© 2000 American Association for Clinical Chemistry, Inc.
Technical Briefs |
Colorimetric Microwell Plate Detection of the Factor V Leiden Mutation
1
Quest Diagnostics, Research and Development Department, 7600 Tyrone Ave., Van Nuys, CA 91405
a author for correspondence: fax 818-376-6387, e-mail bert.m.gold{at}questdiagnostics.com
Venous thrombosis is a serious disease that affects ~1 in 1000
individuals in the US (1). Although thrombophilia has many
causes, among the most frequent is activated protein C
resistance (2)(3)(4). Activated protein C resistance manifests
as an abnormally reduced activated partial thromboplastin time
(5). Among nonidiopathic mechanisms, the most common is
genetic (6), in particular, a G-to-A substitution at
nucleotide 1691 of the factor V gene on chromosome 1 (7).
This transition, located in exon 10 of the gene, produces an Arg-to-Gln
substitution at amino acid 506 of the protein. Individuals bearing this
genetic risk factor for venous thrombosis comprise ~5% of the
general population (8). Amino acid 506 is the less
consequential of two important sites for proteolytic cleavage mediated
by activated protein C (9). Mutant Va, the
-thrombin-activated membrane-bound cofactor product of the factor V
Leiden (FVL) locus, is kinetically inactivated at one-tenth the rate of
wild-type Va (5). In addition, wild-type factor V is cleaved
at another site, Arg 306 (8)(9), fully
inactivating it (6). The combined effects of kinetically
slowed inactivation and full inactivation through Arg 306 cleavage are
believed to account for the FVL heterozygous effect (6).
Screening for the FVL risk factor is considered an important part of a
thrombophilia evaluation. Heterozygotes are 5- to 10-fold more likely
to suffer a deep vein thrombosis than the general population, whereas
individuals with a homozygote FVL genotype are estimated to be at 50-
to 100-fold greater risk than those with two wild-type alleles
(10)(11)(12).
Our aim was to accommodate testing several thousand samples per month at a 25% cost savings over conventional PCR-restriction fragment length polymorphism (RFLP). We developed a reverse allele-specific oligonucleotide (ASO) hybridization assay in a microwell plate format. Our method for genotyping is a major modification of that published by Zehnder et al. (13). In our assay, a different primer pair is used, oligonucleotide probes are covalently attached directly to the microwell plate surface, and biotinylated amplicons are visualized using a chromogenic amplification system (14).
Because the mutation responsible for FVL is located in close proximity to the intron-exon border, both our group and Bertina et al. (7) chose PCR forward and reverse primers in exon 10 and intron 10 regions, respectively. In our initial studies, a 265-bp fragment spanning the exon-intron junction of the factor V gene was PCR amplified from genomic DNA using primers described previously (7), except that the upstream exon 10 primer was shorter by two bases at the 5' end. With these primers, observed amplification reaction yields were less than what we desired. In response, we decided to improve our results through the use of optimized amplification primers. The interrogated base is at position 1691; therefore, we used a pair of PCR primers, 5'-ATCAGAGCAGTTCAACCAGG-3' (upstream primer; bp 1490–1509 in GenBank Accession No. M14335.1 HUMVA) and 5'-ACACTCTAGACTTGCCTTCG-3' (downstream primer; bp 62701–62720 in emb PAC86F14 Accession No. 99572.1 HS86F14), to amplify a 423-bp fragment encompassing the same junction. Amplification primers were biotinylated at their 5' ends and synthesized using the cyanoethylphosphoramidite method on a Perseptive Biosystems Expedite Nucleic Acid Synthesis System (Perkin-Elmer). With our new PCR primers, the amplifications were more robust and consistent.
The capture probes had the following sequences: wild-type probe (bp 1667–1684 in GenBank Accession No. M14335.1 HUMFVA), 5'-H2N-CTGGACAGGCGAGGAATA-3'; mutant probe, 5'-H2N-CTGGACAGGCAAGGAATA-3'. The probes differ from each other by a single base, the 11th base from the 5' end. Oligonucleotide probes were synthesized using a C-12-amino-link phosphoramidite as the final ligand. Controls were synthetic single-stranded DNA fragments biotinylated at their 5' ends and had the following nucleotide sequences: wild-type control, 5'-biotin-AGGACAAAATACCTGTATTCCTCGCCTGTCCAGGGATCTGCTCTTACA-3' (bp 62907–62954 in Z99572.1, which is PAC 86F14 or HS86F14); and mutant control, 5'-biotin-AGGACAAAATACCTGTATTCCTTGCCTGTCCAGGGATCTGCTCTTACA-3', which differs from the wild-type control at the 23rd base from the 5' end.
