Alternative Macrophage Activation Is Increased in Asthma

Abstract

The immune responses of type 2 T helper cells (Th2) play an important role in asthma and promote the differentiation of alternatively activated (M2) macrophages. M2 macrophages have been increasingly understood to contribute to Th2 immunity. We hypothesized that M2 macrophages are altered in asthma and modulate Th2 responses. The aim of this study was to characterize the phenotype and function of human monocyte-derived M2 and bronchoalveolar lavage fluid (BALF) macrophages from healthy control subjects and subjects with asthma. Phenotypic characteristics and effector function of M2 macrophages were examined using monocyte-derived and BALF macrophages obtained from subjects with asthma (n = 28) and healthy volunteers (n = 9) by flow cytometry and quantitative PCR. Resting monocyte-derived (M0) and M2 macrophages were generated by the addition of macrophage colony–stimulating factor or macrophage colony–stimulating factor plus IL-4, respectively. M2 macrophage cytokine expression and their impact on dendritic and CD4⁺ T cell activation were examined in vitro. High levels of CD206 and major histocompatibility complex class II expression identify macrophages with an M2 phenotype that are increased 2.9-fold in the BALF of subjects with asthma compared with control subjects. M2 macrophages have elevated IL-6, IL-10, and IL-12p40 production compared with conventional macrophages and modulate dendritic and CD4⁺ T cell interactions. Histamine receptor 1 and E-cadherin expression identify M2 macrophage subsets associated with increased airflow obstruction. M2 macrophages have a distinct cell surface and effector phenotype and are found in increased numbers in subjects with asthma. These findings suggest that M2 macrophages may play an important role in allergic asthma through their bidirectional interactions with immune and structural cells, and inflammatory mediators.

Clinical Relevance

Lung macrophages, which are the most abundant immune cell in the lung, have been increasingly understood to play a significant role in modulating immune responses. This study demonstrates that a subset of lung macrophages in subjects with asthma have an altered immune phenotype that may promote allergic inflammation and contribute to disease severity. Therapies directed at this important cell subset may prove to be of benefit in asthma.

Asthma is a chronic inflammatory disease characterized by airway obstruction and bronchial hyperresponsiveness. Initial allergen exposure to lung dendritic cells (DCs) causes up-regulated type 2 T helper cells (Th2) responses and mast cell and eosinophil activation, but additional subsets, including Th1, Th17, Th9, innate lymphoid type 2 (ILC2), regulatory T cells, and macrophages, may modulate allergic inflammation (1, 2). Macrophages are frequently divided into classically activated (M1) macrophages and alternatively activated (M2) macrophages (AAMs). In contrast to M1 macrophages that are activated by IFN-γ and Toll-like receptor signaling pathways, IL-4 and IL-13 lead to M2 macrophage differentiation through a signal transducer and activator of transcription 6–mediated transcriptional program (3). M2 macrophages have cell surface, intracellular enzyme, and chemokine profiles that are distinct from M1 macrophages. Although macrophages are well appreciated for their critical role in host defense and tissue homeostasis and repair, an important role for macrophages in Th2 immunity is becoming more clearly defined (4). Eosinophils and ILC2 cells were recently described to drive the development of AAMs (4, 5). The local proliferation of M2 macrophages has been shown to occur under the control of Th2 immune processes (6). In addition, AAMs directly contributed to allergic airway inflammation and can produce autocrine Th2 cytokines that can further amplify type 2 immune responses (7–10).

