|
|
||||||||
Ann Thorac Surg 1996;61:252-258
© 1996 The Society of Thoracic Surgeons
Department of Thoracic Surgery, University Hospital of Strasbourg, and Tissue Typing Laboratory, Centre Régional de Transfusion Sanguine, Strasbourg, France
| Abstract |
|---|
|
|
|---|
. Dendritic cells most probably play a major role in lung diseases such as histiocytosis, primary and secondary cancers, and both acute and chronic lung graft rejection. Identification of the precise functional pathways might lead to therapeutic use of modulation of dendritic cell function. | Introduction |
|---|
|
|
|---|
As an opening remark, we should point out that the ideas presented in this article have been selected according to the scientific credit given to the senior authors of the referenced publications. However, any experimental fact has been observed under very particular experimental conditions and should not be extrapolated to different specific conditions. On the other hand, a review article would not be readable if every experimental condition were to be reported. Because comparison of isolated experimental data is subjected to some confusion and misinterpretation, we suggest that the interested reader go back to the referenced article to complete his or her information.
| Common Characteristics of the Dendritic Cell Lineage |
|---|
|
|
|---|
Origin of Dendritic Cells
Dendritic cells have a common progenitor cell together with the monocyte-macrophage lineage, because both express the medullary antigens CD33 and CD34 [2, 3, 9]. Further, bone marrow cultures sampled from patients recovering from aplasia after chemotherapy grow both lineages spontaneously [10]. However, there is a lack of consensus whether the dendritic and monocyte-macrophage lineages are really distinct, or whether they represent two different functional states of the same cell lineage. Steinman [3] argues for a distinct dendritic lineage, because DCs do not express any macrophage-monocyte marker. Thus, it appeared with CD34+ precursor cells isolated from umbilical cord blood that differentiation might be directed toward the macrophage lineage with granulocyte-macrophage colony stimulating factor (GM-CSF), and toward the DC-lineage with tumor necrosis factor
[11, 12]. In contrast, the fact that DCs have been obtained from cultures of CD14+ cells, presumed to be purified monocytes, argues for a common lineage. Under specific in vitro conditions, monocytes shed their specific CD14 marker and increase the expression of major histocompatibility complex (MHC) class II molecules [13, 14]. The humoral mediators for this metamorphosis of monocytes to DCs are mainly interleukin-4 (IL-4) and IL-13. The effect of IL-4 is amplified with GM-CSF, and antagonized by interferon
[15]. Interleukin-13 further increases survival of monocytes in culture [15, 16].
Migration of Dendritic Cells
The ubiquitous distribution of DCs all over the organism requires migration patterns leading the DCs from their medullary origin via the blood or lymphatic stream to sites of antigen uptake and subsequently to lymphoid effector organs. The sites of antigen uptake are most likely epithelial surfaces such as skin, bronchi, alveoli, or bowel, whereas the effector organs are lymphatic organs containing T cells. Besides well-delimited lymphoid organs such as thymus, nodes, and spleen, the gut-associated lymphoid tissue and the bronchus-associated lymphoid tissue are possible sites of antigen presentation. As a matter of fact, experimental studies have identified two different migration pathways. A hematogenous migration directed toward the spleen has been illustrated in a murine model. Isotopically labeled splenic DCs have been injected intravenously: the cells first migrated to the lung, where a short sequestration of about 1 hour occurred; a secondary migration then happened toward the liver, and finally to the spleen. However, these intravenously injected cells were unable to penetrate lymphatic nodes, even after splenectomy [17]. Similarly, in a rat cardiac allograft model, donor DCs migrated to the spleen [18].
On the other hand, splenic DCs injected subcutaneously migrated toward the paracortical region of the draining lymph nodes [19, 20]. The latter example illustrates the pathway of migration generally accepted for cutaneous Langerhans cells: these cells at first internalize antigens in the skin, and subsequently migrate to the lymph nodes, where they present the antigen to T lymphocytes [19]. Comparably, splenic DCs injected intratracheally migrate to the mediastinal nodes, and intraperitonerally injected DCs migrate to the mesenteric nodes [20, 21].
