with neutravidin-agarose beads. For precipitations with primary antibody, the complexes were collected by immunoprecipitation with protein G or A beads (Invitrogen), according to the manufacturer’s instructions. Immunoprecipitates were separated on SDS-PAGE and analyzed by Western blotting as described above. The membranes were stripped and reprobed with control antibodies as appropriate.

Blocking of MNG formation by anti-E-cadherin BM cells were seeded at 2 ⫻ 106 cells/ml. After 2 h, 20 ng/ml rmM-CSF and 10 ng/ml rmIL-4 to generate MNG or rmM-CSF and RANKL (150 ng/ml) to generate osteoclasts were added in the presence or absence of increasing concentrations of anti-E-cadherin antibody (ECCD-1, Takara Bio Inc.). In

Moreno et al. STAT6-dependent generation of MNG

1543

other experiments, the blocking antibody was added at different times after the initiation of culture with cytokine as indicated. The media, cytokines, and reagents were added fresh every other day. After 5 days in culture, the cells were fixed and stained in one step with 1% Crystal violet in 95% ethanol for MNG or were stained for TRAP as described [4] for the osteoclasts, washed with water, and air-dried before analysis.

Confocal microscopy For the large bead uptake studies, BM cells were isolated from STAT6⫹/⫹ and STAT6⫺/⫺ mice and were cultured on glass coverslips (Nalge Nunc International) for 5 days in the presence of 20 ng/ml rmM-CSF, in the presence or absence of 10 ng/ml rmIL-4. Fluoresbrite AE YG™ microspheres with a diameter of 25 ␮m (1⫻103 microspheres/ml, Polysciences Inc.) were added at the beginning of culture. After 5 days, some of the wells were fixed and stained in one step with 1% Crystal violet in 95% ethanol, washed with PBS, and mounted for light and fluorescent microscopy. Others were fixed in 4% formaldehyde in PBS for 10 min at room temperature and stained with rhodamine phalloidin (Molecular Probes), according to the manufacturer’s instructions, and mounted for confocal microscopy analysis. In some cases, the beads were added after 5 days of culture when MNG were already present; these cultures were allowed to proceed for 3 more days after the addition of the beads before analysis. At the end of the 3 days, the cells were labeled with CellTrackerTM CM-DiI C7001 (Molecular Probes), fixed, and permeabilized, according to the manufacturer’s instructions. The nuclei were stained with the nuclear marker TO-PRO3 (Molecular Probes). The images were captured with a confocal microscope Radiance 2000 (Zeiss/BioRad, Thornwood, NY, USA), and the three-dimensional (3D) reconstructions of the confocal images were generated using Volocity software (Improvision, London, UK). For the cell– cell fusion studies, BM-derived macrophages were prepared from STAT6⫹/⫹ or STAT6⫺/⫺ mice as described above. STAT6⫹/⫹ macrophages were stained with the CellTrackerTM C7025 (green fluorescence), and STAT6⫺/⫺ macrophages were stained with CellTrackerTM CM-DiI C7001 (red fluorescence), according to the manufacturer’s instructions (Molecular Probes). The labeled cells were mixed at a 1:1 ratio and cultured at 2 ⫻ 106 cells/ml in the presence of 20 ng/ml rmM-CSF and 10 ng/ml rmIL-4 for 96 h. Subsequently, the cells were fixed, and the nuclei were stained with the nuclear marker TO-PRO3 (Molecular Probes). The images were captured with a confocal microscope, Radiance 2000 (Zeiss/BioRad), using a 60⫻ oil objective. Fifteen fields from three independent experiments were analyzed for the total number of cells, total number of MNG (defined as a cell containing greater than or equally to three nuclei), number of green-only MNG, number of red-only MNG, and number of MNG with both colors. The percentage of MNG in each category was calculated by dividing each number by the total number of MNG per field. The unpaired Student’s t-test was used for statistical analysis between groups.

RESULTS MNG formation is dependent on IL-4 concentration and cell density We showed previously that IL-4 inhibited the RANKL-induced differentiation of myeloid progenitors to osteoclasts [4]. However, we noted that IL-4 promoted the formation of multinucleated, TRAP-negative cells in these cultures. These cells became highly evident if progenitors were cultured in M-CSF and IL-4 in the absence of RANKL Therefore, we first examined the effect of cell density and IL-4 concentration on MNG formation (Fig. 1). The formation of MNG was dependent on both parameters but especially on cell density. At initial cell densities of 2.5 ⫻ 105 cells/ml or 1 ⫻ 106 cells/ml, we observed few, small MNG, although the cells eventually reached confluence after 5 days in culture. Extended culture for 5 additional days still did not result in the formation of MNG (data not shown). However, at initial cell densities of 2 ⫻ 1544


Fig. 1. MNG formation is dependent on IL-4 concentration and cell density. BM mononuclear cells were isolated from BALB/c mice and plated in 48-well plates at increasing cell densities in the presence or absence of various IL-4 concentrations in the continuous presence of 20 ng/ml M-CSF. Media and cytokines were replaced every other day. After 5 days in culture, the plates were fixed and stained with 1% Crystal violet (original magnification, 10⫻).

106 cells/ml, we observed the formation of MNG with as little as 5 ng/ml IL-4 (fusion index ranging from 50% to 70%). We observed the most consistent results using 2 ⫻ 106 cells/ml and 10 ng/ml IL-4 (Fig. 1). Initial cell concentrations above 2 ⫻ 106/ml failed to generate more or larger MNG, even in the presence of higher IL-4 concentrations (data not shown). These http://www.jleukbio.org

experiments demonstrate that IL-4 acts directly on myeloid progenitors to drive MNG formation in a concentration and cell density-dependent manner.

