Post on 12-Sep-2021
Quantification strategies in real-time RT-PCRwith special focus on RNA quality,
relative expression and data normalisation
qPCR Satellite SymposiumBio City Leipzig, 10 - 11th March 2005
PD Dr. Michael W. PfafflPhysiology - WeihenstephanCenter of Life and Food SciencesTechnical University of MunichWeihenstephaner Berg 385350 Freising-Weihenstephan; Germanymichael.pfaffl@wzw.tum.dewww.gene-quantification.info
Genotype -> Phenotype -> Function
DNA => pre-mRNA => mRNA => Protein => Function↓ ↓ ↓
Transcription Splicing Translationpost-translatoric modifications
Genom Transcriptome Proteome Metabolome& Splicome
TranscriptomeRNA content of a cellT.A. Brown, GENOMES 2nd Edition
• ribosomal RNA rRNA 80-85% ( 5S, 18S und 28S )• transfer RNA tRNA 10-15%• messenger RNA mRNA 2% (1-5%) ( ∅ length 1930 bases)
• high abundant < 100 genes > 10,000 copies/cell• intermediate abundant ~ 500 - 1,000 genes 200 - 400 copies/cell• low abundant ~ 27,000 genes < 10 - 50 copies/cell
• RNA quantity & RNA quality
Quantification Strategies in real time qRT-PCRM.W. Pfaffl, BioSpektrum 2004 (Sonderausgabe PCR)
relative quantificationabsolute quantification
external calibration curve
one color detectionsystem
SYBR Green I
external calibration curve
two color detectionsystem
e.g. Probes, ROX
external calibration
curve withoutany referenece
gene
normalisationvia one
reference gene
via reference gene index
>3 HKG
external calibration curve
• RT-PCR product
• plasmid DNA
• in vitro transcribed RNA
• synthetic DNA Oligos
• synthetic RNA Oligos
without real-time PCR efficiency correction
2 (-∆∆ CP) REST, qGeneLC software
with real-time PCR efficiency correction
ROXROX
Quantification strategies in real-time RT-PCR
Absolute quantification using calibration curves
• recombinant DNA (recDNA) calibration curve (Pfaffl & Hageleit, Biotechnol.Lett. 2001)
• recombinant RNA (recRNA) calibration curve (Pfaffl & Hageleit, Biotechnol.Lett. 2001)
• calibration curve using a synthetic DNA oligo-nucleotide (Bustin, JME 2000)
• calibration curve using a synthetic RNA oligo-nucleotide (Bustin et al. 2000)
• calibration curve using a purified RT-PCR product (Einspanier et al. 1999)
• „Copy & Paste“ of previously performed calibration curves (LC software)
• Calibration curve using a purified RT-PCR product or a synthetic ss/ds oligo-nucleotidetwo-step RT-PCRadvantages: quick, highly defined DNA content for the synthetic oligodisadvantages: instable, “often problems with re-amplification”, „short“ templates
Absolute quantification using calibration curves
• Calibration curve using a recombinant DNA (recDNA), e.g. plasmid DNAtwo-step RT-PCRadvantages: very stabile, no problems with re-amplification, „mimic of mRNA“disadvantages: cloning, linearization and purification of recDNA
• Calibration curve using a recombinant RNA (recRNA)one-step RT-PCR => recRNA and native mRNA undergoing RT and PCR in paralleladvantages: mimics the natural mRNA situation best (recRNA = native mRNA) disadvantages: very instable recRNA, complicate cloning, linearization,
purification of recRNA, storage problems, reproducibility (???)=> storage of recRNA !!!
• „Copy & Paste“ of previously performed calibration curves (e.g. LightCycler Software)advantages: very easy and very high reproducibility (at least for the calibration curve)disadvantages: do not covers variations in real RT-PCR experiment: RNA quality, slope,
qPCR efficiency ??? batch to batch variations, etc. => truth ???
