DCs are pivotal antigen presenting cells and the prime suspects for initiating granuloma formation and T cell alveolitis characteristic for pulmonary sarcoidosis. Paradoxically, local DCs have been suggested to be phenotypically and functionally immature [25
]. In contrast, we provide in this report the involvement of mature, functional DCs in pulmonary sarcoidosis. First, pulmonary sarcoidosis patients have increased numbers of mDCs in BAL, granuloma containing mucosal biopsies and blood. Second, mDCs in granuloma-containing biopsies show increased expression of a maturation marker. Third, mDCs from BAL are very well capable of inducing T cell proliferation and differentiation and show no signs of anergy. Finally, mo-DCs from sarcoidosis patients induce more TNFα in co-cultures with allogeneic CD4+
T cells, compared to mo-DCs from healthy controls. Taken together these results indicate that pulmonary sarcoidosis is associated with increased numbers of mature, functionally competent DCs that intrinsically induce increased levels of a central mediator in sarcoidosis, TNFα.
Previous studies and ours report different results in numbers of mDCs in blood or BAL from sarcoidosis patients [21
]. These differences may reflect variations in obtaining (e.g. volume of total BAL fluid), isolating (enriched cell populations vs. flow cytometric sorting), calculating (number per ml vs. proportions) and markers used for defining DC subsets. E.g we found increased proportions and numbers of mDCs in blood from sarcoidosis patients compared to healthy controls by calculating mDCs as the proportion of PBMCs from the same sample, while others reported (similar) numbers of mDCs as a proportion of total leucocytes, obtained by lysis of red blood cells, and calculated from a concurrent drawn blood sample [23
], or even decreased numbers of mDCs calculated by using proportions of a DC-enriched cell population [21
]. We performed BAL with 200 ml saline, compared to e.g. 100 ml in other reports [24
]. It is conceivable that an increase in BAL fluid amount reduces the relative contribution of the larger conducting airways to the retrieved amount of fluid, thus increasing the cellular yield from the alveolar compartment, increasing the yield of DCs.
To the best of our knowledge we report for the first time functional tests on mDCs from the site of disease in pulmonary sarcoidosis. DCs isolated from BAL from sarcoidosis patients did not induce enhanced T cell proliferation, a skewed T cell differentiation or significant differences in the induction of several cytokines in a MLR with allogeneic naïve CD4+ T cells, compared to controls. Several explanations related to the micro-environment of the lung where DC-T cell interactions take place, are possible: First, DCs from BAL from sarcoidosis patients may have already interacted with T cells, and are beyond their functional maximum upon isolation from BAL. This is supported by studies showing that after initial exposure to a stimulus, DCs produce IL-12, important for inducing T cell proliferation, for a limited period of 10-18 hours [36
]. Second, it is perhaps the sheer number of DCs and not the maturation status that determines the in vivo
outcome of T cell proliferation in sarcoidosis. Although the number of DCs in BAL is in general very low, DCs in the BAL are thought to reflect only a small percentage of pulmonary DCs [37
]. We found increased numbers of mDCs per ml of BAL in sarcoidosis patients, perhaps indicating increased numbers of interstitial DCs that may travel to the draining lymph node for antigen presentation. Future investigations in the lymph node compartment may shed light on this issue. Third, perhaps an intrinsic T cell factor is (additionally) required to induce the exaggerated T cell response observed in sarcoidosis. Interestingly, a single nucleotide polymorphism (SNP) in the IL-23 receptor was recently associated with sarcoidosis [38
]. IL-23 is a cytokine that is essential for the induction of IL-17 producing CD4+
T cells (Th17 cells) that were recently associated with granuloma formation in sarcoidosis by us and others [38
]. Interestingly, we did observe a modest but not significant increase in IL-17 production in the MLR supernatant of sarcoidosis patients. Finally, influx of CD4+
T cells into the bronchoalveolar space may primarily be determined by chemotactic factors. Our data on enhanced levels of MIG and IP10 in BAL, well-known chemotactic factors for Th1 cells, support this notion and confirm earlier reports [32
TNFα is a pivotal mediator of granuloma formation and maintenance, and is thought to play an important role in sarcoidosis pathogenesis [41
]. Indeed, enhanced TNFα secretion by BAL macro-phages is observed in sarcoidosis [3
]. In addition, TNFα is also expressed by Th1 and Th17 cells and both T helper cell subsets are likely involved in sarcoidosis pathogenesis [38
]. Polymorphisms in the TNFα
locus were associated with sarcoidosis phenotype and prognosis and have been linked to altered TNFα expression [42
]. Importantly, TNFα is an essential target for treatment [44
] and we found that mo-DCs from patients with sarcoidosis, not influenced by the micro-environment of the lung, induced increased TNFα release upon interaction with naïve CD4+
T cells when compared to controls. Our results indicate that DCs are intrinsically different in sarcoidosis patients. Nevertheless, we found that mo-DCs from sarcoidosis patients were equally capable of inducing proliferation and differentiation in allogeneic naïve T cells, compared to healthy controls. In the current study we did not explore whether mo-DCs from sarcoidosis patients lead to skewing of T cell differentiation towards a Th1/Th17 phenotype compared to controls, and this remains to be investigated. In co-cultures with PBMCs, mo-DCs were previously reported to demonstrate a reduced capacity to induce T cell proliferation (26). However, it is conceivable that other cells in the PBMC fraction, e.g. regulatory T cells (Tregs), influenced T cell proliferation. Indeed, we found increase numbers of Tregs in the PBMC fraction from sarcoidosis patients' blood (unpublished observations).
Granulomas are thought to arise upon failure of the immune system to clear invading pathogens. Earlier reports on diminished DC maturation and immune reactivity led to the notion that DC anergy may contribute to granuloma formation in sarcoidosis, e.g. due to diminished antigen presenting capabilities or ineffective induction of a T helper cell response [25
]. However, we found that both mDCs in BAL and in granuloma-containing airway biopsies were increased in number and had enhanced expression of maturation marker CD86 in the vicinity of granulomas. Furthermore, mDCs isolated from BAL displayed normal immune reactivity compared to healthy controls. Intriguingly, also in non-granulomatous mucosal airway biopsies from sarcoidosis patients there was a tendency towards increased numbers and maturation of mDCs, compared to healthy controls, although this was not statistically significant. It is tempting to speculate that these mucosal DCs are activated upon acquiring antigen from the airway lumen and subsequently present the antigen in a draining lymph node or are involved in mucosal granuloma formation. Taken together our results strongly support the notion that mDCs are involved in granuloma formation and maintenance in sarcoidosis, rather than the alternative that DCs are defective in supporting the adaptive immune system in antigen clearance. Our study included patients with newly diagnosed stage I/II sarcoidosis, we did not include patients with stage III/IV disease, and our results are therefore limited to patients presenting with early disease. We did not find a correlation between stage I or II and number of mDCs or pDCs. However, chest X-ray stage often poorly correlates with disease activity or progression. Thus, the role of DCs in granuloma formation or maintenance in subjects with advanced disease remains to be determined.
In conclusion, we provide evidence for the involvement of DCs in antigen presentation and granuloma formation at the site of disease in pulmonary sarcoidosis patients. Intrinsic genetic alterations in key APCs may underlie the exaggerated immune response to a hard to discern antigen that is characteristic of sarcoidosis. Immunological measurements and functional examination of DC and T cell subsets from large groups of carefully genotyped patients should prove very interesting.