(HFE) Gene Mutations - Semantic Scholar

Clinical Chemistry 47:8 1384 –1389 (2001)

Molecular Diagnostics and Genetics

Solid-Phase Amplification for Detection of C282Y and H63D Hemochromatosis (HFE) Gene Mutations Mark S. Turner,1 Sarah Penning,1* Angela Sharp,2 Valentine J. Hyland,2 Ray Harris,3 C. Phillip Morris,1† and Angela van Daal1

Background: There is a need for simple, rapid, and inexpensive methods for the detection of single-nucleotide polymorphisms. Our aim was to develop a singletube ELISA-like PCR assay and evaluate it by detecting the common C282Y and H63D mutations found in the hemochromatosis gene (HFE) by use of clinical samples. Methods: The method, termed solid-phase amplification (SPA), involves dual liquid- and solid-phase amplification of a target sequence by the use of two PCR primers, one of which is in two forms: the first is covalently immobilized to the wall of a microwell, and the second is free in solution. During allele-specific amplification, both the free and solid-phase amplicons are labeled by incorporation of digoxigenin (DIG)dUTP. The amount of surface-bound amplicon is determined colorimetrically by the use of an alkaline phosphatase-anti-DIG-Fab conjugate and p-nitrophenyl phosphate. Results: Two different amplicon-labeling methods were evaluated. Analysis of 173 clinical samples for the C282Y and H63D HFE point mutations with SPA revealed that only one sample was incorrectly diagnosed, apparently because of operator error, when compared with conventional restriction fragment length polymorphism assay results. Conclusions: The SPA assay has potential for mediumscale mutation detection, having the advantage of being manipulatively simple and immediately adaptable for

1 Cooperative Research Centre for Diagnostic Technologies, Queensland University of Technology, GPO Box 2434, Brisbane, Qld 4001, Australia. 2 Molecular Genetics Laboratory, Queensland Health Pathology Service, Royal Brisbane Hospitals Complex, Herston, Brisbane, Qld 4006, Australia. 3 School of Pharmacy and Medical Sciences, University of South Australia, GPO Box 2471, Adelaide, SA 5000, Australia. *Current address: Institute for Molecular Bioscience, University of Queensland, Brisbane, Qld 4072, Australia. †Author for correspondence. Fax 617-3864-1534; e-mail [email protected]. Received February 14, 2001; accepted May 2, 2001.

use in clinical laboratories with existing ELISA instrumentation. © 2001 American Association for Clinical Chemistry

Hereditary hemochromatosis (HH)4 is a common genetic disorder with ⬃1 in 300 – 400 Caucasians being homozygous and ⬃1 in 8 –10 individuals being heterozygous for the C282Y HH mutation (1 ). HH is a disease caused by abnormally high iron absorption that causes progressively increasing iron overload of the major organs (primarily the liver, heart, and pancreas), which can lead to organ failure and premature death if left untreated. If the disease is identified, then treatment is simple, merely involving regular phlebotomy to remove the excess iron (2 ). The hemochromatosis gene, HFE, encodes a major histocompatibility complex class 1-like molecule (3 ). Two missense mutations have been identified in patients with HH: a G3 A transition at nucleotide 845, leading to a cysteine-to-tyrosine change at amino acid position 282 (C282Y), and a C3 G transversion at nucleotide 187, leading to a histidine-to-aspartate change at amino acid 63 (H63D). Of patients exhibiting typical clinical features of HH, 85–90% have the C282Y mutation, whereas the involvement of the H63D mutation in HH has not been confirmed (4 ). There is a need for simple, rapid, and inexpensive methods for mutation detection. Amplification procedures have become the method of choice because they are potentially more sensitive. These can be divided into those requiring postamplification manipulation and those, termed “homogeneous” DNA diagnostic assays, that do not. Common techniques requiring the application of postamplification steps include restriction fragment length polymorphism (RFLP) analysis of PCR products 4 Nonstandard abbreviations: HH, hereditary hemochromatosis; RFLP, restriction fragment length polymorphism; SPA, solid-phase amplification; TBST, Tris-buffered saline with Tween 20; and DIG, digoxigenin.

