Coagulation and Transfusion Medicine / SNPS BY NANOCHIP, LIGHTCYCLER, AND RFLP

Diagnostic Single Nucleotide Polymorphism Analysis of Factor V Leiden and Prothrombin 20210G>A A Comparison of the Nanogen Electronic Microarray With Restriction Enzyme Digestion and the Roche LightCycler
Iris Schrijver, MD,1,2 Marla J. Lay,2 and James L. Zehnder, MD1,2
Key Words: Thrombophilia; Factor V; Prothrombin; Factor II; Microarray; LightCycler; NanoChip; Restriction fragment length polymorphism; RFLP
DOI: 10.1309/3VTR7TL2X7TXL0QY

Abstract
Genetic thrombosis risk factors include a sequence variant in the prothrombin gene (20210G>A) and factor V Leiden (1691G>A). These single nucleotide polymorphisms can be diagnosed with restriction fragment length polymorphism (RFLP) analysis, fluorescent genotyping on the LightCycler (Roche Diagnostics, Indianapolis, IN), and microarray-based testing on the novel NanoChip electronic microarray (NanoChip Molecular Biology Workstation, Nanogen, San Diego, CA). We compared these methods for accuracy, time to results, throughput, and interpretation. Results from 789 of 800 individual amplicons analyzed on the NanoChip were in complete agreement with the other assays. Eleven were no calls (uninterpreted by the NanoChip system) resulting from failed polymerase chain reaction amplifications. Although the NanoChip System, when used in a lowthroughput setting, requires more overall time than the LightCycler, it is nearly equivalent per genotyping call. Owing to minimal sample handling, assay results are more reliable on the NanoChip platform and on the LightCycler than with RFLP. The NanoChip assay is reliable and may be especially valuable to laboratories with a large volume of thrombophilia test requests.

Thromboembolic events have a combined frequency of 1 to 2 per 1,000 individuals per year and demonstrate an increased incidence with age.1 Mutations in genes that encode blood coagulation factors can predispose to thrombotic disorders. A major inherited risk factor is mutation 1691G>A in the factor V gene on chromosome 1q23 (factor V Leiden). This mutation results in resistance to factor Va cleavage by activated protein C2,3 and results in persistence of the active state, increased thrombin synthesis, and higher levels of prothrombin fragment 1.2.4,5 The mutant protein confers a 5to 10-fold risk of deep venous thrombosis (DVT) in heterozygous individuals and a 50- to 100-fold risk in homozygotes.6 Factor V Leiden is identified in approximately 20% of patients diagnosed with venous thromboembolism.7 Another risk factor, associated with a 3-fold increased risk of DVT, is sequence variant 20210G>A in the 3'-UTR (untranslated region) of the prothrombin gene on chromosome 11. The general carrier frequency of this mutation is 1% to 3% but approaches 6% to 18% in patients with DVT.7 For reasons that remain obscure, factor V Leiden and prothrombin 20210G>A are coinherited more often than expected. 8 Genetic analysis of factor V Leiden may be used in cases with equivocal results by a modified activated protein C resistance screening assay9 or to differentiate heterozygosity from homozygosity to provide appropriate clinical management.10 In addition, DNA analysis for factor V Leiden and other DVT risk factors including prothrombin 20210G>A is increasingly performed routinely in conjunction with coagulation assays to definitively determine the underlying risk factors for DVT. To evaluate a newly developed electronic microarray assay (Nanogen, San Diego, CA) for factor V Leiden and prothrombin 20210G>A, we performed a comparison for
American Society for Clinical Pathology

