| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
a Author for correspondence. Fax +44 1 606 49366.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
F508 mutation status of the cystic fibrosis
transmembrane conductance regulator gene. The genotypes determined with
the method described herein were in full agreement with those obtained
by the conventional application of ARMS. We also demonstrated the
simultaneous detection of two PCR products in a single reaction. The
assay method described is homogeneous and so obviates the necessity to
open reaction vessels after amplification. This therefore eliminates
PCR carryover contamination.
Key Words: indexing terms: DNA amplification • cystic fibrosis • genetic variants
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Crossover contamination prevention is usually addressed by isolating each stage of the PCR process so that setup, amplification, and detection are carried out in separate facilities with dedicated protective clothing and equipment for each area (3). Another physical precaution is the use of laminar airflows (3). Biochemical means of preventing PCR product carryover contamination include UV photoinactivation after psoralen (4)(5)(6) or isopsoralen (7)(8)(9) treatment, or more commonly, substitution of dUTP for dTTP during PCR with subsequent PCRs being pretreated with uracil DNA glycosylase (6)(9)(10)(11).
An alternate way of preventing carryover contamination is to detect PCR products without opening the reaction tube after amplification. If product detection is by spectroscopic means in a sealed vessel, then the tube can be discarded after analysis, removing the risk of release of amplicon. Another advantage is that homogeneous amplicon detection avoids electrophoretic separations and should be easily adaptable to automated analysis.
Fluorescence polarization (FP) detects changes in the molecular volume of a fluorophore (12) and is also capable of detecting nucleic acid hybrids in solution (A. J. Garman, Zeneca Pharmaceuticals, personal communication, (13)(14)(15)(16)).1 We show that there is a measurable increase in the FP of a fluorescently labeled oligonucleotide probe when specifically hybridized to a PCR amplicon.
The amplification refractory mutation system (ARMSTM) is a
powerful technique for detecting mutations and polymorphisms in DNA
(17) and is in part a significant improvement of the PCR.
Primers that are mismatched at their 3' termini relative to the
template genomic DNA are not readily extended by, e.g., Taq
DNA polymerase. Two ARMS reactions are usually performed on a DNA
sample. These involve primer sets that contain a common primer in both
reactions and a primer specific for one or the other allele in the
separate reactions. The three possible results, depending on the
genotype, are: (a) DNA that is homozygous for one allele
gives a product in the reaction with the primer specific for that
allele, (b) DNA from a homozygote for the other allele gives
a product with the other allele-specific primer, and (c)
there is product in both of the reactions when DNA from a heterozygote
is tested. Here, we demonstrate that the inclusion of a fluorescently
labeled probe with the ARMS primers allows the homogeneous genotyping
of DNA samples. Detection of the ARMS products is achieved by measuring
the change in FP of the probe bound to the amplicon. Because the
sequence of the amplicon produced by each test differs only at the
location of the allelic variation, one fluorescent probe can be used to
detect both amplicons. To demonstrate the use of FP in the homogeneous
detection of ARMS products we adapted an ARMS test for the cystic
fibrosis transmembrane conductance regulator (CFTR) gene
F508 mutation (18).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reagents and solutions.
