Angiopoietin (Ang) -1 and -2 and their receptor Tie2 play critical roles in regulating angiogenic processes during development, homeostasis, tumorigenesis, inflammation and tissue repair. Tie2 signaling is best characterized in endothelial cells, but a subset of human and murine circulating monocytes/macrophages essential to solid tumor formation express Tie2 and display immunosuppressive properties consistent with M2 macrophage polarization. However, we have recently shown that Tie2 is strongly activated in pro-inflammatory macrophages present in rheumatoid arthritis patient synovial tissue. Here we examined the relationship between Tie2 expression and function during human macrophage polarization. Tie2 expression was observed under all polarization conditions, but was highest in IFN-γ and IL-10 –differentiated macrophages. While TNF enhanced expression of a common restricted set of genes involved in angiogenesis and inflammation in GM-CSF, IFN-γ and IL-10 –differentiated macrophages, expression of multiple chemokines and cytokines, including CXCL3, CXCL5, CXCL8, IL6, and IL12B was further augmented in the presence of Ang-1 and Ang-2, via Tie2 activation of JAK/STAT signaling. Conditioned medium from macrophages stimulated with Ang-1 or Ang-2 in combination with TNF, sustained monocyte recruitment. Our findings suggest a general role for Tie2 in cooperatively promoting the inflammatory activation of macrophages, independently of polarization conditions.
Citation: García S, Krausz S, Ambarus CA, Fernández BM, Hartkamp LM, van Es IE, et al. (2014) Tie2 Signaling Cooperates with TNF to Promote the Pro-Inflammatory Activation of Human Macrophages Independently of Macrophage Functional Phenotype. PLoS ONE 9(1): e82088. https://doi.org/10.1371/journal.pone.0082088
Editor: Paul Proost, University of Leuven, Rega Institute, Belgium
Received: June 21, 2013; Accepted: October 27, 2013; Published: January 3, 2014
Copyright: © 2014 García et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This research was supported by Dutch Arthritis Association project grants (NR 04-1-301 and NR 09-1-405) to Dr. K.A. Reedquist. Dr. S. García is supported by the Anxeles Alvariño postdoctoral program (I+D+I Xunta de Galicia and the European Social Fund (ESF)). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have the following interests: Paul P Tak is employed by GlaxoSmithKline and in addition to employment, he holds stock options in GlaxoSmithKline. This does not alter the authors' adherence to all the PLOS ONE policies on sharing data and materials.
The tyrosine kinase receptor Tie2 makes essential contributions to vascular development and blood vessel remodeling through its interaction with angiopoietin (Ang) ligands, of which Ang-1 and Ang-2 are the best characterized , . Ang-1 binding to Tie2 induces autophosphorylation of Tie2 on multiple tyrosine residues and activation of downstream signaling pathways. Tie2 signaling has been most extensively studied within the context of endothelial cell (EC) biology and vascular development and homeostasis. Ang-1 promotes Tie2-dependent EC survival, stability of the endothelial barrier, vascularization, and lymphangiogenesis –. The outcome of Ang-1 signaling to ECs is context-dependent, as trans signaling of Tie2 via Ang-1 presented by adjacent ECs strengthens endothelial barriers, while Ang-1 deposited on extracellular matrix components promotes EC proliferation and migration , . Overexpression of Ang-2 during development antagonizes Ang-1 function , . Ang-2 can compete with Ang-1 to prevent phosphorylation of Tie2, and this antagonistic effect is most readily observed in blocking Tie2 activation at EC cell-cell junctions , . In the absence of Ang-1, or when Ang-2 is present in high concentrations, Ang-2 can stimulate Tie2 signaling . Ang-2 can also initiate EC signaling cascades via direct binding to integrins, as evidenced by the ability of Ang-2 to promote sprouting angiogenesis of Tie2-negative ECs in vitro and in vivo . Ang-2 effects on the vascular are often pro-inflammatory, promoting vascular leakage and sensitizing ECs to tumor necrosis factor (TNF)-dependent induction of adhesion protein expression important for leukocyte extravasation –.
Myeloid cells expressing Tie2 also make requisite contributions to angiogenic processes in health and disease. A subpopulation of circulating monocytes and tumor-associated macrophages (TAMs) express Tie2 in humans and mice , . Tie2-expressing monocytes (TEMs) are recruited to solid tumors in the earliest stages of tumor development, and targeted depletion studies in mice have shown that TEMs are needed to allow vascularization and growth of solid tumors . TEMs represent a small subset of the total human peripheral blood mononuclear cell (PBMC) population, are readily detected in human tumors, and human tumor xenograft models indicate that TEMs play similar pro-angiogenic, pro-tumorigenic roles in both mice and humans . Gene expression analyses of murine TEMs suggest that they represent a distinct committed myeloid lineage derived from embryonic myeloid cells and circulating monocytes . Ang-2 stimulation serves to recruit TEMs and induce TEM expression of pro-angiogenic genes such as TP and CTSB , , . Gene expression analyses also suggest a relationship between TEMs and polarized macrophages with immunosuppressive/immunoregulatory phenotypes , . Indeed, Ang-2 induces interleukin (IL)-10 expression in TEMs, promoting the expansion and suppressive capacity of regulatory T cells .
The phenotypic characteristics and functional capacities of macrophages are tightly regulated by environmental cues they receive during differentiation, a process referred to as macrophage polarization. Cytokines and pathogens, among other factors, can guide macrophage differentiation into (classically activated) pro-inflammatory M1 or (alternatively activated) M2 wound-healing and immunosuppressive macrophages . In vivo, tissues can be populated by both M1- and M2–like macrophages, as well as macrophages of mixed phenotype. To date, TEMs have been described to display M2-like properties in mice and humans , . However, we have recently found that Tie2 is highly activated in synovial macrophages of patients with rheumatoid arthritis (RA) . In RA, pro-inflammatory activation of these macrophages is very closely linked with disease activity and patient response to therapy . These observations suggested that Tie2 function in myeloid biology may not be limited to immunosuppressive functions associated with TEMs, and in the present work, we examined the relationship between Tie2 expression and function in human macrophages within the context of macrophage polarization.
