Background
Cytokines are a family of low-molecular weight proteins secreted by various cell types, with pleiotropic functions and constitute a tightly regulated network that plays a central role in the immune system. Cytokines, classified into different groups such as interleukins (IL), interferons (IFN), colony-stimulating factors (CSF), tumour necrosis factors (TNF), tumour growth factors (TGF) and chemokines are implicated in the differentiation, proliferation, migration and effector functions of immune cells. Interacting one with the others, they have polarizing effects on the target cells and are pivotal in tuning immune responses [
1]. Therefore, it is rather the make-up of cytokines milieu that influences the immune response rather than the action of a single cytokine. Numerous studies indicate that the clinical and/or immunological status depends on the balance between pro-inflammatory cytokines and their regulatory counterparts [
2]. Thus, cytokine profiling should be achieved through analysis of simultaneous quantification of a pattern of cytokines including pro and anti-inflammatory cytokines [
2,
3]. Moreover, recent reports have highlighted the need for clinical immuno-monitoring of patients to adapt treatment or prevent relapses [
4‐
6]. Thus, analysis of the cytokine pattern is central not only in the definition of the immunological status of patients but also in the study of the pathophysiological pathways as well as the cellular subpopulations involved [
7,
8].
Cytokines are often produced locally so that the concentration of circulating cytokines in the plasma is usually low. Their half-life and turnover may vary complicating the delineation of informative cytokine profiles. Although transcription of messenger RNA is not strictly correlated to protein secretion and activity, detection of cytokine RNA by real time PCR is now considered a reference technique for analysis of small-size samples with high sensitivity [
9]. It can be used on its own or to validate and complement information obtained with other techniques such as micro-arrays [
10,
11].
The already available techniques, which offer a so-called "absolute quantification" of the target cytokine mRNA, achieve quantification by reference to an external standard curve based on serial dilutions of a known amount of the corresponding cDNA [
12]. Moreover, to allow for comparison between experiments, data are normalized by reference to an internal standard, which is an endogenous gene for which the number of copy per cell is supposed constant under different experimental conditions [
13,
14]. The term of "absolute" quantification is not completely appropriate since these techniques neither control for the variable efficiency of the RT step nor take it into account in their measurements [
15,
16].
In the present study, we first show that the efficiency of the RT step depends on the target mRNA and on the experiments and that these variations have critical impact on the reliability of mRNA quantification. To overcome this, we describe here CyProQuant-PCR, a new technique for absolute measurement of cytokine mRNA based on an external RNA standard curve. Primer pairs have been designed for allowing amplification of a set of cytokine mRNA using the same conditions both in terms of thermocycling parameters and master mix components, a prerequisite for multiple cytokine mRNA measurements with high throughput.
In the present paper, we describe i) the construction of the synthetic RNA standard, ii) the primer pairs specific for the following human cytokines IL-1β, IL-4, IL-10, IL-12p40, IL-13, IL-15, IL-18, IFN-γ, and for the tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, zeta polypeptide (YWHAZ), the β2-microglobulin (β2-MG) and the ubiquitin-C (UBC) to be used as internal standards and iii) the conditions for efficient real time amplification of multiple cytokine specific mRNA. The technique was validated using in vitro stimulated PBMC and its intra and inter-experimental variability were assessed. Finally, CyProQuant-PCR was used to quantify TGF-β1 transcripts in small blood samples from children with acute Plasmodium falciparum malaria.
Discussion
In this paper we have described a new technique, CyProQuant-PCR, for absolute quantitative profiling of human cytokine mRNA using real time RT-PCR. Although real time PCR is now becoming a popular technique, it still requires improvement for proper quantification. All the techniques for absolute quantification available so far use a DNA standard curve assuming that RT efficiency is constant and approaches 100% [
12]. We show here that RT is not 100% efficient, and more importantly, that its efficiency changes from one gene to another and from one experiment to another. We thus designed primers and standard RNA to be used in such a way that all cytokine mRNA of interest as well as three housekeeping genes could be amplified using the same conditions (thermocycling parameters and buffer). This technique is rapid, reliable and reproducible. When compared to the TaqMan
® commercial kit, which represents the "gold standard", CyProQuant-PCR was as sensitive but less expensive and more flexible. Indeed, for a given gene, the design of the probe might be quite difficult and, with judicious selection of primer pairs, comparable sensitivity can be achieved with the use of SYBR Green [
20].