The surface of the DNA-BindTM 96-well plate (Costar) is covalently coated with a layer of reactive N-oxysuccinimide esters, which will react with nucleophiles such as primary amines. Probes (50 pmol) were dissolved in phosphate buffer, pH 8.5 (100 µL/well), and added to the coated wells to allow specific coupling of amino-labeled oligonucleotides to the DNA-Bind plate during incubation for 18 h at 4 °C. Unbound oligonucleotides were removed by washing three times with wash buffer (50 mmol/L Tris, 0.45 mol/L NaCl, 1 mL/L Tween 20, pH 7.7). The unreacted DNA-Bind active groups were blocked by incubating the plates for 2 h with wash buffer containing 10 g/L bovine serum albumin and 100 mg/L spermidine (both from Sigma). The microtiter plates were washed three times as described above and then were dried under reduced pressure and sealed in Mylar bags with packs of desiccant. Plates were stored at 4 °C and protected from moisture and light.
Total genomic DNA was extracted from peripheral blood leukocytes using a Gentra Generation Capture Column Kit (Gentra Systems) and stored at 4 °C in a buffer containing 10 mmol/L Tris-HCl, 1 mmol/L disodium EDTA, pH 8.0. The DNAs used in this study were archived anonymized samples. All studies adhered to our company’s policies and to those of the American College of Medical Genetics guidelines for research on archived DNA (15).
Samples containing 50 ng of genomic DNA were amplified in 100-µL reactions containing 10 mmol/L Tris-HCl, pH 8.3 (at 25 °C), 0.5 µmol/L each PCR primer, 250 µmol/L each dNTP, 2.5 mmol/L MgCl2, 50 mmol/L KCl, and 5 U of AmpliTaq Gold DNA polymerase. All PCR reagents were obtained from Perkin-Elmer. Amplification was performed in a GeneAmp PCR System 9600 (Perkin-Elmer) using the following thermal profile: an initial hold at 95 °C for 10 min; 35 cycles at 94 °C for 1 min, 62 °C for 30 s, and 72 °C for 30 s; and a final extension hold at 72 °C for 10 min.
Clinical specimens (n = 256) were used for validation. Anonymized and blinded samples were also assayed by PCR-RFLP: Samples were PCR amplified, digested with MnlI endonuclease, and electrophoresed on 6% polyacrylamide vertical gels, followed by ethidium bromide staining.
We modified the microtiter plate ASO hybridization assay for FVL detection described by Zehnder et al. (13) as follows: After PCR amplification, double-stranded DNA was melted to allow hybridization with the capture probe. A 50-µL aliquot of each PCR sample was mixed with an equal volume of denaturation solution (0.8 mol/L NaOH, 0.04 g/L m-cresol purple). A series of experiments demonstrated that complete amplicon denaturation is a critical step in this assay (data not shown). Furthermore, we found that there was an inadequate description of the denaturation and neutralization conditions for our purposes in Zehnder et al. (13). Therefore, we conducted a series of experiments and instituted some controls (such as including the m-cresol purple as a pH indicator dye) to provide assurance that adequate denaturation followed by neutralization had been achieved (data not shown). Aliquots (25 µL) of the denatured, neutralized amplicon were then combined with 100 µL of hybridization solution [1.5x standard saline citrate (1x = 0.15 mol/L NaCl and 0.015 mol/L sodium citrate), 1 g/L sodium dodecyl sulfate, and 150 mL/L deionized formamide in 0.5 mol/L sodium phosphate buffer, pH 4.2] in each of four wells coated with the two probes, in duplicate. Hybridization was allowed to proceed for 2 h at 37 °C. Amplification products that did not hybridize were removed by washing the plates five times with wash buffer at room temperature.
Generation of colorimetric signal was achieved by adding to each well
100 µL of a 1:2 000 dilution of ExtrAvidin-Alkaline Phosphatase
conjugate (Sigma) in wash buffer containing 10 g/L bovine serum
albumin, 500 mg/L bovine 
-globulins, and 100 mg/L spermidine. After
a 60-min incubation at 37 °C, the unbound conjugate was washed five
times as described above. The signal was detected using the ELISA
Amplification System reagents (Life Technologies) according to
the manufacturer’s instructions. Absorbance was read at 490 nm in a
ELx 800 spectrophotometer (BioTek Instruments). In our assay,
four microwells were run for each sample: two wells (duplicates) were
coated with the wild-type capture probe, and two with the mutant probe.