IL-4 induction of monocyte fusion is STAT6dependent IL-4 activates several signal transduction pathways in monocytic cells, which influence cellular responses, including the insulin receptor substrate 2 (IRS-2), Shc, and STAT6 pathways [3], and the STAT6 pathway is well-known to modulate the expression of genes in response to IL-4. However, the IRS-2 pathway regulates the expression of c-myc and the phosphorylation of many cell substrates, including nuclear DNA-binding proteins, which could participate in macrophage fusion [3]. Therefore, to test directly whether the IL-4-induced formation of MNG is STAT6-dependent, we used BM cells from STAT6⫺/⫺ mice. Under optimal conditions for inducing MNG from STAT6⫹/⫹ BM, BM cells from STAT6⫺/⫺ mice failed to generate large MNG (Fig. 2). We performed histomorphometrical analysis of these cultures, analyzing the number of MNG per field, the area covered by the MNG, and the fusion index (Fig. 2). The number of MNG in the STAT6⫹/⫹ cultures averaged 32.13 MNG/field, and in the STAT6⫺/⫺ cultures, there were only 5.61 MNG/field (Fig. 2). Measurement of the area occupied by the MNG showed that the few MNG present in the STAT6⫺/⫺ cultures were significantly smaller (4394.58 ␮m2/field) than those observed in the STAT6⫹/⫹ cultures (84,559.72 ␮m2/field; Fig. 2). The fusion index for the STAT6⫺/⫺ cultures (2.8%) was significantly less than the fusion index calculated for the STAT6⫹/⫹ cultures (46.23%; Fig. 2). These results clearly demonstrate that IL-4 induces MNG formation by a STAT6-dependent mechanism.

Timing of IL-4 effect on MNG formation We previously showed that IL-4 regulated osteoclast development by acting only during a defined period in culture [4]. IL-4

acted on osteoclast precursors only during the initial 48 h of culture; afterwards, they were no longer suppressed by IL-4 and continued along the osteoclast pathway. We also found that treatment with IL-4 only during this initial 48 h was sufficient to suppress further osteoclast development, even if IL-4 were not present for the rest of the culture. To determine if the differentiation of MNG cells from BM precursors required the presence of IL-4 during a specific time period, BM cells were cultured with rmM-CSF for various times before the addition of IL-4 (Fig. 3). We found that BM precursors retained the ability to differentiate into MNG in response to IL-4, even after a preliminary culture with rmM-CSF for 120 h (Fig. 3A). We found that longer times of culture in M-CSF alone lead to the formation of larger MNGs in response to IL-4 (Fig. 3C). In the reverse experiment, BM precursors were cultured with rmMCSF and IL-4 for an initial period, after which IL-4 was removed, and the cells were cultured further with rmM-CSF alone (Fig. 3B). The presence of IL-4 for the first 48 h or 72 h yielded few MNG. However, 96 –120 h of IL-4 treatment were able to stimulate the formation of MNG efficiently, and average area values were similar to those cells cultured in the presence of IL-4 for the full culture period. These results suggest that M-CSF-derived macrophages retain the ability to form MNG cells in response to IL-4. They indicate further that the efficient formation of MNG depends on the presence of IL-4 for relatively long culture periods (96 –120 h).

Effect of STAT6 and IL-4 on the phagocytosis of small particles IL-4 promotes the formation of MNG in vitro and participates in the development of the MNG reaction at the tissue-material interface in vivo, where they phagocytose and destroy foreign materials [19]. It has been proposed that the basic phagocytosis machinery is also used during cell– cell fusion, leading to the generation of giant cells [28]. As IL-4 was not able to induce fusion and MNG formation in STAT6⫺/⫺ cells, we investigated

Fig. 2. STAT6 requirement for MNG formation. BM cells were isolated from STAT6⫹/⫹ and STAT6⫺/⫺ mice and plated on glass coverslips at 2 ⫻ 106 cells/ml for 5 days with 20 ng/ml M-CSF and 10 ng/ml IL-4. Subsequently, they were washed with PBS, fixed with 4% formaldehyde in PBS, and stained with (A) 1% Crystal violet or (B) rhodamine phalloidin and Hoechst’s dye. Digital pictures were taken at low power (10⫻ and 20⫻ objectives). The images from three independent experiments were analyzed for the quantification of the number, area, and fusion index of the MNG using Bioquant software. The area of the cells and the fusion index were determined as described in Materials and Methods. Data represent the average ⫾ SEM. The unpaired Student’s t-test was used for statistical analysis.


1545

Fig. 3. BM precursors in the presence of M-CSF retain their ability to differentiate into MNG cells. BM mononuclear cells from STAT6⫹/⫹ animals were seeded on glass coverslips at 2 ⫻ 106 cells/ml and remained in the continuous presence of 20 ng/ml MCSF for the entire length of the experiment. (A) At the indicated times after the beginning of the culture, 10 ng/ml IL-4 was added to the wells. After 120 h total in the presence of IL-4, the cells were fixed and stained with 1% Crystal violet (original magnification, 10⫻). (B) The cells were exposed to 10 ng/ml IL-4 for the indicated times. Subsequently, the IL-4 was removed by washing the wells three times with complete media and then cultured with M-CSF alone for the remainder of the culture period up to 120 h. Cells were fixed, stained, and photographed as in A. (C) The quantification of the average MNG area in each panel was performed using Bioquant software. The data represent the average ⫾ SEM. These results are representative of three independent experiments.

whether STAT6⫺/⫺ macrophages were able to phagocytose small particles. BM cells from STAT6⫹/⫹ and STAT6⫺/⫺ animals were cultured in the presence of M-CSF alone for 48 h. After this period, the cells were cultured with M-CSF in the presence or absence of IL-4 for an additional 48 h. Small fluorescent beads (1 ␮m size) were added for 30 min before bead uptake was analyzed by FACS (Fig. 4). At this time, the cells were still mononuclear. We found no substantial difference in the ability of STAT6⫹/⫹- and STAT6⫺/⫺-derived macrophages to take up beads (95% and 92% positive for FITCbead uptake, respectively). However, the addition of IL-4 to STAT6⫹/⫹ cultures suppressed the uptake of beads by 20 – 40%, although this was not statistically significant (Fig. 4, A and B). By analyzing the MFI of the positive cells, we found that the amount of beads taken up per cell within 30 min was reduced significantly by IL-4 (Fig. 4B). These IL-4-induced effects on bead uptake were absent in STAT6⫺/⫺ macrophages. These results indicate that the STAT6⫺/⫺-derived macrophages are able to phagocytose small, foreign particles, despite the inability to fuse in response to IL-4 stimulation. Further, they suggest that the IL-4/STAT6 pathway partially limits the phagocytosis of small particles and promotes cell fusion and formation of MNG.

