3.3.2.2.3
PCR Methods
3.3.2.2.3.1
Qualitative PCR
PCR also exploits the specificity of DNA polymerase to allow the selective
amplification of specific DNA segments occurring at low frequency in a complex
mixture of other DNA sequences. In a standard PCR test, two pairs of primers are used:
(1) forward, sense or 5′→3′,
(2) reverse, antisense or 3′→5′.
These primers are designed to hybridize on opposite strands of the sequence of
interest, and through a series of repetitive cycles, of 2–3 thermal steps, amplify the
sequence between the primers millions of times. Amplified pieces can be subjected to
agarose gel electrophoresis to separate amplified DNA according to size, although other
separation methods, such as high performance liquid chromatography (HPLC) and
capillary electrophoresis (CE), have been used. Several food ingredients (e.g. soy,
wheat, canola, potatoes, rice, maize, celery and tomatoes) have been analyzed using
PCR. A crucial rate-limiting step is DNA extraction and purification. Currently, several
methods of extraction are available, with two widely used:
(1) the CTAB method, based on incubating food sample in the presence of the
detergent cetyltrimethylammonium bromide,
(2) the Wizard method, employing DNA-binding silica resins (Promega Corp.,
Madison, WI).
Both methods produce satisfactory DNA isolation without unacceptable DNA
degradation and are cost effective. Factors such as excessive heat, nuclease activity and
low pH (quite common in food processing) contribute to DNA degradation. This is most
likely for products with long shelf lives, such as prepared meatballs in tomato sauce and
beefburgers. Compounds present in foods (e.g. proteins, fats, polysaccharides,
polyphenols, cocoa extracts and caramelized sugar) can inhibit DNA polymerase, and
data suggest that the critical minimum average DNA size for successful PCR analysis is
~400 bp.
1
Most currently available GMOs in the EU contain any of three genetic elements:
the cauliflower mosaic virus (CaMV) 35S promoter, the nopalin synthase (NOS)
terminator, or the kanamycin-resistance marker gene (nptII) and others (Table 4). These
elements also occur naturally in some plants and soil microorganisms, and can thus be
detected using PCR, giving false-positive results. If the PCR assay gives a positive
result, product-specific PCR methods that have been developed for a range of different
GM foods can be carried out. These methods exploit a set of primer pairs that spans the
boundary of two adjacent genetic elements (e.g. promoters, target genes and
terminators), or that are specific for detection of the altered target gene sequence.
Detection limits are in the range of 20 pg to 10 ng of target DNA and 0.0001–1% of the
mass fraction of GMOs.
Table 4
Different methods can be used to confirm the PCR results:
(1) specific cleavage of the amplified product by restriction endonuclease
digestion
(2) hybridization with a DNA probe specific for the target sequence
(3) direct sequencing of the PCR product
2
(4) nested PCR ,
in which two sets of primer pairs bind specifically to the amplified target
sequence
(Fig. 4e).
Fig. 4. Differences between traditional PCR (a,b) and real-time PCR (c,d) (courtesy of
J. Fagan). The concentration of products present at point A (a,c) are plotted as a
function of percent GMO present in the initial sample, demonstrating that
proportionality between DNA concentration and PCR products (dynamic range) is
limited for conventional PCR (b). By contrast, there is a linear relationship between
PCR products and DNA concentration during exponential phase of real-time PCR. (e)
Schematic illustration of nested-PCR method using two sets of primers selected to
bridge the 35S promoter sequence and the petunia-derived CTP sequence. A two-step
approach – an outer primer set followed by an inner primer set – will improve
selectivity and sensitivity of the reaction. Detection limits of the available DNA were
between 0.01 and 0.1% depending on the type of product examined (soybean meal,
protein, lecithin, oil or several processed products) and the size of test portion.
3
Abbreviations: CaMV, cauliflower mosaic virus; CTP, cells transit peptide; EPSPS, 5enol-pyruvyl-shikimate-3-phosphate synthase; NOS, nopalin synthase.
