Methods and reagents: Quantitative PCR: an accurate measure of mRNA?

by Paul N. Hengen, Ph.D. *

Methods and reagents is a unique monthly column that highlights current discussions in the newsgroup bionet.molbio.methds-reagnts, available on the Internet. This month's column discusses quantitative measurements of messenger RNA levels by using reverse transcription polymerase chain reaction (RT-PCR). For details on how to partake in the newsgroup, see the accompanying box.

Quantitation of gene expression requires that an accurate measurement of a specific transcript is made, and netters recently had a battle over whether the polymerase chain reaction (PCR) can be used for this.

One technique, called reverse transcription polymerase chain reaction (RT-PCR), has become much more popular of late. RT-PCR is a multiple-step PCR accomplished by isolating total RNA from a cell line, using reverse transcriptase to create a pool of cDNA, and then amplifying a predetermined DNA fragment through the use of specific primers and thermocycling. The amount of a specific gene transcript can theoretically be measured by estimating the intensity of the amplified DNA band on an ethidium-bromide-stained agarose gel or X-ray film by densitometry.

In comparison to a standard co-amplified mRNA control sequence, the band intensity is used to estimate the initial amount of mRNA available for the reaction and, thus, the amount of gene expression from the particular gene of interest.

Some netters were recently complaining that although the use of PCR for quantitation has been uncritically accepted by many scientists, it really cannot be relied upon for quantitative measurements of nucleic acids owing to the very nature of the amplification process.

When DNA is amplified by PCR, the efficiency of each extension reaction per cycle is not always 100%. Given an initial amount of template and more than 20 cycles to complete the PCR, the amount of DNA product at the end of the reaction is not always as much as one might expect. This is due to a plateau effect after the exponential increase of product, which is dependent on the initial amount of target. [1] One netter summarized what can be considered the sentiments of many scientists when he said `No matter how you do it and how good your controls are, your results from quantitative PCR will be questionable.'

This prompted a discussion about what exactly makes the results of RT-PCR questionable and how some problems inherent to the process might be avoided.

Back to basics
One person pointed out that an article by Luc Raeymaekers (Luc.Raeymaekers@med.kuleuven.ac.be) suggests that many have ignored some basic theory and predictions. Raeymaekers [2] defines the amplification of DNA within a RT-PCR in terms of a mathematical model, which can be used to explain why quantitative PCR can lead many researchers to think that they have a handle on gene expression levels in their experiments.

In a situation where the target sequence (T) and the standard sequence (S, the competitor) would amplify with different efficiencies, the model reveals that even the smallest difference between the efficiencies of amplification (E) can lead to very different quantities of end products of T and S, which can result in an erroneous estimate of the amount of initial material. The argument is made that S and T sequences are often wrongfully assumed to have equal amplification efficiencies, the result of which can be disastrous misjudgements made from the amount of product seen on a gel after 20 to 30 amplification cycles. [3]

According to netters, even with the most carefully designed primers, the two efficiencies in most reactions involving co-amplified sequences are almost never the same. An exception might be when the T and S sequences differ by a single base pair, for example, as in the case of an engineered mRNA with a restriction site planned within it so that the standard DNA amplified from it can be distinguished from DNA amplified from the native mRNA, and both templates are amplified in the same reaction using the same set of primers.

However, some people thought that Raeymaekers' model might be a poor one, not being a completely true representation of the in vitro situation. They said that if it cannot be confirmed through further molecular experimentation, then the argument is moot.

Effective controls
How different are the efficiencies of amplification in different experiments, and how might they be controlled or avoided? One solution would be to use a more carefully selected control. Some people feel that a `housekeeping' gene, i.e. one that is expressed in all cells and does not vary much in expression during the cell cycle, should be used as competitor. But this is still considered an external control that might suffer from the same problem. Another idea is that, by using internal controls, the competitor could be designed so that it amplifies at the same efficiency as the target. [4,5]

In addition, it was mentioned that a large number of cycles should not be run to avoid taking any measurements after the plateau phase of the amplification has been reached. Someone else proposed that, as an added precaution, serial dilutions of the product should be made after the cycling to avoid the problem of band saturation.

