Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Clinical Chemistry 51:3 522–531 (2005) Molecular Diagnostics and Genetics Determination of CYP2D6 Gene Copy Number by Pyrosequencing Erik Söderbäck,1† Anna-Lena Zackrisson,2 Bertil Lindblom,2 and Anders Alderborn1* Background: Identification of CYP2D6 alleles *5 (deletion of the whole CYP2D6 gene) and *2xN (gene duplication) is very important because they are associated with decreased or increased metabolism of many drugs. The most commonly used method for analysis of these alleles is, however, considered to be laborious and unreliable. Methods: We developed a method to determine the copy number of the CYP2D6*5 and CYP2D6*2xN alleles by use of PyrosequencingTM technology. A single set of PCR and sequencing primers was used to coamplify and sequence a region in the CYP2D6 gene and the equivalent region in the CYP2D8P pseudogene, and relative quantification between these fragments was performed. The CYP2D8P-specific Pyrosequencing peak heights were used as references for the CYP2D6specific peak heights. Results: Analysis of 200 pregenotyped samples showed that this approach reliably resolved 0 – 4 genome copies of the CYP2D6 gene. In 15 of these samples, the peak pattern from one analyzed position was unexpected but could be solved by conclusive results from a second position. The method was verified on 270 other samples, of which 267 gave results that corresponded to the expected genotype. One of the samples could not be interpreted. The reproducibility of the method was high. Conclusions: CYP2D6 gene copy determination by Pyrosequencing is a reliable and rapid alternative to other methods. The use of an internal CYP2D8P control as well as generation of a sequence context ensures a robust method and hence facilitates method validation. Cytochrome P450 (CYP) is a family of proteins involved in the degradation of many toxic compounds. The human genome contains more than 60 unique CYP genes, of which 6 are considered the most metabolically active (1 ). Many drugs, being chemically directly or closely related to naturally toxic compounds, are targets for CYP enzymatic activities. It is well known that polymorphisms in the CYP genes affect the activities and/or specificities of the encoded enzymes, which causes differences in the response to drugs degraded by these enzymes. One of the most important CYP enzymes with respect to drug degradation is encoded by the CYP2D6 gene [GenBank accession no. M33388 (2 )], which has been shown to be involved in the degradation of ⬃100 drug compounds (3 ). Approximately 80 different alleles of the CYP2D6 gene have been identified (4 ), and their presence varies between ethnic populations. The enzymatic activities in persons carrying these allelic differences vary from total absence to different degrees of ultrarapid metabolism, which causes considerable variability in the response to certain drug treatments. The CYP2D6 gene is located on chromosome 22, together with the two pseudogenes CYP2D7P [GenBank accession nos. X58467 and X58468 (5 )] and CYP2D8P [GenBank accession no. M33387 (2 )], which are localized in tandem. Low or eliminated CYP2D6 enzyme activity is usually caused by point mutations, small deletions, or insertions. The CYP2D6*5 allele, however, carries a deletion of the complete gene. This allele is one of the most frequent CYP2D6 alleles, and homozygous carriers totally lack metabolic activity. On the other hand, rapid metabolizers often carry a duplication, or even multiplication (up to 12 copies have been found), of the CYP2D6 gene, most commonly the CYP2D6*2 allele [CYP2D6*2xN (1, 6, 7 )]. The increased enzymatic activity in individuals carrying duplications enhances the degradation of drugs, which may cause nontherapeutic drug concentrations at typical dosages. Many antidepressants, for example, are metabolized by CYP2D6, and there is a distinct relationship between the number of CYP2D6 genes and the rate of metabolism of antidepressants (8 ). The presence of the CYP2D6 duplication shows an interesting north–south © 2005 American Association for Clinical Chemistry 1 Biotage AB, Uppsala, Sweden. National Board of Forensic Medicine, Department of Forensic Genetics, Linköping, Sweden. †Current address: Biology Education Centre, Uppsala University, Box 592, SE-751 24 Uppsala, Sweden. *Address correspondence to this author at: Biotage AB, Kungsgatan 76, SE-753 18 Uppsala, Sweden. Fax 46-18-591922; e-mail anders.alderborn@ telia.com. Received September 20, 2004; accepted December 27, 2004. Previously published online at DOI: 10.1373/clinchem.2004.043182 2 522 Clinical Chemistry 51, No. 3, 2005 gradient in the European population, with a frequency of ⬃1% in Scandinavia (9 ) to 7–10% in Italy (10 ) and as high as 29% in Ethiopia in northern Africa (11 ). Several methods for determination of CYP2D6 gene copy number have been reported. Although considered as relatively laborious and unreliable, the most commonly used method is a long-range PCR spanning