RAFT-Generated Polyacrylamide-DNA Block Copolymers for Single

Feb 27, 2009 - After a capillary tube was filled with the running buffer solution of PAAm-b-ODN, a mixture of normal and mutant ssDNA was subjected to...
0 downloads 0 Views 1MB Size
Download PDF
Biomacromolecules 2009, 10, 805–813

805

RAFT-Generated Polyacrylamide-DNA Block Copolymers for Single-Nucleotide Polymorphism Genotyping by Affinity Capillary Electrophoresis Naoki Kanayama,† Hideaki Shibata, Ayumi Kimura, Daisuke Miyamoto, Tohru Takarada,* and Mizuo Maeda Bioengineering Laboratory, RIKEN Advanced Science Institute, 2-1 Hirosawa, Wako, Saitama 351-0198, Japan Received November 12, 2008; Revised Manuscript Received January 22, 2009

Capillary electrophoretic separation of a mixture of 5′-fluorescein isothiocyanate-labeled single-stranded DNA (normal ssDNA) and its single-base-substituted one (mutant ssDNA) was achieved by using a RAFT-generated polyacrylamide-oligodeoxyribonucleotide block copolymer (PAAm-b-ODN) as an affinity polymeric probe. PAAmb-ODN was synthesized through the Michael addition of thiol-terminated PAAm (PAAm-SH) to 5′-maleimidemodified ODN. PAAm-SH was derived from dithiobenzoate-terminated PAAm prepared via RAFT polymerization. The number-averaged molecular weight (Mn) and the molecular weight distribution were determined by aqueous size exclusion chromatography. After a capillary tube was filled with the running buffer solution of PAAm-bODN, a mixture of normal and mutant ssDNA was subjected to electrophoresis and detected by a laser-induced fluorescent detector. Because the base sequence of PAAm-b-ODN was complementary to part of the mutant ssDNA, including a single-base substitution site, the electrophoretic migration of mutant ssDNA was retarded due to the formation of the equilibrium complex with PAAm-b-ODN. Increasing Mn of the PAAm segment enhanced this retardation. On the other hand, normal ssDNA was unable to form the complex owing to a single-base mismatch, which was proved by melting curve measurements. The Lineweaver-Burk-type analysis of the mobility of mutant ssDNA revealed that the binding constants for the complexes with different PAAm-b-ODN probes were almost identical to each other. The analysis also demonstrated that the ratio of the hydrodynamic radius of the complex to that of the free mutant ssDNA increased with increasing Mn of the affinity polymeric probe’s PAAm segment. This means that the PAAm segment indirectly provides mutant ssDNA with an additional hydrodynamic friction force via the affinity interaction of the ODN segment. Optimization of the salt concentration of the running buffer and the capillary temperature improved the resolution of the separation. This affinity polymeric probe will be useful for developing a simple and highly reliable single-nucleotide polymorphism genotyping method.

Introduction Bioconjugate materials between DNA and synthetic polymer with well-controlled chain length, composition, and morphology are receiving considerable attention by virtue of the unique synergistic characteristics of base complementarities of DNA and the diverse physicochemical properties of synthetic polymers.1 In particular, DNA (or RNA) block copolymers, where the end of the synthetic polymer is covalently linked to the 5′or 3′-terminal of the oligonucleotide, have been exploited in the various research fields. They have included the controlled delivery of antisense DNA,2-4 siRNA,5-7 and drugs8 into a living cell, as well as the fabrication of nanometer-sized materials, such as multiblock copolymers,9-11 micelles,12-18 rods,14,15 and vesicles,18 through self-assembly of DNA block copolymers. Furthermore, Barron’s group19-27 and our group28-30 have reported that DNA block copolymers can serve as powerful tools in a single-nucleotide polymorphism (SNP) genotyping based on capillary electrophoresis. SNPs are genetic changes resulting from single-nucleotide substitution, which occurs in the human genome once every kilobase. SNP analysis is expected to enable us to predict the * To whom correspondence should be addressed. Phone: +81-48-4675489. Fax: +81-48-462-4658. E-mail: [email protected]. † Present address. Graduate School of Pure and Applied Science, University of Tsukuba, 1-1-1 Ten-noudai, Tsukuba, Ibaraki 305-8573, Japan.

