Glucocorticoids (GCs) have been used for more than 50 y as immunosuppressive drugs, yet their efficacy in macrophage-dominated disorders, such as chronic obstructive pulmonary disease, is debated. Little is known how long-term GC treatment affects macrophage responses in inflammatory conditions. In this study, we compared the transcriptome of human macrophages, matured in the presence or absence of fluticasone propionate (FP), and their ability to initiate or sustain classical activation, mimicked using acute LPS and chronic IFN-γ stimulation, respectively. We identified macrophage gene expression networks, modulated by FP long-term exposure, and specific patterns of IFN-γ– and LPS-induced genes that were resistant, inhibited, or exacerbated by FP. Results suggest that long-term treatment with GCs weakens adaptive immune signature components of IFN-γ and LPS gene profiles by downmodulating MHC class II and costimulatory molecules, but strengthens innate signature components by maintaining and increasing expression of chemokines involved in phagocyte attraction. In a mouse model of chronic obstructive pulmonary disease, GC treatment induced higher chemokine levels, and this correlated with enhanced recruitment of leukocytes. Thus, GCs do not generally suppress macrophage effector functions, but they cause a shift in the innate–adaptive balance of the immune response, with distinct changes in the chemokine–chemokine receptor network.

Macrophages are innate immune cells with well-established roles in tissue homeostasis, primary response to pathogens, coordination of adaptive immunity, and even wound repair (1, 2). Macrophages accomplish these varied roles by adapting their gene and protein expression programs in response to endogenous and exogenous environmental cues, such as cytokines and pathogen-associated molecular patterns. The ability of macrophages to change their gene and protein signatures falls under the umbrella of a developing concept: macrophage plasticity (3). Plasticity is naturally the basis of macrophage heterogeneity in basal and inflammatory conditions; it is also the basis of worldwide efforts to treat diseases by subverting aberrant macrophage activation. Macrophage-mediated inflammation is increasingly recognized as contributing to chronic inflammatory disorders, such as chronic obstructive pulmonary disease (COPD), severe asthma, rheumatoid arthritis, and multiple sclerosis (4).

The best-characterized macrophage activation pathway has been called classical macrophage activation, and it is triggered by the recognition of pathogen-associated molecular patterns by TLRs, supported by further elicitation of IFN pathways (2). Classically activated macrophages produce microbicidal enzymes, such as inducible NO synthase, inflammatory cytokines, such as TNF, IL-6, IL-1β, IL-12, and IFN-β, various chemoattractants, and matrix metallopeptidases (MMPs). These mediators support IFN-γ–mediated Th1 responses, which are important for sustained antimicrobial activity but are deleterious in chronic diseases, because they lead to tissue damage and chronic inflammation.

To limit macrophage activation and restore homeostasis, endogenous glucocorticoids (GCs) induce anti-inflammatory programs via the GC receptor (GCR) (5, 6). GCR agonists are prescribed for a wide range of inflammatory disorders, and their use is crucial to treat diseases of the respiratory tract, especially asthma (6). Drugs such as prednisolone, dexamethasone, budesonide, and fluticasone propionate (FP) share great GCR specificity albeit different pharmacodynamics (7). The remarkable efficacy of GCs in suppressing inflammation and decreasing lymphocyte activation, proliferation, and survival is well appreciated. However, our knowledge of the interactions of GC and inflammatory pathways in human macrophages remains limited (8), and recent studies have shown that GCs may have little effect in controlling inflammation in macrophage-dominated diseases, and their long-term use is associated with persisting complications (916).

In this study, we investigated from a whole genome point of view the effects of chronic GCR ligation on the maturation of human macrophages and on their ability to initiate and sustain classical activation that is important for Th1 inflammation, dissected by acute and chronic stimulation with the TLR4 ligand LPS and IFN-γ. We show that long-term treatment with FP modulates classical activation in human inflammatory macrophage models in a complex and not solely suppressive manner by maintaining or even increasing the expression of chemokines involved in leukocyte recruitment, while clearly affecting their ability to initiate adaptive immune response programs by decreasing expression of Ag-presenting MHC class II, costimulatory, and cell adhesion molecules. To confirm these results in vivo, we evaluated the effects of GCR ligation in a mouse model of COPD-like airways disease, induced by chronic exposure to tobacco smoke. In this model, GCR ligation led to higher levels of pulmonary chemokines, which correlated with more cellular infiltration, including macrophages. With this global approach, we have identified pathways that remain active or worsen upon GC treatment and have confirmed these pathways to play a role in vivo as well. These results could explain in part the lack of efficacy of GC treatment in macrophage-dominated disorders.

Monocytes were isolated from buffy coats of healthy blood donors (Sanquin, Amsterdam, The Netherlands) by Lymphoprep (Axis-Shield, Oslo, Norway) density gradient centrifugation followed by plastic adherence in IMDM (Lonza, Basel, Switzerland) containing 10% human pool serum, 1% penicillin/streptomycin (Invitrogen, San Diego, CA), and 0.25% ciprofloxacin (Bayer, Leverkusen, Germany). PBMCs were incubated, and nonadherent cells were removed after 2 h incubation by extensive washing with PBS. Subsequently, adherent monocytes were detached using PBS containing 10 mM EDTA at room temperature and plated for macrophage maturation at a density of 1 × 105 cells/cm2 in IMDM containing 10% human pool serum, 1% penicillin/streptomycin, and 0.25% ciprofloxacin.

For protein measurement, monocyte-derived macrophages (MDMs) were generated over a period of 3–7 d (at least n = 3). Where relevant, the media was supplemented with FP (100 nM unless indicated otherwise; GlaxoSmithKline, Brentford, U.K.), dexamethasone (100 nM unless indicated otherwise; Sigma-Aldrich, St. Louis, MO), IFN-γ (12.5 ng/ml; Sigma-Aldrich), IL-4 (50 ng/ml; Peprotech, Rocky Hill, NY), or IL-10 (50 ng/ml; Peprotech). For microarray analysis of the acute endotoxin challenge model, MDMs were cultured for 7 d with FP (100 nM) or no addition. At day 7, LPS (10 ng/ml; Salmonella minnesota R595; Alexis Biochemicals, Lausen, Switzerland) was added 6 h prior to harvesting the cells (n = 3). In the chronic inflammation model, MDMs were cultured for 5 d with FP (100 nM), IFN-γ (12.5 ng/ml), both, or no addition, followed by an additional 2 d with the same culture conditions or FP plus IFN-γ (n = 2). The FP profiles were extracted from n = 5.

RNA was isolated using a Macherey-Nagel (Düren, Germany) total RNA isolation kit NucleoSpin II, following the manufacturer’s instructions. Quality control, RNA labeling, hybridization onto the Illumina HumanHT-12 v4 Expression BeadChips covering 31,000 annotated genes with >47,000 probes (Illumina, San Diego, CA), and data extraction was performed at ServiceXS (Leiden, The Netherlands).

Expression results were normalized by quantiles, using R (Bioconductor, Seattle, WA). Genes differentially expressed were selected using ANOVA, with a p value, based on permutations, <0.01 and a false discovery rate <0.05, using Multiple Experiment Viewer (MEV; Dana-Farber Cancer Institute, Boston, MA). Principal component analysis was performed on the samples using MEV. Genes were filtered for a fold ≥2 in at least one condition. The transcripts were further organized with K-means clustering after median centering using MEV. Pathways and gene ontology overrepresentation were assessed using Ingenuity Pathways Analysis (IPA; Ingenuity Systems, Redwood City, CA) (17). IPA database collects data from literature; the database covers all cell types and species, and for this study we focused our analysis on data generated on human cells backed up by experimental evidence. Data compliant for Minimum Information About a Microarray Experiment for the datasets used in this study are deposited in Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo/) profiles dataset GSE49852. This dataset is part of a superseries—GSE35495—that covers microarrays over the spectrum of macrophage activation, including IFN-γ, IL-4, IL-10, IL-13, and dexamethasone (18).

For comparison, CEL files generated by Ehrchen et al. (19) in a microarray analysis of monocytes stimulated with FP for 16 h were summarized as Robust Multiarray Averaging, and quantiles were normalized using Affy package, R.

For flow cytometric determination of cell surface marker expression, cells were fixed in 1% paraformaldehyde for at least 10 min at room temperature and resuspended in cold FACS buffer consisting of PBS, 0.5% BSA, 10% 0.13 M (pH 7) trisodium citrate dehydrate, and 4 mM EDTA. Immunostaining was performed using mouse mAbs of the IgG1 isotype, unless stated otherwise, specific for: CD32 (3D3-PE), CD45 (2D1-PerCP), CD80 (L307.4-PE), CD86 (FUN-1-FITC), CD163 (GHI/61-PE), CD206 (19.2-APC; all from BD Biosciences, San Jose, CA), CD16 (LNK16-AF647), CD200R (OX108-AF647; both AbD Serotec, Oxford, U.K.), CD64 (10.1-AF488), HLA-DR (IgG2a, L243-AF488; both Biolegend San Diego, CA), CD14 (61D3-PerCP-Cy5.5; eBioscience, San Diego, CA). Mouse isotype control Abs were IgG1 (clone MOPC-21-PerCP-Cy5.5, PE, or APC; BD Biosciences) and IgG2a (MOPC-173-AF488; BioLegend). Flow cytometric analysis was performed using a FACSCanto II (BD Biosciences) and FlowJo (Tree Star, Ashland, OR). Statistics were calculated using one-way or two-way ANOVA with Bonferroni correction for multiple comparisons.

Male C57BL/6j mice (aged 8–10 wk, n = 6 per group) were obtained from Harlan (Zeist, The Netherlands) and were held under specific pathogen-free conditions. The mice were fed ad libitum with standard food and water and were held under specific pathogen-free conditions in groups of 6 mice per cage. The animal procedures, approved by the Institutional Animal Care and Use Committee of the University of Groningen (application number 2625), were performed under strict governmental and international guidelines.

For the induction of a COPD-like phenotype, we used a model of cigarette smoke–induced lung inflammation as published previously (20, 21). In short, mice were exposed nasally to mainstream smoke for 5 d/wk, two sessions per day for 4 mo. As a break-in period, three puffs of smoke two times from 2R1 Reference Cigarettes (University of Kentucky, Lexington, KY) were administered on the first 3 d. Each day thereafter, the number of puffs was increased with one puff up to 24 puffs two times. For the remainder of the 4 mo, 24 puffs two times were administered daily, which equals four cigarettes. The smoking machine was checked for the delivery of total particulate matter as described by Griffith et al. (22, 23) and calibrated before every smoking session to ensure accurate and standardized smoke delivery. On average, each mouse was exposed to 6.4 ± 0.1 mg of total particulate matter per smoking session, which is comparable to what has been described in other studies using this setup (22, 23). Actual smoke exposure was assessed by measuring carboxyhemoglobin levels. Smoking mice on average had carboxyhemoglobin levels of 23.0 ± 1.7% directly after smoking, which is common in mice directly after smoking (24). Control nonexposed mice had levels of 2.0 ± 0.5%.

