Reliable gene expression measurements from degraded RNA by quantitative real-time PCR depend on short amplicons and a proper normalization

• Published in: Laboratory investigation Vol. 85; no. 8; pp. 1040 - 1050
• Main Authors: ANTONOV, Janine, GOLDSTEIN, Darlene R, OBERLI, Andrea, BALTZER, Anna, PIROTTA, Marco, FLEISCHMANN, Achim, ALTERMATT, Hails J, JAGGI, Rolf
• Format: Journal Article
• Language: English
• Published: New York, NY Nature Publishing 01-08-2005; Nature Publishing Group
• Subjects: Base Sequence; Biological and medical sciences; Biotechnology; Breast Neoplasms - genetics; Cell Line, Tumor; DNA Primers; DNA Probes; Female; Fundamental and applied biological sciences. Psychology; Gene Expression Profiling; Humans; Investigative techniques, diagnostic techniques (general aspects); Medical sciences; Reproducibility of Results; Reverse Transcriptase Polymerase Chain Reaction - methods; RNA, Neoplasm - genetics; RNA, Neoplasm - metabolism; Biotechnology; RNA; Materials; Breast cancer; Malignant tumor; Gene expression; Paraffin; Real time; Polymerase chain reaction; Mammary gland diseases; breast cancer diagnosis; formalin-fixed; Gene amplification; paraffin-embedded material; quantitative real-time PCR; Clinical biology; Diagnosis; Quantitative analysis
• ISSN: 0023-6837; 1530-0307
• DOI: 10.1038/labinvest.3700303

Quantitative reverse transcriptase real-time PCR (QRT-PCR) is a robust method to quantitate RNA abundance. The procedure is highly sensitive and reproducible as long as the initial RNA is intact. However, breaks in the RNA due to chemical or enzymatic cleavage may reduce the number of RNA molecules that contain intact amplicons. As a consequence, the number of molecules available for amplification decreases. We determined the relation between RNA fragmentation and threshold values (Ct values) in subsequent QRT-PCR for four genes in an experimental model of intact and partially hydrolyzed RNA derived from a cell line and we describe the relation between RNA integrity, amplicon size and Ct values in this biologically homogenous system. We demonstrate that degradation-related shifts of Ct values can be compensated by calculating delta Ct values between test genes and the mean values of several control genes. These delta Ct values are less sensitive to fragmentation of the RNA and are unaffected by varying amounts of input RNA. The feasibility of the procedure was demonstrated by comparing Ct values from a larger panel of genes in intact and in partially degraded RNA. We compared Ct values from intact RNA derived from well-preserved tumor material and from fragmented RNA derived from formalin-fixed, paraffin-embedded (FFPE) samples of the same tumors. We demonstrate that the relative abundance of gene expression can be based on FFPE material even when the amount of RNA in the sample and the extent of fragmentation are not known.