Histologically normal adjacent larynx tissue samples. only the epithelial component was used for RNA isolation
Treatment protocol
Larynx tissue samples were obtained at the time of surgery and were immediately snap-frozen in liquid nitrogen. Prior to RNA extraction, samples were re-examined, and Giemsa-stained micro-sections obtained from each side of the frozen block were used to delimit the spatial distribution of the tumor mass or non-tumor tissue. Microdissections were performed to ensure that more than 70% of the isolated RNA was derived from cancer cells. In the case of surgical margins, only the epithelial tissue was used. Microdissected tumor samples and non-tumor samples were returned to liquid nitrogen until use.
Extracted molecule
total RNA
Extraction protocol
Snap-frozen tissue samples were pulverized using a mortar pestle. Next, total RNA was isolated from tissue specimens and Hep2 cell culture using TRIzol reagent (Gibco BRL Life Technologies). Total RNA was precipitated by incubating with 0.5 mL of isopropyl alcohol for 10 C. Themin, followed by centrifugation at 12,000 xg for 10 min at 4 pellet was washed with 75% ethanol, solved in RNAse-free water, passed through an RNeasy spin column (Qiagen) for purification and stored at C until further use.-80 RNA quantity and quality were evaluated using spectrophotometer, 1% agarose gel electrophoresis and also microelectrophoresis on a Bioanalyzer 2100 (Agilent Technologies). RNA from all samples was of appropriately high quality for cDNA microarray analysis.
Label
NHS-ester Cy5
Label protocol
Labeled targets for hybridizations were generated from total mRNA in reverse transcription reactions using oligo-dT primers (CyScribe First-Strand labeling kit - Amersham Biosciences). 15 ug of total RNA from each sample were mixed with 4 uL anchored oligo (dT) C(500µg/mL), in a total volume of 11 uL, subsequently denatured at 70 for 5 min, put on ice for 30 seconds, spun down and placed at room temperature for 10 minutes. Next, 1 uL of dNTP Mix, 1 µL of aminoallyl-dUTP, 2µL of 0.1 M DTT, 1µL of RNaseOUT (40U/µL), 4µL of 5x CyScript buffer and 1µL of CyScript reverse transcriptase (200U/µL) L with water. Afterwere added. The volume was adjusted to 20 incubation for 3 h at 42ºC, RNA was hydrolyzed by adding 2 µL of 2.5N NaOH for 15 min at 37ºC. Samples were then neutralized with 10 µL of 2 M HEPES Free Acid, and reactions were purified using 96-well Millipore Multiscreen filter plates as follows: 5 volumes of 5.3 M Guanidine-HCl: 150 mM KOAc were added to labeling reactions. The mixture was applied onto the plate and washed 4x with 80% EtOH by centrifugation at 3500 rpm for 5 min. Residual ethanol was spun out by an additional centrifugation at 3,500 rpm for 5 minutes. Labeled targets were eluted in 50 µL 10 mM Tris pH 8.5, by spinning at 3,000 rpm for 5 minutes, dried on a SpeedVac and kept at -20°C, protected from light until use. Then, cDNA was ressuspended in 40 µL of 0.1 M NaHCO3 and reacted with monoreactive NHS-ester Cy5 dye. The reaction was incubated for 2 h at room temperature and purified.
Hybridization protocol
Labeled targets were ressuspended in 250 µl of 1x hybridization buffer [25% formamide, 12.5% of proprietary Microarray Hybridization Buffer Version 2 (Amersham Biosciences)] denatured for 2 min at 92ºC and centrifuged at 13,000 rpm for 5 min. The Cy5 (tumor or non-tumor sample) labeled cDNAs were hybridized with microarrays on an Automated Slide Processor (ASP, GE Healthcare) and incubated for 16 h at 42ºC. Following hybridization, slides were washed (1.0× SSC, 0.2% SDS 10 min at 55°C; 0.1× SSC, 0.2% SDS 10 min at 55°C; 0.1× SSC, 0.2% SDS 10 min at 55°C; 0.1× SSC 1 min at RT; 0.1× SSC 1 min at RT; dH2O 10 sec. at RT), and dried with a N2 stream.
Scan protocol
Processed slides were scanned with a 700 V PMT setting (GenIII Scanner – Amersham Biosciences).
Description
Experiment described in detail in the previous sections.
Data processing
Background-subtracted artifact-removed median intensities of Cy5 emissions were extracted for each spot from raw images, using the ArrayVision V.7.2 software (Imaging Research Inc., Ontario, Canada).
An array grid was first automatically aligned to locate the position of each spot in the array, and then manually adjusted to obtain the best possible alignment.
To make the experiments comparable, intensity data derived from Cy5-labeled test samples from different hybridizations were normalized by LOWESS, local weighted scatter-plot smoothing.
Intensity data from a sample with total energy comparable to the average intensity of all samples was used as a reference in the LOWESS normalization procedure.
