| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
JDF Center for Prevention of Type 1 Diabetes in Finland, and Departments of
2
Virology and
3
Biotechnology, University of Turku, Turku, Finland.
a Address correspondence to this author at: MediCity Research Laboratories, Tykistökatu 6 A, 4th floor, FIN-20520 Turku, Finland. Fax 358-2-333-7000; e-mail
minna.sjoroos{at}utu.fi.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Methods: A multiplex solid-phase PCR to amplify the HLA-B27 alleles together with ß-actin as an amplification control gene was performed on the NucleoLinkTM (Nunc) surface. PCR was followed by hybridization and detection with time-resolved fluorescence. For the covalent capture of the PCR primers onto the solid support via a 1-(3-dimethylamino-propyl)-3-ethylcarbodiimide hydrochloride-mediated reaction, different 5'-end modifications of oligonucleotides were tested [amination, phosphorylation, and a poly(dT)10 linker].
Results: For covalent immobilization of the primers, amination of the 5' end combined with use of the poly(dT)10 linker was superior. At least 19.5% of the primer added per well was attached via a stable bond. When the standard time-resolved, fluorescence-based HLA-B27 detection system was compared with the newly developed method in a sample series of 82 genomic DNAs and the corresponding dried-blood spots, all results were in full agreement.
Conclusions: The new solid-phase PCR approach can be applied for multiple-target DNA detection. PCR followed by hybridization can be accomplished in a few hours using precoated strips and dried-blood spot PCR templates.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The microtiter plate format is widely used for the analysis of clinical samples, and a polymer surface allowing the covalent bonding of oligonucleotides (2)(3) has been widely applied for hybridization reactions (4)(5)(6)(7)(8)(9)(10). The thermostability of the first solid supports was a problem in PCR applications, but a second-generation surface has been reported to support PCR amplification (11)(12).
Here we describe a combination of multiplex PCR and a hybridization reaction performed on the NucleoLinkTM (Nunc) surface. For the solid-phase PCR, the wells were coated with the reverse primer for HLA-B27 alleles and the forward primer for the ß-actin gene. PCR was followed by subsequent detection of the amplicons by hybridization with lanthanide-labeled oligonucleotide probes and time-resolved fluorometry (TRF)1 (13)(14)(15). This procedure simplifies typing for the spondyloarthropathy-associated HLA-B27 alleles (16).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
) were modified with either 5'-phosphate (MedProbe) or
5'-amino functions (constructed in-house); for determining the amount
of nonspecific coating, they were not modified at all. Primers (Table 1
) with a poly(dT)10 linker between the
terminal amino group and the specific sequence were also tested. All
solution-phase PCR primers (Table 1
) were unmodified (CyberGene AB).
|
The lanthanide-labeled probes (Table 1
) described by Välimaa et
al. (17) were designed to detect HLA-B27 alleles
(europium-labeled probe, Eu-761) and the human
ß-actin gene (samarium-labeled probe, Sm-799).
In addition to these probes, a europium-labeled oligonucleotide
(Eu-1307) complementary to the solid-phase primer (1303) was used to
study the amount of the attached primer obtained and the stability of
the coating (i.e., nature of the bond linking the surface and the
primer).
The detection sensitivities of the Eu-761 and Sm-799 probes were
evaluated with the help of the synthetic targets 1320 and 1321 (Table 1
), which were complementary to the detection-probe sequences. Targets
carried the additional 10-thymidine tail at their 5' ends for optimal
coating. The amounts of the coated targets were 1 x
107 to 5 x 1011
molecules per well.
Oligonucleotides prepared in-house were synthesized, labeled, and purified as described previously (18). To introduce amino functions into the 5'-terminal end of an oligonucleotide during its synthesis for coating and lanthanide-labeling purposes, a diaminohexane-modified deoxycytidine (modC) (19) was used.
coating
For covalent capture, HLA-B27 reverse (3') and
ß-actin forward (5'; 100 ng of each)
solid-phase primers diluted in water were combined with freshly
prepared 10 mmol/L 1-(3-dimethylamino-propyl)-3-ethylcarbodiimide
hydrochloride (Acros Organics) in 10 mmol/L fresh
1-methylimidazole buffer (pH 7.0; Acros Organics). Eighty
microliters of the coating solution was distributed into each well,
which was then sealed with adhesive tape. Subsequently, incubation was
carried out at 55 °C for 5 h. After coating, the wells were
washed (DELFIA® Platewash; Perkin-Elmer Life
Sciences, Wallac Oy) four times with 100 mmol/L Tris (pH 7.5), 150
mmol/L NaCl, 1 mL/L Tween 20. The wells were then washed once
with deionized sterile water. The empty strips were stored at 4 °C
in a plastic bag until used.
