Gene expression studies in tumour tissue in comparison with its corresponding normal tissue counterpart open up prospects for identifying new biomarkers and targets characteristic of the respective tumour entity. For that purpose, the relative quantification of reverse transcription-PCR (RT-PCR) data is the method of choice to ascertain gene expression results [1,2]. This method is based on the normalization of the target gene expression on any stably expressed internal reference gene, a so-called "housekeeping" gene, measured in the same biological material. A crucial problem involved here is finding such a suitable reference gene which has to be tested and verified under defined study conditions. The general conditions a reference gene must meet is that its expression in the samples studied be stable, non-regulated and constant and not be influenced by biological (e.g., age, gender, metabolism, disease stage) or experimental (e.g., addition or deprivation of physiological, physical or chemical agents) conditions or treatments [3-6]. In addition, the reference gene and the target gene should have similar ranges of expression to avoid analytical problems.
In literature, several single housekeeping genes or housekeeping gene indexes summarizing two or more housekeeping genes have been used for relative quantification [4,5,7,8]. However, these genes were often adopted without exact knowledge of their individual expression behaviour under the special study conditions. Recently, we discussed that issue in two studies concerning the identification of candidate reference genes for the relative quantification of expression data in the urological tumours of the prostate and bladder [8,9].
Renal cell carcinoma (RCC) is another important urological tumour. In 2007, RCC is estimated to cause 51.190 new cases and 12.890 deaths in the USA [10]. RCC is, in most cases, clinically asymptomatic and casually detected by routine ultrasonographic follow-ups in persons otherwise in generally good health [11]. The clear cell subtype of RCC (ccRCC) is the most frequent malignant RCC showing an incidence of about 75% [12]. It has a worse prognosis in comparison with the papillary and chromophobe RCC subtypes that account for 10% and 5% of RCC, respectively [12]. Although numerous tumour markers in RCC were tested in the past, to date no definitive biomarkers are available for diagnosis, monitoring, and predicting the outcome of RCC [13]. Thus, there is an urgent need to search for new markers. To identify potential new candidates gene expression profiles using the DNA microarray technique and subsequent RT-PCR assays are helpful tools [14]. Therefore, we searched for suitable normalization genes for gene expression studies on RCC tissue samples. Our strategy was the same as that used in studies of prostate and bladder normalization gene findings [8,9]. Using the MeSH terms "renal cell carcinoma", "gene expression", and "RT-PCR" combined with the Boolean operator "AND" we performed a PubMed search of articles published from October 1993 to August 2006. We had access to 156 articles that used 19 various reference genes (Table 1). It was remarkable that beta-actin (ACTB; 64 times; 37%) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 63 times; 36%) were used as normalizer genes in about three quarters of all studies. The use of these two genes for normalization is historically grown. Although their regulated expression shown under different conditions is inconsistent with their use as normalizers [1,15], this fact has often been disregarded till now or at least it has not been checked under the study conditions [15-18]. All other housekeeping genes cited accounted for only 1 to 6%. Only a few studies compared or used more than one reference gene, often in other organisms [19-21] or under other clinical conditions [7,22,23]. This literature search clearly proves that an univocal reference gene for gene expression studies in renal cell carcinoma does not exist. The search results additionally emphasize the need for a systematic study to identify suitable reference genes for gene expression studies in renal cell carcinoma.
The aim of this study was to identify suitable reference genes for the purpose of normalization in RCC gene expression studies. Therefore, we examined a panel of 10 candidate reference genes listed in Table 2 with regard to their expression behaviour in 25 matched malignant and non-malignant RCC tissue samples. This investigation was focused on the clear cell RCC subtype as the most frequent malignant RCC. We selected as potential reference genes either genes taken from the above-mentioned literature search or genes that were shown to be appropriate for normalization in previous studies of other tissues [6,8,9]. The suitability as reference gene was estimated by comparing the expression stability of the respective gene in malignant and non-malignant tissue using different mathematical procedures and computer programs. The use of unsuitable reference genes in the normalization procedure could result in serious errors in gene expression studies, which could lead to wrong conclusions being drawn [8,24-26]. These aspects are illustrated by the example of the relative gene quantification of "a disintegrin and metalloproteinase domain 9" (ADAM9) using up- and down-regulated housekeeping genes as normalizers.
Answers to all your Biology Questions