the CYP-REP regions flanking the CYP2D6 gene, followed by gel-based size analysis of the PCR products (12 ). Other methods are based on quantitative amplification of a small PCR fragment from the CYP2D6 gene, in which the amount of PCR product reflects the number of CYP2D6 genes in the analyzed genome (13, 14 ). The amount of CYP2D6-specific PCR product is assessed in relation to a coamplified region from an unrelated housekeeping gene, which is assumed to consistently be a single-copy gene. PyrosequencingTM is a nonelectrophoretic, real-time DNA sequencing technology (15, 16 ). A primer is hybridized to a single-stranded PCR template, and the sequencing analysis is started by addition of nucleotides. The nucleotides are added sequentially, and through coupled enzymatic reactions, the polymerase-catalyzed incorporation of nucleotides can be monitored as light peaks in a PyrogramTM. Pyrosequencing has been used for many applications, such as genotyping of single-nucleotide polymorphisms (17, 18 ), mutation detection (19, 20 ), and sequence detection (21, 22 ). The peak heights in the Pyrogram are directly proportional to the emitted light from the Pyrosequencing reaction. This nature of the light peaks enables quantitative analyses such as assessment of allele frequency in a mixed sample population (23–25 ) and degree of DNA methylation (26, 27 ). Several assay protocols have been developed for analysis of CYP alleles by Pyrosequencing (28 –32 ). In this report, we describe an easy single-well assay for determination of CYP2D6 gene copy number, based on relative quantifications from Pyrosequencing reactions using the pseudogene CYP2D8P as reference. Materials and Methods dna samples for assay development and verification A total of 200 DNA samples from different laboratories were used as a key for development of an assay based on Pyrosequencing technology. These DNA samples were extracted by different methods and stored frozen until use. An additional set of DNA samples from 270 Swedish blood donors were analyzed to verify the assay. The Regional Ethics Committee had given permission for the analysis, and all donors had given their consent. These DNA samples were extracted by the dodecyltrimethylammonium bromide/cetyltrimethylammonium bromide (DTAB/CTAB) method (33 ) and pregenotyped for the CYP2D6*2, *3, *4, and *6 alleles by a specific Pyrosequencing method (29 ). Individuals hypothetically ho- 523 mozygous for CYP2D6*1, *2, *3, *4, or *6 were tested for CYP2D6*5 by a multiplex long PCR (7 ) with modifications (29 ). Analysis for gene duplication was performed in 26 of the 270 samples with an established long-fragment PCR method (34 ) with several modifications: Primers were manufactured (Invitrogen AB) according to published primer sequences (34 ). A 25-L PCR reaction was carried out with the Expand Long Template PCR System (Roche Molecular Biochemicals). The reaction mixture contained 2.5 L of 10⫻ PCR buffer 3 (22.5 mM MgCl2), 5 L of deoxynucleotide triphosphate mixture (2.5 mM each nucleotide), 0.36 L of enzyme mixture (3.5 U/L), 0.4 L of each primer (cyp-17F, cyp-32R, and cyp-207 F; 20 M), 11.3 L of water, and 5 L of genomic DNA (10 ng/L). The conditions for amplification with the primer pairs cyp-17F/cyp-32R and cyp-207F/cyp-32R were as follows: an initial denaturing step of 94 °C for 2 min, followed by 35 cycles of 94 °C for 15 s, 64 °C for 30 s, and 68 °C for 6 min, with a final elongation step of 68 °C for 7 min. The PCR products were separated and detected in ethidium bromide-containing 0.8% agarose gels. The number of tandem copies of the CYP2D6 gene could not be determined by the long-PCR method used. assay design The three genes CYP2D6, CYP2D7P, and CYP2D8P were aligned to find suitable regions for PCR primer positioning (to amplify CYP2D6 and CYP2D8P and to avoid amplification of CYP2D7P). PCR designs in which only the CYP2D6 and the CYP2D8P genes were amplified were further analyzed for optimization of Pyrosequencing specificity. One or several sequencing primers for each PCR fragment were located upstream of regions with nucleotide differences between the CYP2D6 and CYP2D8P genes. In the Pyrosequencing analysis, gene-specific signals were achieved by design of an assay-specific nucleotide addition scheme. The CYP2D6*5/CYP2D6*2xN assay was thus based on relative quantification of Pyrosequencing peak signals between a CYP2D6 gene-specific amplicon and a CYP2D8P amplicon. The Pyrosequencing reactions were designed so that possible unwanted amplification of the highly similar CYP2D7P gene did not interfere with any of the CYP2D6/CYP2D8Pspecific peaks. primer sequences Primers from three different vendors were successfully tested in the assay. The upstream PCR primer used was A1058FP (5⬘-GGTGGCTGACCTGTTCTYTG-3⬘). The sequence includes three internal sites where the common D6/D8 sequence diverges from the D7 sequence. Keeping the cycling annealing temperature above 58 °C excludes amplification of the CYP2D7P gene. The degeneration