dna samples
Genomic DNAs extracted from a few routine HLA-B27 whole
blood (EDTA) samples, according to a standard phenol-chloroform-isoamyl
alcohol method, were used to optimize assay conditions. To
validate the performance of the new HLA-B27 assay in
comparison with the routine assay in use in our laboratory
(17), an additional 82 blood samples were collected for
analysis with both methods. These additional samples were either
spotted on filter paper (Schleicher & Schuell; 60 µL of blood per
spot) and dried, or the genomic DNA was isolated from the blood sample
with a commercially available method (QIAamp DNA Blood Mini Kit;
Qiagen).
stability of the covalent linkage
To evaluate the stability of the covalent linkage between
solid-phase primer 1303 and the surface after coating, the wells were
incubated for 2 h (unless otherwise stated) with 100 µL of the
following aqueous reagents: 1 mol/L NaOH, 1 mol/L HCl, 10 mL/L
Triton X-100, 100 mL/L acetic acid, 1x standard saline citrate
(0.15 mol/L NaCl and 0.015 mol/L sodium citrate, pH 5.0), 1x
phosphate-buffered saline (pH 7.5), 0.1 mol/L carbonate (pH
10.0), 100 mL/L ethanol, 100 mL/L methanol, 10 mL/L dimethyl sulfoxide
(DMSO), 10 mL/L N,N-dimethylformamide, absolute ethanol, 1
mol/L NaCl (24-h incubation), 10 mmol/L HCl (37 °C, 4 h) wash
solution used in coating, or deionized sterile water. After treatment,
wells were washed three times with sterile water followed by Eu-1307
hybridization.
To study the resistance of the coated surface to the increased temperatures used in PCR, wells coated with solid-phase primers were subjected to routine thermocycling conditions with an artificial PCR mixture lacking the solution-phase primers and the DNA polymerase. After PCR, the amount of the 1303 primer remaining on the surface was detected by hybridization to the Eu-1307 probe. To control the study, hybridization with uncycled strips was performed simultaneously.
pcr
Genomic DNA was amplified in precoated wells in a 50-µL reaction
mixture containing 1.5 U of DNA polymerase
(DyNAzymeTM II, F-501L; Finnzymes Oy), 200 µM
each dNTP (Amersham Pharmacia Biotech), 1 g/L bovine serum
albumin (Sigma), 20 mL/L DMSO (Merck), 0.4 µM primer 798, 0.03 µM
primer unmod-490, 0.15 µM primer 759, 0.01 µM primer unmod-760, and
1x buffer [10 mM Tris (pH 8.8), 1.5 mM MgCl2,
150 mM KCl, 1 mL/L Triton X-100; Finnzymes Oy] supplemented
with 2 mM MgCl2. For the multiplex amplification,
a Hybaid Touchdown thermocycler with a hot lid was programmed for the
initial incubation at 94 °C for 5 min, 56 °C for 4 min, and
72 °C for 2 min. The next 35 cycles consisted of 1 min at 94 °C,
50 s at 56 °C, and 50 s at 72 °C. To ensure complete
extension, a final incubation at 72 °C for 5 min was used. When
dried-blood spots were used as the source of DNA in PCR, disks
(diameter, 3 mm) from the spots were punched directly into the wells
and incubated for one cycle (4 °C for 30 s, 94 °C for 3 min,
4 °C for 30 s, and 94 °C for 3 min), after which the
temperature was held at 80 °C while the PCR mixture was added into
the wells. The wells were then sealed with a
CycleSeal® (Robbins Scientific), and the program
mentioned above was started. Duplicate samples were used in every PCR
run. HLA-B27-positive and -negative controls were included
in all series.
hybridization
Hybridization and detection of TRF emissions after pulsed-light
excitation and a label-specific delay time were carried out as
described previously (17) with emission signals expressed as
cps. To block the background, various amounts of nonfat dry milk (1,
10, and 30 g/L; Valio Oy), denatured salmon sperm DNA (50–100
ng/well; Life Technologies), and polyvinylpyrrolidone (PVP-360; 1, 10,
and 20 g/L; Sigma) were added to the hybridization mixture. To enhance
the samarium signal, especially after the blood-spot PCR, polyethylene
glycol (PEG-6000; 20, 50, and 10 g/L; Fluka) and dextran sulfate (20,
50, and 100 g/L; Amersham Pharmacia Biotech) additives were tested. To
test whether a rehybridization could be applied to our assay format,
two subsequent hybridizations were performed, separated from each other
by washes (three times with 1x DELFIA Wash Concentrate) at room
temperature, followed by denaturation.
