The present study describes for the first time the influence of SP-D on cellular uptake of allergen particles such as SPP within a complex lung cell culture model and the secretion of cytokines/chemokines. Our data show that SP-D increased the number of MDM and EC, which participated in allergen particle uptake. Interestingly, the number of SPP per single cell did not increase upon incubation with increasing SP-D concentration. Using an uptake index (UI) that considered both, the percentage of cells containing particles as well as the number of SPP inside single cells, we demonstrated that the total number of SPP stayed constant in MDM, was increased in EC, and was decreased in MDDC. Incubation with SPP increased secretion of (pro-)inflammatory cytokines and chemokines. SP-D inhibited the IL-8 release from the cells.
Little is known about the uptake of naturally occurring allergen particles by resident lung cells. In addition, although SP-D is one of the first proteins which comes into contact with inhaled particles and thereby modulates immune responses , only few data exist describing the effects of a SP-D-particle interaction mainly leading to a increased percentage of alveolar macrophages which took up allergen particles [6, 11]. However, the same study also showed that SP-D reduced the number of SPP-positive tissue macrophages and DCs 24 hours after in vivo instillation of the particles . These results highlight the potential differences between effects observed in vitro and in vivo and further emphasize the need of complex in vitro systems which may much better mimic specific in vivo situations. In fact, our results show a decreased UI for the MDDC which would be in line with the results of the study of Winkler and co-workers considering a substantial amount of particle clearance during the longer observation period they used . Within the present study, it is important to note, that the measurement of particle quantification was only performed by a single observer. Although the software Diacount® does hardly allow for a subjective counting, this should be taken into consideration. Importantly, we were able to reliably determine SPP uptake in contrast to binding and focused only on the intracellular SPP. Recently we showed that SP-D increased the amount of human primary bronchial epithelial cells with attached SPP . It is further known that the uptake of particle-like surfactant-aggregates in type II pneumocytes is increased upon co-incubation with SP-D . In accordance, we found an increased amount of EC which took up SPP and also an increased UI upon incubation with SP-D. Importantly, in a previous study , we found that SP-D was able to modulate the interaction of SPP with human primary bronchial epithelial cells but not with A549 cells. Interestingly, the A549 cells, incorporated in the present epithelial airway model, reacted much more sensitive upon contact with SP-D and SPP compared to the A549 in vitro monocultures . Hence we may speculate that contact to other lung cells such as immune cells modulates the A549 cells to react more similar than human primary epithelial cells.
Incubation with 10 million SPP led to an increased (pro-)inflammatory response within the epithelial airway model after further 72 hours incubation with fresh medium compared to cultures without SPP. This clearly indicates an inflammatory effect of SPP that may also occur in the lung after a single inhalation of allergen particles and that may be independent of allergic sensitization to the particles. The increased secretion of several cytokines (e.g. IL-8) determines a pro-inflammatory action of the SPP which was previously observed for various other allergens in different experimental systems [26–30]. Interestingly, some of these mediators are even released by cells upon contact to inert particles. It is e.g. known, that polystyrene particles induce the secretion of TNF-alpha  or IL-8 . Furthermore, SPP are a natural material and are consequently contaminated by LPS. This contamination is of course at least partly responsible for the secretion of inflammatory cytokines. However, it was the aim of the study to investigate the allergen particles as they occur in the nature and not only by means of a purified laboratory situation. Hence we conclude that the secretion of (pro-)inflammatory mediators belongs to all three parameters: the nature of the allergen itself, the particulate body of the allergen particles, and possible LPS contamination.
Importantly, an allogenic response within the co-culture model can be excluded. With respect to IL-8, we observed cytokine secretion after measuring just the cell culture system without any stimulus. However, after incubating the cell-culture system with SPP, we found a significant increase of IL-8. Hence we conclude that this increase is not the result of an allogenic stimulation. The addition of SP-D decreased this SPP-modulated increase of IL-8 which further supports the notion of being no allogenic effect.
SP-D led to a significantly decreased level of IL-8 in the supernatant in the cell cultures after incubation. This is in contrast to a former study with primary bronchial epithelial cells in monocultures. It has been observed that epithelial mono-cultures and triple cell co-cultures react differently upon exposed to various particles e.g. diesel exhaust particles, titanium dioxide as well as single-wall carbon nanotubes [33, 34]. We hypothesize that there is a synergistic effect due to the interaction of the three cell types (EC, MDM and MDDC) that reduce the adverse effects of the xenobiotica. It is important to note that different cell types, included in a complex cell culture model, are better at simulating the real situation in the lung than mono-cultures. Hence we believe, that the present results simulate more realistic the in vivo situation. In a former study performed with fixed co-cultures of MDM placed on top and MDDC placed below a monolayer of EC and exposed to 1 μm polystyrene particles, we were able to visualize frequent interactions between these two cell types. MDM and MDDC were found to extend cytoplasmic processes across the epithelial barrier building cell-cell contacts, and particles were found in MDM and MDDC, some of them near cell-cell contacts . Another study revealed the expression of tight junction and adherens junction protein in MDM and MDDC which was suggested to preserve the epithelial integrity in a trans-epithelial network maintained by both immune cells in order to capture and translocate inhaled particulate antigen through the epithelial lung barrier . These ways of communication have to be investigated in more detail in further studies in order to understand cellular interactions in cell culture models.
Importantly, the cell cultures within these experiments were observed under a light microscope after the end of each experiment. Thereby, we could clearly see the normal cell shape with no signs of apoptotic blebbing or disturbance in the confluent A549 cell layer. In addition, our positive control Concanvalin A led to a secretion of most of the measured cytokines and chemokines. From dead cells, the secretion would not be so pronounced.
T-cell-dependent mediators were measured 72 hours after adding of the CD4+-T-cells, too (see additional file 1, table S1). Although our positive control (Concancavalin A) induced a release of most of the mediators (see additional file 1, table S1), we were not able to measure a significant release after incubation with SPP. For us, the most likely explanation of the low concentrations is the low proportion of MDDC to allergen (Phleumpratense)-specific T-cells. However, it is important to note that up till now we were only able to cultivate the T-cells within the co-culture model for 72 hours in order to guarantee viability of all cells. Crucial parameters of the immunological responses within the body normally occur over longer time periods. This limitation has to be considered. Thereby, the epithelial airway model should be further optimized so that a standardized use of the four different cell types is warranted and T-cell-dependent mediators and behaviour can be evaluated.