- Research article
- Open Access
Exercise induced stress in horses: Selection of the most stable reference genes for quantitative RT-PCR normalization
© Cappelli et al; licensee BioMed Central Ltd. 2008
- Received: 22 November 2007
- Accepted: 19 May 2008
- Published: 19 May 2008
Adequate stress response is a critical factor during athlete horses' training and is central to our capacity to obtain better performances while safeguarding animal welfare.
In order to investigate the molecular mechanisms underlying this process, several studies have been conducted that take advantage of microarray and quantitative real-time PCR (qRT-PCR) technologies to analyse the expression of candidate genes involved in the cellular stress response.
Appropriate application of qRT-PCR, however, requires the use of reference genes whose level of expression is not affected by the test, by general physiological conditions or by inter-individual variability.
The expression of nine potential reference genes was evaluated in lymphocytes of ten endurance horses during strenuous exercise. These genes were tested by qRT-PCR and ranked according to the stability of their expression using three different methods (implemented in geNorm, NormFinder and BestKeeper). Succinate dehydrogenase complex subunit A (SDHA) and hypoxanthine phosphoribosyltransferase (HPRT) always ranked as the two most stably expressed genes. On the other hand, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), transferrin receptor (TFRC) and ribosomal protein L32 (RPL32) were constantly classified as the less reliable controls.
This study underlines the importance of a careful selection of reference genes for qRT-PCR studies of exercise induced stress in horses. Our results, based on different algorithms and analytical procedures, clearly indicate SDHA and HPRT as the most stable reference genes of our pool.
- Reference Gene
- Candidate Reference Gene
- Stable Reference Gene
- Hypoxanthine Phosphoribosyltransferase
- Good Reference Gene
Knowledge of the molecular mechanisms underlying the stress response in athlete horse is a fundamental prerequisite for planning an appropriate training schedule to obtain better performances, preserve animal welfare and avoid overtraining-syndrome [1, 2].
It is universally accepted that moderate physical activity may have beneficial effects in terms of general health conditions and could favour the functioning of the immune system. Conversely, strenuous exercise, like exhaustive endurance races, may have detrimental effects on the immune system, determine changes in the cellular composition of peripheral blood and induce the expression of genes that appear to be related to the overtraining-syndrome [3–5]. The list of candidate genes is nevertheless far from being complete, as the athlete's reaction to exercise is a coordinated response of multiple organ systems, and likely involves multiple and complex regulatory changes: induction of heat shock proteins, inflammatory response modulation (pro and anti-inflammatory cytokines) and generation of reactive oxygen and nitrogen species (ROS and RSN) that, besides their damaging potential, play a crucial role in cellular signalling [6, 7].
Since exercise has been shown to be an important factor in regulating immune cells and their functions, and considering that stress evokes inflammatory reactions, lymphocytes are considered the best candidate cell type to study physiological changes associated with exhaustive exercise .
Quantitative real-time PCR (qRT-PCR) is the technique of choice when trying to detect modifications in transcription levels in a reliable and reproducible manner. Nevertheless, there are some technical issues that must be taken into account, such as quality and quantification of the starting material, enzyme efficiency, and primer design. Different approaches have been proposed to normalize measurements of expression levels , but this is generally done using an internal control gene, known as a reference gene or as housekeeping gene (HKG), under the assumption that this has a constant level of expression in the chosen tissue, is not affected by the treatment and has no inter-individual variability. In addition, the reference gene and the target gene should have similar ranges of expression to avoid analytical problems.
Widely expressed genes like ACTB, GAPDH or R18S are generally preferred, without preliminary analysis of their expression profiles under the specific study conditions [5, 6, 9]. Nevertheless, a number of studies report how commonly accepted HKGs do not always constitute reliable controls [10–14], because of unexpected variation in their expression profiles.
More appropriately, multiple HKGs should be evaluated before their employment, and their stability should be measured in the context of the relevant experimental conditions.
The aim of this paper is to identify the best reference genes for qRT-PCR experiments investigating horse lymphocyte gene expression in exercise induced stress. Statistical algorithms implemented in geNorm , BestKeeper , and NormFinder  were used.
Details of the nine genes evaluated.