Smoking mice were treated nasally with nebulized budesonide (Pulmicort Respules; a gift from AstraZeneca, Zoetermeer, The Netherlands) once per day, five times per week for 4 mo. Pulmicort Respules were diluted to 0.125 mg/ml, and an aerosol was delivered to a Perspex exposure chamber (9 l) with a Aeroneb nebulizer (type Lab; Aerogen, Galway, Ireland) for 85 s. Six mice were then allowed to breath in the aerosol for 1 min by nasal exposure. This procedure was repeated for each batch of six mice. The estimated dose given to the animals was 0.2 μg/mouse, which is equivalent to a 500-μg dose for an adult person. Untreated smoking mice were exposed to saline in a similar manner.

Single-cell suspensions were obtained from lungs for flow cytometric analysis as described previously (25). Cells were quantified using a Coulter Counter Z1 (Coulter, Hialeah, FL). Expression of CD3, CD4, B220, GR1, F4/80, and CD11c was examined with flow cytometry to determine the frequencies of T and B lymphocytes, neutrophils, and macrophages. All Abs were obtained from BD Biosciences and were conjugated with fluorochromes. T cells were defined by positive staining for CD3 and B cells by positive staining for B220. Neutrophils were defined by bright, positive staining for Gr1 and intermediate staining for CD11c. Macrophages were defined by positive staining for F4/80 and CD11c. Cell populations (4 × 104 events) were analyzed using an Epics Elite flow cytometer (Coulter), and data analysis was performed using FlowJo.

Lung tissue was homogenized (20% w/v) in 50 mm Tris-HCl buffer (pH 7.5), containing 150 mm NaCl, 0.002% Tween-20, and protease inhibitor (Sigma-Aldrich), and was subsequently centrifuged at 12,000 × g for 10 min to remove any insoluble material. The concentrations of cytokines were measured by multiplex ELISA (Lincoplex Systems, St. Charles, MO) on a Luminex 100 system using Starstation software (Applied Cytometry Systems, Sheffield, U.K.).

Statistical analyses were done with Mann–Whitney U test unless stated otherwise.

GC analogs are especially important for the treatment of chronic diseases (6); however, the effects of chronic GCR stimulation at the whole genome level in human macrophages are hitherto not described. We matured human peripheral monocytes of five donors into macrophages for 7 d in the presence of FP, which has the longest GC half-life in vitro (7). At day 7, we conducted whole genome analysis using microarrays covering 31,000 annotated genes and found 165 unique genes significantly regulated by FP (Table I, Supplemental Table I) (2631). Among the genes most upregulated by FP were GLDN (gliomedin), RNASE1 (RNase 1), and ADORA3 (adenosine A3 receptor), the latter was also upregulated in monocytes (Table II) (19), whereas CCL22, FCN1 (ficolin), and MMP7 were the most downregulated. Both gliomedin and ficolin are collagen domain proteins. Although ficolin can recognize common carbohydrate residues found in microbes (26), a role for gliomedin in macrophages is not described. Secreted RNase 1 is able to regulate the metabolism of granulation tissue fibroblasts by increasing RNA degradation, affecting metabolic state (27), which can also have consequences for the plasticity of the macrophage response. ADORA3 also is one of the most strongly upregulated genes in smoking-dependent reprogramming of alveolar macrophages in COPD (28). Adenosine A3 receptor activation results in anti-inflammatory effects with a reduction in IL-1β, TNF, and IL-6, and decreased superoxide production in other phagocytes (29, 30). MMP7 (matrilysin) sheds proteoglycan–chemokine complexes from the mucosal surface and confines neutrophil influx to sites of acute lung injury (31). These highly regulated biomarkers may be of relevance to monitor drug efficacy in patients.

Table I.
Selection of genes modulated by chronic FP treatment in human macrophages
SymbolGene NameTypeFold Change
GLDN Gliomedin Other 15.467 
RNASE1 RNase, RNase A family, 1 (pancreatic) Enzyme 9.149 
ADORA3 Adenosine A3 receptor GPCR 7.422 
TFPI Tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor) Other 7.049 
FKBP5 FK506 binding protein 5 Enzyme 5.429 
IL1R2 IL-1 receptor, type II TM receptor 4.781 
CD163 CD163 molecule TM receptor 3.716 
HTRA1 HtrA serine peptidase 1 Peptidase 3.671 
ADAMDEC1 ADAM-like, decysin 1 Peptidase 3.018 
CCL8 CCL8 Cytokine 2.231 
CCL7 CCL 7 Cytokine 2.222 
PTGER2 PGE receptor 2 (subtype EP2) GPCR 2.152 
MMP19 MMP19 Peptidase 2.057 
HLA-F MHC, class I, F TM receptor −2.071 
HLA-DRA MHC, class II, DR α TM receptor −2.107 
CD4 CD4 molecule TM receptor −2.212 
PLAU Plasminogen activator, urokinase Peptidase −2.392 
CLEC4A C-type lectin domain family 4, member A TM receptor −2.417 
HLA-DPB1 MHC, class II, DP β 1 TM receptor −2.417 
HLA-DOA MHC, class II, DO α TM receptor −2.764 
HLA-DPA1 MHC, class II, DP α 1 TM receptor −3.145 
CD74 CD74 molecule, MHC, class II invariant chain TM receptor −3.185 
TNFSF13B TNF (ligand) superfamily, member 13b Cytokine −4.029 
HLA-DRB4 MHC, class II, DR β 4 TM receptor −4.079 
MMP9 MMP9 (gelatinase B) Peptidase −4.871 
HLA-DQA1 MHC, class II, DQ α 1 TM receptor −5.158 
MMP7 MMP7 (matrilysin) Peptidase −5.686 
FCN1 Ficolin (collagen/fibrinogen domain containing) 1 Other −6.933 
CCL22 CCL22 Cytokine −13.365 
SymbolGene NameTypeFold Change
GLDN Gliomedin Other 15.467 
RNASE1 RNase, RNase A family, 1 (pancreatic) Enzyme 9.149 
ADORA3 Adenosine A3 receptor GPCR 7.422 
TFPI Tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor) Other 7.049 
FKBP5 FK506 binding protein 5 Enzyme 5.429 
IL1R2 IL-1 receptor, type II TM receptor 4.781 
CD163 CD163 molecule TM receptor 3.716 
HTRA1 HtrA serine peptidase 1 Peptidase 3.671 
ADAMDEC1 ADAM-like, decysin 1 Peptidase 3.018 
CCL8 CCL8 Cytokine 2.231 
CCL7 CCL 7 Cytokine 2.222 
PTGER2 PGE receptor 2 (subtype EP2) GPCR 2.152 
MMP19 MMP19 Peptidase 2.057 
HLA-F MHC, class I, F TM receptor −2.071 
HLA-DRA MHC, class II, DR α TM receptor −2.107 
CD4 CD4 molecule TM receptor −2.212 
PLAU Plasminogen activator, urokinase Peptidase −2.392 
CLEC4A C-type lectin domain family 4, member A TM receptor −2.417 
HLA-DPB1 MHC, class II, DP β 1 TM receptor −2.417 
HLA-DOA MHC, class II, DO α TM receptor −2.764 
HLA-DPA1 MHC, class II, DP α 1 TM receptor −3.145 
CD74 CD74 molecule, MHC, class II invariant chain TM receptor −3.185 
TNFSF13B TNF (ligand) superfamily, member 13b Cytokine −4.029 
HLA-DRB4 MHC, class II, DR β 4 TM receptor −4.079 
MMP9 MMP9 (gelatinase B) Peptidase −4.871 
HLA-DQA1 MHC, class II, DQ α 1 TM receptor −5.158 
MMP7 MMP7 (matrilysin) Peptidase −5.686 
FCN1 Ficolin (collagen/fibrinogen domain containing) 1 Other −6.933 
CCL22 CCL22 Cytokine −13.365 

GPCR, G protein-coupled receptor; MMP, matrix metallopeptidase; TM, transmembrane.

Table II.
Thirty-five genes modulated by FP treatment in human monocytes (16 h) and macrophages (5 d)
SymbolGene NameTypeFold Change MonocytesFold Change Macrophages
ADM Adrenomedullin Other 2.1 3.7 
ADORA3 Adenosine A3 receptor GPCR 4.7 7.4 
ALOX15B Arachidonate 15-lipoxygenase, type B Enzyme 4.8 2.7 
C1QB Complement component 1, q subcomponent, B chain Other 3.1 2.2 
CD163 CD163 molecule TM receptor 2.4 3.7 
DDIT4 DNA-damage-inducible transcript 4 Other 5.1 4.0 
FBLN5 Fibulin 5 Other 2.2 4.5 
FKBP5 FK506 binding protein 5 Enzyme 6.8 5.4 
HPGD hydroxypg dehydrogenase 15-(NAD) Enzyme 2.2 3.1 
HTRA1 HtrA serine peptidase 1 Peptidase 3.2 3.7 
IL1R2 IL-1 receptor, type II TM receptor 12.5 4.8 
IRS2 Insulin receptor substrate 2 Enzyme 3.8 2.4 
MERTK c-mer proto-oncogene tyrosine kinase Kinase 4.3 3.0 
METTL7A Methyltransferase like 7A Other 2.3 2.4 
MFGE8 Milk fat globule-EGF factor 8 protein Other 2.4 6.4 
PDK4 Pyruvate dehydrogenase kinase, isozyme 4 Kinase 4.8 3.1 
RNASE1 RNase, RNase A family, 1 (pancreatic) Enzyme 2.2 9.1 
SERPINE1 Serpin peptidase inhibitor, clade E (nexin, plasminogen activator inhibitor type 1), member 1 Other 2.7 2.6 
SESN1 Sestrin 1 Other 4.6 2.6 
SLA Src-like-adaptor Other 2.1 2.2 
SLC16A6 Solute carrier family 16, member 6 (monocarboxylic acid transporter 7) Transporter 2.9 3.4 
SLC1A3 Solute carrier family 1 (glial high affinity glutamate transporter), member 3 Transporter 5.2 2.2 
SRPX Sushi-repeat containing protein, X-linked Other 5.3 5.9 
TBC1D16 TBC1 domain family, member 16 Other 7.7 2.6 
TCN2 Transcobalamin II Transporter 2.9 2.1 
TFPI Tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor) Other 4.1 7.0 
TPST1 Tyrosylprotein sulfotransferase 1 Enzyme 4.5 3.0 
TSC22D3 TSC22 domain family, member 3 Other 2.4 4.1 
VSIG4 V-set and Ig domain containing 4 Other 11.2 6.0 
ZCCHC6 Zinc finger, CCHC domain containing 6 Enzyme 2.6 2.3 
CHI3L1 Chitinase 3-like 1 (cartilage glycoprotein-39) Enzyme −4.0 −2.4 
FCN1 Ficolin (collagen/fibrinogen domain containing) 1 Other −2.3 −6.9 
IFIT1 IFN-induced protein with tetratricopeptide repeats 1 Other −2.1 −2.3 
MMP9 MMP 9 (gelatinase B, 92 kDa gelatinase, 92 kDa type IV collagenase) Peptidase −2.7 −4.9 
RARRES1 Retinoic acid receptor responder (tazarotene induced) 1 Other −2.1 −4.8 
SymbolGene NameTypeFold Change MonocytesFold Change Macrophages
ADM Adrenomedullin Other 2.1 3.7 
ADORA3 Adenosine A3 receptor GPCR 4.7 7.4 
ALOX15B Arachidonate 15-lipoxygenase, type B Enzyme 4.8 2.7 
C1QB Complement component 1, q subcomponent, B chain Other 3.1 2.2 
CD163 CD163 molecule TM receptor 2.4 3.7 
DDIT4 DNA-damage-inducible transcript 4 Other 5.1 4.0 
FBLN5 Fibulin 5 Other 2.2 4.5 
FKBP5 FK506 binding protein 5 Enzyme 6.8 5.4 
HPGD hydroxypg dehydrogenase 15-(NAD) Enzyme 2.2 3.1 
HTRA1 HtrA serine peptidase 1 Peptidase 3.2 3.7 
IL1R2 IL-1 receptor, type II TM receptor 12.5 4.8 
IRS2 Insulin receptor substrate 2 Enzyme 3.8 2.4 
MERTK c-mer proto-oncogene tyrosine kinase Kinase 4.3 3.0 
METTL7A Methyltransferase like 7A Other 2.3 2.4 
MFGE8 Milk fat globule-EGF factor 8 protein Other 2.4 6.4 
PDK4 Pyruvate dehydrogenase kinase, isozyme 4 Kinase 4.8 3.1 
RNASE1 RNase, RNase A family, 1 (pancreatic) Enzyme 2.2 9.1 
SERPINE1 Serpin peptidase inhibitor, clade E (nexin, plasminogen activator inhibitor type 1), member 1 Other 2.7 2.6 
SESN1 Sestrin 1 Other 4.6 2.6 
SLA Src-like-adaptor Other 2.1 2.2 
SLC16A6 Solute carrier family 16, member 6 (monocarboxylic acid transporter 7) Transporter 2.9 3.4 
SLC1A3 Solute carrier family 1 (glial high affinity glutamate transporter), member 3 Transporter 5.2 2.2 
SRPX Sushi-repeat containing protein, X-linked Other 5.3 5.9 
TBC1D16 TBC1 domain family, member 16 Other 7.7 2.6 
TCN2 Transcobalamin II Transporter 2.9 2.1 
TFPI Tissue factor pathway inhibitor (lipoprotein-associated coagulation inhibitor) Other 4.1 7.0 
TPST1 Tyrosylprotein sulfotransferase 1 Enzyme 4.5 3.0 
TSC22D3 TSC22 domain family, member 3 Other 2.4 4.1 
VSIG4 V-set and Ig domain containing 4 Other 11.2 6.0 
ZCCHC6 Zinc finger, CCHC domain containing 6 Enzyme 2.6 2.3 
CHI3L1 Chitinase 3-like 1 (cartilage glycoprotein-39) Enzyme −4.0 −2.4 
FCN1 Ficolin (collagen/fibrinogen domain containing) 1 Other −2.3 −6.9 
IFIT1 IFN-induced protein with tetratricopeptide repeats 1 Other −2.1 −2.3 
MMP9 MMP 9 (gelatinase B, 92 kDa gelatinase, 92 kDa type IV collagenase) Peptidase −2.7 −4.9 
RARRES1 Retinoic acid receptor responder (tazarotene induced) 1 Other −2.1 −4.8 