genetic risk of a certain disease, diagnose a disease more accurately, and predict a therapeutic response to a drug.31,32 Therefore, the development of analytical methods for discriminating genetic sequences containing SNPs is of great importance in gene diagnostics. Although capillary electrophoresis-based DNA analysis has already been employed in genetic, medical, and forensic sciences,33 a more facile and general methodology for SNP discrimination is required for practical uses. Barron and co-workers have designed a DNA-polyamide blocktype conjugate consisting of a single-stranded DNA (ssDNA) and a monodispersed and nonionic polypeptoid19-21,23,24,26 or a genetically engineered repetitive polypeptide.22,24,26,27 They demonstrated the size-based electrophoretic separation of DNApolyamide block-type conjugates in molecular sieving matrixfree systems using capillary tubes19-27 or microchannels:26 endlabeled free-solution electrophoresis (ELFSE). By combining ELFSE with a single-base extension method, they succeeded in SNP genotyping.20,26,27 Our group has recently developed a novel affinity capillary electrophoresis using a well-defined DNA block copolymer as an affinity polymeric probe for sequence-specific separation of ssDNA.28-30 We achieved a baseline separation of a 60-mer ssDNA (target ssDNA) and its single-base-substituted one (nontarget ssDNA).30 Scheme 1 illustrates the separation. Because ssDNA is a uniformly charged polyelectrolyte, its mobility in free-solution electrophoresis (µD) is independent of

10.1021/bm801301b CCC: $40.75  2009 American Chemical Society Published on Web 02/27/2009

806

Biomacromolecules, Vol. 10, No. 4, 2009

Kanayama et al.

Scheme 1. Schematic Diagram Illustrating Affinity Capillary Electrophoresis of Target and Nontarget ssDNA Using an ODN-Synthetic Polymer Diblock-Type Conjugate as an Affinity Probea

a The difference in base sequence between target and nontarget ssDNA is only the single base located near the midpoint of each sequence (see Scheme 3).

Scheme 2. Synthetic Scheme of PAAm-b-ODN

Scheme 3. Sequences of Sample ssDNA (Normal and Mutant ssDNA) and the ODN Segment of PAAm-b-ODN Used in this Study

the sequence and the length. Hence, the observed electrophoretic mobility of target ssDNA is identical to that of nontarget ssDNA. To separate a mixture of target and nontarget ssDNA, we used a poly(ethylene glycol)-oligodeoxyribonucleotide block copolymer (PEG-b-ODN) as an affinity polymeric probe in earlier studies.28-30 The base sequence of the ODN segment (7-10 bases) is designed to be complementary to part of the target ssDNA, including a single-base substitution site. Therefore, the target ssDNA forms an equilibrium fully matched duplex with PEG-b-ODN during electrophoresis. Because the long PEG segment of the complex receives a large hydrodynamic friction force, the mobility of the complex (µC) is much smaller than that of the free target ssDNA (µD). On the other hand, nontarget ssDNA cannot form the complex with PEG-b-ODN due to a single-base mismatch and migrates faster without binding to the affinity polymeric probe. Consequently, a mixture of target and nontarget ssDNA is separated, and each ssDNA is detected at the anode end by a detector; the electropherogram shows two distinct peaks. It should be noted that the peak area ratio allows

