NANO LETTERS
Self-Assembly of Enzymes on DNA Scaffolds: En Route to Biocatalytic Cascades and the Synthesis of Metallic Nanowires
2009 Vol. 9, No. 5 2040-2043
Ofer I. Wilner, Simcha Shimron, Yossi Weizmann, Zhen-Gang Wang, and Itamar Willner* Institute of Chemistry, The Hebrew UniVersity of Jerusalem, Jerusalem 91904, Israel Received January 29, 2009; Revised Manuscript Received February 26, 2009
ABSTRACT DNA strands consisting of programmed sequence-specific domains were synthesized by the rolling circle amplification (RCA) process. The spatial positioning of glucose oxidase (GOx) and of horseradish peroxidase (HRP) on the RCA-synthesized DNA template via hybridization enabled the activation of the bienzyme cascade. The GOx-catalyzed oxidation of glucose yielded gluconic acid and H2O2, and the resulting H2O2 oxidized 2,2′-azino-bis[3-ethylbenzthiazoline-6-sulfonic-acid] (ABTS2-) in the presence of HRP. The enzyme cascade could not be activated in the absence of the organizing DNA template or in the presence of a foreign DNA. Also, Au NPs-functionalized GOx was hybridized with the RCA-synthesized single-stranded DNA. The biocatalytic growth of the NPs through the oxidation of glucose, in the presence of AuCl4-, enabled the synthesis of 1-5 µm long Au wires, exhibiting a width of ca. 150 nm.
The programmed assembly of proteins on nucleic acid nanostructures attracts recent research efforts.1-3 The ordered binding of avidin to biotin-tagged DNA nanostructures,4,5 the specific association of thrombin to aptamer units associated with DNA arrays,6-9 and the programmed association of proteins to multiplexed aptamer chains synthesized by the rolling circle amplification (RCA) process10 were reported. Also, the selective linkage of RecA to a homologous nucleic acid that interacts with a duplex nucleic acid structure was suggested as a means to chemically pattern double-stranded DNA nanostructures with metallic nanowires11 and to fabricate nanoscale transistor devices.12 Similarly, the assembly of metal nanoparticles on DNA scaffolds attracts research efforts directed to the “bottom-up” synthesis of metallic nanowires and nanocircuitry.13-15 Although substantial progress was accomplished in the fabrication of protein-DNA, or nanoparticle (NP)-DNA nanostructures,16-21 the design of functions that emerge from these hybrid architectures remains a future challenge in nanotechnology. For example, recently we demonstrated the enhanced generation of photocurrents by the programmed organization of semiconductor NPs and relay units on a DNA scaffold.22 Here we wish to report on the use of DNA chains, synthesized by the rolling circle amplification, as templates for the activation of an enzyme cascade, and as a scaffold for the binding of a Au NP-functionalized enzyme that * Corresponding author,
[email protected]. 10.1021/nl900302z CCC: $40.75 Published on Web 03/26/2009
2009 American Chemical Society
biocatalyzes the synthesis of Au nanowires on the DNA. The two systems represent examples where chemical functionalities emerge from the ordered protein-DNA hybrid structures. Figure 1A depicts schematically the RCA synthesis of the DNA template designed to activate the biocatalytic cascade, consisting of the two enzymes, glucose oxidase (GOx) and horseradish peroxidase (HRP). The circular DNA 1 includes two regions, I and II, that generate the sequence-specific domains in the RCA products that selectively bind GOx and HRP. The domain III separates the domains I and II and yields a spacer nucleic acid sequence between the GOx and HRP binding domains in the RCA product. The RCA process was initiated by hybridization of a primer to 1 and the interaction of the ligated hybrid with polymerase and the oligonucleotide mixture dNTPs. Long DNA chains (up to 30 µm long strands) were synthesized. Figure 1B shows the AFM image of a typical DNA chain. The cross-section analysis of the chain indicates a height of ca. 0.6 nm, consistent with the reported height of single-stranded DNA.10 The formation of the RCA product was further confirmed by confocal microscopy imaging, Figure 1C. The dye 5-carboxyfluorescein N-succinimidyl ester (FAM) was covalently tethered to the amino-functionalized nucleic acid 2 that is complementary to domain I in the RCA product. Similarly, the amino-modified nucleic acid 3, complementary to the domain II of the RCA products, was functionalized
Figure 1. (A) Synthesis of sequence-specific programmed DNA strands using the rolling circle amplification (RCA) and the activation of an enzyme cascade by the spatial positioning of the two enzymes (GOx and HRP) on the DNA template. (B) AFM image and respective cross-section analysis of the RCA-synthesized DNA. (C) Confocal microscopy images of the DNA strands with the immobilized 2-modified FAM and 3-modified TAMRA on the domains I and II, respectively: (I) excitation at 488 nm, fluorescence is imaged at 505-525 nm (FAM emission); (II) excitation at 488 nm, fluorescence is imaged at 560-660 nm (TAMRA FRET emission). (D) AFM images of the DNA strands with the hybridized 2-GOx and 3-HRP, and the respective cross-section analysis of the structures. (E) Time-dependent formation of ABTS•- upon the activation of the biocatalytic GOx/HRP cascade, in the presence of glucose, 0.1 M under O2, and ABTS2-, 4 × 10-4 M, in 0.1 M phosphate buffer solution pH ) 7.4: (a) in the presence of the RCA-synthesized template, 1.5 × 10-12 M, and hybridized 2-GOx/3-HRP, each enzyme 0.3 × 10-9 M; (b) in the absence of the DNA template and the presence of 2-GOx/3-HRP, each 0.3 × 10-9 M; (c) in the presence of the foreign calf thymus DNA, 2.5 × 10-4 mg/mL, and 2-GOx/3-HRP, each 0.3 × 10-9 M.
