Ferrocenylnaphthalene Diimide-Based Electrochemical Hybridization

Ferrocenylnaphthalene Diimide-Based Electrochemical Hybridization Assay for a Heterozygous Deficiency of the Lipoprotein Lipase Gene. Kenichi Yamashit...
0 downloads 0 Views 273KB Size
Download PDF
Bioconjugate Chem. 2002, 13, 1193−1199

1193

Ferrocenylnaphthalene Diimide-Based Electrochemical Hybridization Assay for a Heterozygous Deficiency of the Lipoprotein Lipase Gene Kenichi Yamashita,† Atsuko Takagi,§ Makoto Takagi,† Hiroki Kondo,| Yasuyuki Ikeda,‡ and Shigeori Takenaka†,* Department of Applied Chemistry, Faculty of Engineering, Kyushu University, Fukuoka 812-8581, Japan; Department of Biochemical Engineering and Science, Kyushu Institute of Technology, Iizuka 820-8502, Japan; Departments of Etiology and Pathophysiology, and Pharmacology, National Cardiovascular Center Research Institute, Fujishirodai, Suita, Osaka 565-8565, Japan. Received February 25, 2002; Revised Manuscript Received July 5, 2002

An electrochemical hybridization assay has been devised that enables the rapid analysis of a heterozygous deficiency of the human lipoprotein lipase (LPL) gene. PCR products of 350 base pairs (bp) containing the wild-type sequence, a mutated G818 f A transition or a G916 deletion of the LPL gene were subjected to hybridization with a probe DNA of 13 or 15 bases that represented either the wild-type or the mutated sequence immobilized on a gold electrode. The differential pulse voltammetry of the electrode before and after hybridization was determined in the presence of ferrocenylnaphthalene diimide (FND) at 460 mV. The measured change in peak current, ∆i, was defined by (i - io)/io × 100%, where io and i represent the current before and after hybridization, respectively. Matched combinations of sample and probe gave ∆i values of 40-90%, whereas mismatched combinations gave values of 20-35%, enabling the discrimination of matched hybrids from mismatched ones across a slim margin. Because the heterozygote contains both the wild-type and mutated sequences, however, it alone gives large ∆i values with both the wild- and mutant-type probes. This system was validated on 10 unknown samples of each of the two types of LPL mutation, which were correctly identified in every case.

INTRODUCTION

The analysis of single nucleotide polymorphisms (SNPs) in genes is important both for diagnosing the risk of common diseases such as hypertension and diabetes and for determining the optimum tailor-made medications (1, 2). Each human gene is classified by two identical alleles in homozygotes or two different alleles in heterozygotes. Because the differences in SNPs are so subtle, it is essential that any analytical method for detecting SNPs is sensitive and efficient in order to be practical. Currently, SNP analysis is carried out by one of two techniques that are based on enzymatic reactions (3, 4) and mismatch detection on the DNA duplex (5, 6). The latter technique is especially useful, as it is an extension of a routine DNA probe method. Recently, mismatch detection by means of an electrochemical DNA sensor has been reported (7-12). The advantages of this technique lie in its high analytical sensitivity, speed, and low cost. Two notable variations of this technique have been reported: Willner and co* To whom correspondence should be addressed. Tel and Fax: +81-92-642-3603, e-mail: [email protected]. † Department of Applied Chemistry, Faculty of Engineering, Kyushu University. ‡ Department of Etiology and Pathophysiology, National Cardiovascular Center Research Institute. § Department of Pharmacology, National Cardiovascular Center Research Institute. | Department of Biochemical Engineering and Science, Kyushu Institute of Technology, Iizuka 820-8502, Japan.

workers (9) carried out heterozygote detection using enzyme amplification, and Marrazza and co-workers (12) carried out heterozygote detection using the chronopotentiometric stripping analysis (PSA) developed by Wang (7). We have also succeeded in highly sensitive detection of DNA using an electrochemical hybridization assay (11, 13, 14). Our breakthrough was the design and synthesis of ferrocenylnaphthalene diimide (FND) as an effective indicator of hybridization. In this paper, we have assessed our electrochemical hybridization assay using two electrodes with different immobilized-DNA probes for the analysis of human lipoprotein lipase (LPL) gene. Human LPL protein is expressed mainly in adipose tissue and hydrolyzes triglycerides in chylomicrons and in very low-density lipoprotein (VLDL) particles in order to supply the resulting fatty acids and triglycerols to adipose tissue, where they are stored effectively (15, 16). A malfunctional form of LPL expressed by homozygotes carrying a mutated LPL gene is known to lead to hypertriglyceridemia with a high probability of pancreatitis (17). People with heterozygous LPL deficiency usually manifest with a normolipidemic state, but they can develop type IV hyperlipoproteinemia through the superposition of factors that stimulate triglyceride synthesis, such as high alcohol intake and/or a hyperinsulinemic state (18-22), which is thought to be a risk for cardiovascular heart disease (23). The identification of heterozygous mutations in the LPL gene as an early determination of the etiology underlying type IV hyperlipoproteinemia is therefore important for preventing the progress of hypertriglyceridemia and the subsequent

