IL-13 (50 ng/ml) was added starting day -2 (two days before ALI), and twice weekly for 21 days thereafter.
Growth protocol
Human tracheal epithelial cells (hTECs) were isolated and placed into culture in growth factor enriched medium as described previously. For the present experiments, cells were seeded at 2 x 105 cells per well (24-well Transwell, Corning, Corning, NY) and cultured under submerged conditions until confluence. Cultures were then maintained in DMEM-Ham’s F12 medium with 2% NuSerum (BD, Franklin, NJ), Primocin (100 µg/ml, InvivoGen, San Diego, CA), and retinoic acid (1 x 10–8 M, Sigma), and were switched to air-liquid interface conditions for 3 weeks. Media was changed twice weekly, cells were fed from the bottom reservoir.
Extracted molecule
total RNA
Extraction protocol
Total RNA was isolated from hTECs using the Qiagen RNEasy kit (Qiagen)
Label
biotin
Label protocol
RNA was amplified and biotinylated using the Ambion Illumina TotalPrep Kit.
Hybridization protocol
Hybridization was performed according to the manufacturer’s instructions at the Washington University School of Medicine Genome Sequencing Center Microarray Core Facility.
Scan protocol
Scanning, including background correction, was performed per manufacturer's instructions using BeadStudio 3.0 software (Illumina) at the Washington University School of Medicine Genome Sequencing Center Microarray Core Facility.
Description
d21 NT rep 4 NT_504_32
Data processing
Microarray normalization and statistical analysis was performed using packages from the Bioconductor project executed in the R programming environment. Raw image data were imported into Bioconductor using readIllumina as implemented in the beadarray package with background subtraction and image sharpening. The resulting bead-level data was then normalized for intensity at the bead level using the HULK algorithm, to adjust for local spatial effects (i.e. cross-array gradients). The data were then summarized for each bead type. A model-based variance stabilizing transformation was applied to the bead-type summaries, followed by quantile normalization across the experiment using functions in the beadarray package. Bead-types not detected on at least 1 array (detection p<0.01) were then filtered to improve power to detect differentially expressed genes. We assessed differential expression after 21 days of IL-13 treatment (IL-13 versus control) using linear models and empirical Bayes moderated F statistics as implemented in the LIMMA package. We considered differences in gene expression significant if P values were <0.05 after adjustment for multiple testing, so that false discovery rate was <5%. Bead-types were annotated to genes using a combination of the manufacturer’s annotation, transcript level annotations obtained from AceView, and annotations from the University of Cambridge Computational Biology Group in order to optimize interpretation of the gene expression data.
