Detection of DNA Point Mutation by Atom Transfer Radical

Anal. Chem. 2005, 77, 4698-4705

Detection of DNA Point Mutation by Atom Transfer Radical Polymerization Xinhui Lou, Matthew S. Lewis, Christopher B. Gorman, and Lin He*

Department of Chemistry, North Carolina State University, Raleigh, North Carolina 27695

We report here a new DNA detection method in which polymer growth in atom transfer radical polymerization (ATRP) is used as a means to amplify detection signals. In this method, DNA hybridization and ligation reactions led to the attachment of ATRP initiators on a solid surface where specific DNA sequences were located. These initiators subsequently triggered the growth of poly(hydroxyethyl methacrylate) (PHEMA) at the end of immobilized DNA molecules and formed polymer brushes. The formation of PHEMA altered substrate opacity, rendering the corresponding spots readily distinguishable to the naked eye. A second ATRP reaction to form branched polymers on the surface drastically improved the visibility of DNA hybridization and significantly shortened the detection time. The resulting polymer film was characterized using infrared spectroscopy, ellipsometry, contact angle measurements, and atomic force microscopy. Direct visualization of 1 fmol of target DNA molecules of interest was demonstrated. A proof-of-principle experiment to detect DNA point mutation was conducted. The perfectly matched DNA targets were distinctively differentiated from those with mutations. The demonstrated capability to detect DNA mutation with direct visualization laid the groundwork for the future development of detector-free testing kits in single-nucleotide polymorphism screenings.

Single-nucleotide polymorphisms (SNPs) are the DNA sequence variations that occur when a single nucleotide (A, T, C, or G) in the genome sequence is changed, reportedly at a rate of approximately once every 100-300 bases.1,2 Many believe that these genetic variations are directly related to patients’ individual responses to diseases, drugs, and other environmental factors. Multiple SNPs have been linked to Alzheimer’s, Parkinson’s, diabetes, and various cancerous diseases.3-8 Development of diagnostic techniques to enable SNP detection in a point-of-care * To whom correspondence should be addressed. E-mail: lin_he@ ncsu.edu. Fax: 919-515-8920. (1) Kwok, P.-Y.; Chen, X. Curr. Issues Mol. Biol. 2003, 5, 43-60. (2) Wang, W. Y. S.; Barratt, B. J.; Clayton, D. G.; Todd, J. A. Nat. Rev. Genet. 2005, 6, 109-118. (3) Kehoe, P. G.; Katzov, H.; Feuk, L.; Bennet, A. M.; Johansson, B.; Wiman, B.; de Faire, U.; Cairns, N. J.; Wilcock, G. K.; Brookes, A. J.; Blennow, K.; Prince, J. A. Hum. Mol. Genet. 2003, 12, 859-867. (4) Martin, E. R.; Lai, E. H.; Gilbert, J. R.; Rogala, A. R.; Afshari, A. J.; Riley, J.; Finch, K. L.; Stevens, J. F.; Livak, K. J.; Slotterbeck, B. D.; Slifer, S. H.; Warren, L. L.; Conneally, P. M.; Schmechel, D. E.; Purvis, I.; Pericak-Vance, M. A.; Roses, A. D.; Vance, J. M. Am. J. Hum. Genet. 2000, 67, 383-394.