1384

Clinical Chemistry 47, No. 8, 2001

and sequencing. The necessity for transfer of amplification products for further analysis makes these techniques more complex and time-consuming, and increases the risk of contamination of subsequent tests with amplicons. Homogeneous DNA diagnostic assays include techniques such as the molecular beacon hybridization assay (5 ), the Invader assay (6 ), and the TaqMan assay (7 ). These methods are simpler in that they do not require transfer of the amplified product for analysis. However, many of these techniques are expensive and require the purchase of specialized, costly equipment; therefore, these techniques have not yet been adopted widely. The objective of our study was to develop a sensitive, specific, and inexpensive assay for the small- to medium-scale detection of single-nucleotide polymorphisms that could be carried out using nonspecialized equipment. We describe a near-homogeneous DNA diagnostic method (World Patent WO93/09250), termed solid-phase amplification (SPA), a method that involves the covalent attachment of one PCR primer to a microwell before amplification. Previous reports have shown that covalent attachment of a PCR primer to a solid support such as nylon (8 ), glass (9 ), or a microwell (10 ) before thermal cycling simplifies detection of the bound amplicon after washes. We demonstrate the potential of the SPA microwell assay for the detection of the C282Y and H63D hemochromatosis mutations from clinical samples by the use of allele-specific amplification.

Materials and Methods covalent oligonucleotide binding to microwells Primer (5 pmol) containing a 10-thymidine monophosphate 5⬘ spacer region was covalently linked to a Nunc NucleoLink microwell (Nalge Nunc International) by incubation in 100 ␮L per well of 10 mmol/L 1-ethyl-3-(3dimethylaminopropyl)-carbodiimide (Sigma), 10 mmol/L 1-methylimidazole (pH 7.0; Sigma) for 16 –20 h at 50 °C. Results in our laboratory have shown that coating with 5 pmol of primer gives optimum accessibility to the solidphase primer for hybridization to complimentary sequences (11 ). The wells were washed seven times with Tris-buffered saline with Tween 20 [(TBST); 100 mmol/L Tris-Cl (pH 7.5), 150 mmol/L NaCl, and 1 mL/L Tween 20] and three times with water and then air dried. The wells were blocked with 200 ␮L of 10 g/L bovine serum albumin (Sigma) for 1 h at room temperature. Wells were rinsed three times with TBS [100 mmol/L Tris-Cl (pH 7.5), 150 mmol/L NaCl], and then the PCR reaction constituents were added to the wells. Oligonucleotides were obtained from GeneWorks Pty Ltd and Genset Pacific Pty Ltd.

dna extraction Human genomic DNA was extracted from clinical samples consisting of 250 ␮L of EDTA blood spotted onto Whatman BFC180 paper (Schleicher and Schuell 903 sheets). The clinical samples were submitted specifically

1385

for hemochromatosis testing. The blood was dried overnight, and then fixed by soaking in methanol and airdried for a further 12–16 h. A 3-mm-diameter circle was punched from the card and added to 45 ␮L of buffer [20 mmol/L TrisCl (pH 8.4), 50 mmol/L KCl). The sample was mixed and incubated at 60 °C for 30 min and then 100 °C for 10 min. After cooling to 4 °C, the sample was centrifuged at 16 000g for 2 s in a microcentrifuge, and the supernatant was retained and stored at ⫺20 °C. Accurately quantified human genomic DNA was also purchased from Promega to determine the sensitivity of the assay.

spa pcr The C282Y mutation was amplified with 1.25 pmol of primer C282Y-cminv2 (5⬘-phosphate-TTTTTTTTTT ACT ACC CCC AGA ACA TCA CCA T-3⬘) and 10 pmol of either the wild-type-specific primer C282Y-wt-inv (5⬘GGC CTG GGT GCT CCA CCT GAC-3⬘) or the mutantallele-specific primer C282Y-mt-inv (5⬘-GGC CTG GGT GCT CCA CCT GAT-3⬘). The length of the amplified product was 187 bp. Similarly, the H63D mutation was amplified with 1.25 pmol of primer H63D-com2 (5⬘-phosphate-TTTTTTTTTT TCT TCC CTG CTC CCA CAA GAC CT-3⬘) and 10 pmol of either the primer specific for the wild-type allele, H63D-wt (5⬘-GAC CAG CTG TTC GTG TTC TAT GAA C-3⬘), or the primer specific for the mutant allele, H63Dmut (5⬘-GAC CAG CTG TTC GTG TTC TAT GAA G-3⬘). The length of the amplified product was 297 bp. Genomic DNA template (2.5 ␮L; equivalent to 5% of the extract from a 3-mm-diameter blood spot) was amplified in the following: 25 ␮L of 20 mM Tris-Cl, pH 8.4; 50

Fig. 1. Amplicon detection on the basis of either incorporated DIG-dUTP or fluorescein-labeled primer by the use of SPA. The top panel shows a 2% agarose gel containing the liquid-phase SPA products, whereas the graph shows the absorbance readings at 405 nm of the corresponding wells. SPA products from wells 1– 4 were detected by use of incorporated DIG-11-dUTP, whereas wells 5– 8 were detected by use of the labeled primer. Even-numbered wells contained the wild-type primer, whereas odd-numbered wells contained the mutant primer. Wells 1, 2, 5, and 6 contained 75 pg of plasmid containing the wild-type HFE sequence, whereas wells 3, 4, 7, and 8 contained no template.

1386

Turner et al.: Single-Nucleotide Polymorphism Detection Method

mM KCl; 1.5 mM MgCl2; 200 ␮M each of dATP, dCTP, and dGTP; 190 ␮mol/L of dTTP; 10 ␮mol/L of digoxigenin (DIG)-11-dUTP (Roche); and 0.75 U of Platinum Taq DNA polymerase (Life Technologies). The cycling reactions were done in Nunc Nucleolink wells in an MJ Research PTC-200 thermal cycler (GeneWorks). The reagents were overlaid with a drop of mineral oil, and the wells were sealed with Tape 8 (Nalge Nunc International). After an initial denaturation of 3 min at 96 °C, 40 cycles of 96 °C for 30 s, 60 °C for 30 s (for C282Y detection) or 52 °C for 30 s (for H63D detection), and then 72 °C for 30 s were carried out followed by a final extension step of 3 min at 72 °C. To compare amplicon-labeling methods, the following SPA reactions were set up. In 50 ␮L, the final volume contained 30 pmol of HHW (5⬘-6-carboxyfluorescein-TGG GGA AGA GCA GAG ATA TAC GAG-3⬘) or HHM (5⬘-6-carboxyfluorescein-TGG GGA AGA GCA GAG ATA TAC GAA-3⬘), 3.75 pmol of HHcom (5⬘-phosphateTTTTTTTTTT CAG ACG GTG AGG GCT CTA AAA CA-3⬘), 10 mM Tris-Cl, pH 8.3; 50 mM KCl; 1.5 mM MgCl2; 1.5 U of Taq DNA polymerase (Roche); 200 ␮M each of dATP, dCTP, dGTP, and dTTP; or 200 ␮mol/L dTTP was replaced with 190 ␮M of dTTP and 10 ␮M of DIG-11-dUTP. The template DNA (75 pg) consisted of a plasmid containing an insert spanning the C282 encoding region of wild-type HFE. The cycling reaction was performed as above except that the annealing temperature was 55 °C. The length of the amplified product was 254 bp. Detection of the DIG-dUTP- or fluorescein-labeled

amplicons bound to the well was performed as described below.

agarose gel electrophoresis The solution-phase amplified products from the SPA PCR were analyzed on 2% agarose gels in Tris-borate-EDTA buffer (90 mmol/L Tris-base, 90 mmol/L boric acid, 1 mmol/L EDTA) at 120 V for 30 min.

spa detection After amplification, the well contents were removed, and the wells were washed six times with 200 ␮L TBST with a 5-min soak step after the third wash. Anti-DIG- or antifluorescein-alkaline phosphatase conjugate (50 ␮L; Roche; 1:2500 by volume in TBST containing 10 g/L bovine serum albumin) was added to the wells, and they were shaken for 1 h. The wells were then washed six times with 200 ␮L TBST with a 5-min soak step after the third wash. Immobilized product was detected using p-nitrophenyl phosphate substrate (0.05 mg in 50 ␮L of 0.2 mol/L Tris-Cl per well; Sigma). Color development was monitored at 405 nm using a Biomek microwell plate-reader (Beckman Coulter) up to 1 h. The mean absorbance at 405 nm of an empty microwell was 0.042, and this value was subtracted from all absorbance readings.

rflp analysis One-half of a 3-mm-diameter circle from a blood-spot card was placed in a 0.6-mL PCR tube with 46 ␮L of 1⫻ PCR buffer (Roche) and treated at 60 °C for 30 min, then

Fig. 2. Sensitivity of SPA DIG-11-dUTP. (A), scatterplot of log A/G ratios determined from the use of various amounts of plasmid template containing part of the HFE gene in SPA assay. Templates consisted of linearized plasmid containing the wild-type HFE sequence alone (⽧), an equal amount of plasmids containing wild-type and mutant HFE sequences (〫), or plasmid containing the mutant HFE sequence alone ( ). Log values of ⫺0.5 and 0.5 were applied as cutoff points for each genotype. (B), detection of the wild-type HFE gene by use of various amounts of human genomic DNA as template. The top panel shows a 2% agarose gel containing the liquid-phase SPA products, and the graph shows the absorbance readings at 405 nm of the corresponding wells. Various template copy numbers of human genomic DNA (Promega), as well as 5% and 0.5% of a 3-mm-diameter blood spot, were assayed.

1387


Fig. 3. SPA interrogation of the HFE C282Y (A) and H63D (B) sites by use of DNA from blood spots by detecting DIG-dUTP labeled amplicon. The top panels show 2% agarose gels containing the liquid-phase SPA products, whereas the graphs show the absorbance readings at 405 nm of the corresponding wells. Odd-numbered wells contained the wild-type-specific primer, whereas evennumbered wells contained the mutant-specific primer. Templates added to the wells were as follows: wells 1 and 2, no template; wells 3 and 4, wild-type genomic DNA; wells 5 and 6, heterozygous genomic DNA; and wells 7 and 8, homozygous mutant genomic DNA.

99.9 °C for 10 min. Mastermix (4 ␮L) containing dNTPs, primers, MgCl2, and Taq polymerase was added to the same tube. Primers were those described in 1996 by Feder et al. (3 ), except for H63DMF1. The use of H63DMF1 enabled the multiplexed PCR products to be digested together with RsaI and BclI. The C282Y mutation region was amplified with 40 pmol of primers HCF (5⬘-TGG CAA GGG TAA ACA GAT CC-3⬘) and HCR (5⬘CTC AGG CAC TCC TCT CAA CC-3⬘). The H63D mutation region was amplified with 20 pmol of primers H63DMF1 (5⬘GCT TTG GCC TAC GTG GAT GAT CA-3⬘) and H63DR (5⬘-CTT GCT GTG GTT GTG AAT TTC-3⬘). DNA template was amplified in a final concentration of 10 mM Tris-Cl, pH 8.3; 50 mM KCl; 1.5 mM MgCl2; 200 ␮M each of dATP, dCTP, dGTP, dTTP; and 2 U of Platinum Taq DNA polymerase (Life Technologies). The reactions were done in a Perkin-Elmer GeneAmp PCR System 9700 thermal cycler (Perkin-Elmer). After an initial denaturation of 4 min at 94 °C, 10 cycles (touchdown) of 94 °C for 1 min, 65 °C for 30 s with a 1 °C drop per successive cycle, and 72 °C for 30 s, followed by 25 cycles of 94 °C for 1 min, 53 °C for 30 s, and 72 °C for 30 s were carried out followed by a final extension step of 7 min at 72 °C. Amplified products (15 ␮L) were then digested at 37 °C for 2 h with 5 units of RsaI and BclI restriction enzymes in NEB buffer number 2 (New England Biolabs). The resulting fragments were separated using 10% polyacrylamide gel electrophoresis and visualized with ethidium bromide to determine the genotype. Further details about this RFLP assay, such as the banding profiles obtained for the different genotypes, may be obtained on request.

