MC3T3 cells were grown in 50 µg/ml ascorbic acid, 10 mM ß-glycerol phosphate and one of the following compounds: 20 nM TSA (Sigma, St. Louis, MO), 500 nM MS-275 (Calbiochem, San Diego, CA), 500 mM VPA (Sigma), or DMSO (vehicle) for 18 hours.
Growth protocol
MC3T3-E1 preosteoblasts were plated at 4x104 cells per well of a 12-well plate and differentiated in Minimal Essential Medium (Invitrogen, Carlsbad, CA) containing 10% FBS (Invitrogen and Cambrex Bioscience/Lonza, Basel. Switzerland), 100 U/ml penicillin, 100 μg/ml streptomycin, 50 µg/ml ascorbic acid, 10 mM ß-glycerol phosphate and one of the following compounds: 20 nM TSA (Sigma, St. Louis, MO), 500 nM MS-275 (Calbiochem, San Diego, CA), 500 mM VPA (Sigma), or DMSO (vehicle) for 18 hours.
Extracted molecule
total RNA
Extraction protocol
Total RNA was isolated from MC3T3-E1 osteoblasts with Trizol reagent (Invitrogen).
Label
biotin
Label protocol
Biotin-labeled cRNA was prepared from each culture according to the Affymetrix protocol. Briefly, RNA was denatured at 70oC with T7-oligo (dT) primer and reverse transcribed using Superscript II at 42oC for 1 hour. Second strand cDNA synthesis was performed by incubating first strand cDNA with Escherichia coli DNA polymerase I, E. coli. DNA ligase, RNase H, and dNTPs for 2 hours at 16oC. Biotin-labeled cRNA was prepared from double-stranded cDNA using the Affymetrix GeneChip IVT Labeling kit (Affymetrix, Santa Clara, CA), then purified and fragmented using the Affymetrix Sample Cleanup Module.
Hybridization protocol
Hybridization of the biotinylated-cRNA to the Affymetrix GeneChip Mouse Genome 430 2.0 Array was performed by the BioMedical Genomics Center’s microarray facility at the University of Minnesota using Affymetrix Genechip® Hybridization Oven 640 and Fluidics Station 450.
Scan protocol
Scanning was done with the Affymetrix Genechip® Scanner 3000.
Description
Comparison of control mice and mice treated with HDAC inhibitors.
Data processing
Raw data measurements were imported into Genedata Expressionist® Refiner (EPro1.0.32) for overall chip hybridization quality assessment, correction and condensation of probe sets intensity values. Robust Multichip Average (RMA) method was applied for global background subtraction and cross array normalization. Processed expression data were imported into Genedata Expressionist Analyst. General assessment of the data distribution was performed with Principle Component Analysis, boxplot and log-log data plots. Following LOESS normalization, differential expression measures were calculated by Welch test comparisons of control DMSO (vehicle) versus each HDAC inhibitor.
Significant genes from each comparison were selected using the Benjamini-Hochberg method to control for a maximum false discovery rate (FDR) of 95% or 99% as indicated.