IL-4-induced MNG can incorporate large foreign bodies To analyze the capacity of macrophages and MNG to engulf larger foreign bodies, we cultured BM cells from STAT6⫹/⫹ and STAT6⫺/⫺ animals with M-CSF in the presence of IL-4 and 25 ␮m size fluorescent beads for 5 days (Fig. 5, A–G). 1546


The large MNG generated in the STAT6⫹/⫹ culture in the presence of IL-4 were able to incorporate several of these larger beads per cell (Fig. 5, A and B). In the STAT6⫺/⫺ cultures, IL-4 did not stimulate the incorporation of the larger beads (note that the beads in Fig. 5, C and D, are clearly outside of the cells). Nonetheless, it was possible to observe single cells attached to the surface of the bead in an apparent attempt to contain the foreign particle (Fig. 5E). To determine whether the large beads were actually inside of the MNG, we performed confocal microscopy (Fig. 5, F and G). We found that the large FITC beads were indeed inside the cell adjacent to actin filaments. In these experiments, the beads were added at the beginning of culture, and it was possible that the cells were fusing around the bead. To determine whether fully formed MNG could internalize large beads, we added the beads after 5 days of culture in M-CSF and IL-4. After 3 days in the presence of the beads, the membrane of the cells was labeled with a lipophilic tracer (red) to show the full internalization of the FITC beads. We found that fully formed MNG could indeed internalize 25 ␮m beads (Fig. 5, H–J). Note that the FITC beads are inside the cell adjacent to the nuclei, which show pink in the merged 3D reconstruction (Fig. 5J). These results indicate that the IL-4-induced, STAT6-dependent formation of MNG is required to engulf a larger foreign body.

Mechanism by which IL-4 promotes cell fusion As our data indicate, the IL-4/STAT6 axis controls cell fusion. A number of soluble factors and membrane-bound molecules have been shown to regulate the cell– cell fusion, which occurs during the formation of multinucleated osteoclasts or macrohttp://www.jleukbio.org

Fig. 4. Phagocytosis of small particles. BM mononuclear cells were isolated from STAT6⫹/⫹ and STAT6⫺/⫺ mice and cultured in the presence of 20 ng/ml M-CSF for 48 h. After this initial period, the cells were cultured with 20 ng/ml M-CSF in the presence or absence of 10 ng/ml IL-4 for an additional 48 h. At the end of the 48-h period, the cells were cultured in the presence (heavy line) or absence (light line) of Fluoresbrite AE YG™ microspheres with a diameter of 1 ␮m for 30 min at 37ºC. The cells were subsequently washed extensively with PBS, lifted, and analyzed by flow cytometry for bead content. (A) Representative FACS histograms are shown, and the percentage of cells containing FITC beads is indicated. (B) The effect of IL-4 on the percentage of cells containing beads and on the mean fluorescence intensity (MFI) of cells containing beads was calculated as a percentage of the wild-type (WT), untreated control. The values from three independent experiments are shown in the scatter plots. The average value is indicated by the bar. The unpaired Student’s t-test was used for statistical analysis (P⬍0.05).

phages (reviewed in refs. [29, 30]). Therefore, to analyze the mechanism by which IL-4 promoted macrophage fusion, we first tested whether the addition of soluble factors, which have been shown in the literature to regulate the generation of MNG, could overcome the poor fusion rate observed in the STAT6⫺/⫺ cultures [7, 29]. None of the cytokines tested, including IL-3, GM-CSF, or CCL2, or their combinations was able to rescue the low fusion rate observed in the STAT6⫺/⫺ cultures, although they were able to enhance MNG formation in STAT6⫹/⫹ cultures (data not shown). Furthermore, cell-free supernatants derived from IL-4-treated STAT6⫹/⫹ macrophages were not able to promote fusion in STAT6⫺/⫺ cultures (data not shown).