Results amplicon-labeling methods We first compared two different ways of labeling amplimers during the SPA reaction. The first was the incor-

poration of DIG-dUTP into the amplicon during synthesis, and the second was by using a fluorescein-labeled primer. Using the HFE gene, we found that the background readings where no amplicon was obtained were unacceptably high when the fluorescein-labeled primer was used, but not when DIG-dUTP was used (Fig. 1). There was a ⬃4- or ⬃43-fold increase in signals in wells that had amplicon compared with those that did not when the fluorescein-labeled primer or DIG-dUTP was used, respectively (Fig. 1). It is likely that primer-dimer formation was the cause of the high background when the fluorescein-labeled primer was used. Therefore, detection of amplimers was determined from this point forward by the incorporation of DIG-dUTP.

sensitivity and specificity Using linearized plasmid templates, we found that the assay could reliably discriminate among the three C282Y genotypes from 1 ⫻ 106 copies to between 1 ⫻ 103 and 1 ⫻ 102 copies (Fig. 2A). The results are presented as log transformations of the signal from the mutant-specific reaction divided by the signal from the wild-type-specific reaction, and arbitrary limits were drawn at log values of ⫺0.5 and 0.5 to separate the three genotype clusters as has been used previously (12 ). Using 1 ⫻ 102 copies or less of plasmid template, we obtained little or no amplicon after examining the liquid phase by agarose gel electrophoresis (data not shown). The assay was more sensitive when we used a commercially available, purified genomic-DNA template (Promega), where as few as 1 ⫻ 102 copies could be reliably detected (Fig. 2B). Because this assay was being developed to analyze human clinical samples, it is important to determine whether it is sensitive and specific in the detection of HFE genotypes from blood spots, which are a common source of genomic DNA for genetic analysis. The SPA assay is

1388

Turner et al.: Single-Nucleotide Polymorphism Detection Method

Fig. 4. Scatterplots of log A/G ratios (A) and G/C ratios (B) determined from clinical samples screened for the C282Y and H63D mutations, respectively. From left to right, the three distinct clusters represent the following: ⽧, homozygous wild type; 〫, heterozygous; , homozygous mutant. When applying log values of ⫺0.5 and 0.5 as cutoff points for each genotype, one C282Y heterozygous sample (sample number 83) was incorrectly diagnosed as a homozygous mutant.

sufficiently sensitive to generate adequate signal from 5% of the DNA from a 3-mm blood spot and also a 10-fold dilution of this extract (i.e., 1/200 of a blood spot; Fig. 2B). The SPA assay is also allele specific at the determined annealing temperature with the use of blood spots because it can easily discriminate the homozygous wild type from heterozygous from homozygous mutant for both the C282Y and H63D mutations (Fig. 3).

specific SPA well read 0.505). After subsequent retesting, it scored incorrectly as a heterozygote and then correctly as wild type. Examination of the liquid-phase PCR products from the SPA well revealed that this sample was indeed wild type for H63D, which agreed with the RFLP result. It is not known what caused this sample to score once as a heterozygote.

