Competitive RT-PCR vs. real-time RT-PCR
The theory is straightforward,
but a number of technical caveats are associated with the use of conventional end-point
methodologies for quantitative RT-PCR. In these techniques, PCR results are monitored after
a given number of cycles, by which point factors such as limiting reagent
concentrations and side reactions may have played a significant
role in effecting final product concentration. Quantitative
competitive PCR was developed in response to some of these
difficulties. In this approach, the starting amount of target is calculated based on the ratio
target to competitor after amplification. However, quantitative
Real time PCR or real-time RT-PCR offers numerous advantages over previous attempts at quantitating RT-PCR. Other methods typically rely on end-point measurements, when often the reaction has gone beyond the exponential phase because of limiting reagents. To compensate for such problems, competitive PCR was devised, which allows for normalization of the end product based on the ratio between target and competitor. Because this method is cumbersome, requiring a carefully constructed competitor target for each (RT-)PCR reaction and a series of dilutions to ensure that there is a suitable ratio of target to competitor, it is seldom used successfully. In contrast, with real time (RT-)PCR, the dynamic range is much greater than that of competitive (RT-)PCR - up to 8 orders of magnitude as compared to one with competitive (RT-)PCR -, post-reaction processing is eliminated, and the measurements are taken from the exponential range of the reaction, where component concentrations are not limiting. And best of all, the entire process is automated.
by D Sugden (Endocrinology & Reproduction Research Group, King's College, London)
Workshop 10th November 1999
Competive PCR Guide
by TaKaRa Biochemicals
Principales and Difficulties of competitive PCR
a omparison of primer-dropping, competitive, and real-time RT-PCRs
Wall SJ, Edwards DR.
Anal Biochem 2002 Jan 15;300(2):269-73
School of Biological Sciences, University of East Anglia, Norwich, NR4 7TJ, United Kingdom.
Although all three
types of quantitative RT-PCR produce a similar mRNA expression
profile, the real-time and competitive RT-PCR methods are the most reliable
as they produce
significantly similar results. Of these techniques real-time RT-PCR is
most favorable mainly due to the easier methodology.
Background: Competitive PCR of reverse transcribed mRNA sequences is used to quantify transcripts, but the usual approaches are labor-intensive and time-consuming. We describe the non-gel-based quantification of competitive reverse transcription (RT)-PCR products with use of microparticles and flow cytometry.
Methods: PCR products of a target sequence and an internal control sequence (competitor) were labeled during PCR using digoxigenin (DIG)- and dinitrophenol (DNP)-labeled primer, respectively, allowing specific binding to microparticles coated with the corresponding antibody. Both amplification products were biotinylated to enable fluorescence labeling with streptavidin-R-phycoerythrin. The mean fluorescence intensity of each microparticle population, corresponding to the amount of bound PCR product, was measured in a flow cytometer. We constructed microparticles coated with antibodies against DIG and DNP to specifically capture PCR products derived from target and competitor sequences, respectively.
Results: As required for a reliable competitive PCR assay, nearly identical kinetics were found for the amplification of target and competitor sequences when using only one competitive primer. The method was applied to examine interleukin-8 expression in human lymphocytes after x-irradiation. One hour after irradiation, the concentration of transcripts decreased by half.
Conclusions: The flow cytometric assay for the quantification of competitive RT-PCR products avoids additional hybridization steps and antibody labeling. The use of paramagnetic microparticles would also enable the complete automation of this method.
Development and validation of an internally standardised competitive
Reverse Transcription-Polymerase Chain Reaction (compRT-PCR)