Materials and Methods
Monocyte purification and macrophage differentiation
Human PBMCs were isolated from volunteer donor blood buffy coats (Sanquin) by gradient centrifugation with Lymphoprep (Axis-Shield PoPAS) and monocytes were further isolated by Percoll gradient separation (GE Healthcare). Monocytes were differentiated into macrophages in Iscove's modified Dulbecco's medium/10% fetal calf serum supplemented with 100 µg/ml gentamycin (Invitrogen), in the presence of granulocyte-macrophage colony-stimulating factor (GM-CSF, 5 ng/ml), macrophage colony-stimulating factor (M-CSF, 25 ng/ml), IL-4 (10 ng/ml), interferon (IFN)-γ (10 ng/ml) or IL-10 (10 ng/ml) (all from R&D Systems) for 7 d.
Macrophage purity, differentiation and Tie2 expression was assessed by flow cytometry (FACS Canto Flow Cytometer, BD Biosciences). Fluorochrome-labeled monoclonal antibodies against CD16 (clone 61D3, eBiosciences), CD64 (Biolegend), CD163, CD200R (both from BD Pharmingen) and Tie2 (R&D Systems), and equivalent concentrations of isotype control antibodies, were used. Before staining, Fc receptors were blocked with 10% human serum (Lonza). Data were analyzed with Flow Jo Flow Cytometry Analysis software (Tree Star). Values were expressed as the ratio of the geometric mean fluorescence intensity (geomean) of the marker of interest over that of the isotype control.
RT-PCR and quantitative (q)PCR arrays
Macrophages were left unstimulated or were stimulated with TNF (10 ng/ml, Biosource International) in the presence or absence of recombinant human Ang-1 or Ang-2 (200 ng/ml, R&D Systems) for 4 h. Where indicated, macrophages were preincubated for 1 h in the absence or presence of the Janus kinase (JAK) inhibitor AG490 (10 µM, Calbiochem-Merck). RNA was isolated using the RNeasy Kit and RNase-Free DNase Set (Qiagen). Total RNA was reverse-transcribed using SuperScript™ II RT (Invitrogen). Duplicate PCR reactions were performed using SYBR green (Applied Biosystem) with an ABI Prism® 7000 sequence detection system (Applied Biosystems). cDNA was amplified using specific primers (Tie2 forward (for), 5′-3′, ACAATGGTGTCTGCCATGAA; Tie2 reverse (rev), TTCACAAGCCTTCTCACACG; Tie1 for, CAGATTGCGCTACAGCTAGG; Tie1 rev, CCGCGTAAGTGAAGTTCTCA; IL-12B, ACGTTTCACCTGCTGGTGGCT; IL-12B rev, CTCCGCACGTCACCCCTTGG; IL-6 for, GACAGCCACTCACCTCTTCA; IL-6 rev, CCTCTTTGCTGCTTTCACAC; GAPDH for, GCCAGCCGAGCCACATC; GAPDH rev, TGACCAGGCGCCCAATAC, all from Invitrogen). Relative levels of gene expression were normalized to GAPDH housekeeping gene. The relative quantity (RQ) of mRNA was calculated by using the formula: 2−ΔΔCt. Alternatively, total RNA was subjected to cDNA synthesis using the RT2 First Strand Kit (Qiagen) and mRNA expression of 84 angiogenic factors was analyzed by RT-PCR using low density qPCR Array (Human Angiogenic Growth Factors & Angiogenesis Inhibitors PCR Array, Qiagen). Relative levels of gene expression were normalized to 5 housekeeping genes and RQ values determined as above.
Measurement of cytokine production
Macrophages were stimulated 24 h with TNF (10 ng/ml) in the presence or absence of Ang-1 or Ang-2 (200 ng/ml). Cell-free supernatants were analyzed by ELISA for IL-6, IL-10 IL-8 and thrombospondin (TSP)-2 (PelKine Compact™ ELISA kits, Sanquin Reagents). CXCL-5, -6, CCL-3, -7, and IL-12B were measured using human single-plex assays (Bio-Rad) and read on a Bio-Plex 200 system (Bio-Rad).
Macrophages total cell lysates were subjected to electrophoresis on 4–12% gradient Bis-Tris SDS NuPAGE® gels (Invitrogen), and proteins were transferred to polyvinylidene membranes (Millipore). Membranes were incubated overnight at 4°C in primary antibodies specific for nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor alpha (IκBα), protein kinase B (PKB), phospho(p)-PKB (Ser473), extracellular signal-regulated kinase (ERK), p-ERK, p38, p-p38, and histone 3 (H3) (all from Cell Signalling), washed, and incubated in TBS/T containing horseradish peroxidase–conjugated secondary antibody. Protein was detected with Lumi-lightplus Western Blotting Substrate (Roche Diagnostics) using an ImageQuant LAS 4000 system (GE Healthcare). Densitometry analysis was performed with Image J software. Relative protein phosphorylation or expression was normalized to total protein or H3, respectively.
Monocyte chemotaxis assay
Macrophages were stimulated with TNF in the presence or absence of Ang-1 or Ang-2 for 24 h and cell-free supernatants were collected. Monocytes were purified from human PBMCs using Monocyte Isolation Kit II (Miltenyi Biotec). 1×105 monocytes were transferred into the upper chamber of 5 µm pore-size transwell plates (96 well ChemoTX®, NeuroProbe). Fresh medium alone, or containing Ang-1 or Ang-2 (200 ng/ml), or conditioned macrophage supernatants were added to the lower chamber. After 2 h at 37°C, cells migrating to the lower chamber were quantified by staining with Calcein-AM (1 µM, BD Bioscience) and analysis using a multi-label reader Victor3™ (PerkinElmer Inc.). Data were expressed as signal (arbitrary units), following subtraction of signal of empty wells containing no cells.
Measurement of STAT family member DNA binding
Nuclear fractions of macrophages were isolated using Nuclear Extract Kit (Active Motif) and DNA binding activities of STAT-1, -3, -5A, and -5B were determined using a TransAM transcription factor ELISA (Active Motif) according to the manufacturer's instructions.
Statistical analysis was performed using Windows GraphPad Prism 5 (GraphPad Software, Inc.). Potential differences between experimental groups were analyzed by non-parametric, Kruskal-Wallis test or Friedman test, as appropriate. P-values <0.05 were considered statistically significant.