Phenotype of Dendritic Cells and Its Changes During Culture
As one may expect with respect to the antigen-presenting function, DCs strongly express molecules involved with antigen presentation and T-cell activation on their cytoplasmic membrane: these are mainly the MHC class II molecules on the one hand, and a variety of adhesion molecules on the other hand. After contact with T cells, the density of adhesion molecules is even enhanced [22]. However, DCs consistently lack the monocyte marker CD14. The expression of immunoglobulin receptors is controversial. Controversy regarding particular aspects of the phenotype is not surprising because it appears that the phenotype may change during culture. For example, native, freshly isolated cutaneous Langerhans cells only moderately express MHC class II molecules, and are recognized owing to a high expression of CD1a; in addition, they are positive for the Fc receptor CD32. After some days in culture, the expressions of CD1a and CD32 disappear. The expression of both MHC class II and class I molecules is intensified, and various activation and adhesion molecules appear: IL-2 receptor (CD25), LFA-3 (CD58), and ICAM-1 (CD54) [23]. Comparable findings have been demonstrated with peripheral blood DCs [24], with the restriction that blood DCs do not express the CD1a molecule. The explanation for this maturation is probably that the function of native DCs is to internalize and process antigens toward a presentable state; as a matter of fact, native cells are weak stimulators in a mixed lymphocyte reaction. While antigen processing is achieved, the phenotype matures for effective T-cell activation. The main mediator of this maturation seems to be GM-CSF, which prolongs viability of Langerhans cells or peripheral blood DCs in vitro and amplifies their accessory activity [25]. Tumor necrosis factor
, on the other hand, only prolongs viability [26, 27]. Lymphatic DCs of the rat mature by a combined activity of GM-CSF and IL-1ß [28].
Activation of T Cells
Dendritic cells are capable of inducing an immune response in naive T lymphocytes, which have never been in contact with a given antigen. The T-cell subtype in immediate contact with DCs are the T helper cells, recognized by the CD4 molecule, although the cytotoxic T cells or CD8 might be directly activated by DCs. According to the precise function of CD4 cells, we distinguish two subtypes: the Th1 cell, which induces type IV cellular immune reactions, and the Th2 cell, which initiates humoral immune response [29]. The common pathway of antigen presentation is the MHC molecule. If a foreign MHC class II is directly recognized, ie, if a foreign DC is in contact with a self T cell, one speaks about direct presentation. Indirect presentation occurs when a self DC presents with its MHC class II a foreign peptide to self T cells. Both mechanisms may lead to acute allograft rejection [30]. The contact between APC and T cell is established through adhesion of the MHC class II molecule and the T-cell receptor complex CD3, completed by the coreceptor CD4. However, this single contact between MHC class II molecule and T-cell receptor is not sufficient to trigger an immune response, and a second signal is required. Although it was formerly believed that this second signal was mediated by cytokines such as IL-1 or IL-6 secreted at the time of contact [31], there is increasing evidence that this signal is initiated by the linkage of various intercellular adhesion molecules. The most prominent couples that have been identified to date are summarized in Table 1
[32].
|
Lack of the second signal leads to absence of immune response, and probably to tolerance. Induction of allograft tolerance through blockade of the CD48-CD2 couple has been achieved in a murine heart transplant model [40].
| Particular Implications Within the Lung |
|---|
|
|
|---|
Histologic examination provides information on the relations of DCs with neighbor cells within the general architecture of bronchial or pulmonary tissue. Previous investigations have shown DCs within the bronchi, peripheral lung, and pleura in histologic preparations. In these experiments, DCs were recognized through revealing of the S-100 protein, which is also contained in neuroglial cells [41]. More recent studies have relied on immunohistochemical identification of the HLA-DR molecule [7]. Fresh DCs may also be identified by staining of the cytoplasmic CD68 content; this molecule is spread all over the cytoplasm in macrophages, and forms typical perinuclear spots in DCs [42]. Frozen tissues and cytospin preparations may further be stained with anti-CD1a and anti-CD1c antibodies [43]. Observation of bronchial DCs should ideally be made on tangential sections [44].
Both analysis of the phenotype and function studies require previous preparation of a cell suspension. This is achieved with enzymatic digestion of peripheral lung tissue, and subsequent cell separations [4547]. Deletion of T lymphocytes, macrophages, B lymphocytes, and natural killer cells may be obtained either on physical grounds (adhesion for macrophages, sheep red cell rosetting for T lymphocytes, density gradient for B lymphocytes) or in a more sophisticated manner with cell sorting on a flow cytometer. Because no specific marker for DCs is available, a negative selection has to be made: known leukocytes (eg, T lymphocytes, macrophages) are labeled, and unlabeled cells are selected [24]. Cells with a low autofluorescence might be primarily DCs [48, 49].
The phenotype may be detailed with monoclonal antibody labeling either on cytospin preparations or with flow cytometry. Flow cytometry, however, requires a large number of cells (at least 250,000 per antibody), and retrieval of cells after several manipulations results quite often in a small number of cells.
