|
|
||||||||
Molecular Diagnostics and Genetics |
1 Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology, Stuttgart, Germany.
2 Division of Clinical Pharmacology, University Hospital Tuebingen, Tuebingen, Germany.
aAddress correspondence to this author at: Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology, Auerbachstr. 112, 70376 Stuttgart, Germany. Fax 49-(0)711-85-92-95; e-mail uli.zanger{at}ikp-stuttgart.de.
| Abstract |
|---|
|
|
|---|
Methods: We used multiplex PCR to amplify relevant gene fragments while avoiding amplification of the CYP2B7P1 pseudogene. Polymorphic sites were analyzed by allele-specific primer extension followed by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Method evaluation was performed on a panel of 287 genomic DNA samples previously genotyped by other methods.
Results: Five multiplex assays were developed, comprising the following 15 SNPs: 82T
C (*22); 86G
C (R29T, *17); 136A
G (M46V, *11); 296G
A (G99E, *12); 415A
G (K139E, *8, *13); 419G
A (R140Q, *14); 516G
T (Q172H, *6, *7, *9, *13, *19, *20), 547G
A (V183I); 769G
A (D257N); 785A
G (K262R, *4, *6, *7, *13, *16, *19, *20); 983T
C (I328T, *16, *18); 1006C
T (R336C, *19); 1172T
A (I391N, *15); 1282C
A (P428T, *21); 1459C
T (R487C, *5, *7). In 9 DNA samples showing discrepant genotypes, correctness of the MALDI-TOF MS result was confirmed by direct sequencing.
Conclusions: This genotyping method enabled sensitive, specific, accurate, and comprehensive determination of 15 relevant SNPs of CYP2B6. The assay design allows analysis of SNP subsets, incorporation of additional SNPs, and performance of high-throughput genotyping.
| Introduction |
|---|
|
|
|---|
The CYP2B6 gene is located, together with a closely related pseudogene CYP2B7P1, in a CYP2 gene cluster on chromosome 19 (16). In contrast to the above-mentioned CYP genes, variations in CYP2B6 have been discovered through reverse genetics studies, i.e., by initial sequencing and variation scanning studies and subsequent analysis of the functional consequences of the identified variations (17)(18)(19)(20)(21)(22)(23)(24)(25). More than 100 DNA variations, including numerous nonsynonymous variations, as well as silent, promoter, and intronic changes, were found within the CYP2B6 gene, many of them showing extensive linkage disequilibrium giving rise to distinct haplotypes (26). The spectrum of functional consequences of these variations is wide and includes null alleles with no detectable function and/or expression (alleles CYP2B6*8, *12, *15, *18, *21), alleles with partially reduced function/expression (CYP2B6*5, *6, *7, *11, *14, *19, *20, *21) (17)(19)(21)(23), and alleles with increased expression [CYP2B6*22; (24)]. Some of these variations are rare, but many are common, with allele frequencies between 10% and almost 50%, depending on the population studied (22)(23).
The clinical relevance of CYP2B6 variations has recently been demonstrated for the anti-HIV drug efavirenz. The common clinical practice of administering the same dose to all patients leads to profound differences in drug plasma concentrations, which are correlated with patient genotype (27)(28)(29). Patients with high drug concentrations are at risk of developing concentration-related central nervous system toxicity, including insomnia, fatigue, and headache and often leading to discontinuation of the therapy. Thus, for a drug such as efavirenz, dose adjustment based on CYP2B6 genotype could prevent administration of too-high doses, thus increasing the safety and efficacy of the therapy.
