Experimental comparison of relative RT-qPCR quantification approaches for gene expression studies in poplar
© Regier and Frey; licensee BioMed Central Ltd. 2010
Received: 22 March 2010
Accepted: 11 August 2010
Published: 11 August 2010
RT-qPCR is a powerful tool for analysing gene expression. It depends on measuring the increase in fluorescence emitted by a DNA-specific dye during the PCR reaction. For relative quantification, where the expression of a target gene is measured in relation to one or multiple reference genes, various mathematical approaches are published. The results of relative quantification can be considerably influenced by the chosen method.
We quantified gene expression of superoxide dismutase (SOD) and ascorbate peroxidase (APX) in the roots of two black poplar clones, 58-861 and Poli, which were subjected to drought stress. After proving the chosen reference genes actin (ACT), elongation factor 1 (EF1) and ubiquitin (UBQ) to be constantly expressed in the different watering regimes, we applied different approaches for relative quantification to the same raw fluorescence data. The results obtained using the comparative Cq method, LinRegPCR, qBase software and the Pfaffl model showed a good correlation, whereas calculation according to the Liu and Saint method produced highly variable results. However, it has been shown that the most reliable approach for calculation of the amplification efficiency is using the mean increase in fluorescence during PCR in each individual reaction. Accordingly, we could improve the quality of our results by applying the mean amplification efficiencies for each amplicon to the Liu and Saint method.
As we could show that gene expression results can vary depending on the approach used for quantification, we recommend to carefully evaluate different quantification approaches before using them in studies analysing gene expression.
RT-qPCR is a widely used method for analysing gene expression. It has been developed by combining PCR with fluorescent techniques [1, 2]. It depends on collecting data throughout the PCR amplification, which is achieved by monitoring the increase in fluorescence intensity of a specific fluorescence dye, which correlates to the increase in PCR product concentration. The major progress of qPCR is that quantification does not have to be done in the plateau phase of amplification, which is a disadvantage of previous quantification methods .
PCR can be divided into four major phases: linear ground phase, early exponential phase, log-linear phase and plateau phase . During the linear ground phase, only background fluorescence is detected. The early exponential phase starts when the amount of fluorescence is significantly higher than the background. During the log-linear phase, when PCR has reached its optimal amplification period, the amount of fluorescence rises exponentially. In an ideal reaction the PCR products double after every cycle. Finally, when the reaction components become limited, the plateau phase is reached and the fluorescence does not increase anymore .
Relative quantification of RT-qPCR is used to detect changes in expression of the genes of interest relative to a reference gene, which is usually a housekeeping gene. Early RT-qPCR studies have assumed that housekeeping genes are expressed constantly over a wide range of conditions. Later studies have shown that their expression stability should be proven before choosing them as references, and proposed to use not only one, but several reference genes as internal controls [5, 6].
Many approaches are available for relative quantification of gene expression. Most of them depend on the principle to define a threshold at which the PCR product fluorescence rises over the background fluorescence. The number of cycles needed until this threshold is reached, depending on the amount of template in a sample, is usually called Cq - the higher the template amount, the lower is the Cq value. Widely used approaches depending on this principle are the 2-ΔΔCq or comparative Cq method , the Pfaffl model , or qBase software . While the comparative Cq method assumes the same amplification efficiency for all amplicons, other methods use serial dilutions of the samples to determine the amplification efficiencies from the increase in the Cq value with decreasing cDNA input . Another approach is to determine the amplification efficiency or starting template amount from the increase in fluorescence during the PCR reaction [11, 12].
The aim of this study was to evaluate, whether the different approaches for relative quantification of RT-qPCR generate comparable results. We wanted to find suitable reference genes for normalization of gene expression in a study investigating drought responses of poplar [13, 14] and to test whether the tested methods differ in reliability and suitability for our approach of quantifying gene expression. For this test, ascorbate peroxidase (APX) and superoxide dismutase (SOD) were used as target genes, as they play an important role in oxidative stress defence in plants subjected to drought. Actin (ACT), elongation factor 1 (EF1) and ubiquitin (UBQ) were used as reference genes.
Results and Discussion
Evaluation of expression stability of the reference genes
As third approach for calculation of reference gene expression stability we used the NormFinder software . It revealed expression stabilities of 0.105 for ACT, 0.142 for EF1 and 0.099 for UBQ. Comparison with other studies using this program [16–18] revealed that our reference genes had comparable or even higher expression stability than the ones reported in the literature. This high stability revealed with all applied evaluation approaches [5, 6, 15] might be due to the low number of sample types (expression analysis only in roots, two clones and three treatments, which gives a final number of 6 sample types) as compared to the other studies.