us to readily and precisely estimate the SNP allele frequency. This is one of the principal advantages over other SNP genotyping methods. The affinity electrophoretic method is sensitive enough to detect and quantify approximately 1 mol % target ssDNA in the sample.30 Synthetic polymers for the affinity probe segment are required to exhibit three properties. First, the synthetic polymer segment needs to be electrically neutral. If it is a polyelectrolyte, the resulting block copolymer will migrate in the capillary tube upon application of a voltage to entail its time-dependent concentration change. A nonconstant concentration of the affinity probe during electrophoresis is disadvantageous for analyzing the mobility of target ssDNA on the basis of the equilibrium complex formation with the affinity probe. Second, the synthetic polymer segment should have no interaction with DNA. Strong nonspecific interaction between the synthetic polymer segment and ssDNA may conceal the highly sequence-specific interaction of the ODN segment to target ssDNA. Third, a narrow molecular weight distribution of the polymer segment is preferred,

DNA Block Copolymers for Affinity Electrophoresis

otherwise the complex between the affinity polymeric probe and target ssDNA would receive a nonuniform hydrodynamic friction force during electrophoresis, and the resulting distribution of the complex’s mobility could lead to the broadening of the target ssDNA peak. We achieved the electrophoretic separation of target and nontarget ssDNA by using PEG-b-ODN as the affinity polymeric probe because PEG with a narrow molecular weight distribution met all of the requirements mentioned above.28-30 However, other synthetic polymers have not yet been examined. A similar sequence-specific separation of ssDNA with a different affinity polymeric probe would strongly support the separation scheme. We describe here the application of polyacrylamide (PAAm) generated by reversible addition-fragmentation chain transfer (RAFT) polymerization to the synthetic polymer segment of the affinity probe. RAFT polymerization is a powerful method for preparing a wide variety of vinyl polymers with controlled molecular weights and low polydispersities.34 PAAm has no electric charge and no interaction with DNA. Therefore, the RAFT-generated PAAm satisfies the requirements. Moreover, the RAFT-generated polymers possess a great advantage in sitespecific conjugation with other molecules. Because a dithiobenzoate- or trithiocarbonate-based chain transfer agent (CTA) is used in RAFT polymerization, the generated polymers have the dithiobenzoate or trithiocarbonate group at the ω-terminal. These polymers are easily transformed into the thiol-end functionalized polymers,34-36 which can be directly coupled with unsaturated esters (e.g., maleimides and acrylates), R-haloacetamides, or thiols under mild conditions. In this study we employed the Michael addition reaction of the thiol-terminated PAAm to 5′maleimide-modified ODN to obtain a PAAm-ODN block copolymer (PAAm-b-ODN). The present study not only demonstrates the validity of our separation scheme but also may extend the scope of application of RAFT-generated polymers to analytical research fields. To our knowledge, there is only one example, that of He and co-workers in which the RAFT polymerization technique was applied to analytical science.37 The polymerization step was utilized as the signal amplification for the detection of analyte in their pioneering work, which thus fundamentally differs from this study.

Experimental Section Materials. All reagents were purchased from Wako Pure Chemical Industries (Osaka, Japan) unless otherwise noted. 5′-Aminohexyl ODN and 5′-fluorescein isothiocyanate (FITC)-labeled ssDNA were purchased from Tsukuba Oligo Service (Ibaraki, Japan). The DNA concentration was determined by measuring the absorbance at 260 nm. N-(4Maleimidobutyloxy)sulfosuccinimide sodium salt (Sulfo-GMBS) was obtained from Dojindo Laboratories (Kumamoto, Japan). Tris(2carboxyethyl)phosphine hydrochloride (TCEP) was purchased from Pierce (Rockford, IL). 4-Cyanopentanoic acid dithiobenzoate was synthesized according to a reported method.38 Deionized water (>18.1 MΩ), purified with a Milli-Q instrument (Millipore, Billerica, MA), was used for all of the experiments. Synthesis of 5′-Maleimide-Modified ODN. To a solution of 5′aminohexyl ODN (480 µmol) in 100 mM sodium carbonate-sodium bicarbonate buffer (pH 9.0, 500 µL) was added a solution of N-(4maleimidobutyloxy)sulfosuccinimide sodium salt (1 mg) in deionized water (100 µL). The obtained mixture was allowed to react at ambient temperature for 2 h. The reaction was followed by reversed phase highperformance liquid chromatography (RP-HPLC). The measurement system consisted of the following components from Shimadzu Scientific Instruments (Kyoto, Japan): an LC10AT HPLC pump, a CTO-10AS column oven, an SPD-10A UV-vis detector, and an RID-10A refraction index detector controlled by LC solution software. For an