A collaborative validation study involving 29 laboratories in 13 countries
employing qualitative PCR to determine 35S promoter and NOS terminator from GM
soybeans, maize and other processed components showed that samples containing 2%
RRS or MM were unequivocally identified by all laboratories, and correct classification
was achieved by analysing the CaMV 35S promoter in samples containing 0.5% GM
soybeans. The method for the detection of NOS terminator was less sensitive, giving
higher false-negative results (3 and 7 at the 0.5 and 0.1% GMO, respectively, compared
with 0 and 5 for the CaMV promoter). Because of the large size of the maize genome,
this qualitative PCR method was somewhat less sensitive for the detection of transgenic
maize in raw material (i.e. the number of false negatives for 35S promoter in maize at
the 0.1% GMO was 14).
Recently, GeneScan Europe has introduced a test kit for the detection of GMOs
in food products, which allows a multiplex PCR for the specific detection of DNA
sequences from plant species and GM traits. The procedure starts with the isolation and
purification of DNA from the sample. Then, specific DNA sequences from both plant
species and GM traits (Table 4) are amplified by two separate multiplex PCR reactions,
products of both reactions mixed, and single-stranded (ss) DNA is created by digestion
with an exonuclease. After mixing with hybridization buffer, the sample is spread on the
chip and amplified sequences that will hybridize with cDNA probes covalently bound in
the chip are stained with the fluorescent dye Cy5 and analyzed by a biochip reader (e.g.
4
Biodetect 654™). The detection limit for the GMO Chip Kit is in the range of 250
copies of each of the target DNA sequences in the PCR.
3.3.2.2.3.2
Inverse PCR
Conventional PCR is used frequently to amplify DNA segments lying between
the two priming oligonucleotides. But PCR can also be used to amplify unknown DNA
sequences that lie outside the boundary of known sequences for which primer pairs can
be designed. This is referred to as inverse PCR (iPCR). The technique for amplifying
unknown flanking DNAs was developed by different research groups. In this technique,
genomic DNA is digested with restriction enzyme(s) to produce DNA fragments of
appropriate size (1 to 4 kb). It must be noted that there should be no sites of cleavage
within the core target sequence. The DNA fragments are then ligated to form
monomeric circular molecules, followed by PCR using oligomer primers that are
designed to extend outwardly from the known core target sequence, i.e. in the direction
opposite to those used for normal PCR (Fig.5). The major product of the amplification
reaction is a linear double-stranded DNA molecule that consists of a head-to-tail
arrangement of sequences flanking the original target DNA. Inverse PCR is therefore
useful as a method of chromosome crawling to explore the chromosomal sequences that
are contiguous to a known segment of DNA. But the length of DNA that car be
amplified by PCR is limited to relatively short steps along the chromosome.
3.3.2.2.3.3
Reverse Transcriptasemediated Pcr (RT-PCR)
Combining the process of cDNA synthesis by reverse transcriptase application
leads
trainscriptase PCR (RT-PCR). Thus, amplification of double-stranded DNA by PCR is
not limited to genomic DNA targets, but can also be applied to double-stranded cDNA
that has been synthesized from RNA using a reverse transcription reaction. It is possible
to detect, and even isolate, very rare mRNA transcripts from cell samples. RT-PCR acts
as a means to obtain material for sequence determination and a step in cloning a cDNA
copy of the RNA.
5
As described earlier, there are thermostable DNA polymerases such as rTth,
which can utilize RNA templates for cDNA synthesis and thus permit single enzyme
RT-PCR. Alternatively, one of the viral reverse transcriptase enzymes (AMV RTase or
MMLV RTase) can be used for the cDNA synthesis reaction, and then a second
reaction is performed using one of the thermostable DNA polymerases. There are
advantages and disadvantages to each of these methods, and the deciding factor often
depends on the whether the RT-PCR reaction is being performed for a specialized
research application or an automated diagnostic assay. At present, both classes of
enzymes (viral RTases and rTth DNA polymerase) are used for RT-PCR.