Our results, indicating the dependency of MNG formation on cell density, suggest the importance of cell– cell contact in the fusion process. Thus, we examined a potential role for cell surface molecules. Osteoclasts and MNG are derived from a common mononuclear monocyte-macrophage progenitor and form as the result of cell– cell fusion [16]. E-cadherin has been implicated extensively in cell– cell contact [27] and also in osteoclast progenitor fusion, as antibodies to E-cadherin block osteoclast development and activity [31]. Therefore, we tested the role of E-cadherin on the IL-4-induced MNG formation. We analyzed the levels of E-cadherin during MNG formation on STAT6⫹/⫹ macrophages by Western blotting with two different antibodies (Fig. 6A). We observed a clear induction of E-cadherin protein starting at 8 h after the addition of IL-4, reaching maximum levels after 16 h. The level of E-cadherin remained high after 48 h in the presence of IL-4 (Fig. 6A). To verify that the increase in E-cadherin protein observed by Western blot translated into higher levels of E-cadherin protein on the cell surface, we performed cell surface biotinylation (Fig. 6B). Macrophages from STAT6⫹/⫹ mice were treated in the presence or absence of IL-4 before the cell surface proteins were labeled with biotin. Precipitation of cell lysates with neutravidin beads pulled down E-cadherin only from cells treated with IL-4. In the reverse experiment, precipitation of cell lysates with anti-E-cadherin pulled down biotinylated E-cadherin only from cells treated with IL-4. These results indicate that IL-4 induced expression of E-cadherin and that it was expressed on the cell surface. In the next experiment, we analyzed the induction of E-cadherin on STAT6⫺/⫺-derived macrophages (Fig. 6C). IL-4 failed to up-regulate E-cadherin expression in the absence of STAT6. In addition, we analyzed the expression of another protein, DC-STAMP, recently shown to be essential for the formation of osteoclasts induced by M-CSF plus RANKL treatment and MNG induced by M-CSF plus IL-3 and IL-4 treatment [32]. Similar to E-cadherin, we found that expression of DC-STAMP protein was induced by IL-4 in macrophages derived from STAT6⫹/⫹ mice, but not in macrophages derived from STAT6⫺/⫺ mice (Fig. 6C). Taken together, these results suggest that IL-4 acts to modulate E-cadherin and DC-STAMP expression in a STAT6-dependent manner and thus, facilitates cell– cell fusion. To clarify further the role of E-cadherin on the formation of MNG cells, we treated osteoclast and MNG cultures with a mAb against E-cadherin known to block E-cadherin–E-cadherin interaction [27] (Fig. 7A). Anti-E-cadherin antibody treatment had only a modest effect on the number of MNG or osteoclasts present, using three nuclei per cell as the inclusive criteria (Fig. 7B). These results are in contrast to our studies in STAT6⫺/⫺ cells (Fig. 2), where we observed an 85% decrease in MNG number as compared with STAT6⫹/⫹ cells, suggesting that E-cadherin is not the only STAT6-regulated target involved in fusion. However, concentrations of the anti-E-cadherin antibody ⱖ3 ␮g/ml dramatically reduced the area of the MNG or osteoclast and the number of nuclei contained in each cell. We noticed that anti-E-cadherin did not suppress expression of TRAP in these cultures, in contrast to IL-4 treatment [4]. These results suggest an important role for E-cadherin in the IL-4-induced formation of MNG and the RANKL-induced formation of osteoclasts from mononuclear precursor cells. Moreno et al. STAT6-dependent generation of MNG

1547

Fig. 5. STAT6⫹/⫹-derived MNG can incorporate large foreign bodies. BM mononuclear cells, isolated from STAT6⫹/⫹ and STAT6⫺/⫺ mice, were cultured on glass coverslips at 2 ⫻ 106/ml for 5 days in the presence of 20 ng/ml M-CSF, in the presence or absence of 10 ng/ml IL-4. Fluoresbrite microspheres with a diameter of 25 ␮m were added at the beginning of culture. After 5 days, some of the wells were stained with 1% Crystal violet (A–E) and pictured with light (A and C) or fluorescent microscopy (B, D, and E). Others were fixed in 4% formaldehyde and stained with rhodamine phalloidin for confocal microscopy analysis (F and G). After 5 days, when mature MNG cells were present, Fluoresbrite microspheres with a diameter of 25 ␮m were added to the culture. After 3 days, the membrane of the cells was labeled with CellTracker™ CM-DiI C7001 (red fluorescence). The cells were fixed with 4% formaldehyde and permeabilized with acetone, and the nuclei were stained with TO-PRO3. The slides were analyzed by confocal microscopy. (H) The FITC bead and membrane labeling, (I) the nuclei stain with TO-PRO3, and (J) all three stains in a 3D reconstruction of the shaded area.

The kinetics of E-cadherin induction by IL-4 (Fig. 6) indicates that the E-cadherin protein is induced quickly after the addition of IL-4 to macrophages cultures. This suggests the blockade of E-cadherin homotypic interactions may need to

occur before the cells start to participate in cell– cell adhesion. To test this possibility, we added the anti-E-cadherin-blocking antibody at different times after the addition of IL-4 (Fig. 8A). As we observed previously (Fig. 7), we did not observe a large

Fig. 6. IL-4 induced E-cadherin expression in a STAT6-dependent manner. (A) STAT6⫹/⫹-derived macrophages were cultured with 20 ng/ml M-CSF in the presence or absence of 10 ng/ml IL-4 for different times as indicated. After treatment, the cells were lysed, and the lysates were analyzed for E-cadherin expression by SDS-PAGE and Western blotting as described in Materials and Methods with a mAb (MAB36) or a rabbit polyclonal antibody against E-cadherin. After stripping, the blots were reprobed with anti-HSP-86. (B) STAT6⫹/⫹-derived macrophages were cultured with 20 ng/ml M-CSF in the presence or absence of 10 ng/ml IL-4 for 48 h. After this, the cells were subjected to cell surface biotinylation and lysed. The lysates were subjected to immunoprecipitation (IP) with Neutravidin beads or anti-E-cadherin antibodies and analyzed by SDS-PAGE and Western blotting (WB) with anti-biotin or anti-E-cadherin antibodies as indicated. (C) STAT6⫹/⫹- and STAT6⫺/⫺-derived M-CSF-dependent macrophages were cultured with 20 ng/ml M-CSF in the presence or absence of 10 ng/ml IL-4 for the indicated times. At the end of the treatment, the cells were lysed, and the lysates were analyzed for E-cadherin expression by SDS-PAGE and Western blotting with a mAb (MAB36) against E-cadherin or a polyclonal antibody against DC-STAMP. After stripping, the blots were reprobed with anti-␣ tubulin or anti-HSP-86.