10.1021/bc025519u CCC: $22.00 © 2002 American Chemical Society Published on Web 11/05/2002

1194 Bioconjugate Chem., Vol. 13, No. 6, 2002

Yamashita et al.

Figure 1. Organization of the human LPL gene with the focus on exon 5, where the two types of mutations studied here are located (sites 1 and 2) (19,24). The partial nucleotide sequences of the WT for sites 1 and 2 are depicted in (A), together with their complementary probe sequences (13 bases each). The asterisk in (B) indicates the G to A transition at base 818 (GG818G f GAG), which results in the amino acid substitution Gly188 f Glu (G188E), together with the G188E mutant type (MT) probe (15 bases). The asterisk in (C) indicates the deletion of G at base 916 (G916CT; first position of Ala221), which is designated the Arita mutation, which results in premature termination by a frameshift, together with the Arita MT probe (13 bases).

outbreak of cardiovascular disease (23), as the patient can be advised to change to a more healthy life style. The heterozygous LPL deficiencies studied here are a missense mutation carried on the G188E (GG818G f GAG/Gly188 f Glu, site 1) allele (24) and a one-base deletion carried on the LPL arita allele (G916 f deletion, site 2) (19), which are both found on exon 5 (Figure 1). We subjected 350-bp PCR products comprising these mutated sites to our electrochemical assay. Because these two mutations are the major forms of mutation in the LPL gene, their successful analysis should lead to the analogous analysis of most other LPL mutations. RESULTS

Principle of the Electrochemical Hybridization Assay for Heterozygotes. The principles underlying the electrochemical hybridization assay using FND on a gold electrode with a wild-type (WT) probe immobilized for discriminating the WT allele from a mutant-type (MT) allele are illustrated in Figure 2. Before hybridization, the differential pulse voltammetry (DPV) of this electrode is measured at 460 mV, and the current obtained is designated as io. The same gold electrode is then allowed to hybridize with the PCR products prepared from a WT/ WT homozygote, a WT/MT heterozygote, and an MT/MT homozygote under conditions that have been optimized (e.g., the salt concentration of the hybridization buffer and the temperature) for hybridization of the WT probe with the WT allele. The electrode gives rise to a larger signal, designated as i, for PCR products from the WT/ WT homozygote and the WT/MT heterozygote, owing to the intercalation of FND into the DNA duplex that forms on it; however, the signal increase is much smaller with

PCR products from the MT/MT homozygote because of the one-base mismatch. This signal change is converted into the net increase in current, ∆i, which is defined by the equation: (i - io)/io × 100. In this assay system, a significant increase in current is observed for a combination of the WT probe and the WT allele from the WT/WT homozygote and the WT/MT heterozygote, but a marginal increase is observed for the WT probe and the MT allele from the MT/MT homozygote. Conversely, a gold electrode with an MT probe immobilized produces a significant increase in current for the MT allele from a WT/MT heterozygote and an MT/ MT homozygote, but a marginal increase for the WT allele from a WT/WT homozygote. Thus, only the heterozyote produces a large ∆i with both electrodes which allows its identification. If the differences observed are large enough, the WT allele can be discriminated from the MT allele with certainty by this assay system. In other words, a heterozygous genotype can be identified correctly by a significant value of ∆i obtained with two electrodes carrying either a WT probe or an MT probe. Experimentally, a gold electrode carrying a DNA probe was prepared by immobilizing 0.1 pmol of oligonucleotide on a 2 mm2 gold surface. This electrode gave rise to a current of 1 µA at 460 mV in the presence of 0.05 mM FND (100 pmol), 80% of which seemed to derive from nonspecific binding and/or diffusion of the ligand (unpublished observation). The current increased by 0.30.9 µA upon hybridization of 0.2 pmol of a denatured PCR product. An increase in current above 0.3 µA was ascribed to the FND bound to the duplex formed on the electrode, whereas that below 0.3 µA was ascribed to nonspecific binding to the electrode of sample DNA and FND in

