doi:10.1016/j.ab.2005.06.026 
Copyright © 2005 Elsevier Inc. All rights reserved.
Four-color multiplex reverse transcription polymerase chain reaction—Overcoming its limitations
Kent Persson, Keith Hamby and Luis A. Ugozzoli
, 
Gene Expression Division, Bio-Rad Laboratories, Hercules, CA 94547, USA
Received 7 March 2005.
Available online 29 June 2005.
References and further reading may be available for this article. To view references and further reading you must
purchase this article.
Abstract
Quantitative reverse transcription polymerase chain reaction (qRT-PCR) conducted in real time is a powerful tool for measuring messenger RNA (mRNA) levels in biological samples. Multiplex PCR is defined as the simultaneous amplification of two or more DNA (cDNA) targets in a single reaction vessel and may be carried out only using uniquely labeled probes for each target. Up to four genes can be detected in a multiplex 5′ nuclease assay when using the appropriate instrument and the right combination of fluorophores. One of the more important advantages of multiplexing is a reduced sample requirement, which is especially important when sample material is scarce. Additional benefits are saving time on reaction setup and lower cost compared to singleplex reactions. Although multiplexing has several advantages over singleplex qRT-PCR, limited work has been done to show its feasibility. Few publications on four-color multiplex qRT-PCR have been reported, and to our knowledge no work has been done to explore the assay’s limitations. In this paper, we report the first in-depth analysis of a four-gene multiplex qRT-PCR. To achieve a better understanding of the potential limitations of the qRT-PCR assay, we used in vitro transcribed RNA derived from four human genes. To emulate gene expression experiments, we developed a model system in which the in vitro transcripts were spiked with plant total RNA. This model allowed us to develop an artificial system closely resembling differential gene expression levels varying up to a million fold. We identified a single “universal” reaction condition that enabled optimal amplification in real time of up to four genes over a wide range of template concentrations. This study shows that multiplexing is a feasible approach applicable to most qRT-PCR assays performed with total RNA, independent of the expression levels of the genes under scrutiny.
Keywords: Real-time PCR; RT-PCR; PCR; Gene expression
Fig. 1. Schematic illustration of the in vitro transcript RNA assembly sets. To perform qRT-PCR experiments, we generated 15 in vitro assembly sets by mixing predefined amounts (Ct 
15, 25, and 35) of four different transcripts (AZI, OAZ, ADC, and AdoMetDC). The concentration of the transcripts in the assemblies was either equal or differed by up to one millionfold. To emulate a real gene expression assay based on the use of total RNA as a template, the cDNA synthesis reactions were spiked with Arabidopsis total RNA prior to reverse transcription.
Fig. 2. Evaluation of primers–probe pairs. The performance of the primers–probe pairs was analyzed using a fivefold dilution series of human cDNA spanning three orders of magnitude. The performance data were obtained from the standard curve generated by the iCycler iQ software. As an example, the figure shows the evaluation of the AZI primers–probe. Fluorescence (RFU) was measured every cycle and, after normalization, plotted against cycle number. All the primers–probe sets included in our multiplex assay generated PCR efficiency, correlation coefficient, and slope values within our specifications.
 |
Fig. 3. Optimization of multiplex condition with respect to dNTP and Taq polymerase. While analyzing assemblies 1–4 we found that our original multiplex condition could not support multiplexing beyond a 5-Ct difference in template concentration. While more concentrated templates amplified satisfactorily, the template present at the lowest concentration failed to show exponential amplification. Shown here are the multiplex optimization data generated with the least concentrated in vitro transcript included in assembly 4. The plots (A–F) show how different reaction conditions affect the performance of the four-color multiplex assay. Multiplex reactions using 0.075 U/μl of iTaq DNA polymerase and 200 μM dNTPs showed a Ct difference greater than 0.5 U compared to the singleplex reaction. Moreover, the exponential phases of both singleplex and multiplex reactions could not be superimposed (A). B–F show the optimization of the multiplex assay. For the development of a universal condition, both the concentrations of iTaq DNA polymerase and dNTPs were optimized in parallel. Multiplex reactions using 0.1 U/μl of iTaq polymerase and 400 μM dNTPs showed no difference in Ct values (within 0.5 U) and generated identical exponential phases compared to the singleplex reaction (F).
Fig. 4. Verification of the “universal” multiplex condition. The multiplex “universal” condition allowed for optimal multiplex performance over the entire in vitro assembly series 1–15. This figure shows the amplification plots generated with assembly 1 (all templates had a Ct 35) using the multiplex “universal” condition. The data generated with the singleplex reactions are shown as black amplification traces with the multiplex qRT-PCR in color amplification traces. The Ct values are shown within each graph. Singleplex Ct values are in the left column and multiplex Ct values are included in the right column. The Ct for both the singleplex and the multiplex reactions were within 0.5 U of each other, and both reactions showed similar exponential phases.
Fig. 5. Verification of the “universal” multiplex condition. This figure shows the data generated with the multiplex “universal” condition when applied to the in vitro assembly 9. All templates in the reaction were present at maximal concentrations (Ct 15), a millionfold more concentrated than those used for the experiment shown in Fig. 4. Singleplex reactions are shown as black amplification traces and the multiplex reactions as color amplification traces. The Ct values are shown within each graph. Singleplex Ct values are in the left column and multiplex Ct values in the right column. The Ct values for both the singleplex and the multiplex reactions were within specifications (maximum Ct difference no greater than 0.5). Notice that the exponential phase of both singleplex and multiplex reactions can be superimposed.
Fig. 6. Verification of the “universal” multiplex condition. This figure shows the amplification plots generated with in vitro assembly 15 using the “universal” multiplex condition. The concentration of three of the four transcripts included in this set (ODC, OAZ, and AZI) was one millionfold higher than that present at the lowest concentration (AdoMetDC). The Ct for ODC, OAZ, and AZI was 15 while the Ct for AdoMetDC was 35. Singleplex reactions are shown as black amplification traces and multiplex reactions as color amplification traces. The Ct values are shown within each graph where singleplex values are in the left column and multiplex values in the right column. All the Ct values corresponding to the single and multiplex reactions were found with in 0.5 U of each other, and both reactions showed identical exponential phases.
Table 1.
Primer pairs used for in vitro transcript template generation
Table 2.
Primers and probes used for 5′-nuclease real-time RT-PCR assay
Abstract
Purchase PDF (600 K)
E-mail Article
Add to my Quick Links
Cited By in Scopus (11)
Purchase PDF (173 K)