1548


http://www.jleukbio.org

Fig. 7. Blocking MNG formation by anti-E-cadherin. BM mononuclear cells from STAT6⫹/⫹ animals were cultured in 48-well plates at 2 ⫻ 106 cells/ml in the continuous presence of 20 ng/ml M-CSF and 10 ng/ml IL-4 or M-CSF (20 ng/ml) and RANKL (150 ng/ml) as indicated. Increasing concentrations of a blocking mAb against murine E-cadherin (ECCD1) were added to the cultures. Media, cytokines, and blocking antibody were exchanged every other day. (A) After 5 days, the cultures were fixed and stained with 1% Crystal violet for MNG or for TRAP in the case of the osteoclasts (original magnification, 4⫻). (B) The quantification of the average number, average area, and average number of nuclei per MNG or osteoclasts (OC) in each panel was performed using Bioquant software. The average ⫾ SEM is shown. These results are representative of three independent experiments.

effect on the number of MNG present, using three nuclei per cell as the inclusive criteria (Fig. 8B). However, there was dramatic reduction of the area of the MNG cells if the antibody were added at Time 0 (75% reduction relative to the control), which diminished progressively at later time-points (Fig. 8B). If the antibody were added 48, 72, or 96 h after the addition of IL-4, the reduction in the area of the MNG cells relative to the control diminished to 50%, 32%, and 17%, respectively. This indicates that the blockade of E-cadherin homotypic interactions must occur early in culture to prevent fusion. E-cadherin promotes homotypic cell adhesion, selective cell– cell interaction, and the formation of adherens junctions in epithelial tissue [27, 33], and we show herein that it participates in formation of MNG. It was shown recently that another cell surface molecule, DC-STAMP, also participates in cell– cell fusion. These authors found that regulation of cell fusion by DC-STAMP occurred in a heterotypic manner [32].

That is, the authors found that only one cell partner needed to express DC-STAMP for efficient fusion to proceed. Therefore, to determine whether IL-4-induced changes lead to homotypic or heterotypic interactions, we labeled STAT6⫹/⫹ and STAT6⫺/⫺ macrophages independently and then mixed them in equal proportions. The relative ability of STAT6⫹/⫹ cells to fuse with STAT6⫹/⫹ cells and form MNG in the presence of M-CSF and IL-4 was compared with the ability of STAT6⫹/⫹ cells to fuse with STAT6⫺/⫺ cells using confocal fluorescent microscopy (Fig. 9). We found that the vast majority of MNG cells was derived solely from green-labeled STAT6⫹/⫹ cells (75% of total MNG). As expected, we did not find any MNG derived solely from red STAT6⫺/⫺ cells. This was also the case if the ratio of STAT6⫺/⫺:STAT6⫹/⫹ cells were 3:1 (data not shown). However, we did observe a small proportion (⬃25%) of green MNG cells containing some red label (example marked with arrows, showing as yellow on the merged images; Fig. 9, A Moreno et al. STAT6-dependent generation of MNG

1549

Fig. 8. Time-dependent effect of E-cadherin blockade on MNG formation. BM mononuclear cells from STAT6⫹/⫹ animals were cultured in 48-well plates at 2 ⫻ 106 cells/ml in the continuous presence of 20 ng/ml M-CSF and 10 ng/ml IL-4. (A) At the indicated times after the beginning of the culture, 5 ␮g/ml blocking mAb against murine E-cadherin (ECCD1) was added to the wells. After 120 h total in the presence of IL-4, the cells were fixed and stained with 1% Crystal violet (original magnification, 4⫻). (B) The quantification of the average number and MNG area was performed using Bioquant software. The average ⫾ SEM is shown. These results are representative of two independent experiments.

and B). The green STAT6⫹/⫹ cells and the red STAT6⫺/⫺ cells remained healthy and viable during the course of the experiment (Fig. 9, and data not shown). It is interesting that we noticed green MNG in close proximity to single red STAT6⫺/⫺ cells (Fig. 9C). In addition, we frequently observed single green STAT6⫹/⫹ cells in close interaction with single red STAT6⫺/⫺ cells. In many cases, the STAT6⫹/⫹ cells with a single nucleus contained red and green dye. These results suggest that homotypic STAT6⫹/⫹ to STAT6⫹/⫹ interactions are necessary for the efficient formation of MNG in response to IL-4 but that heterotypic interactions may proceed less efficiently. It is also possible that once formed, the STAT6⫹/⫹ MNG can phagocytose a labeled STAT6⫺/⫺ macrophage (a process termed cellocytosis [30]) or that they can take up released red dye.

DISCUSSION The basic ability of IL-4 to induce MNG formation in vitro was described over 10 years ago [7, 34]. Furthermore, IL-4 was shown to participate in the formation of MNG in vivo in response to biomaterials; anti-IL-4 antibodies reduced the numbers of MNG formed in response to implanted poly (etherurethane urea) cages, and addition of exogenous IL-4 enhanced their numbers [35]. However, the mechanism by 1550


which IL-4 regulates MNG formation has not been delineated. In this study, we investigated the mechanism by which IL-4 promotes MNG formation, identifying a critical role for Ecadherin in the formation of MNG, likely by modulating homotypic fusion of precursors. We found that the IL-4-induced generation of MNG is STAT6-dependent, as precursors derived from STAT6⫺/⫺ mice failed to develop efficiently into MNG in response to IL-4. The few MNG observed in the STAT6⫺/⫺ cultures contained less than five nuclei per cell (data not shown) and may form as a result of background fusion in response to M-CSF, as has been described for other cytokines and systems [7]. A recent report by McNally and Anderson [28] showed that the process of macrophage fusion exhibits several features of phagocytosis, as it required the participation of microtubules, V-type ATPase, and independent phospholipase A2. The authors demonstrated that components of the endoplasmic reticulum (ER) were present at fusion interfaces and on macrophage surfaces during MNG formation. Furthermore, it has been shown that many cytoskeletal components known to participate in adhesion are enriched in the phagocytic cup [36 –38]. These results indicate that the cell uses similar proteins and mechanisms for phagocytosis and cell– cell fusion, which may be limited in quantity. In keeping with this idea, we observed that IL-4 treatment suppressed the number of beads phagocytized per cell by STAT6⫹/⫹ macrophages over a period of 30 min. http://www.jleukbio.org

