Anal. Chem. 2008, 80, 6038–6044
Assembly-Disassembly of DNAs and Gold Nanoparticles: A Strategy of Intervention Based on Oligonucleotides and Restriction Enzymes I-Im S. Lim,† Uma Chandrachud,‡ Lingyan Wang,† Susannah Gal,*,‡ and Chuan-Jian Zhong*,† Departments of Chemistry, and Department of Biological Sciences, State University of New York at Binghamton, Binghamton, New York 13902 The ability to manipulate and intervene in the processes of assembly and disassembly of DNAs and nanoparticles is important for the exploitation of nanoparticles in medical diagnostics and drug delivery. This report describes the results of an investigation of a strategy to intervene in the assembly and disassembly processes of DNAs and gold nanoparticles based on two approaches. The first approach explores the viability of molecular intervention to the assembly–disassembly-reassembly process. The temperature-induced assembly and disassembly processes of DNAs and gold nanoparticles were studied as a model system to illustrate this approach. The introduction of a molecular recognition probe leads to intervention in the assembly-disassembly process depending on its specific biorecognition. This process was detected by monitoring the change in the optical properties of gold nanoparticles and their DNA assemblies. The second approach involves the disassembly of the DNAlinked assembly of nanoparticles using restriction enzymes (e.g., MspI). The presence of the double stranded DNAs in the nanoparticle assembly was also substantiated by a Southern blot. Implications of the results to exploration of the molecular intervention for fine-tuning interfacial reactivities in DNA-based bioassays are also discussed. The pioneering work on DNA mediated assemblies of gold nanoparticles by Mirkin and co-workers1 has opened the door to a host of potential applications in biological sensing, medical diagnostics, and drug delivery. Extensive research efforts have aimed at understanding the interactions and reactivities involved in the DNA-based nanoparticle assembly.2–13 Examples include DNA-gold nanoparticles with fluorescence signatures,2d sequencedependent stability of DNA-modified gold nanoparticles,2e quenching of single stranded DNA-linked nanoparticles,3 a DNAhybridization assay using bar-coded metal nanowires,4,5 single base mismatched associated fluorescent quenching,6 SERS-based * To whom correspondence should be addressed. E-mail: cjzhong@ binghamton.edu (C.J.Z.);
[email protected] (S.G.). † Department of Chemistry. ‡ Department of Biological Sciences. (1) (a) Mirkin, C. A.; Letsinger, R. L.; Mucic, R. C.; Storhoff, J. J. Nature 1996, 382, 607. (b) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078. (c) Storhoff, J. J.; Elghanian, R.; Mucic, R. C.; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 1998, 120, 1959.
6038
Analytical Chemistry, Vol. 80, No. 15, August 1, 2008
Scheme 1. Schematic Diagram (Not to Scale) Illustrating Two Different Approaches for Intervening in the Assembly and Disassembly Processes of DNAs and Au Nanoparticlesa
a (I) Molecular intervention with excess free DNA and (II) restriction enzyme cutting (2 ) increase in temperature; 1 ) decrease in temperature).
multiplexed detection,2i grazing-angle-FT-IR probing of orientations in DNA-nanoparticle hybridization,8 improvement of the sensitivity and photostability of DNA-hybridizations using dyedoped nanoparticles9 and restriction enzyme disassembly and DNA ligase reassembly of a DNA-nanoparticle network.10 While much has been learned about the assembly process, relatively little is known about the disassembly process. The ability to control both assembly and disassembly using biorecognition capabilities of DNA or proteins in combination with nanoparticles is important in biological processes, especially in drug delivery, gene therapy, immunotherapy, and a wide range of biological probes and sensors. Gold nanoparticle-based assays for detecting certain enzymes and proteins provide a sensitive means for rapid detection and analysis of disease and pathogens through changes in the optical properties of the nanoparticles. In this report, we describe the preliminary results of an investigation of two approaches to intervene in the assemblydisassembly process of the DNA-gold nanoparticles (Scheme 1). The first approach explores the viability of molecular intervention 10.1021/ac800813a CCC: $40.75 2008 American Chemical Society Published on Web 07/09/2008