Cytoskeletal structural protein
Cytoskeletal protein involved in cell locomotion
Metabolic salvage of purines in mammals
ribosomal RNA 18S
Member of ribosome RNA
ribosomal protein L32
Member of ribosomal proteins
succinate dehydrogenase complex subunit A
Electron transporter in the TCA cycle and respiratory chain
Assay conditions for each of the nine genes evaluated.
Primers 5'-3' (forward, reverse)
Amplicon length, bp
PCR efficiency, %
Correlation with dilution series (R2)
Expression levels of candidate reference genes
Profiles obtained for each horse and HKG were analysed using three different methods, implemented in the software geNorm, NormFinder and BestKeeper.
Candidate reference genes ranking according to geNorm.
Candidate reference genes ranking according to NormFinder.
Statistical output from BestKeeper analysis.
G Mean [Ct]
A Mean [Ct]
SD [± Ct]
CV [% Ct]
coeff. of corr. [r]
The nine control genes tested in our analysis correlated well one with one another and with the BestKeeper index (Table 5). The best correlation between one HKG and the BestKeeper index was obtained for SDHA (r = 0.983), followed by HPRT, R18S and UBB. The statistically significant correlation shown by SDHA with the BestKeeper index appeared to be consistent with the good performance of this gene as assessed by geNorm and NormFinder. It is also remarkable how GAPDH, RPL32 and TFRC, are again classified as the least reliable HKGs, showing the worst correlations with the determined BestKeeper index (Table 5).
A number of authors have studied gene expression profiles in exercise induced stress using forefront technologies, like gene chips and qRT-PCR. This study is the first solid contribution in assessing which reference genes have to be used to validate and normalize qRT-PCR outcomes.
Several methods have been proposed to allow accurate normalization of gene expression using qRT-PCR [9, 19–22] but at present there is no consensus on which algorithm should be used to measure reference gene stability. A comparison of different methods of reference gene selection allows a better identification of the most reliable controls and reduces the risk of artificial selection of co-regulated transcripts .
We compared three different statistical approaches (geNorm, NormFinder and BestKeeper) to evaluate nine potential HKGs, in order to select the best reference gene to be used in studying exercise-induced stress in horses.
The uniformity in gene ranking between the three software packages was generally high: SDHA is the most stable HKG according to all the three methods. HPRT similarly displays a constant significant stability. R18S always ranks third, B2M fourth and UBB fifth, and can be therefore considered plausible HKGs, even if the addition of supplementary reference genes would not significantly enhance the reliability of the normalization according to the geNorm analysis (V value, Table 3).
Regarding ACTB, it is difficult to formulate a final judgement because, as already reported in a previous study , its classification is not consistent between the three software packages (6th in geNorm and BestKeeper, 2nd in NormFinder). Nevertheless, this gene shows an overall reduced variability, as attested by the M value calculated by geNorm and by its good correlation with the BestKeeper index (r = 0.890).
GAPDH, TFRC and RPL32 were classified as the least stable genes and they are not likely to be useful in this given experimental system. Notably, the expression of GAPDH, that has been used as HKG in a previous exercise induced stress study , appears to be the least stable.
In contrast with what reported elsewhere [23–26], R18S appears to be a good potential reference gene. Despite its good performance, the usefulness of this gene as a control is often doubted: some authors  tend to consider it unsuitable for normalization because its transcription is carried out by RNA polymerase I and because of its well known over-expression in comparison with mRNAs (as confirmed even in our experiments, Figure 1). Considering that HKGs that have expression levels comparable to the gene of interest are generally preferred , its usage should be carefully considered if used for the normalization of genes that exhibit low level of expression.
Our results indicate SDHA and HPRT as the most stable reference genes with a very good statistical reliability according to all the three software employed. Moreover, the use of only two genes (SDHA and HPRT) appears to be sufficient for a reliable normalization of the genes of interest; this result is of special interest for future high throughput applications of the technique.