490 490

Am J Clin Pathol 2003;119:490-496

DOI: 10.1309/3VTR7TL2X7TXL0QY

Coagulation and Transfusion Medicine / ORIGINAL ARTICLE

accuracy, assay duration, hands-on time, and performance complexity with 2 diagnostic assays used in the molecular pathology laboratory. The first, a multiplexed polymerase chain reaction (PCR), followed by restriction digestion and polyacrylamide gel electrophoresis, was used before implementation of separate factor V Leiden and prothrombin 20210G>A assays on the Roche LightCycler (Roche Diagnostics, Indianapolis, IN), our current method.11,12 The LightCycler assay enables amplification of genomic DNA in thin borosilicate glass capillaries under very rapid cycling conditions. During PCR, the number of amplicons generated is measured directly by increasing fluorescence, which is created by fluorescence resonance energy transfer from one labeled probe to another. The 2 probes, one of which overlaps the interrogated single nucleotide polymorphism (SNP), are hybridized to neighboring areas of the amplicon during PCR. After the amplification reaction, slowly increasing temperatures cause dissociation of the hybridized probes and subsequent reduction in the fluorescent signal. This provides a melting curve, which demonstrates earlier dissociation owing to a nonperfect match in the presence of the mutation.13 The dissociation temperatures are sequencespecific and highly reproducible, so that wild-type samples can be distinguished readily from heterozygous or homozygous sequences. The NanoChip Molecular Biology Workstation (Nanogen) consists of a loader that electronically addresses a previously amplified biotinylated PCR product to specified locations on the NanoChip array. Subsequently, wild-type and mutant reporter oligonucleotides with different fluorescent labels (Cy3 and Cy5, respectively) are passively hybridized to the NanoChip array. The hybridization is strengthened with stabilizer oligonucleotides. Detection of fluorescent signals occurs on the reader module, and computerized results are displayed in histograms, tables, and charts. Wild-type, homozygous, and heterozygous samples are identified based on relative red-green fluorescence.14

Valencia, CA). Samples from the other 200 patients were assayed using the LightCycler instrument. Blood samples that were assayed on the LightCycler were extracted on the MagnaPure instrument according to the manufacturers recommendations (Roche Diagnostics). PCR Amplification and Restriction Enzyme Analysis A 267-base-pair (bp) PCR product, containing a 121-bp segment of factor V exon 10 and a fragment of intron 10, was generated with a primer pair, originally described by Bertina et al,2 that flanks the factor V Leiden mutation. The primers for amplification of a part of exon 14 and the partial 3'-UTR of the prothrombin gene were derived from Poort et al 15 and encompass the prothrombin sequence variant 20210G>A. The factor V and prothrombin fragments were amplified in a multiplexed PCR reaction with AmpliTaq Gold DNA polymerase, 250 U, 5 U/L (Applied Biosystems, Foster City, CA) on a 9600 or 9700 Perkin Elmer thermocycler (Perkin Elmer, Shelton, CT). PCR cycling conditions comprised a touchdown procedure16 and included 1 cycle at 95C for 10 minutes, 20 cycles at 95C for 30 seconds, 62C for 30 seconds with a 1C decrease in temperature per cycle, and 72C for 30 seconds. This was followed by 15 cycles of 95C for 30 seconds, 42C for 30 seconds, and 72C for 30 seconds. There was a 5-minute extension at 72C. After PCR amplification, 5 U of restriction enzyme MnlI and 20 U of HindIII (New England Biolabs, Beverly, MA) were added to each PCR tube.17 Following a 60-minute incubation at 37C, 2 L of each sample was electrophoresed for 60 minutes on a 10% polyacrylamide gel at 200 V. The gels were stained with the GelStar staining solution (BioWhittaker, Walkersville, MD), and bands were photographed under UV illumination. The expected size for a homozygous wild-type prothrombin amplicon is 345 bp, whereas the homozygous 20210G>A is cut by HindIII and expected at 322 and 23 bp. Restriction analysis of the wild-type factor V fragment yields wild-type bands of 163, 67, and 37 bp. Factor V Leiden, however, abandons 1 of the 2 restriction sites, resulting in fragments of 200 and 67 bp18 Image 1A. LightCycler Amplification The LightCyclerfactor V Leiden and the LightCyclerprothrombin mutation detection sets (Roche) were used in conjunction with the MagnaPure instrument to prepare samples for subsequent amplification on the LightCycler instrument according to the manufacturers guidelines. In brief, approximately 250 ng of patient DNA was added to a 5-L PCR mix provided in the set, which, among other proprietary reagents, contains the forward and reverse primers for either the factor V or the prothrombin amplification reaction and fluorescently labeled donor and acceptor hybridization probes. The LightCycler conditions for both
Am J Clin Pathol 2003;119:490-496
491
DOI: 10.1309/3VTR7TL2X7TXL0QY

