Label-free genotyping of cytochrome P450 2D6*10 using ligation-mediated strand displacement amplification with DNAzyme-based chemiluminescence detection

Analytica Chimica Acta 710 (2012) 111–117

Contents lists available at SciVerse ScienceDirect

Analytica Chimica Acta journal homepage: www.elsevier.com/locate/aca

Label-free genotyping of cytochrome P450 2D6*10 using ligation-mediated strand displacement amplification with DNAzyme-based chemiluminescence detection Hong-Qi Wang a,b , Zhan Wu a , Yan Zhang a , Li-Juan Tang a,∗ , Ru-Qin Yu a , Jian-Hui Jiang a,∗ a b

State Key Laboratory of Chemo/Bio-Sensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082, PR China Research Center of Agricultural Quality Standards and Testing Techniques, Henan Academy of Agricultural Sciences, Zhengzhou 450002, PR China

a r t i c l e

i n f o

Article history: Received 13 September 2011 Received in revised form 19 October 2011 Accepted 25 October 2011 Available online 6 November 2011 Keywords: Genotyping Cytochrome P450 monooxygenase 2D6*10 DNA ligase Strand displacement amplification Single nucleotide polymorphism

a b s t r a c t Genotyping of cytochrome P450 monooxygenase 2D6*10 (CYP2D6*10) plays an important role in pharmacogenomics, especially in clinical drug therapy of Asian populations. This work reported a novel label-free technique for genotyping of CYP2D6*10 based on ligation-mediated strand displacement amplification (SDA) with DNAzyme-based chemiluminescence detection. Discrimination of single-base mismatch is firstly accomplished using DNA ligase to generate a ligation product. The ligated product then initiates a SDA reaction to produce aptamer sequences against hemin, which can be probed by chemiluminescence detection. The proposed strategy is used for the assay of CYP2D6*10 target and the genomic DNA. The results reveal that the proposed technique displays chemiluminescence responses in linear correlation to the concentrations of DNA target within the range from 1 pM to 1 nM. A detection limit of 0.1 pM and a signal-to-background ratio of 57 are achieved. Besides such high sensitivity, the proposed CYP2D6*10 genotyping strategy also offers superb selectivity, great robustness, low cost and simplified operations due to its label-free, homogeneous, and chemiluminescence-based detection format. These advantages suggest this technique may hold considerable potential for clinical CYP2D6*10 genotyping and association studies. © 2011 Elsevier B.V. All rights reserved.

1. Introduction Adverse drug reactions (ADRs) induced by individual differences in drug susceptibility are major clinical problems and important causes for mortality and morbidity [1]. Although factors that determine susceptibility are unclear in most cases, there is great interest in the genetic factors. For example, there are increasing studies of cytochrome P450 2D6 (CYP2D6), a highly polymorphic gene from the chief drug metabolizing enzyme family cytochrome P450 [2–5]. CYP2D6 is responsible for 40% of first-pass hepatic metabolism and active in the enzymatic breakdown of 20–25% of all medicines prescribed now [6,7]. In Asians, the most universal mutant CYP2D6 allele is CYP2D6*10, of which the mutation frequency is nearly 50% [8]. The single nucleotide change of CYP2D6*10 causes CYP2D6 encoding an unstable enzyme with reduced catalytic activity in drug metabolism, resulting in varied drug metabolizing abilities between individuals [9,10]. Since CYP2D6 is responsible for the metabolism of most drugs com-

∗ Corresponding authors. Tel.: +86 731 88664085; fax: +86 731 88821916. E-mail addresses: [email protected] (L.-J. Tang), [email protected] (J.-H. Jiang). 0003-2670/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2011.10.052

monly prescribed, including antidepressant, neuroleptics, opiate, antihypertensive and antiarrhythmic agents, the genotyping of CYP2D6*10 for Asian populations, thus, seems to be clinically essential. Developing methods for CYP2D6*10 genotyping is of considerable important due to the clinical significance of CYP2D6*10 in pharmacogenomics. Many strategies are known for the genotyping of CYP2D6, such as single-strand conformation polymorphism (SSCP) [11], allele-specific polymerase chain reaction (AS-PCR) [12], PCR with restriction fragment length polymorphism (PCRRFLP) analysis [13], real-time PCR [14], pyrosequencing [15], oligonucleotide microarray technology [16], and multiplex singlebase extension assay [17]. Though these techniques have been demonstrated to be useful tools for the assay of CYP2D6 typing, the shortages such as limited throughput, poor selectivity, low sensitivity, and the requirement of costly instruments and specific immobilizing reagents, may limit their use in genotyping of CYP2D6*10 in clinics. The development in chemical biology provides the opportunity for establishing novel CYP2D6*10 genotyping technologies. For example, enzymatic methods using certain enzyme-aided biochemistry have been good choices for allele genotyping. Enzyme-catalyzed reactions, benefitting from their fidelity in