Biomacromolecules, Vol. 10, No. 4, 2009

807

Table 1. Characterization of PAAm-SH code PAAm(3K)-SH PAAm(6K)-SH PAAm(10K)-SH PAAm(16K)-SH

feed molar ratio [M]/[CTA]/[I]a yield (%) Mnb (g mol-1) MWDb 350/5/1 700/5/1 1400/5/1 1750/5/1

14.4 30.9 41.8 62.8

3500 5700 9800 16100

1.05 1.08 1.12 1.17

a Concentration ratio of the monomer [M], the chain transfer agent [CTA], and the initiator [I]. b Determined by ASEC in 10 mM Tris-HCl buffer (pH 7.4) containing 0.2 M NaCl with the calibration using PEG standards.

analytical column, an Inertsil ODS-3 column (radius of silica-gel particle: 5 µm, 4.6 × 250 mm) from GL Science (Tokyo, Japan) was used. The linear gradient (30 min) was 0-30% B in A (A: 0.1 M TEAacetate buffer (pH 7.0), B: acetonitrile) at a flow rate of 1 mL/min at room temperature. The wavelength used for the detection was 260 nm. The product was purified using the RP-HPLC system under similar conditions. Appropriate fractions were collected and dialyzed against deionized water (Mw cutoff, 1000). The final solution was lyophilized to obtain 5′-maleimide-modified ODN (ODN-MAL). The yield was determined to be 83% by measuring the absorbance at 260 nm. Mass Spectrometry. MALDI-TOF mass spectrometric analysis of ODN-MAL was conducted by using a Bruker REFLEX mass spectrometer (Bruker Daltonics, Billerica, MA) in the negative-ion mode. The sample was prepared by mixing 0.2 µL of an aqueous solution of the purified ODN-MAL and 0.8 µL of the matrix consisting of saturated 3-hydropicolinic acid, 0.1 M ammonium citrate, and acetonitrile (1:1: 1, v/v/v). RAFT Polymerization of Acrylamide Monomer. Dithiobenzoateterminated PAAm (PAAm-DTB) was prepared via RAFT polymerization of acrylamide monomer by employing 4,4′-azobis(4-cyanopentanoic acid; V-501) as the primary source of radicals and 4-cyanopentanoic acid dithiobenzoate as the chain transfer agent (CTA). The acrylamide monomer, V-501, and CTA were dissolved in 10 mL of dimethyl sulfoxide (DMSO) and deoxidized by Ar gas bubbling for 20 min. The polymerization was conducted at 70 °C with a [CTA]0/[V-501]0 ratio of 5 under an Ar atmosphere. The resulting polymer was then purified by repeated precipitation into acetone, followed by lyophilization from deionized water. Synthesis of PAAm-b-ODN. The terminal dithiobenzoate group of PAAm-DTB was transformed to the thiol group by adding a freshly prepared aqueous solution of NaBH4 (0.1 M, 2 mL) to an aqueous solution (10 mL) of PAAm-DTB (100 mg). The mixture was allowed to react overnight at room temperature with stirring and then dialyzed against deionized water (Mw cutoff, 1000). The final solution was lyophilized to obtain a series of thiol-terminated PAAm: PAAm(3K)SH, PAAm(6K)-SH, PAAm(10K)-SH, and PAAm(16K)-SH (Table 1). The number in parentheses denotes the number-averaged molecular weight (Mn) of PAAm determined experimentally. The conjugation of PAAm-SH and ODN-MAL was performed as follows. PAAm(3K)SH (3.3 mg) was dissolved in 100 mM Tris-HCl buffer (pH 7.4, 500 µL) and then deoxidized by Ar gas bubbling for 15 min. To the solution of PAAm(3K)-SH was added a solution of TCEP in 100 mM TrisHCl buffer (pH 7.4, 10 mg/mL, 82 µL). After 15 min, the same buffer solution of ODN-MAL (1 mM, 100 µL) was added and allowed to react overnight at 4 °C. The reaction mixture was loaded on a preparative anion exchange column (Q-Sepharose Fast Flow, GE Healthcare, Buckinghamshire, U.K.) and then on a preparative gel filtration column (Sephadex G-50, GE Healthcare). The final solution was dialyzed against deionized water (Mw cutoff, 1000) and then lyophilized to obtain PAAm(3K)-b-ODN. The yield was determined to be 74% by measuring the absorbance at 260 nm. Other PAAm-bODN copolymers were obtained by the same method. Fourier Transform Infrared Spectroscopy. Fourier transform infrared (FT-IR) spectroscopic measurements of PAAm(3K)-DTB and PAAm(3K)-SH were performed with a TravelIR FT-IR spectrophotometer (SensIR Technologies, Danbury, CT). Spectra were collected