at the third position from the 3⬘ end of the primer (bolded Y) is attributable to the variation between the CYP2D6 and CYP2D8P sequences (C/T). The downstream PCR primer was biotinylated A1051RPB (5⬘-GGGCTCACGCTGCACATC-3⬘). One of the sequencing primers used was 524 Söderbäck et al.: CYP2D6 Gene Copy Number Fig. 1. Alignment of the CYP2D6 gene at the end of exon 6 with the CYP2D7P and CYP2D8P pseudogenes (GenBank accession nos. M33388 for CYP2D6, X58467 for CYP2D7P, and M33387 for CYP2D8P). Arrows indicate PCR primers (A1058FP and A1051RPB) and sequencing primers (A685FP and A1050FP). The shaded areas cover the bases generating peaks in the Pyrosequencing reactions. A685FP (5⬘-CGGGATGGTGACCACCTC-3⬘). The C at the 3⬘ end of the primer is homologous in the D6 and the D8 sequences but differs from the D7 sequence. The other sequencing primer used was A1050FP (5⬘-GGCCTCCTGCTCATGATCC-3⬘). pcr conditions The PCR reagents used were all from the Qiagen HotStarTaq reagent set (cat. no. 203205; Qiagen) except for the deoxynucleotide triphosphates (100 mM solution; cat. no. 27-2035-02; Amersham Biosciences), and autoclaved milliQ water was used for diluting the reagents. In the final PCR mixture, the concentrations were 1.5 mM MgCl2, 2.5 mM each nucleotide, 200 nM each primer, and 0.2 ng/L template DNA. Addition of Q solution, added according to the Qiagen manual, was necessary for PCR product specificity. The PCR cycling was initiated with 15 min of enzyme activation at 95 °C (according to the Qiagen instructions) followed by 45 cycles at 95 °C for 30 s, 59 °C for 30 s, and 72 °C for 30 s. The reaction was finished with a 10-min final extension at 72 °C. The quality of the PCR products was checked on agarose gels. PCR instruments from Hybaid (Thermo-Hybaid Model MBS 0.26) and Applied Biosystems (Models 6700 and 9600) generated PCR products successfully used in the subsequent Pyrosequencing analysis. To test the influence of PCR annealing temperature on Pyrosequencing peak heights, measured as relative light units, we performed a temperature gradient PCR on a Mastercycler gradient (Eppendorf) with annealing temperatures ranging from 50 to 65 °C over the entire PCR block. analysis The PCR products were prepared for Pyrosequencing analysis by use of a Vacuum Prep Workstation (Biotage AB), and the sequencing reactions were performed on either a PSQ 96MA System or a PSQ 96HS System as described by the manufacturer (Biotage AB). Nucleotide dispensation orders were chosen manually to generate specific peaks for the targeted CYP2D6 and CYP2D8P PCR products and to reveal any false amplification from the CYP2D7P pseudogene. For the sequencing primer A685FP, the dispensation order CGACACTC generated CYP2D6-specific peaks at dispensations 5 (A) and 6 (C), which were compared with the CYP2D8P-specific peaks at dispensations 7 (T) and 8 (C) (see Fig. 2). For sequencing primer A1050FP, the dispensation order CTACATGCT generated CYP2D6-specific peaks at dispensations 5 (A) and 6 (T), which were compared with the CYP2D8Pspecific peak at dispensation 7 (G). The Sequence Analysis software (Biotage AB) was used for measurement of peak heights, and the results were exported to MS Excel for further calculations and analysis. In the A685FP primer assay, the CYP2D6-specific peak at dispensation 6 (C) and the CYP2D8P-specific peak at dispensation 8 (C) were used for the calculations. In the A1050FP primer assay, the CYP2D6-specific peak at dispensation 6 (T) and the CYP2D8P-specific peak at dispensation 7 (G) were used. Results assay design and optimization Quantitative amplification assays were designed such that the CYP2D6 gene copy number was examined with a CYP2D6 pseudogene as reference. The choice of a reference gene with a high homology allowed one-tube amplification as well as sequencing of both genes with the same PCR and sequencing primers. The pseudogene had two important functions: (a) as a fixed and stable reference for the quantification of the CYP2D6 copy number (having the same amplification efficiency as the CYP2D6 target gene); and (b) as a positive internal control for the assay and the primers used for analysis of the CYP2D6 gene. The pseudogene CYP2D7P was rejected as a reference because it has been reported to be relatively frequently duplicated (1 ). In contrast, neither deletions nor duplications have been reported for the CYP2D8P gene (which is Clinical Chemistry 51, No. 3, 2005 525 Fig. 2. Typical Pyrograms from assays using the two sequencing primers A685FP and A1050FP on PCR products from samples differing in the number of CYP2D6 genome copies. The x axes are time axes showing the addition of nucleotides to the sequencing reaction. The y axes show light emission obtained as relative light units. nxD6 indicates the number of CYP2D6 genome copies: (A), 0xD6; (B), 1xD6; (C), 2xD6; (D), 3xD6; (E), 4xD6. The left panels show Pyrograms from sequencing primer A685FP; the right panels show Pyrograms from sequencing primer A1050FP. Arrows indicate peaks