112

H.-Q. Wang et al. / Analytica Chimica Acta 710 (2012) 111–117

single-base differentiation and ability in efficient target amplification, typically display extremely high specificity and sensitivity in genotyping [18–21]. These characteristics are highly preferable for clinical applications. However, most enzymatic methods may be limited by their requirements of expensive labeling reagents or multi-step washing and separation. Therefore, developing selective and sensitive enzymatic methods for genotyping of CYP2D6*10 that are greatly robust, cost-efficient, easily automated, scalable for parallel assays is still of great significance in analytical biochemistry. We have recently reported a label-free genotyping technique which comprises a ligation-based allele discrimination reaction followed by two consecutive nickase-based strand displacement amplification (SDA) reactions [22]. This strategy has displayed high sensitivity and selectivity in discriminating single-base mismatches, but two consecutive SDA reactions may increase the risk of non-specific amplification. In the present study, to avoid this problem we propose a modified label-free ligation-mediated strand displacement amplification strategy for genotyping of CYP2D6*10. The modified CYP2D6*10 genotyping strategy is proposed on the basis of ligase chain reaction for allele discrimination and target amplification, and only one-step nickase-based strand displacement amplification (SDA) reaction for signal amplification. Our previous studies have repeatedly demonstrated the fidelity of DNA ligases in highly specific typing of single nucleotide polymorphism (SNP) for genetic diseases and cancers [23,24]. DNA ligase is employed to provide high specificity in genotyping of CYP2D6*10 and its allele ligation reaction selectively activates the one-step nickase-based SDA reaction. The SDA reaction, which can generate abundant peroxidase-mimic DNAzyme sequences, enables the formation of hemin–aptamer complex, a catalyzer of luminol–H2 O2 reaction [25], and sensitive label-free chemiluminescence-based quantification of the allele-specific products in CYP2D6*10 genotyping. Because the SDA reaction is initiated by the ligase chain reaction, we call the procedure as ligation-mediated SDA. Although only one-step SDA was performed in this strategy, the results of our assays demonstrated the proposed strategy still exhibited adequate sensitivity in CYP2D6*10 genotyping assays with high specificity. The proposed technique, which can be carried out in a homogeneous format without any tedious washing and separation steps, permits simplified operations for genotyping of CYP2D6*10 with superb robustness. In addition, its label-free, homogeneous and chemiluminescence-based format makes the proposed technique cost-efficient and easily automated as desired in common use of CYP2D6*10 genotyping in clinics.

2. Experimental 2.1. Materials and instrumentation The oligonucleotides used in this work were designed according to the target of cytochrome P450 monooxygenase CYP2D6*10. They were synthesized from Takara (Dalian, China). The sequences of the synthesized oligonucleotides are given in Table 1. Thermodynamic parameters and secondary structures of all oligonucleotides were calculated using bioinformatics software (http://mfold.rna.albany.edu/). Bst DNA polymerase (8000 U mL−1 ), Nt.Bst NBI nickase (10,000 U mL−1 ), Thermostable Taq ligase (40,000 U mL−1 ), Taq PCR Kit and 10× NEB buffer 3 (1 M NaCl, 500 mM Tris–HCl, 100 mM MgCl2 , and 10 mM dithiothreitol (DTT), pH 7.9) were purchased from New England Biolabs (Ipswich, MA, USA). The deoxyribonucleoside 5 -triphosphates (dNTPs) solution mixture (10 mM) and Triton X-100 of analytical grade were provided by Sangon Biotech (Shanghai, China). H2 O2 with analytical grade was obtained from Shanghai Chemical Reagent Company (Shanghai, China). Nicotinamide adenine dinucleotide (NAD+ ),