Human macrophage Tie2 expression is regulated by polarizing cytokines but is not restricted to immunoregulatory M2 macrophages
To determine the influence of polarization on macrophage Tie2 expression, we differentiated macrophages in the presence of cytokines resulting in the generation of pro-inflammatory M1 macrophages (GM-CSF, IFN-γ) or wound healing/immunoregulatory M2 macrophages (M-CSF, IL-4 and IL-10) , . FACS analysis of polarized macrophages revealed that although each cytokine differentially regulated Tie2 surface expression, there was no relationship between Tie2 expression and M1 or M2 macrophage functional classification (Figure 1A). Compared to medium alone, IFN-γ, M-CSF and IL-10 significantly enhanced macrophage Tie2 surface expression, while little Tie2 was detected on macrophages exposed to IL-4 (Figure 1A–B). Tie2 mRNA expression mirrored results obtained in examining protein expression, with the highest levels of Tie2 mRNA expressed in macrophages differentiated in IFN-γ (MΦIFN), IL-10 (MΦIL10) and GM-CSF (MΦGM-CSF) (Figure 1C). Expression of Tie1, reported to act as an antagonist of Tie2 signaling in ECs, was also differentially induced in macrophages during polarization. Compared to medium alone, GM-CSF significantly enhanced Tie1 mRNA expression, and trends toward enhanced expression were observed in macrophages differentiated in M-CSF (MΦM-CSF) and IL-4 (MΦIL4) (Figure 1D). Tie2 mRNA expression relative to Tie1 was significantly enhanced in MΦIFN and MΦIL10 compared to medium alone (Figure 1E). FACS analysis of recently validated human macrophage polarization markers CD16, CD64, CD163 and CD200R ,  confirmed that macrophage culture conditions resulted in the polarization of macrophages with expected phenotypic characteristics (Figure 2A–D). Thus, in human macrophages, while macrophage polarization conditions regulate Tie2 and Tie1 expression levels, Tie2 expression is not limited to M2-like macrophages.
(A) FACS analysis of macrophages polarized for 7 d with the indicated cytokines, using Ig control (filled grey area) and Tie2 antibodies (open area). Experiments shown are representative of 7 independent experiments. (B) Quantification of relative Tie2 surface expression (geomean) and Tie2 mRNA expression (C) by macrophages differentiated in medium alone or with the indicated cytokine. (D) Quantification of relative Tie1 mRNA expression and (E) Tie2 mRNA expression relative to Tie1 by macrophages differentiated in medium alone or with indicated cytokine. qPCR data is shown as relative quantity, as described in material and methods Values are the mean ± SEM of 5–7 independent experiments. *P<0.05. **P<0.01 versus macrophages differentiated in medium alone. Kruskal-Wallis test was used for statistical analyses.
(A–D) FACS analysis of expression of macrophage surface markers (A) CD16, (B) CD163, (C) CD200R and (D) CD64 in macrophages differentiated for 7 d in medium alone, or medium containing the indicated polarizing cytokines, Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml). Data in A–D are presented as the geomean normalized to values obtained for macrophages cultured in medium alone, and represent the mean ± SEM of 4–5 independent experiments per marker. (E) Surface protein expression (geomean) and mRNA expression (F) of Tie2, as determined by FACS analysis (n = 5) and qPCR (n = 3), respectively, in human macrophages cultured for 7 d in the absence (medium) or presence of 200 ng/ml Ang-1 or Ang-2. *P<0.05, **P<0.01, ***P<0.001, versus macrophages differentiated in medium alone. Kruskal-Wallis test was used for statistical analyses.
Ang-1 and Ang-2 do not influence macrophage polarization
We next examined the possibility that Ang-1 and Ang-2 might directly regulate macrophage polarization, as acute stimulation of freshly derived human monocytes with Ang-2 can further enhance Tie2 expression and reinforce immunoregulatory gene expression patterns . Macrophages were differentiated in the absence or presence of Ang-1 or Ang-2 and specific markers of polarized macrophages were analyzed by flow cytometry. Unlike classical polarizing cytokines, neither Ang-1 nor Ang-2 influenced expression of CD16, CD64, CD163, or CD200R, surface markers strongly up-regulated by IFN-γ, IL-10 and IL-4, respectively (Figure 2A–D) , . Additionally, neither Ang-1 nor Ang-2 significantly altered Tie2 protein or mRNA expression (Figure 2E–F). Together, these data observed suggest that macrophage Tie2 signaling is not sufficient to drive human macrophage polarization or initiate feedback loops regulating Ang or Tie2 expression.
Ang-1 and Ang-2 regulate cytokine production in M1 and M2 macrophages
We next examined the influence of Tie2 signaling on macrophage cytokine production. MΦIFN and MΦIL10 were stimulated in the absence or presence of Ang-1 or Ang-2, alone or in combination with TNF. In MΦIFN, levels of IL-6 production were elevated by Ang-1 in the absence of TNF (P<0.01) and by Ang-1 and Ang-2, in the presence of TNF (P<0.05 in both cases) (Figure 3A). In MΦIL10, both Ang-1 (P<0.05) and Ang-2 (P<0.05) enhanced TNF-induced IL-6 production (Figure 3B). We also analyzed expression of the anti-angiogenic factor TSP-2, as we had previously demonstrated that Ang-2, but not Ang-1, suppressed TSP-2 production in MΦGM-CSF) . TSP-2 levels were low in unstimulated MΦIFN, but were not influenced by Ang-1 or Ang-2, and stimulation of these cells with TNF suppressed TSP-2 production below detection levels (Figure 3C). Trends were observed toward suppressed spontaneous TSP-2 production by MΦIL10 in the presence of either Ang-1 or Ang-2 (Figure 3D). We also considered the influence of Ang-1 and Ang-2 on immunoregulatory cytokine IL-10, as previous work has shown that Ang-2 increased IL-10 production in TEMs . In MΦIFN, neither Ang-1 nor Ang-2 influenced IL-10 production by themselves or following stimulation with TNF (Figure 3E). However, in MΦIL10, Ang-1 significantly enhanced IL-10 production, while a trend was observed towards suppressed IL-10 production in the presence of Ang-2 (Figure 3F). Together, based on the parameters assessed, these data fail to indicate a general immunosuppressive role for Ang-1 and Ang-2 in the biology of differentiated macrophages in vitro.