In this section, three basic molecular genetic strategies that utilize RT-PCR as a
tool are described.
6
Figure 5. a. Major steps involved in inverse PCR b. generalized amplification scheme
for unknown DNA flanking the TR DNA border of pTIT37 by IPCR.
3.3.2.2.3.4
RACE: Rapid amplification of cDNA ends
RT-PCR can be possibly used to amplify a complete cDNA coding
region. But in many applications there may be insufficient information for a
straightforward strategy. In general, cDNA synthesis and library screening can lead to
the isolation of gene sequences that represent all or part of a processed mRNA transcript.
There are always problems when trying to isolate the 5' end of low abundance transcripts
because cDNA libraries often do not contain a full representation of all mRNA
7
sequences. This is especially true of oligo-dT-primed libraries. RACE protocols
generate cDNA fragments by using PCR to amplify sequences of a known gene
between a single region in the mRNA and either the 3' or the 5' end of the transcript.
Thus it is necessary to know or to deduce a single stretch of sequence within the mRNA to
choose a one gene specific primer complementary to the sequence of cDNA segment and
a second "anchored" primer that anneals to a sequence that has been covalently attached
to the newly created cDNA terminus.
In the two RACE protocols, extension of the cDNAs from the ends of the
transcript to the specific primers is accomplished by using primers that hybridize either
at the natural 3'poly (A) tail of the mRNA, or at a synthetic poly (dA) tract added to the 5'
end of the first-strand cDNA. In the RACE 3' end protocol, mRNA is reverse transcribed
using an oligo (dT17) primer, which has a 17-nucleotide extension at its 5' e because of the
natural 3' poly(A) tail of the mRNA. This extension, the anchor sequence designed
to contain the restriction sites for subsequent cloning. Amplification is performed
using the anchor 17-mer and a primer specific the sought-after cDNA.
In the RACE 5' end protocol, the mRNA is reverse transcribed from a genespecific primer. The resultant cDNA is then extended by terminal deoxytidyl transferase
(TdT) to create a poly (dA) tail at the 3' end of the cDNA. Amplification is performed
with the oligo (dT17)/anchor system as used for the 3, protocol and the specific primer. 5'
RACE is used to amplify uncloned upstream transcript sequences
using
a gene-
specific primer.
Several
modifications
of
the
original RACE protocol have been
described. One these, called RNA ligase-mediated RACE (RLM- RACE), involves the
use of bacteriophage T4 RNA ligase to attach covalently a single-stranded RNA anchor
molecule to the decapped 5' end of the mRNA. First-strand cDNA synthesis can be
performed using a gene-specific primer to produce a pool of cDNAs encoding tf
anchored primer sequence. Variation of this method is referred to as ligationanchored PC (LA-PCR).
Quantitative end-point PCR; a crucial aspect of analysis of GMOs in food is
quantitation, because maximum limits of GMOs in foods are the basis for labeling in the
8
EU. Therefore, more quantitative PCR approaches were needed. PCR was shown to be
quantitative if an internal DNA standard was coamplified with target DNA. In systems
such as the quantitative-competitive (QC)-PCR method (Fig. 6 ), the presence of PCR
inhibitors will be noticed immediately because the amplification of both internal
standard and target DNA will be simultaneously affected. QC-PCR consists of four
steps:
(1) coamplification of standard- and target-DNA in the same reaction tube;
(2) separation of the products by an appropriate method, such as agarose gel
electrophoresis and staining the gel by ethidium bromide;
(3) analysis of the gel densitometrically;
(4) estimation of the relative amounts of target and standard DNA by regression
analysis.
At the equivalence point [Fig. 6b(iv)], the starting concentration of internal
standard and target are equal [i.e. the regression coefficient is >0.99 and the slope of the
regression line ≈ 1; Fig. 6d(iv)]. In the QC-PCR, the competition between the
amplification of internal standard DNA and target DNA generally leads to loss of
detection sensitivity. Nevertheless, the Swiss example allows as little as 0.1% GMO
DNA to be detected, which is within the threshold limits specified by the European
Novel Food Regulations.