808

Biomacromolecules, Vol. 10, No. 4, 2009

Kanayama et al.

in the diamond ATR mode with an unpolarized beam at a resolution of 1 cm-1 with 256 scans. Aqueous Size Exclusion Chromatography. The Mn and molecular weight distribution (MWD) of PAAm-SH and PAAm-b-ODN were determined by aqueous size exclusion chromatography (ASEC). The measurements were performed on the HPLC system described above. For an analytical column, an OHpaq SB-804HQ column (8.0 × 300 mm) from Shodex (Tokyo, Japan) was used. For a mobile phase, 10 mM Tris-HCl buffer (pH 7.4) containing 0.2 M NaCl was used at a flow rate of 0.5 mL/min at 40 °C. Calibration was conducted using PEG standards purchased from Polymer Laboratories (Shropshire, U.K.). Affinity Capillary Electrophoresis. All capillary electrophoresis was performed using a P/ACE MDQ capillary electrophoresis system (Beckman-Coulter, Fullerton, CA) equipped with a laser-induced fluorescent detector (excitation, 488 nm; emission, 520 nm). A CEPcoated capillary tube (internal diameter, 75 µm; total length, 49 cm; effective length, 39 cm), whose inner surface was precoated with polymer, was purchased from Agilent Technologies (Wilmington, DE). This capillary tube was employed to suppress the electroosmotic flow through the electrophoretic process. For a running buffer, 50 mM Trisborate buffer (pH 7.5) containing MgCl2 (0-0.5 mM) was used. The temperature of the capillary tube was held constant (25-40 °C) with a recirculating liquid coolant system. Prior to electrophoresis, the running buffer solution of PAAm-bODN (5 µM) was injected into the capillary tube from the cathode end by positive pressure (20 psi for 30 s). Next, the running buffer solution of 5′-FITC-labeled sample ssDNA (50 nM) was introduced into the capillary tube using a similar method (0.5 psi for 10 s). Electrophoresis was conducted with a constant voltage of -15 kV for 20 min. The sample ssDNA was detected with a laser-induced fluorescence detector at the anode end of the capillary tube. Between runs, the capillary tube was sufficiently washed with deionized water (20 psi for 2 min) and then substituted with the running buffer (20 psi for 1 min). The electrophoretic mobility (µ) of the sample ssDNA was calculated by