Electrochemical Hybridization Assay

Bioconjugate Chem., Vol. 13, No. 6, 2002 1195

Figure 2. Principles underlying the electrochemical hybridization assay with FND for heterozygotes. (A) An electrode with an immobilized wild-type (WT) probe is allowed to hybridize with PCR products from WT/WT, WT/MT (mutant), or MT/MT genotypes. Differential pulse voltammetry (DPV) of the electrode is measured in the presence of FND at 460 mV before and after hybridization. Samples containing the WT allele (WT/WT and WT/MT) give rise to a large signal increase after hybridization. (B) The peak current due to FND before (io) and after hybridization (i) is converted to ∆i, which represents a net increase in the current on duplex formation. (C) Identification of the state of zygosis. The same experiment as in (A) is carried out with an electrode carrying an MT probe, which gives a large signal increase with samples containing the MT allele (WT/MT and MT/MT). In other words, only heterozyogous samples give a large ∆i with both types of electrode, thereby allowing their identification.

succession (unpublished observation). It thus follows that 3.7 pmol of FND binds to the electrode in a desired fashion, that is, one FND molecule binds to the DNA duplex every five base pairs. Given a large difference in these systems, this estimate is roughly consistent with previous data showing that FND binds to duplex DNA at every other base pair in solution (14). Evaluation of the Electrochemical Hybridization Assay System. To establish optimal conditions for hybridization between the PCR product of exon 5 (350 bp) in the human LPL gene and an allele-specific DNA probe (Figure 1) immobilized on a gold electrode, both the salt concentration of the buffer and the temperature were surveyed for the following four combinations: a site 1 G188 WT probe (13 bases) and a WT allele of exon 5; a G188E MT probe (15 bases) and a G188E allele; a site 2 WT probe (13 bases) and a WT allele; and an Arita MT probe (13 bases) and an Arita allele. The best conditions were found to be 0.25 x SSC (0.3 M sodium citrate and 3 M NaCl) and 35 °C for the G188 WT probe and the WT allele; 0 x SSC and 35 °C for the G188E MT probe and the G188E allele; 0.25 x SSC and 40 °C for the site 2 WT probe and the WT allele; 0 x SSC and 40 °C for the Arita MT probe and the Arita allele after several experiments under the varied salt conditions (0-0.5 × SSC) and temperature (15-60 °C). Under the best conditions, the PCR products for site 1 of exon 5 (350 bp), prepared from a WT/WT homozygote, a WT/G188E heterozygote and a G188E/G188E homozygote were examined by the electrochemical hybridization assay in the presence of FND on a gold electrode with either the G188 WT probe or the G188E MT probe immobilized. As shown in Figure 3A, the assay with the G188 WT probe gave a ∆i (%) of 52 ( 19 (mean ( standard deviation) for WT/WT homozygotes (n ) 17), 47 ( 15 for WT/G188E heterozygotes (n ) 17), and 17 (

10 for G188E/G188E homozygotes (n ) 17) in 0.25 x SSC at 35 °C. The difference between the G188E/G188E and the WT/WT homozygotes was statistically significant as was the difference between the G188E/G188E homozygote and the WT/G188E heterozygote (p < 0.0001 in both cases). These findings demonstrate that the G188 WT probe hybridizes with the G188E allele only weakly because of the one-base mismatch, and that the G188E allele can be identified with certainty under the assay conditions employed. The threshold value that distinguishes the WT allele from the G188E allele was estimated from the ∆i values (mean ( SD) of the above three groups as follows: [(52 - 19) + (47 - 15) + (17 + 10)]/3 ) 31%. The electrochemical hybridization assay with the G188E MT probe gave a ∆i (%) of 20 ( 13 (mean ( SD) for WT/ WT homozygotes (n ) 20), 60 ( 15 for WT/G188E heterozygotes (n ) 20) and 67 ( 31 for G188E/G188E homozygotes (n ) 20) in 0 x SSC at 35 °C (Figure 3B). The differences between the WT/WT homozygote and the G188E/G188E homozygote and those between the WT/ WT homozygote and the WT/G188E heterozygote were statistically significant, indicating that the WT allele can also be discriminated by using the G188E MT probe. The threshold value that distinguishes the WT allele from the G188E allele was estimated from the ∆i values (mean ( SD) of the above three groups as follows: [(20 + 13) + (60 - 15) + (67 - 31)]/3 ) 38%. Likewise, the electrochemical hybridization assay with the site 2 WT probe gave a ∆i (%) of 44 ( 17 (mean ( SD) for WT/WT homozygotes (n ) 20), 41 ( 11 for WT/ Arita heterozygotes (n ) 20), and 16 ( 11 for Arita/Arita homozygotes (n ) 20) in 0.25 x SSC at 40 °C, as shown in Figure 3C. The differences between the Arita/Arita homozygote and the WT/WT homozygote and those between the Arita/Arita homozygote and the WT/Arita