The conditions of the assay were optimized for minimal hybridization
between probe-target containing a mismatch, at the same time favoring
efficient hybridization of fully complementary fragments. Arrays of
hybridization buffer compositions were tried, including some with
tetramethylammonium chloride, deionized formamide, and various
salt conditions. Tetramethylammonium chloride was tried because of its
known property of eliminating the dependence of melting point
(Tm) on G-C content. In our work, the
addition of this quaternary ammonium salt did not substantially improve
the signal-to-noise ratio.
Through optimization, wild-type samples yielded a strong signal in the
wells coated with the wild-type capture probe and a low, background
signal after hybridizing with the mutant probe. The opposite was
obtained for samples obtained from FVL homozygous individuals.
Heterozygous samples yielded signals of similar intensity when
hybridized with both allele-specific probes. A typical plate is shown
in Fig. 1
A. We analyzed 256 specimens with this reverse ASO hybridization
assay and by PCR-RFLP. The results were 100% concordant. Ratios of the
signal absorbance generated in the mutant probe well divided by the
absorbance generated in the wild-type probe well were calculated for
these same specimens. This is a simplified form of discriminant
analysis. Of the patient samples included in statistical calculations,
187 were wild type, 26 were FVL homozygous, and 43 were
heterozygous. Table 1
shows the mean values, the minimum and maximum ranges, and the
standard deviations for ratios of each genotype observed. Visual
inspection revealed no overlap between the genotypes called (see Fig. 1C
).
 View larger version (51K): [in a new window] |
Figure 1. FVL genotyping by reverse ASO hybridization assay
(A), plate map (B), and scattergram of
raw absorbance values (C).
(A), rows A, C,
E, and G are coated with wild-type ASO
capture probes; rows B, D,
F, and H are coated with mutant ASO
capture probes. A wild-type individual yields a high assay signal after
testing with the wild-type probe and a low signal after testing with
the mutant probe. The reverse is true for individuals who are
homozygous for the FVL allele. For individuals who are heterozygous for
the FVL allele, both allele-specific probes would yield signals of
similar intensity. (B), samples: 1,
negative control; 2, wild-type control;
3, heterozygous control; 4, homozygous
control; 5–8, wild-type patients; 9,
heterozygous patient; 10–17, wild-type patients;
18, heterozygous patient; 19–24,
wild-type patients. (C), scattergram of 490 nm light
absorbance (OD) values for wild-type, FVL heterozygous,
and FVL homozygous patients using the reverse ASO hybridization assay.
Raw absorbance read by the BioTek ELx 800 (BioTek Instruments) was
plotted from paired microwells: one coated with wild-type
(Normal) probe, and one coated with mutant probe (e.g.,
A1, B1).
|
, 23 pairs of negative (containing no DNA) samples; •,
wild-type samples; 
, heterozygotes, 
, homozygous mutant samples.
Kmeans cluster analysis results are described in the text. Complete
concordance was observed between microwell plate genotypes and those
assigned by PCR-RFLP.|
View this table: [in a new window] |
Table 1. Statistics for
ratios.1
|
Occasional differences in the colorimetric signals and ratios between microwells containing samples of identical genotypes were mostly attributable to differential amplification as can be seen on PCR-RFLP gels. To eliminate the need for microwell duplicates, pilot tests were conducted that called genotypes on the basis of one mutant-probe-coated microwell and one wild-type-probe-coated microwell. These experiments demonstrated that single wells worked as well as duplicate wells (data not shown). We used a relatively wide input range (40–300 ng) of extracted genomic DNA for simplicity because it is not practical to rigorously measure the DNA concentration of each sample. Input variability appeared to be smoothed through the use of a quantitative approach for genotype reporting.
Graphing corrected mutant to wild-type absorbance ratios and their
multiplicative inverse allowed us to infer that there was differential
hybridization between the PCR products and their ASOs (data not shown).
We speculated that batch-to-batch variation in coating of microwell
plates accounted for some fluctuation in signal intensity
(shown graphically in Fig. 1C
). To obtain similar colorimetric signals
from different batches of plates and reagents, we designed and
synthesized a set of controls. The wild-type and mutant controls were
synthetic oligonucleotides, each with a length of 49 bp and
biotinylated at the 5' end. The oligonucleotides behaved like denatured
amplified patient samples in that the wild-type control hybridized
specifically to the wild-type capture probe and the mutant control
specifically hybridized to the mutant capture probe. We also created a
heterozygous control by mixing equimolar solutions of wild-type and
mutant controls. From these experiments, we confirmed that differential
hybridization of the capture probes existed. Using equimolar solutions
of controls should theoretically yield equal colorimetric signals for
each capture probe. In reality, we obtained slightly lower signals with
the wild-type capture probe than with the mutant capture probe. By
adjusting the molar ratios of the wild-type and mutant controls, one
can obtain a perfect signal and ratio for the heterozygote control.