Results of this study were compared with a previous study by Ehrchen et al. (18). Genes modulated with a fold ≥ 2 and t test p value < 0.05 in both datasets were considered for the comparison.

GPCR, G protein-coupled receptor; TM, transmembrane.

The 165 genes modulated by chronic FP exposure could be organized into 13 functional and regulation-related networks (Supplemental Table I). Upstream regulator analysis detected as theme, as expected, GCR activation with 48 genes previously associated with dexamethasone; also highlighting the novelty of these gene signatures, revealing 117 new GC-target genes in human macrophages (Fig. 1A). Interestingly, many of these genes can also be linked to IL10, TGFBR2, IL1RN, and HIF1A, pointing to putative autocrine subpathways, all previously associated to resolution of inflammation (Supplemental Table I). Many of the FP-regulated genes are also linked to IFN receptor (38 genes) and TLR4 pathways (56 genes), and the analysis suggests inhibition of the basal levels of STAT3 (21 genes), IRF1 (8 genes), and IRF7 (8 genes), among other mediators, which can affect macrophage ability to initiate classical activation (Fig. 1A).

FIGURE 1.

Remodeling of the transcriptome of human macrophages by chronic FP treatment. Human peripheral monocytes from five donors were matured into macrophages in the presence or absence of FP. At day 7, RNA was extracted for transcriptome analyses using Illumina microarray. (A) Interactome analysis of genes modulated by FP performed using IPA. Genes appear pseudocolored according to the regulation by FP: red = upregulated, green = downregulated. Genes marked with an asterisk indicate that multiple identifiers in the dataset file map to a single gene in the IPA Global Molecular Network. The selection of the interactome shown here highlights that many FP-regulated genes have not been associated with dexamethasone regulation previously in human macrophages or other cells, which are visible because of the lack of connecting lines, and shows that a great proportion of the genes have been associated previously with LPS and IFN signaling. (B) Selection of IPA canonical pathways, which are modulated by FP in human macrophages, corroborates FP effects on important immune pathways, such as Ag presentation, LPS response, and IFN response (Supplemental Table I). (C) Dose response (upper panels) and time kinetics (lower panels) of the regulation of selected surface markers in peripheral blood monocytes, stimulated for 5 d with FP (dark blue lines) or dexamethasone (light blue lines) at a concentration ranging from 0.1–1000 nM and for 3, 5, and 7 d with 100 nM FP (dark blue lines), 100 nM dexamethasone, or left untreated (gray lines). Protein expression, measured using flow cytometry, is provided relative to isotype control. **p < 0.01, ***p < 0.001.

FIGURE 1.

Remodeling of the transcriptome of human macrophages by chronic FP treatment. Human peripheral monocytes from five donors were matured into macrophages in the presence or absence of FP. At day 7, RNA was extracted for transcriptome analyses using Illumina microarray. (A) Interactome analysis of genes modulated by FP performed using IPA. Genes appear pseudocolored according to the regulation by FP: red = upregulated, green = downregulated. Genes marked with an asterisk indicate that multiple identifiers in the dataset file map to a single gene in the IPA Global Molecular Network. The selection of the interactome shown here highlights that many FP-regulated genes have not been associated with dexamethasone regulation previously in human macrophages or other cells, which are visible because of the lack of connecting lines, and shows that a great proportion of the genes have been associated previously with LPS and IFN signaling. (B) Selection of IPA canonical pathways, which are modulated by FP in human macrophages, corroborates FP effects on important immune pathways, such as Ag presentation, LPS response, and IFN response (Supplemental Table I). (C) Dose response (upper panels) and time kinetics (lower panels) of the regulation of selected surface markers in peripheral blood monocytes, stimulated for 5 d with FP (dark blue lines) or dexamethasone (light blue lines) at a concentration ranging from 0.1–1000 nM and for 3, 5, and 7 d with 100 nM FP (dark blue lines), 100 nM dexamethasone, or left untreated (gray lines). Protein expression, measured using flow cytometry, is provided relative to isotype control. **p < 0.01, ***p < 0.001.

Close modal

In terms of functional pathways (Fig. 1B), FP treatment downregulated eight members of the MHC II complex and affected coagulation pathways with inhibition of prothrombotic CCL22 and increased levels of TFP1 (tissue factor pathway inhibitor) and SERPINF1 (serine peptidase inhibitor F1). The synthesis, transport, or recognition of lipids also appears decreased with downregulation of ABCG1, ACAT2, CYP27B1, HSD11B1, PPARGC1A, and PTGER2. The detailed analysis also showed strong changes in molecules for the recruitment of phagocytes (e.g., upregulation of the chemokines CCL7 and CCL8, downregulation of CCL22) and changes in the phagocytosis capacity of macrophages, exemplified by increased expression of ADORA3, BIN1, CAMP, MERTK, MFGE8, SERPINE1, and THBS1, and corresponding with previous reports (32).

Gene expression remodeling by FP in human macrophages is of functional relevance, and the effects observed at the transcriptome level translated into phenotype changes as demonstrated by flow cytometric analysis of CD163 and HLA-DR protein expression, which were upregulated and slightly downregulated respectively (Fig. 1C). Given the suggestion that IRF3, IRF7, and STAT3 basal activity may be diminished, we also investigated the expression of two important inflammation-mediated molecules regulated by TLR4 and IFN pathways, namely CD64 and CD86. The first is responsible for immune complex responses, and the second is an important costimulatory signal for T cell activation. For both molecules, we could confirm protein level changes despite a lack of regulation at the mRNA level. The kinetics of FP effects on CD64 and CD86 protein expression were comparable to CD163 and HLA-DR, with clearly visible effects at 10 nM and a saturated response at 100 nM—levels achieved in the tissue of patients inhaling FP (33).

The effect on these four genes, CD163, HLA-DR, CD64, and CD86, was further corroborated in a time course. We cultured monocytes with FP or in medium alone for 3, 5, or 7 d, followed by flow cytometric analysis, and we found that the pattern of regulation for every gene was slightly different, suggesting differential regulation from gene expression to protein modification and expression. FP increased the expression of CD163 until day 3, when it reached a plateau, whereas HLA-DR followed a bell-shaped curve. The regulation was subtle for CD64 and CD86 (Fig. 1D). GCR involvement was confirmed in all experiments because dexamethasone had comparable effects to those of FP, with the only difference being a slight superiority of FP in the dose and time response (Fig. 1C–D).

Classical macrophage activation involves the exposure to pathogens, which induces IFN-β and IL-12 expression that primes Th1 cells to undergo expansion, while activating strong microbicidal programs in macrophages (2). This pathogen stimulation has also been termed “innate activation” and can be recapitulated by TLR4 triggering (34). To investigate the effects of GCR ligation during innate activation, we defined the transcriptome of MDMs from three donors, matured for 7 d with or without FP, after challenging them for 6 h—concurring with an inflammatory peak (35)—with 10 ng/ml LPS (Fig. 2A). Principal component analysis (Fig. 2B) showed that FP-treated macrophages are able to respond to LPS but only reach an intermediate state, with a significant shift toward an FP activation-like state, represented by negative movements in the y-axis. The graph clustered 12 samples in four groups, which consist of triplicates of each condition. The largest difference is seen in LPS-treated and untreated macrophages; however, a clear distinction remains between untreated and FP-treated cells. A total of 1685 transcripts representing 1500 genes were significantly regulated in the data set (Fig. 2C). Of 630 genes upregulated or downregulated by LPS, 193 were significantly influenced by FP, which corresponds to observations in mouse peritoneal macrophages exposed for 6 h to LPS and dexamethasone (36). Moreover, 409 genes, not significantly changed by LPS or FP alone, were regulated by FP plus LPS. A summary of genes regulated by the combination of FP and LPS is provided in Table III.

FIGURE 2.

