Phenotypic alterations in type II alveolar epithelial cells in CD4⁺ T cell mediated lung inflammation

The epithelium constitutes the interface between the internal milieu and the external environment, and the respiratory epithelium is the initial point of contact for respiratory viruses, airborne allergens and environmental pollutants [1]. The major function of the respiratory epithelium was at one time felt to be primarily that of a physical barrier, but recent studies clearly indicate that its cells are metabolically very active with the capacity to modulate a variety of inflammatory processes through the action of an array of receptor-mediated events. Upon activation, epithelial cells have the capacity to produce a number of pro-inflammatory or regulatory mediators, including arachidonic acid products, nitric oxide, endothelin-1, transforming growth factor (TGF)-β, tumour necrosis factor (TNF)-α, and cytokines such as interleukin (IL)-1, IL-6 and IL-8 [2].

Alveolar type II epithelial cells (AEC II, granular pneumocyte, type II pneumocyte, giant corner cell) represent a highly specialized subpopulation of the respiratory epithelium. AEC II consist of about 15% of the distal lung cells and occupy 5% of the alveolar surface [3]. They perform a variety of important functions within the lung, including regulation of surfactant metabolism, ion transport and alveolar repair in response to injury [4‐7]. AEC II synthesize and secrete lung surfactant, a protein-lipid complex and surface-active material [8]. Ultrastructural criteria used to identify alveolar type II epithelial cells are the presence of lamellar bodies, apical microvilli and specific junctional proteins. AEC II also maintain the integrity of alveolar epithelium by proliferation (and differentiation to type I cells) in response to injury, and tightly regulate alveolar fluid by a variety of mechanisms.

AEC II express a number of molecules necessary for the transduction as well as the generation of signals involved in cell-cell as well as in cell-matrix interactions. Cell-cell interactions may be direct via contact of tight junction proteins, or indirect via secreted and diffusible signals [9]. Consequently, AEC II have been described as integrative units of the alveolus [10]. Interactions of AEC II with leukocytes have also been the subject of intense investigation and there is evidence supporting a role of AEC II in accessory function in T lymphocyte activation [11, 12]. Moreover AEC II chemokine expression is induced upon antigen-specific CD8⁺ T cell recognition and plays a critical role in the perpetuation of experimental interstitial pneumonia [13, 14].

In order to study the pathophysiology of chronic T cell-mediated lung injury, we established a novel model in which a model antigen (influenza A/PR8/34 HA) is expressed under the control of the SP-C promoter, resulting in AEC II cell-specific expression and bred these animals with mice expressing a transgenic T cell receptor, specific for a class II-restricted epitope of HA, leading to a chronic interstitial pneumonitis [15]. Initial characterization of these mice focussed on self-antigen specific T cell function and revealed the induction of peripheral T cell tolerance at the site of inflammation. In this study we demonstrate altered AEC II cell morphology in mice with CD4⁺ T cell-mediated pulmonary inflammation suggesting a state of activation that we wanted to explore at a molecular level. As such, we established a method to isolate highly pure primary AEC II for the purpose of performing ex vivo expression profiling in the context of acute and chronic interstitial pneumonitis. An important role of AEC II gene expression in the orchestration of inflammatory infiltration of the lung parenchyma is suggested by a wide array of inflammatory genes and chemoattractants expressed upon CD4⁺ T cell recognition of antigen presented by the AEC II cells, and this model may prove extremely useful in dissecting the mechanisms involved in the perpetuation of chronic autoimmune pulmonary processes.