1196 Bioconjugate Chem., Vol. 13, No. 6, 2002

Figure 3. Discrimination between wild-type (WT) and mutanttype (MT) alleles of the human LPL gene by the electrochemical hybridization assay. In A and B, 350-bp PCR products, prepared from a WT/WT homozygote, a WT/G188E heterozygote, and a G188E/G188E homozygote, were analyzed with a gold electrode carrying a G188 WT probe (A) or a G188E MT probe (B) in the presence of 0.05 mM FND. In C and D, 350-bp PCR products, prepared from a WT/WT homozygote, a WT/Arita heterozygote, and an Arita/Arita homozygote, were analyzed with a gold electrode carrying a site 2 WT probe (C) or an Arita MT probe (D) in the presence of 0.05 mM FND. The currents obtained are shown as mean ( standard deviation (SD). The difference between the two combinations marked by the asterisk is statistically significant at p < 0.0001.

heterozygote were significant statistically (p < 0.0001). The threshold value that distinguishes the WT allele from the Arita MT allele was estimated from the ∆i values (mean ( SD) of the above three groups as follows: [(44 - 17) + (41 - 11) + (16 + 11)]/3 ) 28%. The electrochemical hybridization assay with the Arita MT probe gave a ∆i (%) of 20 ( 8 (mean ( SD) for WT/ WT homozygotes (n ) 20), 53 ( 13 for WT/Arita heterozygotes (n ) 20), and 53 ( 16 for Arita /Arita homozygotes (n ) 20) in 0 x SSC at 40 °C (Figure 3D). Again, the differences between the WT/WT homozygote and the Arita/Arita homozygote and those between the WT/WT homozygote and the WT/Arita heterozygote were significant statistically. These findings indicate that the Arita MT probe hybridizes with the WT allele only weakly because of the one-base mismatch, and that it can discriminate the WT allele with certainty under the conditions employed. The threshold value that distinguishes the WT allele from the Arita allele was estimated from the ∆i values (mean ( SD) of the above three groups as follows: [(20 + 8) + (53 - 13) + (53 - 16)]/3 ) 35%. Blind Test of the Electrochemical Hybridization Assay System. The validity of the electrochemical hybridization assay for discriminating between a WT allele and an MT allele was examined in a blind test.

Yamashita et al.

Figure 4. Verification of the electrochemical hybridization assay with FND by testing unknown samples for their LPL genotypes. The 350-bp PCR products, prepared from 10 unknown blood samples, were analyzed with a gold electrode carrying a site 1 G188 WT probe (A) or a G188E MT probe (B) in the presence of 0.05 mM FND. Analysis was carried out for four distinct samples per sample number (n ) 4), and the currents shown are the mean ( standard error of the mean (SEM). The genotype of each sample was judged on the basis of the threshold values at which the WT allele can be discriminated from the MT allele taken from the data shown in Figures 3A and 3B; that is, 31% for the site 1 G188 WT probe (A) and 38% for the G188E MT probe (B). The threshold values are represented by the dotted horizontal lines in A and B.