dimethylsulfoxide (DMSO), luminol standard powder, hemin, and 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) were purchased from Sigma Aldrich Chemical Co. All other chemicals were of analytical grade and obtained from Sinopharm Chemical Reagent Co. Ltd. A luminol working solution (25 mM) was prepared using 0.5 mM NaOH. A stock solution of hemin (25 ␮M) was prepared with DMSO and stored at −20 ◦ C in the dark. All other solutions were prepared using ultrapure water, which was obtained through a Millipore Milli-Q water purification system (Billerica, MA, USA) and had an electric resistance of >18.3 M. The experiments of ligation-mediated SDA reaction and PCR amplification of five genomic samples were performed using a CFX96 Real-Time PCR Detection System (Biorad, Germany). Timedependent chemiluminescence signal was recorded on an FL3-P spectrofluorometer (Horiba Jobin YVON). 2.2. Procedure for ligation-mediated SDA reaction The ligation reaction was performed in 30 ␮L 1× NEB buffer 3 (100 mM NaCl, 50 mM Tris–HCl, 10 mM MgCl2 , and 1 mM DTT, pH 7.9) containing 0.1% (w/v) Triton X-100, 1 mM NAD+ , 10 nM probe 1 or 2, 10 nM probe 3, 0.3 U ␮L−1 Thermostable Taq ligase, and the mutant target, wild-type target or non-complementary target of a given concentration. The mixture was subjected to a thermal cycling treatment to perform the ligation reaction. Thermal cycling was carried out in a CFX96 Real-Time PCR Detection System under the following condition: 20 cycles of 95 ◦ C for 1 min and 60 ◦ C for 5 min. After the ligation reaction, 2 ␮L dNTPs (10 mM), 6 ␮L Bst DNA polymerase (8 U ␮L−1 ), and 2 ␮L Nt. Bst NBI (10 U ␮L−1 ) nickase were added into the reaction solution followed by incubation at 60 ◦ C for 2 h to allow SDA reaction in a buffer solution. The final 50 ␮L reaction buffer (pH 7.9) was consisting of 100 mM NaCl, 50 mM Tris–HCl, 10 mM MgCl2 , and 1 mM DTT. 2.3. Procedure for chemiluminescence detection In the mixture of SDA reaction, 16 ␮L H2 O, 20 ␮L 5× HEPES–NH4 OH buffer (125 mM HEPES, 100 mM KCl, and 1 M NaCl, pH 8.0), 2 ␮L hemin (25 ␮M), and 2 ␮L luminol (25 mM) were added. The mixture was incubated at room temperature for 30 min to allow the folding of DNAzyme (hemin with its aptamer) produced in SDA reaction. On adding 10 ␮L H2 O2 (300 mM) in the mixture, time-dependent chemiluminescence signal was recorded on an FL3-P spectrofluorometer with a time interval of 1 s, and an emission slit of 2.5 nm at 425 nm. The chemiluminescence spectra were recorded immediately after the addition of H2 O2 with an emission slit of 2.5 nm from 350 nm through 500 nm with an interval of 2.5 nm. 2.4. Procedure for genomic samples and PCR amplification Human genomes isolated from five individuals with SNP genotypes identified by DNA sequencing were kindly provided by clinical pharmacogenetic center at Xiangya hospital. PCR amplification was performed in 50 ␮L of 20 mM Tris–HCl (pH 8.8) with 10 mM (NH4 )SO4 , 2 mM MgCl2 , 0.1% Triton X-100 and 10 mM KCl, 250 ␮M dNTPs, and 1 ␮M forward and reverse primers (50 pM for each primer), as well as ∼20 ng of genomic DNA. The sequence of forward primer was 5 -CCA TTT GGT AGT GAG GCA GGT AT-3 and the sequence of reverse primer was 5 -CAC CAT CCA TGT TTG CTT CTG GT-3 . Amplification was achieved by thermal cycling for 40 cycles at 94 ◦ C for 30 s, 60 ◦ C for 20 s, 72 ◦ C for 1 min and final extension at 72 ◦ C for 10 min. The resulting products were confirmed by agarose gel assay and directly used for subsequent SNP typing without purification.