generated from nucleotide incorporations specific for CYP2D6 (thin arrows) and CYP2D8P (thick arrows). These peaks were used in the gene copy number determination. ⬃95% homologous to CYP2D6 at the nucleotide level), which qualified this gene as the reference. The relative amounts of amplification products were determined by use of the Pyrosequencing technology to sequence through short regions that differed between the CYP2D6 target gene and the reference gene. The signal peak pattern, presented in a Pyrogram, was thus a product of two sequencing reactions that were performed 526 Söderbäck et al.: CYP2D6 Gene Copy Number simultaneously. Three types of peaks occurred in the Pyrogram: peaks specific for CYP2D6 (i.e., free from interference by CYP2D8P); peaks specific for CYP2D8P (i.e., free from interference by CYP2D6); and peaks consisting of a combined light signal from both genes. We determined the number of CYP2D6 genes by calculating the ratio between CYP2D6-specific peaks and peaks that were specific for the CYP2D8P gene. All peaks in the Pyrograms were, however, used to ensure the specificity of the assay. Several potential target regions in the CYP2D6 and CYP2D8P genes were amplified by PCR and analyzed. The majority of regions investigated were unstable and caused various problems when tested on the DNA sample panel, most probably because of variability within the pseudogene sequence. One region obtained with PCR primers A1058FP and A1051RPB at the end of exon 6, however, was robust for the majority of DNA samples tested (Fig. 1); the ratio was found to be proportional to the number of CYP2D6 genes in the sample. Typical results of genotyping using the sequencing primers A685FP and A1050FP on this PCR product are shown in Fig. 2. The designation “nxD6” indicates the number of CYP2D6 genome copies. 0xD6 was a genotype that was homozygous for the CYP2D6*5 allele, i.e., CYP2D6 was deleted on both chromosomes and only the sequence from CYP2D8P was seen. 1xD6 indicates a sample with one CYP2D6 gene copy (heterozygous for CYP2D6*5), and the expected peak ratio was ⬃0.5. 2xD6 indicates a wild-type sample with two copies of CYP2D6, i.e., one on each chromosome and a ratio of 1. pcr annealing temperature The effect of PCR annealing temperature on the specific peak heights from CYP2D6 and CYP2D8P in a wild-type genome (CYP2D6*1/*2) was tested in a gradient PCR with annealing temperatures in the range 50 – 65 °C. The PCR products were analyzed with the two sequencing primers, and specific peak heights from the CYP2D6 and CYP2D8P genes as well as the ratio between them were plotted as a function of temperature (Fig. 3). Skewed peak heights indicated differences in PCR efficiency between CYP2D6 and CYP2D8P at all PCR annealing temperatures except for a relatively narrow window of 58 – 60 °C. At annealing temperatures below this window, the CYP2D8P-specific peak heights were higher than those for CYP2D6, whereas the CYP2D6-specific peak heights were higher than those for CYP2D8P at higher temperatures. The results were as expected based on the PCR primer A1058FP sequence, which has a C/T degeneracy between the CYP2D6 and CYP2D8P sequences. An annealing temperature of 59 °C was chosen to obtain a ratio close to 1. pyrosequencing analysis Peak heights exported from the Pyrosequencing software were further analyzed in Microsoft Excel, where the relative peak heights between CYP2D6- and CYP2D8P- Fig. 3. Effect of PCR annealing temperature on peak heights. (A and B), peak heights (in relative light units) specific for CYP2D6 (Œ) and CYP2D8P (f) in a wild-type genome (CYP2D6*1/*2) as an effect of a gradient annealing temperature from 50 to 65 °C. (A), sequencing primer A685FP; (B), sequencing primer A1050FP. (C), ratios of specific peak heights (CYP2D6/ CYP2D8P) from the gradient PCR products. ⽧, primer A685FP; Œ, primer A1050FP. The ratio is close to 1 at ⬃59 °C. specific peaks were calculated. Panels A1 and A2 in Fig. 4 are scatter plots of the peak-height ratios for the respective sites for 40 selected samples of different genotypes, showing conclusive centering of ratios for both positions around values of 0 (0xD6), 0.5 (1xD6), 1 (2xD6), 1.5 (3xD6), and higher ratios for three samples. According to our analysis, two of these samples carried 4 gene copies, but that has not been possible to verify by any other technique. The third sample carried even more than 4 gene copies, indicated by the even higher peak-height ratio. This is visualized more clearly if the ratios are sorted from minimum to maximum (see panels B1 and B2 in Fig. 4). Note that samples exhibiting any of the variants described in the section below were excluded from these plots. The peak-height ratios of these samples were not relevant, and Clinical Chemistry 51, No. 3, 2005 527 . . . . . . . . . . . . Fig. 4. Scatter plots of CYP2D6-/CYP2D8-specific peak-height ratios. The y axes show the calculated ratio between CYP2D6- and CYP2D8P-specific peaks. The x axes are the serial numbers of