Background
Pulmonary dendritic cells, activated Th2 effector cells, and their respective cytokine and chemokine networks in the sensitization and initiation phase of allergic airway inflammation have been explored intensively[
1‐
3]. The question, whether the most abundant resident cell type in the alveolar space, the macrophage, plays a significant role in these processes has been neglected in this setting for years[
4]. Increasing knowledge about macrophage polarization brought alveolar macrophages back into focus. In the context of allergic airway inflammation the contribution of the alternatively activated M2 phenotype with its specific functionality appears to be of special interest[
5].
M2 polarization encompasses at least two subtypes M2 and M2-like macrophages dependent on the cytokine milieu macrophages are subjected to. M2 macrophages differentiate in response to IL-4 and IL-13 whereas M2-like macrophages acquire their phenotype in response to TGF-b, IL-10 or PGE and additional TLR activation[
6]. M2-like macrophages release high amounts of IL-10 and are thus considered to be anti-inflammatory. However, M2 macrophages are functionally considered to promote clearance of parasite infections[
7], they are involved in tumor progression[
8], and they contribute to tissue remodeling[
9]. Markers for M2 macrophages differ between mice and man, with Ym1, Fizz1, Arg specifically described to be up-regulated in mice[
10], whereas expression of the mannose receptor CD206 was defined for M2 macrophages from both species[
11,
12]. In humans, several markers for M2 polarization have been described such as up-regulation of HLA-DR and increased expression of Th2 chemokines CCL17, CCL18, CCL22 and CCL24[
13]. Importantly, most data on macrophage polarization are derived from animal studies. M2 markers between mice and man are not identical, and our understanding of the characteristics of M2 polarization in human asthma is still incomplete[
14]. Recently, Staples and colleagues published an important paper in which they phenotypically characterized M2 polarization of alveolar macrophages in a group of non-asthmatic subjects and asthmatic patients[
15]. Using the currently accepted panel of human M2 markers, they showed a partial M2 polarization of sputum and BAL macrophages in symptom-free asthmatics. Interestingly, the CCR4 ligand CCL17, a key chemokine for the recruitment of CCR4
+ effector cells, was prominently induced, indicating that M2 macrophages might also facilitate the inflammatory immune response in humans.
With the current study we extended the phenotypic and functional characterization of alveolar macrophages in asthmatic patients, by paying special attention to the late phase of allergic airway inflammation. Therefore, we investigated macrophage polarization before and after endobronchial allergen challenge. As significant levels of IL-4 and IL-13 are predominantly present in the acute effector phase of the allergen response, we first assessed the kinetics of macrophage polarization in a murine asthma model beyond the acute phase of inflammation. To study M2 polarization, we further established a human in vitro model with monocyte-derived macrophages (MDM) of atopic donors. Subsequently, human alveolar macrophages isolated from non-asthmatic and mild asthmatic subjects undergoing an endobronchial allergen challenge were analyzed. Two doses of allergen were used in separate lobes of the subjects in order to investigate the impact of the severity of the inflammatory response on the extent of polarization. Our findings confirm and extend previous findings and provide data about the plasticity of macrophages in mild asthmatic subjects under stable non-inflamed conditions and during the late phase of the allergic airway inflammation.
Discussion
This study provides evidence for an augmented expression of CCL13, CCL17, and CLEC10A in AMΦ from patients with mild asthma during episodes of acute allergic airway inflammation. The expression of macrophage CCL17 and CLEC10A was related to the degree of eosinophilic inflammation. As increased amounts of CCL17 have been shown to attract CCR4
+ effector cells like eosinophils or specific Th2 lymphocyte subsets[
17,
18], it is tempting to speculate that macrophages play an important role in modulating the degree of allergic inflammation. Our data is in line with and confirms recent results of Staples and co-workers, who showed that human AMΦ from asthmatic patients display a partial M2 polarization already under non-inflamed conditions with increased mRNA expression of CCL17, CLEC10A and protein release of CLL17 and CCL22[
15].
It has been demonstrated in mouse models that macrophages undergo M2 polarization during the acute phase of allergic airway inflammation. However, the time course of polarization markers after resolution of the acute inflammation has not been investigated. Our data demonstrate that the up-regulated expression of established murine M2 polarization markers (Ym1, Fizz1, and Arg) outlive the inflammatory response. The decline of polarization markers over time might be due to replacement by new monocytes entering the alveolar space after resolution of the cellular inflammation, which is also supported by the concordant and continuous decline of all 3 investigated M2 related genes. An impaired phagocytosis of
E. coli by murine M2 macrophages supports the hypothesis that clearance mechanisms might be altered, making the airways more susceptible for bacterial or viral infection during this resolution period[
19]. Therefore, polarized macrophages seem to be capable to play a role in sustaining an inflammatory condition after acute allergen exposure.
We found several markers of M2 polarization to be induced in response to IL-4 in human MDM such as CCL17 and CLEC10A. These polarized macrophages induced allergen-specific T-cell proliferation and Th2 cytokine secretion. Importantly, IL-4 polarized human MDM do not fully reflect the AMΦ phenotype in asthmatics, as the cytokine milieu in vivo is not restricted to solely M2 stimuli.