Blood collection, RNA extraction and cDNA synthesis
Ten horses were chosen among participants to national endurance races (90–120 km). Blood samples were taken from the jugular vein and collected at three different time points: before, at the end of the race, and 24 hours after the race. Immediately after collection, peripheral blood mononuclear cells (PBMCs) were isolated by the Ficoll-Hypaque method (GE Healthcare, Pollards Wood, United Kingdom) from 8 ml of blood. Total RNA was extracted from approximately 1 × 107 PBMCs, using the Aurum Total RNA Fatty and Fibrous Tissue kit (Bio-Rad, Herculers CA, USA) according to the manufacturer's instructions. Genomic DNA was eliminated by a DNase treatment supplied with the kit. Extracted RNA was quantified using the Quant-It RNA assay (Invitrogen, Dorset, United Kingdom) in a VersaFluor fluorometer (Bio-Rad) and checked for integrity in a denaturing agarose gel electrophoresis with ethidium bromide staining. Successful removal of DNA contaminants was tested by absence of PCR amplification in the MC1R gene (GenBank accession number X98012, primers from ). 1.5 μg of total RNA were retro-transcribed using random hexamers and Superscript III Reverse Transcriptase (Invitrogen) according to the manufacturer's specifications. A PCR with ACTB primers (Table 2) was performed on each cDNA to check for successful retro-transcription.
Reference genes selection and primer design
Nine widely used reference genes were evaluated: β-actin (ACTB), glyceraldehyde-3P-dehydrogenase (GAPDH), hypoxanthine ribosyltransferase (HPRT1), β-2-microglobin (B2M), succinate dehydrogenase complex subunit A (SDHA), transferrin receptor (TFRC), ubiquitin B (UBB), ribosomal protein L32 (RPL32) and 18S ribosomal rRNA (R18S). In order to minimize the possibility of co-regulation, genes were selected from different functional classes.
Primers were designed based on available sequences using the Primer3 software.
Mfold  was used to check the chosen sequences to avoid designing primers in the region of template secondary structure; amplicon lengths were optimized to 68/138 bp to ensure optimal polymerization efficiency. Specificity of amplification was confirmed by sequencing.
For each primer pair, a preliminary real-time assay was performed to evaluate the amplification of non-specific products or primer dimer artefacts (Additional files 1, 2, 3, 4, 5, 6, 7, 8, 9). Efficiency of RT-PCR (E), slope values, and correlation coefficients (R2) were determined (Table 2) using serial 1:4 dilutions of a template cDNA (pooled from the ten individuals studied, previously diluted 1:5).
PCR products were subsequently resolved on 2% agarose gel to check for size specificity of the amplicon.
Real-time quantitative PCR
Five microliters of cDNA template (previously diluted 1:10) were added to the master mix FastStart SYBR Green Master (Roche Applied Science, Penzberg, Germany) with the ROX fluorochrome internal check. PCR reactions, in a volume of 25 μl were performed on a MX3000P machine (Stratagene, La Jolla CA, USA). PCR conditions were the same for all primer pairs: initial denaturation at 95°C for 10' followed by 40 cycles of denaturation at 95°C for 30", annealing at 58°C for 30" and extension at 72°C for 30". Fluorescence data were collected at the end of the extension step. Following cycling, the melting curve was determined in the range 58°–95°C, with a temperature slope of 0.01°C/sec. Each reaction was run in triplicate with appropriate negative controls.
Baseline and threshold values were automatically determined for all plates and genes using the MxPro software ver. 3.20 (Stratagene). In order to ensure comparability between data obtained from different experimental plates, threshold values for each gene were manually set to the arithmetic mean between the thresholds as automatically determined following each run. Corrected Ct values were transformed to quantities based on the comparative Ct method. Following appropriate formatting, values were imported into geNorm (version 3.4), NormFinder (version 0.953) and BestKeeper (version 1) VBA applets.
Work sponsored by MIUR – PRIN 2006 (coordinator Prof. Maurizio Silvestrelli)
The authors thank Mr Gianluca Alunni for his valuable technical support and Dr. Francesco Nardi for his suggestions in writing this manuscript. We are also grateful with the horse breeders: Mr Giuseppe Sereni, Mr. Carlo Formica, Miss Chiara Rosi for allowing collection of samples and Dr. Marcello Conte for his kind help.