Fig. 9. Analysis of cell– cell fusion. BM-derived macrophages were prepared from STAT6⫹/⫹ or STAT6⫺/⫺ mice and cultured in the presence of 20 ng/ml rmM-CSF for 48 h. STAT6⫹/⫹ macrophages were stained with the probe CellTracker™ C7025 (green fluorescence), and STAT6⫺/⫺ macrophages were stained with CellTracker™ CM-DiI C7001 (red fluorescence). The labeled cells were mixed at a 1:1 ratio and cultured at 2 ⫻ 106 cells/ml in the presence of 20 ng/ml rmM-CSF and 10 ng/ml rmIL-4 for 96 h. Subsequently, the cells were fixed, and the nuclei were stained with the nuclear marker TO-PRO3 (Molecular Probes). The images were captured with a confocal microscope, Radiance 2000 (Zeiss/BioRad), using a 60⫻ oil objective. (A) Representative panels illustrating typical staining for nuclei (blue, TO-PRO3), STAT6⫹/⫹ cells (green, C7025), and STAT6⫺/⫺ cells (red, C7001) and all three images merged are shown. Optical-section thickness was 0.8 ␮m for green and red and 1.5 ␮m for blue. MNG are highlighted with a dotted line. An example of a MNG containing red and green dye is highlighted with arrows. (B) Fifteen fields originating from three independent experiments were analyzed for the total number of cells, total number of MNG (defined as a cell containing greater than or equal to three nuclei), number of green-only MNG, number of red-only MNG, and number of MNG with both colors (showing as yellow staining). The percentage of MNG in each category was calculated by dividing each number by the total number of MNG per field. The unpaired Student’s t-test was used for statistical analysis between the green and yellow groups (**, P⬍0.001). (C) Representative images of MNG and mononuclear cells.

These cells are programmed to fuse and differentiate into MNG. Perhaps the machinery available for rapid and efficient phagocytic uptake of beads is diverted to the fusion process in STAT6⫹/⫹ cells responding to IL-4. An in-depth analysis of the contribution of the phagocytic machinery to IL-4-induced, STAT6-dependent fusion mechanisms will require further study. When we challenged the macrophage cultures with large particles (25 ␮m diameter, larger than a single macrophage), we found that STAT6⫹/⫹ MNG were able to internalize numerous particles per cell, indicating these cells were functionally active. However, in contrast to small beads (1 ␮m), STAT6⫺/⫺ cells were not able to incorporate any large particles. Nonetheless, STAT6⫺/⫺ cells were detected, attached to the surface of the beads, suggesting a frustrated attempt to

internalize the particles. In agreement with this data, others have reported that the size of the cell correlates with the size of the internalized particle [39 – 41]. Von Knoch et al. [42] have shown that the mean intracellular polyethylene particle size is significantly greater in FBGC than in macrophages. Thus, the lack of IL-4-induced fusion in the absence of STAT6 likely prevents the incorporation of large particles, although the phagocytic capacity of the STAT6⫺/⫺-derived cells remained intact. As IL-4 inhibited the formation of multinucleated osteoclasts but promoted the formation of MNG, we suspected originally that the fusion mechanisms used by osteoclasts and MNG would be different. However, recent studies suggest that there is a common fusion mechanism used by these cell types [29, 32]. We found evidence for an important role for EMoreno et al. STAT6-dependent generation of MNG

1551

cadherin in the formation of MNG and osteoclasts from mononuclear precursor cells. Furthermore, we found that IL-4 clearly induced the membrane expression of E-cadherin in a STAT6-dependent manner on BM macrophages. It was shown recently that expression of the ER and cell surface protein, DC-STAMP, is critical for the formation of MNG and osteoclasts [32]. We also observed a STAT6-dependent induction of DC-STAMP protein in response to IL-4, suggesting the participation of E-cadherin and DC-STAMP in the IL-4-induced fusion. The antibody-blocking experiments revealed a critical role for E-cadherin in the fusion process for MNG and osteoclasts. The addition of the blocking antibody at later time-points showed less-efficient inhibition of MNG formation. This may be a result of a loss of antibody epitope caused by homotypic interaction of E-cadherins on neighboring cells or by lateral interaction of E-cadherin molecules on the same cell [43]. The analysis of the number of MNG generated in the antibodyblocking experiments using the classic definition of MNG (containing at least three nuclei per cell) did not reveal a clear impact of E-cadherin on the fusion process (Figs. 7 and 8). This is in contrast with the profound effect of STAT6 deficiency on the numbers of MNG in IL-4-treated cultures (Fig. 2). However, analysis of cell area and nuclei content per cell clearly indicated a suppressive effect of anti-E-cadherin antibodies on MNG formation. The difference between STAT6 deficiency and E-cadherin blockade on MNG number may be a result of the contribution of other STAT6-regulated molecules, such as DCSTAMP, in the formation of small (approximately three nuclei) MNG. The STAT6 dependency of MNG formation does not appear to be mediated by production of soluble factors. We found that cell-free supernatants from IL-4-treated STAT6⫹/⫹ cells did not induce MNG formation from STAT6⫺/⫺ precursors (data not shown). Furthermore, the addition of the chemokine CCL2, shown to be required for MNG formation [44, 45], was not able to stimulate MNG formation in the STAT6⫺/⫺ cultures (data not shown). Using cell-mixing experiments, we found that STAT6⫹/⫹ cells would fuse efficiently with STAT6⫹/⫹ cells in response to IL-4 and form MNG but that STAT6⫹/⫹ cells would rarely form MNG containing STAT6⫺/⫺ cells (Fig. 9). These results indicate a requirement for cell intrinsic homotypic STAT6⫹/⫹ to STAT6⫹/⫹ interaction for efficient formation of MNG. However, Yagi et al. [32] found that expression of DC-STAMP was only required on one of the cell partners to drive efficient fusion induced by treatment with IL-3 plus IL-4; DC-STAMP⫹/⫹ cells fused efficiently with DC-STAMP⫺/⫺ cells. Thus, the mechanism by which IL-4 promotes homotypic fusion in the presence of M-CSF likely involves the STAT6dependent regulation of several cell surface molecules, including E-cadherin and DC-STAMP. Furthermore, several other candidate molecules known to play a role in the fusion of alveolar macrophages may participate in the IL-4-induced fusion [29, 46]. A complete understanding of the detailed molecular interactions activated by IL-4 to promote macrophage fusion and MNG formation will require additional investigation. 1552