Phenotypic markers for these macrophage subsets in mice have been well established, with M2-phenotype macrophages expressing arginase-1, nitric oxide synthase, YM1, and found in inflammatory zone 1. However, these markers fall short in humans, and either have not been observed in macrophages or do not have human homologs, as in the case of YM1 and found in inflammatory zone 1 (3). Although several markers are induced on human monocyte-derived M2 macrophages (11, 12), they have not been validated on ex vivo human macrophage subsets at steady state, at mucosal sites, such as the lung, or in the setting of allergic inflammation. In asthma, previous studies have shown that human alveolar macrophages up-regulate chemokine CCL17 production (13, 14). Increased CD206 expression, which is a described marker of alternative activation, was observed on macrophages in bronchial biopsy specimens from subjects with asthma and correlated with asthma severity (15). Production of cytokines, such as IL-12p40 and IL-10, was also increased in macrophages from subjects with atopic asthma (16). Despite these advances, reliable identification of human AAMs has remained a challenge. Given their increasingly appreciated role in Th2 immunity, we hypothesized that M2 macrophages play an immunomodulatory role in allergic airway immune responses. We set out to demonstrate, for the first time, an expression profile that allows for phenotypic identification of human macrophages with an alternative activation phenotype. We used this profile to demonstrate that AAMs are found with increased frequency in human subjects with asthma.

Materials and Methods

Generation of Monocyte-Derived Macrophages

Peripheral blood mononuclear cells from healthy donors were isolated by Ficoll (GE Healthcare Life Sciences, Pittsburgh, PA) density gradient centrifugation. Positive selection of monocytes was performed using CD14 MicroBeads (Miltenyi Biotec, San Diego, CA). CD14⁺ cells were cultured for 7 days in RPMI 1,640 (Cellgro, Manassas, VA) supplemented with 100 ng/ml macrophage colony–stimulating factor (M-CSF; Peprotech, Rocky Hill, NJ) (11). Macrophage polarization was obtained by adding Escherichia coli LPS (Sigma, St. Louis, MO) at 10–100 ng/ml (for M1 polarization) (17, 18) or recombinant human IL-4 (Peprotech) at 100 U/ml (for M2 polarization) for 48 hours (17). Three cell types were obtained: resting macrophages (9 days of culture, M0); classical activated macrophages (M1); and AAMs (M2). Human DCs were generated from CD14⁺ monocytes stimulated with granulocyte-macrophage CSF (Peptrotech) and IL-4 for 7 days (19).

Patients and Samples

The Brigham and Women’s Hospital Asthma Research Center (Boston, MA) recruited adults 18–60 years of age with asthma. Severe asthma was defined per the criteria developed by the Severe Asthma Research Program (20). Adults, without a prior history of chronic respiratory disease, including asthma, were recruited as healthy control subjects.

Bronchoalveolar lavage fluid (BALF) samples were obtained from 28 subjects with asthma and 9 healthy control subjects. Cytospins were prepared with Diff Quick (Harleco, Gibbstown, NJ) stain for differential cell count. Additional measurements included Asthma Control Questionnaire score, serum IgE levels, blood eosinophil counts, and fraction of exhaled nitric oxide levels.

Flow Cytometry

Cells from BALF and macrophage culture were incubated with the following monoclonal antibodies (mAbs): anti–E-cadherin FITC, anti–major histocompatibility complex (MHC) II allophycocyanin (APC)–cyanine 7 (BD Biosciences, San Jose, CA), anti-CD206 (macrophage mannose receptor) phycoerythrin (PE), anti-CD209 PE, anti-CX3C chemokine receptor 1 PerCP-eFluor 710, anti-CD16 PE–cyanine 7, anti-CD1c APC (from eBioscience, San Diego, CA), anti–CD200 receptor PE (Thermo Scientific, Waltham, MA), anti-CD32 PE, anti-CCR7 PE, anti-CD11c APC (BD Pharmingen, San Jose, CA), anti–histamine receptor H1 (HRH1) A647 (Bioss, Woburn, MA), and the appropriate fluorescently conjugated isotype controls (BD Biosciences, eBioscience) for each experimental condition. Data were acquired on a Canto II flow cytometer (Becton Dickinson, Franklin Lakes, NJ) and analyzed with FlowJo software (Tree Star, Ashland, OR). For alveolar macrophage analysis, doublet exclusion was followed by live cell gating (see Figure E3 in the online supplement). Results are expressed as positive cell percentage and median fluorescent intensity (MFI) index. MFI index is the MFI with the tested mAbs divided by the MFI of the appropriate isotype-matched control mAbs. The MFI threshold was used to identify and quantify cells with high cell surface expression from low expression, using the formula: MFI threshold = mean MFI in healthy subjects + twofold 95% confidence interval (Figure E4).