(I) in the assembly-disassembly-reassembly process, whereas the second involves the use of restriction enzymes (II) to cut the interparticle DNAs in the nanoparticle assemblies. In both cases, the temperature-induced assembly and disassembly processes of DNAs with gold nanoparticles, as demonstrated first by Mirkin and co-workers,1 served as a model system for demonstrating the proof-of-concept of our intervention strategies. While the DNA base sequences in the first approach follow those reported by Mirkin et al.,1c the use of a free oligonucleotide to intervene the reassembly process of the DNA-nanoparticle system is a new concept.13 For the second approach, restriction enzymes are known to cut DNA at sequence specific sites and are often used by some microorganisms to protect against invasion. In view of the many commercially available enzymes that have been used in a variety of molecular and biological manipulations, the study of the viability of this approach in the intervention of the assembly disassembly processes should provide useful information about the detailed molecular interactions and mechanisms. As detailed in this report, the 30 base-pair DNA sequences used for the demonstration of this approach were designed specifically for the restriction enzymes to recognize specific target sites. EXPERIMENTAL SECTION Chemicals. Hydrogen tetrachloroaurate(III) hydrate (HAuCl4), sodium citrate (99%), sodium chloride (NaCl, 99%), sodium dodecyl sulfate (SDS), ethylenediaminetetraacetic acid (EDTA), Tris, borate, ethidium bromide, and dithiothreitol (DTT) were purchased from Sigma-Aldrich (Milwaukee, WI) and used as received. Phosphate buffer (0.05 M, pH 7) and bovine serum albumin (BSA; fraction V) were purchased from Fisher Scientific (2) (a) Jin, R.; Wu, G.; Li, Z.; Mirkin, C. A.; Schatz, G. C. J. Am. Chem. Soc. 2003, 125, 1643. (b) Reynolds, R. A., III; Mirkin, C. A.; Letsinger, R. L. J. Am. Chem. Soc. 2000, 122, 3795. (c) Stoeva, S. I.; Huo, F.; Lee, J.; Mirkin, C. A. J. Am. Chem. Soc. 2005, 127, 15362. (d) Demers, L. M.; Mirkin, C. A.; Mucic, R. C.; Reynolds, R. A., III; Letsinger, R. L.; Elghanian, R.; Viswanadham, G. Anal. Chem. 2000, 72, 5535. (e) Storhoff, J. J.; Elghanian, R.; Mirkin, C. A.; Letsinger, R. L. Langmuir 2002, 18, 6666. (f) LyttonJean, A. K. R.; Han, M. S.; Mirkin, C. A. Anal. Chem. 2007, 79, 6037. (g) Hurst, S. J.; Han, M. S.; Lytton-Jean, A. K. R.; Mirkin, C. A. Anal. Chem. 2007, 79, 7201. (h) Xu, X.; Georganopoulou, D. G.; Hill, H. D.; Mirkin, C. A. Anal. Chem. 2007, 79, 6650. (i) Cao, C.; Jin, R.; Mirkin, C. A. Science 2002, 297, 1536. (3) (a) Li, H.; Rothberg, L. J. J. Am. Chem. Soc. 2004, 126, 10958. (b) Li, H.; Rothberg, L. J. Anal. Chem. 2004, 76, 5414. (4) (a) Nicewarner-Pena, S. R.; Freeman, R. G.; Reiss, B. D.; He, L.; Pena, D. J.; Walton, I. D.; Cromer, R.; Keating, C. D.; Natan, M. J. Science 2001, 294, 137. (b) Keating, C. D.; Natan, M. J. Adv. Mater. 2003, 15, 451. (c) Mbindyo, J. K. N.; Reiss, B. D.; Martin, B. R.; Keating, C. D.; Natan, M. J.; Mallouk, T. E. Adv. Mater. 2001, 13, 249. (5) Braun, E.; Eichen, Y.; Sivan, U.; Ben-Yoseph, G. Nature 1998, 391, 775. (6) Maxwell, D. J.; Taylor, J. R.; Nie, S. J. Am. Chem. Soc. 2002, 124, 9606. (7) Vidal, B. C.; Deivaraj, T. C.; Yang, J.; Too, H.; Chow, G.; Gan, L. M.; Lee, J. Y. New J. Chem. 2005, 29, 812. (8) Sauthier, M. L.; Carroll, R. L.; Gorman, C. B.; Franzen, S. Langmuir 2002, 18, 1825. (9) Zhou, X.; Zhou, J. Anal. Chem. 2004, 76, 5302. (10) (a) Kanaras, A. G.; Wang, Z.; Bates, A. D.; Cosstick, R.; Brust, M. Angew. Chem., Int. Ed. 2003, 42, 191. (b) Wang, Z.; Kanaras, A. G.; Bates, A. D.; Cosstick, R.; Brust, M. J. Mater. Chem. 2004, 14, 578. (c) Kanaras, A. G.; Wang, Z.; Brust, M.; Cosstick, R.; Bates, A. D. Small 2007, 3, 590. (d) Kanaras, A. G.; Wang, Z.; Hussain, I.; Brust, M.; Cosstick, R.; Bates, A. D. Small 2007, 3, 67. (11) Li, M.; Mann, S. J. Mater. Chem. 2004, 14, 2260. (12) McIntosh, C. M.; Esposito, E. A., III; Boal, A. K.; Simard, J. M.; Martin, C. T.; Rotello, V. M. J. Am. Chem. Soc. 2001, 123, 7626. (13) Lim, I-I. S.; Wang, L.; Chandrachud, U.; Gal, S.; Zhong, C. J. Res. Lett. Nanotechnol. 2008, Article ID 527294.