In recent years, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) proved to be a superior technology for the detection of single-nucleotide variations (SNPs). This method has several advantages over other methods, including high accuracy because of the direct measurement of molecular masses, high sensitivity in the detection of both homozygous and heterozygous base changes, and high-throughput and cost-effective multiplex capability (30)(31)(32)(33)(34)(35)(36)(37)(38). We developed a MALDI-TOF MS genotyping method to determine 15 SNPs in the CYP2B6 gene.
| Materials and Methods |
|---|
|
|
|---|
maldi-tof mass spectrometric cyp2b6 genotyping assay
The CYP2B6 promoter sequence (AC 24497632), gene sequence (NG_000008.5 and AC023172), and the CYP2B7 pseudogene sequence (AC008537) were used for specific CYP2B6 primer design with the AlignX program of the VectorNTI Suite 9 package (InforMax, Inc.). All reactions, including PCR amplification, shrimp alkaline phosphatase treatment, and base extension, were performed in 384 microtiter plates (ABgene, Epsom) with a Puredisk pipette robot (Cybio). PCR amplification and primer extension reaction were carried out in a DYADTM PCR thermal cycler (MJ Research), and no-template controls were carried along in every plate to exclude contaminations.
Multiplex PCR.
PCRs (final volume, 8 µL) contained 50 ng of DNA, 0.2 units of HotStarTaq polymerase, 1 mmol/L MgCl2 (Qiagen GmbH), and the desired primers (MWG Biotech AG) at their optimized concentrations (see Table 1
in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue1). A tag (5'-ACGTTGGATG) was included in the primer sequence to equalize extreme relative percentage-GC contents (39). PCR conditions were 95 °C for 15 min followed by 45 cycles at 95 °C for 30 s, 60 °C or 61 °C (assay 5) for 1 min, and 72 °C for 1 min; and finally 72 °C for 10 min. After dephosphorylation of excess dNTPs with 0.3 units of shrimp alkaline phosphatase in 1x Rx-Buffer (Amersham) at 37 °C for 20 min, 85 °C for 10 min, and 20 °C for 1 s, the PCR products were used as templates for the primer extension reactions. CYP2B6-specific amplification was verified for all primer pairs by sequence analysis of products obtained in single reactions.
|
Primer extension reactions.
Extension reactions (final volume, 16 µL) contained 1 µL of buffer C (500 mmol/L Tris-HCL, pH 9.5; 100 mmol/L MgCl2), 6 mmol/L MgCl2, 1 U of Termipol DNA polymerase I (all from Solis Biodyne), and extension primers at optimized concentrations (see Table 2
in the online Data Supplement). Assay-specific deoxynucleoside triphosphates (dATP, dGTP, dTTP; Amersham) and dideoxynucleoside triphosphates (ddA, ddC, ddG, ddT; BioLog Life Science Institute) were added at 0.12 mmol/L. In assays 1, 2, 3, and 5, dCTP (Amersham) and dGTP were added at 0.25 mmol/L. Extension reactions were performed at 94 °C for 4 min followed by 55 cycles at 94 °C for 30 s and 52 °C (assays 1 and 3), 56 °C (assay 2), or 60 °C (assay 4 and 5) for 30 s, and 72 °C for 30 s; and finally 72 °C for 2 min. The final nucleotide extension products were treated with a cationic exchange resin (AG® 50W-X8 Resin; Bio-Rad Laboratories, Inc.) for 30 min to remove salts.
|
MALDI-TOF MS measurement.
We spotted and air dried 1 µL of 3-hydroxypicolinic acid matrix (45 mmol/L and 3 mmol/L diammonium hydrogen citrate solution) onto a 384-format MTP AnchorChipTM 400/384TF target plate (Bruker Daltonik, spot size 400 µm). The extension reaction products (0.5 µL) were dispensed on the matrix and air dried. The target plate was then inserted into the Ultraflex MALDI-TOF mass spectrometer (Bruker Daltonik), and analysis was performed with 180 nitrogen laser shots for each sample. The mass range of the MS instrument was set at 40009000 Da. Genotyping calls were made with the assistance of the GENOTOOLS software (Bruker Daltonik). DNA samples not automatically assigned were determined manually.