RT-PCR supporters
While some were calling for the need to create PCR police to help keep the rampant misuse to a minimum, others felt that, as well as being accurate, it is a well-accepted practice and the issues of this debate are old news. They think that researchers should be able to use this technique without reservation because the issues have already been thoroughly debated and there is no longer any concern that RT-PCR is not a legitimate way to quantitate messenger.

Netters said that they have had little trouble publishing articles in which the technique had been used, and that the reviewers agreed that the technique can be used to successfully demonstrate levels of specific transcripts. They felt that if rigorous conditions of the standard are carefully met, RT-PCR can be more accurate than using a northern blot analysis.

RNase protection and northern blotting
Others disagreed completely, insisting that the older, but more time-consuming methods of RNase protection assay (RPA) and northern blotting are far more accurate and precise, and that these methods are clearly a better choice than RT-PCR in most cases.

For example, one netter reported that by using RPA, he could detect approximately 1x10^6 target transcripts, which he felt was sensitive enough. In addition, it has been estimated that a northern blot is about tenfold less sensitive than RPA, with minimum detection at approximately 1x10^7 target transcripts. [6]

Another person said he would only be confident to conclude a result if an experiment that was repeated three times by conventional northern and/or in situ hybridization analysis matched that obtained by RT-PCR. Not only is this a good idea for confirming data found through RT-PCR, but it is also scientifically sound.

Unfortunately, because both RPA and northern blotting are more labor intensive and require much more starting material (10-20 ug of RNA per lane), they are limited to studying those genes that are relatively highly expressed, excluding many interesting genes that might be expressed at much lower levels, or rare pathological samples.

Most feel that PCR quantitation can and has been done poorly in many studies, and that this, not the method itself, has given it a poor reputation. If done carefully, RT-PCR has the potential to be one of the most powerful of techniques. One day perhaps RT-PCR will become the method of choice, but for all but a few exceptions, the old tried-and-true methods provide more solid footing.

References

[1] Chelly, J. et al. (1990) Eur. J. Biochem. 187, 691-698

[2] Raeymaekers, L. (1993) Anal. Biochem. 214, 582-585

[3] Raeymaekers, L. (1995) Genome Res. 5, 91-94

[4] Celi, F. S., Zenilman, M. E. and Shuldiner, A. R. (1993) Nucl. Acids Res. 21, 1047

[5] Cottrez, F. et al. (1994) Nucl. Acids Res. 22, 2712

[6] Ferre', F. (1994) in The Polymerase Chain Reaction (Mullis, K. B., Ferre', F. and Gibbs, R. A., eds), p. 67, Birkhauser


Any statements made by the author are not meant to advocate the use of a particular commercial product or endorse any company. All opinions are those of the author and do not reflect the opinion of the National Cancer Institute or the National Institutes of Health.

Copyright: This manuscript is not copyrighted by Elsevier Publishing Company. However, you may not reproduce any portion for resale or edit the text for redistribution, sale, or otherwise without written permission from the author.


You found this at the World Wide Web (WWW) Uniform Resource Locator (URL):

ftp://ftp.ncifcrf.gov/pub/methods/TIBS/nov95.txt

Any reference to this column must be cited as the following published article:

@article{Hengen1995Novtibs,
author = "P. N. Hengen",
title = "Methods and reagents - Quantitative {PCR} - an
accurate measure of {mRNA}?",
journal = "Trends in Biochemical Sciences",
volume = "20",
number = "11",
pages = "476-477",
month = "November",
year = "1995"}



*******************************************************************************
* Paul N. Hengen, Ph.D.                           /--------------------------/*
* National Cancer Institute                       |Internet: pnh@ncifcrf.gov |*
* Laboratory of Mathematical Biology              |   Phone: (301) 846-5581  |*
* Frederick Cancer Research and Development Center|     FAX: (301) 846-5598  |*
* Frederick, Maryland 21702-1201 USA              /--------------------------/*
* - - -  Methods FAQ list -> ftp://ftp.ncifcrf.gov/pub/methods/FAQlist - - -  *
* -  TIBS column archive -> http://www-lmmb.ncifcrf.gov/~pnh/readme.html - -  *
* - The BEST Molecular Biology HomePage -> http://www-lmmb.ncifcrf.gov/~pnh/  *
*******************************************************************************