Analyses of (A,B) IL-6, (C,D) TSP-2, and (E,F) IL-10 production in supernatants of macrophages differentiated in (A,C,E) IFN-γ or (B,D,F) IL-10 after 24 h incubation in medium alone or TNF-α (10 ng/ml) in the absence (white bars) or presence of Ang-1 (200 ng/ml, gray bars) or Ang-2 (200 ng/ml, black bars). Bars represent the means and SEM of 4–7 independent experiments. N.D., not detectable. *P<0.05, **P<0.01 versus cells not exposed to Ang-1 or Ang-2. Friedman test was used for statistical analyses.
Ang-1 and Ang-2 enhance macrophage expression of TNF-induced chemokines and pro-inflammatory cytokines
To gain further insight into the effects of Tie2 signaling on inflammatory properties of polarized macrophages, we explored the effects of macrophage stimulation with Ang-1 and Ang-2 on mRNA expression of 84 genes involved in angiogenic processes. MΦIFN, MΦGM-CSF, MΦM-CSF, and MΦIL10 displayed distinct profiles of gene expression, although unsupervised clustering of data revealed the highest relationship of gene expression patterns between M1 MΦIFN and MΦGM-CSF, and between M2 MΦM-CSF and MΦIL-10 (Figure 4A). Results obtained were in general agreement with previously published studies of gene expression during macrophage polarization as, for example, MΦIFN displayed the highest expression of CXCL9, CXCL10, CXCL11, and IL12A (Figure 4A) .
(A) mRNA expression profiles of 84 angiogenesis related genes in macrophages differentiated in GM-CSF, M-CSF, IFN-γ and IL-10 7 d (n = 3). Data is presented as an unsupervised clustergram. (B) Heat map analysis of mRNA levels of angiogenesis-related genes in macrophages differentiated in GM-CSF after 4 h incubation in medium alone or Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml).
We chose MΦGM-CSF, a commonly studied macrophage, and M1 MΦIFN and M2 MΦIL-10, as three functionally distinct types of macrophages expressing high levels of Tie2, for further analysis. Surprisingly, we were unable to identify additional genes which were reproducibly regulated by more than 2-fold in these macrophages following stimulation with either Ang-1 or Ang-2 alone (Figure 4B and data not shown). However, recent studies have indicated that while Ang-1 and Ang-2 are relatively weak regulators of gene expression in ECs and macrophages, these cytokines can cooperate in a synergistic fashion with TNF to regulate inflammatory gene induction , . We observed that of the 84 genes examined, 24 were induced by at least 2-fold by TNF in MΦIFN in 3 independent experiments, 19 in MΦGM-CSF, and 20 in MΦIL10 (Figure 5A). Interestingly, of the 30 different genes induced by TNF in the 3 polarization conditions, 11 were induced in all 3 macrophage types, although quantitative differences in basal and induced gene expression levels were readily apparent. When macrophages were stimulated with Ang-1 or Ang-2 in combination with TNF, we observed that Ang-1 and Ang-2 primarily influenced the subset of genes already regulated independently by TNF: 17 of 27 in MΦIFN, 11 of 19 in MΦGM-CSF, and 11 of 20 in MΦIL10 (Figure 5A and Table S1). Many of these genes were regulated in multiple types of macrophages. Ang-1 cooperated with TNF to significantly (P<0.05) increase mRNA levels of CXCL6, CXCL9, and IL6 in MΦGM-CSF, CXCL6, CXCL8, IL6 and IL12B in MΦIFN, CXCL3, CXCL5, CXCL6, CXCL8, IL6 and IL12B in MΦIL10. Ang-2 cooperated with TNF (P<0.05) to induce CXCL3, CXCL5, and CXCL9 in MΦIFN (Figure 5B). This synergism was independent of the polarization conditions, although the effects of Ang-1 and Ang-2 were most robust in MΦIFN. In both MΦIFN and MΦIL10, a dose-dependent cooperation of Ang-1 and Ang-2 (2–200 ng/ml) with TNF in the induction of IL6 and IL12B was observed (data not shown). Other genes were targeted by Ang-1 in a polarization-restricted fashion, such as CXCL2 in MΦIL10, and BMP2, EREG, SerpinE1 and BTG1 in MΦIFN (Figure S1A and Table S1). In almost all cases Ang-1 was more effective than Ang-2 in cooperating with TNF to induce gene expression (Figure 5B and Table S1), but in MΦIFN we identified a small subset of genes, CHGA, IL17F, NPPB, PROK1 and Tie1, which were upregulated only in the presence of TNF in combination with Ang-2 (Figure S1B).
(A) mRNA expression profiles angiogenesis related genes in macrophages differentiated with GM-CSF, IFN-γ or IL-10 after 4 h incubation in medium alone or TNF-α (10 ng/ml) in the absence or presence of Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml) (n = 3). Data is presented as a heat map where lowest mRNA expression is showed in dark blue and highest in yellow. (B) Analyses of chemokines and cytokines mRNA expression levels of selected genes analyzed in (A). Data is shown as log2 relative quantity respect to unstimulated cells, as described in material and methods. Bars represent the means and SEM of 3 independent experiments. *P<0.05, between stimulatory conditions. Friedman test was used for statistical analyses. (C) Multiplex analysis of protein production by macrophages differentiated in GM-CSF, IFN-γ, or IL-10 after 24 h incubation in medium alone or TNF (10 ng/ml) in the absence or presence of Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml). Bars represent the means and SEM of 6 independent experiments. *P<0.05, * P<0.01, between stimulatory conditions, #P<0.05, ##P<0.01, ### P<0.001, compared to unstimulated cells. Friedman test was used for statistical analyses.
We next sought to confirm the effects of Ang-1 and Ang-2 on TNF-induced gene expression in polarized macrophages at the protein level, selecting a subset of secreted products. Of the analytes assessed, only CCL-3 was significantly induced by TNF, in MΦGM-CSF and MΦIL10. However addition of Ang-1 in combination with TNF significantly induced secretion of CCL-7 and IL-12B in MΦGM-CSF, IL-8, CCL-3, CCL-7, and IL-12B in MΦIFN, and CXCL-5, CXCL-6, IL-8, CCL-3, CCL-7, and IL12-B in MΦIL10. Ang-2 cooperated with TNF to significantly induce production of IL-8, CCL-7, and IL-12B in MΦGM-CSF, IL-8, CCL-3, and IL-12B in MΦIFN, and CXCL-5, CXCL-6, IL-8, CCL-3, CCL7, and IL-12B in MΦIL10 (Figure 5C). Together these results suggest that Ang-1 and Ang-2 play highly overlapping roles in cooperating with TNF to induce pro-inflammatory cytokines and chemokines, independently of macrophage polarization conditions.