Quantitative PCR

Total RNA from cultured macrophages was extracted using the RNeasy Mini Kit (Qiagen, Germantown, MD). Primer sequences are listed in Table E1.

CD4⁺ T Cell Activation Assay

CD4⁺ T cells were isolated from human peripheral blood mononuclear cells obtained from healthy blood donors using indirect magnetic labeling (Miltenyi Biotec). Autologous macrophages and DCs were cocultured with allogeneic CD4⁺ T cells for 48 hours (21).

Cytokine ELISAs

IL-6, IL-10, IL-12p40, IL-12p70, IL-13, and IFN-γ cytokines were measured in the cell culture supernatants by ELISA (eBioscience and Thermo Scientific).

Statistical Analysis

The data were analyzed using NCSS 2001 software (Kaysville, UT). Values are reported as mean (±SEM). A P value less than 0.05 was considered statistically significant.

Results

Identification of M2 Phenotypic Markers on Monocyte-Derived Macrophages

We first characterized the cell surface phenotype of in vitro–derived M2 macrophages generated from human monocytes obtained from healthy donors stimulated with M-CSF and IL-4 and compared them to resting macrophages generated with M-CSF (M0). We examined an array of markers, including mannose and lectin receptors, chemokine receptors, MHC-II, inhibitory receptors, Fc receptors, and CD11c. In response to IL-4, human monocyte-derived macrophages up-regulate CD11c, CD16, CD200 receptor, CD206, and MHC-II (Table E2). Comparison of M0 and M2 cell surface expression reveals a highly significant increase in the MFI index of CD206 (6.1 ± 0.8 versus 24.6 ± 3.8; P = 0.001) and MHC-II (184.2 ± 29.1 versus 244.3 ± 32.2; P = 0.001) on M2 macrophages (Figure 1A). CD206 and MHC-II are expressed on both resting M0 and M2 macrophages, with differences in the level of expression, but, as individual markers, they did not allow for clear delineation of M0 versus M2 macrophages. However, we found that the coexpression of CD206 and MHC-II reliably phenotypically identified in vitro–derived M2 macrophages compared with the M0 subset (81.9 ± 5.7% versus 25.5 ± 6.6%; P < 0.001; Figures 1B and 1C). Similar results were obtained using M1 polarized monocyte-derived macrophages (Figure E1A).

Figure 1.

Phenotypic characterization of human M2 macrophages. (A) Representative histogram plots demonstrate CD206 and major histocompatibility complex (MHC) class II molecule expression on monocyte-derived alternatively activated (M2, black solid line) versus resting macrophages (M0, black dotted line). Light gray line, isotype control for M0; dark gray line, isotype control for M2. (B) Representative dot plots of CD206⁺MHC-II⁺ cell populations from M0 (left panel) and M2 (right panel) monocyte-derived macrophages. (C) Percentage of CD206⁺MHC-II⁺ M0 and M2 monocyte-derived macrophages from n = 12 subjects (independent experiments). Values are mean ± SEM. *P < 0.05 using paired t test.

Alternatively Activated Macrophages Have a Distinct Cytokine Profile

We subsequently examined the cytokine profile of M0 and M2 macrophages. M2 macrophages produce more IL-6 (81.4 ± 16.8 pg/ml versus 6.3 ± 3.0 pg/ml), IL-10 (288.0 ± 76.1 pg/ml versus 84.0 ± 25.9 pg/ml), and IL-12p40 (162.3 ± 44.3 pg/ml versus 27.0 ± 18.0 pg/ml) compared with M0 macrophages. Conversely, M2 macrophages produce significantly less IL-12p70, which is critical for Th1 polarization (Figure 2A). Concomitant analysis of cell surface phenotype and cellular function demonstrated that co-expression of CD206⁺MHC-II⁺ cells correlated with IL-6 and IL-12p40 release (Figure 2B). As expected (22), macrophages polarized with LPS (M1) release large amounts of IL-6 and IL-10 in a concentration-dependent manner (Figure E1B).