9
Fig. 6. Schematic illustration of milestones carried out in a Swiss QC-PCR. Standard
and target DNAs are coamplified in the same reaction tube (a). Following the PCR, the
products are separated by gel electrophoresis (b), which distinguishes the standard DNA
from the amplified target by the size of the product. Gels stained with ethidium
bromide. At the equivalent point (iv), the starting concentrations of internal standard
and target are equivalent. Densitometric analysis of the various bands (c) can be used to
calculate the linear regression (d). Reproduced, with permission, from.
10
Twelve European control laboratories attempted to validate QC-PCR. Each of
eight coded samples that contained GMOs was processed twice and compared with
external standards containing 0.5 and 2% RRS. No false-negative results were found
among 246 determinations for six samples containing RRS, and no false-positive results
for the negative control sample were reported. Interlaboratory differences were smaller
for QC-PCR than for qualitative PCR, and were mainly caused by insufficient sample
homogenization. Furthermore, the calibration of QC-PCR can be controlled by certified
reference material available commercially.
3.3.2.2.3.5
Quantitative RT-PCR
It is often important to know relatively steady-state levels of specific gene
transcripts as a means to investigate the role of a gene function in a particular cell
phenotype. Northern blots and RNase protection assays are the two common ways to
measure steady-state mRNA levels. However, both of these assays are labor
intensive and may require up to 25 µg of total RNA for each assay. Therefore, most
gene expression studies cannot be performed using conventional RNA-based assays.
It is possible to develop a quantitative RT-PCR assay that takes into account
variation in the number of target molecules as well as amplification efficiency.
Quantitative RT-PCR measurements require a standard RNA template against
which the experimental RNA can be measured. Two types of RNA standards have
been used for this purpose. One type is an endogenous gene product that is
ubiquitously expressed in all cell types, for example, actin or glyceraldehyde-3phosphate dehydrogenase (GAPDH), which are the same standards used in northern
blots and RNase assays. The other type of standard is an exogenously added cRNA
(complementary RNA) that is synthesized by in vitro transcription. These exogenous
cRNAs are sometimes called RT-PCR "mimics" because they contain the same RTPCR priming sites and overall sequence as the target RNA, but produce a PCR
product that differs from the target RNA by a unique restriction site or a shift in
molecular weight. This makes it possible to compare directly the PCR product of the
target RNA to the cRNA standard on the same agarose gel. Mimic cRNAs are
transcribed from plasmids containing SP6 or T7 bacteriophage promoters and are
constructed by in vitro mutagenesis of the cloned target cDNA.
11
Figure 7. Rapid amplification of cDNA ends (RACE) a- 3’ Protocol b- 5’ Protocol
12
There are two ways to use mimic cRNAs to measure the amount of target
mRNA in a sample,
(1) Noncompetitive RT-PCR: This is performed by adding mimic cRNA to the
RNA sample at approximately the concentration of the target mRNA. By limiting PCR
amplification to the linear range (midexponential phase), usually 25 cycles or less, it is
possible to establish a direct correlation between the known amount of cRNA and target
mRNA. However, the absolute number of target mRNA molecules in the sample greatly
affects the number of cycles needed to detect the product within the linear range. This is
a significant problem if this method is used for diagnostic purposes.
(2) Competitive RT-PCR a series of RT-PCR reactions that contain the same
amount of sample RNA to which various known amounts of mimic cRNA have be
added. The concentration of mimic cRNA that produces the same amount of mimic
PCR product as the target RNA in the sample (a mimic: target product ratio of 1.0)
represents the point at which both templates are competing equally for the primers.