4698 Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

fashion is therefore of paramount importance for early disease diagnosis and treatment. Numerous approaches have been exploited to improve SNP detection sensitivity and specificity. The first set of existing methods is based on the use of sequence-specific enzymatic reactions, such as cleavage, ligation, and replication, to recognize allele-specific sequences and amplify the recognition events. A prominent example is the use of rolling circle amplification (RCA) in which a short DNA primer is attached at the end of a detection probe.9,10 Upon DNA hybridization that brings the detection probe close to the surface, this primer is replicated to generate an excess amount of DNA repeating units for detection. Coupling with an invader technique, 1000 DNA molecules in the biological matrix have been successfully detected.11,12 Another example is taking advantage of enzyme-specific RNA digestion upon the formation of RNA-DNA heteroduplexes.13,14 Specifically, surface-immobilized RNA probes are efficiently removed by enzyme RNase H after DNA hybridization that forms RNA-DNA duplexes. The then-released DNA binds to the next available RNA probe until complete consumption of particular RNA probes on the surface. A detection limit of 1 fM has been achieved, corresponding to a remarkable 106 enhancement to conventional detection sensitivity. The other set of approaches to improve DNA detection employs highly responsive detection tags to amplify transducer signals.15-18 Nie and co-workers have used quantum dots to boost (5) Martin, E. R.; Scott, W. K.; Nance, M. A.; Watts, R. L.; Hubble, J. P.; Koller, W. C.; Lyons, K.; Pahwa, R.; Stern, M. B.; Colcher, A.; Hiner, B. C.; Jankovic, J.; Ondo, W. G.; Allen, F. H.; Goetz, C. G.; Small, G. W.; Masterman, D.; Mastaglia, F.; Laing, N. G.; Stajich, J. M.; Ribble, R. C.; Booze, M. W.; Rogala, A.; Hauser, M. A.; Zhang, F.; Gibson, R. A.; Middleton, L. T.; Roses, A. D.; Haines, J. L.; Scott, B. L.; Pericak-Vance, M. A.; Vance, J. M. J. Am. Med. Assoc. 2001, 286, 2245-2250. (6) Noble, J. A.; White, A. M.; Lazzeroni, L. C.; Valdes, A. M.; Mirel, D. B.; Reynolds, R.; Grupe, A.; Aud, D.; Pletz, G.; Erlich, H. A. Diabetes 2003, 52, 1579-1582. (7) Ren, Z.; Cai, Q.; Shu, X.-O.; Cai, H.; Cheng, J.-R.; Wen, W.-Q.; Gao, Y.-T.; Zheng, W. Cancer 2004, 101, 251-257. (8) Listgarten, J.; Damaraju, S.; Poulin, B.; Cook, L.; Dufour, J.; Driga, A.; Mackey, J.; Wishart, D.; Greiner, R.; Zanke, B. Clin. Cancer Res. 2004, 10, 2725-2737. (9) Qi, X.; Bakht, S.; Devos, K. M.; Gale, M. D.; Osbourn, A. Nucleic Acids Res. 2001, 29, e116/1-e116/7. (10) Hatch, A.; Sano, T.; Misasi, J.; Smith, C. L. Genet. Anal. 1999, 15, 35-40. (11) Lyamichev, V.; Neri, B. Methods Mol. Biol. 2003, 212, 229-240. (12) Chen, Y.; Shortreed, M. R.; Olivier, M.; Smith, L. M. Anal. Chem. 2005, 77, 2400-2405. (13) Goodrich, T. T.; Lee, H. J.; Robert, C. M. Anal. Chem. 2004, 76, 61736178. (14) Goodrich, T. T.; Lee, H. J.; Corn, R. M. J. Am. Chem. Soc. 2004, 126, 40864087. (15) Han, M.; Gao, X.; Su, J. Z.; Nie, S. Nat. Biotechnol. 2001, 19, 631-635. 10.1021/ac050706h CCC: $30.25

© 2005 American Chemical Society Published on Web 07/07/2005

DNA detection sensitivity by providing high quantum yield, improving photostability, and reducing spectral overlaps.15 Mirkin and co-workers have described a colorimetric polynucleotide detection method based on the formation of gold nanoparticle networks.19 Detection sensitivity has been further improved in the consecutive reports by the same group with a point mutation selectivity of ∼100,000:1 achieved.17 While most methodologies to date have significantly improved detection sensitivity and specificity in SNP genotyping, the reported protocols are yet to be amenable for clinical settings. The main obstacle lies in the requirements for special laboratory skills in detection label preparation and for the complex instruments in signal readout. A new approach that enables sensitive detection of SNPs with off-shelf chemistry in an instrument-free detection fashion is desirable for the development of any portable screening devices. Polymers have long been used in biological applications as blocking reagents to reduce nonspecific adsorption and scaffolding frames to support immobilization of biomolecules. There are a few reports in which polymers of unique electrical or optical properties have been used in a preformed macromolecular form for DNA detection by electrostatically attracting cationic polymers to highly negatively charged DNA or by chemical coupling to attach DNA on the polymer side chains.20-23 The dynamic growing process of macromolecules has yet to be exploited as an amplification tool. We note that polymerization is essentially a highly efficient signal amplification process with an enhancement power controllable at will. Similar to RCA but using chemical reaction in the place of enzyme-based replication, connecting the same small monomers in a head-to-tail fashion forms a long polymer chain that contains hundreds to millions of repeating units. Consequently, the detection signal kept in one small monomer is amplified hundreds to millions of times through the chain propagation. Atom transfer radical polymerization (ATRP) is a new class of controlled/“living” radical polymerization that was first reported by the groups of Sawamoto24,25 and Matyjaszewski.26,27 Compared with other living polymerization techniques, ATRP is based on the repetitive addition of monomers to radicals that are generated from dormant alkyl halides in a reversible redox process.28 It has been a popular means to graft polymer brushes on a solid support because of the broad selection of monomers, good control over the product molecular weight and dispersity, and high tolerance (16) Wang, J.; Liu, G.; Merkoci, A. J. Am. Chem. Soc. 2003, 125, 3214-3215. (17) Nam, J.-M.; Stoeva, S. I.; Mirkin, C. A. J. Am. Chem. Soc. 2004, 126, 59325933. (18) Cao, Y. C.; Jin, R.; Mirkin, C. A. Science 2002, 297, 1536-1540. (19) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607-609. (20) Liu, B.; Bazan, G. C. Proc. Natl. Acad. Sci. U.S.A. 2005, 102, 589-593. (21) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2003, 125, 896-900. (22) Gibbs, J. M.; Park, S.-J.; Anderson, D. R.; Watson, K. J.; Mirkin, C. A.; Nguyen, S. T. J. Am. Chem. Soc. 2005, 127, 1170-1178. (23) Kim, W. J.; Sato, Y.; Akaike, T.; Maruyama, A. Nat. Mater. 2003, 2, 815820. (24) Sawamoto, M.; Kamigaito, M. Trends Polym. Sci. 1996, 4, 371-377. (25) Sawamoto, M.; Kamigaito, M. J. Macromol. Sci. 1997, A34, 1803-1814. (26) Matyjaszewski, K.; Patten, T. E.; Xia, J. J. Am. Chem. Soc. 1997, 119, 674680. (27) Wang, J.-S.; Matyjaszewski, K. Macromolecules 1995, 28, 7572-7573. (28) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921-2990.