H.-Q. Wang et al. / Analytica Chimica Acta 710 (2012) 111–117

113

Table 1 Synthesized oligonucleotide sequences (5 → 3 ) used in the experiments.a

mutant target

5’-CAC CGG CGC CAA CGC TGG GCT GCA CGC TAC TCA CCA GGC CCC CTG CCA CTG CCC GGG CT-3’

wild-type target

5’-CAC CGG CGC CAA CGC TGG GCT GCA CGC TAC CCA CCA GGC CCC CTG CCA CTG CCC GGG CT-3’

probe 1

5’-CCC AAC CCG CCC TAC CCA ACA GAC TCG GGT AGG GCG GGT ACG TGG CAG GGG GCC TGG TGA -3’

probe 2

5’-CCC AAC CCG CCC TAC CCA ACA GAC TCG GGT AGG GCG GGT ACG TGG CAG GGG GCC TGG TGG -3’

probe 3

5’-PO32-- GTA GCG TGC AGC CCA GCG TTC ATC GGC CTT TTT TTT TTG GCC GAT G -3’

non-cognate target

5’-GAG CAT AAA CAG GTG GGA GTT GTC TTA CCA ACT CTG AGA GGC CAA TTA ATT AAG AGA AAA -3’

a

Probes 1 and 2 are the discriminating probes, probe 3 is the common probe, and non-cognate target is designed for a control experiment. The recognition sequence for Nt.Bst NBI nickase is highlighted in blue. The mutation site is highlighted in red in target sequences and discriminant probes. Underlined sequences in a single probe are complementary in order to form hairpin structure. (For interpretation of the references to color in text, the reader is referred to the web version of the article.)

3. Results and discussion 3.1. Probe design and analytical principle As illustrated in Scheme 1, the strategy for genotyping of CYP2D6*10 comprises a ligation-based allele discrimination reaction (LDR) followed by a nickase-based SDA reaction. The latter can produce abundant aptamer sequences against hemin to form

Scheme 1. Illustration of label-free genotyping of CYP2D6*10 using ligationmediated strand displacement amplification with DNAzyme-based chemiluminescence detection.

hemin–aptamer complex. In the proposed strategy, two discriminating probes, 1 and 2, are designed with a downstream sequence (near 3 end) complementary to the mutant CYP2D6*10 and its wild-type targets, respectively, with 3 termini overlapping the polymorphic bases. The 5 ends of probes, 1 and 2, are complementary to hemin aptamer and have a half recognition site (5 -GACTC-3 ) for Nt.Bst NBI nickase. Moreover, the two probes are both designed to have a hairpin structure with an 11-basepair stem region at their 5 ends, to inhibit the excess probes in the reaction system from capturing the aptamer sequences produced in the SDA reaction. Probe 3 is a common probe with the 5 -terminal sequence complementary to target upstream to the SNP site. On the downstream part of probe 3, it has two complementary sequences that can form a hairpin structure. In allele genotyping, discriminating probe 1 or 2 is separately annealed with probe 3 on DNA target. For the probe perfectly complementary to the target at the junction, it can be covalently joined with probe 3 by DNA ligase to form a ligation product. Oppositely, no ligation product can be obtained for the discriminating probe that is mismatched with the target at the junction. Using thermal cycling of thermostable DNA ligase, it allows the allele discrimination reaction to proceed repeatedly, and then offers a route for the amplification of the ligated products in the perfectly complementary system to enhance the detection sensitivity. The 3 end of probe 3, where a hairpin structure is, can behave as a primer to initiate an extension reaction in the presence of Bst DNA polymerase and dNTPs. For ligated product, such a self-primed extension reaction can proceeded across the polymorphic base and then the half recognition site of Nt.Bst NBI nickase on the discriminant probe, to yield a full recognition site for nickase with a nicking site generated on the replicated strand. Thus, a cycle comprised of nickase cleavage, polymerase extension and subsequent replicated aptamer sequence release is created, which renders a SDA for the ligated product. Via such SDA, the target-specific ligation product is greatly amplified into a great amount of hemin–aptamer complex, a peroxidase-mimic DNAzyme that can be readily probed from its strong chemiluminescence signal [25–27]. Note that the ligation-based allele discrimination reaction yields ligated product only in a perfectly complementary system, probe 1 and the mutant

114

H.-Q. Wang et al. / Analytica Chimica Acta 710 (2012) 111–117

Fig. 1. Time-dependent chemiluminescence responses in triplet repetitive assays for CYP2D6*10 genotyping. Luminescence signals were recorded at 425 nm with an emission slit of 2.5 nm. (a) Control experiment using probes 1 and 3 with no DNA target; (b) control experiment using probes 1 and 3 with 700 pM wild-type DNA target; (c) control experiment using probes 1 and 3 with 700 pM non-cognate DNA target; (d) responses using probes 1 and 3 with 700 pM mutant target.