Interestingly, the three approaches did not identify the same reference gene to have the most stable expression. According to Reid et al (2006), EF1 seemed to be the best reference gene (Figure 1) , according to Brunner et al (2004) ACT (Figure 2) , and with NormFinder UBQ . Nevertheless, the differences in expression stability between the genes were very small and thus we conclude that all three genes are suitable to be used for normalizing gene expression in our study, even when used as reference gene alone. However, it has been shown that the use of several reference genes, which are not regulated under the different conditions between different treatments or tissue types, leads to an even higher reliability of quantitative gene expression studies .
Importance of amplification efficiency for relative quantification
When using the approach from Liu and Saint (2002) , we found amplification efficiencies of 0.91 ± 0.01 for ACT, 0.81 ± 0.01 for EF1, 0.87 ± 0.02 for UBQ, 0.87 ± 0.02 for APX and 0.94 ± 0.02 for SOD. For this calculation, two arbitrary thresholds within the exponential phase have to be chosen. During evaluation of the method we found that minor variations in the choice of the thresholds can lead to large differences in the calculated amplification efficiency. When using the approach presented by Ruijter et al (2009) , we found amplification efficiencies of 1.97 ± 0.03 for ACT, 1.91 ± 0.01 for EF1, 1.94 ± 0.01 for UBQ, 1.95 ± 0.05 for APX and 1.92 ± 0.05 for SOD. In this method, the data were log-transformed before calculation of the efficiency from the log-linear phase, and in contrast to the Liu and Saint method , all data points within the log-linear phase were used, which makes it more reliable to detect the correct data points. Nordgård et al (2006) have shown that all approaches using the amplification plots of each individual reaction for calculation of amplification efficiencies produce large errors in quantification of gene expression . They recommended using rather serial dilutions than individual reactions in order to increase precision. However, in recent years, the approach to use individual samples to determine amplification efficiencies has become more common in relative qPCR studies than using serial dilutions . It has been shown that the most reliable approach is to assume the same amplification efficiency for all reactions with the same primer pair , to calculate the amplification efficiency from each individual amplification plot, check for outliers and use the mean efficiency for all samples .
Evaluation of the quantification approaches and comparison of gene expression
To summarize, we were able to show that the different approaches available for relative quantification of RT-qPCR data differ in their reliability, and that the results computed from the same dataset can differ considerably. The genes ACT, EF1 and UBQ selected to normalize expression of the target genes APX and SOD have been proven to be non-regulated in the different experimental conditions. In accordance with the literature, the approach calculating PCR efficiencies for each individual reaction  produced highly variable results. However, as it has been shown that the use of the mean amplification efficiency for each gene, computed from the individual reactions, is the most reliable approach, we conclude that the use of the Liu and Saint method is suitable to determine the amplification efficiencies, if the mean value for each primer pair is used for further analysis . The good estimation of the template amount of a cDNA dilution series with known relative concentrations by the comparative Cq method, LinRegPCR, the Liu & Saint method with average amplification efficiencies for each amplicon, the Pfaffl model and qBase software suggest that these approaches produce reliable results. We do not offer a universal recommendation which approach should be used for relative gene expression studies, but we suggest that investigators should carefully evaluate different quantification approaches before using them in studies analysing gene expression.
Plant material, RNA extraction and cDNA synthesis
We used fine roots of two clones of black poplar (Populus nigra L.), Poli and 58-861, which had been subjected to different drought treatments [13, 14]. Well-watered (WW) plants, which were watered to field capacity, were used as controls. Drought treatments were 50% water limitation (WL 50%) and 75% water limitation (WL 75%). All samples were used for evaluation of reference gene expression stability, whereas expression of the target genes was analysed only in WW and WL 75% samples. RNA was extracted using the Agilent Plant RNA Isolation Mini Kit according to the manufacturer's instructions (Agilent Technologies AG, Basel, Switzerland). RNA concentration was measured using a NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Wilmington DE, USA) in 1 μL volume. RNA quality was assessed using the Agilent Bioanalyzer (Agilent Technologies AG, Basel, Switzerland) with the Agilent RNA 6000 Pico Kit. First strand cDNA synthesis was performed with the QuantiTect Reverse Transcription Kit (Qiagen) using 200 ng total RNA according to the manufacturer's instructions. Reverse transcription was initiated using the RT-primer mix supplied with the kit, consisting of oligo-dT and random primers.
