Gene expression profiling of isogenic cells with different TP53 gene dosage reveals numerous genes that are affected by TP53 dosage and identifies CSPG2 as a direct target of p53 -- Data Supplement - Supporting Information -- Proceedings of the National Academy of Sciences

Supporting information for Yoon et al. (2002) Proc. Natl. Acad. Sci. USA, 10.1073/pnas.242597299

Supporting Materials and Methods
Microarray Data Validation.
We used the TaqMan real-time assay (1) to validate the expression differences for seven genes. Relative expression levels initially determined by oligonucleotide-array were correlated with the TaqMan results for all samples. We chose TP53, HLA-E, CAPL, CSPG2, RGS2, and ICERelII, which were down-regulated in samples expressing p53 (P < 0.008, t test), and RTP which was up-regulated in p53 +/+ cells (P < 0.0001, t test). A comparison of the oligonucleotide-array and TaqMan data is shown in Fig. 3. For these genes the median correlation coefficient between the averaged GeneChip and TaqMan values for both 0-h and 12-h conditions was r = 0.908. We conclude that the GeneChip expression estimates correspond well to the actual gene expression level.

To perform TaqMan real-time assays, total RNA (2 mg) was treated with DNAse I to remove contaminating DNA (DNA-free, Ambion, Austin, TX), and was then reverse transcribed with both random hexamers and oligonucleotide-dT primers by using the Superscript system (Invitrogen). cDNA samples were pooled and the amount of cDNA was quantified with a spectrophotometer. The same amount of cDNA was used for all PCRs. All TaqMan probes were labeled with 6-carboxyfluorescein (6-FAM) at the 5' end and the tetramethylrhodamine (TAMRA) quencher at the 3' end. Sequences of PCR primers and TaqMan probes are available upon request. The comparative C_T (threshold cycle) method was used to determine the ratio of the target gene and the endogenous control according to specifications (Applied Biosystems). The threshold cycle was determined for the target and the internal control, and the cycle number difference (DC_T) was calculated. To determine the cDNA copy number difference among the samples, the difference between the samples, DC_T value, and a control sample (p53 +/+ untreated), was determined and noted as DDC_T. The cDNA copy number was then calculated by the expression 2^-DDCT, which correlates the cycle difference DDC_T to the cDNA copy doubling per each PCR cycle.

Two-Way Hierarchical Clustering.
To verify that gene expression profiling could discriminate among p53 +/+, p53 +/–, and p53 –/– cells, we performed two-way hierarchical clustering. Cluster analysis was performed by using cluster software and visualized by using treeview software (http://rana.stanford.edu/software) (2). A variation filter was applied to all genes to select those with a standard deviation above 100 and present in at least 80% of the samples (total of 5,548 genes). Filtered genes were log transformed, median centered, and normalized as recommended in the software manual. The hierarchical clustering organizes samples on the basis of overall similarities in their gene expression profile patterns (Fig. 4 a and c). In the dendrogram, the pattern and the length of the branches reflect the relatedness of the samples (Fig. 4 b and d).

1. Heid, C. A., Stevens, J., Livak, K. J. & Williams, P. M. (1996) Genome Res. 6, 986–994.

2. Eisen, M. B., Spellman, P. T., Brown, P. O. & Botstein, D. (1998) Proc. Natl. Acad. Sci. USA 95, 14863–14868.