the samples. (A1 and B1), ratios for sequencing primer A685FP. (A2 and B2), ratios for sequencing primer A1050FP (same samples as in panels A1 and B1). Panels A1 and A2 show the DNA samples in order of collection; panels B1 and B2 show the samples when sorted from lowest to highest ratios based on ratios from the A1050FP assay. Observe that the frequencies of the different genotypes are from a nonrepresentative subset of samples. the two peak ratio values from such samples would segregate drastically, indicating the variation and the requirement for visual examination of the Pyrograms. reproducibility To study the reproducibility of the assays, we repeatedly analyzed four samples with known CYP2D6 gene copy numbers (1xD6, 2xD6, 3xD6, and 4xD6) on nine separate occasions (on different days) and plotted the D6/D8P ratio against the expected number of CYP2D6 copies (Fig. 5). Data points from samples analyzed on a specific occasion (day) are connected by a line, and the linear regression coefficients (R2) were calculated for each such series. For the A685FP assay, the linear regression coefficients varied from 0.9731 to 0.9994, and for A1050FP, they varied between 0.9632 and 0.9979. The lines confirmed a good linearity in the analysis of copy numbers, and the comparison between lines indicated high reproducibility. This analysis also indicated that the assay using sequencing primer A685FP gave a somewhat greater day-to-day variation than the assay using the A1050FP primer. samples generating atypical pyrosequencing data The CYP2D8P gene deviated from the expected sequence in 13 of the ⬃200 test samples. One relatively common allelic type with a C3 T transition at the equivalent of 528 Söderbäck et al.: CYP2D6 Gene Copy Number (compare panel B1 in Fig. 6 with panel B1 in Fig. 2). This sample exhibited a typical 1xD6 pattern when analyzed with the A1050FP sequencing primer (compare panel B2 in Fig. 6 with panel B2 in Fig. 2). Another allelic variant was found in one 3xD6 sample when analysis was performed with the A685FP sequencing primer. This variant was more difficult to interpret. The Pyroseqencing data (Pyrogram) from the A685FP assay exhibited a pattern indicating multiple copies (6xD6) (panel C1 in Fig. 6). In contrast, the Pyrogram from the A1050FP assay exhibited a typical 3xD6 pattern (panel C2 in Fig. 6). The only plausible explanation is that the sequencing primer A685FP did not anneal to one of the CYP2D8P alleles in this sample, leading to CYP2D8Pspecific peaks being generated from only one of the two CYP2D8P alleles, giving reference peaks one half the height of those obtained from the same sample analyzed with the A1050FP primer. assay verification Fig. 5. Analysis of reproducibility of the CYP2D6 gene copy number assays. (A), A685FP assay; (B), A1050FP assay. Data points for samples analyzed on a specific occasion (day) are connected by a line. position 2933 in CYP2D6 was found downstream of the A1050FP sequencing primer. This was readily revealed, however, by the unexpected Pyrogram pattern (compare panel A2 in Fig. 6 with panel C2 in Fig. 2), which was affected by the signals from the variants of the CYP2D8P gene. All of these samples exhibited a typical 2xD6 pattern when analyzed with the A685FP sequencing primer (compare panel A1 in Fig. 6 with panel C1 in Fig. 2). One 1xD6 sample with a base transition in one of the alleles, rendering an equivalent effect for the A685FP primer, was also found among the 200 test samples The two sequencing primers A685FP and A1050FP gave interpretable results in 269 of the 270 pregenotyped DNA samples. A few samples had to be reanalyzed because of discrepancies between the A685FP and A1050FP assays. Thirteen samples (4.8%) were identified as carrying one allele with a duplication of CYP2D6 as indicated by the gene copy number, 3xD6 (Table 1). Long-range PCR analysis verified these findings. The gene copy number 0xD6 was not found in any of the samples from blood donors. Pyrograms corresponding to a gene copy number of 1xD6 were generated by 23 of 24 samples from individuals previously genotyped as heterozygous carriers of the CYP2D6*5 allele. The aberrant sample had a gene copy number of 2xD6, which did not compare with the previously found genotype, CYP2D6*4/*5. The sample was reanalyzed with the same result. A typical allelic distribution of the CYP2D6 gene (2xD6) was found in 231 samples. In the first set of analyses, three samples generated results that deviated between the two primers. The A685FP primer indicated 3xD6, whereas the A1050FP primer indicated 2xD6 in two of these samples. The third sample generated the opposite aberrant result. After reanalysis, all three samples gave the result 2xD6. This was also verified by the long-range PCR for multiple genes. We could find no explanation for the aberrant results in the first analysis. One sample pregenotyped as wild-type, CYP2D6*2/*2, was identified as 1xD6 by the Pyrosequencing assay. However, reinvestigation using long-range PCR did not