Ang-1 and Ang-2 cooperate with TNF to promote monocyte recruitment by macrophages
To determine if the effects we observed of Ang-1 and Ang-2 on macrophage chemokine production were sufficient to influence biological responses, we performed monocyte chemotaxis assays. In control experiments, inclusion of Ang-1 or Ang-2 alone in chemotaxis buffer resulted in a trend towards induced monocyte migration, which did not reach statistical significance (Figure 6A). Additionally, inclusion of TNF alone or in combination with Ang-1 or Ang-2 in conditioned medium from MΦGM-CSF, MΦIFN, or MΦIL10 failed to influence monocyte chemotaxis compared to conditioned medium alone (data not shown). Conditioned medium from macrophages stimulated with TNF increased monocyte migration compared to conditioned medium from control macrophages, although differences were not statistically significant (Figure 6B–D). Conditioned medium from macrophages stimulated with TNF in combination with Ang-1 induced significantly higher migration compared to medium from unstimulated macrophages, and macrophages stimulated with TNF alone in MΦIFN (P<0.05) and MΦIL10 (P<0.01). Ang-2 also enhanced TNF-induced monocyte migration compared to control conditioned medium (P<0.05) in MΦIFN and MΦIL10, and compared to TNF alone in MΦIFN (P<0.05).
Monocyte migration assays in response to (A) chemotaxis buffer alone (medium) or in combination with Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml), or chemotaxis buffer supplemented with conditioned medium of macrophages polarized in (B) GM-CSF, (C) IFN-γ or (D) IL-10 in the absence (unstimulated) or presence of 24 h stimulation with TNF (10 ng/ml) alone or in combination with Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml). Bars represent the means and SEM of 7–8 independent experiments. Experiments with each polarization condition were performed in parallel, but for ease of analysis are presented as 3 independent graphs. *P<0.05, * P<0.01, between stimulatory conditions, #P<0.05, ##P<0.01, ### P<0.001, compared to conditioned medium. Friedman test was used for statistical analyses.
Ang-1 and Ang-2 increase TNF-induced cytokine production via JAK/STAT signaling
To better understand the molecular basis by which Ang-1 and Ang-2 cooperate with TNF to drive gene expression in macrophages, we analyzed activation of key TNF-regulated signaling pathways in polarized macrophages. Ang-1 and Ang-2 alone failed to activate ERK, p38 or PKB in MΦIFN (Figure 7A). However, Ang-1 and Ang-2 did slightly reduce IκBα expression compared to unstimulated macrophages, although difference were only significant in macrophages stimulated with Ang-1 (P<0.05) (Figure 7A and Figure S2). In MΦIl-10, we found that Ang-1 and Ang-2 induced ERK activation at 60 min, but differences were only significant with Ang-1 (P<0.05) (Figure 7A and S2). Both Ang-1 and Ang-2 failed to modulate activation of p38, PKB or the IkBα expression in MΦIL10. TNF- induced activation of ERK was significantly enhanced (P<0.05), while both Ang-1 and Ang-2 cooperatively induced ERK activation in MΦIL10 (Figure 7B and S3). Ang-1 also extended TNF-dependent p38 activation in both MΦIFN and MΦIL10 (Figure 7B and Figure S3). No significant impact of Ang-1 or Ang-2 on TNF-induced IkBα degradation or PKB activation was observed in either macrophage population.
(A) Immunoblot analyses of cellular lysates obtained from IFN-γ or IL-10 differentiated macrophages stimulated for the indicated times (minutes) with TNF (10 ng/ml), Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml) for expression and phosphorylation (p) of the indicated signaling proteins. All data shown are from independent immunoblots performed on macrophage lysates obtained from a single donor and are representative of 4 individual experiments. (B) Immunoblot analyses of cellular lysates obtained from IFN-γ or IL-10 differentiated macrophages stimulated for the indicated times (minutes) with TNF (10 ng/ml) in the presence or absence of Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml) for expression and phosphorylation (p) of the indicated signaling proteins. All data shown are from independent immunoblots performed on macrophage lysates obtained from a single donor and are representative of 4 individual experiments. (C) STAT family DNA-binding activity in nuclear lysates obtained from macrophages stimulated for 2 h with TNF (10 ng/ml) in the absence or presence of Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml). Bars represent the means and SEM of 4 independent experiments. *P<0.05, **P<0.01. Friedman test was used for statistical analyses. (D) IL12B (upper panels) and IL6 mRNA (lower panels) expression levels in IFN-γ (left panels) and IL-10 –differentiated macrophages (right panels) after stimulation with medium alone or TNF (10 ng/ml) in the absence or presence of Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml), in the presence or absence of AG490 (10 µM). Values in are represented as percentage mRNA expression in the presence of AG490 in the defined conditions relative to expression in the absence of the inhibitor. Bars represent the means and SEM of 5 independent experiments. *P<0.05, **P<0.01. Friedman test was used for statistical analyses.
As the above experiments failed to give a clear indication as to how Ang-1 and Ang-2 might cooperate with TNF to regulate macrophage gene expression, we next examined a potential role for JAK/STAT signaling. TNF stimulation alone failed to influence activation of STAT-1, -3, -5A and 5B transcription factors in MΦIFN, but Ang-1 and Ang-2, in combination with TNF, significantly induced activation of STAT-1 and STAT-3, and enhanced binding activity of STAT-5A was also observered in the presence of Ang-1 (Figure 7C). In MΦIL10, TNF by itself could stimulate STAT-5A, while Ang-1 and Ang-2 both cooperated with TNF to induce activation of STAT-3 and STAT-5B (Figure 7C). Pharmacological inhibition of JAK/STAT signaling selectively and significantly abrogated contributions of Ang-1 and Ang-2 to macrophage IL-12B and IL-6 mRNA expression, while leaving basal and TNF-induced expression largely intact (Figure 7D). Collectively, these experiments provide evidence that while Ang-1 and Ang-2 similarly and slightly enhance many intracellular pathways common to TNF signaling in MΦIFN and MΦIL10, unique capacities of Tie2 to promote JAK/STAT activation contribute strongly to cooperative activation of macrophages by Ang-1 and Ang-2.