A major reason for not using an RNA standard curve is its poor stability due to its sensitivity to RNase degradation. In our hands, we did not find any detectable degradation when keeping the standard in concentrated aliquots (stock solution at 1000 μg/mL) in RNase-free water at minus 80°C and avoiding freeze-thawing cycles. However, for a greater stability, we incorporated modified dNTP, 2'-Fluoro-dCTP and 2'-Fluoro-dUTP, which have been reported to decrease the sensitivity of the
in vitro transcribed RNA to specific RNases [
21,
22], and tested the effect of this incorporation on the RT-PCR efficiency. We did not find significant difference in the efficiencies (100% for the non-modified IL-4 standard versus 104% for the 2'-Fluoro-standard) (data not shown).
The methodology described here was easily and successfully applied to the quantification of several cytokine genes. The quantification is reliable on 7 to 8 logs with a sensitivity ranging from 1000 to 100 copies depending the cytokine. It was first used to determine the magnitude and the kinetics of early induction of cytokines mRNA upon PBMC stimulation using bacterial derived materials. This showed that CyProQuant-PCR is powerful enough to detect early on modest cytokine induction such as IL-4. Such a tool is useful to decipher the kinetics of cytokine response involved in physiopathological pathways but also as a read-out to measure minor immune responses to specific antigens [
23].
We finally applied CyProQuant-PCR to measure the level of TGF-β1 transcripts in asymptomatic controls and malaria patients after LPS stimulation. Interestingly, we observed that cells from malaria patients have a significant higher capacity to respond to LPS compared to controls. The increased TGF-β1 transcript accumulation by patients' cells, suggests that an inadequate production of TGF-β1 might play a role in malaria pathogenesis as already proposed [
24]. Further work is needed to elaborate on this finding. This first study provided the proof that CyProQuant-PCR is readily applicable to small clinical samples from paediatric cases. This opens the possibility to further quantify the cytokine imbalance associated with malaria pathogenesis and generate a disease cytokine signature, a prerequisite for novel therapeutic interventions targeting cytokine gene expression. Areas of application include both infectious and non-infectious diseases, as well as chronic inflammatory diseases such as rheumatoid arthritis or sarcoidosis and acute diseases such as sepsis or malaria. Beside its importance in patient immuno-monitoring, cytokine profiling is also a major tool to study specific immune response against antigens for design and testing of immuno-modulatory drugs or vaccines.
Methods
Healthy human PBMC: isolation and in vitro stimulation
Blood was collected from healthy donors at the French Blood Bank. Peripheral blood mononuclear cells (PBMC) were isolated by density separation over Ficoll Hypaque and washed two times in RPMI 1640 (Gibco BRL, Invitrogen, Cergy Pontoise, France). Cells were re-suspended in RPMI 1640 (Gibco BRL, Invitrogen, Cergy Pontoise, France) supplemented with 2 mM glutamine (Gibco BRL, Invitrogen, Cergy Pontoise, France) and 10% AB+ human serum (French Blood Bank) at 2.106 cells/mL and either used directly for RNA extraction or cultured in duplicate with or without LPS (10 ng/mL, E. Coli O111: B7, Sigma, L'Isle d'Abeau Chesnes, France) or SAC (0,0075%, PanSorbine Cells, Calbiochem, La Jolla, CA, USA). After incubation, cells were washed with PBS and re-suspended in RNA-PLUS (Q-Biogene, Illkirch, France) for RNA isolation.
Malaria patients and asymptomatic controls
Twenty children admitted during the high malaria transmission season of 2001 to the emergency room at the Department of Child Health, Korle-Bu Teaching Hospital, Ghana were included. Five asymptomatic controls matched to patients for age, residence location and time of sample collection were enrolled. The general inclusion and exclusion criteria were as described by Kurtzhals
et al. [
25]. Parents or guardians signed informed consent forms. The study received ethical clearance from The Ethics and Protocol Review Committee at the university of Ghana Medical School and the Ministry of Health. Total cellular RNA was extracted from PBMC recovered from 500 μL of blood following supplier's instructions (RNA PLUS, Q-Biogene, Illkirch, France) after 22 hours of incubation at 37°C, 5% CO2 with or without LPS (10 ng/mL,
E. Coli O111: B7, Sigma, L'Isle d'Abeau Chesnes, France).
RNA isolation
RNA was extracted following supplier's instructions, re-suspended in 60 μL of RNase-free water (Ambion, Huntingdon, UK) and quantified spectrophotometrically at 260 nm.
