Increased TREM-2 expression on the subsets of CD11c+ cells in the lungs and lymph nodes during allergic airway inflammation

Dendritic cells (DCs) are professional APCs that traffic to the draining lymph nodes where they present processed antigens to naïve T-cells. The recently discovered triggering receptor expressed on myeloid cells (TREM)-2 has been shown to be expressed on DCs in several disease models, however, its role in asthma is yet to be elucidated. In the present study, we examined the effect of allergen exposure on TREM-2 expression in the airways and on DC subsets in the lung and lymph nodes in murine model of allergic airway inflammation. Sensitization and challenge with ovalbumin reproduced hallmark features of asthma. TREM-2 mRNA expression in the whole lung was significantly higher in the OVA-sensitized and -challenged mice which was associated with increased protein expression in the lungs. Analysis of CD11c+MHC-IIhi DCs in the lung and draining lymph nodes revealed that allergen exposure increased TREM-2 expression on all DC subsets with significantly higher expression in the lymph nodes. This was associated with increased mRNA expression of Th2 and Th17 cytokines. Further analyses showed that these TREM-2+ cells expressed high levels of CCR-7 and CD86 suggesting a potential role of TREM-2 in mediating maturation and migration of DC subsets in allergic airway inflammation.

pro-inflammatory signals based on the micro-environment. TREM-2 expression was found to be upregulated on bronchoalveolar lavage fluid cells of patients with pulmonary sarcoidosis 20 , and its deficiency on alveolar macrophages resulted in augmented bacterial clearance, decreased bacteremia and improved survival compared to wild type animals 21 . Viral and bacterial infections, which are known to exacerbate inflammation and impair the lung response in respiratory diseases like asthma, have been associated with increase intracellular and cell surface levels of TREM-2 on macrophages 22,23 . Studies using human DCs have shown that TREM-2 activation, via the DAP-12 pathway, promoted upregulation of CCR-7, partial DC maturation and DC survival through activation of protein tyrosine kinase (PTK) and extracellular signal-regulated kinase (ERK)-mediated signaling 24 . In the intestine, TREM-2 was shown to contribute to mucosal inflammation during colitis with TREM-2 −/− DCs displaying lower production of inflammatory cytokines in response to TLR ligands 25 .
Although TREM-2 provides both activating and inhibitory signals in several disease models, little is known about its expression in the airways and it is still unclear if the receptor plays a role in the pathogenesis of asthma. Given that TREM-2 has been shown to drive inflammation in other disease models, we hypothesized that TREM-2 might play a role in the onset and progression of allergic airway inflammation. Here, we examined the effect of allergen sensitization and challenge on TREM-2 expression in the airways and on DC subsets isolated from the lungs and mediastinal lymph nodes in a murine model of allergic airway inflammation.

Results
Sensitization and challenge with ovalbumin reproduced hallmark features of asthma. To study the effects of allergen exposure on TREM-2 expression, BALB/c mice were sensitized and challenged with ovalbumin or sterile PBS as described in Fig. 1A. OVA-sensitized and challenged mice displayed significantly higher airway response to all concentrations of methacholine (3.125 mg/ml-100 mg/ml) as determined non-invasively using whole body plethysmography and invasively using anesthesia, tracheal intubation and mechanical ventilation to determine specific airway resistance (R L ) (Fig. 1B). Following euthanasia, blood and BALF cells were collected. Analysis of BALF revealed that OVA-sensitized and challenged mice showed greater total number of leukocytes (Fig. 1CI) as well as significantly higher percentages of eosinophils, neutrophils and lymphocytes when compared to the PBS controls (Fig. 1CII). OVA-sensitization and challenge resulted in significant increases in total IgE levels in BALF and serum when compared to the PBS-sensitized and challenged groups (p < 0.01). When we analyzed OVA-specific IgE in the serum, OVA-sensitized and challenged groups had significantly higher levels of OVA-specific IgE, which was not detected in control groups ( Fig. 1DI-III). To visualize morphological changes in the airways, lung sections were stained with hematoxylin and eosin, periodic acid-Schiff (PAS) reaction and Masson's trichrome stain (Fig. 1E). Lung sections from mice sensitized and challenged with OVA showed observable structural changes including narrowing of the lumen and increased cellular infiltration (Fig. 1EIV-VI), increased mucus secretion (Fig. 1EV) and collagen deposition (Fig. 1EVI) which were absent in PBS-sensitized and challenged groups ( Fig. 1I-III). These results show that our mouse model can generate hallmark features of allergic asthma.