ACKNOWLEDGMENTS We acknowledge Ms. Rebecca Kurnat, Ms. Xiulan Qi, and Ms. Elena Semenova for excellent technical assistance and Dr. Joseph Stains for helpful discussions. This work was supported by PHS grant AI59775 (A.D.K.).

REFERENCES 1. Paul, W. E. (1991) Interleukin-4: a prototypic immunoregulatory lymphokine. Blood 77, 1859 –1870. 2. Gessner, A., Mohrs, K., Mohrs, M. (2005) Mast cells, basophils, and eosinophils acquire constitutive IL-4 and IL-13 transcripts during lineage differentiation that are sufficient for rapid cytokine production. J. Immunol. 174, 1063–1072. 3. Nelms, K., Keegan, A. D., Zamorano, J., Ryan, J. J., Paul, W. E. (1999) The IL-4 receptor: signaling mechanisms and biologic functions. Annu. Rev. Immunol. 17, 701–738. 4. Moreno, J. L., Kaczmarek, M., Keegan, A. D., Tondravi, M. (2003) IL-4 suppresses osteoclast development and mature osteoclast function by a STAT6-dependent mechanism: irreversible inhibition of the differentiation program activated by RANKL. Blood 102, 1078 –1086. 5. Abu-Amer, Y. (2001) IL-4 abrogates osteoclastogenesis through STAT6dependent inhibition of NF-␬B. J. Clin. Invest. 107, 1375–1385. 6. Wei, S., Wang, M. W., Teitelbaum, S. L., Ross, F. P. (2002) Interleukin-4 reversibly inhibits osteoclastogenesis via inhibition of NF-␬ B and mitogen-activated protein kinase signaling. J. Biol. Chem. 277, 6622– 6630. 7. McNally, A. K., Anderson, J. M. (1995) Interleukin-4 induces foreign body giant cells from human monocytes/macrophages. Differential lymphokine regulation of macrophage fusion leads to morphological variants of multinucleated giant cells. Am. J. Pathol. 147, 1487–1499. 8. Anderson, J. M. (2000) Multinucleated giant cells. Curr. Opin. Hematol. 7, 40 – 47. 9. Guironnet, G., Dezutter-Dambuyant, C., Vincent, C., Bechetoille, N., Schmitt, D., Peguet-Navarro, J. (2002) Antagonistic effects of IL-4 and TGF-␤1 on Langerhans cell-related antigen expression by human monocytes. J. Leukoc. Biol. 71, 845– 853. 10. Watanabe, K., Tanaka, Y., Morimoto, I., Yahata, K., Zeki, K., Fujihira, T., Yamashita, U., Eto, S. (1990) Interleukin-4 as a potent inhibitor of bone resorption. Biochem. Biophys. Res. Commun. 172, 1035–1041. 11. Lacey, D. L., Erdmann, J. M., Teitelbaum, S. L., Tan, H. L., Ohara, J., Shioi, A. (1995) Interleukin 4, interferon-␥, and prostaglandin E impact the osteoclastic cell-forming potential of murine bone marrow macrophages. Endocrinology 136, 2367–2376. 12. Suk, K., Somers, S. D., Erickson, K. L. (1993) Regulation of murine macrophage function by IL-4: IL-4 and IFN-␥ differentially regulate macrophage tumoricidal activation. Immunology 80, 617– 624. 13. Gordon, S. (2003) Alternative activation of macrophages. Nat. Rev. Immunol. 3, 23–35. 14. Ma, J., Chen, T., Mandelin, J., Ceponis, A., Miller, N. E., Hukkanen, M., Ma, G. F., Konttinen, Y. T. (2003) Regulation of macrophage activation. Cell. Mol. Life Sci. 60, 2334 –2346. 15. Miyamoto, T., Ohneda, O., Arai, F., Iwamoto, K., Okada, S., Takagi, K., Anderson, D. M., Suda, T. (2001) Bifurcation of osteoclasts and dendritic cells from common progenitors. Blood 98, 2544 –2554. 16. Vignery, A. (2000) Osteoclasts and giant cells: macrophage-macrophage fusion mechanism. Int. J. Exp. Pathol. 81, 291–304. 17. Dadsetan, M., Jones, J. A., Hiltner, A., Anderson, J. M. (2004) Surface chemistry mediates adhesive structure, cytoskeletal organization, and fusion of macrophages. J. Biomed. Mater. Res. A 71, 439 – 448. 18. Anderson, J. M. (2001) Biological responses to materials. Annu. Rev. Mater. Res. 31, 81–110. 19. Anderson, J. M. (1988) Inflammatory response to implants. ASAIO Trans. 34, 101–107. 20. Hu, W. J., Eaton, J. W., Ugarova, T. P., Tang, L. (2001) Molecular basis of biomaterial-mediated foreign body reactions. Blood 98, 1231–1238. 21. Picha, G. J., Goldstein, J. A., Stohr, E. (1990) Natural-Y Meme polyurethane versus smooth silicone: analysis of the soft-tissue interaction from 3 days to 1 year in the rat animal model. Plast. Reconstr. Surg. 85, 903–916. 22. Sutherland, K., Mahoney II, J. R., Coury, A. J., Eaton, J. W. (1993) Degradation of biomaterials by phagocyte-derived oxidants. J. Clin. Invest. 92, 2360 –2367.