PCR products amplified from site 1 of exon 5 (350 bp) in the human LPL gene were prepared from 10 unknown samples and examined by the electrochemical hybridization assay in the presence of FND on a gold electrode with an immobilized G188 WT probe or a G188E MT probe. The data obtained were judged on the basis of the threshold values (31 and 38%) derived from the data shown in Figures 3A and 3B. The genotypes of these samples determined by the hybridization assay were found to be completely consistent with the results from DNA sequencing, as shown in Figure 4. In other words, it appears that this assay system is effective for analyzing unknown or practical samples. Another blind test was carried out for site 2 of exon 5 of the human LPL gene. PCR products amplified from this site (350 bp) were prepared from 10 unknown samples and analyzed by the electrochemical hybridization assay in the presence of FND on a gold electrode with an immobilized site 2 WT probe or an Arita MT probe. The data obtained were judged on the basis of the threshold values (28 and 35%) derived from the data shown in Figures 3C and 3D. The genotypes of these samples determined by the hybridization assay were completely consistent with results from DNA sequencing, as shown in Figure 5. Again, the usefulness of this assay system was proved by these site 2 samples. In future studies it would be worth investigating this utility of this electrochemical method for the practical analysis of a wider range of genes with known mutations. DISCUSSION

The analysis of SNPs has become more important since it has been closely associated with the prediction and prevention of various syndromes. Currently, SNPs are analyzed by direct DNA sequencing or new techniques

Electrochemical Hybridization Assay

Figure 5. Further verification of the electrochemical hybridization assay with FND by testing unknown samples for their LPL genotypes. The 350-bp PCR products, prepared from 10 unknown blood samples, were analyzed with a gold electrode carrying a site 2 WT probe (A) or an Arita MT probe (B) in the presence of 0.05 mM FND. Analysis was carried out as in Figure 4. The genotype of each sample was judged on the basis of the threshold values at which the WT allele can be distinguished from the MT allele taken from the data shown in Figures 3C and 3D; that is, 28% for the WT probe (A) and 35% for the Arita MT probe (B). The threshol values are represented by the dotted horizontal lines in A and B.

such as dynamic allele-specific hybridization (DASH) (5) and template-directed dye-terminator incorporation (TDI) (3). The demand for simpler methods never ceases, however, as normally several samples need to be processed at any one time during SNP analysis. An electrochemical method seems to be a promising alternative to the existing fluorescence methods owing to its speed, low operational costs, and amenability to automation. In fact, the highly sensitive detection of gene mutations has been achieved electrochemically by using Co(phen)33+ or dounomycin as a redox active marker (9, 12). As described above, we have been able to analyze two types of mutations present on exon 5 of the human LPL gene electrochemically with FND. The key to this success lies in the use of two different DNA probes, comprising WT and mutant DNA, immobilized on an electrode. Only heterozygous samples produce a large increase in current with both electrodes, thereby allowing identification of the sample as heterozygous. Samples containing as little as 0.2 pmol of PCR product can be analyzed by this method, which corresponds to a sensitivity that is 1000-fold higher than that of existing methods. Once the electrodes with the immobilized probes have been prepared, the analysis is straightforward. First, chromosomal DNA is extracted from blood donated by individuals and subjected to PCR. Second, after denaturation, the PCR products are subjected to the electrochemical analysis. This method is based on the fact that the DNA duplex does not form with DNA sequences containing a mismatch(es). Currently, there is only a modest 3-4-fold difference in the value of ∆i between the WT and MT alleles. If this difference can be magnified, however, the analysis will become more sensitive, and efforts toward this goal are being made in our laboratories. Another area that needs future investigation concerns the data values, which fluctuate rather significantly even

Bioconjugate Chem., Vol. 13, No. 6, 2002 1197

among given genotypes (ca. 50% of the mean). At present, the exact reason for this fluctuation is not known, but it is possible that material affecting the electrochemical reaction may not have been removed completely by the methods used to isolate the chromosomal DNA. In regard to the background current, its magnitude tends to increase with increasing sizes of DNA; that is, nonspecific adsorption to the electrode becomes marked with DNA larger than 40 bases. Nonspecific binding can be eliminated by warming, but this process can also partially detach the probe DNA from the electrode. A solution to this problem may be the use of other masking agents such as mercaptohexanol instead of 2-mercaptoethanol (unpublished observation). Despite these limitations, the SNPs of the human LPL gene could be analyzed successfully and practically by our electrochemical hybridization assay with FND. Although only the two types of known mutations were examined in this study, other types of mutations may be analyzed easily if electrodes carrying suitable probes are prepared. Moreover, an electrochemical DNA microarray system based on this technique may be even more useful. Studies are now under way to attain this goal in our laboratories. EXPERIMENTAL SECTION