A significant number of lung diseases are presumed to be T cell mediated based in part on the observation of T cell accumulation at sites of disease activity, particularly the interstitial lung diseases (ILD). The ILD represent a broad group of heterogeneous disorders and the participation of CD4⁺ T cells in various forms of ILD has been suggested. Sarcoidosis, idiopathic interstitial pneumonias, autoimmune connective tissue diseases and pulmonary hemorrhage syndromes represent some of the major categories of ILD. Sarcoidosis, for example, appears to be associated with an exaggerated cellular immune response to an unknown antigen and CD4⁺ Th1 lymphocytes are important effectors of pulmonary injury in this disease [20, 21]. In addition to ILD, it has been postulated that T cells are important contributors in other pulmonary disorders such as chronic obstructive pulmonary disease (COPD) and asthma [22, 23]. In these, it is hypothesized that cigarette smoke or allergen induced immune responses can, under certain conditions, progress to T cell mediated autoimmune disease. Recently it has been suggested that smoking-induced emphysema may represent an autoimmune disease of sorts, in which the presence of Th1 responses to a specific lung antigen correlates with emphysema severity [21]. Furthermore, oligoclonal CD4⁺ T cell expansion has been suggested to contribute to the pathogenesis of obliterative bronchiolitis [24]. Although there is growing evidence that CD4⁺ T cells contribute to various pulmonary disorders, little is known concerning the role of AEC II cells in T cell mediated lung injury. To expand our understanding of the roles of selected cell types in the induction and progression of inflammatory pulmonary processes, animal models represent tools of extraordinary value. To explore the contribution of AEC II gene expression in T cell mediated lung inflammation, we made use of a transgenic mouse model of chronic T cell mediated lung inflammation that mimics some of the features of the interstitial lung discussed above, and that was previously established [15]. We report here the application of flow cytometry to efficiently isolate alveolar type II epithelial cells from mouse lungs by negative selection followed by whole genome transcriptome analysis. Gene expression profiling has emerged as an important tool in the characterization of complex molecular responses in inflammation and disease. The use of isolated cellular subpopulations has proven to be more informative than whole tissues in dissecting the roles of individual cell types in disease development in general, and immune regulation in particular. Comparative genetic fingerprinting of AEC II isolated from healthy mice and mice suffering from severe lung inflammation promises to be extremely informative regarding the role of AEC II in the induction and regulation of pulmonary immunity and inflammation.

Though confirmation of protein expression is essential, morphological changes in AEC II phenotype and array data suggest very active participation of alveolar epithelial cells in inflammatory processes in the lung. Using Affymetrix GeneChip experiments we identified a heterogeneous set of more than 322 genes differentially expressed in AEC II under pathophysiologic conditions. Variations in signal intensities between experimental repetitions may account for slight differences in the disease progression in individual pooled mice as well as for differences in cRNA synthesis and hybridization efficiencies between two array experiments. To exclude as far as possible that changes in gene expression occur as a consequence of the isolation procedure, care was taken to purify AEC II from the different mouse pools strictly following the described protocol, i.e. avoiding variations of incubation times or temperature, etc. Therefore, the influence of cell isolation procedure on gene expression in AEC II cells from healthy versus inflamed lungs will subtract from each other and account for changes in the molecular signature of AEC II as a consequence of CD4⁺ T cell mediated lung inflammation.

The differential expression of several immune modulating molecules like TGF-β3 or the various chemokines and chemokine ligands observed, suggests that in an inflamed environment AEC II may interact with resident and mobile neighbour cells via secreted and diffusible signals [9]. Members of the transforming growth factor-beta family are linked to proliferation or secretory activities of AEC II. It has been shown that TGF-β3 production by AEC II is dynamically down-regulated during the proliferative phase of recovery from acute hyperoxic injury [25]. Consistent with this, TGF-β3 expression was down-regulated in AEC II from the inflamed lung, and since AEC II represent the stem cells for alveolar type I epithelial cells (AEC I), this suggests a role of the TGF-β family in AEC II proliferative responses and/or the cellular hypertrophy of AEC II observed in the inflamed lung.

In addition to TGF-β3, the CXC chemokines CXCL2, CXCL13 and CXCL12 were also differentially expressed in AEC II from inflamed compared to healthy lungs (Figure 4, 5, 6, Table 1). These chemokines praticipate in the process of attracting various cell populations into the lung. CXCL12 and CXCL13 bind to CXCR4 and CXCR5, which are primarily expressed on T lymphocytes or on circulating fibrocytes [26]. Interestingly, CXCL12 and CXCL13 expression was induced shortly after T cell recognition of epithelial antigen (Figure 5, 6 and data not shown) and massive lymphocytic infiltrates were observed shortly after T cell transfer (data not shown). Furthermore, down-regulation of T cell chemoattractants was evident at later stages of inflammation (Figure 4 and Table 1) and could contribute to a more controlled infiltration of specific T cells into the lung. Accordingly it has been shown that CXCL13 plays an important role in the development of inducible bronchus associated lymphoid tissue (iBALT) in respiratory immunity [27] by attracting T lymphocytes. It has been suggested that infection or inflammation triggers the organization of lymphoid structures in the lung of both mice and humans [28, 29], though this is somewhat controversial. These structures do not fit the classical definition of BALT, as they are not formed independently of antigen [30, 31]. Because the iBALT appears in the lung only after infection or inflammation, it is generally assumed that iBALT is simply an accumulation of effector cells that were initially primed in conventional lymphoid organs. The neo-formation of iBALT is caused by inflammatory responses which directly promote the recruitment, priming and expansion of antigen-specific lymphocytes [27]. It is interesting to speculate that AEC II in SPC-HA/TCR-HA double transgenic mice, after the initial inflammatory responses, down-regulate CXCL13 expression in order to counteract new formation of iBALT and infiltration of specific T cells.