CYP2D6*10 target, or probe 2 and the wild-type target. Hence, the mutant CYP2D6*10 target can be quantified by the chemiluminescence signal using probe 1, and the wild-type target can be determined by the luminescence readout using probe 2, thereby allowing immediate genotyping of DNA target. Compared with the previously developed method which has two SDA reactions, the modified technique can be lower in cost. In addition, using thermostable DNA ligases in the modified method theoretically can provide higher specificity in single-base mismatch discrimination. 3.2. Typical characteristics of CYP2D6 SNP assay Fig. 1 depicts typical time-dependent chemiluminescence response curves of the proposed strategy. In the chemiluminescence measurements, sharp peak-shaped time-dependent signal

Fig. 2. Chemiluminescence spectral responses in CYP2D6*10 genotyping assays. Luminescence spectra were recorded immediately after addition of H2 O2 with an emission slit of 2.5 nm from 350 nm through 500 nm with an interval of 2.5 nm. (a) Control experiment using probes 1 and 3 with no DNA target; (b) control experiment using probes 1 and 3 with 700 pM wild-type CYP2D6 target; (c) control experiment using probes 1 and 3 with 700 pM non-cognate DNA target; (d) responses using probes 1 and 3 with 700 pM mutant CYP2D6*10 target.

profiles were obtained, typical flash-type chemiluminescence of luminol–H2 O2 reaction system [28]. In the system without any target DNA, we obtained a chemiluminescence response of ∼1.45 × 104 cps with a relative standard deviation (RSD) of 3.8% in triplex repeated experiments, as shown in curve (a). Such background signal was generated by a weak hemin-catalyzed luminol–H2 O2 reaction at a very low reaction rate. For the system added 700 pM mutant CYP2D6*10 target and probes 1 and 3, we obtained a much stronger signal ∼6.86 × 105 cps with a RSD of 5.5% in triplex repeated experiments, as shown in curve (d). The ratio of the signal to background was ∼47, a noticeable evidence for the successful ligation of probe 1 with probe 3 and the formation of

Fig. 3. (A) Dependency of discriminant ratio values on hemin concentration using 700 pM mutant and wild-type CYP2D6 target. Discriminant ratio is defined as the ratio of luminescence signal from the mutant target to that from the wild-type. (B) Dependency of chemiluminescence signals on SDA reaction temperature. DNA target concentration is 700 pM. (C) Dependency of chemiluminescence signals on SDA reaction time. DNA target concentration is 700 pM. (D) Dependency of discriminant ratio values on luminol concentration using 700 pM mutant and wild-type CYP2D6 target. Error bars are the standard deviation of three repetitive experiments.

