Abstract We used paired samples to address whether long-term storage of cells in TRIzol has a significant effect on microarray-based gene expression profiling (GEP). The TRIzol reagent is widely used for the preservation and isolation of RNA, but there is suspicion that prolonged storage of tissues prior to RNA isolation, even at −80 C., can cause chemical modification (depurination) of RNA. This could result in early termination during reverse transcription (RT) of mRNA molecules, potentially affecting GEP more strongly for transcripts with probes located farther from the 3’ end. The Myeloma and Leukemia Tissue Banks at M. D. Anderson Cancer Center routinely collect and store primary tumor samples for research under informed consent. From these banks we selected primary tumor samples of multiple myeloma (MM; CD138+) and acute myeloid leukemia (AML; Ficoll-purified, CD3- and CD19-depleted), for which paired aliquots had been stored frozen in TRIzol (“Tri”) or cryopreservation medium (RPMI + 20% FCS + 10% DMSO, “Cryo”) since the time of initial isolation (range, 2-9 years). Cryo aliquots were quickly thawed, washed, and placed in TRIzol, then total RNA was isolated from all aliquots as per the TRIzol manufacturer's instructions. RNA quality was assessed with a Bioanalyzer 2100 (Agilent); the RNA integrity number (RIN) was > 9 for 12/12 AML Tri, 6/12 AML Cryo, 6/6 MM Tri, and 3/6 MM Cryo samples. RNA from 12 sample pairs (6 AML, 6 MM, all with RIN > 7) was further purified with Qiagen RNAeasy columns, and the Illumina® TotalPrep RNA Amplification Kit (Ambion) was used to generate amplified, biotinylated cRNA after RT by the Eberwine procedure. Bioanalyzer-generated histograms of cRNA were similar for all sample aliquots, suggesting no early termination of RT due to TRIzol. GEP was performed with cRNA on Illumina HT-12 BeadArrays. Analysis of “probe-level” data with GenomeStudio software (Illumina), after median normalization, showed excellent concordance between sample Tri/Cryo pairs for all but probes with low intensities (<50). For the 48795 probes on the arrays, no more than 320 gave DiffScores < −13 or > 13 for any sample pair, and most pairs had <50 probes that differed. After exclusion of low-intensity values, hierarchical clustering associated all Tri aliquots with their corresponding Cryo aliquot. We conclude that long-term TRIzol storage has little effect on microarray-based GEP, and preserves RNA quality better than does cryopreservation. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 101st Annual Meeting of the American Association for Cancer Research; 2010 Apr 17-21; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2010;70(8 Suppl):Abstract nr 2177.
Read full abstract