Scheme 1. Schematic Drawing of ATRP-Assisted DNA Detection

of surrounding functional groups.28-31 The demonstration of surface-initiated ATRP in aqueous media at room temperature further renders it feasible to be used as a signal amplification method in biosensing.32-35 In this report, we describe a simple polymerization-facilitated detection strategy to amplify DNA hybridization events using an ATRP reaction. Taking advantage of this monomer-replicating concept, the detection of specific DNA sequences is accomplished by the formation of polymer brushes on the surface at ambient temperature (Scheme 1). Because the formed polymers change the refractive index near the surface and subsequently affect surface opacity, the spots where DNA hybridization occurred are directly visible. In this report, we also examined a two-stage polymerization strategy that further improved the visibility of DNA detection by forming branched polymers on the surface. To our knowledge, this is the first time the dynamic molecular growth has been used as an amplification means in bioanalytical detection. The demonstrated direct visualization of DNA hybridization also marks a promising step toward the development of SNP home testing kits. EXPERIMENTAL SECTION Materials. Au substrates (50-Å chrome followed by 1000-Å gold on float glass) were purchased from Evaporated Metal Films (Ithaca, NY). Dithiothreitol (DTT), triethylamine (TEA), mercaptohexanol (MCH), bromoisobutyryl bromide, N-hydroxysuccinimide (NHS), 2-hydroxyethyl methacrylate (HEMA), CuCl, CuBr2, and 2,2′-bipyridyl (bpy) were purchased from Sigma-Aldrich (St. Louis, MO). All oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA). The sequences of DNA used in this report are listed in Table 1. T4 DNA ligase was purchased from Stratagene (La Jolla, CA). A NAP-5 column from Amersham Pharmacia Biotech was used for DNA purification. Capture Probe Immobilization. The gold substrates were cleaned in a piranha solution (70% H2SO4, 30% H2O2, hazard (29) Jeyaprakash, J. D.; Samuel, S.; Dhamodharan, R.; Ruhe, J. Macromol. Rapid Commun. 2002, 23, 277-281. (30) Zheng, G.; Stoever, H. D. H. Macromolecules 2002, 35, 7612-7619. (31) Chen, X.; Randall, D. P.; Perruchot, C.; Watts, J. F.; Patten, T. E.; von Werne, T.; Armes, S. P. J. Colloid Interface Sci. 2003, 257, 56-64. (32) Coullerez, G.; Carlmark, A.; Malmstroem, E.; Jonsson, M. J. Phys. Chem. A 2004, 108, 7129-7131. (33) Huang, W.; Kim, J.-B.; Bruening, M. L.; Baker, G. L. Macromolecules 2002, 35, 1175-1179. (34) Wang, X. S.; Armes, S. P. Macromolecules 2000, 33, 6640-6647. (35) Chatterjee, U.; Jewrajka, S. K.; Mandal, B. M. Polymer 2005, 46, 15751582.

Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

4699

Table 1. Summary of the DNA Sequences Used in This Report name

sequence

description

A

5′ NH2-CH2)6-A18 TCC TTA TCA ATA TT-(CH2)3-S-S-(CH2)2CH3

dual functional probe used to study the kinetics of PHEMA formation on the gold surface