Function of DCs is assessed in a mixed lymphocyte reaction. There are two types of reaction that are currently used. Direct antigen presentation is tested in an allogenic model, where stimulating DCs are cultured with allogenic responding peripheral blood mononuclear cells or peripheral blood lymphocytes. Indirect presentation is simulated in an autogenic model, where stimulating DCs present a soluble antigen (eg, ovalbumin) to presensitized autogenic lymphocytes [48, 49].
Origin of Pulmonary Dendritic Cells
The immunologic incompetence of the bronchial mucosa in newborns is correlated to an absence of APC in the fetal or newborn bronchial mucosa. In rodents, a progressive colonization of the tracheobronchial mucosa with DCs occurring during the time span between birth and weaning has been documented [50]. There seems to be a precursor cell that does not express the MHC class II molecules. These cells acquire the class II expression progressively during the neonatal period. At the same time, antigen-presenting cells appear in the alveolar tissue. This colonization is accelerated by interferon
and antagonized by aerosolized steroids [50].
Histology
Pulmonary DCs are mainly located in the airway mucosa and in the alveolar septa. Airway DCs are located either in close contact to the basal membrane with their digitations floating in the lumen or in the submucosa close to serous acini [7, 51]. The morphologic features of bronchial DCs are similar in mice, rats, and humans. Distribution of these cells is variable according to the level of the bronchial tree, and decreases considerably below the third generation of bronchi: the concentration is greater than 600 cells/mm2 at the tracheal level and less than 300 cells/mm2 below the third generation of bronchi [44]. When local inflammation occurs, the number of DCs increases, and the number and size of dendrites increase as well; the phenotype changes toward an increased expression of adhesion molecules (in particular CD11a:CD18) [44]. Alveolar DCs are mainly located in the alveolar septa, close to the areas of septal fusion. A possible interaction between alveolar DCs and alveolar macrophages is suggested by a close anatomic relationship, because the alveolar macrophages are tagged to the alveolar wall close to the fusion line of alveolar septa [52]. Alveolar DCs are not retrievable with bronchoalveolar lavage in normal animals or humans [47]. However, as soon as 4 hours after induction of an inflammatory response with intratracheal Calmette-Guérin bacillus in rats, DCs appear in the lavage [53].
The bronchial and alveolar DCs are not identical in their phenotype. In the rat, alveolar cells are positive for the ED2 marker and are negative for ED1, in contrast to bronchial cells, which are positive for ED1 and negative for ED2 [8]. A similar dichotomy between bronchial and alveolar DCs has been reported in humans [43]. Comparison of function shows that rat bronchial DCs are more efficient at presenting antigens to sensitized T cells, whereas alveolar cells are more potent stimulators for naive T lymphocytes [54]. The former is related to asthma and bronchial hypersensitivity, whereas the latter fact might be involved with acute lung allograft rejection.
Dendritic Cell Function and Modulation by Alveolar Macrophages
Although the methodology employed for retrieval of lung DCs varies from one author to another, results converge to show a dramatic stimulatory activity in a mixed lymphocyte reaction, which is consistently superior to the stimulation observed with macrophages or B lymphocytes [47, 49]. Macrophages are usually weak accessory cells, but their capacities to present antigens are enhanced in pathologic circumstances such as sarcoidosis [55]. The close anatomic relationship between DCs and macrophages noticed on histologic studies suggests a functional interaction. As a matter of fact, presentation of the ovalbumin antigen by autogenic DCs to sensitized T lymphocytes is inhibited in the presence of macrophages [56]. In contrast, depletion of macrophages in a living animal with aerosolized dichloromethylen-diphosphonate allows isolation of DCs with a fivefold increased antigen-presenting activity. Inhibition of the antigen-presenting function is maintained in a co-culture model, where DCs are separated from macrophages with a semipermeable membrane. This observation suggests a humoral modulation by various cytokines, and mainly by tumor necrosis factor
[52]. Besides macrophages, alveolar pneumocytes may down-regulate the immune response [57].