Sensitization and challenge with OVA increases TREM-2 expression in the airways. To ascertain whether TREM-2 was expressed in the airways, we first examined mRNA expression of TREM-2 in the lungs isolated from OVA-sensitized and challenged mice as well as PBS control groups. Using RT-PCR, it was found that TREM-2 mRNA transcripts were significantly upregulated in OVA-sensitized and challenged groups when compared to control mice ( Fig. 2A) (p < 0.05). To determine if TREM-2 was expressed on CD11c + cells in the airways, fixed sections were stained for dual expression of CD11c (pan DC marker) and TREM-2 and co-localization determined by fluorescent microscopy. Lung sections from OVA-sensitized and challenged groups had more than twice the number of cells expressing both CD11c and TREM-2, as indicated by yellow-orange staining, when compared to PBS controls (Fig. 2B,C), suggesting that TREM-2 might contribute to the inflammatory response associated with allergic asthma.
OVA-sensitization and challenge generates distinct subsets of CD11c + cells in the airways and lymph nodes. Given that there was colocalization of TREM-2 with CD11c in the airways, our next step was to determine the phenotype of CD11c + cells generated after exposure to ovalbumin. Lung lobes from PBS and OVA-sensitized and -challenged mice were isolated, processed and analyzed by flow cytometry to determine the phenotype of DCs in the airways and mediastinal lymph nodes. After gating out debris and doublets, live cells were selected and gated based on expression of CD64, CD24 and high levels of MHC-II to determine mature DCs and eliminate alveolar macrophages (macrophages are CD64 + and CD24 − ). For our analyses, we selected the subsets of cells that were CD64 − (blue histogram), CD11c + MHC-II hi and CD24 + (red histogram). These cells were further analyzed based on the expression of CD11b and CD103 (Fig. 3A,B). In the airways, it was found that the OVA-sensitized and -challenged groups had greater percentage of mature CD11c + MHC-II hi cells when compared to the PBS group (Fig. 3C). Further analyses of this subset of cells revealed that there were four populations of CD11c + cells in the airways: CD11b + CD103 − (CD11b hi ), CD11b lo/− CD103 + (CD103 + ), CD11b + CD103 + and a subset of cells that did not express either markers (CD11b − CD103 − ) (Fig. 3A,B). Most of the cells isolated from the PBS-sensitized and challenged groups were negative for both CD11b and CD103. In the OVA-sensitized and -challenged groups however there were significant increases in the percentages and total number of cells which were CD11b + CD103 − (CD11b hi ) ( Fig. 3B and D). As seen in the airways, four populations of CD11c + cells were identified in the mediastinal lymph nodes: CD11b + CD103 − (CD11b hi ), CD11b − CD103 + (CD103 + ) CD11b + CD103 + and CD11b − CD103 − (Fig. 3E). Interestingly, there was an increase in the percentage of CD11b + CD103 + when compared to the lungs and a decrease in the CD11b hi subset. Overall, the CD11b + CD103 − (CD11b hi ) subsets had significantly higher percentage of cells when compared to the CD11b − CD103 + (CD103 + ) (p < 0.01) and CD11b + CD103 + subset (p < 0.05) (Fig. 3F). When we examined DC populations in the lymph SCiENTifiC RepoRts | 7: 11853 | DOI:10.1038/s41598-017-12330-6 nodes of PBS-sensitized and challenged mice, very few cells were found (data not shown). These data show that CD11c + cells in the lungs and lymph nodes consists of a heterogeneous population of cells with increased CD11b subsets in the airways and increased double positive subsets in the lymph nodes.
CD11c + cells in the lung and lymph nodes express TREM-2. The immunofluorescence data showed that TREM-2 was found to be expressed on CD11c + cells in the airways and our previous results showed that this is a heterogeneous population. To confirm that TREM-2 was expressed on the subsets identified, each population was further analyzed for TREM-2 expression by FACS. All subsets of cells isolated from the lungs and lymph nodes express some amount of TREM-2. In the airways, TREM-2 expression was found to be higher on the subsets of cells isolated from OVA-sensitized and challenged mice ( Fig. 4A and C) when compared to the PBS controls (data not shown). Of the three DC subsets identified, TREM-2 expression was highest on the double positive (CD11b + CD103 + ) and CD11b hi subsets (Fig. 4C). Similar results were found on the subsets of cells isolated from the mediastinal lymph nodes (Fig. 4B,C), confirming that CD11c + cells in the lungs and lymph nodes express TREM-2. When we compared TREM-2 expression among the subsets in the lungs and lymph nodes, it was found that there was significantly increased TREM-2 expression on all subsets of cells in the lymph nodes when compared to those in the airways (p < 0.05). The CD11b hi and CD103 + subsets had the greatest overall increase in TREM-2 expression in the lymph nodes, with approximately twice the percentage of TREM-2 + cells (p < 0.01 and p < 0.001 respectively) when compared to the lungs (Fig. 4C). Results suggest that there is increased TREM-2 expression on CD11c + cells migrating from the lungs to the lymph nodes after exposure to ovalbumin. TREM-2 + cells express higher CCR-7 and CD86 when compared to their TREM-2 − counterparts. Given that we saw increased TREM-2 expression on subsets of cells in the lymph nodes and CCR-7 is crucial for migration of dendritic cells, we next determined expression of CCR-7 and CD86 on TREM-2 positive and negative subsets isolated from the lungs and lymph nodes of OVA-sensitized and challenged mice. In the lungs, TREM-2 positive cells had greater CCR-7 and CD86 expression when compared to their TREM-2 − counterparts ( Fig. 5A-C). CCR-7 expression was highest on the CD103 + and CD11b + CD103 + TREM-2 + subsets (p < 0.001 and p < 0.01 respectively) with no significant differences seen in the CD11b hi TREM-2 + subsets (Fig. 5B). Though not significant, CD86 expression was higher on TREM-2 + subsets with the CD11b hi subset of cells having the greatest CD86 expression (Fig. 5C). In the lymph nodes, both CCR-7 and CD86 expression was increased on all TREM-2 + subset of cells (Fig. 5D-F). CCR-7 expression was significantly higher on all three TREM-2 + subsets when compared to the TREM-2 − cells (p < 0.05), with the double positive subset having the greatest CCR-7 expression (Fig. 5E). These data show that there is increased CCR-7 and CD86 expression on TREM-2 + cells in the lungs and lymph nodes highlighting a potential role of TREM-2 in the migration and maturation of CD11c + cells.

Increased TREM-2 expression in the lymph nodes is associated with increased Th2 and Th17 responses.
To determine the T helper cell subsets that were being generated by increased TREM-2 expression on DCs migrating to the lymph nodes, we examined mRNA expression of Th1, Th2, Th17 and Treg cell cytokine secretion using RT PCR as well as expression of transcription factors using FACS in the mediastinal lymph nodes. Our results showed significant increases in mRNA expression of IL-4, IL-6, IL-17 and TGF-β in the lymph nodes isolated from OVA-sensitized and challenged mice compared to control groups (p < 0.05; Fig. 6A). Of the cytokines examined, IL-4 and IL-17 had overall greatest increase Relative expression of IL-10, IL-12 and IFN-γ were only marginally higher in the OVA-sensitized and challenged groups and showed no significant differences when compared to the controls (Fig. 6A). Analysis of FACS data revealed that approximately 30% of cells isolated were CD4 + T-cells. Further analyses of these cells showed increased expression of GATA-3 when compared to T-bet (p < 0.001) and increased RORγt expression when compared to FoxP3 (p < 0.05; Fig. 6B-C). These results suggest that increased TREM-2 expression on DCs in the lymph nodes play a role in driving Th2 and Th17 responses.