The detection sensitivities of the Eu-761 and Sm-799 probes were determined by direct hybridization to the precoated targets, omitting the denaturation step because of the single-stranded nature of the targets. Direct hybridization was also applied to the Eu-1307 probe hybridization used to study the amount of primer 1303 attached and the stability of the coating. A concentration of 5 µg/L of the probe was used together with 500 mL/L formamide (Amresco). Hybridizations were performed on the precoated wells (as described in "Amounts of Primers Attached"), as well as on precoated and pretreated wells (as described in "Stability of the Covalent Linkage"), thus removing the variation of results caused by PCR.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
were obtained. The optimized coating protocol, as described in
Materials and Methods, and the routine hybridization
protocol (17) were used. Subsequently, the same experiment
was repeated with different samples and the fully optimized assay
(Table 2B
).
|
In both studies, a marked effect of the poly(dT)10 can be seen. When we compared the results obtained with phosphorylated primers with those obtained with the aminated primers, the signal-to-noise (S/N) ratio (e.g., signal is defined as the specific signal obtained, and noise is defined as the nonspecific signal of the PCR background control, which is the reaction without DNA template) for the aminated primers was substantially higher, especially with the samarium labeled ß-actin-specific probe detection. In addition, the S/N ratio of the phosphorylated primers in our system was almost equal to the S/N ratio of the unmodified counterparts.
pcr
Several reaction conditions, with
MgCl2 concentration, primer concentrations,
annealing temperature, and additives (e.g., bovine serum albumin and
DMSO) as the most important ones, were tested to optimize the PCR
reactions. The conditions leading to optimal performance of the PCR are
indicated in Materials and Methods.
stability of the covalent linkage
After coating, stability tests were conducted to determine the
strength of the bond immobilizing the oligonucleotide onto the surface.
The results of the treatment of the 1303-precoated surface with a
variety of reagents are indicated in Table 3
. When the fluorescence counts obtained were compared with the
signals of the untreated (water incubation) wells and expressed as
residual signals (%), the surface was found to be very stable. It was
affected only by treatment with a strong acid (1 mol/L HCl) or alkali
(1 mol/L NaOH).
|
When the wells precoated with primers 1302 and 1303 were subjected to PCR temperatures followed by Eu-1307 hybridization, almost 100% of the attached 1303 primer could still be detected as remaining on the surface. Because we could detect very little difference between the treated and untreated wells, we concluded that our PCR conditions did not inactivate the primer or detach it from the surface.
amount of primers attached
Only a rough estimate of the amount of the 1303 primer attached to
the wells during the coating could be made. The labeled Eu-1307 probe
used in the direct hybridization was not able to detect >20 ng of the
primer (1.2 x 1012 molecules). Because the
optimum amount of the primer added to each well in coating was
100 ng (6.15 x 1012 molecules), it could
only be concluded that at least 19.5% of the input primer,
corresponding to 1.2 x 1012 molecules, was
attached during the coating.
hybridization
We tried to reduce the background fluorescence signal of our
HLA-B27 assay by adding common blockers, such as nonfat dry
milk, salmon sperm DNA, and polyvinylpyrrolidone to the hybridization
mixture. In contrast, to enhance the signal, hybridizations with
dextran and polyethylene glycol were carried out. The background signal
for Eu-761 decreased by an average of 33% and that of Sm-799 decreased
by 13% when 1 g/L nonfat dry milk was added to the hybridization
reaction. Increasing the amount of dry milk or adding other additives
did not improve the performance of the assay.