Myeloid Tie2 signaling plays an essential role in integrating angiogenesis and tissue remodeling to permit the growth of solid tumors , . Gene expression profiles have indicated that murine TEMs represent a distinct lineage commitment during monocyte differentiation, and restriction of myeloid Tie2 expression to a distinct subset of circulating human monocytes expressing genes associated with alternatively activated, M2-like macrophages has also been reported , . Prominent secreted gene products expressed by human TEMs include cathepsin B, MMP-9, and IL-10, and expression of these factors, important for angiogenesis and immune regulation, is reinforced by Ang-2 stimulation , , . Also, Ang-2 supports TEM activation of immunosuppressive Tregs, potentiating tumor growth in vivo, again indicating an anti-inflammatory function for Tie2 signaling in myeloid cells . However, we now provide direct evidence that human macrophage Tie2 expression is not restricted to cells derived from a distinct, preprogrammed circulating monocyte subset, but instead is highly dependent upon the cytokine milieu present during macrophage differentiation. Tie2 expression is readily detected on pro-inflammatory M1 (IFN-γ, GM-CSF), alternatively activated/wound healing M2 (M-CSF) and immunosuppressive M2 (IL-10)-differentiated macrophages, indicating that Tie2 expression is dissociated from macrophage functional classification.
Although Ang-1 and Ang-2 by themselves have weak to negligible effects on macrophage gene expression, these cytokines interact strongly with TNF to promote gene expression in both M1 and M2 macrophages. Thus, as is the case with ECs, myeloid Tie2 signaling appears to play an important role in integrating angiogenesis with local inflammatory cues . One relevant physiological consequence of macrophage Tie2 stimulation that we identify here is the generation of chemokines which attract peripheral blood monocytes. Disparate studies have previously indicated that Ang-1 and Ang-2 can directly promote monocyte and TEM recruitment by acting as intrinsic chemotactic factors , . Although we do observe trends towards monocyte recruitment by Ang-1 and Ang-2 alone, cultured medium from either M1 or M2 macrophages stimulated by Ang-1 or Ang-2 in combination with TNF reproducibly induced monocyte chemotactic responses. Thus in vivo, Ang-1 and Ang-2 might perpetuate inflammation both through direct effects on ECs, increasing vascular permeability, as well as stimulating chemokine production by macrophages in the tissue .
TEMs, especially within the context of Ang-2 stimulation, display an immunoregulatory phenotype, characterized by low expression of IL-12 and high expression of IL-10 –. In contrast, we find that in both M1 and M2 macrophages, Ang-2 synergizes with TNF to induce IL-12 expression, and by itself can suppress IL-10 production in M2 macrophages. The differences we observe between polarized macrophages and reported data on TEMs may simply reflect differences in gene regulation between monocytes and differentiated macrophages, or intrinsic properties of TEMs as a committed myeloid subpopulation. Remarkably, we find that for most genes studied, Ang-1 synergizes more potently with TNF than Ang-2 to drive macrophage gene expression. In contrast, TEMs are generally unresponsive to Ang-1, although limited effects of Ang-1 on CCL17 suppression and EGF induction have been reported . We find that Ang-1 and Ang-2 cooperate with TNF in a similar fashion to regulate gene expression in both M1 and M2 macrophages. However, we do note that Ang-2 can selectively regulate expression of some genes, such as CHGA, COL4A3, IL17F, TIE1, NPPB, and PROK1, indicating that as yet unidentified signaling pathways may be differentially regulated by Ang-1 and Ang-2.
Importantly, we show that the influence of Ang-1 and Ang-2 on TNF-induced gene expression in macrophages is largely dependent upon activation of JAK/STAT signaling pathways. It has previously been shown that Ang-1 can stimulate phosphorylation of STAT-1, STAT-3, STAT-5 and STAT-6 in human acute myeloid leukemia cells, although the physiological consequences of this were not identified . We observe that Ang-1 and Ang-2 activate STAT-1 and STAT-3 in MΦIFN, and STAT-3 and STAT-5 in MΦIL10. Although we have not assessed the specific contributions of each STAT member to IL-6 and IL-12 production in polarized macrophages, pharmacological inhibition of JAK proteins selectively inhibits the ability of Tie2 to cooperate with TNFR stimulation to regulate these gene products. We also note that both Ang-1 and Ang-2 slightly enhance TNF-dependent ERK and p38 activation. Activation of these kinases might also contribute to the effects of Ang-1 and Ang-2 on macrophage gene expression, either through signaling to additional transcription factors, or more direct interactions with JAK/STAT signaling. In this regard, mitogen-activated protein kinase cascades have been reported to directly phosphorylate STAT1 and STAT3 on serine residues, enhancing their DNA-binding activity –. Alternatively, JAK2 and JAK3 can stimulate ERK activation, providing another possible mechanism of interaction between TNF and Ang-1/Ang-2 signaling , .
We find that functional expression of Tie2 is a general feature of differentiated macrophages that, although regulated by polarization conditions, is not related to the functional phenotype of the macrophage. The data presented here offer not only an explanation for the dichotomous role of myeloid Tie2 in supporting both immunosuppression in tumor biology and immune activation in RA, but also highlights how specific pathological conditions might contribute to macrophage plasticity , . TAM and TEM M2-like functional characteristics are reinforced by tumor-derived cytokines , –, . Tumor-derived Ang-2, for example, increases TEM IL-10 production, while suppressing IL-12, together directly and indirectly hindering tumor rejection by effector T cells , . However, IFN-γ can convert TAMs into M1-like macrophages which promote tumor regression, and it would be of interest to determine if myeloid Tie2 signaling can modulate this effect . We have previously observed Tie2 activation in synovial macrophages of patients with RA and psoriatic arthritis, a form of spondyloarthritis (SpA) , and in RA and other forms of arthritis, distinct macrophages with M2-like and mixed M1/M2 phenotypes are observed , , . In SpA, expression of the M2 marker CD163 is elevated in intimal lining layer macrophages compared to RA, consistent with a relative lack of M1-polarizing cytokines in SpA synovial fluid , . Despite this, in both RA and SpA TNF plays a prominent role in driving disease, and synovial macrophage numbers and cytokine products are closely associated with disease activity and patient responses to treatment, indicating contributions to inflammation and joint destruction independent of M1/M2 phenotype , –. Our observation that Ang-1 and Ang-2 alone have negligible effects on macrophage polarization and gene expression is reminiscent of studies showing that immune complexes, generally associated with M2 polarization in murine monocytes, also fail to drive human macrophage polarization, while enhancing TLR-induced IL-10 production in both M1 and M2 macrophages . Attempts to understand macrophage plasticity in vivo have provided evidence that initiation and resolution of inflammation, and tissue damage and repair, are accompanied by sequential replacement of tissue macrophages with M1 and M2 macrophages , . Our data suggest that otherwise neutral “costimulatory” macrophage agonists, such as Ang-1/Ang-2 and immune complexes, accompanied by changing expression of prominent macrophage agonists, such as TNF, IFNγ, TLR ligands, IL-10 and IL-4, may override macrophage phenotypic and functional characteristics to regulate tissue damage and repair.