Materials and Methods

Samples Four hundred individual patients who were tested previously for the presence of factor V Leiden (1691G>A) and prothrombin gene variant 20210G>A in the Stanford University Medical Center Molecular Pathology Laboratory, Stanford, CA, were included in the study. Samples from 200 of these patients were analyzed initially by restriction enzyme analysis of multiplexed factor Vprothrombin PCR amplicons. DNA was extracted from peripheral blood leukocytes using the Puregene DNA purification kit (Gentra Systems, Minneapolis, MN) or the Qiagen spin column (Qiagen,
American Society for Clinical Pathology

491 491

Schrijver et al / SNPS BY NANOCHIP, LIGHTCYCLER, AND RFLP

assays are identical. The samples were denatured at 95C for 10 minutes, followed by 40 cycles of 95C for 15 seconds, 55C for 10 seconds, and 72C for 10 seconds. This was followed by 1 melting curve cycle, which begins with an increase in temperature to 95C to denature the amplicons, followed by a 30-second pause at 40C and a slow increase at a temperature rate of 0.1C per second to 80C. The final cooling cycle is at 40C for 30 seconds. The cycles are followed by a display on the computer screen of the fluorescence intensity vs temperature and derived melting curves Image 1B . The typical temperature for prothrombin 20210G is 58C, and for 20210A it is 49C. The probe melting temperature for wild-type factor V is 65C, whereas factor V Leiden it is 57C. PCR Amplification and NanoChip Platform Analysis The extracted DNA, which previously was used with the restriction enzyme assay or the LightCycler analysis, now was amplified with the Nanogen primer sets for multiplexed amplification of the factor V and prothrombin fragments that encompass the SNPs of interest. PCR amplification with forward primers and biotinylated reverse primers was performed using AmpliTaq Gold DNA polymerase 250 U, 5 U/L, and the four 2'-deoxynucleotide 5'-triphosphates (Amersham Pharmacia, Piscataway, NJ) on a 9600 Perkin Elmer instrument. Amplification conditions were as follows: 95C for 10 minutes, 40 cycles of 94C for 20 seconds, 55C for 20 seconds, and 72C for 45 seconds, and a final extension of 72C for 5 minutes. Following amplification, the PCR samples were desalted on a Millipore Multi Screen 96-well desalting plate (Millipore, Bedford, MA) and mixed with a 50-mmol/L concentration of histidine. A volume of 60 L of each sample was transferred into a Nunc 96-well plate (Nalge Nunc, Rochester, NY) and prepared for the NanoChip loader according to the manufacturers guidelines. The samples were addressed electronically to pads on the NanoChip cartridge. Each sample was run in duplicate and placed on adjacent pads. On completion of the loader map file, the cartridges were rinsed with a high-salt buffer (50-mmol/L concentration of sodium phosphate, pH 7.4, 500-mmol/L concentration of sodium chloride). Two reporter-stabilizer mixes for the factor V and prothrombin assays were prepared as specified in the manufacturers protocol. First, the factor V mix was pipetted onto the microarray and incubated for 3 minutes, followed by a wash with the high-salt buffer. Next, the cartridge was placed onto the NanoChip reader instrument and analyzed at the factor V Leiden discrimination temperature (31C). The hybridized factor V Leiden probe was removed from the cartridge by a series of washes with 0.1N NaOH, deionized H2O and high-salt buffer. The procedure was repeated using the prothrombin reporter mix. Results were interpreted at a discrimination temperature of 36C Image 1C.
492 492