Discussion
Dendritic cells are the primary antigen presenting cells that control the initiation of antigen-induced immune response in the airways. Lung DCs are a heterogeneous population of cells which express CD11c and are further divided into two major subsets based on the expression of CD11b and CD103 5,6 . Several studies have shown that the phenotype of lung DC subsets plays a role in determining the nature of the immune response to inhaled antigens. Results from these studies suggest that both CD11b + and CD103 + DCs can drive inflammation based on the dose and route of administration of the allergen as well as the duration of the antigen-sensitization and challenge protocol [7][8][9][10][11] .
In the steady state, we identified three populations of cells in the airways: CD11b + CD103 − (CD11b hi ) and CD11b − CD103 + (CD103 + ), as reported in the literature, as well as another subset of cells that expressed both CD11b and CD103 (CD11b + CD103 + ). Exposure to ovalbumin resulted in an increase in the percentage of cells that were CD11b hi and slight decreases in the percentage of cells that were CD11b + CD103 + and CD103 + , though total cell numbers of each subset was higher in the OVA-sensitized and challenged groups. It has been shown that exposure to allergen leads to rapid recruitment of monocyte-derived CD11b hi DCs to the airways 26,27 . CD11b hi DCs also secrete a host of pro-inflammatory chemokines which mediate chemotaxis of immune cells driving allergic airway inflammation 5,6 . Additionally, CD11b hi DCs have been shown to be more efficient in OVA uptake than CD103 + subset of cells 7 which could account for the vast increase in this population observed in the airways. Our results, as well as the finding from other groups, suggest that exposure to ovalbumin increases CD11b hi DCs in the airways and this subset of cells is possibly the one driving localized inflammation observed in the airways of these animals. As seen in the airways, three populations of DC subsets were also identified in the draining lymph nodes. When compared to the lungs, there was an increase in percentage of CD103 + and CD11b + CD103 + cells and a decrease in the percentage of CD11b hi cells. It should be noted that total numbers of cells in this population was still the highest, further confirming that the CD11b hi subset is the major contributor in the generation and progression of the inflammatory response seen in the OVA-sensitized and challenges group.
TREM-2 belongs to a family of recently discovered cell surface receptors which have been shown to play a role in fine tuning innate responses induced by TLRs 17,28 . Although its ligand is yet to be identified, TREM-2 mediates signaling via association with the DAP-12 adaptor protein 16 . It is generally regarded as an anti-inflammatory mediator but has been shown to amplify the production of pro-inflammatory cytokines and upregulate CCR7 expression, which is crucial for migration of dendritic cells to the draining lymph nodes 18,24,25,29 . Recently, TREM-2 has also been shown to be upregulated on DCs that contribute to bone destruction in a murine model of acquired cholesteatoma 30 .
Although the receptor has been implicated in a number of disease models 17,21,31,32 , its expression and function has not been clearly elucidated in allergic airway inflammation. As seen with other inflammatory conditions 20,22,25,30,31 , TREM-2 was found to be upregulated in the airways of mice exposed to OVA when compared to the control groups. In steady state, TREM-2 was expressed on all subsets of DCs identified in the lungs confirming that DCs do express TREM-2. OVA-sensitization and challenge led to upregulation of TREM-2 expression on the three subsets of CD11c + cells in both the lungs and mediastinal lymph nodes. Increased TREM-2 expression in the lymph nodes suggest that TREM-2 might be involved in the mechanism driving migration of DC subsets.