µ)

lL l ) tE tV

(1)

where l is the effective length of the capillary tube, t is the migration time, E is the electric field, L is the total length of the capillary tube, and V is the applied voltage. Each CE experiment was performed in triplicate to confirm reproducibility. Melting Temperature Measurement. The melting curve of doublestranded DNA (dsDNA) between PAAm-b-ODN (3 µM) and sample ssDNA (3 µM) in 50 mM Tris-borate buffer (pH 7.5) containing 0.5 mM MgCl2 was obtained by measuring the change of absorbance at 260 nm as a function of temperature with a UV-2550 spectrophotometer equipped with a TMSPC-8 temperature controller unit (Shimadzu). The heating and cooling ramp was 1 °C/min. The sample-holding chamber was kept flushed with dry N2 gas during the measurements. The melting temperature (Tm) was determined as an average of the maximum values in the first derivative of the melting curves obtained from the heating and cooling processes. Each melting experiment was performed in triplicate to confirm reproducibility, and the obtained Tm values agreed to less than (1 °C from the mean.

Results and Discussion Synthesis of Affinity Polymeric Probe. The synthetic procedure of PAAm-b-ODN is shown in Scheme 2. First, we obtained ODN-MAL by coupling N-(4-maleimidobutyloxy)sulfosuccinimide and 5′-aminohexyl ODN (5′-GCAGCCC-3′). After purification by RP-HPLC (Figure S1), we characterized the product by MALDI-TOF mass spectrometry (Mfound ) 2408.9 g mol-1, Mtheory ) 2409.2 g mol-1). Second, we

Figure 1. FT-IR spectra of (a) PAAm(3K)-DTB and (b) PAAm(3K)SH.

Figure 2. ASEC traces of PAAm-b-ODN with different PAAm lengths obtained by using a UV-vis detector (detection wavelength: 260 nm). For a mobile phase, 10 mM Tris-HCl buffer (pH 7.4) containing 0.2 M NaCl was used at a flow rate of 0.5 mL/min at 40 °C.

synthesized PAAm-DTB by RAFT polymerization of acrylamide monomer using V-501 as the free radical initiator and 4-cyanopentanoic acid dithiobenzoate as the chain transfer agent (CTA). Varying the feed molar ratio of the monomer, the initiator, and CTA, we obtained PAAm-DTB with a different polymerization degree. Third, we hydrolyzed the ω-terminal dithiobenzoate group of PAAm-DTB with NaBH4 to yield PAAm-SH.34,36 Upon the addition of NaBH4 to the pale red solution of PAAm-DTB, the solution became colorless, indicating that the reaction proceeded well. The elimination of the thiobenzoyl moiety was further confirmed by FT-IR spectroscopy (Figure 1). We observed the disappearance of a peak at 1043 cm-1 (a stretching vibration of the CdS bond), as well as peaks at 872 and 760 cm-1 (an out-of-plane bending vibration of the aromatic C-H bond). Table 1 shows the feed molar ratio of the monomer, CTA, and the initiator in the RAFT polymerization and the total yield of PAAm-SH, along with Mn and MWD of PAAm-SH, as determined by ASEC measurements. Finally, we coupled ODN-MAL with PAAm-SH in the presence of TCEP to obtain PAAm-b-ODN, which was followed by the ASEC system equipped with a UV detector. We employed TCEP as a reductant to prevent the undesired dimerization of PAAm-SH through the disulfide bond formation to give PAAmS-S-PAAm (Figure S2),35 which was inert to the Michael addition reaction. We removed residual PAAm-SH and ODNMAL by anion exchange chromatography and gel filtration chromatography, respectively. As shown in Figure 2, we observed a unimodal peak in the ASEC trace of each PAAmb-ODN. We determined Mn of PAAm-b-ODN using ASEC (Table 2). Furthermore, we confirmed that MWD of PAAm-b-

DNA Block Copolymers for Affinity Electrophoresis

Biomacromolecules, Vol. 10, No. 4, 2009

Table 2. Characterization of PAAm-b-ODN code

Ma (g mol-1)

Mnb (g mol-1)

MWDb

PAAm(3K)-b-ODN PAAm(6K)-b-ODN PAAm(10K)-b-ODN PAAm(16K)-b-ODN

5900 8100 12200 18500

5400 7700 11900 18700

1.04 1.09 1.14 1.19

a The calculated value obtained by adding the Mn value of PAAm (Table 1) to the Mfound value of ODN-MAL determined by MALDI-TOF mass spectrometry. b Determined by ASEC in 10 mM Tris-HCl buffer (pH 7.4) containing 0.2 M NaCl with the calibration using PEG standards.