ER-alpha intra-assay variation CV = 18.7% (n = 3)
input ss cDNA molecules
1.65 e+91.65 e+81.65 e+71.65 e+61.65 e+51.65 e+41.65 e+3
dete
cted
mol
ecul
es
1e+3
1e+4
1e+5
1e+6
1e+7
1e+8
1e+9
1e+10 31.4% 14.4% 7.3% 7.9% 33.6% 13.7% 22.8%
input ss cDNA molecules
1.65 e+91.65 e+81.65 e+71.65 e+61.65 e+51.65 e+41.65 e+3
dete
cted
mol
ecul
es
1e+3
1e+4
1e+5
1e+6
1e+7
1e+8
1e+9
1e+10 41.2% 31.9% 46.2% 20.6% 15.1% 21.4% 23.8%
ER-alpha inter-assay variation CV = 28.6% (n = 7)
ERα intra-assay & inter-assay variationvariation on the basis of detected molecules using a recombinant plasmid DNA calibration curveintra-assay variation: within one runinter-assay variation: between different runs
Validation: absolute quantification of steroid receptors
25.7% (n = 4)29.7% (n = 4)28.6% (n = 4)24.3% (n = 7)inter-assay variation
83.5[ 82.9 ]
83.983.883.183.5
[ 87.9 ]89.0
[ 89.9 ]90.190.590.2
86.085.0
--85.385.486.0
85.484.485.085.5
--84.5
Species specific Tmelt (°C)Homo sapiens
Rattus norvegicusCallithrix jacchus (primate)
Bos taurusOvis ariesSus scrofa
5.7% (n = 4)17.6% (n = 4)18.7% (n = 4)31.2% (n = 3)intra-assay variation
93.9%81.3%81.2%90.7%PCR efficiency
760 – 7.60*109
molecules(r = 0.998)
106 - 1.06*1010
molecules(r = 0.996)
165 - 1.65*109
molecules(r = 0.995)
120 - 1.20*1010
molecules(r = 0.998)
quantification range(test linearity)
760 molecules106 molecules165 molecules120 moleculesquantification limit
14 molecules10 molecules2 molecules12 moleculesdetection limit227 bp262 bp234 bp172 bpproduct length
PRERβERαAR
ER-alpha [molecules/25 ng RNA]1e+3 1e+4 1e+5 1e+6
ER
-bet
a [m
olec
ules
/25
ng R
NA
]
1e+2
1e+3
1e+4
1e+5
Col 22 vs Col 26: 977883,75
Col 22 vs Col 26: 201818,75
Col 22 vs Col 26: 5421,71
Col 22 vs Col 26: 79348,75
Col 22 vs Col 26: 100003,75
Col 22 vs Col 26: 60479,88
Col 22 vs Col 26: 145339,88
Col 22 vs Col 26: 6035,36
Col 22 vs Col 26: 12027,5
Col 22 vs Col 26: 204646,5
Col 22 vs Col 26: 4061,57
Col 22 vs Col 26: 4591,62
Col 22 vs Col 26: 1535,57
Col 22 vs Col 26: 10212,38
Col 22 vs Col 26: 4160,5
ratio = 1
uterusliverlunglongissimus dorsihind leg musclesshoulder musclesneck musclesheartspleenmammary glandrumenabomasumjejunumkidney medullakidney cortex
Estrogen receptors (ERα & ERβ) expression pattern in cattle tissues
Pfaffl et al., APMIS 2001
Relative QuantificationThe mRNA expression is relative to WHAT ???
• relative to a non treated control
• relative to a time point zero
• relative to another gene of interest
• relative to the mean expression of a target gene
• relative to an universal calibration curve
• relative to the expression of one constant expressed HKGGAPDH, tubulins, various actins, albumins, cyclophilin, micro-globulins, histone subunits, ribosomal units (18S or 28S rRNA), ………….etc.
• relative to a HKG Index containing more HKGs (> 3)geNorm (Vandesompele et al., Genome Biology, 2002)BestKeeper (Pfaffl et al.; Biotechnology Letters 2004)Normfinder (Andersen et al., Cancer Research 2004)statistical modeling (Szabo et al., Genome Biology 2004)
• etc. ???
Normalisation strategiesAccording to known amounts of extracted RNA
(RIN quality check, molecules/ng RNA; ag transcript/ng RNA)
According to mass or volume of extracted tissue(molecules/mg tissue; ag transcript/mg tissue; transcript/cells)
According to one known and NOT regulated HKGGAPDH, tubulins, actins, albumins, cyclophilin, micro-globulins, histone subunits, ribosomal units (18S or 28S rRNA),.............
According to a HKG Index containing more HKGs (> 3)geNorm (Vandesompele et al. 2002, Genome Biology)Accurate normalization of real-time quantitative RT-PCR data by geometric veraging of multiple internal control genes.
BestKeeper (Pfaffl et al. 2004; Biotechnology Letters 2004)Determination of most stable housekeeping genes, differentially regulated target genes and sample integrity: BestKeeper Excel based tool using pair-wise correlations.