Figure 2.

Macrophage cytokine production and modulation of dendritic cell (DC) and CD4⁺ T cell responses. (A) Differential cytokine production of IL-6, IL-10, IL-12p40, and IL-12p70 by monocyte-derived M0 and M2 macrophages was assessed by ELISA (n = 8 healthy donors). Values are mean ± SEM. *P < 0.05 using paired Student’s t test. (B) Monocyte-derived macrophage secretion of IL-6 and IL-12p40 correlates with CD206⁺MHC-II⁺ cell surface expression. Open circles, M0; solid circles, M2; r, Spearman’s correlation coefficient. (C) M2 macrophages suppress type 1 T helper cell cytokine production. DCs and CD4⁺ T cells were co-cultured without (shaded bars) or with macrophages (M) (left panel). Open bars, M0; solid bars, M2. Values are mean ± SEM from n = 11 independent experiments. *P < 0.05 using Student t test. The percent reduction of IFN-γ production induced by macrophages correlates with CD206⁺MHC-II⁺ cell surface expression (right panel). Open circles, M0; solid circles, M2; r, Spearman’s correlation coefficient.

M2 Macrophages Suppress DC and CD4⁺ T Cell Immune Responses

Macrophages are not very effective antigen-presenting cells in comparison to DCs (23). However, macrophages have been shown to alter lymphocyte responses (24, 25). We investigated this role by adding M0 or M2 macrophages to DCs and CD4⁺ T cells to assess their impact on adaptive immune responses. Neither M0 nor M2 macrophages stimulate IFN-γ or IL-13 cytokine production by CD4⁺ T cells (data not shown). In contrast, and consistent with the antigen-presentation capacity of DCs, there is significant IFN-γ production when CD4⁺ T cells are cultured with DCs (Figure 2C). However, M2 macrophages suppress Th1 immune responses, significantly decreasing IFN-γ production by DCs and CD4⁺ T cells (Figure 2C). Furthermore, this inhibitory effect, expressed as the percent decrease in IFN-γ production induced by macrophages, correlates with CD206 and MHC-II cell surface expression (r = 0.53; P = 0.03; Figure 2C), and suggests an immunomodulatory role for M2 macrophages in adaptive immunity.

Baseline Characteristics of Subjects with Asthma and Healthy Control Subjects

We next characterized the alveolar macrophage populations in the BALF of subjects with asthma and healthy control subjects. Subject characteristics are described in Table 1. In the asthma group, 71.4% of patients had mild-to-moderate asthma and 28.6% were affected with severe disease. As expected, patients with severe asthma had lower lung function and received higher doses of inhaled corticosteroid (ICS) than patients with mild-to-moderate asthma (74.0 ± 7.1% of predicted value versus 89.4% ± 3.1; P = 0.008 and 1,189 ± 201 μg beclomethasone equivalent versus 545 ± 112 μg; P = 0.002, respectively). None of the patients with severe asthma was receiving oral corticosteroids or anti-IgE therapy.

Table 1.