Extraction of Chromosomal DNA. Human chromosomal DNA was isolated from whole cells taken from the peripheral blood of a healthy male YI, patient MN (24) carrying a heterozygous LPL G188E allele (GG818G f GAG/Gly188 f Glu; corresponding to a G-to-A transition at position 818 of exon 5 of the LPL gene), patient EN (19) carrying a heterozygous LPLarita allele (G916 f deletion; corresponding to deletion of G at position 916 on exon 5), and patient TN (19) who was homozygous for the LPLarita allele, using a QIAamp DNA Blood kit (QIAGEN GmbH, Germany). With respect to the LPL gene, the genotypes of these individuals were WT/WT, WT/G188E, WT/arita and arita/arita, respectively. PCR Amplification and Purification of PCR Products. Exon 5 (base numbers 730-963, 234 bp) and its flanking regions (40 bp each side) of the LPL gene were amplified from chromosomal DNA by PCR using the following primer pair: forward (5′-TGT AAA ACG ACG GCC AGT AAA TTT ACA AAT CTG TGT TCC TGC TTT TT-3′, No. 259) and reverse (5′-CAG GAA ACA GCT ATG ACC GAT AAG AGT CAC ATT TAA TTC GCT TCT A-3′, No. 260). Note that each primer contains an extra sequence of 18 bases at their 5′-terminus for direct sequencing, making the size of the PCR product 350 bp. PCR was performed in a 100 µL volume consisting of 1 x PCR buffer containing 1.5 mM MgCl2, 10 pmol of each of the forward and reverse primers, 50 ng of chromosomal DNA, 200 µM dNTPs, and 2.5 U of AmpliTaq Gold (Perkin-Elmer). The PCR reaction protocol was 95 °C for 10 min, 40 cycles of 94 °C for 0.5 min, 55 °C for 1 min, 72 °C for 2 min, and finally an incubation at 72 °C for 7 min in a GeneAmp PCR System 9600 (Perkin-Elmer). Each PCR product was purified by a QIAquick spin column in a QIAquick PCR Purification Kit (QIAGEN), and its integrity was assessed by agarose gel electrophoresis. In Vitro Preparation of the Homozygote of LPL G188E. Exon 5 and its flanking regions of the LPL gene were amplified from the chromosomal DNA of patient MN carrying a heterozygous LPL G188E allele by PCR using the primer pair described above, and the PCR product (350 bp) was purified with a QIAquick spin

1198 Bioconjugate Chem., Vol. 13, No. 6, 2002

column. After ligation into a pCR2.1 vector, the recombinant was transformed into transfection-competent E. coli cells using an Original TA Cloning Kit (Invitrogen, Carlsbad, CA). After verifying the presence of the G188E mutation by Ava II digestion and DNA sequencing, the 350-bp DNA fragment containing the G188E mutation was amplified from this subclone by PCR using the primer pair described above, purified with a QIAquick Spin column, and used for hybridization experiments as a genotype of G188E/G188E. DNA Probes. DNA probes for site 1 G188 WT and G188E MT, site 2 WT and arita MT with a 5′-thiol moiety (Figure 1) were custom synthesized by Hokkaido System Science Co. Ltd. (Sapporo, Japan), and their concentrations were estimated from their molar absorptivities at 260 nm (25): 119 200 cm-1 M-1, 142 100 cm-1 M-1, 119 000 cm-1 M-1, and 120100 cm-1 M-1, respectively. Preparation of Probe DNA-Immobilized Electrodes. Hybridization of the PCR Product with Immobilized Probe DNA. A solution of the PCR product in 10 mM Tris HCl buffer (pH 8.5) was mixed with a small volume of 20 x SSC (0.3 M sodium citrate and 3 M NaCl) to bring the final concentration to 0 x SSC or 0.25 x SSC for hybridization. Before hybridization, DNA was denatured in boiling water for 10 min and then cooled quickly on liquid nitrogen. The denatured PCR product was kept at 0 °C until use. One microliter of the denatured PCR product (ca. 0.2 pmol) was placed on the electrode and hybridization with the DNA probe was allowed to proceed for 1 h at 20 °C. Electrochemical Measurements. Differential pulse voltammetry (DPV) was measured with an ALS model 900 electrochemical analyzer (CH Instruments Inc., Austin, TX), using a normal three-electrode configuration consisting of an Ag/AgCl reference electrode, a Pt counter electrode, and the electrode with the immobilized DNA probe, before and after the hybridization reaction in 2 mL of 0.1 M AcOH-AcOK buffer (pH 5.6) containing 0.1 M KCl and 0.05 mM FND. Measurement temperatures were optimized at 40 °C for the site 1 DNA probes and at 35 °C for site 2 DNA probes. The conditions for DPV measurements were as follows: initial potential, 0 V; final potential, 0.6 V; scan rate, 100 mV/s; pulse amplitude, 0.05 V; sample width, 16.7 ms; pulse period, 0.2 s; pulse width, 0.05 s; quiet time, 2 s; sensitivity, 10-6 A/V. Peak currents due to FND were measured at 460 mV before and after hybridization with the sample DNA. Data Treatment. Data were converted into ∆i values, which were defined as (i - io)/io × 100%, where io and i refer to the current before and after hybridization, respectively. ∆i represents a net increase in the current of FND bound to the duplex DNA formed per immobilized DNA probe on the electrode. Nearly 30% of the observed current could be accounted for by the background as noted previously (26). The statistical significance of the differences in ∆i among three groups was evaluated by Scheffe’s F-test in a one-factor factorial ANOVA (27). ACKNOWLEDGMENT

