Pulmonary antigen-presenting cells, especially dendritic cells (DCs), express the surface marker CD11c [13] and have a potent ability to induce the proliferation and activation of naïve T cells and to secrete inflammatory and T-helper cell polarizing cytokines [14-16]. CD11c⁺ cells are critical for initiating and shaping the antigen-specific adaptive immune response and are critical during the reactivation of CD4⁺ T cells in vivo [17]. CD11c⁺ DCs are capable of these activities because they possess multiple unique characteristics. First, DCs are strategically located beneath the airway epithelium and continually take up antigen under steady-state conditions [15]. Second, DCs can undergo maturation upon exposure to inflammatory stimuli and travel to draining lymph nodes, presenting antigens in the context of both MHCI and MHCII. Finally, DCs express co-stimulatory molecules and secrete polarizing cytokines necessary to initiate and shape the T cell mediated immune response [16,18]. However, defining DCs via surface marker expression remains complicated, especially in non-lymphoid tissues such as the lung, due to the number of different methods described in the literature and the shared cell surface markers expressed by several cell subsets. The myeloid DC subset is attributed with T cell stimulatory capacity, having the ability to induce Th1, Th2, or Th17 type responses [19], as well as non-inflammatory T regulatory (Treg) responses [20]. Myeloid DCs in the lung have been defined as CD11c⁺CD11b⁺ [19,21], CD11c⁺MHCII⁺ [22], or CD11c⁺ alone or in combination with low FITC auto-fluorescence [23,24]. This variation is further complicated by the overlap of markers with multiple other cell types, the most prominent of which in the lung is CD11c⁺ macrophages [21]. Plasmacytoid DCs (pDCs) are also present within the lung [25] and have been shown to exert an anti-inflammatory role, decreasing both the ability of mDCs to generate effector T cells as well as inducing Treg cell proliferation [26]. This pDC population expresses B220 (CD45RB) as well as low levels of Gr-1, making them more difficult to delineate from B cells and granulocytes than myeloid DCs [18]. Recently, another population of dendritic cells called inflammatory DCs has been described (CD11b⁺Gr-1^loF4/80^lo), which traffic from the blood to sites of inflammation, upregulate CD11c upon arrival in the tissue, and acquire the DC characteristics of migration to the draining lymph node and induction of T cell proliferation [22,27].

DCs are capable of skewing the T helper cell response through their ability to express distinct patterns of co-stimulatory molecules as well as to produce cytokines that create an environment for differential T cell polarization [19]. Expression of the co-stimulatory molecules CD86 and OX40L have been shown to promote naïve CD4⁺ T cells to develop a Th2 phenotype [19,28,29]. Importantly, OX40L-deficient mice are protected from allergic sensitization and Th2 mediated inflammation, indicating that OX40L plays a critical role in the generation of Th2 immune responses [29]. DCs also regulate Th2 cell differentiation and expansion by producing IL-6 [30,31]. Th17 cells may also be induced by DC production of IL-6 in combination with TGFβ or IL-23 [30,32,33], or by IL-1β alone [34], while IL-12 alone promotes a Th1 response [35]. Thus, activated DCs as a cellular source of co-stimulatory molecules and polarizing cytokines function as important regulators of CD4⁺ mediated T cell responses.

Since activation of pathogenic CD4⁺ Th2 cells is believed to be dependent on CD11c⁺ DCs [36], we tested whether CD11c⁺ cells are necessary for NO₂-promoted allergic sensitization. We used transgenic mice in which CD11c⁺ cells can be temporarily depleted due to the incorporation of a transgene encoding a simian Diphtheria Toxin Receptor (DTR) and Green Fluorescent Protein (GFP) fusion protein under the control of the murine CD11c promoter (CD11c-DTR mice) [37,38]. As murine cells lack the DT receptor [17] and only CD11c⁺ cells in CD11c-DTR Tg⁺ mice express the DT receptor, administration of DT to these mice results in inducible ablation of 90% of CD11c⁺ cells within the mouse lasting for approximately 48 hours [38]. Using CD11c-DTR mice, it has been demonstrated in models of allergic asthma that CD11c⁺ DCs are important during allergic sensitization induced by the i.p. injection of ova in combination with the adjuvant Alum [22] and are also critical during allergen challenge [17]. Importantly, the use of these mice to assess the role of CD11c⁺ cells during allergic sensitization via inhalation, and in response to NO₂, is so far unpublished.