(Pittsburgh, PA). Thiol, digoxigenin and biotin modified DNAs were purchased from Integrated DNA Technologies, Inc. with standard desalting purification (Coralville, IA). NAP-5 columns were purchased from Amersham Biosciences (Uppsala, Sweden). Precast polyacrylamide gels were purchased from Biorad Corporation (Hercules, CA) and the restriction enzymes from New England Biolabs (Beverly, MA). Water was purified with a Millipore Milli-Q water system. Instrumentation. UV-visible (UV-vis) spectra were acquired with a Hewlett-Packard 8453 spectrophotometer. The spectra were collected over the range of 200-1100 nm. Transmission electron microscopy (TEM) was performed on Hitachi H-7000 electron microscope (100 kV). TEM samples were taken from the assembly and disassembly solutions of DNAs and nanoparticles. Typically, the TEM samples were prepared by taking a solution sample and casting it onto a carbon-coated copper grid sample holder followed by evaporation in air at room temperature. Experimental Procedures. The detailed procedures for the molecular intervention and the restriction enzymes cutting of the Au-DNA nanoparticle assemblies are as follows. Molecular Intervention Based on Complementary Oligonucleotides (I). The three different DNAs were used. (DNA1, 5′-TCTCAACTCGTA/3ThioMC3-D/3′; DNA2, 5′/5ThioMC6-D/ CGCATTCAGGAT-3′; and DNA3, 5′-TACGAGTTGAGAATCCTGAATGCG-3′). The citrate-capped gold nanoparticles (Aunm 11.4 ± 0.8 nm) were synthesized using the reported procedure.14 DNA1 and DNA2 were first dissolved in 0.2 M phosphate buffer (pH 8) at a concentration ranging from 300 to 370 µM. The disulfide bonds in DNA1 and DNA2 were cleaved using an approach similar to the reported procedure1 where dithiothreitol (DTT) was added at 0.1 M final concentration to ∼10 OD of the nucleotides in a final volume of 400 µL. The solution was allowed to react at room temperature for 2 h and then put through a NAP-5 column, and an aliquot of 1.1 mL of 0.1 M phosphate buffer (pH 8) was added to the column to elute the cleaved oligonucleotide. The final concentration of the cleaved DNAs was 20 µM with an OD260nm of 2.2. The exact concentrations of DNAs varied slightly depending on the specific experiment. The surface of the gold nanoparticles were functionalized with the cleaved 5′ and 3′- thiol modified oligonucleotides similar to the reported procedure1 to form A1 and A2 [with DNA1 and 2, respectively]. Briefly, 1 mL of the cleaved DNAs (1 or 2) separately was added to 5 mL of gold nanoparticles (stock concentration 14 nM). The solution was left standing at room temperature for 16 h, after which it was diluted to 0.1 M NaCl and 10 mM phosphate buffer (pH 7) and allowed to stand for another 40 h at room temperature. The DNA-capped nanoparticles were then centrifuged and washed twice at 14 000 rpm (18 620g) for 25 min (each time the solution was redispersed in a 0.1 M NaCl/10 mM phosphate buffer (pH 7) solution) before being redispersed in its final 0.3 M NaCl/10 mM phosphate buffer (pH 7) and 0.01% sodium azide solution and stored at room temperature. In some cases, the solution of DNA-capped nanoparticles was heated up to 50 °C for 10 min before centrifugation. With dependence on the batch of nanoparticles and DNA used, the concentration of each component varied slightly. (14) Lim, I-I. S.; Goroleski, F.; Mott, D.; Kariuki, N. N.; Ip, W.; Luo, J.; Zhong, C. J. J. Phys. Chem. B 2006, 110, 6673.