M. Pfaffl; F.
Schwarz & H. Sauerwein (1998)
<>To investigate the role of local IGF-1 mRNA expression in various tissues, we developed and validated a method which allows for a specific, sensitive and reliable quantification of IGF-1 mRNA: an internally standardised Reverse Transcription-Polymerase Chain Reaction (RT-PCR). A synthetic competitive template IGF-1 standard cRNA (IGF-1 cRNA) was designed, which contains the same flanking primer sequences used to amplify the wild type IGF-1 mRNA, but differs by 56 bp in length. To obtain the IGF-1 mRNA concentration present in tissue RNA samples, series of 250 ng total-RNA were spiked with three known quantities of the standard IGF-1 cRNA, incubated for competitive RT-PCR reactions and the two amplificates obtained (184 bp from IGF-1 cRNA and 240 bp from the wild type IGF-1 mRNA) were subsequently separated and quantified by HPLC-UV. For every individual tissue RNA sample, the ratio R (R = competitor PCR product / wild type PCR product) was plotted against the number of starting molecules of the competitor IGF-1 cRNA. The initial amount of IGF-1 mRNA present in the sample can then be read off where R = 1. The validated assay had a detection limit of 1600 IGF-1 cRNA molecules/reaction, the intra-assay variation was 7.4% (n = 5) and linearity (r = 0.997) was given between 140 ng to 840 ng total-RNA input. The present method was first applied to study the effect of long term castration on the IGF-1 expression rates in bovine tissues. The hepatic IGF-1 mRNA concentrations were well correlated (r = 0.81) with the plasma concentrationsas quantified by RIA and were higher in intact than in castrated animals. In two skeletal muscles (m. splenius and m. gastrocnemius) IGF-1 mRNA concentrations were 20- and 35- times lower than in liver, respectively, without any differences between steers and bulls. In bulls, the IGF-1 mRNA expression was higher in m. splenius (p<0.01) than m. gastrocnemius, indicating that locally produced IGF-1 might be important for sexually dimorphic muscle growth patterns.
IGF-1 mediates the anabolic growth hormone actions in skeletal tissues. Above that locally expressed IGF-1 is an important growth regulator acting in auto- and paracrine way (Thissen et al., 1994). To investigate the tissue specific expression in low abundant tissues a method is required which allows for a reliable quantification of IGF-1 mRNA. Considering these limitations, RT-PCR offers the most potent instrument to detect low-abundance mRNAs and the detection limit can been increased up to 1000-fold in comparison to other methods, e.g. Northern hybridisation (Saiki et al., 1988). The relationship between the initial amount A of target mRNA present in the tissues and the amount Yn of DNA produced after n PCR cycles can be expressed asYn = A *(1+E)n
where E is the amplification efficiency of one reaction step (Chelly et al., 1988). Small variations in the reaction efficiency, therefore, translate into large differences in the amount of RT-PCR product generated after n cycles. These limitations in quantitative analyses can be compensated by parallel co-amplification of the native mRNA together with known amounts of an internal standard cRNA. The amplification efficiency should affect both templates similarly. Several designs have been used in quantitative RT-PCR to obtain an internal standard cRNA that suits the characteristics of having an identical amplification efficiency as the wild-type mRNA template and of being easy distinguishable from it (Nedelman et al., 1992). Hereby the construction of an internal standard by inserting (Martini et al., 1995) or deleting (Becker-Andrè and Hahlbrock, 1989; Piatak et al., 1993; Malucelli et al., 1996) a relatively small sequence within the wild type template are common practice. Due to the negative relationship between the efficiency of amplification and the length of the amplified sequences, the both templates should be as short as possible (Rolfs et al., 1992). Analysis and quantification of competitive PCR products can be done either by electrophoretic separation with densitometric quantification or by HPLC and following UV detection at 260 nm. HPLC-UV is the most exact quantification method for PCR products in terms of accuracy, precision and linearity (Katz et al., 1990). In consequence we designed, developed and validated an internally standardised IGF-1 mRNA RT-PCR assay with subsequent HPLC-UV quantification for quantitative comparisons in tissues of low IGF-1 mRNA abundance. The method was first applied to investigate the effect of castration on IGF-1 mRNA expression in bovine liver and two different skeletal muscles.