A-IN

5′ Br(CH3)2CCONH-(CH2)6-A18 TCC TTA TCA ATA TT-(CH2)3-S-S-(CH2)2CH3

initiator-coupled A

C

5′-pTAA CAA TAA TCC CTC AA A18-(CH2)3-S-S-(CH2)2CH3

capture probe partially complementary to C′ portion of target C′D′

D

5′-NH2-(CH2)6-A18 AAA TCC TTA TCA ATA TT

detection probe partially complementary to D′ portion of target C′D′

D-IN

5′-Br(CH3)2CCONH-(CH2)6-A18 AAA TCC TTA TCA ATA TT

initiator-coupled D

NC

5′-pGGC AGC TCG TGG TGA AA A18-(CH2)3-S-S-(CH2)2CH3

noncomplementary capture probes to target C′D′

C′D′

5′-GAG GGA TTA TTG TTA AAT ATT GAT AAG GAT

perfectly matched target DNA complementary to probes C and D

C1′D′

5′-GAG GGA ATA TTG TTA AAT ATT GAT AAG GAT

single-mismatch target to probes C, with the mutation site shown in the boldface, underlined letter

C3′D′

5′-GAG GGA AAG TTG TTA AAT ATT GAT AAG GAT

three-mismatch target to probes C, with the mutation sites shown in the boldface, underlined letters

solution, potentially explosive, handle with care) prior to use. The oligonucleotide probes, C and NC (Table 1), had disulfide bonds at the 3′-end for surface attachment. To generate free thiol groups for surface immobilization, 100 µL of stock oligonucleotide solution (C or NC) at 100 µM was mixed with 0.01 mmol of DTT and 4 µL of TEA at room temperature. A Voyager DE STR MALDI-TOF mass spectrometer (Applied Biosystems) was used to confirm the successful reduction of disulfide bonds. After reduction, the excess amount of DTT was removed using a NAP-5 column. The concentrations of the reduced oligonucleotides were determined by the UV absorbance at 260 nm. Freshly reduced capture probes at 1 µM in a KH2PO4 buffer (1 M, pH 4.4) were spotted onto Au substrates at room temperature and incubated in a humid chamber for 16-20 h. The surfaces were then incubated with 1 mM MCH aqueous solution for 1 h, followed by copious rinsing with DI water and dried under Ar.36,37 Initiator Coupling to Detection Probe. Initiator-coupled DNA detection probe was prepared as previously described.38 Briefly, 7 nmol of oligonucleotide D solution and 10 µL of conjugation buffer (1.0 M NaHCO3/Na2CO3, pH 9.0) were added into a centrifuge tube, followed by the addition of freshly prepared bromoisobutyryl NHS ester solution at 10 mg/mL. After 30-min reaction at room temperature, unreacted NHS ester was removed by gel filtration. MALDI-TOF MS was again used to monitor the coupling efficiency by measuring the amount of oligonucleotides before and after the coupling reaction. Three-Strand DNA Hybridization. Probe C and NC-attached Au surface was incubated with 1.5 µL of target oligonucleotides, C′D′ (C1′D′, and C3′D′ for mismatch detection), at various concentrations in a humidity chamber at room temperature. The 1 M NaCl in the Tris-EDTA (TE) buffer was used as the hybridization buffer. The hybridization volume was reduced to 1.5 µL with the help of a sandwich assembly using a glass (36) Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 8916-8920. (37) Peterlinz, K. A.; Georgiadis, R. M.; Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 3401-3402. (38) Lou, X.; He, L. Polym. Prepr. 2004, 45, 455-456.


coverslip. After 1-6 h of hybridization, a stringency wash with a NaCl/phosphate buffer was performed to selectively denature the imperfect duplexes. An additional hybridization with 1 µM initiatormodified detection probes, D-IN, was conducted for 1 h in 1 M NaCl/TE buffer afterward. After hybridization, the surface was rinsed with the ligation buffer containing 50 mM Tris-HCl (pH 7.5) and 7 mM MgCl2 (in the absence of DTT to eliminate possible replacement of thiol-labeled DNA probes). DNA Ligation. After hybridization, the surface was then reacted with 20 µL of a solution containing 4 units of T4 DNA ligase and 15% PEG 8000 in a 1× T4 DNA ligation buffer without DTT. The reaction was allowed to proceed for 2 h at room temperature before rinsing with 5% SDS and subsequently with 1% SDS in 40 mM K2HPO4/KH2PO4 at 45 °C. ATRP Reaction for DNA Detection. In a typical surfaceinitiated ATRP reaction, a solution of HEMA in water was degassed for 30 min to reduce the amount of O2 present in the reaction system, followed by another 15-min degassing upon the addition of the catalyst mixture of CuCl, CuBr2, and 2,2′-bipyridyl at 1:0.3:2.9. The flask containing the DNA-immobilized substrate was also purged with Ar for 15 min. The monomer/catalyst solution was then injected into the flask where the substrate was located. The polymerization was stopped by the removal of the substrate, followed by the subsequent rinse with methanol. The polymer film formed was characterized using polarization modulation-infrared reflection-adsorption spectroscopy (PM-IRRAS), ellipsometry, and contact angle measurements. A second-stage ATRP reaction was conducted by subsequent anchoring additional initiators on PHEMA that was formed during the first-stage polymerization. Direct coupling of 2-bromoisobutyryl bromide to the hydroxyl groups on the PHEMA side chains was achieved using a solution of 2-bromoisobutyryl bromide (0.08 M) and TEA (0.1 M) for 20 min. Additional ATRP reactions were then conducted following the aforementioned protocol. To decrease the background noise from MCH, the OH groups of MCH were blocked using acetyl chloride. The block step was

completed in 20 min, and the substrate was rinsed with methanol and H2O and dried with Ar. Alternatively, methoxy-capped PEG was used in the place of MCH to eliminate the needs for the surface-blocking step. Instrumentation. A Voyager DE-STR MALDI-MS was used to monitor the coupling reaction between amino-labeled DNA and bromoisobutyryl NHS ester and the reduction of disulfide bonds on DNA. A 35 mg/mL solution of 3-hydroxypicolinic acid in 7 mg/mL diammonium citrate and 10% acetonitrile in water was used as the MALDI matrix. DNA solutions were desalted using C18ZipTip (Millipore). The PM-IRRAS spectra were recorded on a Digilab FTS 7000 (Randolph, MA) spectrometer.39 Reflectance FT-IR spectroscopy was performed using a Digilab spectrometer containing a PIKE grazing angle (70°) attachment. The spectra were typically collected with 256 scans using a MCT detector. The film thickness was measured with a VB-250 VASE ellipsometer. The instrument irradiated the substrates at a 70° incident angle. Commercial WVASE software was used for surface thickness calculation. The reflective indexes of 1.5 and 1.46 were used for the polymer films and DNA respectively, according to the literature.40 A three-layer model was used to fit the experimental data. Contact angle measurements were obtained using a CAM 200 optical contact angle meter (KSV instrument Ltd.). HPLC grade water was used for measurement. The contact angle was determined by using the Young/Laplace fitting method. Three replicates were measured for each substrate to calculate the standard deviation of the measurements. The topology and roughness measurements of polymer film formed on the substrates were measured using a Digital Instruments Nanoscope IIIa AFM microscope in tapping mode. The scan resolution was set at 512 points/line, the scan rate at 1 Hz, and the scan area of 2.5 µm × 2.5 µm. The scale bar of z-values was adjusted depending on surface roughness and was placed beside the images. The root-mean-square (rms) values were calculated with the vendor-provided software. All surface measurements were conducted with dried samples. RESULTS AND DISCUSSION As shown in Scheme 1, using ATRP to augment the presence of specific DNA sequences was separated into two consecutive yet independent stages: specific DNA recognition and signal amplification. DNA hybridization and ligation were used to differentiate target DNA sequences from mismatches. Since ATRP initiators attached on the detection probes had negligible impacts on either DNA duplex formation or DNA ligation, standard DNA handling protocols that have been well documented in the literature were directly adapted to maximize sequence recognition specificity. A three-strand synthetic oligonucleotide system was used in our concept-proof experiment. DNA capture probes of complementary (C) and noncomplementary (NC) sequences to the target DNAs were first immobilized on a Au substrate (Scheme 1). To ensure a consistent surface coverage, the thiol groups at the 3′end of both probes were freshly reduced.36 Subsequent surface (39) Brewer, S. H.; Anthireya, S. J.; Lappi, S. E.; Drapcho, D. L.; Franzen, S. Langmuir 2002, 18, 4460-4464. (40) Elhadj, S.; Singh, G.; Saraf, R. F. Langmuir 2004, 20, 5539-5543.

Figure 1. PM-IRRAS spectra of (a) capture-DNA probes on the Au surface; (b) initiator-coupled DNA duplex on the surface after hybridization; and (c) a reflectance FT-IR spectrum of the formation of PHEMA atop DNA molecules on the surface. The scale bar for spectrum c is shown at the up-right corner. Note that the reflectance FT-IR system used is ∼10× less sensitive than PM-IRRAS. Spectra a and b are enhanced five more times to show the spectral details.

blocking with MCH for 1 h was conducted to passivate the surface and remove any physically adsorbed oligonucleotides. The successful immobilization of probes C and NC on the surface was confirmed in a PM-IRRAS spectrum: a small peak near 1700 cm-1 was observed from the carbonyl stretching and exocyclic -NH2 bending vibrations of DNA bases. Additional peaks around 1500 cm-1 from the purine and pyrimidine rings and peaks in the region of 1000-1300 cm-1 from DNA phosphodiester backbones were also observed (Figure 1a).39 Note that no absorption was observed using traditional reflectance FT-IR measurements due to the limited amount of materials attached on the surface. Following surface immobilization, incubating the capture probecoated substrates with the target DNA sequence (C′D′) and then the initiator-labeled detection probes (D-IN) in the hybridization buffer led to the formation of DNA duplex at the desired location. The detection probes, D, were coupled with ATRP initiators (NHS ester of bromoisobutyroyl bromide) before hybridization. MALDIMS was used to examine the coupling efficiency (Supporting Information). A mass increase of 149 was clearly observed, corresponding well to the addition of bromoisobutyroyl group at the end of probe D. The negligible residual peak at the original position confirmed the coupling reaction was near completion in less than 30 min. Upon the hybridization of C′D′ and D-IN on the surface, a slight increase in carbonyl stretching was observed in the corresponding PM-IRRAS spectrum (Figure 1b), supporting the successful hybridization of DNA. A T4 ligase was subsequently used to connect the nicks between C and D-IN probes to permanently affix the detection probes on the surface, along with the ATRP initiators. The substrate was then immersed in an ATRP reaction mixture containing HEMA as the monomer. HEMA was chosen in our study for its good water solubility.33 It also offers an additional benefit in multistage ATRP reactions, which will be discussed later. The reaction catalyst was a mixture of CuCl, CuBr2, and 2,2′bipyridyl. CuBr2 was used as a deactivator to control the polymer Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

4701

Scheme 2. Chemical Formation of PHEMA in DNA Detection

Table 2. Ellipsometric and Contact Angle Measurements of PHEMA Formation in ATRP-Assisted DNA Detectiona

films

film thickness (Å)

contact angle (deg)

bare gold surface ssDNA-coated surface dsDNA-coated surface PHEMA formation on the surface control spots after ATRP

0b 16.4 ( 0.5 18.8 ( 0.5 100.1 ( 0.7 36.2 ( 0.7

89.0 ( 0.5 31.0 ( 0.0c 26.0 ( 0.2 44 ( 2 49 ( 2

a Two replicates were measured in ellipsometry and three replicates for contact angle measurements. The averages of the measurements were calculated. b The bare Au substrate was used as the reference point. No measurement was made at this point. c In this experiment, three measurements yielded the same readouts, with the standard variation of zero.

Figure 2. Photograph of 0.1 µM target DNA C′D′ detected after 5-h ATRP reaction. The growth of the polymer film results in the directly observable opaque spots where the complementary targets hybridized (C), whereas the control spots remain transparent (NC). The dotted circles are merely guides to the eye. ATRP catalysts used: [CuCl] ) mM; [CuBr2] - 1.5 mM; [tpy] ) 7.2 mM. Detailed ATRP reaction conditions see the text.

growth rate and ensure the formation of thick polymer films.29 Immersion of the initiator-attached Au substrate into this mixture immediately triggered ATRP reaction (Scheme 2, A). After a 5-h reaction, a layer of poly(hydroxyethyl methacrylate) (PHEMA) film was formed on the surface at where the capture probes C were located. The formed thick polymer film altered surface reflectivity, changed surface opacity, and rendered the spots directly visible to the naked eye (Figure 2). The spot visibility can be further increased by converting the hydroxyl groups to that of more hydrophobic nature. Similar color change was observed from the other nearby spot (C), prepared as the reaction replicate. In contrast, the control spots where the capture probes NC were immobilized remained unchanged (NC). The reflectance FT-IR measurement of the polymer-occupied spots illustrated the 4702 Analytical Chemistry, Vol. 77, No. 15, August 1, 2005

formation of a new peak between 3000 and 3500 cm-1 (hydroxyl stretching), accompanied with a significant absorption increase of carbonyl content near 1700 cm-1 (Figure 1c). Both changes confirmed the formation of PHEMA on the surface. No reflectance FT-IR signal was observed from the control spots where few HEMA molecules adsorbed nonspecifically. The contact angle measurements were also conducted to monitor surface hydrophobicity changes as surface chemistry varied (Table 1). The bare gold substrate exhibited a highly hydrophobic surface with a contact angle of ∼89°. The assembly of negatively charged DNA onto the surface greatly altered its interfacial properties and significantly reduced this contact angle to ∼31°. After DNA hybridization, the contact angle further decreased to ∼26°. Since both HEMA and PHEMA are more hydrophobic than DNA molecules, an increase in the contact angle was observed after exposing the surface to the ATRP reaction solution. The measured contact angle (∼44°) was similar to our previous measurement of PHEMA film grown atop ω-mercaptoundecyl bromoisobutyrate.41 The contact angle of the control spots was also increased after being exposed to the ATRP mixture. Note that the contact angle measurements alone could not be used to differentiate the formation of PHEMA from the nonspecifically adsorbed HEMA monomers due to their similar hydrophobicity. To clearly distinguish the growth of PHEMA atop DNA from the nonspecifically adsorbed HEMA on the control spots, an ellipsometer was used to measure the film thickness changes after ATRP. Using a bare gold substrate as the reference point, the immobilization of DNA capture probes resulted in a film growth of 16.4 ( 0.5 Å (Table 2). Hybridization with the target and detection probes only slightly increased the film thickness due to the limited hybridization efficiency. The ATRP reaction, however, significantly increased the film thickness to 100.1 ( 0.7 Å. The molecular weight of PHEMA formed cannot be directly measured, due to the limited amount of materials available on the surface. Nevertheless, tens to hundreds of repeating units were suspected in the formed polymer chains.42 On the control spot, on the other hand, the film thickness increased less than 20 Å from the nonspecific adsorption of monomers on the surface. (41) Lou, X.; He, L. Manuscript in preparation. (42) Ramakrishnan, A.; Dhamodharan, R.; Ruhe, J. Macromol. Rapid Commun. 2002, 23, 612-616.

Figure 3. PHEMA film growth atop DNA molecules as a function of polymerization time. A single-stranded DNA, A (Table 1), was used as the model system.

It is interesting to note that the NC spots exhibited less background signal, comparing to the surrounding areas where MCH was used as the blocking reagent. Using DNA and other reagents as the blocking reagents, along with more stringent surface washes, are under investigation to further reduce the background. The sensitivity of this polymerization-assisted DNA detection directly depends on two factors: (1) the number of the polymer chains formed, i.e., the density of the reaction initiators available on the surface; (2) the final molecular weight (MW) of each individual polymer molecule formed. In DNA detection, the electrostatic repulsion of highly charged DNA backbones limits the maximum amount of capture DNA probes immobilized to ∼1013 molecules/cm2.36 The low concentration of target DNAs obtainable for detection and lower than 100% hybridization and ligation efficiencies further reduce the amount of immobilized initiators available. Therefore, the growth of high MW polymer becomes the logical choice to improve DNA detection limit. Our investigation of the formation of DNA-containing polymer brushes on a surface showed the polymer film thickness changed linearly as a function of the reaction time. In this experiment, a singlestranded DNA (A) was used as the model sequence to mimic the ligation product of capture and detection DNA probes (C-D) and used to study PHEMA growth at various ATRP conditions. DNA A had a thiol group at the 3′-end for surface immobilization and a 5′-amino group for initiator coupling. Figure 3 showed that, under certain catalyst concentration, a linear relationship was observed with a PHEMA growth rate at ∼2 nm/h, after the initial rapid change in film thickness. Therefore, it is reasonable to expect that thicker polymer brushes could form to improve DNA detection sensitivity, giving sufficient monomers and reaction time provided. It is acknowledged that this growth rate is strongly dependent on the surface density of radicals, i.e., the initiator coverage. A different growth rate would be expected for the surfaces exposed to target DNA solutions of lower concentrations.43 Alternatively, the formation of branched polymer molecules can be used as a more efficient means to obtain substantial

Figure 4. (A) Two-stage 30-min ATRP reaction to form branched polymers. Note that the polymer structure was significantly simplified, and no unique conformation of linear/branched polymers was suggested in this scheme. (B) Photograph of 1 nM target DNA C′D′ detected after the two-stage ATRP reaction. The spots immobilized with the complementary probes shows significant color change (C), whereas the control spots remain unchanged (NC). The dotted circles are guides to the eye. ATRP catalysts used: [CuCl] ) 69 mM; [CuBr2] ) 20.7 mM; [bpy] ) 98.7 mM. Detailed ATRP reaction conditions see the text.

materials on the surface.44 Taking advantage of the additional hydroxyl groups available on HEMA, we performed a secondary ATRP reaction and formed branched PHEMA to dramatize DNA detection (Figure 4A). Specifically, linear PHEMA grew atop DNA molecules in the first a 30-min ATRP reaction to form an anchor layer. Additional 2-bromoisobutyryl bromide initiators were then coupled to the hydroxyl groups on the side chains of the anchor PHEMA. These initiators were subsequently used in the second round of 30-min ATRP reaction and grew into branched PHEMA (Scheme 2B). Direct surface inspection illustrated the formation of a much thicker layer of polymeric material as the optical clarity of the surface drastically changed (Figure 4B). A 1 nM concentration of target DNA was readily discernible to the naked eye. Considering the small volume of the hybridization solution used (1.5 µL), this change corresponded to the detection of 1.5 fmol of 30-mer oligonucleotide without the aid of any detectors. The formation of an ultrathick film using this two-stage ATRP strategy was also evidenced in surface characterization. The ellipsometer we used did not give an accurate readout, probably due to the significantly increased surface roughness that failed in theoretical model fitting. A local examination of the surface using an atomic force microscope (AFM) revealed the formation (43) Kim, J. B.; Huang, W. X.; Miller, M. D.; Baker, G. L.; Bruening, M. L. J. Polym. Sci. A 2003, 41, 386-394. (44) Matyjaszewski, K. Polym. Mater. Sci. Eng. 2001, 84, 363-364.


4703

Figure 5. AFM images of PHEMA films grown on DNA-coated Au substrates (A) after a 5-h ATRP reaction and (B) after two 30-min ATRP reactions. The images are 2.5 µm × 2.5 µm. The scale bars showing surface roughness are placed on the side of the images.

of polymer islands on the surface with feature heights exceeding 200 nm (Figure 5B). These larger individual features were attributed to the formation of branched polymers, leading to the improved detection sensitivity. The AFM images also revealed a pronounced difference in surface roughness between the substrates prepared from the two-stage ATRP to the one from onetime ATRP reaction (Figure 5A), with a rms roughness of 37.2 nm compared to 1.81 nm, respectively. In addition, it is worth pointing out that the reaction time was also significantly shortened from a 5-h linear reaction to two 30-min ones in the formation of branched polymers. Further improvement to lower detection limit or shorten assay time is under investigation using ATRP monomers with multiple hydroxyl groups to form hyperbranched polymers.45,46 The linear growth of PHEMA under controlled conditions makes quantification of target DNA feasible based on the ellipsometric measurements of film thickness. Nevertheless, this ATRP-amplification scheme is best used in DNA genotyping to provide yes-or-no answers, such as those in SNP screenings. In a proof-of-concept experiment, target DNAs of single or three mismatches were used to evaluate the fidelity of ATRP-assisted SNP detection (Figure 6). Three separate substrates were prepared with the same DNA probes two spots of the capture DNA probes (C), and the other two of noncomplementary probes (NC) for intersubstrate comparison. After surface passivation, three substrates were exposed to the solutions containing perfectly matched DNA target (C′D′), one-base-mismatch (C1′D′), and (45) Narain, R.; Armes, S. P. Macromolecules 2003, 36, 4675-4678. (46) Narain, R.; Armes, S. P. Chem. Commun. 2002, 23, 2776-2777.


Figure 6. Photographs of two-stage ATRP-assisted DNA point mutation detection. Only the perfectly matched target DNA results in the visible spots to the naked eyes (image A, spots C′D′), whereas one-mismatch (image B, spots C1′D′) and three-mismatch (image C, spots C3′D′) show negative responses. The spots where the noncomplementary probes immobilized (images A-C) also show negative responses. The dotted circles around the NC spots are guides to the eye.

three-base-mismatch DNA (C3′D′) solutions, respectively (Table 1). All three sequences were noncomplementary to the control probes on the NC spots, and all three substrates were treated with the same DNA detection probe, followed by DNA ligation. After the 1 × 5-h ATRP reaction, the growth of PHEMA was only visible on the substrate incubated with the perfectly matched target sequences (C′D′). No discernible patterns can be recognized for the substrates incubated with C1′D′, C3′D′, and the same low backgrounds were observed where the noncomplementary capture probes were immobilized (NC) across all three substrates. In conclusion, we have developed a signal amplification technique for the detection of specific DNA sequences based on controlled/“living” radical polymerization, ATRP. It is the first time that a small-molecule-initiated chain growth is used as an amplification method for biomolecular interaction detection.47 It offers three advantages over current DNA detection techniques: (1) small initiators instead of bulky detection tags are prelabeled (47) After the manuscript was accepted, the authors were made aware of a recently filed patent on the similar concept by a different group: Rowlen, K. L.; Birks, J. W.; Bowman, C.; Sikes, H. Use of Microarray and Photopolymerization for Amplification and Detection of Amplified TargetProbe Interaction. PCT Int. Appl. WO 2005024386, 2005.

onto DNA probes that introduce minimal interference to DNA sequence recognition; (2) detection amplification is independently conducted after DNA recognition. Consequently, numerous wellestablished DNA hybridization and ligation protocols can be used to improve single-mismatch detection fidelity. Meanwhile, various chemical procedures can be used to optimize the polymerization reaction without the concerns over the stability of detection tags or DNA duplexes, (3) the growth of high MW polymers eliminates the needs of sophisticated detectors to visualize femtomole target DNA in the sample. The formation of branched polymers in multistage ATRP reaction further improved the detection limit and significantly shortened the assay turnover rate. Together, the demonstration of the instrument-free ATRP-assisted DNA detection simplifies the future sensor construction and opens up

opportunities for the development of point-of-care diagnostic devices. ACKNOWLEDGMENT We thank Profs. Jan Genzer, Stefan Franzen, and Harold Ade for the use of their ellipsometer, FT-IR, and contact angle meter, respectively. SUPPORTING INFORMATION AVAILABLE MALDI-MS measurements of ATRP initiator-coupling reaction. This material is available free of charge via the Internet at http:// pubs.acs.org. Received for review April 24, 2005. Accepted May 13, 2005. AC050706H


4705

Detection of DNA Point Mutation by Atom Transfer Radical

Recommend Documents