clinical sample analysis

Despite the fact that PCR has revolutionized our ability to sensitively and specifically undertake mutation analysis of clinical specimens, the labor intensive, low-capacity process of product detection after PCR amplification has meant that PCR has not been routinely applied in many clinical laboratories. We have developed a microwellplate system, termed SPA, that fully integrates amplification and detection in the same microwell. This is achieved by covalently attaching one primer to the microwell before amplification, thereby allowing colorimetric detection of a solid-bound amplimer carrying a detector tag after washes. The yield of solid-phase amplicons is greatly increased by simultaneous liquid- and solid-phase amplification of targets. The former is affected by the addition of a small amount of the same primer as used in the solid phase. This also offers the advantage that amplicons can be conventionally analyzed if required. We chose HH to validate the SPA assay because of the relatively prevalent nature of these mutations and because this disease can be treated easily. In addition, debate on whether large-scale

Discussion We analyzed 173 clinical samples that had been previously typed for the C282Y and H63D mutations of the HFE gene using a RFLP analysis method. Only those SPA reactions that had an A405nm ⬎0.5 were included in the analysis, which led to four SPA reactions being identified for retesting. The results for the samples are presented as log transformations of the signal from the mutant-specific reaction divided by the signal from the wild-type-specific reaction (Fig. 4). Of the 342 results obtained, only 1 result (sample number 83 tested for the C282Y mutation) was incorrectly diagnosed (Fig. 4). When this sample was tested again, the result agreed with the RFLP analysis. Operator error was the probable contributor to the SPA signal variation of this sample because it appears most likely that template was not added to both SPA wells. The only other problem observed during the screening was with the inconsistent results obtained for detection of the H63D mutation for sample number 112. It was first scored correctly as a wild-type with a weak signal (the wild-type-

1389


screening of populations for HH should take place is currently under way (13 ). Direct solid-phase amplification enabled amplification and detection of the mutations, C282Y and H63D, found in HH patients, and from 173 clinical samples we diagnosed 99.7% of results correctly using SPA. Conventional genetic testing, including that for HH, is based on RFLP analysis. This method is subjective in contrast to SPA, which has quantitative results, thereby allowing arbitrary limits to be set to separate the genotypes. Another advantage of SPA is that it eliminates the need for agarose gel analysis of PCR products, therefore eliminating the use of mutagenic chemicals and, in some cases, an overnight enzymatic digestion of the amplicon for RFLP analysis. In a 96-microwell plate format using precoated wells, the SPA assay allows ⬎44 samples to be analyzed in 5 h, including cycling and amplicon detection. Because only ⬃100 min of “hands-on” time is required to process a 96-microwell plate, several microwell plates could be run simultaneously by one technician. Additionally, according to the manufacturer, coated Nunc NucleoLink microwells have been stable when dried and stored at 4 °C for at least 10 months (Nalge Nunc International). The assay also has the potential of being automated with widely available, existing ELISA equipment, thereby facilitating larger-scale screening by use of 384- and 1536-microwell plate formats. SPA requires no transfer steps because the entire assay from amplification to detection is performed in the same microwell. A similar procedure, termed the sequential nucleic acid amplification and capture assay (11 ), is also carried out entirely in the same microwell, but differs from SPA in that the amplification occurs solely in the solution phase and the amplicon is detected after hybridization to a covalently bound capture probe. SPA and sequential nucleic acid amplification and capture have an advantage over nonhomogeneous assays such as minisequencing methods (12 ) and dynamic DNA denaturation (14 ) assays because they are true single-microwell assays and require a lower number of postamplification manipulations. Solid-phase PCR has also been used in a single-well multiplex format with time-resolved fluorometry for detection of HLA-B27 alleles (15 ). At the present time in Australia, the SPA assay has been calculated to cost approximately US$2.46 in reagents per single-nucleotide polymorphism with no setup cost in a laboratory with a PCR instrument and a microwell-plate reader. In comparison, the TaqMan assay (Applied Biosystems) costs substantially less in reagents at approximately US$0.94 per single-nucleotide polymorphism (2000 assay scale), but has a substantial setup cost of US$77 500 for the ABI PRISM 7700 Sequence Detector (Applied Biosystems). Therefore, SPA would appear to be most cost-effective for clinical laboratories that carry out small- to medium-scale testing. The SPA reaction is broadly applicable for the detec-