It was previously shown that IL-4 stimulation of murine macrophages
in vitro leads to increased CCL17 release[
18,
20]. We could extend these findings for human AMΦ derived from allergen challenged segments as these cells express increased amounts of CCL17 on mRNA and protein level after allergen challenge. However, it remains to be elucidated to which extent macrophages account for the increased CCL17 levels in BAL of asthmatic patients since dendritic cells and epithelial cells have also been identified as prominent sources[
21‐
24]. We did not observe CCL17 release by native T-cells of asthmatic patients in our co-cultures[
25]. In opposite to findings of Staples, we found CCL17 mRNA already increased at baseline in macrophages from mild asthmatic patients compared to healthy subjects and additionally identified CCL13 to be up-regulated. CCL13 is a potent chemokine which attracts CCR3
+ cells like eosinophils, Th2-lymphocytes, basophils and correlates with asthma exacerbation[
26,
27]. CCL13 protein has been shown to be increased in BAL of asthmatic patients compared to healthy controls[
28]. Our mRNA results in AMΦ show that polarized macrophages represent an additional source of this protein beside epithelial cells in response to allergen provocation in the airways.
The macrophage galactose-type C-type lectin (MGL) CLEC10A which we and others found up-regulated on alveolar macrophages from asthmatic patients[
15] recognizes carbohydrate structures with terminal galactose or N-acetylgalactosamine residues, which are present on naturally occurring allergens[
12,
29]. Additionally, the mannose receptor (CD206)[
30] and IgG receptor (CD32)[
31] are described to be increased on macrophages from asthmatic patients. Whether this up-regulation is capable to alter phagocytosis of M2 macrophages remains to be elucidated. Interestingly, phagocytosis of bacteria is impaired in patients with severe asthma[
32,
33] and IgG-opsonized yeast are less effectively cleared by macrophages from asthmatic patients[
34]. Impaired phagocytosis of bacteria was also reported from various
in vitro models of M2 polarization, also supporting the hypothesis that pulmonary clearance of inhaled pathogens might be hampered[
35]. In our study, CD206 was not differentially expressed on macrophages from asthmatic patients compared to healthy subjects and not regulated in response to allergen exposure.
In asthmatic patients chitinases (chitotriosidase, CHIT1) and chitinase-like proteins (YKL-40) in serum and BAL correlate with the severity of the disease, the degree of airway remodeling, and the frequency of asthma exacerbations[
36‐
38]. These molecules are released from activated monocytes and macrophages and are suggested to be linked with remodeling processes[
39,
40]. CCL18, a macrophage-derived chemokine has been described to attract T-lymphocytes and to promote collagen production by human fibroblasts[
41,
42]. Interestingly, we did not observe CCL18 up-regulation in alveolar macrophages 24 h after segmental allergen challenge, while we found CCL18 increased in MDM
in vitro at 48 h after IL-4 stimulation. This is in line with a recent
in vitro study investigating the kinetics of CCL18 induction showing maximum release at 72 h but no significant change at 24 h after stimulation with allergen[
41]. In line with older reports, we found an increased expression of HLA-DR and other co-stimulatory surface markers on M2 macrophages[
31,
43]. Although the results from our
in vitro allergy model show that M2 macrophages induce T-cell proliferation and the release of Th2 cytokines, we were not able to confirm the magnitude of proliferation and cytokine release in similar experiments using AMΦ. Consequently, this questions their relevance for allergen presentation. As a limitation, we did not investigate whether macrophages from asthmatic patients exhibit a suppressive function on DC-induced T-lymphocyte proliferation as has been reported before[
44].
There is increasing evidence that the different models available to study the concept of macrophage polarization are providing controversial results. Murine models of allergic asthma revealed that the course of the allergic airway inflammation was not influenced by the presence or absence of IL-4 receptor-α bearing macrophages which become polarized during the allergic inflammation[
45]. In contrast, the transfer of IL-4 receptor-bearing macrophages aggravated the eosinophilic inflammation in IL-4 receptor-deficient mice[
46]. However, in an animal model of acute exacerbation M2 macrophages triggered Th2 cytokines in CD4
+ T-lymphocytes through the interaction with CD80/CD86, which was not seen in a model of mild chronic asthma[
47]. A rhinovirus-induced exacerbation study identified M2 macrophages after allergen challenge which subsequently released more Th2 cytokines and CCL11 after additional rhinovirus infection. On the other side, depletion of macrophages before allergen/rhinovirus challenge lead to reduced eosinophilia, CCL11 and IL-13 levels[
48]. Thus, functional consequences of macrophage polarization might become more apparent during asthma exacerbation phases.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
CW has made substantial contributions to conception and design, acquisition and interpretation of data and drafting the manuscript. LW carried out the sample preparation and analysis (RT-PCR, FACS and cytokines) of human pulmonary cells. NM carried out the mouse model and the MDM in vitro data. CF and FS performed the bronchoscopy study. MM contributed to the study conception and design. OH performed the statistical analysis and contributed to draft the manuscript. JMH has made substantial contributions to the study design, drafting the manuscript and funding acquisition. All authors read and approved the final manuscript.