Establishment and Validation of the quantitative IGF-1 mRNA RT-PCR
Considering the described criteria we designed a short internal standard IGF-1 cRNA, for which the same flanking primers are used as for the wild-type IGF-1 mRNA. The conditions for the RT-PCR as described in Materials and Methods were optimised with regard to PCR buffer pH, primer and MgCl2 concentration in the PCR reaction, dNTPs concentration and annealing temperatures. To ensure a parallel start in all individual reaction tubes and to increase specificity, yield and precision of the PCR, a "hot-start" amplification with a melting wax barrier between RT reagents and PCR master-mix was applied. The quantification of wild-type IGF-1 mRNA in different tissues required a preliminary estimation of the IGF-1 cRNA start-molecule concentration range to be used for individual tissues. This was performed by 7 titration steps from 1.6 * 108 to 1.6 * 1011 cRNA start-molecules versus 250 ng total tissue RNA. For routine comparisons, three standard concentrations covering the range in which equal amounts of the two amplification products are to be expected for a certain tissue were selected.
The amplification efficiencies for the wild-type and the standard template were recorded during the exponential and the plateau phase of the PCR. Figure 3a shows the results of the competitive co-amplification for the two amplificates. Until cycle numbers 23-25 there was an exponential increase in the amount of both products, followed by the plateau phase. In order to compare the amplification efficiencies of target IGF-1 mRNA and standard IGF-1 cRNA, the 10log of the HPLC integrals (10log Yn) was plotted versus the number of PCR cycles (abscissa) and the linear regressions were then calculated for the exponential and the plateau phase (Figure 3b). The relationship Yn = A * (1+E)n, in which E is the amplification efficiency of interest, can be transformed to 10log Yn = n * 10log (1+E) + 10log A, yielding a linear equation: y = x * a + t. The resulting efficiencies of competitive IGF-1 RT-PCR during the exponential phase were nearly identical (Figure 3b; E (IGF-1 cRNA) = 0.66 ; r = 0.98 and E (IGF-1 mRNA) = 0.65 ; r = 0.98). Similarly, during the plateau phase the amplification efficiencies were parallel with E (IGF-1 cRNA) = 0.05 (r = 0.78) and E (IGF-1 mRNA) = 0.04 (r = 0.75). The initial ratio R of the both products remained constant throughout the amplification cycles.
The sensitivity of the RT-PCR was evaluated using different starting amounts of IGF-1 cRNA from 2.8 ag (16 IGF-1 cRNA molecules) to 28 ng (1.6 * 1011 cRNA molecules). The minimal detectable amount of IGF-1 cRNA using the HPLC-UV detection modus was 1600 molecules/tube.
Linearity and variability
The precision of the HPLC-UV quantification of PCR products was initially established by quantifying 28 individual DNA samples at 7 different concentrations from 5 to 325 ng DNA. A linear relationship between the DNA concentration injected (d) onto the DEAE column and the respective peak integral (i) could be demonstrated (i = 1.13 * d + 3.62; r = 0.99). The linearity of the RT-PCR was determined by quantifying the IGF-1 mRNA in serial dilutions of a liver RNA preparation (140, 280, 560 and 840 ng). Each RNA dilution was assayed together with four different IGF-1 cRNA standard concentrations. Figure 4 shows the resulting ratio plots for the four individual RNA input concentrations. The IGF-1 mRNA molecule numbers initially present were read off at R = 1. In Figure 5 the amount of IGF-1 mRNA molecules (a) measured in the different RNA dilutions is plotted versus the total-RNA input (t) into the RT-PCR assay. A linear relationship between the amount of analyte and the measured IGF-1 mRNA concentration could thus be demonstrated (a = 2.0 * 107 * t + 5.4 * 106; r = 0.997). To confirm the reproducibility of the competitive IGF-1 RT-PCR, the assay variation was determined: five identical RT-PCR experiments were set up; each with three different standard dilutions and 250 ng liver RNA. Quantification resulted in 1.069 ± 0.079 * 109 IGF-1 mRNA molecules (n = 5) and thus in an assay variation of 7.4%.
Quantification of DNA,
total-RNA and IGF-1 mRNA in bovine tissues
Table 1 summarises the DNA concentration, total-RNA concentration and IGF-1 mRNA expression rates in liver and in two different muscles from steers and bulls. No significant changes in total transcriptional activity (total-RNA / DNA) within different tissue types could be observed. Hepatic IGF-1 mRNA concentrations were higher in intact than in castrated males (p < 0.05) and were correlated with the mean IGF-1 plasma concentrations recorded one week before slaughter (Figure 6). In bulls higher (p<0.01) IGF-1 mRNA expression was observed in m. splenius than in m. gastrocnenius. In both muscles the difference between bulls and steers did not reach the level of significance (p = 0.07 for m. splenius and p = 0.22 for m. gastrocnemius, respectively).