H.-Q. Wang et al. / Analytica Chimica Acta 710 (2012) 111–117

abundant hemin–aptamer complex with substantially enhanced catalytic activity. In contrast, for the system added 700 pM wildtype target and probes 1 and 3, we observed a chemiluminescence response as weak as the background response, ∼1.48 × 104 cps with a RSD of 4.9% in triplex repeated assays as shown in curve (b). It evidenced there were few hemin aptamer sequences in the system, due to the unsuccessful ligation between probes 1 and 3 arousing from single mismatch between probe 1 and wild-type target. These observations exhibited the high specificity of the proposed strategy. Single mismatch between the discriminant probe and the target DNA can be well differentiated by DNA ligase, and the ligation product can only be generated in a perfectly complementary system. Furthermore, a control experiment was performed using 700 pM noncognate DNA target (noncomplementary sequence for probes 1–3) in place of mutant CYP2D6*10 target. Compared with the background response, the similar weak luminescence response was obtained (∼1.42 × 104 cps with a RSD of 3.6% in triplex repeated assays, curve (c)). This control experiment also demonstrated the specificity of the proposed technique. The corresponding chemiluminescence spectra for the proposed allele genotyping assay were also inspected. As shown in Fig. 2, a broad emission peak with a strong response was obtained in the perfectly complementary system of probe 1 and the mutant CYP2D6*10 target, typical luminescence spectra for luminol–H2 O2 reaction in wavelength range from 350 to 500 nm with a maximum at 428 nm. In contrast, very weak luminescence peaks were obtained in the systems of blank, single-base mismatched and noncomplementary target DNA, respectively. The obvious differences in luminescence responses revealed that the amplification of hemin aptamer sequence was specific for the perfectly complementary system, where DNA target can mediate the ligation between the discriminant and the common probes. 3.3. Optimization of assay conditions The assay conditions including hemin concentration, reaction temperature and time of SDA, luminol concentration, have great impact on the luminescence response of the assay. Fig. 3 depicts the relationships between the chemiluminescence responses and these assay conditions. We observed that the luminescence signals of the mutant CYP2D6*10 and wild-type targets, were both enhanced with increasing hemin concentrations. Since the proposed technique was to improve the selectivity of the proposed genotyping assay in detecting DNA target perfectly complementary to the discriminant probe, we chose discrimination ratio, signal of the mutant CYP2D6*10 target divided by that of the wild-type target, to optimize the concentration of hemin. The dependency of the discriminant ratio values on hemin concentration is depicted in Fig. 3A. With the increase of hemin concentration, the discriminant ratio increased at first and then decreased with a maximum achieved at the hemin concentration of 0.5 ␮M. Thus, this hemin concentration was used throughout the subsequent experiments. Reaction temperature had great effect on the efficiency of SDA via altering the activities of Bst DNA polymerase and Nt. Bst NBI nickase. Therefore, different reaction temperatures were investigated. The dependency of luminescence signals on SDA reaction temperature is depicted in Fig. 3B. The maximum luminescence response was achieved while the reaction temperature was 60 ◦ C. Thus, the subsequent experiments were performed at 60 ◦ C. The intensity of luminescence signals also depended on the SDA reaction time. Fig. 3C reveals the dependency of luminescence signals on the SDA reaction time. Obviously, with the time prolonging, luminescence intensity gradually increased and then turned into level off at 2 h. Thus, the reaction time for SDA was set to 2 h throughout subsequent experiments. Discrimination ratio was chosen to optimize the concentration of luminol for the same reason

115

Fig. 4. Time-dependent chemiluminescence signals of the CYP2D6*10 genotyping assay in response to mutant target of varying concentrations. Luminescence signals were recorded at 425 nm with an emission slit of 2.5 nm. Typical chemiluminescence spectra (A) and corresponding chemiluminescence peak intensities at 425 nm (B) versus different target concentrations. Error bars are the standard deviation of three repetitive experiments. Insert: linear relationship between chemiluminescence peak intensity and target concentration.

as hemin concentration optimizing. Fig. 3D depicts the dependency of the discriminant ratio values on luminol concentration. With the increase of luminol concentration, the discriminant ratio increased at first and then turned into level off at the luminol concentration of 0.5 mM. Thus, this luminol concentration was used throughout the subsequent experiments. 3.4. Analytical performance of the SNP typing assay Fig. 4A depicts the typical dependency of the time-dependent chemiluminescence signals of the genotyping assay in response to the mutant CYP2D6*10 target at different concentrations. With the increase of target concentration, the luminescence response became stronger. The peak intensity was increasing with CYP2D6*10 concentration varying from 1 pM to 50 nM as shown in Fig. 4B. Within the range from 1 pM to 1 nM, chemiluminescence responses are in linear correlation to the concentrations of DNA target with a correlation coefficient of 0.9979. The detection limit was 0.1 pM estimated in terms of the 3 rule (three times standard deviation over the blank). Besides, the proposed technique provided a high signal-to-background ratio of ∼57 at a CYP2D6*10 target concentration of 1 nM, implying superb robustness in assay. The assay also exhibited an excellent reproducibility benefitting from a homogeneous format. The RSDs of the luminescence peak intensities were 3.3%, 4.8%, 5.5%, and 4.1% in three repetitive assays of 1 pM, 400 pM, 700 pM, and 1 nM CYP2D6*10 target. Overall, the proposed technique was highly selective and sensitive in CYP2D6*10 genotyping. The fidelity of the ligase

116

H.-Q. Wang et al. / Analytica Chimica Acta 710 (2012) 111–117

Fig. 6. Time-dependent luminescence signals of the CYP2D6*10 genotyping assay in response to five genomic DNA samples. The samples are numbered alphabetically. Each sample is assayed using probes 1 and 3 as well as probes 2 and 3, respectively, in order to determine the genotypes. Each assay is performed twice in parallel.

sample b was heterozygous (DNA was mutant on one chromosome and wild-type on the other chromosome). The identification results were consistent with the sequencing data. This suggested that the proposed strategy might be suitable for related genomic research. 4. Conclusions Fig. 5. Gel electrophoresis analysis of five genomic DNA samples after PCR amplification in 2.5% agarose. Lane M is DNA size marker, and the other lanes a–e are the PCR products of five genomic DNA samples.

furnished the high specificity for the assay, and the amplification reactions offered it desirable sensitivity. What is more, because of the choices of label-free, homogeneous format, simple assay operations, and chemiluminescence-based signal detection, the assay could be robust, cost-efficient, readily automated, and scalable for parallel assays of hundreds of samples, suggesting the proposed technique to be an ideal alternative to existing techniques. 3.5. SNP typing of genomic DNA The proposed technique was further applied to the assay of human genomic DNA samples. Five genomic DNA samples were collected with the genotypes characterized by DNA sequence analysis. PCR was performed on these samples to obtain 271 bp amplicons. The PCR amplicons were confirmed by the 2.5% agarose gel image as shown in Fig. 5. Bright bands can be observed for the PCR amplicons near the position of DNA marker with a length about 300 bp. The amplicons of PCR were directly used for SNP assay using the proposed technique. Fig. 6 depicts the characteristic time-dependant chemiluminescence signals obtained from these amplicons in the present of discriminant probe 1 or 2. We observed that, for samples, a and d, chemiluminescence signals significantly stronger than the background response were obtained by using the discriminant probe 2, while chemiluminescence signals as weak as the background response were obtained by the use of discriminant probe 1, indicating the DNA targets were perfectly complementary to the discriminant probe 2 at the 3 terminus. In other words, the DNA targets in the two samples were identified as homozygous wild-type. For samples, c and e, chemiluminescence signals much stronger than the background response were only obtained by using the discriminant probe 1, so, they can be identified as homozygous mutant CYP2D6*10. For sample b, the use of the discriminant probes 1 and 2 both gave stronger luminescence signals than the background response, suggesting the DNA target in the

We reported a highly specific and sensitive label-free technique for the genotyping of CYP2D6*10 based on ligation reaction and isothermal amplification reaction of SDA. The use of allele ligation reaction furnished a highly specific identification of single nucleotide mutation in CYP2D6. The use of nickasebased SDA enabled efficient amplification to generate abundant peroxidase-mimic signal-reporter sequences, which provided desirable sensitivity for the technique. The proposed technique was demonstrated to have many advantages such as superb robustness, high selectivity and sensitivity, low cost, and simple operations. The combination of microtitre plate formats would facilitate the applications of this technique in high-throughput studies. These features supported its considerable potential in general clinical application. In view of these advantages, the proposed technique was expected to provide an intrinsically robust, highly specific and sensitive genotyping platform for association studies in pharmacogenomics. Acknowledgements The work was supported by NSFC (21025521, 21035001, 20875027), “973” National Key Basic Research Program (2011CB911000), European Commission FP7-HEALTH-2010 Programme-GlycoHIT (260600), National Grand Program on Key Infectious Disease (2009ZX10004-312), CSIRT Program and NSF of Hunan Province (10JJ7002). References [1] M. Pirmohamed, B.K. Park, Trends Pharmacol. Sci. 22 (2001) 298–305. [2] D. Serrano, M. Lazzeroni, C.F. Zambon, D. Macis, P. Maisonneuve, H. Johansson, A. Guerrieri-Gonzaga, M. Plebani, D. Basso, J. Gjerde, G. Mellgren, N. Rotmensz, A. Decensi, B. Bonanni, Pharmacogenomics J. 11 (2011) 100–107. [3] Y. Katoh, S. Uchida, M. Kawai, N. Takei, N. Mori, J. Kawakami, Y. Kagawa, S. Yamada, N. Namiki, H. Hashimoto, Biol. Pharm. Bull. 33 (2010) 285–288. [4] M.C. Rebsamen, J. Desmeules, Y. Daali, A. Chiappe, A. Diemand, C. Rey, J. Chabert, D. Hochstrasser, M.F. Rossier, Pharmacogenomics J. 9 (2009) 34–41. [5] S.F. Zhou, Clin. Pharmacokinet. 48 (2009) 689–723. [6] M. Ingelman-Sundberg, Pharmacogenomics J. 5 (2005) 6–13. [7] R.H. van Schaik, Drug Resist. Updat. 11 (2008) 77–98.