CD11c⁺ DCs are critical in activating naïve CD4⁺ T cells [16], the products of which are primarily responsible for the symptoms of allergic asthma [36,50]. Multiple inhaled environmental agents have adjuvant activities and lead to allergic sensitization, including endotoxin [51,52], triacylated lipopeptides [53], residual oil fly ash [54,55], diesel exhaust particles [56-58], cigarette smoke [59], ultrafine particles [60], and house dust mite [61,62]. Proposed mechanisms for this inappropriate immune response promoted by environmental agents include disruption of the epithelial barrier, leading to cytokine and chemokine production by epithelial cells, as well as antigen access to underlying DCs [25,63] and altered DC function [60,64-69]. Because CD11c⁺ DCs are crucial for mounting naïve T cell responses, we hypothesized that eliminating CD11c⁺ cells during NO₂-promoted allergic sensitization would minimize features of allergic asthma in mice. To test this hypothesis, we utilized transgenic CD11c-DTR mice, in which temporary depletion of CD11c⁺ DCs is possible.

To use the CD11c-DTR mice, we adjusted our model of NO₂-promoted allergic sensitization to include only one sensitization. Our data demonstrate that even a single exposure to 15 ppm NO₂ for 1 hour followed by 30 minutes of aerosolized ova immediately after is sufficient to induce allergic sensitization to ova. Interestingly, we measured substantial increases in IL-4, IL-5, IL-13, and IL-17 from restimulated CD4⁺ T cells as well as substantial increases in eosinophilia and neutrophilia in the BAL, indicating that our one sensitization model polarizes CD4⁺ T cells down both Th2 and Th17 pathways. These pathways overlap as the mediator IL-6 is important in promoting both pathways [30-32] and was secreted at elevated levels by DCs from NO₂-exposed mice. Systemic administration of DT to CD11c-DTR transgenic mice prior to sensitization resulted in a decrease in the total white blood cell count and eosinophilia in the BAL of Tg⁺ animals versus Tg^- animals. Similar differences were measured for cytokine production from in vitro restimulated CD4⁺ T cells. Thus, systemic depletion of CD11c⁺ cells provides mechanistic insight into the important role of CD11c⁺ cells in NO₂-promoted allergic sensitization. However, our findings suggest that while CD11c⁺ cells are important in NO₂-promoted allergic sensitization, depleting these cells during sensitization does not abolish all parameters of the allergic asthma phenotype, as we were unable to measure a difference in the levels of ovalbumin-specific immunoglobulins from the serum of Tg⁺ and Tg^- animals. This may be due, in part, to the low levels of antigen-specific immunoglobulins generated as a consequence of our modified sensitization regimen in which mice were only exposed to inhaled NO₂ and ova on a single occasion compared to our previously-published model that involved two sensitizations [6]. Importantly, when DT was administered via i.p. injection, depletion of CD11c⁺ cells within the lung occurred as expected in Tg⁺ animals [17]. However, this depletion does leave some CD11c⁺ cells present in the lung, inducing approximately an 80% decrease from 5.8% to 1% of total lung cells. Thus, it is possible that this depletion minimizes the initial number of naïve T cells activated, but that after repeated exposure to antigen, clonal expansion allows for the development of some of the features of the Th2 response, including the expansion of the eosinophil population, their recruitment to the airway, and the generation of antigen-specific immunoglobulins. Additionally, depleting CD11c⁺ cells during sensitization may alter rather than reduce the immune response, changing the outcomes associated with sensitization. However, we did not measure any qualitative alterations in the type of immune response provoked in the CD11c-DTR Tg⁺ mice.

It is important to consider other CD11c⁺ cell populations in addition to mDCs that are also affected by DT administration to Tg⁺ animals. Plasmacytoid DCs (pDCs) and alveolar macrophages also express CD11c, albeit to a lesser extent than mDCs [21]. Plasmacytoid DCs promote tolerance by decreasing the capability of myeloid DCs to induce T cell proliferation as well as promoting CD4⁺ T regulatory cell proliferation [26]. Thus, by depleting CD11c⁺ cells within the lung, depletion of both pro-inflammatory mDCs as well as anti-inflammatory pDCs occurs. Despite this caveat, CD11c-DTR mice have revealed the importance of CD11c⁺ cells in promoting allergic sensitization in other models.

Our current studies demonstrate that one mechanism by which NO₂ may act as an adjuvant is through the activation and recruitment of pulmonary CD11c⁺ DCs, a cell type believed to act as the bridge between innate and adaptive immunity [25]. We measured increased numbers of CD11c⁺CD11b^- cells and minimal changes in CD11c⁺CD11b⁺ cells in the lungs 2 hours following NO₂ exposure, all of which have the potential to be DCs. Other adjuvants have been shown to work similarly, including the well known adjuvant alum, which causes the recruitment of inflammatory DCs to the site of injection (peritoneum) and subsequently to the draining lymph nodes [22]. Additionally, the CD11c⁺CD11b^- population in the lung significantly upregulated expression of MHCII, CD40, and OX40L 2 hours post-NO₂, thereby enhancing its capability for antigen presentation and providing co-stimulation to promote Th2 responses. Myeloid DCs pulsed with ova and administered directly into mouse lungs also result in allergic sensitization to ova, indicating that these cells have the capability to induce T cell responses [24]. Furthermore, our data demonstrate that inhalation of 15 ppm NO₂ for just 1 hour is sufficient to induce the production of MCP-1 within the lung, a chemokine associated with DC recruitment [70]. IL-6 is also increased in the BAL fluid 2 hours following NO₂ exposure, which can induce a Th2 polarized environment [31].

From NO₂-exposed mice, we measured increased protein in the BAL, indicative of cellular damage that may promote the activation of epithelial cells capable of secreting factors such as IL-6 [71] and other pro-inflammatory mediators [25]. Agents that cause damage to the lung epithelium, such as the proteases produced by house dust mites, have been shown to promote allergic sensitization through allowing access of the antigens to resident antigen presenting cells and inducing cytokine and chemokine production from epithelial cells [63]. Previous findings have demonstrated that 10 ppm NO₂ induces the release of intracellular molecules such as HSP70 [6], which can act as an endogenous danger signal [72] and activate the transcription factor NF-κB, a key modulator in both innate and adaptive immune responses. Interestingly, NF-κB is capable of inducing IL-6 and MCP-1 gene expression [73] and is activated following NO₂ exposure [6], providing a potential mechanism by which NO₂ promotes CD11c⁺ pulmonary cell activation and Th2 polarization. Our in vitro co-culture data indicate than an effect of NO₂ solely on lung DCs is insufficient to explain the creation of a Th2/Th17-polarizing environment in the lung. In vivo, additional cell types, including epithelium, contribute to the effects of NO₂ inhalation on CD4⁺ T cell polarization. We have recently reported that NF-κB activation in the airway epithelium, as is also induced by NO₂ exposure [6,74], is sufficient to activate mediastinal lymph node DCs and to promote a mixed Th2/Th17 response to an innocuous inhaled antigen [75].

Allergic asthma is primarily dependent upon the activities of CD4⁺ T cells [2]. CD11c⁺ DCs regulate the initiation of naïve CD4⁺ T cell responses in multiple ways, including carrying antigens to the draining lymph node, processing and presenting these antigens, providing co-stimulation, and secreting cytokines [16]. In our studies, CD11c⁺ pulmonary cells secreted increased amounts of IL-1α, IL-1β, IL-12p70 and IL-6 ex vivo following NO₂ exposure, demonstrating that these cells alone produce pro-inflammatory mediators. When co-cultured with CD4⁺ OTII cells, NO₂-exposed pulmonary CD11c⁺ cells promoted the production of IL-5 from OTII cells more so than air-exposed controls, implicating that NO₂-exposed CD11c⁺ cells induced increased Th2 cell activity. Although we measured increases in Th1- (IL-12p70), Th2- (IL-6), and Th17-(IL-1) polarizing cytokines, it is important to remember that these co-cultures did not include epithelial cells. It is likely that epithelial cells also influence the outcome of DC and CD4⁺ T cell interactions, as epithelial cells are capable of secreting multiple mediators that affect T cell polarization, promoting the development of a Th2 immune response [25]. This may also be one of the reasons why we observed no differences in the secretion of IL-17 by the CD4⁺ T cells co-cultured with DCs from NO₂-exposed mice compared to those from the air-exposed mice.