with tetramethylrhodamine N-succinimidyl ester (TAMRA) and hybridized to the RCA chains. The hybridization of the dye-modified 2 with the RCA product allowed us to image the DNA chains by following their fluorescence (λex ) 488 nm, λem ) 505-525 nm), Figure 1C, panel I. Also, upon excitation of the system at 488 nm, the red fluorescence of TAMRA (λem ) 560-660 nm) was generated by the chains, Figure 1C, panel II. It should be noted that the hybridization of only the FAM-modified 2 with the RCA products did not yield any significant red fluorescence upon excitation at 488 nm, but resulted in the green fluorescence, upon photoexcitation at 488 nm. These results imply that, upon the hybridization of the FAM-functionalized 2 and TAMRAmodified 3 with the regions I and II of the RCA products, a fluorescence resonance energy transfer (FRET) process occurs. The enzymes GOx and HRP were then modified by the nucleic acids 2 and 3, respectively. The average loading of GOx with 2 corresponded to 1.5 and the nucleic acidfunctionalized GOx retained g90% of the native enzyme activity. Similarly, HRP was modified with the nucleic acid 3. The average loading of 3 on HRP was 3, and the activity of the enzyme was similar to that of the unmodified biocatalyst. The 2-functionalized GOx and 3-modified HRP were then hybridized with the domains I and II of the RCAsynthesized chains. Figure 1D shows the atomic force microscopy (AFM) image of the bienzyme/DNA hybrid Nano Lett., Vol. 9, No. 5, 2009
chains. The cross-section analysis of the chains reveals a height of ca. 5.5 nm (as compared to a height of 0.6 nm of the base DNA chains), consistent with the association of the proteins on the DNA scaffolds.10 The activity of the two enzymes, GOx and HRP, linked to the DNA scaffold was then examined by the activation of the bienzyme system with glucose, Figure 1A. The GOxmediated oxidation of glucose yields gluconic acid and H2O2. The latter product is the substrate of HRP, and the HRPmediated oxidation of 2,2′-azino-bis[3-ethylbenzthiazoline6-sulfonic-acid] (ABTS2-), to the colored product, ABTS•-, allows the probing of the enzyme cascade. Figure 1E, curve a, shows the time-dependent formation of ABTS•-, upon the addition of glucose to the system. The bienzyme cascade is effectively activated. For comparison, Figure 1E shows the activity of the two enzymes 2-GOx and 3-HRP (at a comparable concentration, 0.3 × 10-9 M), in the absence of the RCA products, curve b, or in the presence of the foreign calf thymus DNA, curve c. Evidently, no ABTS•- is formed. These results clearly demonstrate that the localization of the two enzymes on the DNA scaffold enables the communication between the two biocatalysts and the activation of the enzyme cascade. This effective activity of the two enzymes on the DNA scaffold is attributed to the proximity of the two enzymes on the DNA template. The H2O2 generated by GOx forms a high local concentration adjacent to the HRP, resulting in the effective activation of the second enzyme. It 2041
Figure 2. (A) Assembly of Au-NP-functionalized GOx on the RCA-synthesized template, and the biocatalytic enlargement of the NPs to Au nanowire structures. (B) Absorption spectra corresponding to the growth of the NPs on the GOx-DNA hybrid nanostructures. System was composed of DNA template 3 × 10-9 M, with hybridized Au-NP-functionalized 2-GOx, 7 × 10-7 M, glucose, 8 × 10-2 M, AuCl4-, 3.75 × 10-3 M, under oxygen, in 0.1 M phosphate buffer solution, pH ) 7.4. Spectra were recorded at time intervals of 1 min, and the absorbance reached saturation after 6 min. (C) AFM images of the resulting Au nanostructures: (I) a large 12 µm2 image; (II and III) images of different single wires and the respective cross-section analyses. (D) (I) Scanning electron microscopy image of the biocatlyticaly generated Au nanowires and (II) the enlargement of one domain of the wire. (III) Energy-dispersive spectrometry analysis of the resulting Au nanowire.
