Sample #: 45 Survival Period: 8 status after 5 years: Dead HIST: SQ AGE: 59 SEX: Male pStage: IIIB pT: 3 pN: 1
Extracted molecule
total RNA
Extraction protocol
On average, 50 7 um-thick cryostat sections were prepared for the extraction of total RNA from tumor cell-rich areas, which were identified by a surgical pathologist (YY) using every 10th section stained by May-Giemsa. Careful attention was paid to the microdissection of only such tumor cell-rich areas, yielding an average of 75.4% tumor cell content (Figure 1i). RNAs were then isolated using RNAeasy (Quiagen, Valencia, CA, USA) according to the manufacturer's instruction, and their quality was checked with the RNA 6000 Nano Assay kit and the 2100 Bioanalyzer (Agilent, Palo Alto, CA, USA).
Label
33P
Label protocol
Five mg of total RNA was reverse-transcribed using oligo-dT primers (Invitrogen) and SuperScript II reverse transcriptase (Invitrogen) in the presence of 100 mCi of [33P]dCTP (Amersham Bioscience, Piscataway, NJ) according to the instructions for the GeneFilters (Invitrogen) with slight modification.
Hybridization protocol
The GeneFilters were prehybridized for 2 h at 51°C with 0.5 mg/ml of poly-dA (Invitrogen) and 0.5 mg/ml Cot-1 DNA (Invitrogen) in 10 ml of AlkPhos DIRECT hybridization buffer (Amersham Bioscience) and then hybridized for 17 h at 51°C with the denatured radio-labeled probe. Hybridization of the arrays was followed by two washings with a solution containing 2 M of urea, 0.1 % of SDS, 50 mM of Na phosphate buffer (pH 7.0), 150 mM of NaCl, 1 mM of MgCl2 and 0.2 % of AlkPhos DIRECT blocking reagent (Amersham Bioscience). The arrays were then washed twice with a solution containing 2 mM of MgCl2, 50 mM of Tris and 100 mM of NaCl, and with a solution containing 2 mM of MgCl2, 50 mM of Tris and 15 mM of NaCl. All procedures were carried out with the aid of custom-made AutoHybridizers (Fuji Photo Film, Tokyo, Japan).
Scan protocol
The arrays were then exposed for 2 hours to an Imaging Plate and scanned at 25-mm resolution with a BAS5000 phosphoimager (Fuji Photo Film), images of the hybridized arrays were processed with L Process (Fuji Photo Film) and signal intensities quantified with ArrayGauge software (Fuji Photo Film).
Description
Five mg of total RNA was reverse-transcribed using oligo-dT primers (Invitrogen) and SuperScript II reverse transcriptase (Invitrogen) in the presence of 100 mCi of [33P]dCTP (Amersham Bioscience, Piscataway, NJ) according to the instructions for the GeneFilters (Invitrogen) with slight modification. The GeneFilters were prehybridized for 2 h at 51°C with 0.5 mg/ml of poly-dA (Invitrogen) and 0.5 mg/ml Cot-1 DNA (Invitrogen) in 10 ml of AlkPhos DIRECT hybridization buffer (Amersham Bioscience) and then hybridized for 17 h at 51°C with the denatured radio-labeled probe. Hybridization of the arrays was followed by two washings with a solution containing 2 M of urea, 0.1 % of SDS, 50 mM of Na phosphate buffer (pH 7.0), 150 mM of NaCl, 1 mM of MgCl2 and 0.2 % of AlkPhos DIRECT blocking reagent (Amersham Bioscience). The arrays were then washed twice with a solution containing 2 mM of MgCl2, 50 mM of Tris and 100 mM of NaCl, and with a solution containing 2 mM of MgCl2, 50 mM of Tris and 15 mM of NaCl. All procedures were carried out with the aid of custom-made AutoHybridizers (Fuji Photo Film, Tokyo, Japan). The arrays were then exposed for 2 hours to an Imaging Plate and scanned at 25-mm resolution with a BAS5000 phosphoimager (Fuji Photo Film), images of the hybridized arrays were processed with L Process (Fuji Photo Film) and signal intensities quantified with ArrayGauge software (Fuji Photo Film). After each hybridization, the arrays were stripped by boiling them in 0.5% SDS solution for 60 min and scanned for residual hybridization before the next round of hybridization with a newly labeled probe of the same sample to acquire their expression profiles in duplicate.
Data processing
The raw data were re-scaled to account for the differences in individual hybridization intensities. We employed a rank-invariant scaling method to select genes (Tseng et al., 2001), which were then used for fitting of a non-linear normalization curve.Total genomic DNA was spotted on 192 spots for positive control. Nothing was spotted on 40 spot with the aim of negative control. Therefore, a total of 232 spot data were already eliminated from the whole 5584 spot data on the GF200 array.
