Gold-Nanoparticle-Based Assay for Instantaneous Detection of

Mar 3, 2010 - Gold-Nanoparticle-Based Assay for Instantaneous Detection of Nuclear Hormone Receptor−Response Elements Interactions .... This approac...
3 downloads 8 Views 341KB Size
Download PDF
Anal. Chem. 2010, 82, 2759–2765

Gold-Nanoparticle-Based Assay for Instantaneous Detection of Nuclear Hormone Receptor-Response Elements Interactions Yen Nee Tan,† Xiaodi Su,*,† Edison T. Liu,‡ and Jane S. Thomsen‡ Institute of Material Research and Engineering, ASTAR (Agency for Science, Technology and Research), 3 Research Link, Singapore 117602, and Genome Institute of Singapore, ASTAR (Agency for Science, Technology and Research), 60 Biopolis Street, Singapore 138672 Gold nanoparticles (AuNPs) are widely used as colorimetric probes for biosensing, relying on their unique particle size-dependent and/or interparticle distancedependent extinction spectrum and solution color. Herein, we describe an AuNP-based colorimetric assay to detect binding interactions between nuclear hormone receptors and their corresponding DNA-binding elements, particularly the human estrogen receptors (ERr and ERβ) and their cognate estrogen response elements (EREs). We found that the protein-DNA (ER-ERE) complexes can stabilize citrate anion-capped AuNPs against salt-induced aggregation to a larger extent than the protein (ER) or the DNA (ERE) alone, due to their unique molecular size and charge properties that provide a strong electrosteric protection. Moreover, our results show that the extent of stabilization is sequence-dependent and can distinguish a single base variation in the ERE associated with minor changes in protein-DNA binding affinity. With this assay, many important parameters of protein-DNA binding events (e.g., sequence selectivity, distinct DNA binding properties of protein subtypes, binding stoichiometry, and sequence-independent transient binding) can be determined instantly without using labels, tedious sample preparations, and sophisticated instrumentation. These benefits, in particular the high-throughput potential, could enable this assay to become the assay of choice to complement conventional techniques for large scale characterization of protein-DNA interactions, a key aspect in biological research. Estrogen receptors (ERs) are DNA-binding proteins that belong to the nuclear receptor superfamily of transcriptional factors. ERs regulate estrogen gene expression by binding to specific DNA sequences known as estrogen response elements (EREs). Slight variations in the nucleotide composition of an ERE can affect the binding characteristics of ERs, which ultimately could affect the transcriptional rate of target genes. An important line of research in estrogen regulation of target gene expression is dependent on reliable techniques capable of measuring sequence* Corresponding author. E-mail: [email protected]. Tel: 65-68748420. Fax: 65-68720785. † Institute of Material Research and Engineering. ‡ Genome Institute of Singapore. 10.1021/ac9026498  2010 American Chemical Society Published on Web 03/03/2010