should be noted that the programmed activation of intracellular enzyme cascades originates from the precise structural and spatial localization of the coupled biocatalysts, and this provides the essence of the holy grail of “systems biology”. We believe that our system provides a simple model system for mimicking the complexity of natural systems. For the structures of 1, 2, and 3, and the detailed experimental conditions to prepare the 2- or 3-functionalized enzymes, see Supporting Information. The directed assembly of enzymes on the RCA products was further implemented to synthesize Au metallic nanowires. The biocatalytic growth of metallic nanoparticles was used to follow the activity of enzymes and to sense their products.23,24 Similarly, enzymes (such as glucose oxidase or alkaline phosphatase) were modified with Au NPs seeds, and the enzyme-NP conjugates were used as “catalytic inks” for the dip-pen nanolithographic patterning of biocatalytic templates that grow metallic nanowires.15 The major advantage of growing metallic nanowires on enzyme templates 2042
rests on the fact that the growth of the nanoparticles by the biocatalytic process introduces a self-inhibition path for the growth of the nanowires. When the enzymes are coated with the metal layer, the biocatalytic transformation is blocked, and the growth of the nanowires is terminated. Figure 2A shows the principle for growing the Au nanowires. The enzyme units were reacted with N-hydroxysuccinimide (NHS)-modified Au NPs (1.4 nm) to yield the Au NP-GOx hybrid. Then, the GOx-Au NPs hybrids were modified with DNA 2 and 3. The GOx/Au-NP/DNA units were further hybridized with the RCA products generated by the circular DNA 1. The resulting composite was then reacted with glucose, and the resulting H2O2 acted as reducing agent to reduce AuCl4- and enlarge the seed Au NPs. Figure 2B shows the time-dependent absorption spectra upon the treatment of the Au-NP-GOx/DNA composite with glucose, in the presence of AuCl4-. The plasmon absorbance in the range of 520-560 nm is intensified, and a red shift in the plasmon absorbance is observed as the reaction proceeds. Nano Lett., Vol. 9, No. 5, 2009
After ca. 6 min, the absorbance reaches a saturation value. These results are consistent with the biocatalytic growth and enlargement of the Au NPs seeds resulting in intensified absorbance values that are red shifted. The saturated absorbance is attributed to the coating of the enzyme with the enlarged NPs, a process that prohibits the biocatalytic generation of H2O2, and the further growth of the NPs. Figure 2C, panels I-III, shows the AFM image of numerous Au wires generated by this method (panel I) and typical images of single Au nanowires. The height of the nanowires corresponds to 40-60 nm, whereas the width of the wires is ca. 150 nm. Figure 2D (panels II and III) shows the scanning electron microscopy image of the nanowires. The wires consist of enlarged, and contacted, Au-NPs (diameter ca. 16 nm). The energy dispersive spectrum of the wire confirms that it is, indeed, composed of Au. For the detailed procedures for enlargement of the Au NPs and the synthesis of the nanowires, see Supporting Information. Further support that the biocatalytic growth of the Au-NPs, indeed, occurred, was obtained by comparing the Au-NP-GOx-functionalized DNA scaffolds before and after the biocatalytic enlargement of the NPs. While the height of the Au-NP-GOx-modified DNA scaffolds prior to the enlargement process corresponds to ca. 6.5 nm (see Figure 1S in Supporting Information), the biocatalytically enlarged Au nanowires exhibit a height that corresponds to ca. 50 nm. It should be noted that the majority of the nanowires are shorter (ca. 2-3 µm) and often wider than the bare DNA scaffolds generated by the RCA process or the enzyme-modified DNA scaffolds (ca. 5-8 µm). This might originate from the flexibility of the enzyme/ DNA scaffold that results in the bridging of the Au NPs associated with the flexible chains upon the