- Hamlin MJ, Shearman JP, Hopkins WG: Changes in physiological parameters in overtrained Standardbred racehorses. Equine Veterinary Journal 2002, 34: 383-388. 10.2746/042516402776249146View ArticlePubMedGoogle Scholar
- Angeli A, Minetto M, Dovio A, Paccotti P: The overtraining syndrome in athletes: A stress-related disorder. Journal of Endocrinological Investigation 2004, 27: 603-612.View ArticlePubMedGoogle Scholar
- Cappelli K, Verini-Supplizi A, Capomaccio S, Silvestrelli M: Analysis of peripheral blood mononuclear cells gene expression in endurance horses by cDNA-AFLP technique. Research in Veterinary Science 2007, 82: 335-343. 10.1016/j.rvsc.2006.08.009View ArticlePubMedGoogle Scholar
- Barrey E, Mucher E, Robert C, Amiot F, Gidrol X: Gene expression profiling in blood cells of endurance horses completing competition or disqualified due to metabolic disorder. Equine Veterinary Journal Supply 2006, 36: 43-49.View ArticleGoogle Scholar
- Buttner P, Mosig S, Lechtermann A, Funke H, Mooren FC: Exercise affects the gene expression profiles of human white blood cells. Journal of Applied Physiology 2007, 102: 26-36. 10.1152/japplphysiol.00066.2006View ArticlePubMedGoogle Scholar
- Zieker D, Fehrenbach E, Dietzsch J, Fliegner J, Waidmann M, Nieselt K, Gebicke-Haerter P, Spanagel R, Simon P, Niess AM, Northoff H: cDNA microarray analysis reveals novel candidate genes expressed in human peripheral blood following exhaustive exercise. Physiological Genomics 2005, 23: 287-294. 10.1152/physiolgenomics.00096.2005View ArticlePubMedGoogle Scholar
- Connolly PH, Caiozzo VJ, Zaldivar F, Nemet D, Larson J, Hung SP, Heck JD, Hatfield GW, Cooper DM: Effects of exercise on gene expression in human peripheral blood mononuclear cells. Journal of Applied Physiology 2004, 97: 1461-1469. 10.1152/japplphysiol.00316.2004View ArticlePubMedGoogle Scholar
- Huggett J, Dheda K, Bustin S, Zumla A: Real-time RT-PCR normalisation; strategies and considerations. Genes and Immunity 2005, 6: 279-284. 10.1038/sj.gene.6364190View ArticlePubMedGoogle Scholar
- Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F: Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology 2002, 3: research0034-research0034. 10.1186/gb-2002-3-7-research0034PubMed CentralView ArticlePubMedGoogle Scholar
- Selvey S, Thompson EW, Matthaei K, Lea RA, Irving MG, Griffiths LR: Beta-Actin - An unsuitable internal control for RT-PCR. Molecular and Cellular Probes 2001, 15: 307-311. 10.1006/mcpr.2001.0376View ArticlePubMedGoogle Scholar
- Barber RD, Harmer DW, Coleman RA, Clark BJ: GAPDH as a housekeeping gene: Analysis of GAPDH mRNA expression in a panel of 72 human tissues. Physiological Genomics 2005, 21: 389-395. 10.1152/physiolgenomics.00025.2005View ArticlePubMedGoogle Scholar
- Bas A, Forsberg G, Hammarstrom S, Hammarstrom ML: Utility of the housekeeping genes 18S rRNA, beta-actin and glyceraldehyde-3-phosphate-dehydrogenase for normalization in real-time quantitative reverse transcriptase-polymerase chain reaction analysis of gene expression in human T lymphocytes. Scandinavian Journal of Immunology 2004, 59: 566-573. 10.1111/j.0300-9475.2004.01440.xView ArticlePubMedGoogle Scholar
- Dheda K, Huggett JF, Bustin SA, Johnson MA, Rook G, Zumla A: Validation of housekeeping genes for normalizing RNA expression in real-time PCR. Biotechniques 2004, 37: 112-119.PubMedGoogle Scholar
- Banda M, Bommineni A, Thomas RA, Luckinbill LS, Tucker JD: Evaluation and validation of housekeeping genes in response to ionizing radiation and chemical exposure for normalizing RNA expression in real-time PCR. Mutation Research 2007.Google Scholar