Normfinder (Andersen et al., 2004 Genome Biology)Normalization of Real-Time Quantitative Reverse Transcription-PCR Data: A Model-Based Variance Estimation Approach to Identify Genes Suited for Normalization.
Statistical modeling for selecting housekeeper genes(Szabo et al. 2004, Genome Biology)
Relative Quantification in real time qRT-PCR
without real-time PCR efficiency correction
2 (-∆∆ CP)
relative quantification
external calibration
curve withoutany referenece
gene
normalisationvia one
reference gene
via reference gene index
>3 HKG
ROX
“Delta-delta method” for comparing relative expression results between treatments in real-time PCR
presented by PE Applied Biosystems (Perkin Elmer, Forster City, CA, USA)
ABI Prism 7700 Sequence detection System User Bulletin #2 (2001)
Relative quantification of gene expression.http://docs.appliedbiosystems.com/pebiodocs/04303859.pdf
expression ratio = 2
- [ ∆CP sample - ∆CP control]
- ∆∆CPexpression ratio = 2
Livak KJ, Schmittgen TD. (2001) Analysis of Relative Gene Expression Data Using Real-Time Quantitative PCR and the 2 (- delta deltaC(T)) method.Methods. 2001 25(4): 402-408.
“Delta-delta method” for comparing relative expression results between treatments in real-time PCR
presented by ABI (Applied Biosystems Inc.)
assumptions: E = 2 & ∆CP is constant over a wide range
expression relative to the time point zero & normalised by a HKG:TNFα mRNA expression in cultured leukocytes after LPS stimulation:
white blood cells [WBC] vs. somatic milk cells [SMC] isolated blood monocytes vs. isolated milk macrophages
PCR Efficiency = 2 n = 6 mean ± sem
TNFalpha
Time[h]
0 2 4 6 8
TNFa
lpha
[del
taC
P]
0
2
4
6
8 WBCSomatic Cells
***
***
******
***
+++
++
+
+
******
***
***
*
+++
+++
TNFalpha
Time[h]
0 2 4 6 8
TNFa
lpha
[del
taC
P]
0
2
4
6
8 MonocytesMacrophages
*** +***
++ ******
+++***
***+++
***
+++ ***
+++
Prgomet et al., 2004
Relative Quantification in real time qRT-PCR
without real-time PCR efficiency correction
2 (-∆∆ CP) REST, qGeneLC software
with real-time PCR efficiency correction
relative quantification
external calibration
curve withoutany referenece
gene
normalisationvia one
reference gene
via reference gene index
>3 HKG
ROX
Tissue “background” interfere with real-time PCR efficiency and amplification fidelity
IGF-1 mRNA amplification in three cattle tissues
liver
m. splenius m. gastrocnemius
Efficiency variation in real-time RT-PCR
Roche Diagnostics, LC rel. Quantification software, March 2001
PCR inhibitors:Hemoglobin, Urea, Heparin
Organic or phenolic compoundsGlycogen, Fats, Ca2+
Tissue matrix effectsLaboratory items, powder, etc.