Primer design and evaluation
Amplicon size (bp)
ACC CTC CAA TCC AGA CAC TG
TTG CTG ACC GTA TGA GCA AG
AAG CCA TGG GAT GAT GAG AC
ACT GGA GCC AAT TTT GAT GC
CGT GGA GGA ATG CAG ATT TT
GAT CTT GGC CTT CAC GTT GT
TCT TGC GAG GAA GTG AAG GT
AAT GGT TGG ACC TCC AGT GA
GGG TCT CGT CCA ACA CAC TT
AGC CAT GGC GAT AGA TTG AC
RT-qPCR reactions of 15 μL total volume contained 7.5 μL 2× FastStart Universal SYBR Green Master Mix (Roche, Basel, Switzerland), 5 μM of forward and reverse primers and 1 μL of 1:10 or 1:100 diluted cDNA. RT-qPCR was performed on an ABI 7500 Fast real-time PCR system (Applied Biosystems) with the following conditions: 10 min 95°C initial denaturation; 40 × 15 sec 95°C denaturation, 60 sec 60°C primer annealing/elongation. The fluorescence was recorded during the annealing/elongation step in each cycle. A melting curve analysis was performed at the end of each PCR by gradually increasing the temperature from 60 to 95°C while recording the fluorescence. A single peak at the melting temperature of the PCR-product confirmed primer specificity. To be able to compare between different runs, we used a fixed fluorescence threshold for derivation of the Cq value for all runs. We performed three technical replicates for each of three biological replicates per clone/treatment combination to evaluate the relative quantification approaches.
Expression stability of the reference genes was tested by calculating the standard deviations of the Cq for each gene between all treatments . To confirm expression stability, it was also tested by a second approach which uses a regression of Cq against all sample types , and additionally, the NormFinder application for Microsoft Excel was used . Dilution series for all primer/sample combinations were prepared to evaluate amplification efficiencies. For relative quantification of our gene expression data, we tested five widely used approaches.
The comparative Cq method is based on the differences in Cq between target and reference genes and normalizes gene expression to a calibrator sample . We tested for differences in amplification efficiencies of target and reference genes by producing a cDNA dilution series and plotting the ΔCq (Cq reference gene - Cqtarget gene) against the dilution, which should result in a slope of the regression line of close to zero. In a second step, the difference between the samples and a calibrator ΔΔCq (ΔCq sample - ΔCq calibrator) was calculated and used to determine the relative expression rate (r). In an ideal reaction, the amplification efficiency is close to one, which leads to the equation r = 2-ΔΔCq for relative quantification of gene expression. qBase software depends on the same principle as the comparative Cq method, but allows to include correction for amplification efficiencies and multiple reference genes for normalization . For quantification with qBase we used the mean amplification efficiencies calculated by LinRegPCR .
Pfaffl presented a mathematical model which also takes into account that the amplification of different genes may have different efficiencies . Amplification efficiency (E) was calculated from the plot of the Cq values against cDNA input according to the equation E = 10(-1/slope) . The relative expression ratio of a target gene in comparison with a reference gene was calculated according to the equation r = EtargetΔCq target (calibrator - sample)/EreferenceΔCq reference (calibrator - sample).
In the method presented by Liu and Saint the amplification efficiency is determined from the increase in fluorescence (R) of each individual reaction during PCR . We selected two arbitrary thresholds (RA and RB) within the exponential phase of the PCR curve from a graph of the amplification plots (log fluorescence against cycle number). Efficiency was determined according to the equation E = (RB/RA)1/(CqB-CqA) - 1. For quantification of gene expression we used two approaches: first, we used individual amplification efficiencies for each sample, and second, we used the average efficiency for each amplicon. The starting content of the target sequence (R0) was calculated using the equation R0 = RCq/(E+1)Cq and normalized to the starting amount of the reference gene.
With LinRegPCR software we determined gene expression from the slopes of the amplification curves . The data were log-transformed and a regression line determined from the log-linear phase of PCR. The starting concentration of the template was then directly computed from the intercept of the regression line.
With all quantification approaches we quantified a cDNA dilution series with known relative concentrations and compared the results with the expected values .
One-way ANOVA was applied using SPSS 16.0 for Windows statistical software package (SPSS Schweiz AG, Zurich, Switzerland) to determine differences in gene expression between clones and treatments. All results are presented as mean values ± standard errors.
The work was funded by the Swiss Secretariat for Education and Research, COST Action E38 (woody root processes, Grant No. C04.0256).