generation of control samples
Control samples are needed for each SNP in heterozygous and homozygous form to exclude artifacts such as pausing peaks or other DNA fragments. To generate a homozygous variant control sample from an available heterozygous sample, a PCR product was generated with the appropriate PCR primers (see Table 1 in the online Data Supplement). The 2 alleles were separated by subcloning the fragments in TOP10 cells by positive selection with a TOPO TA Cloning® Kit vector (Invitrogen) containing the F plasmid ccdB killer gene (40). DNA was isolated from bacteria and sequenced to verify the presence of the SNP. If no heterozygous sample was available, we used a previously developed touchdown PCR method to introduce the variation in a wild-type (wt) sample. A primer located within the amplified region and containing the variation of interest was designed with a melting temperature
8 °C above that of the amplification primers (see Table 1 in the online Data Supplement), as described previously (41). The primer sequences are available on request. Sequence verification was performed with AB Big Dye® Terminator Mix v1.1. (Applied Biosystems) and the ABI Prism 310 Genetic Analyzer (Applied Biosystems).
allele nomenclature and statistics
Base numbering and allele definitions were according to the published recommendations of the CYPallele Nomenclature Committee (http://www.imm.ki.se/CYPalleles/criteria.htm). Numbering was based on the cDNA with the full-length cDNA sequence published by Yamano et al. (42) defined as the wt (CYP2B6*1). We used the DeFinetti program (http://ihg.gsf.de/cgi-bin/hw/hwa1.pl) to test genotype frequencies for conformance with HardyWeinberg equilibrium. Data were compiled according to the genotype and allele frequencies estimated from the observed numbers of each specific allele and with the assistance of the PHASE program v.2.0.2 (43)(44). The frequency of each SNP in our study participants is given with the 95% confidence interval (CI).
| Results |
|---|
|
|
|---|
C promoter SNP, which was recently shown to cause enhanced transcription and to result in increased expression (24).
The c.64C
T (R22T) variation, defining the CYP2B6*2 allele, was thus far not included because it was shown to be comparable to the wt in expression and function (17)(20).
development of maldi-tof ms assays
For the specific amplification of CYP2B6 gene fragments, it was important to (a) effectively discriminate between CYP2B6 and the pseudogene CYP2B7, and (b) minimize the possibility that nucleotide variations within the PCR- and extension-primer sequences interfered with amplification. To meet these requirements, the PCR primer pairs (see Table 1 in the online Data Supplement) were designed to bind to CYP2B6 in regions of low homology to CYP2B7, mostly found in intronic or exon/intron boundary regions. The only exon not included in the multiplex PCR assays is exon 6, for which no relevant variations had been described to date. All primer pairs were shown to be functional in various combinations up to a single-tube multiplex amplification reaction for all 8 amplicons. However, a lower degree of multiplexing was finally applied because not all extension reactions could be carried out simultaneously. The compatibility of the extension primers is strongly restricted by their sequence, molecular masses, and annealing temperatures. Individual extension primers were optimized with respect to primer length and concentration and tested with positive controls for all 3 genotypes (homozygous wt, homozygous variant, and heterozygous). The control DNAs were selected from previously genotyped genomic DNA samples or generated recombinantly as described in Materials and Methods. By testing various combinations, we arrived at a final design comprising 5 multiplex assays (assays 15; see Table 2 in the online Data Supplement). Representative MALDI-TOF MS spectra for all 5 assays are shown in Fig. 1
.