Ang-1 and Ang-2 enhance the TNF-induced expression of angiogenesis related genes. (A–D) mRNA expression of angiogenesis related genes in macrophages differentiated with GM-CSF, IFN-γ or IL-10 after 4 h incubation in medium alone or TNF-α (10 ng/ml) in the absence or presence of Ang-1 (200 ng/ml) or Ang-2 (200 ng/ml) (n = 3). Data is shown as relative quantity respect to unstimulated cells, as described in material and methods. Bars represent the means and SEM of 3 independent experiments. *P<0.05, between stimulatory conditions. Friedman test was used for statistical analyses.
Effects of Ang-1 and Ang-2 on signaling pathways in polarized macrophages. Data represents densitometric analysis of immunoblots shown in Figure 7A and are shown as relative expression (a.u.) with respect to unstimulated cells, as described in material and methods. Bars represent the means and SEM of 4 independent experiments. #P<0.05, ##P<0.01, compared to medium. Friedman test was used for statistical analyses.
Effects of Ang-1 and Ang-2 on TNF-dependent signaling pathways in polarized macrophages. Data represents densitometric analysis of immunoblots shown in Figure 7B and are shown as relative expression (a.u.) with respect to unstimulated cells, as described in material and methods. Bars represent the means and SEM of 4 independent experiments. *P<0.05, between stimulatory conditions, #P<0.05, ##P<0.01, ### P<0.001, compared to unstimulated conditions. Friedman test was used for statistical analyses.
We would like to thank T. Dekker, B.S. Dierdorp, and Dr. R. Lutter (AMC, Amsterdam) for performing multiplex analyses, and Prof. Dr. M. de Winther (AMC, Amsterdam) for critical reading of the manuscript.
Conceived and designed the experiments: SG SK JH DLB PPT KAR. Performed the experiments: SG SK CAA BMF LH IEvE. Analyzed the data: SG SK CAA BMF LH IEvE JH DLB PPT KAR. Wrote the paper: SG SK KAR.
- 1. Thomas M, Augustin HG (2009) The role of the Angiopoietins in vascular morphogenesis. Angiogenesis 12: 125–137.
- 2. Huang H, Bhat A, Woodnutt G, Lappe R (2010) Targeting the ANGPT-TIE2 pathway in malignancy. Nat Rev Cancer 10: 575–585.
- 3. Suri C, McClain J, Thurston G, McDonald DM, Zhou H, et al. (1998) Increased vascularization in mice overexpressing angiopoietin-1. Science 282: 468–471.
- 4. Thurston G, Suri C, Smith K, McClain J, Sato TN, et al. (1999) Leakage-resistant blood vessels in mice transgenically overexpressing angiopoietin-1. Science 286: 2511–2514.
- 5. Tammela T, Saaristo A, Lohela M, Morisada T, Tornberg J, et al. (2005) Angiopoietin-1 promotes lymphatic sprouting and hyperplasia. Blood 105: 4642–4648.
- 6. Fukuhara S, Sako K, Minami T, Noda K, Kim HZ, et al. (2008) Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1. Nat Cell Biol 10: 513–526.
- 7. Saharinen P, Eklund L, Miettinen J, Wirkkala R, Anisimov A, et al. (2008) Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell-cell and cell-matrix contacts. Nat Cell Biol 10: 527–537.
- 8. Yuan HT, Khankin EV, Karumanchi SA, Parikh SM (2009) Angiopoietin 2 is a partial agonist/antagonist of Tie2 signaling in the endothelium. Mol Cell Biol 29: 2011–2022.
- 9. Felcht M, Luck R, Schering A, Seidel P, Srivastava K, et al. (2012) Angiopoietin-2 differentially regulates angiogenesis through TIE2 and integrin signaling. J Clin Invest 122: 1991–2005.
- 10. Chen JX, Chen Y, DeBusk L, Lin W, Lin PC (2004) Dual functional roles of Tie-2/angiopoietin in TNF-alpha-mediated angiogenesis. Am J Physiol Heart Circ Physiol 287: H187–H195.
- 11. Roviezzo F, Tsigkos S, Kotanidou A, Bucci M, Brancaleone V, et al. (2005) Angiopoietin-2 causes inflammation in vivo by promoting vascular leakage. J Pharmacol Exp Ther 314: 738–744.
- 12. Fiedler U, Reiss Y, Scharpfenecker M, Grunow V, Koidl S, et al. (2006) Angiopoietin-2 sensitizes endothelial cells to TNF-alpha and has a crucial role in the induction of inflammation. Nat Med 12: 235–239.
- 13. De Palma M, Venneri MA, Galli R, Sergi SL, Politi LS, et al. (2005) Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8: 211–226.
- 14. Venneri MA, De Palma M, Ponzoni M, Pucci F, Scielzo C, et al. (2007) Identification of proangiogenic TIE2-expressing monocytes (TEMs) in human peripheral blood and cancer. Blood 109: 5276–5285.
- 15. Pucci F, Venneri MA, Biziato D, Nonis A, Moi D, et al. (2009) A distinguishing gene signature shared by tumor-infiltrating Tie2-expressing monocytes, blood “resident” monocytes, and embryonic macrophages suggests common functions and developmental relationships. Blood 114: 901–914.
- 16. Coffelt SB, Tal AO, Scholz A, De Palma M, Patel S, et al. (2010) Angiopoietin-2 regulates gene expression in TIE2-expressing monocytes and augments their inherent proangiogenic functions. Cancer Res 70: 5270–5280.
- 17. Coffelt SB, Chen YY, Muthana M, Welford AF, Tal AO, et al. (2011) Angiopoietin 2 stimulates TIE2-expressing monocytes to suppress T cell activation and to promote regulatory T cell expansion. J Immunol 186: 4183–4190.