Results
Accuracy A novel diagnostic system for identification of factor V Leiden 1691G>A and prothrombin 20210G>A was compared with restriction enzyme digestion and the LightCycler assays. In our comparison with the 200 samples originally genotyped by restriction enzyme analysis, results were obtained in 197 of 200 samples for factor V and 199 of 200 for prothrombin. All obtained results were identical between the assays. Failure of amplification for both factor V and prothrombin occurred in a single sample and may have been because the DNA was of low concentration and several years old. Testing of this sample could not be repeated because no DNA was left after the original comparison. Whereas the prothrombin segment in the multiplex PCR reaction was amplified, 2 samples failed factor V amplification, even when a separate factor V primer set was used. In our comparison with the 200 samples originally genotyped by the LightCycler, results were obtained in 195 of 200 samples for factor V and 198 of 200 for prothrombin. All results obtained on the NanoChip platform were entirely congruent with the LightCycler results. To assess whether failure was run-specific or due to sample compromise resulting from DNA degradation or the presence of PCR inhibitors, we attempted reamplification on both the NanoChip System and on the LightCycler of 5 samples that initially failed using the Nanogen primers. Of these 5 samples, testing on 2 could not be repeated because no sample was left, and 3 were amplified successfully only with the LightCycler primer set. In 1 result on the NanoChip System, a nontemplate control sample produced an indeterminate result. Of the 2 sample pads, the pad immediately next to the positive sample was weakly positive, whereas the second pad remained negative. When the sample was tested again, it gave a negative result. Assay Duration and Throughput The time requirement for DNA extraction was similar for all 3 assays. The restriction fragment length polymorphism (RFLP) approach requires a PCR setup in which we use 28 patient samples, a control heterozygous for both mutations of interest, and a control without template. After the PCR amplification, restriction enzymes are added to each tube. The samples are incubated at 37C for 1 hour. During this time, 30 minutes are used for the preparation of 4 polyacrylamide gels Figure 1 Table 1. The samples are loaded onto the gels and electrophoresed, followed by interpretation of the results. The LightCycler analysis is preceded by a PCR preparation for 32 samples. This is the maximum number of samples that can be analyzed per run on a single Light-Cycler instrument (Figure 1, Table 1). We include a DNA negative control and a factor V Leiden and/or a prothrombin 20210G>A
American Society for Clinical Pathology

Am J Clin Pathol 2003;119:490-496

DOI: 10.1309/3VTR7TL2X7TXL0QY

Coagulation and Transfusion Medicine / ORIGINAL ARTICLE

A
1 2 3 4 5 6 7 8 9 10
345 bp 322 bp 200 bp 163 bp

Image 1 A, Restriction fragment length polymorphism analysis. The products of the factor V Leiden/prothrombin 20210G>A multiplex polymerase chain reaction (PCR) reaction are visualized on polyacrylamide gel electrophoresis after restriction digestion. Bands under 100 base pairs (bp) are not shown. Lane 1, 100-bp DNA ladder; lanes 2-5, normal factor V (163 bp) and prothrombin (345 bp); lanes 6 and 7 , homozygous wild-type for factor V Leiden (163 bp) and heterozygous for 20210G>A (345- and 322-bp bands due to digestion with HindIII); lane 8, homozygous for 20210G (345 bp) and heterozygous for factor V Leiden (163 and 200 bp after digestion with MnlI); lane 9, control sample, heterozygous for factor V Leiden and prothrombin 20210G>A; lane 10, negative control. B, LightCycler analysis. Melting profile of a sample that is homozygous wild-type for factor V Leiden (1691G, green), superimposed on a heterozygous control (1691G>A, blue). The typical melting temperature for the wild-type sequence is 65C, but in the presence of factor V Leiden, melting occurs at 57C. C, NanoChip display. One mode of data presentation in the NanoChip prothrombin 20210G>A assay. Heterozygous samples are made up of half red and half green circles, whereas homozygous mutant samples are red only and homozygous wild-type samples are green only. This display is an approximation of the software analysis, which is more precisely summarized in detailed tables and graphs. Differences in signal strength most likely reflect variation in PCR efficiency. Ref, positive heterozygous prothrombin control, supplied by the manufacturer; Pt, patient who is heterozygous for prothrombin 20210G>A and homozygous for factor V Leiden. For proprietary information, see the text.