Patient Demographics

Characteristics	Asthma	Control Subjects
	(n = 28)	(n = 9)	P Value
Age, yr	37.2 ± 2.4	29.4 ± 3.7	0.11
Sex, male/female	11/17	6/3	0.25
Race or ethnic group, n
Caucasian	21	5	0.40
Hispanic	3	0	0.56
African American	3	2	0.58
Asian	1	2	0.14
Smoker
Current smoker, yes/no	0/28	0/9	1.00
Ex-smoker, yes/no	1/27	0/9	1.00
Pack-years, n	0.01	0.0	0.57
Age at onset of symptoms, yr	12.4 ± 2.6	n/a	—
Body mass index, kg/m2	27.5 ± 0.9	26.4 ± 2.4	0.60
Asthma severity, nonsevere/severe	20/8	n/a	—
Positive atopic status, yes/no	24/4	3/6	0.005
Total IgE, U/ml	520 ± 318	n/a	—
Severe exacerbations per subject in previous year, n	0.5 ± 0.2	n/a	—
Previous admission to the intensive care unit for asthma, yes/no	3/25	n/a	—
Score on Juniper Asthma Control Questionnaire	1.2 ± 0.2	n/a	—
Score on Asthma Quality of Life Questionnaire	5.5 ± 0.2	n/a	—
PC20 for methacholine, mg/ml	2.6 ± 0.9	n/a	—
Prebronchodilator FEV1, % of predicted value	84.2 ± 3.3	102.4 ± 3.6	0.005
FEV1:FVC ratio, %	75.0 ± 2.2	83.5 ± 1.9	0.045
Improvement in FEV1 after bronchodilator use, %	14.0 ± 2.2	2.8 ± 0.9	0.001
Eosinophil count in blood, ×109/liter	0.22 ± 0.03	0.07 ± 0.03	0.02
FeNO, ppb	29.1 ± 5.0	n/a	—
Use of ICS, yes/no	25/3	n/a
Dose of ICS–beclomethasone equivalent, μg	758 ± 116	n/a	—
Use of long-acting β-agonists, yes/no	17/11	n/a	—
Regular use of oral prednisolone, yes/no	0/28	n/a	—
Use of montelukast, yes/no	6/22	n/a	—
BALF cell count, ×106/ml	2.3 ± 0.4	0.1 ± 0.05	0.001
Macrophages, %	88.8 ± 2.1	87.1 ± 4.8	0.61
Neutrophils, %	2.1 ± 0.8	2.6 ± 1.8	0.61
Lymphocytes, %	6.5 ± 1.2	9.7 ± 3.7	0.33
Eosinophils, %	2.6 ± 1.4	0.6 ± 0.4	0.37

Definition of abbreviations: BALF, bronchoalveolar lavage fluid; Fe_NO, fraction of nitric oxide in exhaled air; ICS, inhaled corticosteroid; n/a, not applicable; PC₂₀, provocative concentration of inhaled methacholine required to lower the FEV₁ by 20%.

Plus–minus values indicate ± SEM. P values were calculated with the use of a two-sided independent t test for variables with a parametric distribution, Fisher’s exact test for comparison of proportions, and the Mann–Whitney U test for comparison of nonparametric variables.

M2 Phenotype CD206^hiMHC-II^hi Alveolar Macrophages Are Increased in Asthma

Flow cytometry analysis of BALF macrophages demonstrated that co-expression of CD206 and MHC-II on alveolar macrophages is higher in subjects with asthma compared with healthy subjects (Figure 3A). As described in the Materials and Methods, the MFI for CD206 and MHC-II on macrophages obtained from healthy control subjects was used to determine the percentage of cells expressing high levels of both CD206 and MHC-II (CD206^hiMHC-II^hi). The percentage of CD206^hiMHC-II^hi alveolar macrophages was 36.3 (±5.2)% in subjects with asthma as compared with 12.5 (±6.6)% in control subjects (P = 0.0001; Figure 3B), corresponding to a 2.9-fold increase in patients with asthma. There was variability in the expression of CD206 and MHC-II in subjects with asthma. Subjects, usually with mild-to-moderate asthma, had CD206^hiMHC-II^hi and CD206^loMHC-II^lo populations that were clearly demarcated, whereas more subjects with severe asthma had a more homogeneous population, with higher expression of CD206 and MHC-II (Figure 3A). CD206^hiMHC-II^hi alveolar macrophages exhibited greater cell size and overexpression of CD16, E-cadherin, and HRH1 as compared with CD206^loMHC-II^lo cells (Figure E5). The proportion of CD206^hiMHC-II^hi alveolar macrophages, as well as macrophage size and phenotype, were similar in subjects with nonsevere asthma and those with severe asthma (Figure E6). However, the proportion of CD206^hiMHC-II^hi alveolar macrophages was elevated in uncontrolled asthma (Asthma Control Questionnaire score > 1.5) compared with controlled asthma (Figures 3C and 3D). This may be related in part to the association between alveolar macrophage CD206 and MHC-II expression and eosinophilic inflammation in subjects with asthma (Figure E7). Moreover, there was no correlation between ICS dose and the percentage of CD206^hiMHC-II^hi cells in BAL (r = −0.27; P = 0.21).