Cluster analysis was carried out on the absorbance values graphed in
Fig. 1C
after log10 transformation. Gaussian
distribution of values within each given cluster was observed.
Parsimonious construction of a similarity tree through the join
procedure in SYSTAT (SPSS, Inc.), using several quantitative
algorithms, correctly revealed the existence of four data clusters:
(a) no DNA added, (b) wild type, (c)
heterozygous, and (d) homozygous mutant for FVL. Kmeans
analysis demonstrated adequate cluster separation to distinguish 97.6%
of the 256 results (P <0.01); the remaining 6 samples were
not outliers in a repeat PCR and visualization. We believe this
demonstrates the accuracy and laboratory utility of this microwell
plate-based genotyping method.
Acknowledgments
Tables showing (a) cluster analysis, (b) cost analysis, and (c) a FORTRAN program to analyze data are available as a supplement at the Clinical Chemistry Online Web site (http://www.clinchem.org/content/vol46/issue8/).
References
- Gregg JP, Yamane AJ, Grody WW. Prevalence of the factor V-Leiden mutation in four distinct American ethnic populations. Am J Med Genet 1997;73:334-336. [Web of Science][Medline] [Order article via Infotrieve]
-
Griffin JH, Evatt B, Wideman C, Fernandez JA. Anticoagulant protein C pathway defective in majority of thrombophilic patients. Blood 1993;82:1989-1993.
[Abstract/Free Full Text] - Koster T, Rosendaal FR, de Ronde H, Briet E, Vandenbroucke JP, Bertina RM. Venous thrombosis due to poor anticoagulant response to activated protein C: Leiden thrombophilia study. Lancet 1993;342:1503-1506. [Web of Science][Medline] [Order article via Infotrieve]
-
Svensson PJ, Dahlback B. Resistance to activated protein C as a basis for venous thrombosis. N Engl J Med 1994;330:517-522.
[Abstract/Free Full Text] -
Sun X, Evatt B, Griffin JH. Blood coagulation factor Va abnormality associated with resistance to activated protein C in venous thrombophilia. Blood 1994;83:3120-3125.
[Abstract/Free Full Text] - Griffin JH. Control of coagulation reactions. In: Beutler E, ed. Williams hematology, 6th ed. New York: McGraw-Hill, 2000:in press..
- Bertina RM, Koeleman BP, Koster T, Rosendaal FR, Dirven RJ, de Ronde H, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature 1994;369:64-67. [Medline] [Order article via Infotrieve]
- Ridker PM, Miletich JP, Hennekens CH, Buring JE. Ethnic distribution of factor V Leiden in 4047 men and women. Implications for venous thromboembolism screening. JAMA 277:1305–7..
-
Kalafatis M, Rand MD, Mann KG. The mechanism of inactivation of human factor V and human factor Va by activated protein C. J Biol Chem 1994;269:31869-31880.
[Abstract/Free Full Text] -
Ridker PM, Hennekens CH, Lindpaintner, Stampfer MJ, Eisenberg PR, Miletich JP. Mutation in the gene coding for coagulation factor V and the risk of myocardial infarction, stroke, and venous thrombosis in apparently healthy men. N Engl J Med 1995;332:912-917.
[Abstract/Free Full Text] - Lee DH, Henderson PA, Blajchman MA. Prevalence of factor V Leiden in a Canadian blood donor population. Can Med Assoc J 1996;155:285-289. [Abstract]
- Solymoss S. Factor V Leiden. Who should be tested?. Can Med Assoc J 1996;155:296-298. [Abstract]
- Zehnder JL, Benson RC, Cheng S. A microplate allele-specific oligonucleotide hybridization assay for detection of factor V Leiden. Diagn Mol Pathol 1997;6:347-352. [Medline] [Order article via Infotrieve]
- Stanley CJ, Johannsson A, Self CH. Enzyme amplification can enhance both the speed and the sensitivity of immunoassays. J Immunol Methods 1985;83:89-95. [Medline] [Order article via Infotrieve]
- Anonymous. ACMG statement. Statement on storage and use of genetic materials. American College of Medical Genetics Storage of Genetic Materials Committee. Am J Hum Genet 1995;57:1499-1500. [Medline] [Order article via Infotrieve]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||