Further examination of TREM-2 positive subsets in the lungs and lymph nodes revealed that these cells had greater CCR-7 expression compared to their TREM-2 negative counterparts. It is well established that upregulation of CCR-7 is critical for migration of DCs to the draining lymph nodes and DCs migrate under the influence of lymphatic chemokines CCL-19 and CCL-21. We have previously shown that OVA-sensitization and challenge resulted in increased expression of CCL-19 and CCL-21 in the T-cell zone of the cortex and upregulated CCL-21 in the afferent lymphatic vessels and high endothelial vesicles in the mediastinal lymph nodes 12 . Subsets of cells that expressed higher levels of CCR-7 were also shown to be better at migrating towards CCL-19 and CCL-21 in vitro 33 . Of the three subsets of cells identified in the airways, the CD11b hi subset, though the predominant phenotype, was shown to have the lowest overall CCR-7 expression. Studies have shown that while lung resident CD11b hi DCs are more efficient at antigen uptake, they migrate poorly to the lymph nodes and tend to remain in the airways where they mediate the production of pro-inflammatory cytokines 7,34 . If the CD11b hi lung DC population was made up of some infiltration monocyte-derived CD11b hi DCs, these cells have been shown to express little-to-no CCR-7 even after stimulation 35 . Since CCR-7 is crucial for dendritic cell migration, lack of expression on a population of these cells would prevent them from migrating, thus decreasing the percentage of this phenotype observed in the lymph nodes.
Our findings suggest that increased TREM-2 expression on subsets of cells in the lymph nodes may be driving Th2 and Th17 responses given the marked increase in mRNA expression of IL-4, IL-6 and IL-17 in OVA-sensitized and challenged mice. This suggests some division of labor among the subsets of cells identified in our studies. Several studies have shown that CD11b + DCs secrete a host of pro-inflammatory mediators and play a role in priming and re-stimulating effector CD4 + T-cells driving Th2 responses 5,6,9,11,27 . The CD103 + DCs have also been shown to prime Th1 as well as Th17 responses 11 . It is now well known that Th17 cytokines induce mucus cell metaplasia, neutrophil recruitment, AHR and airway remodeling 36 . The increased numbers of CD103 + DCs in the lymph nodes, increased TREM-2 expression on these cells as well as increased expression of IL-6, TGF-β and IL-17 highlight a potential role of these cells in driving Th17 responses in OVA-sensitized and challenged mice. Overall, TREM-2 expression was found to be highest on the double positive subset of cells in both the lungs and lymph nodes with increased total number of these cells in the lymph nodes. This subset of cells has yet to be characterized in the airways, however, studies have shown that during inflammatory conditions, CD11b + DCs can express CD103 5,37 leading to the observed phenotype seen in our results. CD11b + CD103 + dendritic cells have been recently characterized in the intestine as bona fide dendritic cells that are capable of priming Th17 cells 38,39 . Given that our results have also shown that there was increased mRNA expression of Th17 associated cytokines, which are known to play a critical role in airway neutrophilia and AHR, further studies are required to determine if a similar response occurs in the airways.
The current studies were focused on TREM-2 expression on DC subsets in the lungs and lymph nodes. It is important to note that TREM-2 has been found to be expressed on other cell types in the airways, including macrophages, which may also contribute to allergic airway inflammation 22,40 . When we examined TREM-2 expression on the CD64 + CD11c lo CD11b + subset of cells in the airways, there was very mild expression found on these cells. More in-depth studies using macrophage-specific markers to accurately gate this population of cells with high purity would be needed to determine the exact contribution of TREM-2 expressing macrophages to the response seen in OVA-induced lung inflammation.
TREM-2 upregulation in the lungs and lymph nodes along with increased expression of transcripts for Th2 and Th17 cytokines in the lymph nodes of mice exposed to ovalbumin could indicate that the receptor is participating in the onset or progression of allergic airway inflammation. Two recent studies have implicated the receptor as a pro-inflammatory mediator in airway inflammation. Analysis of BALF from patients with asthma showed increased expression of TREM-2 which was associated with increased eosinophil and eosinophilic inflammation 41 . Another study has shown that TREM-2 was upregulated in the lungs of an experimental model of melioidosis and TREM-2 deficient mice had markedly reduced inflammation 40 . Other studies have shown that the receptor plays a role in dampening the immune response. It is, therefore, reasonable to assume that the induction of TREM-2 on DC subsets in the airways and lymph nodes could indicate an impaired defense mechanism originally aimed at decreasing rather than enhancing inflammation.