Figure 3.

Alveolar macrophages with an M2 phenotype are increased in subjects with asthma. (A) Dot plot examples showing the percentage of CD206^hiMHC-II^hi macrophages in bronchoalveolar lavage from healthy control subjects (left panel), subjects with moderate asthma (middle panel), and subjects with severe asthma (right panel). (B) Percent CD206^hiMHC-II^hi macrophages of the total MHC-II⁺ alveolar macrophages in subjects without asthma (n = 9) and those with asthma (n = 28). (C and D) Percent CD206^hiMHC-II^hi macrophages of total MHC-II⁺ alveolar macrophages is greater in patients with uncontrolled versus controlled asthma and correlates with Asthma Control Questionnaire (ACQ) score. Dark gray bars, ACQ < 1.5; light gray bars, ACQ > 1.5. Values are mean ± SEM. *P < 0.05 using Mann–Whitney test. r, Pearson’s correlation coefficient.

HRH1 and E-Cadherin Expression on CD206^hiMHC-II^hi Alveolar Macrophages Correlate with Airway Obstruction in Asthma

Due to their important role in allergic responses and up-regulation on in vitro–derived M2, we also examined HRH1 and E-cadherin expression on in vitro–derived and alveolar macrophages (11, 26, 27). Indeed, HRH1 and E-cadherin mRNA are highly expressed on monocyte-derived M2 macrophages (Figure E1). HRH1 is up-regulated in asthma, as this receptor is overexpressed in CD206^hiMHC-II^hi macrophages from subjects with asthma as compared with CD206^hiMHC-II^hi macrophages from subjects without asthma (Figures 4A and 4B). HRH1 and E-cadherin expression and the percent of E-cadherin–positive cells were significantly increased on CD206^hiMHC-II^hi alveolar macrophages (Figure E5). We then assessed whether there was a correlation between HRH1 and E-cadherin expression on CD206^hiMHC-II^hi macrophages and asthma severity. HRH1 was more highly expressed on CD206^hiMHC-II^hi alveolar macrophages from subjects with asthma with more severe airway obstruction. Indeed, the HRH1 MFI index negatively correlates with the forced expiratory volume in 1 second (FEV₁; r = −0.93; P = 0.003) and the ratio FEV₁:forced vital capacity in subjects with asthma (Figures 4C and 4D). In contrast to HRH1, E-cadherin is not up-regulated in CD206^hiMHC-II^hi macrophages from subjects with asthma as compared with normal subjects (Figure E8). However, we found a negative correlation between the percentage of E-cadherin–positive alveolar macrophages and airway obstruction assessed by FEV₁:forced vital capacity (r = −0.64; P = 0.046) in subjects with asthma (Figure E8).

Figure 4.

Histamine receptor H1 (HRH1) expression is increased on M2 macrophages in asthma. (A and B) HRH1 cell surface expression on CD206^hiMHC-II^hi alveolar macrophages is higher in subjects with asthma (solid bars) than in subjects without asthma (open bars). Median fluorescent intensity (MFI) index is the MFI with anti-HRH1 monoclonal antibody (mAb) divided by the MFI of an isotype control mAb. (C and D) HRH1 expression on CD206^hiMHC-II^hi alveolar macrophages and correlation with lung function in patients with asthma (n = 7). Values are mean ± SEM. *P < 0.05 using Student’s t test; r, Pearson’s correlation coefficient.