Dendritic Cells and Lung Diseases
Dendritic cells contained within the normal lung remain different from the cutaneous Langerhans cell. Although they may at times express the CD1a marker, they never contain Birbeck granules. However, histiocytosis X is characterized by large infiltrations with typical Langerhans cells. These cells further express the CD1c marker associated with antigen uptake, which is lacking in the normal cutaneous Langerhans cell [58]. The origin of histiocytosis X is unknown, but it appears that histiocytosis X occurs exclusively in smokers. Several investigators followed the hypothesis that smoking might induce proliferation of Langerhans cells within the lung. Bronchoalveolar lavages were compared in smokers and in nonsmokers. Bronchoalveolar lavage cytology of smokers displayed 1.1% CD1a positive cells, versus 0.1% in nonsmokers; Birbeck granules were found in 0.4% of the cells in smokers and were constantly absent in nonsmokers [59]. Squamous metaplasia induced by tobacco abuse might create an epithelial environment leading to differentiation of precursor cells in the lung or migration originating from the peripheral blood. There also might be a proliferation of exceptional preexisting Langerhans cells. Histologic studies of lung samples collected during lung resections provided similar data. Normal lung contains CD1a-positive cells within the bronchial epithelium and CD1c-positive cells within the alveolar tissue. In smokers, both the absolute number of cells and the number of CD1a cells within the alveolar tissue increase [43]. The humoral mediator of this CD1a differentiation might be GM-CSF, because this cytokine is released by hyperplastic alveolar pneumocytes [60]. Pulmonary infiltrates with DCs have been described in various other chronic lung diseases: idiopathic interstitial fibrosis, bronchiolitis obliterans with organizing pneumonia, desquamative interstitial pneumonia, and allergic alveolitis [61, 62].
The most stimulating point of interest for future research is that DCs have been observed in tumoral and peritumoral tissues. In particular, such infiltrates have been observed together with adenocarcinoma, and more specifically with bronchioloalveolar carcinoma. Correlation of DC infiltrates has been less consistent with other histologies. The trigger seems to be a local production of GM-CSF by pneumocytes or tumoral cells, as ascertained with in situ hybridization studies [60]. The significance of this infiltrate is unclear. Do tumors develop because tumor antigen presentation is deficient, or are these infiltrates on the contrary a sign of an efficient host response? A study by Furukawa and colleagues [63] demonstrated a positive prognostic value for infiltrates with DCs. In a population of 40 patients with stage I adenocarcinoma, 77.5% were found to have infiltrates with DCs; their 5-year survival rate was 86.4%, which was significantly better than the 38.9% observed in patients without DC infiltrates (p < 0.005;
2 = 4.24) [63]. Finally, are tumor-associated DCs the same variety as those observed with histiocytosis? In contrast to histiocytosis Langerhans cells, tumor-associated DCs did not express the MHC class II molecule DP, the Fc
receptor CD16, the IL-2 receptor CD25, and various adhesion molecules such as CD11b, CD11c, and CD54. Besides, production of IL-1 and interferon
by tumor-associated DCs was significantly lower. The biologic significance of these differences remains unclear [64].
The major challenge in clinical immunology is the comprehension of the mechanisms that determine chronic rejection. Griffith and colleagues [65] demonstrated in 1989 that chronic irreversible graft dysfunction will develop in 30% to 40% of lung transplant recipients due to bronchiolitis obliterans; this syndrome is at least in part explained by immunologic mechanisms. In an autopsy study of lung transplant recipients who died with bronchiolitis, bronchial mucosa specimens submitted to S-100 protein staining disclosed numerous submucosal infiltrates with DCs. The latter were significantly increased when compared with lung transplant recipients who died without bronchiolitis obliterans, with recipients of an extrathoracic graft, and finally with patients who died free of any respiratory disease [66]. The DC infiltrates seem to originate from the recipient, as has been demonstrated in a rat lung transplant model. Migration of recipient DCs is observed starting at the first postoperative week; at 2 months, DCs have been progressively replaced by recipient-derived DCs. After the second month, these cells form clusters, and finally disappear when histologic modifications of chronic rejection have occurred [67].
| Conclusion |
|---|
|
|
|---|
| Footnotes |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
cooperate in the generation of dendritic Langerhans cells. Nature 1992;360:25861.[Medline]
or IL-10. J Immunol 1993;151:637081.[Abstract]
maintains the viability of murine epidermal Langerhans cells in culture, but in contrast to granulocyte/macrophage colony-stimulating factor, without inducing their functional maturation. J Exp Med 1990;171:15971.This article has been cited by other articles:
![]() |
K. R. Cooke, W. Krenger, G. Hill, T. R. Martin, L. Kobzik, J. Brewer, R. Simmons, J. M. Crawford, M. R.M. van den Brink, and J. L.M. Ferrara Host Reactive Donor T Cells Are Associated With Lung Injury After Experimental Allogeneic Bone Marrow Transplantation Blood, October 1, 1998; 92(7): 2571 - 2580. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |