With relative quantitative real-time reverse transcriptase PCR (qPCR), multiple genes across many specimens may be evaluated to measure changes in expression. However, to accurately determine the relative expression levels, and the corresponding fold changes, a reference gene is necessary. Reference genes, frequently termed “housekeeping genes,” are used to normalize the expression results for differences in cDNA quantity between different specimens and thus enable comparisons between genes of interest across treatments. In order to act as a reference, a housekeeping gene’s expression should remain unchanged regardless of treatment. Genes whose expression is generally unchanged with treatment conditions are most often associated with basic cellular processes such as metabolism. Our goal was to identify the most appropriate reference genes for analyses of porcine articular cartilage.
Regardless of the tissue being examined, housekeeping genes have usually been selected based on genes used in previous studies in various human tissues, and typically include beta actin (actb), beta 2 microglobulin (b2m), glyceraldehyde 3 phophate dehydrogenase (gapdh), hydroxymethylbilane synthase (hmbs), hypoxanthine guanine phosphoribosyl transferase (hprt), ribosomal protein L13a (rpl13a), ribosomal protein S18 (s18), succinate dehydrogenase flavoprotein subunit A (sdha), TATA box binding protein (tbp), and tyrosine 3 monooxygenase/tryptophan 5 monooxygenase activation protein—zeta polypeptide (ywhaz). A variety of genes have been used in the past as housekeeping genes in cartilage studies in various species. Gapdh has been used as a housekeeping gene in studies of human, bovine, porcine, and caprine articular cartilage, including both normal and osteoarthritic (OA) samples[1, 2]. Swingler et al. used sdha as a reference gene in their study of human OA cartilage. These genes appeared to be selected based on literature, not selected based on evaluation of a panel of genes to identify the most stable gene. Pombo-Suarez et al. evaluated nine of these same reference genes in addition to ubiquitin C in human cartilage with advanced OA and found the rarely used housekeeping genes TATA box binding protein (tbp), ribosomal protein L13a (rpl13a) and beta 2 microglobulin (b2m) to be the most stably expressed genes while they found the most commonly used genes (gapdh, actb and 18s) to be the least stable. Pombo-Suarez et al. therefore recommended that tbp, rpl13a and b2m be used as housekeeping genes for human cartilage research. In a canine study of normal and OA cartilage, rpl13a and sdha were identified as the most stable reference genes. The pig has been used as a model of human OA disease, cartilage repair, xenotransplantation, and gene transfer research, but no one has yet conducted a study to determine the ideal reference gene(s) for gene expression studies in porcine articular cartilage.
The reason for using a reference gene is to control for differences in the amount of starting material, efficiency of amplification, and differences in expression from cells and the overall level of transcription. Therefore, selecting a stable housekeeping gene presents a circular problem: determining a stable gene when that gene is expressed differently across samples/tissues. Several methods have been developed to identify the best housekeeping gene(s) from an initial panel of potential reference genes. Three of the most commonly used methods are geNorm, BestKeeper, and NormFinder. All of these programs attempt to provide a relative measure of the stability of a panel of genes by comparing their individual stability in relation to that of the entire panel.
In geNorm, the average pairwise gene expression variation of each potential housekeeping gene is compared to all other evaluated reference genes. The genes that demonstrate the least variance in comparison with all other genes are ranked as the most stable genes and are therefore likely to be the best reference genes. The authors developed a Visual Basic Application for Microsoft Excel (geNorm;) to carry out the analysis.
BestKeeper, developed by Pfaffl et al., uses an Excel based application to determine the most stable gene from a panel of up to ten candidate genes. The geometric mean of the cycle threshold values (Ct values) for each sample across all housekeeping genes are combined together to form the BestKeeper index. Subsequently, each individual gene is compared in a pair-wise fashion via Pearson correlation coefficients to the BestKeeper index. The outcome is a ranked order of genes in terms of their stability. The highest ranked gene is the most stable. Rather than using only one housekeeping gene or the impractical method of using all potential housekeeping genes, the authors recommended the use of the best 3 or 4 genes as that provides a realistic number of housekeeping genes while still providing adequate normalization of results.
NormFinder was developed by Ohl et al. and also uses an Excel based application to determine the most stable genes from a panel. This program uses a model-based approach, where all expression values are compared via analysis of variance, and all genes and specimen results are utilized for estimation of the expected expression values. A stability measure is calculated to identify the genes that deviate the least from the calculated values.
Nygard et al. evaluated a panel of nine genes using the geNorm approach to determine the best housekeeping genes across 17 different porcine tissues. That study included tissues such as muscle, adipose, heart, bladder, kidney, liver, skin, intestine, pancreas, bone marrow, and different portions of the brain, but no cartilage. They identified actb, ribosomal protein L4 (rpl4), tpb, and hprt as the most stably expressed housekeeping genes across the 17 tested tissues. Though cartilage was not included in the set of tissue they evaluated, their set of potential housekeepers included all of the genes previously discussed as commonly used in cartilage in other species with the exception of rpl13a and 18s. Rpl13a, like rpl4, encodes a protein of the 60S subunit of ribosomes and is still a good candidate while 18s has been shown to vary in proportion to total RNA and is therefore no longer considered a good candidate for normalization. In this study we propose to determine the best housekeeping genes for use in porcine articular cartilage and to evaluate three software packages, geNorm, BestKeeper, and NormFinder for determining overall gene stability. We used the nine genes identified by Nygard and co-workers as potential housekeeping genes as a starting point with the addition of peptidylprolyl isomerase A (ppia). Ppia was added because it has been used as a normalizing gene in cartilage for both OA-related[13, 14] and non-OA related studies[15, 16] and it exhibited no differential expression in impacted and control cartilage specimens in our previous work.