The mechanism by which TREM-2 might be driving inflammation remains unclear. Results from existing studies seem to suggest that the cell type that expresses the receptor as well as the cytokine milieu might influence its role in inflammation. When expressed on macrophages and microglia, TREM-2 seems to play more of an anti-inflammatory role 29,42,43 . However, based on the results from our studies as well as other groups, when expressed on dendritic cells, the role seems to be reversed 25,30 . It is our speculation that the receptor could be contributing to allergic airway inflammation by amplifying pro-inflammatory signals generated by TLRs or other  Bronchoalveolar Lavage Fluid Analysis and IgE Measurements. Immediately after euthanasia, blood was collected from the left ventricle of the heart and the separated serum was stored at −80 °C for further analysis. Bronchoalveolar lavage fluid (BALF) was collected by lavaging the lungs with 1 ml of warm PBS. The samples were then centrifuged at 400 × g for 10 minutes and the supernatant collected and stored at −80 °C for further analysis. The cell pellet was resuspended in PBS and total cell counts were performed using Countess Automated Cell Counter (Invitrogen, Grand Island, NY). Cells were immobilized on glass slides by cytospin centrifugation (Shandon Cytospin 4, Thermo Fisher Scientific, Waltham, MA), fixed in acetone and stained using Diff-Quik staining reagent (StatLab Medical Products, Lewisville, TX) according to manufacturer's instructions. Differential cell counts were carried out by randomly counting 300 cells as visualized under light microscopy.
Total and OVA-specific IgE in BALF and serum were measured using commercially available ELISA (eBioscience, San Diego, CA and BioLegend, San Diego, CA) according to manufacturers' instructions. Immunofluorescence. Paraffin embedded sections were stained for co-expression of TREM-2 and CD11c.
Briefly, non-specific binding was blocked by incubating sections with 5% rabbit serum in PBS for 1 hour. Sections were washed twice with PBS and incubated with primary mouse antibodies to TREM-2 (1:200; abcam, Cambridge, MA) and CD11c (1:200; abcam, Cambridge, MA) overnight at 4 °C. The sections were then washed and incubated with rabbit anti-goat Alexa Fluor (1:500; abcam, Cambridge, MA) and rabbit anti-hamster FITC (1:500; ImmunoReagents, Raleigh, NC) secondary antibodies for 2 hours at room temperature. Following the final PBS washes, sections were fixed using Vectashield mounting media with DAPI (Vector Laboratories, Burlingame, Figure 7. Schematic diagram of proposed role of TREM-2 in allergic airway inflammation. Exposure to antigen results in either activation of TLRs and/or DAP-12 DC receptors or directly leads to upregulation of TREM-2 on DCs. TREM-2 then amplifies these pro-inflammatory signals leading to upregulation of CCR-7 and CD86 on DC subsets. This promotes migration and maturation of these cells leading to increased antigen presentation and priming of Th2 and Th17 response, thus promoting airway inflammation. Statistical Analyses. Flow cytometric analyses were carried out using FlowJo Data Analysis Software v10.0 (Tree Star Inc, OR). All other data were analyzed using GraphPad Prism version 6.00 (GraphPad Software, La Jolla, CA). Unpaired student's t test was used to determine differences between groups. Multiple group comparisons were made using one-way ANOVA with Tukey's post-hoc tests. Values are expressed as means ± SEM. A value of p < 0.05 was considered significant. Data Availability. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Conclusions
In conclusion, we report that TREM-2 is expressed on all subsets of dendritic cells in the airways and is upregulated after allergen sensitization and challenge with ovalbumin. This upregulation correlated with increased CCR-7 expression which plays a critical role in migration of DCs to secondary lymphoid organs as well as increased mRNA expression of Th2 and Th17 cytokines. Based on increased expression of CCR-7 and costimulatory molecule CD86 on DC subsets in the draining lymph nodes, the receptor could potentially be playing a role in survival, migration and/or maturation of DC subsets. More in depth studies are however required to confirm this. The findings of this study highlight a potential role of TREM-2 in allergic airway inflammation and the receptor might well be a novel target for therapeutic intervention.