Analytical Chemistry, Vol. 80, No. 15, August 1, 2008
6039
To study the assembly of the DNA-capped nanoparticle solution, 180 µL of A1 was added to 180 µL of A2 in the presence of 240 µL of 10 mM phosphate buffer (pH 7)/0.3 M NaCl solution. Then, 6 µL of 10 µM DNA3 (also called T1) was added to the solution and the reaction monitored by UV-vis. For the molecular intervention study, 2 µL of 369 µM of uncleaved DNA1 (P1) was added to the solution of (A1 + A2 + T1) when the temperature of the solution was 80 °C controlled by using a temperature controller (Glas-Col, Digi Trol II) with the liquid flow controlled by a mechanical pump. Actual temperatures in the reaction solution were measured by immersing a thermal couple into the cuvette during the reaction. Restriction Enzyme Cutting. To demonstrate the viability of restriction enzyme cutting, two different DNAs, (5′-/5ThioMC6D/TGCCAAGGCTTGCCCGGGCAGGTCTGGCCT-3′ (top strand) and 5′-/5ThioMC6-D/AGGCCAGACCTGCCCGGGCAAGCCTTGGCA-3′ (bottom stand)) were used. The immobilization of the DNAs on Au nanoparticles was carried out in a similar way as in the above procedures. The surface of gold nanoparticles functionalized with the cleaved 5′ thiol modified oligonucleotides form B1 and B2 with the top and bottom DNAs, respectively. For the assembly of the nanoparticles, 0.7 mL of B1 was added to 0.4 mL of B2 with 0.4 mL of 10 mM phosphate buffer (pH 10)/ 0.1 M NaCl solution. To detect the changes in the surface plasmon resonance band, the solution was heated to 90 °C using a quartz cuvette with a built-in liquid flow cell connected to a temperature controller for 15 min and gradually cooled down to room temperature. For the enzyme cutting experiment, the restriction enzymes MspI (100 units/ µL) and SmaI (20 units/µL) were utilized. Two different approaches were conducted in this study. In the first approach, 1-2 µL of restriction enzyme was added to 17 µL of the assembled solution (theoretically should contain a total of ∼19 pmol of DNA) along with the buffer for the restriction enzyme. The solution was incubated for various times at 25 °C (SmaI) or 37 °C (MspI) with constant shaking. After the incubation, EDTA (final concentration of 0.02 M) was added to the solution to inactivate the enzyme and DTT (final concentration of 0.4 M) was added to detach the nanoparticles from the DNA fragments. The solutions were incubated overnight at room temperature with constant shaking. Alternatively, the DNAs were removed first from the assembled nanoparticles using DTT, then the enzyme was added and reacted for various times as above. In some reactions, the amount of components was increased 3× to obtain sufficient DNA for detection, then after removal from the nanoparticles, the water was removed using evaporation before loading on the gel. These reacted DNAs were separated by polyacrylamide gel electrophoresis in TBE buffer (0.089 M Tris, 0.089 M borate, 2 mM EDTA) and visualized using ethidium bromide (0.5 µg/mL in TBE) and imaged using a Kodak EDAS 290 gel documentation system (Rochester, NY). To confirm the presence of both DNA strands in the nanoparticle assembly, we performed a Southern blot. After the ethidium bromide imaging of the gel, the gel was incubated in a solution (0.5 M NaOH and 1.5 M NaCl solution) for 10 min to denature any double-stranded DNAs to single-stranded oligonucleotides. The gel was then transferred to a positively charged nylon membrane (Roche, Mannheim, Germany) for 30 min in a 6040
Analytical Chemistry, Vol. 80, No. 15, August 1, 2008
semidry transfer machine (Biorad Hercules, CA) using TBE buffer, followed by UV-cross-linking of the DNA to the nylon membrane using a UV cross-linker (Fisher Scientific). Part of the membrane was probed with a digoxigenin-labeled top strand DNA (M1) that recognizes the bottom strand DNA, and the other was probed with a biotin-labeled bottom strand DNA (M2) that recognizes the top strand DNA linked to gold nanoparticles (see Scheme 4). Prehybridization of the membrane was performed using the hybridization solution (0.25 M sodium phosphate, dibasic pH 7.2, 1 mM EDTA, 7% SDS) for 1 h at 35 °C in a hybridization oven. Both of the labeled probes (525 pm of the biotin labeled probe and 540 pm of the digoxigenin labeled probe) were first boiled (5 min at 95 °C) and cooled immediately to make them single stranded, then added to 5 mL of the hybridization solution. Both the membranes were incubated overnight in the hybridization solution at 35 °C, after which the membranes were washed in 2XSSC (0.3 M sodium chloride