Ref Pt

heterozygous control in each analysis. After computer programming of the MagnaPure instrument, the MagnaPure loads the PCR reagents and patient DNA into the appropriate capillaries in about 35 minutes. The samples undergo a
American Society for Clinical Pathology

temperature cycle program on the LightCycler, and the results are interpreted directly after completion. The Nanogen assay begins with sample preparation and a multiplexed PCR reaction. The amplicons are desalted on a 96well plate and transferred to a Nanogen loader plate, after which user-defined computer sample-map files are created and controls prepared. The sample plate and the cartridges are put into the loader and processed for 270 minutes (throughput depends on the number of samples being tested). During this time, the samples are addressed electronically to their designated positions on the NanoChip array. The subsequent hybridization with the factor V probe occurs outside the loader, after which the cartridge is moved into the reader, scanned, and interpreted by the Nanogen software. The hybridization and reader steps are repeated with the prothrombin probe. Loader flush and shutdown are completed during interpretation of the results (Figure 1, Table 1). It should be noted that Nanogen claims 96 samples can be loaded in 295 minutes owing to parallel fluidic processing. However, we did not validate this by testing 96 samples at one time. In our comparison, we tested only 48 samples (96 calls) on 1 cartridge, although the Nanogen loader can hold up to 4
Am J Clin Pathol 2003;119:490-496
493
DOI: 10.1309/3VTR7TL2X7TXL0QY

493 493

Schrijver et al / SNPS BY NANOCHIP, LIGHTCYCLER, AND RFLP

Hours 0 RFLP 1 2 3 4 5 6 7 8 9 10 11

control sample, which was homozygous for factor V Leiden and heterozygous for the prothrombin mutation. Hands-on Time The time of actual involvement by the molecular technologist is 1 hour and 40 minutes for the RFLP assay for 30 samples (60 genotyping calls) and 1 hour and 20 minutes for LightCycler analysis for 32 samples (32 genotyping calls) (Figure 1, Table 1). Per sample, this translates to 3.3 minutes for the restriction enzyme assay and 2.5 minutes for the LightCycler. In the NanoChip System evaluation, 2 hours and 35 minutes of technologist time were required for completion of the assay. Hands-on time for this assay is 3.2 minutes per sample because we used 48 samples (96 genotyping calls) per Nanochip cartridge and 2 electronic pads per sample. The hands-on time per genotyping call translates to 1.7 minutes for the RFLP assay, 2.5 minutes for the LightCycler, and 1.6 minutes for the NanoChip System.

LightCycler NanoChip

Figure 1 Comparison of total assay time and hands-on time for restriction fragment length polymorphism (RFLP) analysis, LightCycler amplification, and the NanoChip assay. The time, displayed in hours, indicates the duration of each assay. Shaded boxes represent hands-on-time, and clear boxes represent the time during which the assay continues without direct involvement. For proprietary information, see the text.

cartridges. Each electronic microarray has 100 locations and can be addressed singly, but we addressed each sample in duplicate for quality control purposes (Figure 1, Table 1). With each run, we included 1 sample without template and heterozygous reference controls for factor V Leiden and prothrombin 20210G>A. These were provided by the manufacturer. In addition, we chose to include 1 known patient
Table 1 Time Study Details*

Discussion
Microarrays can be manufactured by directly generating nucleic acid probes on the solid support, by straight transfer

Restriction Fragment Length Polymorphism Specimen preparation, min PCR MagnaPure programming MagnaPure loading Procedure, min PCR amplification Enzyme addition Incubation Gel loading Electrophoresis Desalting Sample map creation Loader Hybridization with factor V probe Reader Stripping of the microarray Hybridization with prothrombin probe Reader Results, min Interpretation Time and throughput Time to results Hands-on time Throughput used in the present study Genotyping calls Overall time per sample, min Overall time per genotyping call, min Hands-on time per sample, min Hands-on time per genotyping call, min Maximum throughput per run
PCR, polymerase chain reaction. * For proprietary information, see the text.