Complex changes in gene expression induced by LPS in FP-polarized macrophages. (A) Experimental design: Human peripheral monocytes from three donors were matured into macrophages in the presence (red bar) or absence (gray bar) of FP. Prior to RNA extraction for transcriptome analyses using the Illumina microarray at day 7, cells were stimulated for 6 h with LPS (green bars) as indicated. (B) Principal component analysis showed remote locations for medium-, FP-, LPS-, and FP+LPS-treated MDMs, representing distinct phenotypes. Components 1 and 2, comprising 72.8% of the total variance, were used to create the graph. (C) Venn diagrams providing numbers of transcripts upregulated or downregulated greater than twofold over untreated by FP, LPS, or FP→LPS treatment. (D) Median-centered K-means clustering of genes significantly regulated in the dataset revealed six distinctive expression patterns. The y-axis indicates divergence from median values with intensity of gene expression illustrated in colors: low (black), intermediate (blue to green to yellow), high (red). Cluster 1 consists of 396 genes that are upregulated by LPS and fairly resistant to FP. Cluster 2 includes 323 genes that are downregulated by LPS and fairly resistant to FP. Cluster 3 shows 185 genes that are upregulated by LPS and inhibited by FP. In cluster 4, 190 genes were upregulated by LPS in FP-primed MDMs. Alternatively, cluster 5 contains 238 genes that are downregulated by LPS stimulation in FP-primed cells. Finally, cluster 6 shows 168 genes that are induced by FP and mildly counteracted by LPS. Gene clusters were assessed for pathways and upstream regulators using IPA. For details see Supplemental Table II. (E) Hierarchical clustering heat map of Ag presentation–related molecules and chemokines.

FIGURE 2.

Complex changes in gene expression induced by LPS in FP-polarized macrophages. (A) Experimental design: Human peripheral monocytes from three donors were matured into macrophages in the presence (red bar) or absence (gray bar) of FP. Prior to RNA extraction for transcriptome analyses using the Illumina microarray at day 7, cells were stimulated for 6 h with LPS (green bars) as indicated. (B) Principal component analysis showed remote locations for medium-, FP-, LPS-, and FP+LPS-treated MDMs, representing distinct phenotypes. Components 1 and 2, comprising 72.8% of the total variance, were used to create the graph. (C) Venn diagrams providing numbers of transcripts upregulated or downregulated greater than twofold over untreated by FP, LPS, or FP→LPS treatment. (D) Median-centered K-means clustering of genes significantly regulated in the dataset revealed six distinctive expression patterns. The y-axis indicates divergence from median values with intensity of gene expression illustrated in colors: low (black), intermediate (blue to green to yellow), high (red). Cluster 1 consists of 396 genes that are upregulated by LPS and fairly resistant to FP. Cluster 2 includes 323 genes that are downregulated by LPS and fairly resistant to FP. Cluster 3 shows 185 genes that are upregulated by LPS and inhibited by FP. In cluster 4, 190 genes were upregulated by LPS in FP-primed MDMs. Alternatively, cluster 5 contains 238 genes that are downregulated by LPS stimulation in FP-primed cells. Finally, cluster 6 shows 168 genes that are induced by FP and mildly counteracted by LPS. Gene clusters were assessed for pathways and upstream regulators using IPA. For details see Supplemental Table II. (E) Hierarchical clustering heat map of Ag presentation–related molecules and chemokines.

Close modal
Table III.
Selection of genes regulated by the combination of FP and LPS
SymbolGene NameTypeFold Change
FPLPS+FPLPS
CCL20 CCL20 Cytokine 1.346 85.178 20.286 
IL1B IL-1β Cytokine −1.112 83.968 36.773 
CCL4L2 CCL4-like 1 Other 2.014 67.507 47.831 
CCL8 CCL8 Cytokine 2.189 66.669 31.013 
SERPINB2 Serpin peptidase inhibitor, clade B (OVA), member 2 Other 1.288 43.689 5.85 
SLAMF1 Signaling lymphocytic activation molecule family member 1 TM receptor 1.195 39.851 27.412 
EBI3 EBV induced 3 Cytokine 2.5 32.823 19.231 
IL8 IL-8 Cytokine 1.195 28.994 5.7 
ANGPTL4 Angiopoietin-like 4 Other 1.328 26.979 6.613 
CFB Complement factor B Peptidase −1.107 25.849 17.81 
NAMPT Nicotinamide phosphoribosyltransferase Cytokine 1.046 21.729 12.559 
MT1A Metallothionein 1A Other 4.725 19.824 5.709 
MT2A Metallothionein 2A Other 5.367 15.328 6.632 
IL1A IL-1α Cytokine −1.01 11.533 1.32 
MT1× Metallothionein 1× Other 1.769 10.04 1.397 
CXCL1 CXCL1 Cytokine −1.038 8.068 2.074 
TIMP1 TIMP metallopeptidase inhibitor 1 Other 2.92 6.715 1.477 
CCL2 CCL2 Cytokine 2.103 6.574 1.645 
CCL7 CCL7 Cytokine 1.787 6.001 1.826 
CXCL2 CXCL2 Cytokine −1.202 4.35 1.887 
CXCL5 CXCL5 Cytokine 1.008 3.489 1.145 
MDFIC MyoD family inhibitor domain containing Other −1.026 −2.075 1.453 
IRF8 IFN regulatory factor 8 Transcription regulator −1.079 −2.104 1.438 
SGK3 Serum/glucocorticoid regulated kinase family, member 3 Kinase 1.285 −2.155 1.027 
ARHGAP25 Rho GTPase activating protein 25 Other −1.361 −2.175 1.017 
ADAM9 ADAM metallopeptidase domain 9 Peptidase 1.021 −2.403 1.023 
HLA-DPA1 MHC, class II, DP α 1 TM receptor −2.466 −2.491 1.012 
TPM3 Tropomyosin 3 Other −1.228 −2.905 1.25 
CCL22 CCL22 Cytokine −4.953 −3.135 1.022 
SymbolGene NameTypeFold Change
FPLPS+FPLPS
CCL20 CCL20 Cytokine 1.346 85.178 20.286 
IL1B IL-1β Cytokine −1.112 83.968 36.773 
CCL4L2 CCL4-like 1 Other 2.014 67.507 47.831 
CCL8 CCL8 Cytokine 2.189 66.669 31.013 
SERPINB2 Serpin peptidase inhibitor, clade B (OVA), member 2 Other 1.288 43.689 5.85 
SLAMF1 Signaling lymphocytic activation molecule family member 1 TM receptor 1.195 39.851 27.412 
EBI3 EBV induced 3 Cytokine 2.5 32.823 19.231 
IL8 IL-8 Cytokine 1.195 28.994 5.7 
ANGPTL4 Angiopoietin-like 4 Other 1.328 26.979 6.613 
CFB Complement factor B Peptidase −1.107 25.849 17.81 
NAMPT Nicotinamide phosphoribosyltransferase Cytokine 1.046 21.729 12.559 
MT1A Metallothionein 1A Other 4.725 19.824 5.709 
MT2A Metallothionein 2A Other 5.367 15.328 6.632 
IL1A IL-1α Cytokine −1.01 11.533 1.32 
MT1× Metallothionein 1× Other 1.769 10.04 1.397 
CXCL1 CXCL1 Cytokine −1.038 8.068 2.074 
TIMP1 TIMP metallopeptidase inhibitor 1 Other 2.92 6.715 1.477 
CCL2 CCL2 Cytokine 2.103 6.574 1.645 
CCL7 CCL7 Cytokine 1.787 6.001 1.826 
CXCL2 CXCL2 Cytokine −1.202 4.35 1.887 
CXCL5 CXCL5 Cytokine 1.008 3.489 1.145 
MDFIC MyoD family inhibitor domain containing Other −1.026 −2.075 1.453 
IRF8 IFN regulatory factor 8 Transcription regulator −1.079 −2.104 1.438 
SGK3 Serum/glucocorticoid regulated kinase family, member 3 Kinase 1.285 −2.155 1.027 
ARHGAP25 Rho GTPase activating protein 25 Other −1.361 −2.175 1.017 
ADAM9 ADAM metallopeptidase domain 9 Peptidase 1.021 −2.403 1.023 
HLA-DPA1 MHC, class II, DP α 1 TM receptor −2.466 −2.491 1.012 
TPM3 Tropomyosin 3 Other −1.228 −2.905 1.25 
CCL22 CCL22 Cytokine −4.953 −3.135 1.022 

TM, Transmembrane.

The genes regulated in the dataset could be divided into six distinct major gene patterns using K-median clustering (Fig. 2D–E; for comprehensive details about molecular pathways, gene networks, and upstream regulators, see Supplemental Table II). Cluster 1 identified 396 genes modulated by LPS and not FP, and the combination marginally decreased levels for 25% of the genes. This cluster included inflammatory cytokines such as IL-1B, TNF, IL-6, and IL-15. Upstream regulators analysis showed clear links with MyD88 (modulates 45 genes), IFN (modulates > 70 genes), and IL-6 pathways (modulates 51 genes). In this study, FP pretreatment decreased CXCL10 and increased IL-1B, suggesting that FP negatively affected the IFN pathway while positively affecting the MyD88 pathway. This hypothesis is strengthened by the fact that the IL-1B increase was accompanied by increases in MyD88-regulated CCL20, CCL8, IL-8, CXCL1, CCL2, and CCL17, whereas the CXCL10 decrease was accompanied by downregulation of IFN-dependent IFIT2, IFIT1, IFIT3, CXCL9, CD80, and IFIH1. Cluster 2 contained 323 genes downregulated in response to LPS, regardless of the presence of FP. This cluster comprised macrophage markers, and upstream regulator analysis revealed possible inhibition of signaling by receptors for the lineage-determining factors GM-CSF and M-CSF, downregulating CD68, DOK2, GDF15, IFNGR1, PCNA, and PLAU.

Cluster 3 constituted the only LPS-dependent pathway completely abrogated by FP treatment. The cluster contained 185 genes enriched in lymphocyte activation molecules such as CD80, CD83, CD86, IL10, IL15, IL18, CD274, and TNFSF13B. Interestingly, dexamethasone now appeared as upstream regulator of at least 43 genes, which are also linked to IFN-dependent pathways, more than to MyD88. Cluster 4 comprised 190 genes upregulated by LPS stimulation in FP-treated MDM exclusively. Upstream analysis showed that at least 50 of these genes have been associated with dexamethasone treatment. This cluster is enriched in metallothioneins (e.g., MT1M, MT1A, MT1E, MT1G, MT1×, MT1F), which participate in stress responses and apoptosis modulation, and the chemokines CXCL1, CXCL2, CXCL5, CCL2, CCL7, IL-1A, and IL-8, which participate in cell recruitment. Cluster 5 comprised 238 genes expressed in resting MDM, downregulated by the combination of FP and LPS. This group strongly associated with MHC class II–mediated Ag presentation (e.g., HLA-DMA, HLA-DMB, HLA-PA1, HLA-PB1, HLA-DQA1, HLA-DRA, HLA-DRB3, HLA-DRB4, CD74) and appeared to be cluster 2 dependent on M-CSF and GM-CSF. Finally, cluster 6 contained 168 genes induced by LPS and further enhanced by FP, with overrepresentation of complement system pathways (e.g., C1QA, C1QB, C1QC, C3AR1, C5AR1) and cytokine recognition (e.g., IL1R1, IL1R2, IL13RA1, IL27RA). Dexamethasone, retinoic acid, and PPARD appeared as major regulators.

LPS stimulation provided clues about how FP-treated macrophages initiate altered classical activation programs, with the IFN pathways being the most affected. Next, we investigated how FP treatment affected long-term IFN-γ responses, by treating before, during, or after IFN-γ challenge, to mimic treatment in chronic inflammatory settings. We analyzed mRNA expression profiles of MDMs of two donors cultured for 5 d with medium alone, FP, IFN-γ, or both, followed by 2 d during which culture conditions were kept stable or switched from FP or IFN-γ, respectively, to FP plus IFN-γ (Fig. 3A).

