Sputum cell pellets were processed using the All-in-One purification kit (Norgen Biotek, Thorhold, CAN), checked on an Agilent bioanalyzer and, if needed, treated again to remove DNA contamination (Qiagen, Gaithersberg, MD).
Growth protocol
Sputum induction was performed with hypertonic saline, as previously described. Simply, by using a microscope, mucus plugs (cellular parts) from the sputum sample were dissected, separated and collected from aqueous compartments. Processing of the collected cellular part was done by mixing with 1 X Sputolysin Reagent (Calbiochem, cat# 560000) at a ratio of 1:4 (weight:volume), followed by vortexings at high speed for 30 seconds and low speed for 15 min, adding equal volume of PBS and a cell strainer filtration. Total cell count, viability, and differential, were determined by hemocytometer, trypan blue exclusion, and Wright-Geimsa stain, respectively. After centrifuged at 3,500 x g for 5 min, cell pellets were lysed in lysis solution of All-in-One Purification Kit (Product# 24200, Norgen Biotek Co, Thorhold, CAN) and stored at -80C until use.
Extracted molecule
total RNA
Extraction protocol
10 ng of sputum RNA was amplified using the WT-Ovation Pico RNA amplification System (NuGen, San Carlos, CA) and processed per Affymetrix protocols. Total RNA from the blood was isolated using a column-based system (total-RNA kit, Norgen) and if needed, DNA contamination was removed using a DNA clear kit (Qiagen). Hemoglobin reduction of blood samples was used to remove hemoglobin gene transcripts (GLOBINclear Kit, Ambion, Austin, TX) and samples were checked by Agilent bioanalyzer.
Label
biotin
Label protocol
Purified RNA from the sputum or blood was processed for gene expression using the Affymetrix HuGene 1.0 ST gene arrays following manufacturer’s protocols.
Hybridization protocol
See manufacturer's protocol
Scan protocol
See manufacturer's protocol
Description
Gene expression from sputum
Data processing
RMA was used to background correct, quantile normalize and summarize the gene expression levels across all arrays. Blood and sputum arrays were processed separately. Principal component analysis (PCA) was used to visualize the data and remove outliers. The batch effect was adjusted using Combat and the RIN number was adjusted using linear regression model.