Discussion

The division of macrophages into M1 and M2 subtypes provides a useful framework for the study of these major macrophage subsets in Th2 immunity. Recent work has brought to the forefront the distinct role of this innate immune cell in allergic inflammation. Th2 immune responses and IL-4 have been shown to drive macrophage proliferation and alternative activation in vivo (6). Eosinophils and ILC2, which have been implicated in asthma pathogenesis, have also been shown to drive AAM development and activation (4, 5). The evolutionary link between the alternative activation of macrophages and Th2 responses is highlighted by work that demonstrated that the ontogeny of eosinophils and M2 macrophages is directly linked (28, 29).

Macrophages are the most abundant immune cell in the lung, and account for approximately 70% of the immune cells (29, 30). Alveolar macrophages are the predominant macrophage subset. They account for over 55% of lung immune cells. In contrast to interstitial macrophages, alveolar macrophages demonstrate minimal turnover (29–31). Their close interposition with structural and immune cells that have rapid turnover may allow for more sustained contributions by macrophages to immune and local responses.

To identify alternative macrophage activation in asthma, we compared the phenotype and function of resting macrophages (M0) to IL-4–stimulated macrophages (M2). Indeed, IL-4 is a key cytokine in asthma pathophysiology and alternative macrophage activation, and its receptor is a new pharmacological target in severe asthma (32). In contrast, M1 polarizing agents, such as LPS, are not considered to play a primary role in asthma. In addition, bronchoscopy was performed in clinically stable subjects with asthma at least 6 weeks after any asthma exacerbation. We first examined phenotypic markers that have been previously observed to be elevated on in vitro–derived M2 macrophages (3, 11). We found that a combination of high levels of co-expression of the cell surface markers, CD206 and MHC-II, consistently identified macrophages polarized to a M2 phenotype. We next examined BALF alveolar macrophages obtained from healthy subjects and subjects with asthma. As was observed in vitro, macrophages with increased expression of both CD206 and MHC-II were identified in human BALF, and were increased 2.9-fold in subjects with asthma compared with healthy control subjects. There was no correlation between the percentage of CD206^hiMHC-II^hi macrophages and ICS dose. Moreover, surface marker expression in CD206^hiMHC-II^hi cells was similar in subjects with asthma treated with low-dose ICS versus high-dose ICS (data not shown). This would be consistent with prior observations that have found that airway macrophages in asthma are relatively corticosteroid insensitive (13, 33). Increased numbers of CD206^hiMHC-II^hi macrophages may reflect a Th2-high phenotype, greater disease severity, and contribute to corticosteroid resistance and the pathobiology of asthma.

Consistent with prior work, M2 macrophages produced increased levels of IL-10 (3, 16, 34, 35). We also observed increased levels of IL-6 and IL-12p40. M2 macrophages did not produce IL-12p70, which is critical for Th1 immune responses and IFN-γ production. In addition, increased IL-10 production by M2 macrophages may also down-modulate Th1 responses. Although we did not have sufficient cell numbers to explore these findings with BALF macrophages, they are consistent with prior observations showing that both IL-10 and IL-12p40 were significantly elevated in the sputum macrophages from subjects with asthma (16). IL-6 is a cytokine that has been implicated to have both pro- and antiinflammatory properties. Increased IL-6 was observed in the sputum of subjects with severe asthma and correlated with reduced FEV₁ (36). IL-6 also promoted naive CD4⁺ T cells to a Th2 phenotype, induced expression of the IL-4 receptor, and enhanced the response of macrophages to IL-4 (37, 38). As the most numerous immune cells in the lung, alveolar macrophages may be an important source of IL-6 and IL-10.