FIGURE 3.

Distinct yet reversible effects of FP and IFN-γ on gene expression in macrophages. (A) Experimental design: Human peripheral monocytes from two donors were treated with FP (red bar), IFN-γ (blue bar), both (red and blue bar), or no stimulus (gray bar). After 5 d, MDM were exposed to the same or changed stimuli for an additional 2 d as indicated, after which RNA was extracted for transcriptome analyses using Illumina microarray. (B) Principal component analysis showed remote locations for medium-, FP-, IFN-γ–, and FP+IFN-γ–treated cells, representing distinct MDM phenotypes. Components 1 and 2, comprising 79.8% of the total variance, were used to create the graph. (C) Venn diagrams providing numbers of transcripts upregulated or downregulated greater than twofold over untreated by FP, IFN-γ, or FP+IFN-γ treatment. (D) Median-centered K-means clustering of the significantly regulated genes in the dataset revealed six distinctive expression patterns. The y-axis indicates divergence from median values with intensity of gene expression illustrated in colors: low (black), intermediate (blue to green to yellow), high (red). Cluster 1 consists of 87 genes upregulated by FP. Cluster 2 includes 97 genes downregulated by FP. Most genes in cluster 1 and 2 are oppositely regulated by IFN-γ. The 285 genes in cluster 3 are upregulated by IFN-γ and are resistant to FP-mediated downregulation. Clusters 4 and 5 comprise 305 and 128 genes, respectively, which showed an augmented effect with long-term exposure to the combination of FP+IFN-γ. Cluster 6 comprises 217 genes downregulated only in FP-treated MDMs. Gene clusters were assessed for pathways and upstream regulators using IPA. For details see Supplemental Table III. (E) Hierarchical clustering heat map of Ag presentation–related molecules and chemokines.

FIGURE 3.

Distinct yet reversible effects of FP and IFN-γ on gene expression in macrophages. (A) Experimental design: Human peripheral monocytes from two donors were treated with FP (red bar), IFN-γ (blue bar), both (red and blue bar), or no stimulus (gray bar). After 5 d, MDM were exposed to the same or changed stimuli for an additional 2 d as indicated, after which RNA was extracted for transcriptome analyses using Illumina microarray. (B) Principal component analysis showed remote locations for medium-, FP-, IFN-γ–, and FP+IFN-γ–treated cells, representing distinct MDM phenotypes. Components 1 and 2, comprising 79.8% of the total variance, were used to create the graph. (C) Venn diagrams providing numbers of transcripts upregulated or downregulated greater than twofold over untreated by FP, IFN-γ, or FP+IFN-γ treatment. (D) Median-centered K-means clustering of the significantly regulated genes in the dataset revealed six distinctive expression patterns. The y-axis indicates divergence from median values with intensity of gene expression illustrated in colors: low (black), intermediate (blue to green to yellow), high (red). Cluster 1 consists of 87 genes upregulated by FP. Cluster 2 includes 97 genes downregulated by FP. Most genes in cluster 1 and 2 are oppositely regulated by IFN-γ. The 285 genes in cluster 3 are upregulated by IFN-γ and are resistant to FP-mediated downregulation. Clusters 4 and 5 comprise 305 and 128 genes, respectively, which showed an augmented effect with long-term exposure to the combination of FP+IFN-γ. Cluster 6 comprises 217 genes downregulated only in FP-treated MDMs. Gene clusters were assessed for pathways and upstream regulators using IPA. For details see Supplemental Table III. (E) Hierarchical clustering heat map of Ag presentation–related molecules and chemokines.

Close modal

Principal component analysis (Fig. 3B) showed remote locations for medium-, FP-, IFN-γ–, and FP plus IFN-γ–treated macrophages, representing distinct phenotypes. Cells cultured with FP or IFN-γ for 5 d, followed by the combination of both stimuli for 2 d, developed intermediate yet distinct gene expression profiles that closely resembled the profile of cells exposed for 7 d to the combination of FP plus IFN-γ. The graph shows six distinct clusters, correlating with the experimental conditions and comprising donor duplicates. FP and IFN-γ clusters are highly diverse from each other. The FP-plus-IFN–γ cluster is equally distant to FP and IFN-γ alone. Samples in which stimuli were changed after 5 d clustered at intermediate positions; 1327 transcripts, representing 1119 genes, were significantly regulated in the dataset (Fig. 3C). Of the 362 genes regulated by IFN-γ, 145 were significantly affected by FP; 400 genes not significantly changed by IFN-γ or FP alone were regulated by FP plus IFN-γ. A summary of genes regulated by the combination of FP and IFN-γ is provided in Table IV.

Table IV.
Selection of genes regulated by the combination of FP and IFN-γ
SymbolGene NameTypeFold Change
FPIFN-γ+FPIFN-γ
S100A8 S100 calcium binding protein A8 Other 1.127 18.525 2.874 
RNASE2 RNase, RNase A family, 2 (liver, eosinophil-derived neurotoxin) Enzyme 1.019 6.906 1.057 
SERPINA1 Serpin peptidase inhibitor, clade A (α-1 antiproteinase, antitrypsin), member 1 Other −1.274 6.770 4.578 
PLAC8 Placenta-specific 8 Other −1.086 6.457 1.315 
ALOX5 Arachidonate 5-lipoxygenase Enzyme 1.047 5.029 1.771 
SIGLEC10 Sialic acid binding Ig-like lectin 10 Other 1.135 4.986 1.260 
ITGB7 Integrin, β 7 TM receptor −1.369 4.954 2.597 
OLR1 Oxidized low density lipoprotein (lectin-like) receptor 1 TM receptor 1.983 4.356 1.093 
MUC1 Mucin 1, cell surface associated Transcription regulator −1.183 4.325 1.321 
OLFML3 Olfactomedin-like 3 Other −1.261 4.123 1.127 
CLEC12A C-type lectin domain family 12, member A Other −1.328 4.013 2.541 
GAPT GRB2-binding adaptor protein, transmembrane Other −1.156 3.195 1.146 
CSF2RA CSF 2 receptor, α, low-affinity (granulocyte-macrophage) TM receptor 1.089 2.686 1.249 
FCN1 Ficolin (collagen/fibrinogen domain containing) 1 Other −7.093 −1.264 1.795 
CCL7 CCL7 Cytokine 2.213 −2.645 −1.330 
CCL2 CCL2 Cytokine 1.498 −3.881 2.031 
SEPP1 Selenoprotein P, plasma, 1 Other 1.671 −5.247 −1.860 
CD36 CD36 molecule (thrombospondin receptor) TM receptor 1.894 −6.550 −1.607 
CD83 CD83 molecule Other −1.034 −7.126 −1.861 
CHIT1 Chitinase 1 (chitotriosidase) Enzyme −1.622 −11.295 −4.330 
IL1RN IL-1 receptor antagonist Cytokine −1.256 −12.333 −3.028 
EMP1 Epithelial membrane protein 1 Other 1.319 −16.277 −3.414 
LPL Lipoprotein lipase Enzyme 1.804 −21.094 −3.944 
AQP9 Aquaporin 9 Transporter 1.115 −21.650 −3.952 
SPP1 Secreted phosphoprotein 1 Cytokine 1.606 −30.607 −2.708 
FABP4 Fatty acid binding protein 4, adipocyte Transporter 1.457 −46.129 −2.236 
TM4SF19 TM4 L six family member 19 Other −1.164 −60.890 −2.320 
MMP9 MMP9 (gelatinase B) Peptidase −4.934 −132.508 −3.313 
SymbolGene NameTypeFold Change
FPIFN-γ+FPIFN-γ
S100A8 S100 calcium binding protein A8 Other 1.127 18.525 2.874 
RNASE2 RNase, RNase A family, 2 (liver, eosinophil-derived neurotoxin) Enzyme 1.019 6.906 1.057 
SERPINA1 Serpin peptidase inhibitor, clade A (α-1 antiproteinase, antitrypsin), member 1 Other −1.274 6.770 4.578 
PLAC8 Placenta-specific 8 Other −1.086 6.457 1.315 
ALOX5 Arachidonate 5-lipoxygenase Enzyme 1.047 5.029 1.771 
SIGLEC10 Sialic acid binding Ig-like lectin 10 Other 1.135 4.986 1.260 
ITGB7 Integrin, β 7 TM receptor −1.369 4.954 2.597 
OLR1 Oxidized low density lipoprotein (lectin-like) receptor 1 TM receptor 1.983 4.356 1.093 
MUC1 Mucin 1, cell surface associated Transcription regulator −1.183 4.325 1.321 
OLFML3 Olfactomedin-like 3 Other −1.261 4.123 1.127 
CLEC12A C-type lectin domain family 12, member A Other −1.328 4.013 2.541 
GAPT GRB2-binding adaptor protein, transmembrane Other −1.156 3.195 1.146 
CSF2RA CSF 2 receptor, α, low-affinity (granulocyte-macrophage) TM receptor 1.089 2.686 1.249 
FCN1 Ficolin (collagen/fibrinogen domain containing) 1 Other −7.093 −1.264 1.795 
CCL7 CCL7 Cytokine 2.213 −2.645 −1.330 
CCL2 CCL2 Cytokine 1.498 −3.881 2.031 
SEPP1 Selenoprotein P, plasma, 1 Other 1.671 −5.247 −1.860 
CD36 CD36 molecule (thrombospondin receptor) TM receptor 1.894 −6.550 −1.607 
CD83 CD83 molecule Other −1.034 −7.126 −1.861 
CHIT1 Chitinase 1 (chitotriosidase) Enzyme −1.622 −11.295 −4.330 
IL1RN IL-1 receptor antagonist Cytokine −1.256 −12.333 −3.028 
EMP1 Epithelial membrane protein 1 Other 1.319 −16.277 −3.414 
LPL Lipoprotein lipase Enzyme 1.804 −21.094 −3.944 
AQP9 Aquaporin 9 Transporter 1.115 −21.650 −3.952 
SPP1 Secreted phosphoprotein 1 Cytokine 1.606 −30.607 −2.708 
FABP4 Fatty acid binding protein 4, adipocyte Transporter 1.457 −46.129 −2.236 
TM4SF19 TM4 L six family member 19 Other −1.164 −60.890 −2.320 
MMP9 MMP9 (gelatinase B) Peptidase −4.934 −132.508 −3.313 