RT-PCR is a potent and sensitive methodology to detect low amounts of mRNA molecules and offers important insights into the local expression system in low abundant tissues. Using competitive systems with an internal standard, the limitations of quantitative power can be circumvented (Becker-Andrè and Hahlbrock, 1989; Piatak et al., 1993; Martini et al., 1995). The reliability of this approach depends on the condition of identical amplification efficiencies for both, the wild-type and the standard RNA. Wang et al. (1989) postulated that a reliable quantification of PCR products should remain within the exponential phase for both. In the IGF-1 mRNA quantification system described herein a parallel co-amplification of the two fragments could be substantiated, similarly as reported by Bouaboula et al. (1992) and Zimmermann and Mannhalter (1996) for other mRNA species. As demonstrated herein, the ratio between the two products remained constant throughout the amplification, i.e. during the exponential and the plateau phase. Thus the present IGF-1 mRNA quantification is not necessarily limited to the exponential phase of the reaction. In view of the data provided for sensitivity, linearity and reproducibility, the competitive RT-PCR assay developed herein allows for the absolute and accurate quantification of IGF-1 mRNA molecules with a sufficiently high sensitivity even for tissues or cells with low abundancies or for very small amounts of RNA available. We have first applied this IGF-1 mRNA quantification system to compare the IGF-1 expression rates in bovine tissues, however, the method can not only be applied in ruminant tissues but also for comparisons in other species with sufficientlyhigh homologies of the amplified IGF-1 fragment. Besides bovine tissues, we have successfully applied this method in porcine tissues (Pfaffl et al., 1998). Considering the close relationship between the hepatic IGF-1 mRNA concentrations and the IGF-1 plasma concentrations, a biological parallelism of IGF-1 mRNA transcription rates and IGF-1 protein translation might be postulated. The two muscles investigated were selected according to their overproportional (m. splenius) or underproportional (m. gastrocnemius) growth response to testicular steroids (Brännäng, 1971). Castration divergently influenced the IGF-1 mRNA expression rates in liver and in skeletal muscles. Comparing the IGF-1 mRNA quantitiesin the two investigated skeletal muscle, we observed higher concentrations in m. splenius than in m. gastrocnemius in bulls. Total transcriptional activity (RNA/DNA ratio) remained constant within all tissues and the differences in IGF-1 mRNA were specific up- or down-regulations of the IGF-1 gene expression. These findings are in accordance with earlier investigations on the higher growth impetus of m. splenius compared to m. gastrocnemius in bulls (Berg and Butterfield, 1976). The molecular basis for this sexually dimorphic muscle growth pattern might be attributed to relativly higher sensitivities to testicular steroids in neck muscle (Sauerwein and Meyer, 1989). Above that, the present study implies that local differences in IGF-1 expression might be one of the mediators of the allometric growth of these individual muscles in intact males.
insulin and thyroid hormone concentrations between
two pig genotypes with markedly divergent growth rates
and the effects of growth hormone treatment.
The intention of the current study was to gain more insight into the endocrine and molecular controlmechanisms of growth in the pig. For this purpose various growth related parameters were determinedin 4-month-old barrows of two extreme pig genotyes, the small, obese Göttingen Miniature (GM) and thelarge and lean German Landrace (DL). Mean growth hormone (GH) concentration, GH pulse frequencyand GH pulse amplitude did not differ between breeds. Likewise, plasma IGF-1, thyroxine,tri-iodothyronine (T3) concentrations were similar in both breeds. However the plasma GH response(maximum level and area under curve) to a single i.v. injection of GHRH in DL was higher than in GM(P < 0.05). Furthermore, basal plasma insulin and in particular plasma cortisol concentrations were higher in GM compared with DL pigs (P < 0.05 and <0.01 respectively). Analysis of cortisol during 4-hfrequent blood sampling indicated higher cortisol amplitudes in GM compared with DL (P < 0.01).Specific bGH-binding to hepatic membrane preparations was not different between breeds and IGF-1 mRNA concentrations determined by reverse transcription-polymerase chain reaction in liver, m.semimenbranosus and m. longissimus dorsi were similar in both breeds. I.m. treatment with recombinant porcine somatotropin (rpST; 70 mg/ kg live weight) over an 8-day period in contemporary barrowsincreased without any breed difference, plasma IGF-1, T3 and insulin concentrations and hepaticspecific bGH-binding, but did not affect thyroxine or cortisol concentrations in plasma. IGF-1 geneexpression was also elevated in liver and muscle tissues in rpST-treated animals without obvious breedeffects. The observations underline the complexity of the hormonal and molecular control of growth andsupport the notion that differences in growth potential are the consequence of differences at variouslevels of the somatotropic axis and apparently relate to differences in other control systems of energymetabolism such as the pituitary adrenal axis or the endocrine pancreas as well.