PBS and
5-([4,6-dichlorotriazin-2-yl]amino)fluorescein (DTAF) were from Sigma
Chemical Co., Poole, UK; Sephadex G-25, NAP-10 columns, and dNTPs from
Pharmacia Biotechnology, Piscataway, NJ; Taq DNA polymerase
(AmpliTaq) from P. E. Applied Biosystems, Foster City, CA; Nusieve
agarose from FMC BioProducts, Rockland, ME; and ØX174
DNA/HaeIII digest molecular size marker from Life
Technologies, Gaithersburg, MD. All other chemicals were from Aldrich
Chemical Co., Milwaukee, WI. Oligonucleotides with the 5' labels
6-carboxyrhodamine (ROX), fluorescein, and 6-aminohexylphosphate were
from Oswel Research Products (Southampton, UK). The remaining
oligonucleotides were synthesized on a P. E. Applied Biosystems 394
DNA/RNA synthesizer with P. E. Applied Biosystems reagents according to
the supplier's protocols and were purified by size exclusion
filtration through NAP-10 columns. The DTAF labeling buffer was 0.10
mol/L NaCl, 0.10 mol/L sodium carbonate, pH 8.00; HPLC buffer A 0.1
mol/L ammonium chloride; HPLC buffer B 800 mL/L HPLC buffer A, 200 mL/L
acetonitrile; ARMS buffer 1.2 mmol/L MgCl2, 10 mmol/L
Tris-HCl, 50 mmol/L KCl, 0.1 g/L gelatin, pH 8.3; Tris–borate–EDTA
electrophoresis buffer (TBE) 134 mmol/L Tris, 74.9 mmol/L boric acid,
2.55 mmol/L EDTA, 0.1 mg/L ethidium bromide; gel loading buffer 300
mL/L glycerol and 1 g/L bromphenol blue in TBE; and PBS 10 mmol/L
phosphate, 2.7 mmol/L KCl, 0.137 mol/L NaCl, pH 7.4.
Probe design.
Probe sequences were designed with the
primer analysis software Oligo 5 (National Biosciences, Plymouth, MN).
Specifically, probes were to hybridize to one of the strands of the
target amplicon after amplification but not during the extension phase
of the PCR. If probes were hybridized during PCR extension they would
be degraded by the 5'-exonuclease function of Taq DNA
polymerase (19) as the enzyme extended the primer annealed
to the same strand. Probe sequences were selected to have
Tms between 12 and 15 °C below the
temperature of the PCR extension phase. To facilitate probe
hybridization, regions of the gene sequence capable of forming stable
secondary structures were rejected as binding sites. An internal
structure search was performed for candidate probe sequences plus 10
nucleotides up- and downstream. Only sequences that included hairpins
with stems of three bp or less and where the G value for loop
formation was >-1.0 kcal/mol were considered.
Probe labeling.
The 6-aminohexylphosphate-labeled
oligonucleotide probe was labeled with DTAF as follows: One milliliter
of DTAF (500 mg, 940 nmol) in a mixture of acetonitrile and PBS (1:1 by
vol) was added to probe 1 (Table 1
) (21.3 nmol) in DTAF labeling buffer (500 µL). After 1 h
the reaction was quenched by adding diisopropylamine (10 mg, 10 mmol)
and separated on Sephadex G-25 (15 mm x 400 mm) eluted in water.
The faster eluting fraction was resolved by reversed-phase HPLC
[Waters µbondapak C18 (Waters Corp., Milford, MA) gradient elution
profile with buffers A and B: time 0 min (100% HPLC buffer A, 0% HPLC
buffer B), 2.5 min (100% A, 0% B), 7.5 min (0% A, 100% B), 13 min
(0% A, 100% B), 18 min (100% A, 0% B), and 20 min (100% A, 0% B,
stop)]. Several yellow-colored peaks were obtained and desalted on
Sephadex G-25 as above. One peak gave a short retention time and was
designated probe 2. This was confirmed to be the DTAF-labeled
oligonucleotide (UV 
max 260 nm with minor peak at 495
nm). In addition, the FP of free probe 2 = 51 mP, FP of probe 2
annealed to excess probe 3 = 170 mP.
|
DNA samples.