and 0.03 M sodium citrate at pH 7.0) 1% SDS solution for 15 min at 35 °C and then washed with 2XSSC for 10 min at 35 °C. The membranes were then incubated in the blocking solution 1XTBS (0.5 M Tris, 1.50 M NaCl, pH 7.5), 5% BSA, and 0.1% Tween 20 for 1 h. The membrane with the biotin labeled probe was then incubated with streptavidin horse radish peroxidase (1/100 000 in the blocking solution) (Pierce, Rockford, Illinois). The digoxigenin labeled probe was incubated with an alkaline phosphatase linked antidigoxigenin antibody (1/10 000 in the blocking solution) (Roche, Mannheim, Germany). This incubation was done for 1 h. The membranes were then washed with 1XTBS, 0.1% Tween 20 solution three times. The membrane with biotin labeled probe was incubated with the horseradish peroxidase chemiluminiscent substrate (Pierce, Rockford, IL), and the membrane with digoxigenin labeled probe was incubated with Lumi Phos alkaline phosphatase substrate (Pierce). The membranes were then exposed to X-ray film. RESULTS AND DISCUSSION Molecular Intervention Based on Complementary Oligonucleotides. The temperature dependent assembly-disassembly processes (S1) are conceptually illustrated in Scheme 2; these processes have been documented.1 To demonstrate the viability of the molecular intervention process (S2), we used the same oligonucleotides as previously reported.1 The introduction of a molecular recognition probe (P1) into a solution of two types of DNA-anchored nanoparticles, e.g., DNA1-capped Au (A1) and DNA2-capped Au (A2) in the presence of a target DNA (T1) as reported previously,1 lead to the possibility of intervention of the assembly-disassembly process depending on its recognition with A1, A2, or T1. For example, if recognition occurs between P1 and T1, the presence of the former DNA could prevent A1 from reassembling with T1 and A2 as the solution temperature is reduced. This process can be detected by monitoring the change in the optical properties of gold nanoparticles and their DNA assemblies. Thus, the formation of S2 provides opportunities for further fine-tuning of the interfacial reactivities in the DNAnanoparticle bioassay or biomaterial engineering, which constitutes the basic motivation for this work. The gold nanoparticles used in this study display a surface plasmon (SP) resonance band at about 520 nm (Figure 1), the shift of which is correlated to the change in the size, interparticle
Scheme 2. Schematic Diagram (Not to Scale) Illustrating (S1) the Assembly of DNA1-Capped Au (A1) and DNA2-Capped Au (A2) Nanoparticles via a Target DNA (T1) and (S2) an Intervention Technique of the Solution upon Increasing the Temperature and Introducing P1 into the Heated Solution Which Leads to the Formation of S2 upon Returning to Room Temperaturea
a 2 ) Heat to 75-80 °C; 1 ) gradually cooling down to 25 °C (RT). The detailed complementary binding of A2 and P1 in the presence of its complementary DNA (T1) is included.
Figure 1. Spectral evolution of the SP band showing the assembly of DNA capped nanoparticles (A1 and A2) in the presence of a target DNA (T1) (A), the reactivity upon heating and cooling of the solution from A in the absence (B) and presence of an uncleaved DNA (P1) (C). Concentration for various components: [Aunm] ) 3.3 nM; [DNA1 or DNA2] ) 1.0 µM; [T1] ) 0.1 µM; [P1] ) 1.2 µM.
Figure 2. TEM images of samples from A1 (A), A2 (B), and the assemblies upon the addition of T1 (C). (Note that the scale bar for the assemblies is smaller in order to show the degree of clustering).
distance, and dielectric medium.15 As shown in Figure 1A, the solution of the DNA capped nanoparticles (A1 + A2) (red curve) (15) (a) Lim, I-I. S.; Zhong, C. J. Gold Bull. 2007, 40, 59. (b) Njoki, P. N.; Lim, I-I. S.; Mott, D.; Park, H. Y.; Khan, B.; Mishra, S.; Sujakumar, R.; Luo, J.; Zhong, C. J. J. Phys. Chem. C 2007, 111, 14664. (c) Lim, I-I. S.; Vaiana, C.; Zhang, Z.; Zhang, Y.; An, D. L.; Zhong, C. J. J. Am. Chem. Soc. 2007, 129, 5368. (d) Lim, I-I. S.; Ip, W.; Crew, E.; Njoki, P. N.; Mott, D.; Zhong, C. J.; Pan, Y.; Zhou, S. Langmuir 2007, 23, 826. (e) Lim, I-I. S.; Pan, Y.; Mott, D.; Ouyang, J.; Njoki, P. N.; Luo, J.; Zhou, S.; Zhong, C. J. Langmuir 2007, 23, 10715. (16) (a) Heidmann, S.; Seifert, W.; Kessler, C.; Domdey, H. Nucleic Acids Res. 1989, 17, 9783. (b) Pingoud, A.; Jeltsch, A. Nucleic Acids Res. 2001, 29, 3705. (c) Line, P. M.; Lee, C. H.; Roberts, R. J. Nucleic Acids Res. 1989, 17, 3001. (d) Greene, P. J.; Gupta, M.; Boyer, H. W.; Brown, W. E.; Rosenberg, J. M. J. Biol. Chem. 1981, 256, 2143.