H.-Q. Wang et al. / Analytica Chimica Acta 710 (2012) 111–117 [8] I. Johansson, M. Oscarson, Q.Y. Yue, L. Bertilsson, F. Sjoqvist, M. IngelmanSundberg, Mol. Pharmacol. 46 (1994) 452–459. [9] U.A. Meyer, Lancet 356 (2000) 1667–1671. [10] M. Ingelman-Sundberg, S.C. Sim, A. Gomez, C. Rodriguez-Antona, Pharmacol. Ther. 116 (2007) 496–526. [11] A.K. Daly, B.P. King, J.B. Leathart, Methods Mol. Biol. 320 (2006) 193–207. [12] I.K. Litos, E. Emmanouilidou, K.M. Glynou, E. Laios, P.C. Ioannou, T.K. Chrisopoulos, M. Kampa, E. Castanas, A. Gravanis, Anal. Bioanal. Chem. 389 (2007) 1849–1857. [13] P. Dorado, M.C. Cáceres, E. Pozo-Guisado, M.L. Wong, J. Licinio, A. LLerena, Biotechniques 39 (2005) 571–574. [14] B. Arneth, M. Shams, C. Hiemke, S. Härtter, Clin. Biochem. 42 (2009) 1282–1290. [15] J. Sistonen, S. Fuselli, A. Levo, A. Sajantila, Clin. Chem. 51 (2005) 1291–1295. [16] T. Heller, J. Kirchheiner, V.W. Armstrong, H. Luthe, M. Tzvtkov, J. Brockmöller, M. Oellerich, Ther. Drug Monit. 5 (2006) 673–677. [17] E.Y. Kim, S.S. Lee, H.J. Jung, H.E. Jung, C.W. Yeo, J.H. Shon, J.G. Shin, Clin. Chim. Acta 411 (2010) 2043–2048.

117

[18] S. Germer, M.J. Holland, R. Higuchi, Genome Res. 10 (2000) 258–266. [19] F.J. Steemers, W. Chang, G. Lee, D.L. Barker, R. Shen, K.L. Gunderson, Nat. Methods 3 (2006) 31–33. [20] M.R. Lockett, M.R. Shortreed, L.M. Smith, Anal. Chem. 79 (2007) 6031–6036. [21] J.V. Ness, L.V. Ness, D.J. Galas, Proc. Natl. Acad. Sci. U.S.A. 100 (2003) 4504–4509. [22] H.Q. Wang, W.Y. Liu, Z. Wu, L.J. Tang, X.M. Xu, R.Q. Yu, J.H. Jiang, Anal. Chem. 83 (2011) 1883–1889. [23] Y. Huang, Y.L. Zhang, X. Xu, J.H. Jiang, G.L. Shen, R.Q. Yu, J. Am. Chem. Soc. 131 (2009) 2478–2480. [24] J. Li, X. Chu, Y. Liu, J.H. Jiang, Z. He, Z. Zhang, G. Shen, R.Q. Yu, Nucleic Acids Res. 33 (2005) e168. [25] Y. Li, C.R. Geyer, D. Sen, Biochemistry 35 (1996) 6911–6922. [26] V. Pavlov, Y. Xiao, R. Gill, A. Dishon, M. Kotler, I. Willner, Anal. Chem. 76 (2004) 2152–2156. [27] J. Liu, Z. Cao, Y. Lu, Chem. Rev. 109 (2009) 1948–1998. [28] B.A. Tannous, M. Verhaegen, T.K. Christopoulos, A. Kourakli, Anal. Biochem. 320 (2003) 266–272.