Roche LightCycler 40 25 35 65 15 3h 1 h 20 min 32 32 5.6 5.6 2.5 2.5 32

Nanogen Electronic Microarray 30 132 45 35 270 10 20 10 10 20 15 9 h 57 min 2 h 35 min 48 96 12.4 6.2 3.2 1.6 192 (384:2)

30 90 10 60 15 60 15 4 h 40 min 1 h 40 min 30 60 9.3 4.7 3.3 1.7 30

494 494

Am J Clin Pathol 2003;119:490-496

DOI: 10.1309/3VTR7TL2X7TXL0QY

American Society for Clinical Pathology

Coagulation and Transfusion Medicine / ORIGINAL ARTICLE

of prefabricated probes onto the array, or by electronic mobilization of oligonucleotides to pads on an active microelectronic chip.19 In an electronic dot-blot method developed by Nanogen, the use of electric fields and an L-histidine buffer with minimal conductivity permits rapid transport and electric field concentration of DNA to defined locations on the array, as well as optimized hybridization.19 Hybridization stringency is achieved by reversal of polarity at optimized electric field strength, which repulses the charged molecules that lack optimal affinity for the substrate.20 A porous streptavidin-containing permeation layer covers the microelectrodes and facilitates interaction with small ions and water but protects larger molecules such as DNA from damage in the semiconductor microchip.21 Since our evaluation of the NanoChip System, the permeation layer has been modified from an agarose-based to a nonprotein-based hydrogel matrix. The NanoChip Molecular Biology Workstation includes a loader, which electronically addresses the biotinylated PCR amplicon to specified locations on the microchip. Subsequently, a stabilizer oligonucleotide and wild-type and mutant reporter oligonucleotides with different fluorescent labels (Cy3 and Cy5, respectively) are passively hybridized to the chip. Detection of fluorescent signals occurs on the reader module, and computerized results are displayed in histograms, tables, and charts.14 Of the 200 RFLP samples that were compared in our Nanogen evaluation, 19 were heterozygous for factor V Leiden and 2 were homozygous for this mutation. Nineteen samples were heterozygous for prothrombin 20210G>A, and 1 was homozygous. Factor V Leiden NanoChip results could be obtained for 197 of 200 patient samples. These were identical to the RFLP findings, as were the 199 of 200 results for prothrombin 20210G>A. Results for 200 patient samples were compared with results on the Roche LightCycler. For this assay, we included 20 heterozygous and 2 homozygous factor V Leiden samples. The positive prothrombin 20210G>A samples included 1 homozygous and 19 heterozygous specimens. With the NanoChip, we obtained 195 of 200 factor V Leiden and 198 of 200 prothrombin 20210G>A results. Three samples that could not be amplified with the Nanogen primers in a nonmultiplexed reaction were reamplified on the LightCycler to assess sample integrity. Amplification was achieved, indicating that the LightCycler amplification may be better optimized. Two samples, from a woman and her newborn child, displayed an atypical melting curve pattern on LightCycler analysis Image 2. The LightCycler assay will detect other sequence alterations under the mutation probe, with altered melting patterns observed. To confirm that the underlying mutation was not at nucleotide position 20210, the samples were sequenced directly and were called as homozygous wildtype for prothrombin at position 20210 (data not shown).
American Society for Clinical Pathology