ODN was as low as that of PAAm-SH (Tables 1 and 2). These results indicate that well-defined DNA-PAAm diblock copolymers whose PAAm segments were of different molecular weights, namely, PAAm(3K)-b-ODN, PAAm(6K)-b-ODN, PAAm(10K)-b-ODN, and PAAm(16K)-b-ODN, were successfully prepared by this procedure. Affinity Capillary Electrophoresis of ssDNA. For an analytical sample, we used a mixture of an equal amount of a chemically synthesized 5′-FITC-labeled 60-mer ssDNA (normal ssDNA) and its single-base-substituted one (mutant ssDNA). Their sequences are shown in Scheme 3, which correspond to part of the mitochondrial cytochrome b gene of cucumber powdery mildew and its single-base mutant at codon 143 (GGT to GCT). This mutation induces strobilurin-related fungicide resistance in the fungal pathogen.39 The base sequence of PAAm-b-ODN is designed to be complementary to part of the mutant ssDNA including a single-base substitution site of codon 143 (Scheme 3). Prior to electrophoresis, a capillary tube was filled out with a solution of PAAm-b-ODN at a concentration of 5 µM in 50 mM Tris-borate buffer (pH 7.5) containing MgCl2. The same buffer solution of normal and mutant ssDNA at each concentration of 50 nM was then injected into the capillary tube and the voltage was applied. Figure 3 shows the electropherograms obtained with the PAAm-b-ODN probes at 25-35 °C. We detected two peaks in all of the electropherograms at 25 °C; the former peak was identified as that of normal ssDNA and the latter as that of mutant ssDNA in the discrete runs (data not shown). These results demonstrate that PAAm-b-ODN works as an affinity polymeric probe in capillary electrophoresis. Interestingly, the peak of mutant ssDNA was gradually retarded as the molecular weight of the PAAm segment increased, whereas that of normal ssDNA remained almost unchanged. When the capillary temperature was increased from 25 to 35 °C, this sequence-specific retardation was gradually weakened. We account for this temperature-dependence by using the melting temperature (Tm) of the duplex between PAAm-b-ODN and sample ssDNA under the identical buffer condition. The obtained Tm values are summarized in Table 3. All of the Tm values of single-base mismatched duplexes between PAAm-b-ODN and normal ssDNA were determined to be below 5 °C. However, all of the Tm values of fully matched duplex between PAAm-b-ODN and mutant ssDNA were above 30 °C. These results strongly suggest that the PAAm-b-ODN probe forms an equilibrium complex with mutant ssDNA but has no interaction with normal ssDNA during electrophoresis at 25-35 °C. In addition, because Tm of the fully matched duplex between PAAm-b-ODN and mutant ssDNA was virtually independent of the PAAm length, the PAAm-b-ODN probes whose PAAm segments were of different molecular weights from each other were supposed to exhibit almost the same degree of affinity to mutant ssDNA. To demonstrate this clearly, we next calculated the binding constant of the complex using the observed electrophoretic mobility.