biocatalytic growth of the NPs. To conclude, the present study has introduced the use of sequence-specific nucleic acids, generated by the rolling circle amplification (RCA) process, as a scaffold for the ordered assembly of enzymes into hybrid composites that reveal structure-derived functions. This was exemplified with the activation of an enzyme cascade and with the effective preparation of DNA templates for the synthesis of Au nanowires. These concepts could be further expanded to include multienzyme cascade systems and to synthesize, by means of enzymes, other nanowires consisting of organic polymers or semiconductor materials. Although the RCA process that leads to the generation of the DNA scaffold for the biocatalytic growth of metallic nanowires or for the activation of the two-enzyme cascade is simple, the concept suffers from a limitation. The flexibility of the single-stranded DNA scaffold might result in the folding of the chains and, eventually, the cross-linking of chain domains by the tethered
Nano Lett., Vol. 9, No. 5, 2009
nucleic acid-functionalized biocatalysts. This could be eliminated, in the future, by designing duplex DNA scaffolds, exhibiting enhanced rigidity for the synthesis of the metallic nanowires or for the activation of the enzymes cascade. Acknowledgment. This research is supported by the Israel Science Foundation, Converging Technologies Project. Supporting Information Available: Methods for the synthesis of the FAM- and TAMRA-modified nucleic acids, the procedure for preparation of the RCA product, the synthesis of the nucleic acid-functionalized enzymes, the modification of the nucleic acid-functionalized GOx/HRP cascade, and the procedure to biocatalytically grow the Au nanowires. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Seeman, N. C. Annu. ReV. Biophys. Biomol. Struct. 1998, 27, 225– 248. (2) Rinker, S.; Ke, Y.; Liu, Y.; Chhabra, R.; Yan, H. Nat. Nanotechnol. 2008, 3, 418–422. (3) Niemeyer, C. M. Curr. Opin. Chem. Biol. 2000, 4, 609–618. (4) Cohen, J. D.; Sadowski, J. P.; Dervan, P. B. Angew. Chem., Int. Ed. 2007, 46, 7956–7959. (5) Niemeyer, C. M.; Adler, M.; Pignataro, B.; Lenhert, S.; Gao, S.; Chi, L.; Fuchs, H.; Blohm, D. Nucleic Acid Res. 1999, 27, 4553–4561. (6) Li, H.; Park, S. H.; Reif, J. H.; LaBean, T. H.; Yan, H. J. Am. Chem. Soc. 2004, 126, 418–419. (7) Liu, Y.; Lin, C.; Li, H.; Yan, H. Angew. Chem., Int. Ed. 2005, 44, 4333–4338. (8) Weizmann, Y.; Braunschweig, A. B.; Wilner, O. I.; Cheglakov, Z.; Willner, I. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 5289–5294. (9) Beyer, S.; Nickels, P.; Simmel, F. C. Nano Lett. 2005, 5, 719–722. (10) Cheglakov, Z.; Weizmann, Y.; Braunschweig, A. B.; Wilner, O. I.; Willner, I. Angew. Chem., Int. Ed. 2008, 47, 126–130. (11) Keren, K.; Krueger, M.; Gilad, R.; Ben-Yoseph, G.; Sivan, U.; Braun, E. Science 2002, 297, 72–75. (12) Keren, K.; Berman, R. S.; Buchstab, E.; Sivan, U.; Braun, E. Science 2003, 302, 1380–1382. (13) Gazit, E. FEBS J. 2007, 274, 317–322. (14) LaBean, T. H.; Li, H. Nanotoday 2007, 2, 26–35. (15) Basnar, B.; Weizmann, Y.; Cheglakov, Z.; Willner, I. AdV. Mater. 2006, 18, 713–718. (16) Patolsky, F.; Weizmann, Y.; Lioubashevsky, O.; Willner, I. Angew. Chem., Int. Ed. 2002, 41, 2323–2327. (17) Gothelf, K. V.; LaBean, T. H. Org. Biomol. Chem. 2005, 3, 4023– 4037. (18) Willner, I.; Basnar, B.; Willner, B. FEBS J. 2007, 274, 302–309. (19) Park, S. H.; Lytton-Jean, A. K. R.; Lee, B.; Weigand, S.; Schatz, G. C.; Mirkin, C. A. Nature 2008, 451, 553–556. (20) Park, S. Y.; Yan, H.; Reif, J. H.; LaBean, T. H.; Finkelstein, G. Nanotechnology 2004, 15, S525–S527. (21) Deng, Z.; Tian, Y.; Lee, S. H.; Ribbe, A. E.; Mao, C. Angew. Chem., Int. Ed. 2005, 44, 3582–3585. (22) Tel-Vered, R.; Yehezkeli, O.; Yildiz, H. B.; Wilner, O. I.; Willner, I. Angew. Chem., Int. Ed. 2008, 47, 8272–8276. (23) Willner, I.; Baron, R.; Willner, B. AdV. Mater. 2006, 18, 1109–1120. (24) Zayats, M.; Baron, R.; Popov, I.; Willner, I. Nano Lett. 2005, 5, 21–25.
NL900302Z
2043