verify the 1xD6 gene copy number, i.e., no gene deletion was observed. Of the samples from 269 blood donors, four 2xD6 samples and one 3xD6 sample were found to carry one allele with the variant CYP2D8P sequence identified with primer A1050FP (see the Pyrogram in Fig. 6). A typical peak pattern was obtained with the sequencing primer A685FP. 529 Clinical Chemistry 51, No. 3, 2005 Fig. 6. Pyrograms from samples generating a typical pattern in one of the assays but not in the other. The x axes show the addition of nucleotides, and the y axes show the light signal obtained as relative light units. The left panels show the A685FP primer assay; the right panels show the A1050FP primer assay. Sample A (top row) shows a typical 2xD6 pattern in the A685FP assay, but an atypical 2xD6 pattern in the A1050FP assay. Sample B (middle row) shows a typical 1xD6 pattern in the A1050 assay and an atypical pattern in the A685 assay. Sample C (bottom row) shows a typical 3xD6 pattern in the A1050 assay and a pattern resembling 6xD6 in the A685 assay. Arrows indicate peaks generated from nucleotide incorporations specific for CYP2D6 (thin arrows) and CYP2D8P (thick arrows). Discussion Table 1. Numbers of CYP2D6 gene copies found in 269 of 270 blood donor samples tested. Gene copy numbera Genotype *1/*1 *1/*2 *1/*3 *1/*4 *1/*5 *1/*6 *2/*2 *2/*3 *2/*4 *2/*5 *3/*4 *4/*4 *4/*5 *4/*6 *5/*6 Total a 0xD6 1xD6 2xD6 29 68 2 47 3xD6 3 2 6 9 1 2 24 3 41 1 1 7 6 0 1 24 2 11 1 2 232 13 Two samples with disparate results between previous genotyping and Pyrosequencing analysis are indicated in bold. In this study, we describe an assay for determination of the CYP alleles CYP2D6*5 and CYP2D6*2xN. The assay, which is based on competitive PCR between the CYP2D6 and CYP2D8P genes and Pyrosequencing analysis, generated stable patterns and unequivocal genotype determination for samples carrying the CYP2D6*5 allele in both a homozygous and heterozygous state. For samples carrying a multiple-gene-copy allele (CYP2D6*2xN), it was possible to identify carriers of 3 or 4 gene copies. The samples carrying 4 gene copies were not verified with other techniques. However, this conclusion was supported by the high linearity in the reproducibility assay (Fig. 5). The resolution of alleles carrying multiple gene copies may be increased by further optimization of the method. Jansson et al. (35 ) have also successfully applied Pyrosequencing technology for gene copy analysis of the glutathione S-transferase M1 gene and identified an allele (GSTM1*0) with complete deletion of the gene. A related approach has been described by Pielberg et al. (36 ), who used breakpoint analysis, which cannot be applied on CYP2D6 because the breakpoints are located in the 2.8-kb CYP-REP region. The analysis of two positions on the 530 Söderbäck et al.: CYP2D6 Gene Copy Number same PCR product, as described in the current method, offers substantial advantages because it facilitates the resolution of ambiguous results. DNA samples prepared from blood in four different laboratories were analyzed in this study. Interestingly, small differences in the relative peak patterns of the sequence were observed from DNA samples delivered from the different laboratories and purified by different methods (data not shown). Our assumption is that the method for preparation of genomic DNA may influence the first cycles of the PCR reaction, perhaps via remaining material (e.g., histone complexes or regulatory proteins) bound to the DNA after the preparation. This difference in relative peak pattern was more pronounced when we used primer batches from a certain vendor (data not shown). The conclusion is that care must be taken when interpreting variability that may be attributable to low quality of the primers or the DNA preparation used. A few allelic variations found within the CYP2D8P gene generated atypical Pyrogram patterns. A simple solution for these problems would be to use a patternrecognition approach for the genotyping analysis. Because stable patterns were generated for the different genotypes as well as for the atypical samples, this assay is suitable for software-based recognition of Pyrosequencing peak patterns. This would facilitate the analysis step considerably because export of peak heights and Excelbased analysis would be unnecessary. A relatively simple prototype software for such purposes was tested with very good results. Blind tests of DNA from blood donors showed rare and unexplained discrepancies in results generated with the different techniques used. Reanalysis of the samples clarified most of these discrepancies. However, two samples remained divergent. One sample previously typed as CYP2D6*4/*5 gave a gene copy number of 2xD6, which is not usually found for the actual genotype. A possible explanation is that the *4 allele in this case contained a gene duplication. This is also supported by the long-PCR analysis for multiple genes, which showed a duplication of one of the alleles. In the other sample, the