MMP, Matrix metallopeptidase; TM, transmembrane

The regulated genes could be divided into six distinct clusters by K-means clustering (Fig. 3D–E; for comprehensive details about the genes, molecular pathways, gene networks, and upstream regulators in each cluster, see Supplemental Table III). Cluster 1 and 2 comprised 87 and 97 genes, respectively, regulated primarily by FP and overlapping largely with the genes regulated by FP alone that we described in the first section of 11Results and in Supplemental Table I. IFN-γ oppositely regulated these genes, in combination and alone, which adds to the theme of opposed crosstalk between GCR and TLR4 and IFN receptor pathways. Cluster 3 represented 285 genes induced by IFN-γ, refractory to FP treatment in all variations and slightly exacerbated in the cotreatment. These genes support Th1 attraction and activation with inflammatory mediators, such as TNF, the chemokines CXCL9 and CXCL10, and the MHC class I–mediated Ag-presentation cluster HLA-A, HLA-C, HLA-E, HLA-F, and HLA-G. The downregulation of MHC class II in TLR stimulation and preservation of class I activation upon IFN receptor stimulation suggests that FP spares antiviral and not antibacterial responses, which may contribute to the enhanced incidence of pneumonia in patients with COPD treated long term with FP (9, 37, 38). Long-term exposure to the combination of FP plus IFN-γ exclusively downregulated 305 genes, as shown in cluster 4. PPRG was downregulated 2.5-fold in this cluster, and genes that depend on it such as lipid-binding proteins (APOC1, APOC2, CD36, FABP3, FABP4, FABP5, FABP5L2, FABP5L3) were affected. The scavenger receptor CD36 (decreased by FP) is also involved in bacterial phagocytosis during pneumococcal pneumonia (39). Other functional pathways, which appear decreased, are cell adhesion (ITGAE, ITGAX, ITGA3, ITGB, ITGB1, ITGB5, ITGB1BP1) and chemotaxis with downregulation of CCL2, CCL7, and CCL3. Cluster 5 is the opposite of cluster 4, with 128 genes upregulated by long-term exposure to FP plus IFN-γ. This cluster is related to the inflammatory response; its predicted upstream regulators include LPS, dexamethasone, and IFN over MyD88. Among the genes included are complement members C1QC, C1QB, and C5, the multifunctional pair S100A8 and S100A9, and the chemokine receptors for CCL2 and CXCL12, CCR2 and CXCR4, respectively. Cluster 6 was similar to cluster 4 and contained 217 genes, downregulated only in FP-treated macrophages. Forty-three of these genes have previously been associated with TLR4 stimulation and are generally IFN independent; MMP7, MMP9, CCL2, and PLAU are downregulated and add to a vast list of inflammatory genes in the cluster.

To investigate macrophage plasticity further at a protein level, we compared FP effects on IFN-γ and two other important cytokines for macrophages—alternatively activating IL-4 and anti-inflammatory IL-10—by following the behavior of model genes. Of the 10 molecules with a reported differential expression in human macrophages (4042) that we validated in detail (Fig. 4A), three were studied further, namely CD64, CD163, and CD206 (Fig. 4B). CD163 is an interesting marker because it was induced by FP and IL-10, but not modulated by IFN-γ or IL-4. IL-10 and FP cooperated to increase CD163 expression even further. However, IL-10 is not that similar to FP because it increased CD64 expression, which was decreased by FP. Regulation of CD206 (mannose receptor), which is important for complications such as fibrosis (4), depended on the cytokine context. FP appeared to leave IL-4–driven CD206 expression unaffected, but opened possibilities for cytokine cooperation. Combined treatment with FP and IFN-γ further increased IL-4–mediated CD206 expression, whereas IFN-γ does not induce CD206. Moreover, FP with IFN-γ and IL-10 enabled CD206 expression, whereas the separate treatments had no effect. In conclusion, FP effects on macrophage-polarizing cytokines are not simply inhibitory. Through model genes, it is possible to appreciate the many phenotypes elicited by treatment with GCs. Importantly, GCs seem to alter the response to otherwise polarizing cytokines.

FIGURE 4.

FP affects the interaction between classical and alternative macrophage activation. Peripheral blood monocytes were cultured for 5 d in the presence of 12.5 ng/ml IFN-γ, 50 ng/ml IL-4, 50 ng/ml IL-10, 100 nM FP, or combinations thereof, followed by flow cytometric assessment of the expression of surface proteins. (A) Macrophages that are classically activated with IFN-γ upregulate CD64; macrophages that are alternatively activated with IL-4 upregulate HLA-DR, CD86, CD200R, and CD206; and regulatory macrophages that are generated in the presence of IL-10 upregulate CD14, CD16, CD64, and CD163. Macrophages that are generated with FP upregulate CD163. Expression of CD32 and CD80 was not affected by the investigated stimuli (data not shown). (B) Combinations of IFN-γ, IL-4, IL-10, and FP regulate expression of CD64, CD163, and CD206 in a complex manner. FP neutralizes induction of CD64 by IFN-γ and IL-10, but enhances expression of CD163 by IL-10. Notably, FP enabled expression of CD206, even when none of the individual stimuli (IFN-γ and IL-10) had an effect by themselves. Protein expression is provided relative to isotype control. Underscores in (B) indicate a significant difference in expression caused by a combination compared with the respective underlined stimulus. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4.

FP affects the interaction between classical and alternative macrophage activation. Peripheral blood monocytes were cultured for 5 d in the presence of 12.5 ng/ml IFN-γ, 50 ng/ml IL-4, 50 ng/ml IL-10, 100 nM FP, or combinations thereof, followed by flow cytometric assessment of the expression of surface proteins. (A) Macrophages that are classically activated with IFN-γ upregulate CD64; macrophages that are alternatively activated with IL-4 upregulate HLA-DR, CD86, CD200R, and CD206; and regulatory macrophages that are generated in the presence of IL-10 upregulate CD14, CD16, CD64, and CD163. Macrophages that are generated with FP upregulate CD163. Expression of CD32 and CD80 was not affected by the investigated stimuli (data not shown). (B) Combinations of IFN-γ, IL-4, IL-10, and FP regulate expression of CD64, CD163, and CD206 in a complex manner. FP neutralizes induction of CD64 by IFN-γ and IL-10, but enhances expression of CD163 by IL-10. Notably, FP enabled expression of CD206, even when none of the individual stimuli (IFN-γ and IL-10) had an effect by themselves. Protein expression is provided relative to isotype control. Underscores in (B) indicate a significant difference in expression caused by a combination compared with the respective underlined stimulus. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To investigate the effects of GC treatment in vivo, we used a mouse model of cigarette smoke-induced lung inflammation. It consisted of 4 mo of tobacco smoke exposure to generate COPD-like features, such as airway remodeling, mild emphysema, and chronic inflammation (20, 21). This model differs from the human disease, but is useful to investigate cellular and molecular mechanisms underlying the development and progression of COPD. The GC analog used for this experiment was budesonide, which has an intermediate pharmacodynamics activity compared with fluticasone and dexamethasone (7); it is retained to a greater extent than FP in the mucosa of the airways (43) and, unlike FP, shows no differences in plasma level in healthy or diseased groups (44).

As described previously, smoking induced a Th17 type of airway inflammation, characterized by accumulation of neutrophils, macrophages, and B cells, and increased levels of IL-17, IL-6, GM-CSF, G-CSF, and CCL2 (21). After GC treatment, total macrophages and B cells were significantly higher in lung tissue of the smoking mice as compared with saline-treated smoking animals (Fig. 5A). The transcriptome study in human macrophages predicted that GCR ligation would affect chemokine expression. Despite interspecies differences, we found that GC treatment affected the expression of key chemokines in this model, which correlated with higher overall leukocyte numbers in the lungs (Fig. 5B). In lung tissue, these chemokines can be produced by many cell types, and they have a pleiotropic effect. Given the correlation with macrophages, we consider this a significant trend, which will require further investigation.

FIGURE 5.

GC enhances leukocyte recruitment and chemokine expression in a mouse model of COPD-like lung inflammation. Mice (n = 6 per group) were exposed to cigarette smoke 5 d per week for 4 mo and treated five times per week with either budesonide (closed symbols) or saline (open symbols). Lung leukocyte infiltration and expression of cytokines and chemokines were measured with flow cytometry and ELISA, respectively. (A) GC treatment led to significantly higher numbers of macrophages and B lymphocytes, being present in lung tissue of smoking mice. A similar trend was observed for neutrophils and T lymphocytes. (B) GC treatment led to significantly higher levels of key chemokines in lung tissue of smoking mice. These differences were not found for IFN-γ, IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-9, IL-10, IL-12, IL-13, IL-15, IL-17, and GM-CSF. *p < 0.05, **p < 0.01, as tested with Mann–Whitney U test.

FIGURE 5.

GC enhances leukocyte recruitment and chemokine expression in a mouse model of COPD-like lung inflammation. Mice (n = 6 per group) were exposed to cigarette smoke 5 d per week for 4 mo and treated five times per week with either budesonide (closed symbols) or saline (open symbols). Lung leukocyte infiltration and expression of cytokines and chemokines were measured with flow cytometry and ELISA, respectively. (A) GC treatment led to significantly higher numbers of macrophages and B lymphocytes, being present in lung tissue of smoking mice. A similar trend was observed for neutrophils and T lymphocytes. (B) GC treatment led to significantly higher levels of key chemokines in lung tissue of smoking mice. These differences were not found for IFN-γ, IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-9, IL-10, IL-12, IL-13, IL-15, IL-17, and GM-CSF. *p < 0.05, **p < 0.01, as tested with Mann–Whitney U test.

Close modal

The role of macrophages in the etiology and pathology of chronic inflammatory disorders is being recognized increasingly (4). Many of these diseases are treated with GCs, but recent studies have shown that GCs can have little effect in controlling inflammation in macrophage-dominated diseases (9, 16). This lack of efficacy in diseases, like COPD, might be explained by our finding that GCs do not generally suppress macrophage effector functions, but cause a shift in the innate–adaptive balance of immune responses.

FP significantly altered the expression of 165 genes in macrophages matured for 7 d. When comparing the effects of long-term FP exposure on gene expression with the profile obtained by Ehrchen et al. (19) in monocytes stimulated for 16 h with FP, we found a modest correlation (r2 = 0.3850; p < 0.0001), pointing to the fact that FP modulates genes in an acute and a chronic fashion. These differences add to the emergent concept of intermodel and interspecies differences found in the response of macrophages to other immune relevant stimuli, such as GM-CSF, M-CSF, LPS, and IL-4 (18, 35, 45). The genes already associated with FP in monocytes at 16 h, also modulated after 7 d of FP, included CD163, MERTK, and MFGE8. Early transient genes modulated after 16 h, but not after 7 d, included the GC-associated formyl peptide receptor 1 and IL-10 (19, 46). Induction of IL-10 by GCs in monocytes but not MDMs is in line with earlier findings (47). The chronic FP transcriptome comprises functionally relevant genes, such as TFPI, CAMP (cathelicidin antimicrobial peptide), THBS1 (thrombospondin 1), and the transcription factor KLF9, which binds GC-box elements and has not been linked to FP treatment in macrophages. We conclude that FP regulates genes in a time-dependent fashion; therefore, chronic stimulation of macrophages might not be directly translatable to studies in circulating monocytes.

The transcriptome analysis revealed the expected overlap of the FP profiles with dexamethasone, methylprednisolone, prednisolone, and suggested a strong negative relationship between GCR gene signatures and TLR4 and IFN receptor pathways. Anticipating that the true effects of GCR ligation on macrophages are better observed in inflammatory conditions, we explored the consequences of FP treatment on the macrophage transcriptome upon LPS-mediated or IFN-γ–mediated activation. As expected, FP massively affected gene expression induced by LPS or IFN-γ (36, 48). Transcriptome analysis revealed distinct behavioral patterns, of which the most interesting ones comprised induced genes resistant to FP, inhibited by FP, or exacerbated upon FP treatment. Notably, IFN-γ–induced genes related to the development of an innate immune response were fairly resistant to FP. In contrast, FP counteracted IFN-γ–induced genes associated with lymphocyte activation. Illustrative is the ability of GCs to repress Ag-presentation pathways, which is well established (49, 50). We found that FP suppresses both MHC class I and class II genes. Whereas MHC class II genes remained suppressed by FP upon IFN-γ stimulation, genes encoding MHC class I molecules were upregulated by IFN-γ, regardless of the presence of FP. The decreased expression of MHC class II is likely to hamper the initiation of adaptive immune responses, whereas innate immune responses can develop and might be enhanced by FP (51).

