Gene expression analysis by real-time reverse transcription polymerase chain reaction : in X uence of tissue handling

Factors such as warm ischemia and time at room temperature before tissue treatment may inXuence the results of mRNA expression analyses on tissue specimens obtained during surgery. We evaluated the eVect of these factors on RNA integrity and mRNA expression levels by incubating freshly obtained mouse liver tissue at 25 or 37 °C for periods of 0–4 h. Changes in the mRNA expression levels of seven genes, Tbp, Eef1a, Fos, Junb, Myc, Vegf, and Glut2, were determined by real-time reverse transcription-polymerase chain reaction. Incubation at 25 °C for up to 4 h only slightly altered (by a factor of less than 2) levels of mRNA for Tbp, Eef1a, Junb, Myc, Vegf, and Glut2. This result is consistent with limited RNA degradation at this temperature. Incubation at 37 °C strongly aVected the levels of these mRNAs. Four hours of incubation at this temperature resulted in extensive RNA degradation, with mRNA levels falling to 1/10th those before incubation. When relative quantiWcation was performed, i.e., quantiWcation of the target gene transcripts in comparison to an endogenous housekeeping transcript (Tbp or Eef1a), the changes in mRNA levels were reduced to less than 2.5-fold. Fos behaved very diVerently from the other genes tested on incubation, with Fos mRNA levels increasing considerably following incubation at either 25 or 37 °C. Our data suggest that, with the exception of certain genes induced by tissue injury, relative quantiWcation of mRNA, even on degraded RNA samples, can provide a reliable estimate of in vivo mRNA levels.  2004 Elsevier Inc. All rights reserved. Quantitative determination of gene expression at the studies at the mRNA level [12]. The standard surgical mRNA level is a powerful approach to the comparative analysis of normal and pathological states. Various methods including Northern blotting [1], RNase protection assays [2], RT1-PCR [3–6], branched DNA signal ampliWcation [7], serial analysis of gene expression [8], and DNA arrays [9,10] are available for quantifying mRNA levels. Real-time RT-PCR measures product accumulation during the log-linear phase of the reaction and is currently one of the most sensitive, accurate, and reproducible methods for mRNA expression quantiWcation [3–6]. The advantages of this method make real-time RT-PCR a method of choice for the validation of diVerentially expressed genes identiWed using DNA arrays [11]. Banks of frozen normal and pathological tissues obtained from surgical procedures were Wrst established in the 1990s for molecular analyses including expression ¤ Corresponding author. Fax: +33-1-42-34-63-44. E-mail address: francois.radvanyi@curie.fr (F. Radvanyi). 1 Abbreviations used: RT, reverse transcription. 0003-2697/$ see front matter  2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2004.02.004 excision and processing of tissue specimens are geared toward patient care and samples set aside for molecular studies are frequently suboptimal with respect to RNA integrity. The warm ischemia time in vivo before the surgical removal of the specimen and the time ex vivo at room temperature prior to freezing of the specimen or its treatment with an RNA-preserving solution are variable and this may have signiWcant consequences with regard to the mRNA levels observed. To evaluate the eVects of warm ischemia and time delays at room temperature on RNA integrity and mRNA expression levels, we incubated freshly obtained mouse liver tissue at 25 or 37 °C for periods of 0–240 min before homogenization in guanidine isothiocyanate. We used animal tissue so that we could control incubation temperature and the delay before RNA extraction. As liver is a fairly homogenous tissue, it was possible to compare changes in mRNA expression levels and RNA stability in various fragments of liver tissue exposed to various temperatures for various times. We assessed 102 A. Almeida et al. / Analytical Biochemistry 328 (2004) 101–108 RNA degradation by following the decrease in 18S and 28S rRNA species and the appearance of degradation products by microcapillary gel electrophoresis [13]. The eVects of time and incubation temperature on the mRNA levels of seven genes were measured by real-time RT-PCR. The genes included in this study were three immediate early genes (Fos, Junb, and Myc) [14–16], one hypoxia-induced gene (Vegf) [17,18], two housekeeping genes (Tbp, Eef1a) [19–21], and one tissue-speciWc gene (Glut2) [22]. Materials and methods