|
evaluation of the maldi-tof ms genotyping assay
To evaluate the established CYP2B6 genotyping assay, we analyzed 287 genomic DNA samples from individuals of Caucasian origin who had previously been genotyped for some of the SNPs by various other methods. In total, this amounted to 4305 individual genotype determinations. Fig. 2
shows the relative intensities of the mass signals corresponding to the 2 primer extension products of a given CYP2B6 SNP, plotted against each other, for the 4 most frequent variations. As evident from this analysis, the mass spectral data were clearly structured into 3 groups, i.e., homozygous wt/wt and mutant (mut)/mut genotypes found along the axes, and wt/mut heterozygotes scattered in-between (no-template controls with relative signal intensities of 0 are indicated with white points and black from the middle). In total, 2300 genotypes have been previously determined by other methods, including sequencing, allele-specific 5'-nuclease assay (TaqMan), denaturing high-performance liquid chromatography, and/or PCRrestriction fragment length polymorphism (RFLP). In 9 cases (0.4%), MALDI-TOF MS analysis resulted in a genotype discrepant from that of the earlier studies. Retrospective inspection revealed that PCR-RFLP assays had been used for the earlier analyses of these samples and that the fragment profiles were ambiguous and difficult to interpret. After resequencing these samples, the MALDI-TOF MS outcomes were confirmed in all 9 cases.
|
frequency analysis of cyp2b6 variations in caucasians
The allele and genotype frequencies observed in the analyzed population of 287 Caucasians are summarized in Table 2
. All genotypes were shown to be in HardyWeinberg equilibrium. Five variations previously observed in persons of African or African-American origin were not detected in any Caucasian sample. The most frequent SNPs were the amino acid substitutions K262R (27.4%), Q172H (25.2%), and R487C (9.7%), followed by the promoter SNP 82C (2.1%). The variations M46V, G99E, K139E, R140Q, V183I, and I391N were present at a frequency of <1% (Table 2
). Results of haplotype analysis with the PHASE program indicated that the most frequent alleles were the wt allele (*1; 59%) and *6 (24.6%). The other CYP2B6 alleles and their frequencies were *4 (2.3%), *5(7.3%), *11 (0.35%), *12 (0.17%), *13 (0.52), *14 (0.52%), *15 (0.7%), and *22 (2.1%). The alleles *17, *18, *19, and *21 were not found in this population.
| Discussion |
|---|
|
|
|---|
Similar to other cytochrome P450 genes, CYP2B6 represents a particular challenge with respect to developing gene-specific assays because of its high homology to the pseudogene CYP2B7P1 (46). Because no variation analysis has been carried out for CYP2B7, the influence of contaminating CYP2B6 amplicons with CYP2B7 is unpredictable. We designed intronic primers such that each of them harbored at least 1 nucleotide difference from the pseudogene sequence, and we controlled each amplicon for its specificity by direct sequencing. Occasionally contamination was observed and PCR conditions were then modified to improve specificity. It is thus important that such controls are performed when setting up this or similar assays. The use of a common tagging sequence at the 5'-end of the amplification primers to level out differences in GC content proved to be advantageous for multiplexing (39). Under optimized conditions, the entire set of primer pairs gave satisfactory results in a single multiplex amplification reaction.
The combination of the various primer extension reactions into multiplex assays is another critical aspect, and support by suitable software is highly recommended to allow the optimization of several different parameters, i.e., GC content, molecular mass range, annealing temperatures, dimers, and hairpin loop formation. We tested the accuracy of conversion of the unextended primer into allele-specific analyte by use of either control DNAs from previously sequenced/genotyped samples or recombinantly generated controls. To achieve highest multiplexing levels, we tested many primer combinations, leading us to the final assay design consisting of 5 multiplex assays. More than 4 extension primers could not be combined into a single-multiplex extension reaction, mainly because of cross-binding problems and different extension mixes necessary for the 15 SNPs (see Table 2 in the online Data Supplement).
For evaluation, we analyzed all 15 SNPs in 287 genomic DNA samples of white European origin. As shown in Fig. 2
for the most frequent CYP2B6 variations, the MALDI-TOF MS data obtained for the 3 possible genotypes (wt, mut, and heterozygous) were clearly structured into 3 clusters. On average,
5% (equivalent to 14 samples) were not automatically identified and had to be assigned manually, resulting in an overall automated call rate of >95% for all 15 SNPs. Thus, with the assays presented, a reliable high-throughput genotyping of CYP2B6 variations is achievable. The comparison of the MALDI-TOF MS results with those obtained by other methods revealed a few discrepant results, which could be assigned to 2 distinct RFLP assays. These divergent results were retrospectively shown to be caused by erroneous data interpretation of the primary RFLP data, and the MALDI-TOF MS outcomes could be verified in all cases by sequencing, emphasizing the high accuracy of this method.