Although GCs differentially regulate various immune gene signatures (52, 53), the regulation of chemokines by FP emerged as a strong theme; this recapitulates previously identified trends in monocytes (19). FP treatment significantly augmented LPS-induced IL-8 (CXCL8), CCL2, CCL8, and CCL20 expression, and moderately enhanced CXCL1, CXCL2, CXCL5, CCL3L1, and CCL7 expression, whereas it inhibited LPS-induced upregulation of CXCL9, CXCL10, CCL19, CCL22, and (moderately) CXCL11. These chemokines activate distinct sets of receptors. The inhibited chemokines bind CXCR3A, CXCR3B, CCR7, and CCR4, which are mainly expressed on T cells; the overexpressed chemokines bind CXCR1, CXCR2, CCR1, CCR2, CCR3, CCR5, and CCR6, which are primarily found on monocytes and granulocytes, and only some T cells (54). The dysregulation of the chemokine system by GCR ligation in macrophages is gaining interest. Recently, Kent et al. (55) studied the transcriptome of COPD macrophages exposed to LPS and dexamethasone, and found 23 genes that were insensitive to GCR ligation, including IL-1β, IL-18, and CCL5. Importantly, the chromosome 4 chemokine cluster members CXCL1, CXCL2, CXCL3, and CXCL8 were all GC-resistant in our and their study (55).

The contribution of resident versus inflammatory macrophages is not clear in COPD and other lung diseases, and mimicking the milieu of such a complex organ is not straightforward. Studies done on MDMs are important, although it cannot be excluded that these profiles may be more representative of inflammatory, newly recruited monocytes, rather than resident tissue macrophages, nor the fact that in the lung environment GCs may act differently. Furthermore, diseased human tissue has limited possibilities when studying immune responses. To investigate the relevance of our in vitro results in a more physiologic manner, we used a mouse model of COPD, in which resident and inflammatory macrophages are known to play a role (21). In this model, we found sustained increases of macrophages and a set of chemokines and cytokines upon GC treatment with budesonide. The increased chemokine levels correlated with enhanced leukocyte numbers, supporting the predictions of the chemokine–chemokine receptor network. Previously, in two mouse studies using either LPS or Mycoplasma pneumonia in combination with corticosteroid treatment, CCL5, CXCL1, CXCL2, and neutrophils were found to be upregulated by dexamethasone treatment (56, 57). Interestingly, in the bronchoalveolar lavage of patients with COPD treated with budesonide, the number of macrophages was also increased (12). These unexpected effects of corticosteroids are therefore reported more often, and they deserve further attention (58).

This study puts forward a set of GCR/TLR/IFN receptor-regulated macrophage markers that could be used as biomarkers to monitor GC efficacy and further investigate the interaction of GCs and macrophages in human disease. Furthermore, our results add to the emerging concept of GC-insensitive and GC-exacerbated pathways in macrophages that could explain their lack of efficacy in macrophage-dominated disorders, such as COPD. Therapeutically, the innate–adaptive immune response shift induced by GCs in macrophages and its subdivision in pathways could offer novel views on other anti-inflammatory therapeutic strategies.

We thank Marie Geerlings, Nicole Stowell, Anuk Das, and Don Griswold for experimental help; the Laboratory of Viral Immune Pathogenesis (Academic Medical Center, Amsterdam, The Netherlands), Dr. René Lutter, and Prof. Peter Sterk for supporting this project; and Prof. Siamon Gordon for reading the manuscript and helpful discussions.

This work was supported by a grant from the J.K. de Cock Stichting and Prof. Dirkje Postma through a Spinoza grant from the Dutch Government.

The online version of this article contains supplemental material.

Abbreviations used in this article:

COPD

chronic obstructive pulmonary disease

FP

fluticasone propionate

GC

glucocorticoid

GCR

GC receptor

IPA

Ingenuity Pathways Analysis

MDM

monocyte-derived macrophage

MEV

Multiple Experiment Viewer

MMP

matrix metalloproteinase.