exhibited an SP band at 525 nm. Upon the addition of T1 in room temperature, the SP band shifted gradually from 525 to ∼560 nm within a 30 min time frame (blue curve in Figure 1A), after which a decrease in the absorbance began to occur as a result of the precipitation of the assembly solution due to the formation of large clusters. The color of the solution changed from red to purple during this process. Transmission electron microscopy images for samples from the A1 or A2 solution showed scattered and isolated nanoparticles (Figure 2A,B), whereas the sample containing the A1, A2, and the T1 DNA exhibited highly clustered features of the nanoparticle assemblies (Figure 2C). The degree of clustering is dependent on the concentration of the nanoparAnalytical Chemistry, Vol. 80, No. 15, August 1, 2008
6041
Scheme 3. Schematic Diagram (Not to Scale) Illustrating the Disassembly of the DNA-Au Nanoparticle Assemblies Using MspI Restriction Enzyme before Subjecting the DNA to a Gel Analysis
ticles, oligonucleotides, and salts, in addition to the temperature (data not shown). This assembly-disassembly process is reversible as demonstrated by the spectral evolution in the process of heating the solution above the DNA’s melting temperature and the process of returning to room temperature (Figure 1B).1,2 In this example, the solution was heated to 80 °C for 5 min, where the solution displayed a red color and the SP band shifted to 525 nm. As the solution temperature was cooled down, the color of the solution turned purple which is accompanied by the red-shift and the broadening of the SP band. These observations are consistent with those reported previously.1,2 As an example, we used uncleaved DNA1 (5′ TCTCAACTCGTA/ 3ThioMC3-D/-3′) as P1 to demonstrate the viability of process-S2 in Scheme 2. When the solution of the DNA-nanoparticle assembly was heated to 80 °C, a controlled amount of P1 (12X the T1 DNA concentration) was added to the solution. The solution temperature was kept at 65 °C for an additional 5 min (Figure 1C). When the solution was cooled down to room temperature, the SP band remained at 525 nm without any detectable broadening toward longer wavelengths. The color remained reddish. This observation is in sharp contrast to the optical changes for the same process with A1 + A2 + T1 in the absence of P1 (Figure 1B). The contrast in the spectral evolution is indicative of the intervention of P1 in the assembly process, which served as an example demonstrating the viability of manipulating the complementary binding of DNAs in the assembly disassembly processes. The absence of a change in the solution color and the spectral absorbance for this second process suggested that P1 has taken the place of A1 in terms of the complementary binding, which can be explained by the difference in size and mobility between P1 and A1. In comparison with A1 which has DNA1 attached onto the nanoparticle surface, P1 has a smaller size and a faster mobility. Therefore, the reactivity between P1 and T1 upon cooling is favored over A1 in terms of the complementary binding to T1, as illustrated in Scheme 2. On the basis of the concentrations of T1 and A1 or A2 (after centrifugation and assuming 50% of the DNA1 and DNA2 bound to the nanoparticles) and a spherical model for Au nanoparticles, the estimated surface coverage of the thiolate-oligonucleotides, under ideal surface packing conditions, was about 150 per particle2d with ∼25% being bound to T1. The amount of P1 added to the solution (∼350 P1 molecules per nanoparticle and 12× the T1 concentration) is therefore more than sufficient to hybridize to all T1 moieties before they interact with A1 or A2. It is expected that the proportion of intervention should 6042
Analytical Chemistry, Vol. 80, No. 15, August 1, 2008
be dependent on the relative concentration of P1 vs the other constituents in the solution. While the preliminary results call for further investigation into the detailed structures and reactivities of S2, the demonstration of the viability of the intervention strategy for manipulating the assembly disassembly processes of DNA and nanoparticles potentially opens opportunities for expanding this strategy into applications in biorecognition-based assays and biomaterials engineering. For example, the disulfide moieties on the outmost shell structure of S2 provide the binding sites for attachment to different types of nanoparticles in terms of size and functionality. The DNA strands can be immobilized on different nanoparticles or substrates via a gold-thiolate bond as a result of the cleavage of the disulfide bond upon its adsorption on the gold surface. The tailoring of the functional properties of the R-ligand in Scheme 2 could potentially serve as probes in fine-tuning the interfacial reactivity. With dependence on the identity or reactivity of the R group, the surface chemistry of the DNA-anchored nanoparticles can be tailored to meet specific technological applications, which are part of our ongoing quantitative work on the dependence of the reactivity on the concentrations of the DNAs involved and the optimization of the reaction. Molecular Intervention Based on Restriction Enzyme Cutting. As illustrated by the B1-B2 pair in Scheme 3, the complementary bound DNAs in the assembly of