dependent ER-ERE interactions with high accuracy. The electrophoretic mobility shift assay (EMSA) is a commonly used technique to study protein-DNA interactions by measuring the electrophoretic mobility of protein, DNA, and (preformed) protein-DNA complexes in polyacrylamide gels.1 This technique involves lengthy separation procedures and uses labeled DNA for detection. Surface plasmon resonance spectroscopy (SPR) is a technique that measures biomolecular interactions on a solid-liquid interface.2,3 The SPR experiments, however, require specific instrumentation for measurement and involve complex surface chemistry for probe immobilization. Extensive optimizations are often needed to ensure that the immobilized probes are in a favorable orientation and conformation for target binding.4,5 Gold nanoparticles (AuNPs) have a distinct extinction spectrum from the collective excitation of surface electrons by light. They have been extensively used as a colorimetric sensing platform for biological analysis, relying on the unique interparticle distancedependent extinction spectrum and solution color.6-8 The key to the assay design is the control of particle dispersion and aggregation using specific biomolecular interactions and/or biological processes. To date, numerous assays have been developed for detecting a wide range of analytes (e.g., DNA,9-14 metal ions,15-19 (1) Garner, M. M.; Revzin, A. Nucleic Acids Res. 1981, 9, 3047–3060. (2) Homola, J. Anal. Bioanal. Chem. 2003, 377, 528–539. (3) Boozer, C.; Kim, G.; Cong, S.; Guan, H.; Londergan, T. Curr. Opin. Biotechnol. 2006, 17, 400–405. (4) Yang, N.; Su, X.; Tjong, V.; Knoll, W. Biosens. Bioelectron. 2007, 22, 2700– 2706. (5) Knoll, W.; Zizlsperger, M.; Liebermann, T.; Arnold, S.; Badia, A.; Liley, M.; Piscevic, D.; Schmitt, F.; Spinke, J. Colloids Surf., A 2000, 161, 115–137. (6) Liu, J.; Cao, Z.; Lu, Y. Chem. Rev. 2009, 109, 1948–1998. (7) Zhao, W.; Brook, M. A.; Li, Y. ChemBioChem 2008, 9, 2363–2371. (8) Thaxton, C. S.; Georganopoulou, D. G.; Mirkin, C. A. Clin. Chim. Acta 2006, 363, 120–126. (9) Elghanian, R.; Storhoff, J. J.; Mucic, R. C.; Letsinger, R. L.; Mirkin, C. A. Science 1997, 277, 1078–1081. (10) Li, H.; Rothberg, L. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 14036–14039. (11) Li, H.; Rothberg, L. J. J. Am. Chem. Soc. 2004, 126, 10958–10961. (12) Kanjanawarut, R.; Su, X. D. Anal. Chem. 2009, 81, 6122–6129. (13) Su, X. D.; Kanjanawarut, R. ACS Nano 2009, 3, 2751–2759. (14) Sato, K.; Hosokawa, K.; Maeda, M. J. Am. Chem. Soc. 2003, 125, 8102– 8103. (15) Lee, J.-S.; Han, M. S.; Mirkin, C. A. Angew. Chem., Int. Ed 2007, 46, 4093– 4096. (16) Xue, X. J.; Wang, F.; Liu, X. G. J. Am. Chem. Soc. 2008, 130, 3244–3245. (17) Slocik, J. M.; Zabinski, J. S., Jr.; Phillips, D. M.; Naik, R. R. Small 2008, 4, 548–551. (18) Lee, J. H.; Wang, Z.; Liu, J.; Lu, Y. J. Am. Chem. Soc. 2008, 130, 14217– 14226.

Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

2759

and small molecules20-24) and for studying various enzymatic reactions.25-30 These assays utilize either interparticle crosslinking or noncross-linking aggregation mechanisms, with bioconjugated and/or unmodified AuNPs as colorimetric reporters. A detailed description of the versatile assay designs can be found in a review article by Zhao et al.7 In this study, we report a new application of AuNPs (unmodified particles) for measuring interactions between nuclear hormone receptors and their response elements, exemplified by the human recombinant estrogen receptors (R and β subtypes) and their cognate estrogen response elements (ERE). The assay was designed based on our discovery that the protein-DNA complexes (ER-ERE) can stabilize citrate anion-capped AuNPs against salt-induced aggregation to a larger extent compared to either the protein (ER) or the double-stranded DNA (ERE) alone. Zeta potential measurements were conducted to reveal that it is the large molecular size and rich charge of the ER-ERE complexes that provide efficient electrosteric stabilization. Using ERR, ERβ, and their EREs carrying different nucleotide mutations, we have demonstrated that our assay is capable of measuring many important parameters of protein-DNA binding events (e.g., sequence selectivity with single base resolution, binding stoichiometry, distinct binding properties of protein subtypes, and sequence-independent transient binding) without using labels, tedious sample preparations, and sophisticated instrumentation. The simplicity of this assay would allow the setup of a high-throughput investigation of protein-DNA interactions with small sample volumes. Unlike other unmodified AuNP-based assays (e.g., DNA hybridization10-13 and enzymatic processes25-28) that are designed based on the controlled loss/gain of electrostatic forces, the current assay uses both the electrostatic and steric stabilization forces (electrosteric forces) exerted by the analyte to control particle stability. This sensing principle should be applicable to those protein-DNA binding systems where the resulted complexes have significantly larger molecular size (when protein binds DNA at higher orders, like dimer) and distinguishable charge properties relative to DNA and protein before forming complexes. RESULTS AND DISCUSSION Determination of ERr-DNA Complex Formation. Figure 1 is a schematic illustration showing how ER, ERE, and the ER-ERE complex modulate the AuNPs’ stability differently. We first investigated the ability of this assay to detect complex (19) Wang, H.; Wang, Y.; Jin, J.; Yang, R. Anal. Chem. 2008, 80, 9021–9028. (20) Hurst, S. J.; Han, M. S.; Lytton-Jean, A. K. R.; Mirkin, C. A. Anal. Chem. 2007, 79, 7201–7205. (21) Han, M. S.; Lytton-Jean, A. K. R.; Mirkin, C. A. J. Am. Chem. Soc. 2006, 128, 4954–4955. (22) Liu, J.; Lu, Y. Angew. Chem., Int. Ed. 2006, 45, 90–94. (23) Chen, S. J.; Huang, Y. F.; Huang, C. C.; Lee, K. H.; Lin, Z. H.; Chang, H. T. Biosens. Bioelectron. 2008, 23, 1749–1753. (24) Song, G.; Chen, C.; Qu, X.; Miyoshi, D.; Ren, J.; Sugimoto, N. Adv. Mater. 2008, 20, 706–710. (25) Wei, H.; Chen, C.; Han, B.; Wang, E. Anal. Chem. 2008, 80, 7051–7055. (26) Zhao, W.; Chiuman, W.; Lam, J. C. F.; Brook, M. A.; Li, Y. Chem. Commun. 2007, 3729–3731. (27) Oishi, J.; Asami, Y.; Mori, T.; Kang, J.-H.; Niidome, T.; Katayama, Y. Biomacromolecules 2008, 9, 2301–2308. (28) Choi, Y.; Ho, N.-H.; Tung, C.-H. Angew. Chem., Int. Ed. 2007, 119, 721– 723. (29) Zhao, W.; Ali, M. M.; Aguirre, S. D.; Brook, M. A.; Li, Y. Anal. Chem. 2008, 80, 8431–8437. (30) Wang, Z. D.; Lee, J-. H.; Lu, Y. Adv. Mater. 2008, 20, 3263–3267.

2760

Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

Figure 1. Schematic illustration of AuNPs’ stability against saltinduced aggregation in the presence of double stranded DNA, protein, and protein-DNA complex, respectively.

Table 1. Estrogen Response Elements Sequences names (denoted as) wild type (wtERE or vit) mutant (mutERE) pS2 ERE (pS2) scrambled DNA (scrDNA)

sequences of the sense strand (5′ to 3′) GTCCAAAGTCAGGTCACAGTGACCTGATCAAAGT GTCCAAAGTCAGTTCACAGTGATCTGATCAAAGT GTCCAAAGTCAGGTCACAGTGGCCTGATCAAAGT GTCCAAAGTCAATCGCCAGCACGATGATCAAAGT

formation of ERR with its wild-type ERE consensus sequence (wtERE). This wtERE has the optimal ERE core binding sequence (GGTCAnnnTGACC, n ) spacer nucleotides) (Table 1) and is, therefore, bound to the receptor with high affinity.31,32 Figure 2A shows the UV-vis adsorption spectra and color photographs of AuNP solutions in the presence of wtERE, ERR, and their complex (ERR-wtERE, preincubated at a ratio of 2:1), respectively, recorded at 10 min upon mixing with KCl (final concentration 20 mM). With reference to the stable AuNP solutions in red (no salt or biological samples added) that have a sharp surface plasmon peak at 525 nm, all three solutions with salt and biological samples underwent distinct but different degrees of aggregation, characterized by the color change associated with the appearance of adsorption at a longer wavelength (i.e., 650 nm) and the decrease of absorbance at 525 nm in their UV-vis adsorption spectra. The absorbance ratio at these two signature wavelengths (A650/A525) was used to quantify the aggregation extent and was plotted against time to reveal the aggregation kinetics. From the aggregation kinetics in Figure 2B, we observed that AuNPs in the presence of ERR-wtERE protein-DNA complex underwent the least and/or slowest aggregation upon salt addition, followed by ERR protein and wtERE DNA. The corresponding colors of these solutions sampled at 10 min (inset of Figure 2A) showed the least color change from red to deep red for the solution containing the ERR-wtERE complex mixture and more vigorous color change to blue and gray for ERR- and ERE-containing solutions, respec(31) Muramatsu, M.; Inoue, S. Biochem. Biophys. Res. Commun. 2000, 270, 1–10. (32) Klinge, C. M. Nucleic Acids Res. 2001, 29, 2905–291.