Figure 8 illustrates how competitive RT-PCR can be used to quantify the amount of
AMG mRNA in a cellular RNA sample that has been doped with increasing amounts of
an AMG mimic cRNA. Insertional mutagenesis can be used to construct an AMG
mimic cRNA for an internal marker in RT-PCR reactions using AMG gene-specific
primer (P1and P2). The AMG mimic PCR product is 30 bp longer than the normal
AMG PCR product. The amount of products generated in reactions can be seen in
agarose gel. Since competitive RT-PCR quantification is based on ratios, it is not
necessary to restrict PCR amplification to the exponential phase, and therefore
maximum sensitivity can be achieved by performing up to 35 cycles.
Quantitative real-time PCR; to circumvent some of the problems of conventional
quantitative end-point PCR, a real-time Q-PCR was introduced. At present, real-time
PCR is the most commonly used technology for quantification of GMOs. In theory,
production of PCR products should proceed exponentially. However, in practice it
reaches a plateau between 30 and 40 cycles because certain reaction components
become limiting. In conventional PCR, products of the reaction are measured at a single
point in the reaction profile. Plotting the concentration of products present at this point
as a function of the initial amount of DNA present in each of those reactions shows that
13
proportionality between DNA concentration (dynamic range) and PCR products occurs
over a limited range of DNA concentrations, leading to loss of precision in quantitation.
However, it has been shown empirically that the concentration of DNA in real-time
PCR reaction is proportional to PCR cycle number during the exponential phase of PCR
. Therefore, if the number of cycles it takes for a sample to reach the same point in its
exponential growth curve is known, its precise initial DNA (then GMO) content can be
determined. Real-time PCR also allows for detection of low copy DNA number. Several
commercially available real-time PCR thermal cyclers automate the analytical
procedure and allow cycle-by-cycle monitoring of reaction kinetics, permitting
calculation of the target sequence concentration. Several formats are used to estimate
the amount of PCR product:
(1) the ds-DNA-binding dye SYBR Green I;
(2) hybridization probes or fluorescence resonance energy transfer (FRET)
probes;
(3) hydrolysis probes (TaqMan® technology);
(4) molecular beacons.
These systems also permit differentiation between specific- and nonspecific PCR
products (such as primer-dimer) by the probe hybridization or by melt curve analysis of
PCR products, because nonspecific products tend to melt at a much lower temperature
than do the longer specific products.
A total of 179 food products containing GM RRS (e.g. baby food and diet
products, soy drinks and desserts, tofu and tofu products, cereals, noodles, fats, oils and
condiments) were analyzed. The method proved to be sensitive. Amplifiable soy DNA
could not, however, be detected in fats, oils and condiments. The genetic modification
of RRS was detected in 34 samples, eight of which contained >1% RRS.
Including a reference housekeeping gene as an internal standard, which will
amplify in parallel with the gene of interest, provides quantitation in PCR-based assays.
As for double competitive PCR, the quantitative estimate is established by
comparing the relative ratios of a GMO-specific target sequence and a reference target
sequence, usually a species-specific sequence. There is normally a linear correlation
14
between the quantity of GMO and the quantity of genetically modified DNA. This
correlation is less obvious between the quantity of GMO and protein/RNA.
Table 5. Examples of published methods to quantify GMO derivatives grouped into
competitive and real-time PCR methods.
Type of method
Target GMO
Reference
RoundupReady soybean (raw
materials)
Van den Eede & al., 2000
Bt176 maize (raw materials)
Van den Eede & al., 2000
P-35S (screening)
Hardegger & al., 1999
T-Nos (screening)
Hardegger & al., 1999
Epsps gene (RoundupReady)
Studer & al., 1998
Synthetic CryIA(b) gene
Studer & al., 1998
Bt-11 maize (event specific)
Zimmermann & al., 2000
RoundupReady soybean (event
specific)
Berdal & Holst-Jensen, in
press
Double competitive PCR
methods
Real-time PCR methods
Taverniers & al, in press
Terry & Harris, in press
Synthetic CryIA(b) gene
15
Vaïtilingom & al, 1999