by competitive co-amplification of a template
in reverse transcription-polymerase chain reaction.
Sauerwein H, Pfaffl MW, Meyer HHD.
We describe a polymerase chain reaction (PCR)-based method for the quantification of androgen receptor (AR) mRNA in tissues. The amount of PCR products depends on the exponential amplification of the initial cDNA copy number; therefore minor differences in the efficiency of amplification may dramatically influence the final product yield. To overcome these tube-to-tube differences in reaction efficiency, an internal control AR cRNA was reverse transcribed along with the target mRNA using the same primers. This standard was obtained by deleting a 38 bp fragment from an amplified bovine AR sequence, which was then subcloned and transcribed into cRNA. Known dilutions of the competitor cRNA were spiked into a series of RT-PCR reaction tubes containing equal amounts of the target mRNA. Following RT-PCR, the co-amplified specimens obtained were separated by gel electrophoresis and quantified by densitometric analysis of ethidium bromide stain. We applied this method to quantify the AR-mRNA in skeletal muscle of castrated as well as from intact male cattle. The applicability of the quantification system for AR-mRNA described herein was demonstrated for other species, e.g. man.Effects of muscle type, castration, age, and compensatory growth rate on androgen receptor mRNA expression
in bovine skeletal muscle.
Pfaffl MW, Hocquette JF, Gerrard DE, Picard B, Geay Y, Sauerwein H.
The effect of testosterone on sexual dimorphism is evident by differential growth of forelimb and neck muscles in bulls and steers. Divergent hormone sensitivites may account for the differential growth rates of individual muscles. Therefore, the objective of this study was to compare androgen receptor (AR) expression in three different muscles of bulls and steers at various ages and growth rates. Thirty Montbeliard bulls and 30 steers were assigned to four slaughter age groups. Four or five animals of each sex were slaughtered at 4 and 8 mo of age. Animals in the remaining two slaughter groups (12 and 16 mo) were divided into groups of either restricted (R) or ad libitum (AL) access to feed. Five animals of each sex and diet were slaughtered at the end of the restricted intake period at 12 mo of age. To simulate compensatory growth, the remaining animals (R and AL) were allowed ad libitum access to feed until slaughter at 16 mo of age. Total RNA was extracted from samples of semitendinosus (ST), triceps brachii (TB), and splenius (SP) muscles. Androgen receptor mRNA was quantified in 200-ng total RNA preparations using an internally standardized reverse transcription (RT) PCR assay. Data were analyzed using 18S ribosomal RNA concentrations as a covariable. Steers had higher AR mRNA levels per RNA unit than bulls (P < .01). Androgen receptor mRNA levels differed between muscles (P < .05), with lowest expression in the SP. The pattern of AR expression differed (P < .05) for each muscle with increasing age. Between 4 and 12 mo of age, AR mRNA levels increased (P < .05) in SP but remained unchanged in the ST and TB. Feeding regimen had no effect on muscle AR expression, but steers exhibiting compensatory growth had higher AR mRNA levels than AL steers (P < .01) or bulls (P < .01). Our results show that AR expression is muscle-specific and may be modulated by circulating testicular hormones. These data suggest that the regulation of AR expression may be linked to allometric muscle growth patterns in cattle and compensatory gain in steers.