- Spinsanti G, Panti C, Lazzeri E, Marsili L, Casini S, Frati F, Fossi CM: Selection of reference genes for quantitative RT-PCR studies in striped dolphin (Stenella coeruleoalba) skin biopsies. Bmc Molecular Biology 2006., 7:Google Scholar
- Ayers D, Clements D, Salway F, Day P: Expression stability of commonly used reference genes in canine articular connective tissues. BMC Veterinary Research 2007., 3:Google Scholar
- Maccoux LJ, Clements DN, Salway F, Day PJR: Identification of new reference genes for the normalisation of canine osteoarthritic joint tissue transcripts from microarray data. Bmc Molecular Biology 2007., 8:Google Scholar
- Bogaert L, Van Poucke M, De Baere C, Peelman L, Gasthuys F, Martens A: Selection of a set of reliable reference genes for quantitative real-time PCR in normal equine skin and in equine sarcoids. Bmc Biotechnology 2006., 6:Google Scholar
- Pfaffl MW, Tichopad A, Prgomet C, Neuvians TP: Determination of stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper - Excel-based tool using pair-wise correlations. Biotechnology Letters 2004, 26: 509-515. 10.1023/B:BILE.0000019559.84305.47View ArticlePubMedGoogle Scholar
- Andersen CL, Jensen JL, Orntoft TF: Normalization of real-time quantitative reverse transcription-PCR data: A model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Research 2004, 64: 5245-5250. 10.1158/0008-5472.CAN-04-0496View ArticlePubMedGoogle Scholar
- Akilesh S, Shaffer DJ, Roopenian D: Customized molecular phenotyping by quantitative gene expression and pattern recognition analysis. Genome Research 2003, 13: 1719-1727. 10.1101/gr.533003PubMed CentralView ArticlePubMedGoogle Scholar
- Haller F, Kulle B, Schwager S, Gunawan B, Heydebreck AV, ltmann H, zesi L: Equivalence test in quantitative reverse transcription polymerase chain reaction: Confirmation of reference genes suitable for normalization. Analytical Biochemistry 2004, 335: 1-9. 10.1016/j.ab.2004.08.024View ArticlePubMedGoogle Scholar
- Blanquicett C, Johnson MR, Heslin M, Diasio RB: Housekeeping gene variability in normal and carcinomatous colorectal and liver tissues: Applications in pharmacogenomic gene expression studies. Analytical Biochemistry 2002, 303: 209-214. 10.1006/abio.2001.5570View ArticlePubMedGoogle Scholar
- Thellin O, Zorzi W, Lakaye B, De Borman B, Coumans B, Hennen G, Grisar T, Igout A, Heinen E: Housekeeping genes as internal standards: Use and limits. Journal of Biotechnology 1999, 75: 291-295. 10.1016/S0168-1656(99)00163-7View ArticlePubMedGoogle Scholar
- Schmittgen TD, Zakrajsek BA: Effect of experimental treatment on housekeeping gene expression: Validation by real-time, quantitative RT-PCR. Journal of Biochemical and Biophysical Methods 2000, 46: 69-81. 10.1016/S0165-022X(00)00129-9View ArticlePubMedGoogle Scholar
- Bond BC, Virley DJ, Cairns NJ, Hunter AJ, Moore GBT, Moss SJ, Mudge AW, Walsh FS, Jazin E, Preece P: The quantification of gene expression in an animal model of brain ischaemia using TaqMan real-time RT-PCR. Molecular Brain Research 2002, 106: 101-116. 10.1016/S0169-328X(02)00417-5View ArticlePubMedGoogle Scholar
- Radonic A, Thulke S, Mackay IM, Landt O, Siegert W, Nitsche A: Guideline to reference gene selection for quantitative real-time PCR. Biochemical and Biophysical Research Communications 2004, 313: 856-862. 10.1016/j.bbrc.2003.11.177View ArticlePubMedGoogle Scholar
- Rieder S, Taourit S, Mariat D, Langlois B, Guerin G: Mutations in the agouti (ASIP), the extension (MC1R), and the brown (TYRP1) loci and their association to coat color phenotypes in horses (Equus caballus). Mammalian Genome 2001, 12: 450-455. 10.1007/s003350020017View ArticlePubMedGoogle Scholar
- Zuker M: Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Research 2003, 31: 3406-3415. 10.1093/nar/gkg595PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.