http://www.jleukbio.org

23. Zhao, Q., Agger, M. P., Fitzpatrick, M., Anderson, J. M., Hiltner, A., Stokes, K., Urbanski, P. (1990) Cellular interactions with biomaterials: in vivo cracking of pre-stressed Pellethane 2363-80A. J. Biomed. Mater. Res. 24, 621– 637. 24. Zhao, Q., Topham, N., Anderson, J. M., Hiltner, A., Lodoen, G., Payet, C. R. (1991) Foreign-body giant cells and polyurethane biostability: in vivo correlation of cell adhesion and surface cracking. J. Biomed. Mater. Res. 25, 177–183. 25. Chan, S. C., Birdsell, D. C., Gradeen, C. Y. (1991) Detection of toluenediamines in the urine of a patient with polyurethane-covered breast implants. Clin. Chem. 37, 756 –758. 26. Tondravi, M. M., McKercher, S. R., Anderson, K., Erdmann, J. M., Quiroz, M., Maki, R., Teitelbaum, S. L. (1997) Osteopetrosis in mice lacking hematopoietic transcription factor PU.1. Nature 386, 81– 84. 27. Nose, A., Tsuji, K., Takeichi, M. (1990) Localization of specificity determining sites in cadherin cell adhesion molecules. Cell 61, 147–155. 28. McNally, A. K., Anderson, J. M. (2005) Multinucleated giant cell formation exhibits features of phagocytosis with participation of the endoplasmic reticulum. Exp. Mol. Pathol. 79, 126 –135. 29. Vignery, A. (2005) Macrophage fusion: the making of osteoclasts and giant cells. J. Exp. Med. 202, 337–340. 30. Vignery, A. (2005) Macrophage fusion: are somatic and cancer cells possible partners? Trends Cell Biol. 15, 188 –193. 31. Mbalaviele, G., Chen, H., Boyce, B. F., Mundy, G. R., Yoneda, T. (1995) The role of cadherin in the generation of multinucleated osteoclasts from mononuclear precursors in murine marrow. J. Clin. Invest. 95, 2757– 2765. 32. Yagi, M., Miyamoto, T., Sawatani, Y., Iwamoto, K., Hosogane, N., Fujita, N., Morita, K., Ninomiya, K., Suzuki, T., Miyamoto, K., Oike, Y., Takeya, M., Toyama, Y., Suda, T. (2005) DC-STAMP is essential for cell– cell fusion in osteoclasts and foreign body giant cells. J. Exp. Med. 202, 345–351. 33. Nelson, W. J., Nusse, R. (2004) Convergence of Wnt, ␤-catenin, and cadherin pathways. Science 303, 1483–1487. 34. McInnes, A., Rennick, D. M. (1988) Interleukin 4 induces cultured monocytes/macrophages to form giant multinucleated cells. J. Exp. Med. 167, 598 – 611.

35. Kao, W. J., McNally, A. K., Hiltner, A., Anderson, J. M. (1995) Role for interleukin-4 in foreign-body giant cell formation on a poly(etherurethane urea) in vivo. J. Biomed. Mater. Res. 29, 1267–1275. 36. Aderem, A., Underhill, D. M. (1999) Mechanisms of phagocytosis in macrophages. Annu. Rev. Immunol. 17, 593– 623. 37. Allen, L. A., Aderem, A. (1996) Molecular definition of distinct cytoskeletal structures involved in complement- and Fc receptor-mediated phagocytosis in macrophages. J. Exp. Med. 184, 627– 637. 38. Allen, L. H., Aderem, A. (1995) A role for MARCKS, the ␣ isozyme of protein kinase C and myosin I in zymosan phagocytosis by macrophages. J. Exp. Med. 182, 829 – 840. 39. Harris, W. H. (1994) Osteolysis and particle disease in hip replacement. A review. Acta Orthop. Scand. 65, 113–123. 40. Heinemann, D. E. H. (1997) The role of monocytes and macrophages in tissue reaction to wear particles of orthopedic implant materials. Osteologie 6, 112–124. 41. Shanbhag, A. S., Jacobs, J. J., Glant, T. T., Gilbert, J. L., Black, J., Galante, J. O. (1994) Composition and morphology of wear debris in failed uncemented total hip replacement. J. Bone Joint Surg. Br. 76, 60 – 67. 42. Von Knoch, M., Buchhorn, G., von Knoch, F., Koster, G., Willert, H. G. (2001) Intracellular measurement of polyethylene particles. A histomorphometric study. Arch. Orthop. Trauma Surg. 121, 399 – 402. 43. Klingelhofer, J., Laur, O. Y., Troyanovsky, R. B., Troyanovsky, S. M. (2002) Dynamic interplay between adhesive and lateral E-cadherin dimers. Mol. Cell. Biol. 22, 7449 –7458. 44. Kyriakides, T. R., Foster, M. J., Keeney, G. E., Tsai, A., Giachelli, C. M., Clark-Lewis, I., Rollins, B. J., Bornstein, P. (2004) The CC chemokine ligand, CCL2/MCP1, participates in macrophage fusion and foreign body giant cell formation. Am. J. Pathol. 165, 2157–2166. 45. Jay, S. M., Skokos, E., Laiwalla, F., Krady, M., Kyriakides, T. R. (2007) Foreign body giant cell formation is preceded by lamellipodia formation and can be attenuated by inhibition of Rac1 activation. Am. J. Pathol. 171, 632– 640. 46. Helming, L., Gordon, S. (2007) Macrophage fusion induced by IL-4 alternative activation is a multistage process involving multiple target molecules. Eur. J. Immunol. 37, 33– 42.


1553