tion of any gene alteration, whether it be a point mutation, a deletion, an inversion, or an insertion. We have used SPA to detect other common genetic mutations, including those in the medium-chain acyl-CoA dehydrogenase, prothrombin, coagulation factor V, and methylenetetrahydrofolate reductase genes (data not shown). In conclusion, SPA is a rapid, simple, and inexpensive method for the detection of genetic mutations and has potential for automated screening across a wide range of genetic variants.

We would like to thank Yu Pei Tan for technical assistance with SPA. We also appreciate comments from Phil Giffard on the manuscript. This work was supported by the Australian Cooperative Research Centres Program.

References 1. Cullen LM, Summerville L, Glassick TV, Crawford DHG, Powell LC, Jazwinska EC. Neonatal screening for the hemochromatosis defect. Blood 1997;90:4236 –7. 2. Niederau C, Fischer R, Sonnenberg A, Stremmel W, Trampisch HJ, Strohmeyer G. Survival and causes of death in cirrhotic and noncirrhotic patients with primary hemochromatosis. N Engl J Med; 1985;313:1256 – 62. 3. Feder JN, Gnirke A, Thomas W, Tsuchihashi Z, Ruddy DA, Basava A, et al. A novel MHC class I-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet 1996;13:399 – 408. 4. Powell LW, Subramaniam VN, Yapp YR. Haemochromatosis in the new millenium. J Hepatol 2000;32:48 – 62. 5. Tyagi S, Bratu DP, Kramer FR. Multicolor molecular beacons for allele discrimination. Nat Biotechnol 1998;16:49 –53. 6. Kwiatkowski RW, Lyamichev V, de Arruda M, Neri B. Clinical, genetic and pharmacogenetic applications of the invader assay. Mol Diagn 1999;4:353– 64. 7. Livak K, Marmaro J, Todd JA. Towards fully automated genomewide polymorphism screening. Nat Genet 1995;9:341–2. 8. Lockley AK, Jones CG, Bruce JS, Franklin SJ, Bardsley RG. Colorimetric detection of immobilised PCR products generated on a solid support. Nucleic Acids Res 1997;25:1313– 4. 9. Adessi C, Matton G, Ayala G, Turcatti G, Mermod J-J, Mayer P, Kawashima E. Solid phase DNA amplification: characterisation of primer attachment and amplification mechanisms. Nucleic Acids Res 2000;28:e87. 10. Oroskar AA, Rasmussen SE, Rasmussen HN, Rasmussen SR, Sullivan BM, Johansson A. Detection of immobilized amplicons by ELISA-like techniques. Clin Chem 1996;42:1547–55. 11. Somodevilla-Torres M, Timms P, Harris R, Morris CP, van Daal A. Solid-phase amplification and detection: a single-tube diagnostic assay for infectious agents. Mol Diagn 2001;in press. 12. Pecheniuk NM, Marsh NA, Walsh TP. Multiple analysis of three common genetic alterations associated with thrombophilia. Blood Coagul Fibrinolysis 2000;11:183–9. 13. Barton JC, Acton RT. Population screening for hemochromatosis: has the time finally come? Curr Gastroenterol Rep 2000;2:18 –26. 14. Howell WM. Dynamic allele-specific hybridization. Nat Biotechnol 1999;17:87– 8. 15. Sjo¨roos M, Ilonen J, Lo¨vgren T. Solid-phase PCR with hybridization and time-resolved fluorometry for detection of HLA-B27. Clin Chem 2001;47:498 –504.