After the fluorescence detection and denaturation phases of the first hybridization, the rehybridization was successfully performed. No difference in the S/N ratios of the two consecutive hybridizations with either HLA-B27-positive or -negative samples could be seen. Thus, no loss in hybridization capacity of the wells could be observed.
sensitivity of the detection probes
A linear relationship (data not shown) between increasing amounts
of the targets (1320 and 1321) and the fluorescence signals originating
from the complementary lanthanide-labeled probes (Eu-761 and Sm-799,
respectively) was observed. The signals produced by both probes were
linear for 1 x 108 to 5 x
1011 target molecules (i.e., more than
three orders of magnitude with correlation coefficients of 0.991 for
Eu-761 and 0.977 for Sm-799). When the cutoff for a positive reaction
was set as three times the background counts, detection limits of
5.2 x 108 molecules for the Eu-761 probe
and 2.1 x 1010 molecules for the Sm-799
probe were obtained.
reproducibility of the assay
The intra- and interassay CVs for the entire solid-phase assay,
including coating, PCR, and hybridization, were 14% and 18%,
respectively, for isolated DNA samples, and 21% and 16% for blood
spots (n = 8).
clinical samples
At the above-mentioned cutoff, full agreement was found with the
routine assay (17) for the 82 blood-spot samples.
When we qualitatively categorized the S/N ratios into
HLA-B27-positive and -negative groups, no difference between
the genomic DNAs and the blood-spot DNAs was observed (Fig. 1
). The systematically lower S/N ratios obtained with the
blood-spot samples can be explained by the lower amount of the DNA
template and/or DNA polymerase inhibitors present in the PCR. Also
clearly seen in Fig. 1
is the effect of the intrinsic fluorescence
properties of the samarium label (20) on the S/N ratio,
which was much lower than that of europium. Nevertheless, the samarium
probe used in our assay serves as a useful internal control for the
quality of the sample DNA in PCR and the performance of PCR in general.
When the samarium signal originating from the
ß-actin PCR product was low, representing a low
efficiency of PCR for a particular sample, the europium signal
was decreased and vice versa. The more efficient amplification of the
HLA-B27 alleles in our multiplex PCR makes interpretation of
the fluorescence signal of a sample easy, even if combined with a lower
samarium control signal.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The use of the spacer arms of unspecific sequences coupled to an attached oligonucleotide to extend it away from the surface was necessary for optimal PCR performance. This procedure is widely used as a method to create conditions that approximate hybridization performed in solution (24). Thus, the steric hindrance produced by the surface, which restricts the reaction between the primer and the DNA template, can be avoided (25).
The stability of the covalent bond between the 5'-amino primer and the functional group on the surface of the plate was demonstrated by its resistance to treatment with acids, alkali, detergents, organic solvents, and buffers with different pHs. In addition, the bond was able to withstand the PCR cycling conditions. Thus, a rehybridization procedure after measurement was also possible.
The sensitivities of the detection probes Eu-761 and Sm-799 were in
line with previous results using TRF for hybridization detection
(18)(26). The somewhat lower sensitivity of the
samarium probe is attributable to the fivefold higher background
(
1000 cps) caused by the surface material compared with the
background of the wells (200 cps) used in the earlier calculations.
The results of the newly developed HLA-B27 method were consistent with the results of the standard method (17) in each of the 82 clinical samples assayed. HLA-B27-positive samples can be clearly identified, even when combined with a low ß-actin signal (blood spots), reflecting more efficient amplification of the HLA-B27 alleles and a better detection sensitivity of the europium label. The use of blood spots as the source of DNA in PCR allows the cumbersome and time-consuming steps of DNA extraction to be avoided. In addition, numerical values obtained with TRF allow objective interpretation, statistical treatment, and computer-assisted handling of the data. Moreover, the TRF technology allows construction of a triple-label assay (i.e., europium, samarium, and terbium) (18), which is especially suitable for multiple gene detection.
In conclusion, in the newly developed solid-phase assay,
HLA-B27 was used only as a model analyte, although the
method is well suited for our diagnostic service, which handles 50
samples per week. The method will be applied in our
barcode-assisted, large-scale screening program.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  A. Pemov, H. Modi, D. P. Chandler, and S. Bavykin DNA analysis with multiplex microarray-enhanced PCR Nucleic Acids Res., January 20, 2005; 33(2): e11 - e11. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Turner, S. Penning, A. Sharp, V. J. Hyland, R. Harris, C. P. Morris, and A. van Daal Solid-Phase Amplification for Detection of C282Y and H63D Hemochromatosis (HFE) Gene Mutations Clin. Chem., August 1, 2001; 47(8): 1384 - 1389. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
,
europium; 
, samarium. As indicated, HLA-B27-positive
samples can be clearly distinguished from negative ones, independent of
the nature of the PCR template.