- 18. Mosser DM, Edwards JP (2008) Exploring the full spectrum of macrophage activation. Nat Rev Immunol 8: 958–969.
- 19. Krausz S, Garcia S, Ambarus CA, de Launay D, Foster M, et al. (2012) Angiopoietin-2 promotes inflammatory activation of human macrophages and is essential for murine experimental arthritis. Ann Rheum Dis 71: 1402–1417.
- 20. Hamilton JA, Tak PP (2009) The dynamics of macrophage lineage populations in inflammatory and autoimmune diseases. Arthritis Rheum 60: 1210–1221.
- 21. Martinez FO, Sica A, Mantovani A, Locati M (2008) Macrophage activation and polarization. Front Biosci 13: 453–461.
- 22. Ambarus CA, Krausz S, van Eijk M, Hamann J, Radstake TR, et al. (2012) Systematic validation of specific phenotypic markers for in vitro polarized human macrophages. J Immunol Methods 375: 196–206.
- 23. Koning N, van Eijk M, Pouwels W, Brouwer MS, Voehringer D, et al. (2010) Expression of the inhibitory CD200 receptor is associated with alternative macrophage activation. J Innate Immun 2: 195–200.
- 24. Martinez FO, Gordon S, Locati M, Mantovani A (2006) Transcriptional profiling of the human monocyte-to-macrophage differentiation and polarization: new molecules and patterns of gene expression. J Immunol 177: 7303–7311.
- 25. Murdoch C, Tazzyman S, Webster S, Lewis CE (2007) Expression of Tie-2 by human monocytes and their responses to angiopoietin-2. J Immunol 178: 7405–7411.
- 26. Ahmad S, Cudmore MJ, Wang K, Hewett P, Potluri R, et al. (2010) Angiopoietin-1 induces migration of monocytes in a tie-2 and integrin-independent manner. Hypertension 56: 477–483.
- 27. Hatfield K, Oyan AM, Ersvaer E, Kalland KH, Lassalle P, et al. (2009) Primary human acute myeloid leukaemia cells increase the proliferation of microvascular endothelial cells through the release of soluble mediators. Br J Haematol 144: 53–68.
- 28. Haq R, Halupa A, Beattie BK, Mason JM, Zanke BW, et al. (2002) Regulation of erythropoietin-induced STAT serine phosphorylation by distinct mitogen-activated protein kinases. J Biol Chem 277: 17359–17366.
- 29. Wierenga AT, Vogelzang I, Eggen BJ, Vellenga E (2003) Erythropoietin-induced serine 727 phosphorylation of STAT3 in erythroid cells is mediated by a MEK-, ERK-, and MSK1-dependent pathway. Exp Hematol 31: 398–405.
- 30. Zhang Y, Cho YY, Petersen BL, Zhu F, Dong Z (2004) Evidence of STAT1 phosphorylation modulated by MAPKs, MEK1 and MSK1. Carcinogenesis 25: 1165–1175.
- 31. Winston LA, Hunter T (1995) JAK2, Ras, and Raf are required for activation of extracellular signal-regulated kinase/mitogen-activated protein kinase by growth hormone. J Biol Chem 270: 30837–30840.
- 32. Kim S, Choi JH, Lim HI, Lee SK, Kim WW, et al. (2009) EGF-induced MMP-9 expression is mediated by the JAK3/ERK pathway, but not by the JAK3/STAT-3 pathway in a SKBR3 breast cancer cell line. Cell Signal 21: 892–898.
- 33. Murray PJ, Wynn TA (2011) Protective and pathogenic functions of macrophage subsets. Nat Rev Immunol 11: 723–737.
- 34. Sica A, Mantovani A (2012) Macrophage plasticity and polarization: in vivo veritas. J Clin Invest 122: 787–795.
- 35. Duluc D, Delneste Y, Tan F, Moles MP, Grimaud L, et al. (2007) Tumor-associated leukemia inhibitory factor and IL-6 skew monocyte differentiation into tumor-associated macrophage-like cells. Blood 110: 4319–4330.
- 36. Duluc D, Corvaisier M, Blanchard S, Catala L, Descamps P, et al. (2009) Interferon-gamma reverses the immunosuppressive and protumoral properties and prevents the generation of human tumor-associated macrophages. Int J Cancer 125: 367–373.
- 37. Ambarus CA, Noordenbos T, de Hair MJ, Tak PP, Baeten DL (2012) Intimal lining layer macrophages but not synovial sublining macrophages display an IL-10 polarized-like phenotype in chronic synovitis. Arthritis Res Ther 14: R74.
- 38. Vandooren B, Noordenbos T, Ambarus C, Krausz S, Cantaert T, et al. (2009) Absence of a classically activated macrophage cytokine signature in peripheral spondylarthritis, including psoriatic arthritis. Arthritis Rheum 60: 966–975.
- 39. Tak PP, Smeets TJ, Daha MR, Kluin PM, Meijers KA, et al. (1997) Analysis of the synovial cell infiltrate in early rheumatoid synovial tissue in relation to local disease activity. Arthritis Rheum 40: 217–225.
- 40. Haringman JJ, Gerlag DM, Zwinderman AH, Smeets TJ, Kraan MC, et al. (2005) Synovial tissue macrophages: a sensitive biomarker for response to treatment in patients with rheumatoid arthritis. Ann Rheum Dis 64: 834–838.
- 41. Baeten D, Kruithof E, De Rycke L, Boots AM, Mielants H, et al. (2005) Infiltration of the synovial membrane with macrophage subsets and polymorphonuclear cells reflects global disease activity in spondyloarthropathy. Arthritis Res Ther 7: R359–R369.
- 42. Ambarus CA, Santegoets KC, van Bon L, Wenink MH, Tak PP, et al. (2012) Soluble immune complexes shift the TLR-induced cytokine production of distinct polarized human macrophage subsets towards IL-10. PLoS One 7: e35994.
- 43. Lumeng CN, Bodzin JL, Saltiel AR (2007) Obesity induces a phenotypic switch in adipose tissue macrophage polarization. J Clin Invest 117: 175–184.
- 44. Lumeng CN, Delproposto JB, Westcott DJ, Saltiel AR (2008) Phenotypic switching of adipose tissue macrophages with obesity is generated by spatiotemporal differences in macrophage subtypes. Diabetes 57: 3239–3246.