- Gibson UEM, Heid CA, Williams PM: A novel method for real time quantitative RT-PCR. Genome Res. 1996, 6: 995-1001. 10.1101/gr.6.10.995.View ArticlePubMedGoogle Scholar
- Heid CA, Stevens J, Livak KJ, Williams PM: Real time quantitative PCR. Genome Res. 1996, 6: 986-994. 10.1101/gr.6.10.986.View ArticlePubMedGoogle Scholar
- Tichopad A, Dilger M, Schwarz G, Pfaffl MW: Standardized determination of real-time PCR efficiency from a single reaction set-up. Nucleic Acids Res. 2003, 31: e122-10.1093/nar/gng122.View ArticlePubMedPubMed CentralGoogle Scholar
- Wong ML, Medrano JF: Real-time PCR for mRNA quantitation. BioTechniques. 2005, 39: 75-85. 10.2144/05391RV01.View ArticlePubMedGoogle Scholar
- Brunner AM, Yakovlev IA, Strauss SH: Validating internal controls for quantitative plant gene expression studies. BMC Plant Biol. 2004, 4: 14-10.1186/1471-2229-4-14.View ArticlePubMedPubMed CentralGoogle Scholar
- Reid KE, Olsson N, Schlosser J, Peng F, Lund ST: An optimized grapevine RNA isolation procedure and statistical determination of reference genes for real-time RT-PCR during berry development. BMC Plant Biol. 2006, 6: 27-10.1186/1471-2229-6-27.View ArticlePubMedPubMed CentralGoogle Scholar
- Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2-ΔΔCtmethod. Methods. 2001, 25: 402-408. 10.1006/meth.2001.1262.View ArticlePubMedGoogle Scholar
- Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001, 29: e45-10.1093/nar/29.9.e45.View ArticlePubMedPubMed CentralGoogle Scholar
- Hellemans J, Mortier G, De Paepe A, Speleman F, Vandesompele J: qBase relative quantification framework and software for management and automated analysis of real-time quantitative PCR. Genome Biol. 2007, 8: R19-10.1186/gb-2007-8-2-r19.View ArticlePubMedPubMed CentralGoogle Scholar
- Rasmussen R: Quantification on the LightCycler. Rapid Cycle Real-time PCR, Methods and Applications. Edited by: Meuer S, Wittwer C Nakagawara K. 2001, Heidelberg: Springer, 21-34.View ArticleGoogle Scholar
- Liu W, Saint DA: A new quantitative method of real-time reverse transcription polymerase chain reaction assay based on simulation of polymerase chain reaction kinetics. Anal Biochem. 2002, 302: 52-59. 10.1006/abio.2001.5530.View ArticlePubMedGoogle Scholar
- Ruijter JM, Ramakers C, Hoogaars WMH, Karlen Y, Bakker O, van den Hoff MJB, Moorman AFM: Amplification efficiency: linking baseline and bias in the analysis of quantitative PCR data. Nucleic Acids Res. 2009, 37: e45-10.1093/nar/gkp045.View ArticlePubMedPubMed CentralGoogle Scholar
- Cocozza C, Cherubini P, Regier N, Saurer M, Frey B, Tognetti R: Early effects of water deficit on two parental clones of Populus nigra grown under different environmental conditions. Funct Plant Biol. 2010, 37: 244-254. 10.1071/FP09156.View ArticleGoogle Scholar
- Regier N, Streb S, Cocozza C, Schaub M, Cherubini P, Zeeman SC, Frey B: Drought tolerance of two black poplar (Populus nigra L.) clones: contribution of carbohydrates and oxidative stress defence. Plant Cell Environ. 2009, 32: 1724-1736. 10.1111/j.1365-3040.2009.02030.x.View ArticlePubMedGoogle Scholar
- Andersen CL, Jensen JL, Ørntoft 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 Res. 2004, 64: 5245-5250. 10.1158/0008-5472.CAN-04-0496.View ArticlePubMedGoogle Scholar
- Hong SY, Seo PJ, Yang MS, Xiang F, Park CM: Exploring valid reference genes for gene expression studies in Brachypodium distachyon by real-time PCR. BMC Plant Biol. 2008, 8: 112-10.1186/1471-2229-8-112.View ArticlePubMedPubMed CentralGoogle Scholar