The authors thank Ms. N. Fujino, M. Yokoyama, and N. Katakura for expert technical assistance. This work was supported in part by a Grant-in-Aid for Scientific Research on Priority Areas (A) (No. 13031063) from the Ministry of Education, Culture, Sports, Science and Technology and Grants-in-Aid for Scientific Research (A) (No. 12305054), (B)(No. 12555236), (C)(Nos. 13680672, 11670402, 12670384, and 14570376), a Grant-in-Aid for Research (H13-Tiryo-001) on Advanced Medical Technol-

Yamashita et al.

ogy from the Ministry of Health Labor and Welfare of Japan, and a Grant-in-Aid for JSPS Fellows (No. 200000715) from the Japan Society for the Promotion of Science. The authors are also grateful for financial support from the New Energy and Industrial Technology Development Organization (NEDO). LITERATURE CITED (1) Brookes, J. (1999). The essence of SNPs. Gene 234, 177186. (2) McCarthy, J. J., and Hilfiker R. (2000). The use of singlenucleotide polymorphism maps in pharmacogenomics. Nature Biotechnol. 18, 505-508. (3) Chan, X., Zehnbauer, B., Gnirke, A., and Kwok, P.-Y. (1997). Fluorescence energy transfer detection as a homogeneous DNA diagnostic method. Proc. Natl. Acad. Sci. U.S.A. 94, 10756-10761. (4) Dubiley, S., Kirillov, E., Lysov, Y., and Mizabekov, A. (1997). Fractionation, phosphorylation and ligation on oligonucleotide microchips to enhance sequencing by hybridization. Nucleic Acids Res. 25, 2259-2265. (5) Howell, W. M., Jobs, M., Gyllensten, U., and Brookes, A. J. (1999). Dynamic allele-specific hybridization. Nature Biotechnol. 17, 87-88. (6) Pease, A. C., Solas, D., Sullivan, E. J., Cronin, M. T., Holmes, C. P., and Fodor, A. P. A. (1994). Light-generated oligonucleotide arrays for rapid DNA sequence analysis. Proc. Natl. Acad. Sci. U.S.A. 91, 5022-5026. (7) Wang, J., Palecek, E., Nielsen, P. E., Rivas, G., Cai, X., Shiraishi, H., Dontha, N., Luo, D., and Farias, A. M. (1996). Peptide nucleic acid Probes for sequence-specific DNA biosensors. J. Am. Chem. Soc. 118, 7667-7670. (8) Wang, J., Rivas. G., Cai, A., Chicharro, M., Parrado, C., Dontha, N., Bergleiter, A., Mowat, M., Palecek, E., and Nielsen, P. E. (1997). Detection of point mutation in the p53 gene using a peptide nucleic acid biosensor. Anal. Chim. Acta 344, 111-118. (9) Patolsky, F., Lichtenstein, A., and Willner, I. (2001). Detection of single-base DNA mutations by enzyme-amplified electronic transduction. Nature Biotechnol. 19, 253-257. (10) Boon, E. M., Ceres, D. M., Drummond, T. G., Hill, M. G., and Barton, J. K. (2000) Mutation detection by electrocatalysis at DNA-modified electrodes. Nature Biotechnol. 18, 10961100. (11) Yamashita, K., Takagi, M., Kondo, H., and Takenaka, S. (2000) Electrochemical detection of base pair mutation. Chem. Lett. 1038-1039. (12) Marrazza, G., Chiti, G., Maschini, M., and Anichini, M. (2000). Detection of human apolipoprotein E genotypes by DNA electrochemical biosensor coupled with PCR. Clin. Chem. 46, 31-37. (13) Takenaka, S., Uto, Y., Saita, H., Yokoyama, M., Kondo, H., and Wilson, W. D. (1998). Electrochemically active threading intercalator with high double stranded DNA selectivity. J. Chem. Soc., Chem. Commun. 1111-1112. (14) Takenaka, S., Yamashita, K., Takagi, M., Uto, Y., and Kondo, H. (2000). DNA sensing on a DNA probe-modified electrode using ferrocenylnaphthalene diimide as the electrochemically active ligand. Anal. Chem. 72, 1334-1341. (15) Havel, R. J., Goldstein, J. L., and Brown, M. S. (1980). Lipoproteins and lipid transport, in Metabolic Control and Disease, Saunders, Philadelphia. (16) Krauss, R. M., Levy, R. I., and Fredrickson, D. S. (1974). Selective measurement of two lipase activities in postheparin plasma from normal subjects and patients with hyperlipoproteinemia. J. Clin. Invest. 54, 1107-1124. (17) Brunzell, J. D. (1995). Familial lipoprotein lipase deficiency and other causes of the chylomicronemia syndrome. The metabolic and molecular bases of inherited disease, 7th ed, Vol. II, McGraw-Hill, New York. (18) Takagi, A., Ikeda, Y., Mori, A., Tsutsumi, Z., Oida, K., Nakai, T., and Yamamoto, A. (1994). A newly identified heterozygous lipoprotein lipase gene mutation (Cys239 f stop/TGC972 f TGA.; LPL obama) in a patient with primary type IV hyperlipoproteinemia. J. Lipid Res. 35, 2008-2018.