F508/F508 and
F508/+ cell line DNA were obtained from the Coriell
Institute for Medical Research (Camden, NJ), and +/+ DNA was isolated
from peripheral blood leukocytes from an unaffected individual as
described previously (18). Each sample was diluted to a
concentration of 10 mg/L (10 ng µL-1) before use. The
mutant allele was amplified with primers 1 and 2; the normal allele was
amplified with primers 1 and 3. Primers 4 and 5 were included in both
reactions to generate an apolipoprotein (apo) B gene amplification
control amplicon. All primers (Table 2
) were used at a concentration of 1 µmol/L; the PCR mixes
included dNTPs (100 µmol/L each) and probe (30 nmol/L) in ARMS buffer
(150 µL total volume). DNA (150 ng) and a mineral oil overlay were
added. The samples were heated at 94 °C for 5 min; then
Taq DNA polymerase (3 U) was added. Thirty-five rounds of
thermal cycling were carried out (94 °C 1 min, 58 °C 1 min,
72 °C 1 min). An aliquot (10 µL) was removed after PCR for gel
analysis and the remainder was examined by FP.
|
Gel electrophoresis.
Three percent gels (Nusieve
agarose:agarose, 3:1) were prepared and electrophoresed in TBE buffer.
ARMS reaction aliquots were mixed with gel loading buffer (10 µL) and
electrophoresed against ØX174/HaeIII digest DNA size
markers (500 ng). Gels were visualized by UV transillumination and
photographed.
FP.
FP probes were annealed to the target amplicon by
heating at 94 °C for 5 min, snap cooling on ice for 5 min, and
equilibrating at 25 °C for 30 min. The samples were then either
transferred to a black microtiter plate for automated reading on the
Fluorolite FPM-2 or were transferred to quartz microcuvettes (100 µL)
for manual reading on the Fluoromax. Fluorescein- and DTAF-labeled
probes were excited at 495 nm and the fluorescence was detected at 525
nm; ROX-labeled probes were excited at 585 nm and the fluorescence was
detected at 605 nm. Four fluorescence readings were taken to calculate
FPs: IVV, reading with excitation and emission
polarizers vertical; IVH, reading with
excitation polarizer vertical and emission polarizer horizontal;
IHV, reading with excitation polarizer
horizontal and emission polarizer vertical; and,
IHH, reading with excitation and emission
polarizers horizontal.
 |
 |
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
F508 and F508 (+)
alleles of the CFTR gene (20). A panel of DNAs (+/+,
+/F508, and F508/F508)
were used to generate ARMS amplicons (18) in the presence
of fluorescein-labeled probe 5. Control reactions with no DNA were also
run (see Fig. 1
). An apo B-derived amplification control product was also
generated with primers 4 and 5 but was not detected by FP. Gel
electrophoresis of aliquots of the ARMS reactions showed amplicons of
the expected size in all lanes; no additional products were seen (see
Fig. 1
). The remainder of the ARMS reactions were subjected to a
heat–cool cycle to anneal probe 5 to the ARMS amplicons and the FP of
the probe was determined on the SPEX Fluoromax for each of the 24
samples. The results are shown in Fig. 2
. The positive samples give a mean FP of 99 mP (n = 12, SD
2.4); the negative samples give a mean FP of 68 (n = 12, SD 2.1).
By using a t-distribution test, t = 32.3 for
these data sets. For 22 degrees of freedom t0.95
= 1.72 and t0.99 = 2.51, which shows that the
mean FPs of the two data sets are significantly different at a 99%
level of confidence. A 99% confidence interval, rounded outwards, for
a future observation from the positive sample population is (91, 107;
99 ± 8), and similarly from the negative sample population is
(61, 75; 68 ± 7). These intervals do not overlap and therefore
demonstrate a clear separation between the two populations (see Fig. 3
).
|
|
|
Multiplex detection of amplicons.
Simultaneous detection
of several fluorophores in a mixture is possible by selecting
fluorophores with discrete excitation and emission frequencies.
Negative ARMS reactions are therefore distinguishable from PCR failures
by detecting the amplification control in samples where no diagnostic
ARMS amplicon is found. Multiplex, homogeneous detection of PCR
products was demonstrated with the DTAF-labeled probe 2 for the ARMS
amplicon in combination with the ROX-labeled probe 4 for the apo B
amplicon. These two probes were used in three amplification mixes
whereby the first contained both the ARMS and apo B products, the
second only the ARMS product, and the third only the apo B product.