809

Analysis of Electrophoretic Mobility. Because normal ssDNA migrates without forming the complex with PAAm-bODN, the electrophoretic mobility is described as that of free ssDNA. On the other hand, mutant ssDNA forms an equilibrium complex with PAAm-b-ODN during electrophoresis, as follows K

mutant ssDNA + PAAm-b-ODN {\} complex

(2)

where K is the binding constant of the complex. We consider that this equilibrium holds throughout the electrophoresis because we determined the electrophoretic mobility of PAAmb-ODN as 2.41 × 10-4 cm2/V s for PAAm(3K)-b-ODN, 1.54 × 10-4 cm2/V s for PAAm(6K)-b-ODN, 9.73 × 10-5 cm2/V s for PAAm(10k)-b-ODN, and 6.33 × 10-5 cm2/V s for PAAm(16k)b-ODN at 25 °C in the absence of MgCl2, which was significantly lower than that of free ssDNA (3.52 × 10-4 cm2/V s) under the identical conditions. This means that mutant ssDNA migrating rapidly toward the anode encounters the affinity polymeric probes migrating slowly to the same direction in the capillary tube until reaching the detection window at the anode. The electrophoretic mobility of normal and mutant ssDNA, µN and µM, is formulated as follows29

µN ) µD µM )

(3)

( 1 + 1K[P] )µ + ( 1 +K[P]K[P] )µ D

C

(4)

where µD and µC are the electrophoretic mobility of free ssDNA and the complex with PAAm-b-ODN, respectively, and [P] is the equilibrium concentration of the affinity polymeric probe. To determine the binding constant of the complex by adding the affinity polymeric probe to the running buffer and measuring the change in the electrophoretic mobility of mutant ssDNA, eq 4 is transformed into the equation

1 1 1 1 ) + µM - µD (µC - µD)K [P] µC - µD

(5)

The double reciprocal plot (1/(µM - µD) versus 1/[P]), which is known as the Lineweaver-Burk plot in enzyme studies, gives the binding constant (K) and the electrophoretic mobility of the complex (µC). Because the affinity probe (5 µM) was introduced into the capillary tube in large excess over mutant ssDNA (50 nM), [P] can be replaced by the initial concentration of the affinity polymeric probe ([P]0) in the present analysis. Figure 4 depicts representative electropherograms showing the dependence of migration time on [P]0 and the resulting double reciprocal plot. All of the results obtained with the other PAAmb-ODN probes are also shown in Figure S4, Supporting Information. The K and µC values calculated from the double reciprocal plots are summarized in Tables 4 and 5, respectively. In Table 4, we found little differences in the binding constant of the complex between a different PAAm-b-ODN probe and mutant ssDNA at 25-35 °C. This agrees with the fact that we acquired almost the same Tm value of the complex between a different PAAm-b-ODN probe and mutant ssDNA (Table 3). Hence, the nonspecific interaction between the PAAm segment and the mutant ssDNA is negligible. It was also found that the binding constant of the complex between PAAm-b-ODN and

810

Biomacromolecules, Vol. 10, No. 4, 2009

Kanayama et al.

Figure 3. Electropherograms showing the separation of normal and mutant ssDNA when (a) PAAm(3K)-b-ODN, (b) PAAm(6K)-b-ODN, (c) PAAm(10K)-b-ODN, or (d) PAAm(16K)-b-ODN was used as an affinity polymeric probe. Conditions: running buffer, 50 mM Tris-borate (pH 7.5) containing 0.5 mM MgCl2; [PAAm-b-ODN] ) 5 µM; [normal ssDNA] ) 50 nM; [mutant ssDNA] ) 50 nM; capillary temperature, 25, 30, and 35 °C; applied voltage, -15 kV. Table 3. Melting Temperature of the Duplex between PAAm-b-ODN and Sample ssDNA code

T

PAAm(3K)-b-ODN/normal ssDNA PAAm(3K)-b-ODN/mutant ssDNA PAAm(6K)-b-ODN/normal ssDNA PAAm(6K)-b-ODN/mutant ssDNA PAAm(10K)-b-ODN/normal ssDNA PAAm(10K)-b-ODN/mutant ssDNA PAAm(16K)-b-ODN/normal ssDNA PAAm(16K)-b-ODN/mutant ssDNA

a m

(°C)