- Hu R, Fan C, Li H, Zhang Q, Fu YF: Evaluation of putative reference genes for gene expression normalization in soybean by quantitative real-time RT-PCR. BMC Mol Biol. 2009, 10: 93-10.1186/1471-2199-10-93.View ArticlePubMedPubMed CentralGoogle Scholar
- Remans T, Smeets K, Opdenakker K, Mathijsen D, Vangronsveld J, Cuypers A: Normalisation of real-time RT-PCR gene expression measurements in Arabidopsis thaliana exposed to increased metal concentrations. Planta. 2008, 227: 1343-1349. 10.1007/s00425-008-0706-4.View ArticlePubMedGoogle Scholar
- Ramakers C, Ruijter JM, Lekanne Deprez RH, Moorman AFM: Assumption-free analysis of quantitative real-time polymerase chain reaction (PCR) data. Neurosci Lett. 2003, 339: 62-66. 10.1016/S0304-3940(02)01423-4.View ArticlePubMedGoogle Scholar
- Karlen Y, McNair A, Perseguers S, Mazza C, Mermod N: Statistical significance of quantitative PCR. BMC Bioinformatics. 2007, 8: 131-10.1186/1471-2105-8-131.View ArticlePubMedPubMed CentralGoogle Scholar
- Nordgård O, Kvaløy JT, Farmen RK, Heikkilä R: Error propagation in relative real-time reverse transcription polymerase chain reaction quantification models: The balance between accuracy and precision. Anal Biochem. 2006, 356: 182-193. 10.1016/j.ab.2006.06.020.View ArticlePubMedGoogle Scholar
- Rieu I, Powers SJ: Real-time quantitative RT-PCR: Design, calculations, and statistics. Plant Cell. 2009, 21: 1031-1033. 10.1105/tpc.109.066001.View ArticlePubMedPubMed CentralGoogle Scholar
- Cook P, Fu C, Hickey M, Han ES, Miller KS: SAS programs for real-time RT-PCR having multiple independent samples. BioTechniques. 2004, 37: 990-995.PubMedGoogle Scholar
- Peirson SN, Butler JN, Foster RG: Experimental validation of novel and conventional approaches to quantitative real-time PCR data. Nucleic Acids Res. 2003, 31: e73-10.1093/nar/gng073.View ArticlePubMedPubMed CentralGoogle Scholar
- Boyle B, Dallaire N, MacKay J: Evaluation of the impact of single nucleotide polymorphisms and primer mismatches on quantitative PCR. BMC Biotechnol. 2009, 9: 75-10.1186/1472-6750-9-75.View ArticlePubMedPubMed CentralGoogle Scholar
- Rozen S, Skaletzki H: Primer3 on the www for general users and for biologist programmers. Bioinformatics Methods and Protocols: Methods in Molecular Biology. Edited by: Krawetz S, Misener S. 2000, Totowa: Humana Press, 365-386.Google Scholar
- Tuskan GA, DiFazio S, Jansson S, Bohlmann J, Grigoriev I, Hellsten U, Putnam N, Ralph S, Rombauts S, Salamov A, Schein J, Sterck L, Aerts A, Bhalerao RR, Bhalerao RP, Blaudez D, Boerjan W, Brun A, Brunner A, Busov V, Campbell M, Carlson J, Chalot M, Chapman J, Chen GL, Cooper D, Coutinho PM, Couturier J, Covert S, Cronk Q, Cunningham R, Davis J, Degroeve S, Déjardin A, dePamphilis C, Detter J, Dirks B, Dubchak I, Duplessis , Ehlting SJ, Ellis B, Gendler K, Goodstein D, Gribskov M, Grimwood J, Groover A, Gunter L, Hamberger B, Heinze B, Helariutta Y, Henrissat B, Holligan D, Holt R, Huang W, Islam-Faridi N, Jones S, Jones-Rhoades M, Jorgensen R, Joshi C, Kangasjarvi J, Karlsson J, Kelleher C, Kirkpatrick R, Kirst M, Kohler A, Kalluri U, Larimer F, Leebens-Mack J, Leplè JC, Locascio P, Lou Y, Lucas S, Martin F, Montanini B, Napoli C, Nelson R, Nelson C, Nieminen K, Nilsson O, Pereda V, Peter G, Philippe R, Pilate G, Poliakov A, Razumovskaya J, Richardson P, Rinaldi C, Ritland K, Rouzè P, Ryaboy D, Schmutz J, Schrader J, Segerman B, Shin H, Siddiqui A, Sterky F, Terry A, Tsai CJ, Uberbacher E, Unneberg P, Vahala J, Wall K, Wessler S, Yang G, Yin T, Douglas C, Marra M, Sandberg G, Van de Peer Y, Rokhsar D: The genome of black cottonwood, Populus trichocarpa (Torr. & Gray). Science. 2006, 313: 1596-1604. 10.1126/science.1128691.View ArticlePubMedGoogle Scholar