Each experiment was in duplicate. The average FP of each probe in each
amplification mixture is shown in Fig. 4
. These results show that the two FP probes can be used to
detect their target amplicons in the same solution.
|
Stability of probe–product hybrids and specificity of hybrid
formation.
Probe 2 was annealed to both the ARMS and apo B
amplicons and the time course of the change in FP of the probe was
monitored on the FPM-2 as shown in Fig. 5
. Hybridization of the probe to the apo B amplicon was not
observed. These data demonstrate that the probe anneals to its target
selectively where it forms a stable hybrid.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In ARMS, amplification efficiency is dramatically reduced when a primer is mismatched at the 3' terminus relative to the genomic target. Only when primers are matched to the genomic target is a detectable amplicon generated. Detection of a mutation is therefore reduced to simply determining whether or not a sample contains an amplicon because its generation of an amplicon is dependent on the genomic DNA sequence. For FP detection the threshold concentration of product is dependent on the probe concentration. Sufficient target needs to be formed for all of the probe to be able to bind to avoid a mixture of bound and unbound probe. In such a situation there would be lower net FP than in a sample in which the probe was entirely bound. A partially bound sample is still usually distinguishable from an unbound sample but there will be a diminution in the separation of the polarizations of the positive and negative samples. For this reason we set the probe concentration below the expected concentration of any product formed. If only low concentrations of diagnostic ARMS amplicon are formed, giving a low FP, we would expect that a correspondingly small amount of amplification control would be formed. Such a sample would be identified as an amplification failure by reference to the FP of the control probe rather than as a negative sample that could lead to a misdiagnosis.
FP is determined by the measurement of fluorescence intensities. The
higher the probe concentration, the more accurate the measurement;
lower probe concentrations favor reproducible amplicon detection.
Optimum probe concentrations therefore reflect this relation between
accuracy and reproducibility. We found that for ARMS reactions
generating two amplicons, a probe concentration of 30 nmol/L is ideal.
With primer sets at a concentration of 1 µmol/L, only 3% of primers
require conversion to amplicon to produce target at 30 nmol/L. We found
that in such reactions the amount of primer incorporated into amplicon
is usually between 5% and 10% (data not shown). A 30 nmol/L probe
concentration will therefore give a reproducible change in FP in almost
all test samples. In Fig. 4
(top) the difference in FPs for the two
positive samples detected by the apo B probe is probably due to a low
yield of apo B amplicon. This would result in a significant proportion
of probe remaining unhybridized, thereby lowering the net FP value.
Thus, samples that give product at <30 nmol/L will usually be
detectable, but samples in which the amplicon concentration is <10
nmol/L will be difficult to distinguish from negatives and would be
identified as amplification failures to be retested.
Although FP can be used quantitatively (16), it is also
suited to delivering simple "yes or no" answers as is shown in Fig. 2
. Twelve of the samples contain the ARMS amplicon, 12 do not. FP
clearly differentiates between the two groups with little variation
about their respective mean FP. It is important to determine if there
is an overlap of the data sets. We have shown that the two data sets
are distinct and therefore misdiagnoses should be avoided.
FP detection of PCR amplification is reliant on the hybridization of an
oligonucleotide probe to one strand of a denatured PCR product. There
are two competing equilibria in such a reaction. These are the binding
of the probe to one of the amplicon strands and the reannealing of the
two amplicon strands. The thermodynamically more stable state for the
system is for the PCR product to reanneal, leaving the probe unbound.