The chemokine CXCL2 is involved in attraction of polymorphonuclear granulocytes to sites of infection [32]. These neutrophils play an important role as regulators of immune responses through release of cytokines such as IL-1, IL-3, IL-6, IL-12, tumor necrosis factor-α (TNF-α) or TGF-β as well as chemokines such as CCL2 (MCP-1) or CCL20 (MIP-3α) [33, 34].

Elevated expression of CCL20 by AEC II has been shown to attract other pro-inflammatory cells [34, 35]. CCL20, which was dramatically up-regulated in the inflamed lung (Figure 4, 5, 6, Table 1), has been shown to be constitutively produced by AEC II cells and can attract immature dendritic cells (imDC) to the lung [36, 37]. Immature dendritic cells are known to exert immune modulatory functions and may contribute to the establishment of a controlled immune response in SPC-HA/TCR-HA double transgenic mice. In contrast to CCL20, CCL11 (eotaxin), an eosinophil chemoattractant, was dramatically down-regulated in AEC II from the inflamed lung (Figure 4 and 6, Table 1). Not surprisingly, anti-CCL11 reduced eosinophils infiltration of the lungs of RSV-infected mice. In addition, however, anti-CCL11 also caused inhibited CD4-T-cell influx [38]. Together, these data indicate an active immune regulatory function of AEC II in inflammatory pneumonitis involving the expression and secretion of soluble mediators that may affect other immune cells with regulatory features which may amplify, or interfere with, inflammatory responses in the lung.

Although gene expression data provide evidence that AEC II may (either directly or indirectly) exhibit immune regulatory functions, we also identified genes involved in the induction of T cell mediated immunity. In this context it is interesting to note that the expression levels for molecules involved in antigen processing and presentation were up-regulated in AEC II obtained from diseased mice. For instance, increased expression of molecules needed for the MHC class-II restricted antigen presentation, like H2-Ea and H2-Ab1, but also invariant chain (CD74), was observed. Furthermore, expression of genes encoding for the transporter associated with antigen processing (TAP1) and various proteasomal subunits, all related with MHC class I presentation, were increased (Figure 4, 5, 6, Table 1). This effect was observed both in SPC-HA/TCR-HA mice that exhibit chronic inflammation as well as in AEC II from SPC-HA mice shortly after T cell transfer. Up-regulation of MHC encoded genes is likely the result of interferon (IFN)-γ production by the CD4⁺ T cells, and is well known to induce the transcription of genes encoded within the MHC region. Based on our previous observation in an adoptive transfer model for CD8⁺ T cell mediated pulmonary inflammation, as well as in cell culture experiments [13, 14, 39], we have strong evidence that T cell antigen recognition triggers inflammatory gene expression in AEC II cells, a significant portion of which is IFN-γ dependent. Although CD4⁺ T cell derived IL-2 and IFN-γ are likely pro-inflammatory mediators that trigger AEC II gene expression in SPC-HA/TCR-HA mice or SPC-HA mice after adoptive T cell transfer, it is possible that other T cell derived factors contribute to the observed changes in AEC II gene expression, such as TNF-α.

Further genes differentially expressed in AEC II upon airway inflammation are cyclin A2 and cyclin D2, both involved in cell cycle regulation [40, 41] and several matrix metalloproteinases (MMP) and tissue inhibitor metalloproteinases (TIMP), all of which are critical in repair and remodelling in response to injury [42, 43]. In addition to these, genes with roles in adhesion, cytoskeletal function, transport, metabolism, signal transduction, transcription and development suggest that AEC II are active participants in all aspects of immune regulation, inflammation and responses to injury. The impact of these gene products on the ethiopathogenesis of pulmonary inflammation remain to be elucidated in further detail.

We have developed a new AEC II isolation protocol based on flow cytometric negative selection for the isolation of cell populations of high purity and viability. Employing this technique, we determined the genome-wide profile of gene expression in response to T cell-mediated interstitial pneumonitis. Overall, these results provide a detailed description of AEC II gene expression under pathophysiologic, autoimmune conditions. Differentially expressed genes of diverse molecular functions have been identified that may be critical for numerous physiologic activities, some of which may be currently unappreciated. Data obtained by such analysis will help to understand the function of these important immune cells in the respiratory system and may point out strategies for intervention in the progression of chronic inflammatory processes in the lung.