Electrochemical Hybridization Assay (19) Takagi, A., Ikeda, Y., Tsutsumi, Z., Shoji, T., and Yamamoto, A. (1992). Molecular studies on primary lipoprotein lipase (LPL) deficiency. One base deletion (G916) in exon 5 of LPL gene causes no detectable LPL protein due to the absence of LPL mRNA transcript. J. Clin. Invest. 89, 581591. (20) Syvanne, M., Antikainen, M., Ehnholm, S., Tenkanen, H., Lahdenpera, S., Ehnholm, C., and Taskinen, M. R. (1996). Heterozygosity for Asn291 f Ser mutation in the lipoprotein lipase gene in two Finnish pedigrees: effect of hyperinsulinemia on the expression of hypertriglyceridemia. J. Lipid Res. 37, 727-738. (21) Takagi, A., Ikeda, Y., Takeda, E., and Yamamoto, A. (2000). A newly identified lipoprotein lipase (LPL) gene mutation (F270L) in a Japanese patient with familial LPL deficiency. Biochim. Biophys. Acta 1502, 433-446. (22) Emi, M., Wilson, D. E., Iverius, P. H., Wu, L., Hata, A., Hegele, R., Williams, R. R., and Lalouel, J. M. (1990). Missense mutation (Gly f Glu188) of human lipoprotein lipase imparting functional deficiency. J. Biol. Chem. 265, 5910-5916.

Bioconjugate Chem., Vol. 13, No. 6, 2002 1199 (23) Austin, M. A., Hokanson, J. E., and Edwards, K. L. (1998). Hypertriglyceridemia as a cardiovascular risk factor. Am. J. Cardiol. 81, 7B-12B. (24) Takagi, A., Ikeda, Y., Tachi, K., Shinozuka, T., and Yamamoto, A. (1999). Identification of compound heterozygous mutations (G188E/W382X) of lipoprotein lipase gene in a Japanese infant with hyperchylomicronemia: the G188E mutation was newly identified in Japanese. Clin. Chim. Acta 285, 143-154. (25) Cantor, R., and Warshaw, M. M. (1970). Oligonucleotide interactions. 3. Circular dichroism studies of the conformation of deoxyoligonucleotides. Biopolymers 9, 1059-1077. (26) Miyahara, H., Yamashita, K., Kanai, M., Uchida, K., Takagi, M., Kondo, H., and Takenaka, S. Electrochemical analysis of single nucleotide polymorphisms of p53 gene. Talanta, in press. (27) Wonnacott, T. H., and Wonnacott, R. J. (1990). Introductory statistics, 5th ed., John Wiley & Sons, New York.

BC025519U