After completion of the mass spectrometric analysis, 9 genotypes (0.2%) of the 4305 total SNP analyses remained undetermined even after 3 repeated analyses. The reason for this is currently unclear. However, because only 4 different SNPs were affected (see Table 2
) and because PCR amplification was successful in each case, we speculate that unknown sequence variations within primer binding sites may have prevented the binding and/or extension of the respective primers. The overall genotyping success rate for the 287 samples was thus higher than 99% in 3 of the 5 assays, and even 100% in 2 assays, which confirms the high technical reliability of MALDI-TOF MS (47).
The SNP frequencies observed in this study confirm and extend current knowledge on CYP2B6 genetic variation in the middle European population. CYP2B6 variations have been discovered relatively recently, and thus their value for predicting drug response is largely unexplored. However, clinical applications of CYP2B6 genotyping so far include bupropion, used in smoking cessation (48); cyclophosphamide, in lupus nephritis (49) and in hematologic malignancies(50); and efavirenz, an anti-HIV agent (27)(28)(29). In particular, the consistent results of multiple studies on efavirenz pharmacogenetics emphasize the potential of CYP2B6 variations in predicting patient variability in drug side effects, response, and toxicity. To date, only the more common variations of CYP2B6 have been included in these studies, in part because comprehensive genotyping assays have not been available (51)(52)(53). In our assay, we have included numerous recently discovered, less frequent SNPs, which have pronounced effects on expression/function and which together account for 5% to 10% of functionally variant alleles, depending on the ethnic origin of the population studied (Table 1
) (19)(21)(23)(25). Their determination in the course of clinical studies should thus help to improve our understanding of the phenotypegenotype relationship and enhance the predictive value of CYP2B6 genotyping.
|
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Human genes: CYP2B6, cytochrome P450, family 2, subfamily B, polypeptide 6; CYP2B7P1, cytochrome P450, family 2, subfamily B, polypeptide 7 pseudogene 1; CYP2D6, cytochrome P450, family 2, subfamily D, polypeptide 6; CYP2C19, cytochrome P450, family 2, subfamily C, polypeptide 19; CYP2C9, cytochrome P450, family 2, subfamily C, polypeptide 9. ![]()
| References |
|---|
|
|
|---|
C) leading to enhanced transcription and relocation of the transcriptional start site. Mol Pharmacol 2005;67:1772-1782.The following articles in journals at HighWire Press have cited this article:
![]() |
E. Schaeffeler, U. M. Zanger, M. Eichelbaum, S. Asante-Poku, J.-G. Shin, and M. Schwab Highly Multiplexed Genotyping of Thiopurine S-Methyltransferase Variants Using MALDI-TOF Mass Spectrometry: Reliable Genotyping in Different Ethnic Groups Clin. Chem., October 1, 2008; 54(10): 1637 - 1647. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. H. Hofmann, J. K. Blievernicht, K. Klein, T. Saussele, E. Schaeffeler, M. Schwab, and U. M. Zanger Aberrant Splicing Caused by Single Nucleotide Polymorphism c.516G>T [Q172H], a Marker of CYP2B6*6, Is Responsible for Decreased Expression and Activity of CYP2B6 in Liver J. Pharmacol. Exp. Ther., April 1, 2008; 325(1): 284 - 292. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. C.H. Tsang, P. Charoenkwan, K. C.K. Chow, Y. Jin, C. Wanapirak, T. Sanguansermsri, Y.M. D. Lo, and R. W.K. Chiu Mass Spectrometry Based Detection of Hemoglobin E Mutation by Allele-Specific Base Extension Reaction Clin. Chem., December 1, 2007; 53(12): 2205 - 2209. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |