Publication protocol
Tri Reagent (Molecular Research Center Inc., Cincinnati, OH) [65] was used to isolate total RNA with modifications similar to those previously described [24,25]. The final total cellular RNA concentrations (in RNase-free water, 1 mM EDTA) for MAV and choroid plexus ranged from 1.3 to 2 μg/μl, while those of the striatum and parietal cortex ranged from 2.5 to 4.0 μg/μl. The isolated RNA was stored at −70°C until amplification for oligo array analysis. Changes in gene expression were evaluated from total RNA using the Agilent One-Color Microarray-Based Gene Expression Platform. For hybridization onto Agilent-014879 Whole Rat Genome 4 × 44 K 60mer oligonucleotide arrays (G4131F, Agilent Technologies, Palo Alto, CA; NCBI GEO Accession # GPL7294), 500 ng of total RNA was used in the Agilent Quick Amp Labeling Kit (5190–0442) according to the manufacturer's instructions.
As mentioned above, total RNA from individual animals was used for array hybridization (no pooling of samples) for the MAV, striatum, and parietal cortex. Five groups were evaluated, but only four individual animals could be analyzed per slide. To facilitate comparisons and reduce normalization variability, the first five of the 4 × 44 K slides groups were sequentially rotated so that 4 animals in all 5 groups were evaluated. The remaining EIH and AMPH animals needing evaluation were run on the next four slides, each of which contained one normothermic control and selected repeats of previously (first five slides) analyzed samples. A total of nine 4 × 44 K array slides for expression analysis were used for each region (27 slides total). This was done to aid in array normalization and to ensure that minimal variability occurred in expression results across all nine slides. There was minimal variability between duplicate control or treatment arrays, however, in our statistical analysis duplicates were not averaged because this could not be done for the AMPH groups. The same platform and very similar methods were followed to analyze the pools of RNA (two to three rats per pool) derived from the choroid plexus, except that the control pools were not run in duplicate. Five 4 × 44 K array slides were used for gene expression analysis of the choroid plexus.
Total RNA was reverse transcribed into complementary DNA (cDNA) using T7-promotor primer and MMLV reverse transcriptase. The cDNA was transcribed into complimentary RNA (cRNA), during which it was fluorescently labeled by incorporation of cyanine (Cy)3-CTP. After purification, using the RNeasy mini kit (Qiagen), cRNA yield and Cy3 incorporation efficiency (specific activity) into the cRNA were determined using a NanoDrop® ND-1000 Spectrophotometer (NanoDrop Technologies). All cRNAs had a yield >5 μg and a specific activity of 8–15 pmol/μg. For each sample, a total of 1.65 μg cRNA was fragmented and hybridized onto an Agilent 4 × 44 K oligonucleotide microarray in a hybridization oven at 65°C for 17 hr using the Agilent Gene Expression Hybridization Kit (5188–5281). The hybridized microarrays were disassembled at room temperature in Gene Expression Wash Buffer 1 (5188–5325) and then washed in Gene Expression Wash Buffer 1 at room temperature for 1 min. This was followed by a wash for 1 min in Gene Expression Wash Buffer 2 (5188–5326) at an elevated temperature (≈31°C) and air drying. To preserve the fluorescent signal intensity on the microarrays from degradation of the Cy3 dye due to environmental ozone, the entire disassembly and washing procedure was conducted in a low-ozone laboratory (~1 ppb ozone), with the final drying step and scanning conducted inside a low-ozone biobubble (<1 ppb ozone). The arrays were scanned on an Axon 4000B scanner (Molecular Devices Corporation, Sunnyvale, CA) and further processed using GenePix Pro 6.0 software (Molecular Devices). The resulting text files were uploaded into the ArrayTrack database (NCTR, Jefferson, AR) for analysis. The local background was subtracted from the fluorescence values of each spot to obtain signal intensity values.
The working microarray dataset was obtained through gene filtering, background correction, logbase2 (log2) transformation, and normalization. Probes were filtered to include only those genes with official NCBI gene symbols. We included in our analysis each probe for genes with less than 4 probes. For genes with 4 or more probes, we used the average signal intensity after discarding the minimum and maximum values. Raw expression data were corrected for background signal. To symmetrize the distribution of background-corrected intensity values and stabilize the variance, we transformed the gene expression data using log2 transformation and applied quantile normalization [66]. Statistical significance of differential expression was determined for all pairwise treatment comparisons within the 3 hr and 1 day time points. Significance was determined using the Significance Analysis of Microarrays (SAM) method [67]. SAM computes modified gene-specific t-test statistics and estimates the false discovery rate (FDR) using permutation analysis with replicated samples of the expression data. The FDR [68] is a statistical method for multiple hypothesis testing that controls the expected proportion of false positives among all identified genes in microarray analysis. Using SAM, we were able to identify differentially expressed genes while controlling the false discovery rate. We obtained each list of differentially expressed genes using SAM with an estimated FDR of approximately 10% for each pairwise comparison for 3 hr analysis and an FDR of approximately 20% for the 1 day analysis. Further, the lists of significant genes were filtered so that genes meeting the following criteria remained: at least one of the two genes in the pairwise comparison had an average intensity equal to or greater than 150 and the fold-change between the two treatments exceeded 1.3. We used the inverse of the log2 transformation to convert transformed/normalized data back to their original domain and used these values to report average gene intensity and fold-change.
Genes determined to be significantly different at the 3 hr and 1 day time points were analyzed using pathway analysis through Ingenuity Pathway Analysis (IPA) software (Ingenuity System, Inc., Redwood City, CA). The “Core Analysis” function in IPA was used to interp
Full paper
Login or
join for free to view the full paper.