genotype CYP2D6*2/*2 predicted a wild-type 2xD6 gene copy number, but instead we found a 1xD6 profile. This was the only sample of the 269 blood samples tested for which a discrepancy remained between the found copy number and the genotype in combination with long-PCR methods for the CYP2D6*5 allele and multiple CYP2D6 genes. One sample from the blood donors was excluded because of recurrently unclear results from both Pyrosequencing and long-PCR reactions. Considering the relatively high probability for variation in the CYP2D8P pseudogene, it could be assumed that other allelic variants will appear in larger sample groups. However, the variations identified here all indicate the strength of the assay, which generates sequence background from two positions. It is possible to determine the reason for unexpected (atypical) patterns in one position by having as a key the typical appearance of peaks generated at the other position. A possible problem with the assay would be total or partial deletion of one of the two CYP2D8P alleles in a particular genome. To our knowledge, a sample carrying such a deletion has not been reported. Such a sample would generate Pyrograms for both positions with CYP2D8P-specific reference peaks that are one half of the expected height relative the CYP2D6-specific peaks. The gene quantification analysis would suggest an erroneous 4xD6 genotype for wild-type (2xD6) samples. The same type of problem would appear if the variation within CYP2D8P was so pronounced that the PCR reaction does not work for one of the CYP2D8P alleles. As described above, such a nonfunctional primer sequence most probably contributed to the aberrant result obtained during sequencing of one of the tested samples. If both CYP2D8P alleles are deleted in a sample, the specific CYP2D8P peaks will be absent from the Pyrogram, indicating the necessity to use other techniques (i.e., long-range PCR) for the analysis of CYP2D6 gene copy number. No such samples were found among those tested. Except for long-range PCR, all other methods for the determination of CYP2D6 gene copy number that have been reported in the literature use an unrelated gene as reference. To our knowledge, this is the first report describing an analysis using a highly related gene as reference. There are several advantages associated with the use of a related gene and coamplification in a method such as the one described here. Only a single set of primers is used (the same for both target and reference gene), which increases the probability of an accurate determination of gene copy number. The generation of amplicons from unrelated genes may lead to variations in amplification efficiencies between the different genetic regions. The use of a single set of primers also enables more straightforward method validation. Both of these advantages should make the approach described here attractive for pharmacogenetic studies. We are grateful for the receipt of genotyped DNA samples from Drs. Steven Wong and Paul Jannetto (Department of Pathology, Medical College of Wisconsin, Milwaukee, WI), Dr. Inger Johansson (Division of Molecular Toxicology, Karolinska Institute, Stockholm, Sweden), and Dr. Mia Wadelius (Department of Clinical Pharmacology, Akademic Hospital, Uppsala, Sweden). We are also grateful to Ann-Britt Gladh for her excellent work. This work was supported by Grant 03-15 from The National Board of Forensic Medicine. References 1. Online Mendelian Inheritance in Man (OMIM) http://www.ncbi. nlm.nih.gov/entrez/Omim (accessed May 24, 2004). 2. Kimura S, Umeno M, Skoda RC, Meyer UA, Gonzalez FJ. The human debrisoquine 4-hydroxylase (CYP2D) locus: sequence and Clinical Chemistry 51, No. 3, 2005 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. identification of the polymorphic CYP2D6 gene, a related gene, and a pseudogene. Am J Hum Genet 1989;45:889 –904. Bertilsson L, Dahl M-L, Dalen P, Al-Shurbaji A. Molecular genetics of CYP2D6: clinical relevance with focus on psychotropic drugs. Br J Clin Pharmacol 2002;53:111–22. Home Page of the Human Cytochrome P450 (CYP) Allele Nomenclature Committee. CYP2D6 allele nomenclature. http://www. imm.ki.se/CYPalleles/cyp2d6.htm (accessed May 24, 2004). Heim MH, Meyer UA. Evolution of a highly polymorphic human cytochrome P450 gene cluster: CYP2D6. Genomics 1992;14: 49 –58 Johansson I, Lundqvist E, Bertilsson L, Dahl ML, Sjoqvist F, Ingelman-Sundberg M. Inherited amplification of an active gene in the cytochrome P450 CYP2D locus as a cause of ultrarapid metabolism of debrisoquine. Proc Natl Acad Sci U S A 1993;90: 11825–9. Hersberger M, Marti-Jaun J, Rentsch K, Hänseler E. Rapid detection of the CYP2D6*3, CYP2D6*4, and CYP2D6*6 alleles by tetra-primer PCR and of the CYP2D6*5 allele by multiplex long PCR. Clin Chem 2000;46:1072–7. Dalen P, Dahl ML, Ruiz ML, Nordin J, Bertilsson L. 10-Hydroxylation of nortriptyline in white persons with 0, 1, 2, 3, and 13 functional CYP2D6 genes. Clin Pharmacol Ther 1998;63:444 –52. Dahl ML, Johansson I, Bertilsson