DNA-Au nanoparticles can be cut by an enzymatic reaction at a specific site. Upon assembly of the nanoparticles via complementary binding of DNAs, the restriction enzyme was added to the assembly solution and incubated for various times. This was then followed by the addition of EDTA to inactivate the enzyme and DTT to detach the nanoparticles from the DNAs.10 As a control (not shown), DTT was added first to the assembled solution followed by the addition of the restriction enzyme. After various reaction times, EDTA was added to the solution to inactivate the enzyme. These DNAs were then analyzed using polyacrylamide gel separation. Figure 3 shows the spectral evolution of the SP band for the assembly of B1 and B2. In comparison with the control experiments (Figure 3 insert), the absence of any spectral evolution toward the longer wavelength when B1 and B2 are mixed at room temperature (not shown) or as the solution was heated to a higher temperature (red curve) is consistent with the absence of binding between the top and bottom strand DNAs. However, at the higher temperature, the DNA structures were fully single stranded. As the temperature gradually decreased back to room temperature, a spectral change was evident as reflected by the expansion of
Scheme 4. Schematic Diagram (Not to Scale) Illustrating Reactivities of the Assembly of Top Strand DNA-Capped Au (B1) and Bottom Strand DNA-Capped Au (B2) Nanoparticles via Complementary Binding Pairs in Four Processesa
a (1) the addition of DTT to detach the Au nanoparticles from the B1-B2 assembly pair, (2) gel separation, (3) the addition of NaOH to denature the DNA fragments, (4) the transfer of DNA to a membrane, and (5) binding of complementary tagged target probes (M1 or M2), followed by detection of the bound tagged probes using specific recognition molecules (antibodies).
Figure 3. Spectral evolution of the SP band showing the assembly of the DNA capped nanoparticles (B1 and B2) upon heating (90 °C, red) and cooling (25 °C, blue). Insert: control experiment of unmodified Au nanoparticles upon heating (90 °C) and cooling down to room temperature. Concentration for various components: [Aunm] ) 3.9 nM; [top or bottom DNA] ) 1.2 µM assuming about 50% of the initially added DNA is bound to the nanoparticles. The arrows indicate the spectral evolution changes as the DNA (B1 and B2) capped nanoparticles gradually cooled to room temperature.
the SP band toward the longer wavelength (blue curve). This spectral change is indicative of the assembly of the nanoparticles via DNA hybridization. The disassembly of the assembled nanoparticles in the solution was examined next using two types of restriction enzymes, SmaI and MspI. After the incubation of the assembled DNA-nanoparticle solution with 40 units of SmaI, the DNA was removed and separated on a gel, and visualized with ethidium bromide. No significant change in the mobility of the DNA from the digested DNA-nanoparticle assembly was seen (Figure 4A). To substantiate that the DNA in the assemblies was indeed double stranded, a Southern blot was performed on the separated DNA using modified tagged/target probes (M1 or M2) complementary to the bottom and top DNAs, respectively (Scheme 4). The Southern blot experiment showed that both top and bottom strands were present in the DNA isolated from the assembled nanoparticles (Figure 4B). As a control, single stranded bottom and top DNAs isolated from the nanoparticles (lanes 6 and 7) were also probed with the biotin-labeled bottom strand.
Only the top strand was detected as expected which indicates the specificity of the hybridization reaction. These results led to the conclusion that the nanoparticle-DNA assemblies used in the restriction digestion experiment were indeed double-stranded. Thus, it was not clear why the DNA in the assembled nanoparticles was not digested by the SmaI restriction enzyme. We next examined digestion with another restriction enzyme, MspI, recognizing the same region of the assembled DNA as SmaI and observed digestion (Figure 5). As shown in Figure 5, in the presence of MspI, cutting of the double-stranded DNA assembled nanoparticles was observed in a time dependent manner (lanes 3-5). A control experiment without the enzyme shows the DNA band at the expected position of full-length DNA (lane 1). The smaller DNA bands that appear during enzyme digestion correspond to the size of fragments produced when the free oligonucleotides are incubated with a restriction enzyme cutting in the same place (data not shown). The cut DNA appears as two bands because the enzyme produces fragments that are one base different in length, which can be separated using this polyacrylamide gel. Thus, our results showing digestion with MspI of the assembled DNA-nanoparticles are similar to the observation reported previously by Brust and co-workers for the system using the EcoRI and other restriction enzymes.10 We were unsuccessful in seeing the cutting of the DNAnanoparticle assembly when we used another restriction enzyme SmaI. We believe that the DNAs might force the gold particles too close together and block the ability of the restriction enzyme to have access to the DNA molecules. SmaI is an enzyme with monomer molecular weight of approximately 29 kDa,16a although these enzymes work generally as dimers.16b If this enzyme works in a larger complex, it may have difficulty accessing the DNA strands between the gold nanoparticles. This enzyme can clearly cut this double-stranded DNA if it is not attached to gold nanoparticles (lane 1 in Figure 4). It is also interesting to note that the length of the DNA connecting the nanoparticles in our system is smaller than that used in the other published work with the cutting of DNAs (30 versus 50 bp).10 This difference may make it harder for enzymes to access the DNA between the nanoparticles. Part of our ongoing work involves the assembly of nanoparticles using oligonucleotides with different lengths and Analytical Chemistry, Vol. 80, No. 15, August 1, 2008