However, both samples were heterozygous for prothrombin sequence variants near the SNP of interest. The variants in these samples may represent innocent polymorphisms. Despite the proximity of these variants to the SNP of interest, the NanoChip assay correctly identified these 2 samples as homozygous wild-type for prothrombin 20210G>A, indicating that the NanoChip System may not be affected by variants of this type. The results from the 400 multiplexed amplicons (800 genotyping calls) analyzed on the NanoChip System resulted in overall 2.0% and 0.8% no-call rates for factor V and prothrombin, respectively. Eight samples contributed to the no-call rate due to failed PCR: 3 samples failed PCR for both factor V and prothrombin and 5 samples failed PCR for factor V only. There was 1 instance of an indeterminate nontemplate control with the NanoChip assay. A single positive pad was flanked by a strongly positive patient control sample on the previously addressed pad and by the second (negative) nontemplate pad on the other side. By running in duplicate, the discrepancy was identified readily. The DNA concentration of the positive patient control adjacent to the indeterminate nontemplate control sample was 100 ng/L. As with all other PCRs in this assay, 200 ng of DNA was used in the amplification reaction. The negative control

Image 2 LightCycler analysis. The melting profile of a heterozygous control (20210G>A, blue) reveals 1 peak at the typical melting temperature for the wild-type sequence (58C) and 1 peak that corresponds to presence of the mutation, which decreases the melting temperature to 49C. The atypical melting curve for 1 patient sample (green) is superimposed onto this curve and demonstrates 1 wild-type melting temperature (58C) and 1 unknown allele at 53C. Direct DNA sequencing confirmed that the patient is homozygous for the wild-type sequence, prothrombin 20210G. For proprietary information, see the text.

Am J Clin Pathol 2003;119:490-496

495
DOI: 10.1309/3VTR7TL2X7TXL0QY

495 495

Schrijver et al / SNPS BY NANOCHIP, LIGHTCYCLER, AND RFLP

sample was run on a polyacrylamide gel to rule out possible contamination during the PCR reaction. As no PCR product was visualized, the same sample was addressed onto another chip and subsequently read as negative. More extensive evaluations have shown no evidence of cross-contamination between test sites.22 However, performing sample testing in duplicate initially, to verify the absence of cross-contamination, may be indicated for quality assurance. The NanoChip technology of electronic assignment of DNA sequences to pads on a microarray has been applied to the diagnosis of factor V Leiden and prothrombin 20210G>A. After a brief training period, the assay is relatively easy to perform and accurate. Other advantages include the possibility of very high throughput, extensive automation with minimal risk of sample error, and comprehensive results display on the computer screen after completion of the assay. One disadvantage may include a longer time to results compared with the LightCycler assay if the instrument is not used in a high-throughput setting. The NanoChip assay may find application in smaller laboratories once the assay is automated further or completed in a shorter time. Currently, it may be of special interest for molecular diagnostic laboratories with a medium to large test volume.
From the 1Department of Pathology and the 2Molecular Pathology Laboratory, Stanford University Medical Center, Stanford, CA. Supported in part by Nanogen. Address reprint requests to Dr Schrijver: Dept of Pathology, L235, Stanford University Medical Center, Stanford, CA 94305. Acknowledgment: We thank Carol D. Jones for technical assistance.

References
1. Hansson PO, Welin L, Tibblin G, et al. Deep vein thrombosis and pulmonary embolism in the general population: The Study of Men Born in 1913. Arch Intern Med. 1997;157:1665-1670. 2. Bertina RM, Koeleman BP, Koster T, et al. Mutation in blood coagulation factor V associated with resistance to activated protein C. Nature. 1994;369:64-67. 3. Girolami A, Simioni P, Scarano L, et al. Hemorrhagic and thrombotic disorders due to factor V deficiencies and abnormalities: an updated classification. Blood Rev. 1998;12:45-55. 4. Martinelli I, Bottasso B, Duca F, et al. Heightened thrombin generation in individuals with resistance to activated protein C. Thromb Haemost. 1996;75:703-705. 5. Zoller B, Holm J, Svensson P, et al. Elevated levels of prothrombin activation fragment 1 + 2 in plasma from patients with heterozygous Arg506 to Gln mutation in the factor V gene (APC-resistance) and/or inherited protein S deficiency. Thromb Haemost. 1996;75:270-274.