L, Ingelman-Sundberg M, Sjoqvist F. Ultrarapid hydroxylation of debrisoquine in a Swedish population. Analysis of the molecular genetic basis. J Pharmacol Exp Ther 1995;274:516 –20. Scordo MG, Spina E, Facciola G, Avenoso A, Johansson I, Dahl ML. Cytochrome P450 2D6 genotype and steady state plasma levels of risperidone and 9-hydroxyrisperidone. Psychopharmacology (Berl) 1999;147:300 –5. Aklillu E, Persson I, Bertilsson L, Johansson I, Rodrigues F, Ingelman-Sundberg M. Frequent distribution of ultrarapid metabolizers of debrisoquine in an Ethiopian population carrying duplicated and multiduplicated functional CYP2D6 alleles. J Pharmacol Exp Ther 1996;278:441– 6. Johansson I, Lundqvist E, Dahl M-L, Ingelman-Sundberg M. PCRbased genotyping for duplicated and deleted CYP2D6 genes. Pharmacogenetics 1996;6:351–5. Neville M, Selzer R, Aizenstein B, Maguire M, Hogan K, Walton R, et al. Characterization of cytochrome P450 2D6 alleles using the Invader system. Biotechniques 2002;32:S34 –S43. Schaeffeler E, Schwab M, Eichelbaum M, Zanger UM. CYP2D6 genotyping strategy based on gene copy number determination by TaqMan real-time PCR. Hum Mutat 2003;22:476 – 85. Ronaghi M, Uhlen M, Nyren P. A sequencing method based on real-time pyrophosphate. Science 1998;281:363, 365. Berg LM, Sanders R, Alderborn A. Pyrosequencing technology and the need for versatile solutions in molecular clinical research [Review]. Expert Rev Mol Diagn 2002;2:361–9. Ahmadian A, Gharizadeh B, Gustafsson AC, Sterky F, Nyren P, Uhlen M, et al. Single-nucleotide polymorphism analysis by pyrosequencing. Anal Biochem 2000;280:103–10. Alderborn A, Kristofferson A, Hammerling U. Determination of single-nucleotide polymorphisms by real-time pyrophosphate DNA sequencing. Genome Res 2000;10:1249 –58. Garcia CA, Ahmadian A, Gharizadeh B, Lundeberg J, Ronaghi M, Nyren P. Mutation detection by pyrosequencing: sequencing of 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 531 exons 5– 8 of the p53 tumor suppressor gene. Gene 2000;253: 249 –57. Goriely A, McVean GA, Rojmyr M, Ingemarsson B, Wilkie AO. Evidence for selective advantage of pathogenic FGFR2 mutations in the male germ line. Science 2003;301:643– 6. Jonasson J, Olofsson M, Monstein HJ. Classification, identification and subtyping of bacteria based on Pyrosequencing and signature matching of 16S rDNA fragments. APMIS 2002;110: 263–72. Rahim A, Coutelle C, Harbottle R. High-throughput Pyrosequencing of a phage display library for the identification of enriched target-specific peptides. Biotechniques 2003;735:317–24. Gruber JD, Colligan PB, Wolford JK. Estimation of single nucleotide polymorphism allele frequency in DNA pools using Pyrosequencing. Hum Genet 2002;110:395– 401. Neve B, Froguel P, Corset L, Vaillant E, Vatin V, Boutin P. Rapid SNP allele frequency determination in genomic DNA pools by Pyrosequencing. Biotechniques 2002;32:1138 – 42. Lavebratt C, Sengul S, Jansson M, Schalling M. Pyrosequencingbased SNP allele frequency estimation in DNA pools. Hum Mutat 2004;23:92–7. Tost J, Dunker J, Gut IG. Analysis and quantification of multiple methylation variable positions in CpG islands by Pyrosequencing. Biotechniques 2003;35:152– 6. Colella S, Shen L, Baggerly KA, Issa JP, Krahe R. Sensitive and quantitative universal Pyrosequencing methylation analysis of CpG sites. Biotechniques 2003;35:146 –50. Eriksson S, Berg L, Wadelius M, Alderborn A. Cytochrome P450 genotyping by multiplexed real-time DNA sequencing with Pyrosequencing technology. Assay Drug Dev Technol 2002;1:49 –59. Zackrisson AL, Lindblom B. Identification of CYP2D6 alleles by single nucleotide polymorphism analysis using pyrosequencing. Eur J Clin Pharmacol 2003;59:521– 6. Wadelius M, Sorlin K, Wallerman O, Karlsson J, Yue QY, Magnusson PK, et al. Warfarin sensitivity related to CYP2C9, CYP3A5, ABCB1 (MDR1) and other factors. Pharmacogenomics J 2004;4: 40 – 8. Furman KD, Grimm DR, Mueller T, Holley-Shanks RR, Bertz RJ, Williams LA, et al. Impact of CYP2D6 intermediate metabolizer alleles on single-dose desipramine pharmacokinetics. Pharmacogenetics 2004;14:279 – 84. Min DI, Ellingrod VL, Marsh S, McLeod H. CYP3A5 polymorphism and the ethnic differences in cyclosporine pharmacokinetics in healthy subjects. Ther Drug Monit 2004;26:524 – 8. Gustincich S, Manfioletti G, Del Sal G, Schneider C, Carninci P. A fast method for high-quality genomic DNA extraction from whole human blood. Biotechniques 1991;11:298 –300, 302. Lovlie R, Daly AK, Molven A, Idle JR, Steen VM. Ultrarapid metabolizers of debrisoquine: characterization and PCR-based detection of alleles with duplication of the CYP2D6 gene. FEBS 1996;392:30 – 4. Jansson M, Rada A, Tomic L, Larsson L-I, Wadelius C. Analysis of the Glutathione S-transferase M1 gene using pyrosequencing and multiplex PCR—no evidence of association to glaucoma. Exp Eye Res 2003;77:239 – 43. Pielberg G, Day AE, Plastow GS, Andersson L. A sensitive method for detecting variation in copy numbers of duplicated genes. Genome Res 2003;13:2171–7.