6043
Figure 4. Gel electrophoresis data shows lack of digestion by SmaI but presence of double-stranded DNA in the nanoparticle assembly. Double-stranded DNA oligonucleotides (10 pmole) reacted with SmaI overnight, followed by addition of DTT (lane 1), or not reacted with enzyme (10 pmole: lane 2, or 1 pmole: lane 3) were separated on a 10% polyacrylamide TBE gel. Also separated on the gel were the DNAs removed from the DNA-nanoparticle assembly reacted without (lane 4) or with SmaI enzyme (lane 5) overnight and DNA removed from nanoparticles made with either the bottom or top DNA strands (lanes 6 and 7, respectively). The DNAs were detected after separation using ethidium bromide (A) or through hybridization (B) of the modified top digoxigenin labeled DNA (M1) (lanes 1-4) or the modified bottom strand biotin labeled DNA (M2) (lanes 5-7) following a Southern blot. The arrow and arrowhead mark the positions of uncut and cut DNAs, respectively.
is 31 kDa).16d These enzymes both work best at warmer temperatures than SmaI (37 °C versus 25 °C), which might explain part of this discrepancy as explained above. Another major difference is the amount of the enzyme used in the successful reactions. Because of the available concentration of the SmaI and MspI enzymes, 5-fold less of the former enzyme was used in these reactions. The amount of enzyme used in the published work with EcoRI is in the same range as what we used for MspI (300 units).10 Thus, the amount of available enzyme may be a significant barrier to successful cutting.
Figure 5. Gel electrophoresis of cut DNAs from assembled nanoparticles. Assembled DNA/Au (51 µL) treated without (lane 1) and with 300 units of the MspI restriction enzyme for 0 (lane 2), 1 (lane 3), 5 (lane 4), and 7 h (lane 5) followed by addition of EDTA and DTT, and concentrated via evaporation, was separated on a 15% TBE polyacrylamide gel then stained with ethidium bromide. The arrow and arrowhead mark the positions of uncut and cut DNAs, respectively.
base pairs. For example, the addition of spacers to either side of the DNA sequences while keeping the enzyme recognition site intact may provide new insight into the reactivity of the restriction enzymes. The inability for this restriction enzyme to completely cut the assembled DNAs might also be due to the high kinetic barrier in our system. As illustrated in one of our earlier papers in the assembly and disassembly of homocysteine capped nanoparticles,15d the densely packed head-to-head zwitterion electrostatic interactions between the amino acid groups of homocysteine bound to the gold nanoparticles constitutes a significant barrier to the deprotonation of the amine group at ambient temperature, which explains the slow pH-induced disassembly kinetics. This barrier is overcome by elevating the temperature resulting in an interparticle spatial expansion, which increases the pH-induced disassembly rate. We did get cutting with another restriction enzyme, MspI, that has a similar monomeric size (29.8 kDa).16c EcoRI used by Brust and colleagues has a similar size (monomer 6044
Analytical Chemistry, Vol. 80, No. 15, August 1, 2008
CONCLUSIONS In summary, two approaches to intervening in the assemblydisassembly processes of the DNA-gold nanoparticles have been demonstrated. The oligonucleotide-complementary binding in the assembly-disassembly-reassembly processes is shown to be an effective strategy for molecular intervention. This is evidenced by monitoring the change in the optical properties of gold nanoparticles and their DNA assemblies in the temperatureinduced assembly and disassembly processes. Cutting the DNAassembled gold nanoparticles using a restriction enzyme is shown to be another viable strategy for manipulation of the disassembly process. This is evidenced by monitoring the appearance of bands indicative of the cutting of the double-stranded DNAs in the assembled nanoparticles using gel electrophoresis. Further delineation of the quantitative aspects of the molecular intervention and restriction enzyme cutting processes and the related parameters for expanding the strategy for DNA-based bioassays and biomaterial engineering is part of our ongoing efforts. ACKNOWLEDGMENT This work is supported in part by the National Science Foundation (Grant No. CHE0349040), and in part by the U.S. Department of Energy (Grant DE-FG02-06ER64281) as a subcontract from SUNY-Utica. I-Im S. Lim acknowledges the support of the National Science Foundation Graduate Research Fellowship. Received for review April 22, 2008. Accepted June 4, 2008. AC800813A