1
Gordon
S.
,
Taylor
P. R.
.
2005
.
Monocyte and macrophage heterogeneity.
Nat. Rev. Immunol.
5
:
953
964
.
2
Mosser
D. M.
,
Edwards
J. P.
.
2008
.
Exploring the full spectrum of macrophage activation.
Nat. Rev. Immunol.
8
:
958
969
.
3
Sica
A.
,
Mantovani
A.
.
2012
.
Macrophage plasticity and polarization: in vivo veritas.
J. Clin. Invest.
122
:
787
795
.
4
Murray
P. J.
,
Wynn
T. A.
.
2011
.
Protective and pathogenic functions of macrophage subsets.
Nat. Rev. Immunol.
11
:
723
737
.
5
Glass
C. K.
,
Saijo
K.
.
2010
.
Nuclear receptor transrepression pathways that regulate inflammation in macrophages and T cells.
Nat. Rev. Immunol.
10
:
365
376
.
6
Barnes
P. J.
2011
.
Glucocorticosteroids: current and future directions.
Br. J. Pharmacol.
163
:
29
43
.
7
Esmailpour
N.
,
Högger
P.
,
Rohdewald
P.
.
1998
.
Binding kinetics of budesonide to the human glucocorticoid receptor.
Eur. J. Pharm. Sci.
6
:
219
223
.
8
Yona
S.
,
Gordon
S.
.
2007
.
Inflammation: Glucocorticoids turn the monocyte switch.
Immunol. Cell Biol.
85
:
81
82
.
9
Calverley
P. M.
,
Anderson
J. A.
,
Celli
B.
,
Ferguson
G. T.
,
Jenkins
C.
,
Jones
P. W.
,
Yates
J. C.
,
Vestbo
J.
TORCH investigators
.
2007
.
Salmeterol and fluticasone propionate and survival in chronic obstructive pulmonary disease.
N. Engl. J. Med.
356
:
775
789
.
10
Rabe
K. F.
,
Wedzicha
J. A.
.
2011
.
Controversies in treatment of chronic obstructive pulmonary disease.
Lancet
378
:
1038
1047
.
11
Hakim
A.
,
Adcock
I. M.
,
Usmani
O. S.
.
2012
.
Corticosteroid resistance and novel anti-inflammatory therapies in chronic obstructive pulmonary disease: current evidence and future direction.
Drugs
72
:
1299
1312
.
12
Jen
R.
,
Rennard
S. I.
,
Sin
D. D.
.
2012
.
Effects of inhaled corticosteroids on airway inflammation in chronic obstructive pulmonary disease: a systematic review and meta-analysis.
Int. J. Chron. Obstruct. Pulmon. Dis.
7
:
587
595
.
13
Baschant
U.
,
Lane
N. E.
,
Tuckermann
J.
.
2012
.
The multiple facets of glucocorticoid action in rheumatoid arthritis.
Nat. Rev. Rheumatol.
8
:
645
655
.
14
Gomes
J. A.
,
Stevens
R. D.
,
Lewin
J. J.
 III
,
Mirski
M. A.
,
Bhardwaj
A.
.
2005
.
Glucocorticoid therapy in neurologic critical care.
Crit. Care Med.
33
:
1214
1224
.
15
Ciccone
A.
,
Beretta
S.
,
Brusaferri
F.
,
Galea
I.
,
Protti
A.
,
Spreafico
C.
.
2008
.
Corticosteroids for the long-term treatment in multiple sclerosis.
Cochrane Database Syst. Rev.
(
1
):
CD006264
.
16
Sorrells
S. F.
,
Caso
J. R.
,
Munhoz
C. D.
,
Sapolsky
R. M.
.
2009
.
The stressed CNS: when glucocorticoids aggravate inflammation.
Neuron
64
:
33
39
.
17
Martinez
F. O.
2012
.
Analysis of gene expression and gene silencing in human macrophages
.
Curr. Protoc. Immunol.
Chapter 14:
Unit 23
.
18
Martinez
F. O.
,
Helming
L.
,
Milde
R.
,
Varin
A.
,
Melgert
B. N.
,
Draijer
C.
,
Thomas
B.
,
Fabbri
M.
,
Crawshaw
A.
,
Ho
L. P.
, et al
.
2013
.
Genetic programs expressed in resting and IL-4 alternatively activated mouse and human macrophages: similarities and differences.
Blood
121
:
e57
e69
.
19
Ehrchen
J.
,
Steinmüller
L.
,
Barczyk
K.
,
Tenbrock
K.
,
Nacken
W.
,
Eisenacher
M.
,
Nordhues
U.
,
Sorg
C.
,
Sunderkötter
C.
,
Roth
J.
.
2007
.
Glucocorticoids induce differentiation of a specifically activated, anti-inflammatory subtype of human monocytes.
Blood
109
:
1265
1274
.
20
van der Strate
B. W.
,
Postma
D. S.
,
Brandsma
C. A.
,
Melgert
B. N.
,
Luinge
M. A.
,
Geerlings
M.
,
Hylkema
M. N.
,
van den Berg
A.
,
Timens
W.
,
Kerstjens
H. A.
.
2006
.
Cigarette smoke-induced emphysema: A role for the B cell?
Am. J. Respir. Crit. Care Med.
173
:
751
758
.
21
Melgert
B. N.
,
Timens
W.
,
Kerstjens
H. A.
,
Geerlings
M.
,
Luinge
M. A.
,
Schouten
J. P.
,
Postma
D. S.
,
Hylkema
M. N.
.
2007
.
Effects of 4 months of smoking in mice with ovalbumin-induced airway inflammation.
Clin. Exp. Allergy
37
:
1798
1808
.
22
Griffith
R. B.
1984
.
A simple machine for smoke analytical studies and total particulate matter collection for biological studies.
Toxicology
33
:
33
41
.
23
Griffith
R. B.
,
Hancock
R.
.
1985
.
Simultaneous mainstream-sidestream smoke exposure systems I. Equipment and procedures.
Toxicology
34
:
123
138
.
24
Watson
E. S.
,
Jones
A. B.
,
Ashfaq
M. K.
,
Barrett
J. T.
.
1987
.
Spectrophotometric evaluation of carboxyhemoglobin in blood of mice after exposure to marijuana or tobacco smoke in a modified Walton horizontal smoke exposure machine.
J. Anal. Toxicol.
11
:
19
23
.
25
Melgert
B. N.
,
Postma
D. S.
,
Geerlings
M.
,
Luinge
M. A.
,
Klok
P. A.
,
van der Strate
B. W.
,
Kerstjens
H. A.
,
Timens
W.
,
Hylkema
M. N.
.
2004
.
Short-term smoke exposure attenuates ovalbumin-induced airway inflammation in allergic mice.
Am. J. Respir. Cell Mol. Biol.
30
:
880
885
.
26
Liu
Y.
,
Endo
Y.
,
Iwaki
D.
,
Nakata
M.
,
Matsushita
M.
,
Wada
I.
,
Inoue
K.
,
Munakata
M.
,
Fujita
T.
.
2005
.
Human M-ficolin is a secretory protein that activates the lectin complement pathway.
J. Immunol.
175
:
3150
3156
.
27
Aho
S.
,
Lehtinen
P.
,
Kulonen
E.
.
1980
.
Effects of purified macrophage RNase on granuloma fibroblasts with reference to silicosis.
Acta Physiol. Scand.
109
:
275
281
.
28
Shaykhiev
R.
,
Krause
A.
,
Salit
J.
,
Strulovici-Barel
Y.
,
Harvey
B. G.
,
O’Connor
T. P.
,
Crystal
R. G.
.
2009
.
Smoking-dependent reprogramming of alveolar macrophage polarization: implication for pathogenesis of chronic obstructive pulmonary disease.
J. Immunol.
183
:
2867
2883
.
29
Sajjadi
F. G.
,
Takabayashi
K.
,
Foster
A. C.
,
Domingo
R. C.
,
Firestein
G. S.
.
1996
.
Inhibition of TNF-alpha expression by adenosine: role of A3 adenosine receptors.
J. Immunol.
156
:
3435
3442
.
30
Haskó
G.
,
Szabó
C.
,
Németh
Z. H.
,
Kvetan
V.
,
Pastores
S. M.
,
Vizi
E. S.
.
1996
.
Adenosine receptor agonists differentially regulate IL-10, TNF-alpha, and nitric oxide production in RAW 264.7 macrophages and in endotoxemic mice.
J. Immunol.
157
:
4634
4640
.
31
Li
Q.
,
Park
P. W.
,
Wilson
C. L.
,
Parks
W. C.
.
2002
.
Matrilysin shedding of syndecan-1 regulates chemokine mobilization and transepithelial efflux of neutrophils in acute lung injury.
Cell
111
:
635
646
.
32
Zahuczky
G.
,
Kristóf
E.
,
Majai
G.
,
Fésüs
L.
.
2011
.
Differentiation and glucocorticoid regulated apopto-phagocytic gene expression patterns in human macrophages. Role of Mertk in enhanced phagocytosis.
PLoS ONE
6
:
e21349
.
33
Esmailpour
N.
,
Högger
P.
,
Rabe
K. F.
,
Heitmann
U.
,
Nakashima
M.
,
Rohdewald
P.
.
1997
.
Distribution of inhaled fluticasone propionate between human lung tissue and serum in vivo.
Eur. Respir. J.
10
:
1496
1499
.
34
Mukhopadhyay
S.
,
Peiser
L.
,
Gordon
S.
.
2004
.
Activation of murine macrophages by Neisseria meningitidis and IFN-gamma in vitro: distinct roles of class A scavenger and Toll-like pattern recognition receptors in selective modulation of surface phenotype.
J. Leukoc. Biol.
76
:
577
584
.
35
Schroder
K.
,
Irvine
K. M.
,
Taylor
M. S.
,
Bokil
N. J.
,
Le Cao
K. A.
,
Masterman
K. A.
,
Labzin
L. I.
,
Semple
C. A.
,
Kapetanovic
R.
,
Fairbairn
L.
, et al
.
2012
.
Conservation and divergence in Toll-like receptor 4-regulated gene expression in primary human versus mouse macrophages.
Proc. Natl. Acad. Sci. USA
109
:
E944
E953
.
36
Ogawa
S.
,
Lozach
J.
,
Benner
C.
,
Pascual
G.
,
Tangirala
R. K.
,
Westin
S.
,
Hoffmann
A.
,
Subramaniam
S.
,
David
M.
,
Rosenfeld
M. G.
,
Glass
C. K.
.
2005
.
Molecular determinants of crosstalk between nuclear receptors and toll-like receptors.
Cell
122
:
707
721
.
37
Lapperre
T. S.
,
Snoeck-Stroband
J. B.
,
Gosman
M. M.
,
Jansen
D. F.
,
van Schadewijk
A.
,
Thiadens
H. A.
,
Vonk
J. M.
,
Boezen
H. M.
,
Ten Hacken
N. H.
,
Sont
J. K.
, et al
Groningen Leiden Universities Corticosteroids in Obstructive Lung Disease Study Group
.
2009
.
Effect of fluticasone with and without salmeterol on pulmonary outcomes in chronic obstructive pulmonary disease: a randomized trial.
Ann. Intern. Med.
151
:
517
527
.
38
Thornton Snider
J.
,
Luna
Y.
,
Wong
K. S.
,
Zhang
J.
,
Chen
S. S.
,
Gless
P. J.
,
Goldman
D. P.
.
2012
.
Inhaled corticosteroids and the risk of pneumonia in Medicare patients with COPD.
Curr. Med. Res. Opin.
28
:
1959
1967
.
39
Sharif
O.
,
Matt
U.
,
Saluzzo
S.
,
Lakovits
K.
,
Haslinger
I.
,
Furtner
T.
,
Doninger
B.
,
Knapp
S.
.
2013
.
The scavenger receptor CD36 downmodulates the early inflammatory response while enhancing bacterial phagocytosis during pneumococcal pneumonia.
J. Immunol.
190
:
5640
5648
.
40
Martinez
F. O.
,
Sica
A.
,
Mantovani
A.
,
Locati
M.
.
2008
.
Macrophage activation and polarization.
Front. Biosci.
13
:
453
461
.
41
Koning
N.
,
van Eijk
M.
,
Pouwels
W.
,
Brouwer
M. S.
,
Voehringer
D.
,
Huitinga
I.
,
Hoek
R. M.
,
Raes
G.
,
Hamann
J.
.
2010
.
Expression of the inhibitory CD200 receptor is associated with alternative macrophage activation.
J. Innate Immun.
2
:
195
200
.
42
Ambarus
C. A.
,
Krausz
S.
,
van Eijk
M.
,
Hamann
J.
,
Radstake
T. R.
,
Reedquist
K. A.
,
Tak
P. P.
,
Baeten
D. L.
.
2012
.
Systematic validation of specific phenotypic markers for in vitro polarized human macrophages.
J. Immunol. Methods
375
:
196
206
.
43
Petersen
H.
,
Kullberg
A.
,
Edsbäcker
S.
,
Greiff
L.
.
2001
.
Nasal retention of budesonide and fluticasone in man: formation of airway mucosal budesonide-esters in vivo.
Br. J. Clin. Pharmacol.
51
:
159
163
.
44
Harrison
T. W.
,
Tattersfield
A. E.
.
2003
.
Plasma concentrations of fluticasone propionate and budesonide following inhalation from dry powder inhalers by healthy and asthmatic subjects.
Thorax
58
:
258
260
.
45
Lacey
D. C.
,
Achuthan
A.
,
Fleetwood
A. J.
,
Dinh
H.
,
Roiniotis
J.
,
Scholz
G. M.
,
Chang
M. W.
,
Beckman
S. K.
,
Cook
A. D.
,
Hamilton
J. A.
.
2012
.
Defining GM-CSF- and macrophage-CSF-dependent macrophage responses by in vitro models.
J. Immunol.
188
:
5752
5765
.
46
Sawmynaden
P.
,
Perretti
M.
.
2006
.
Glucocorticoid upregulation of the annexin-A1 receptor in leukocytes.
Biochem. Biophys. Res. Commun.
349
:
1351
1355
.
47
Mozo
L.
,
Suárez
A.
,
Gutiérrez
C.
.
2004
.
Glucocorticoids up-regulate constitutive interleukin-10 production by human monocytes.
Clin. Exp. Allergy
34
:
406
412
.
48
Flammer
J. R.
,
Dobrovolna
J.
,
Kennedy
M. A.
,
Chinenov
Y.
,
Glass
C. K.
,
Ivashkiv
L. B.
,
Rogatsky
I.
.
2010
.
The type I interferon signaling pathway is a target for glucocorticoid inhibition.
Mol. Cell. Biol.
30
:
4564
4574
.
49
Snyder
D. S.
,
Unanue
E. R.
.
1982
.
Corticosteroids inhibit murine macrophage Ia expression and interleukin 1 production.
J. Immunol.
129
:
1803
1805
.
50
Schwiebert
L. M.
,
Schleimer
R. P.
,
Radka
S. F.
,
Ono
S. J.
.
1995
.
Modulation of MHC class II expression in human cells by dexamethasone.
Cell. Immunol.
165
:
12
19
.
51
Franchimont
D.
2004
.
Overview of the actions of glucocorticoids on the immune response: a good model to characterize new pathways of immunosuppression for new treatment strategies.
Ann. N. Y. Acad. Sci.
1024
:
124
137
.
52
Warren
M. K.
,
Vogel
S. N.
.
1985
.
Opposing effects of glucocorticoids on interferon-gamma-induced murine macrophage Fc receptor and Ia antigen expression.
J. Immunol.
134
:
2462
2469
.
53
Galon
J.
,
Franchimont
D.
,
Hiroi
N.
,
Frey
G.
,
Boettner
A.
,
Ehrhart-Bornstein
M.
,
O’Shea
J. J.
,
Chrousos
G. P.
,
Bornstein
S. R.
.
2002
.
Gene profiling reveals unknown enhancing and suppressive actions of glucocorticoids on immune cells.
FASEB J.
16
:
61
71
.
54
Viola
A.
,
Luster
A. D.
.
2008
.
Chemokines and their receptors: drug targets in immunity and inflammation.
Annu. Rev. Pharmacol. Toxicol.
48
:
171
197
.
55
Kent
L. M.
,
Smyth
L. J.
,
Plumb
J.
,
Clayton
C. L.
,
Fox
S. M.
,
Ray
D. W.
,
Farrow
S. N.
,
Singh
D.
.
2009
.
Inhibition of lipopolysaccharide-stimulated chronic obstructive pulmonary disease macrophage inflammatory gene expression by dexamethasone and the p38 mitogen-activated protein kinase inhibitor N-cyano-N’-(2-[8-(2,6-difluorophenyl)-4-(4-fluoro-2-methylphenyl)-7-oxo-7,8-dihydropyrido[2,3-d] pyrimidin-2-yl]aminoethyl)guanidine (SB706504).
J. Pharmacol. Exp. Ther.
328
:
458
468
.
56
Aoki
K.
,
Ishida
Y.
,
Kikuta
N.
,
Kawai
H.
,
Kuroiwa
M.
,
Sato
H.
.
2002
.
Role of CXC chemokines in the enhancement of LPS-induced neutrophil accumulation in the lung of mice by dexamethasone.
Biochem. Biophys. Res. Commun.
294
:
1101
1108
.
57
Tagliabue
C.
,
Salvatore
C. M.
,
Techasaensiri
C.
,
Mejias
A.
,
Torres
J. P.
,
Katz
K.
,
Gomez
A. M.
,
Esposito
S.
,
Principi
N.
,
Hardy
R. D.
.
2008
.
The impact of steroids given with macrolide therapy on experimental Mycoplasma pneumoniae respiratory infection.
J. Infect. Dis.
198
:
1180
1188
.
58
Wilckens
T.
,
De Rijk
R.
.
1997
.
Glucocorticoids and immune function: unknown dimensions and new frontiers.
Immunol. Today
18
:
418
424
.

GlaxoSmithKline financially supported this study.