6. Dahlback B. New molecular insights into the genetics of thrombophilia: resistance to activated protein C caused by Arg506 to Gln mutation in factor V as a pathogenic risk factor for venous thrombosis. Thromb Haemost. 1995;74:139-148. 7. Franco RF, Reitsma PH. Genetic risk factors of venous thrombosis. Hum Genet. 2001;109:369-384. 8. Robetorye RS, Rodgers GM. Update on selected inherited venous thrombotic disorders. Am J Hematol. 2001;68:256-268. 9. Svensson PJ, Zoller B, Dahlback B. Evaluation of original and modified APC-resistance tests in unselected outpatients with clinically suspected thrombosis and in healthy controls. Thromb Haemost. 1997;77:332-335. 10. Cumming AM, Shiach CR. The investigation and management of inherited thrombophilia. Clin Lab Haematol. 1999;21:77-92. 11. von Ahsen N, Schutz E, Armstrong VW, et al. Rapid detection of prothrombotic mutations of prothrombin (G20210A), factor V (G1691A), and methylenetetrahydrofolate reductase (C677T) by real-time fluorescence PCR with the LightCycler. Clin Chem. 1999;45:694-696. 12. Nauck M, Marz W, Wieland H. Evaluation of the Roche diagnostics LightCyclerFactor V Leiden Mutation Detection Kit and the LightCyclerProthrombin Mutation Detection Kit. Clin Biochem. 2000;33:213-216. 13. Wittwer CT, Ririe KM, Andrew RV, et al. The LightCycler: a microvolume multisample fluorimeter with rapid temperature control. Biotechniques. 1997;22:176-181. 14. Pollak ES, Feng L, Ahadian H, et al. Microarray-based genetic analyses for studying susceptibility to arterial and venous thrombotic disorders. Ital Heart J. 2001;2:568-567. 15. Poort SR, Rosendaal FR, Reitsma PH, et al. A common genetic variation in the 3'-untranslated region of the prothrombin gene is associated with elevated plasma prothrombin levels and an increase in venous thrombosis. Blood. 1996;88:3698-3703. 16. Don RH, Cox PT, Wainwright BJ, et al. Touchdown PCR to circumvent spurious priming during gene amplification. Nucleic Acids Res. 1991;19:4008. 17. Zehnder JL, Benson RC. Sensitivity and specificity of the APC resistance assay in detection of individuals with factor V Leiden. Am J Clin Pathol. 1996;106:107-111. 18. Liu XY, Nelson D, Grant C, et al. Molecular detection of a common mutation in coagulation factor V causing thrombosis via hereditary resistance to activated protein C. Diagn Mol Pathol. 1995;4:191-197. 19. Edman CF, Raymond DE, Wu DJ, et al. Electric field directed nucleic acid hybridization on microchips. Nucleic Acids Res. 1997;25:4907-4914. 20. Sosnowski RG, Tu E, Butler WF, et al. Rapid determination of single base mismatch mutations in DNA hybrids by direct electric field control. Proc Natl Acad Sci U S A. 1997;94:11191123. 21. Heller MJ, Forster AH, Tu E. Active microelectronic chip devices which utilize controlled electrophoretic fields for multiplex DNA hybridization and other genomic applications. Electrophoresis. 2000;21:157-164. 22. Nagan N, OKane D. Validation of a single nucleotide polymorphism genotyping assay for the human serum paraoxonase gene using electronically active customized arrays. Clin Biochem. 2001;34:589-592.

496 496

Am J Clin Pathol 2003;119:490-496

DOI: 10.1309/3VTR7TL2X7TXL0QY

American Society for Clinical Pathology