Cells were harvested and washed twice in PBS. Then, 2×105 cells were incubated with indicated labelled antibody for 60 min at 4°C. After washing twice with PBS/Gelafusal (Serumwerke Bernburg, Germany)/sodium-acid, antibody binding was analysed by flow cytometry (FC 500, Beckman Coulter). Cryostat sections were incubated with the antibodies indicated. Positive cells were identified BGJ398 cost by biotinylated goat anti-rat IgG and the avidin–biotin complex technique according to the manufacturer’s protocol (supersensitive multilink alkaline phosphatase ready-to-use detection system, Biogenix, San Ramon, CA). The colour reaction of New Fuchsin

substrate (DAKO, Hamburg, Germany) was used for detection of bound proteins. In control sections, primary antibodies were replaced with an isotype control antibody. Tissue sections were photographed using a DP70 CCD camera mounted on a BX41 light microscope (Olympus; Hamburg, Germany). Histological section were stained by H&E, photographed, and thickness of infiltrate was calculated using BZ-9000E analyzer software (Keyence BZ-9000E; Keyence, Neu-Isenburg; Germany). MMP-9 in the BAL and peritoneal

fluid was measured by ELISA (R&D, Wiesbaden, Germany). A set of 48 cytokines/chemokines was detected by a membrane-based cytokine array according manufacture’s protocol (RayBiotech, Norcross GA, USA). We used pooled BAL from two WT or two Thy-1−/− mice, respectively. The experiment was repeated with the BAL of a third mouse of each group. In summary, the array results represent the chemokine/cytokine profile MAPK Inhibitor Library cost of the BAL of three different WT and Thy-1−/− mice, respectively. The densitometric data were adjusted

to negative see more and positive controls on the same membrane. Every chemokine/cytokine was detected by two different spots. The mean of the densitometric signal was used for evaluation. To identify differences in the amount of chemokine/cytokine the quotient of the signal from the BAL of WT mice and Thy-1−/− mice from each membrane hybridization was calculated. To get robust data, an increase of the signal was only accepted when the signal was enhanced over 25% (quotient >1.25) in both hybridizations. Human eosinophils were prepared from granulocytes upon Ficoll-density-gradient centrifugation of whole EDTA blood by depletion of CD16-positive neutrophils by magnetic separation according to manufacturer’s protocol. Efficiency of separation was examined by anti-CD16 staining and flow cytometric analysis. Human monocytes were separated from blood of healthy volunteers by magnetic cell separation using anti-CD14-beads (Miltenyi Biotec) as described previously 39. Total RNA was isolated from human eosinophils or monocytes with the RNeasy Mini Kit (Qiagen, Hilden, Germany) and 0.