The direct contributions of M2 macrophages to allergic airway responses and the pathogenesis of asthma remain unclear. It is well appreciated that macrophages have poor antigen-presentation capacity compared with DCs (23). We also observed minimal cytokine production with incubation of either M0 or M2 with CD4⁺ T cells in the absence of DCs. We co-cultured autologous macrophages and DCs to which allogeneic CD4⁺ T cells were added. There was robust IFN-γ production when DCs were incubated with CD4⁺ T cells. In contrast to M0 macrophages, when M2 macrophages were added to the DC and CD4⁺ T cell cultures, IFN-γ production was significantly down-regulated. The down-modulation of Th1 responses by alternative macrophage activation has also been observed in vivo (24). M2 macrophages may lower the threshold for the development of adaptive Th2 immune responses, and may alter DC effector function in asthma. Further studies are needed to confirm these observations and better define which mechanisms are involved in the modulation of adaptive immune responses.

Although CD206 and MHC-II proved to be the most useful M2 markers in vitro and ex vivo, we observed that two additional markers were elevated on M2 macrophages both in vitro and in BALF. HRH1 levels were increased on M2 macrophages compared with M0, which is consistent with prior transcriptional studies (11). HRH1 was also more highly expressed on alveolar macrophages with M2 phenotype in subjects with asthma compared with normal subjects. In subjects with asthma, cell surface HRH1 levels were highest on the alveolar macrophages with M2 phenotype compared with M0, and negatively correlated with measures of increased airflow obstruction. Another pathognomonic allergic mediator, IgE, has also been implicated in AAM development. Human monocytes incubated with IgE developed into macrophages with increased HRH1 expression that produced high levels of IL-6 and low levels of IL-12p70, consistent with M2 macrophage phenotype (39). Taken together with the observations presented here, alternative macrophage activation may be linked through multiple pathways with Th2 immune mediators that may work synergistically in the promotion of allergic responses.

In addition to HRH1, we also observed an increase in E-cadherin expression on CD206^hiMHC-II^hi alveolar macrophages compared with CD206^loMHC-II^lo cells. In some patients with asthma, there was a distinct E-cadherin–positive population that may represent a unique subset of AAMs. Increased E-cadherin expression that is IL-4 and signal transducer and activator of transcription 6 driven has been previously reported on AAMs (17, 40). E-cadherin–expressing macrophages have been shown to mediate homotypic and heterotypic interactions (17, 40). There was a strong negative correlation between E-cadherin expression on alveolar macrophages and FEV₁, and may be a marker of more severe disease. E-cadherin on M2 macrophages may play a contributory role through homotypic interactions with both immune and structural airway cells that express E-cadherin, such as epithelial cells, DCs, ILC2, and other macrophages, or its heterotypic ligands, the integrin CD103 and the inhibitory receptor killer cell lectin-like receptor subfamily G member 1 (17, 27). Further work will be needed to determine the role of HRH1 and E-cadherin expression on the effector function of lung AAMs and in allergic responses. Additional subsets of AAMs and other markers, such as transglutaminase, which was recently demonstrated to identify M2 macrophages in both mice and humans, may contribute to Th2 responses and asthma phenotypes (12).

There are several limitations with these analyses. The macrophages sampled are those only accessible by BAL. There are likely macrophage subsets that are not sampled, such as interstitial macrophages, that may have distinct phenotypic and functional characteristics (29, 41, 42). These analyses do not address alveolar macrophage recruitment and the potential plasticity of macrophage phenotype in vivo either at steady state or in the setting of acute allergic inflammation (43). These analyses do not preclude an important role for M1 macrophages that have been shown to increase activation in subjects with asthma with a neutrophilic phenotype (44, 45). There were insufficient numbers of alveolar macrophages to characterize the effector properties of the phenotypic subsets that were identified. Additional studies with greater numbers of subjects will be needed to further delineate the role of alveolar macrophage subsets in asthma subtypes at steady state and in the setting of exacerbations.

The data presented here and in prior studies point to a potentially important role for alternative macrophage activation that can be identified by CD206^hiMHC-II^hi expression in asthma. The impact of M2 macrophages on Th2 immune responses is likely wielded through their bidirectional interactions with other immune cells, such as DCs, epithelial cells, and other structural cell types and expression and responses to allergic inflammatory mediators. Pharmacologic interventions that target M2 development and function may be prove to be important and synergistic with current asthma therapies.