Primers
Oligonucleotide primers were synthesized at Eurogentec (Saraing, Belgium). To validate primers, a pool of cDNA from healthy human PBMC stimulated for 6 and 12 hours with LPS (10 ng/mL, E. Coli O111: B7, Sigma, L'Isle d'Abeau Chesnes, France) and PHA-L (10 μg/mL, Sigma-Aldrich, Lyon, France) was used. Analysis of the amplicons was assessed by 4% agarose gel electrophoresis and dissociation curve studies using Dissociation Curve Software (Applied Biosystems, Foster City, CA, USA). PCR products were cloned into pCR2.1 vector using Original TA cloning kit (InVitrogen, Cergy Pontoise, France) and sequenced (Genome Express, Meylan, France).
Construction of the external DNA and RNA standards
PCR products generated by each primer pairs were column-purified (Nucleospin, Macherey-Nagel, Hoerdt, France) and quantified spectrophotometrically at 260 nm. The molecular weight of the standard DNA was calculated by N*487-[(N-1)*175] were N is the number of bases composing the standard DNA. Stock solution of 1012 copies of standard DNA /3,85 μL were made in Tris-EDTA buffer (Ambion, Huntingdon, UK), split in single-use aliquots and stored at -80°C in safe-lock tubes. External DNA standard range was made extemporaneously by 1:10 serial dilutions in water.
Gene specific primers were fused on their 5' end to the sequence of the RNA polymerase T7 promoter to generate modified primers. These primers were used to amplify a gene specific PCR product flanked by transcription initiation sites. Five hundred nanograms of this construct were in vitro transcribed (MegaShortScript, Ambion, Huntingdon, UK). The standard RNA generated was purified (MegaClear, Ambion, Huntingdon, UK) and loaded on a 4% agarose gel for electrophoresis. The concentration of the standard RNA was determined spectrophotometrically at 260 nm. The molecular weight of the transcript was calculated by N*500-[(N-1)*175] were N is the number of bases composing the standard RNA. Stock solution of 1012 copies of standard RNA /3,85 μL were made in RNA storage solution (Ambion, Huntingdon, UK), split in single-use aliquots and stored at -80°C in safe-lock tubes. External RNA standard range was made extemporaneously by 1:10 serial dilutions in water.
Reverse transcription
For CyProQuant-PCR assays, 100 ng of total cellular RNA from PBMC and serial dilution of external RNA standard were reverse transcribed simultaneously in a parallel procedure using Reverse Transcription TaqMan reagents (Applied Biosystems, Foster City, CA, USA) on a MasterCycler Gradient (Eppendorf, Le Pecq, France). The final volumes were set at 100 μL for the cellular RNA samples and 50 μL for the external RNA standard range. The thermocycling parameters were as follows: 25°C, 10 min.; 48°C, 60 min. and 95°C, 5 min. cDNA were immediately used for PCR amplification.
Real-time RT-PCR and quantification of transcripts
Reverse-transcribed standard RNA and cellular RNA were amplified simultaneously on the same PCR plate on an ABI Prism 7700 (Applied Biosystems, Foster City, CA, USA). An aliquot of 5 μL of the RT reaction was amplified in duplicate in a final volume of 30 μL of SYBR Green PCR Master mix (Applied Biosystems, Foster City, CA, USA). Thermocycling conditions were 50°C for 2 min., 95°C for 10 min. and 40 cycles of [95°C/15 sec.; 60°C, 1 min]. The sample target RNA copy numbers were calculated using SDS 1.9 Software (Applied Biosystems, Foster City, CA, USA). The baseline fluorescence was set manually to correct for differences in initial cDNA concentration and the threshold was positioned at a fluorescence level that was 10 times higher than the background signal. Target mRNA copy numbers in cellular samples were calculated based on a standard curve generated by SDS 1.9 Software (Applied Biosystems, Foster City, CA, USA) by plotting cycles at threshold (CT) against the logarithmic values of the starting RNA standard copy number.
ELISA tests
TNF-α and MIF secreted proteins were quantified by sandwich ELISA following supplier's instructions (Bio-Source, Clinisciences, Montrouge, France and R&D, Lille, France respectively). Results are expressed as pg/mL for one million living cells.
Statistical analysis
Tests for significance were done using Stata software (Stata Corporation, College Station, Texas, 77845 USA) by Kruskal-Wallis rank test.
Patent application
Results disclosed in this manuscript have been protected in French patent application FR0408645.
Authors' contributions
PB developed the entire technique. IV did the in vitro stimulation experiments. DJ gave technical assistance. SL helped in RNA extraction of malaria samples. JCB introduced PB to molecular biology techniques and provided critical advices. BDA designed and conducted the study that yielded the malaria samples. OMP revised the manuscript and supported the work. CB conceived the strategy and coordinated the study. PB and CB drafted the manuscript. All authors read and approved the final manuscript.