GENE QUANTIFICATION EUROPE CONFERENCE

COMPARING DIFFERENT RT-PCR APPROACHES TO

MEASURE SPECIFIC CELLULAR TRANSCRIPTS
Pfaffl M.W.; Rojas. P; Meyer H.H.D. & Einspanier R.
Institut für Physiologie, Forschungszentrum für Milch und Lebensmittel
der TU München-Weihenstephan, 85350 Freising-Weihenstephan

Introduction:
Several techniques are currently available to measure changes in gene
expression. Because of its high sensitivity, reverse-transcription with Fig.1: Quantification in ethidium bromide-stained agarose gel
following polymerase-chain-reaction (RT-PCR) is being increasingly used
to quantify physiologically relevant changes in gene expression. The B
r = 0.984
quantification technique of choice depends on the target sequence, the 600

optical density
expected range of the mRNA amount, the degree of accuracy required, A

and whether quantification needs to be relative or absolute. Herein three

different RT-PCR quantification methods are described and compared: 200

• external standardised RT-PCR with quantification on ethidium

bromide stained gels in combination with densitometry; A
60
B
300 4
pg HAS-cRNA
40 400 4 40 pg HAS- 400

• external standardised RT-PCR with online-detection using cRNA

LightCycler CybrGreen technology. SybrGreen binds to newly

synthesised DNA giving a fluorescence signal once per cycle that is
Fig.2: Online-quantification using LightCycler
proportional to the DNA concentration;
• internal standardised competitive RT-PCR measured after HPLC 1.5
separation and UV detection; A reference cRNA standard mutant is r = 0.996

co-amplified in the same reaction tube with the native mRNA

Log Fl vs. cycle

sequence of interest.
200 pg standard HAS-cRNA
1.2
20 pg “
2 pg “
Quantification methodologies: sample A 6 pg
sample B 18 pg

Quantification of RT-PCR products in ethidium bromide stained gels

could be performed using external cRNA standards. Within a limited 0.7

range of linear detection (4 - 400 pg HAS cRNA; r = 0.984) a 5-fold 20 25 30 cycle no.

difference between two distinct cDNA samples could be detected (fig. 1).
When comparing to a LightCycler quantification an increase of the Fig. 3: LightCycler PCR intra-assay variation
reliability and sensitivity was observed (1 - 1000 pg HAS cRNA; r = - comparison of 3 LightCycler standard curves;
- 500-500.000 IGF-1 plasmid-DNA start molecules; one charge LC premix;
0.996) to later technique (fig. 2). - regression, Pearson correlation coefficient (r) and 99% confidence interval;
- coefficient of variation (CV) of LC-PCR at indicated molecule input;
1e+6

LightCycler online-quantification with an external cRNA standard

curve offers high sensitivity (280 ag IGF-1 cRNA = 1600 molecules), low
intra-assay variation (11.8%; fig. 3) and test linearity over a wide range of 1e+5 all std-curves: r = 0.986
detected molecules

IGF-1 cDNA template input (280 ag - 28 ng; r = 0.992). std-curve 1: r = 0.989

std-curve 2: r = 0.990
std-curve 3: r = 0.981
1e+7
intra-assay over all CV = 11.8%
In internal standardised competitive RT-PCR identical PCR 1e+4 1e+6 13.6% 16.3% 11.4% 5.7%

efficiencies of native mRNA and competitive standard cRNA were

detected molecules

1e+5

confirmed. The validated IGF-1 assay has a detection limit of 1600 IGF-1 1e+4

cRNA molecules/reaction, offers low intra-assay variation (7.4%) and

1e+3 1e+3

linearity is given between 140 - 840 ng total-RNA input (r = 0.997). The

1e+2

comparison of RT-PCR results of individual samples quantified by 500 5000 50000 500000

LightCycler vs. competitive RT-PCR (fig. 4) showed high similarity (r = 5e+2 1e+3 5e+3 1e+4 5e+4 1e+5 5e+5 1e+6
plasmid-DNA molecule input
0.908; n = 30).
Conclusions:
Fig.4: Comparison of two quantitative PCRs
- internal standardised competitive IGF-1 RT-PCR (abscissa)
Absolute quantification of specific transcripts still is a high sophisticated - external standardised LightCycler Online-PCR (ordinate)
- calculated on gramme tissue basis of several bovine tissues
approach. Here we could show that three different RT-PCR techniques 1e+13
possess the following qualities: 1e+12
m. splenius (n = 10)
m. gastrocnemius (n = 10)
IGF-1 mRNA quantified with LC-PCR

• ethidium bromide gels are easiest to perform, but with lowest liver (n = 10)
regression
1e+11
reliability of produced data; 99% confidence interval

1e+10
• internal standardised RT-PCR is a very time consuming and
1e+9 r = 0.908
laborious technique yielding the most precise results;
• online detection during RT-PCR, e.g. using LightCycler, combines 1e+8

the ease and necessary exactness to be able to produce most reliable 1e+7

as well as quick results. 1e+6

1e+5
All three techniques are found to be suitable to detect transcriptional 1e+9 1e+10 1e+11
IGF-1 mRNA quantified with comp. RT-PCR
1e+12 1e+13 1e+14

changes depending on the aims and equipment of the researcher.