However, the rate of DNA duplex formation varies with the square root
of the length of the smaller strand forming the duplex
(21). The binding of the probe to one of the amplicon
strands is therefore kinetically favored over the reannealing of the
two amplicon strands as is shown in Fig. 5
. This Fig. also shows that
the specificity of the reaction is very high. To detect a mutation by
ARMS with FP, there must be three specific oligonucleotide annealing
events: the binding of the two PCR primers to the genome to generate
the amplicon, followed by the binding of the fluorescently labeled
probe to an internal region of that amplicon. Thus an amplified region
of the genome will only give rise to a positive FP result if it
contains a sequence complementary to that of the probe. The chance of
such a region being amplified by random priming off the genome is
essentially zero. Nevertheless, confirming the specificity of the first
two hybridization events in the development of an ARMS/FP assay might
be considered prudent. This is conveniently done by monitoring the ARMS
part of the assay with gel electrophoresis as we described.
Our results show that fluorescein, DTAF, and ROX are useful for the FP detection of PCR products, and we demonstrated the multiplexed detection for two amplicons using the two latter fluorophores. Other chromophores can be used to label probes, including 6-carboxy-2', 7'dimethoxy-4', 5'dichlorofluorescein (JOE) and Cy-5 (data not shown). A wide range of the visible spectrum that forms the basis of a powerful multiplex detection capability is covered by these dyes.
In conclusion, we have demonstrated simple, rapid, and reproducible genotyping of DNA samples using FP detection of ARMS amplicons. The detection protocol requires a postamplification heat, cool, and equilibration cycle. Reading is rapid with the microtiter plate reader FPM-2, although this instrument is not suitable for carrying out multiplex detection. The technique can be considered homogeneous in that it requires no separation or wash step, but, in the multiplex format, it must be carried out in a nonhomogeneous manner because the ARMS reactions must be transferred from one vessel to another for the FP measurement. Therefore, for homogeneous multiplex ARMS analyses with FP, a microtiter FP plate reader is needed with two novel features: (a) multiple label detection across the entire visible spectrum, and (b) the ability to carry out assays directly on samples contained in PCR amplification tubes or microplates. This would lend the system to automation and completely obviate the need for complex means of avoiding carryover contamination (3)(4)(5)(6)(7)(8)(9)(10)(11).
|
Acknowledgments |
|---|
|
Footnotes |
|---|
1 Nonstandard abbreviations: FP, fluorescence polarization; ARMS, amplification refractory mutation system; CFTR, cystic fibrosis transmembrane conductance regulator; DTAF, 5-([4,6-dichlorotriazin-2-yl]amino)fluorescein; ROX, 6-carboxyrhodamine; TBE, Tris–borate–EDTA; and apo, apolipoprotein. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  N. Marme, A. Friedrich, M. Muller, O. Nolte, J. Wolfrum, J. D. Hoheisel, M. Sauer, and J.-P. Knemeyer Identification of single-point mutations in mycobacterial 16S rRNA sequences by confocal single-molecule fluorescence spectroscopy Nucleic Acids Res., July 26, 2006; 34(13): e90 - e90. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Beaudet, J. Bédard, B. Breton, R. J. Mercuri, and M. L. Budarf Homogeneous Assays for Single-Nucleotide Polymorphism Typing Using AlphaScreen Genome Res., April 1, 2001; 11(4): 600 - 608. [Abstract] [Full Text]
|
|
|
|
 |
|
 B. Vogelstein and K. W. Kinzler Digital PCR PNAS, August 3, 1999; 96(16): 9236 - 9241. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Chen, L. Levine, and P.-Y. Kwok Fluorescence Polarization in Homogeneous Nucleic Acid Analysis Genome Res., May 1, 1999; 9(5): 492 - 498. [Abstract] [Full Text]
|
|
|
|
 |
|
 L. J. Kricka Nucleic Acid Detection Technologies — Labels, Strategies, and Formats Clin. Chem., April 1, 1999; 45(4): 453 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Whitcombe, J. Brownie, H. L. Gillard, D. McKechnie, J. Theaker, C. R. Newton, and S. Little A homogeneous fluorescence assay for PCR amplicons: its application to real-time, single-tube genotyping Clin. Chem., May 1, 1998; 44(5): 918 - 923. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and to apo B
amplicon (•)