PCR enhancers:DMSO, Glycerol, BSA
Formamide, PEG, TMANO, TMAC etc.Special commercial enhancers:
Gene 32 protein, Perfect Match, Taq Extender, AccuPrime, E. Coli ss DNA binding
real-time PCRefficiency
NA degradation = RIN
PCR reaction components
DNA concentration
tissue degradation
unspecificPCR products
hardware:PCR platform & cups
lab management DNA dyes cycle conditions
Relative quantification of a target gene versus areference gene (housekeeping gene)
Single data (n = 1) e.g. array results:
Etarget∆CPtarget (control - sample)
relativeexpression =
Ereference∆CPref (control - sample)
Pfaffl, Nucleic Acids Research 2001
relativeexpression
( Eref )CPSample
( Etarget )CPSample
÷( Eref )
Calibrator
( Etarget )CPCalibrator
=
Roche Diagnostics, LC relative Quantification software, March 2001
CP
Relative quantification of a target gene versus areference gene (housekeeping gene)
single data (n = 1) e.g. array results:
Etarget∆CPtarget (control - sample)
relativeexpression =
Eref∆CPref (control - sample)
Pfaffl, Nucleic Acids Research 2001
multiple data (1 < n < 100) e.g. experimental groups via REST-XL©:
Etarget∆CPtarget (MEAN control – MEAN sample)
relativeexpression =
Eref∆CPref (MEAN control – MEAN sample)
Pfaffl et al., Nucleic Acids Research 2002
Influence of total RNA quality, quantity and purity on qRT-PCR resultstotal RNA extracted bovine WBC analyzed in Bioanalyzer 2100
ladder RIN: 9.5
RIN: 5.6 RIN: 2.8
V. Walf, S. Huch, MW. Pfaffl, 2005
RIN in different bovine tissues (Box Plot)R
IN
0
2
4
6
8
10
liver (n = 22)heart (n = 17)spleen (n = 17)lung (n = 22)rumen (n = 23)reticulum (n = 26)omasum (n = 17)abomasum (n = 17)ileum (n = 17)jejunum (n = 20)colon (n = 19)caecum (n = 16)mes. lymphnode (n = 26)
S. Fleige & MW. Pfaffl, 2005
RIN liver 1st and 2nd extraction
0
2
4
6
8
10
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
sample number
RIN
Leber 1.Extraktion Leber 2.Extrakion
Enzymatic or UV degradation of tissue extracted total RNA
RIN 10.0 9.2 9.2 8.6 8.1 7.8 7.2 6.7 6.0 5.0 4.4 4.0
Influence of total RNA quality on qRT-PCR results
y = -0,206x + 7,6396R2 = 0,5363
y = -0,887x + 21,398R2 = 0,7165
y = -0,7211x + 27,077R2 = 0,7682
0,0
5,0
10,0
15,0
20,0
25,0
30,0
0 1 2 3 4 5 6 7 8 9 10
RNA Integrity Number
CP
= T
ake
Off
Po
int
28S
beta actin
IL-1
bovine ileum: CP = TOP
y = -0,2036x + 8,6557R2 = 0,7612
y = -2E-15x + 6,00R2 = 1,0
y = -0,1461x + 13,565R2 = 0,4903
R2 = 0,8119
0,0
5,0
10,0
15,0
20,0
25,0
0 1 2 3 4 5 6 7 8 9 10
RNA Integrity Number
CP
= Ta
ke O
ff Po
int
18S 28S beta-actin IL-1 Linear (18S) Linear (28S) Linear (beta-actin) Linear (IL-1)
18S
28S
beta actin
IL-1
CP = constant in relative quantification
y = -0,2583x + 21,787∆
28S n.s.Interleukin-1beta P < 0.001
beta-Actin P < 0.0118S P < 0.001
Influence of total RNA quality on qRT-PCR CP (Ct)IL-1: Crossing Point
14,0
16,0
18,0
20,0
22,0
24,0
26,0
28,0
30,0
32,0
34,0
0 1 2 3 4 5 6 7 8 9 10RNA Integrity Number
Cro
ssin
g P
oint
Reticulum (E) Lymph nodes (E) Lymph nodes (P) Colon (P) Lung (E)Corpus luteum (P) Caecum (P) Spleen (P) Abomasum (P)
Influence of total RNA quality on qRT-PCR efficiency
28S: Amplification
1,5
1,6
1,7
1,8
1,9
2,0
0 1 2 3 4 5 6 7 8 9 10RNA Integrity Number
PCR
Eff
icie
ncy
Reticulum (E) Lymph nodes (E) Lymph nodes (P) Colon (P) Lung (E)Corpus luteum (P) Caecum (P) Spleen (P) Abomasum (P)
Advantages:• each sample derived from identical total RNA (one-step quantification) or cDNA
source (two-step quantification) = identical inhibitors and enhancers• tissue specific matrix effects are given, but not existent in relative quantification• total RNA quantity, RT efficiency and cDNA quantity are minor relevant • effect of total RNA quality and influence on qRT-PCR have to be tested in detail• optimized assays are highly reproducibility (variations only given by researcher, kit,
robot and real-time platform)• each sample will be normalised with HKG or better by a HKG INDEX• no production of a calibration curve: e.g. cloning, linearization or purification of
calibration curve material• no data conversion: molecules, concentrations or molarities• => direct procedure: measured data were inserted in mathematical model• => relative quantification by software applications, e.g. qGene, REST, LightCycler• => automatic calculation => bio-informatics session
Disadvantages:• No information about the effective molecule concentration present in the tissues
(molecules/mg tissue; molecules/ng RNA; molecules /cell, etc.)
Relative quantification using efficiency corrected calculation