Figure 2. (A) UV-visible adsorption spectra and color photographs of AuNPs exposed to wtERE (50 nM), ERR (100 nM), and ERR-wtERE complex (preincubation of 100 nM ERR and 50 nM wtERE), taken at 10 min upon salt addition (final KCl concentration is 20 mM). Result for a stable citrate-anion capped AuNP solution without addition of salt is shown as a reference. (B) Aggregation kinetics of the corresponding AuNP solutions.

tively. A negative control experiment was carried out with bovine serum albumin (BSA) that has no specific binding to wtERE. No distinct difference in particle stability was observed upon salt addition to the AuNPs containing BSA or BSA/wtERE mixtures, respectively (Figure S1, Supporting Information). These data further confirm that sequence-dependent protein-DNA complex formation (ERR-ERE) is responsible for the higher particle stability. With zeta potential experiments (measuring the surface charge properties) as support, our explanation for the associated changes in aggregation are as follows: The wtERE (which is a 34 bp dsDNA) is unable to protect AuNPs due to its double helical structure which exposes highly charged repulsive phosphate backbones and insulates DNA bases from adsorption on the AuNP surface.10,11,13 Consequently, no electrostatic protection is observed following addition of salt. The zeta potential measured for the AuNP solution in gray color (-11.29 ± 5.65 mV) is much lower than the well dispersed AuNPs in the absence of salt (-32.82 ± 2.43 mV), which explains the low stability. In the case where ERR protein is the only constituent in the AuNP solution, the intermediate level of aggregation (blue color solution) may arise from both electrostatic and steric modulation. As ERR at the tested condition (pH 7.4) is slightly positively charged (isoelectric point of ERR is 8.333), the initial uptake of ERR would destabilize the negatively charged AuNPs due to charge neutralization, as similarly reported for other positively charged substances (e.g., peptides27,28). Moreover, the adsorption of this large 66 kDa protein may also provide a steric barrier to prevent the nanoparticles from crowding. It is well-known that some amino acids (e.g., cysteine (C), histidine (H), etc.) show strong binding ability to noble metal via their functional side groups (e.g., thiol or imidazole).34,35 The adsorption of ERR (13 C and 20 H out of 595 amino acids; see full amino acid sequence in Supporting Information) onto AuNPs to provide steric protection can then be rationalized. The mutual compensation between these two effects (electrostatic destabilization and steric stabilization) eventually leads to the intermediate particle stability observed. The relatively low surface charge (zeta potential -10.55 ± 4.90 mV) measured

for the ERR-coated AuNPs confirms the charge neutralization and that the steric protection is attributable to the immediate stability. In contrast, the low AuNP aggregation found when exposed to the ERR-wtERE complex could be attributable primarily to the larger molecular size of the complex (ERR binds ERE as a dimer33,36,37), providing a more effective steric protection compared with ERR. Additionally, the coating of ERR-wtERE complex on the AuNP surface (through the protein’s amino acid side groups) may also allow AuNPs to gain more electrostatic protection due to the presence of heavily charged wtERE in the complex. The higher negative charge density measured by the zeta potential (-24.04 ± 6.45 mV) confirms this speculation. The double protection (electro-static and -steric) could be attributable to the significantly improved salt stability exerted by the ERR-wtERE complex. This principle should be valid for those DNA binding proteins that are not heavily charged and bind to their DNA sites at high orders, such that the complexes are always much larger in molecular size and richer in charges. Determination of ERr-DNA Binding Stoichiometry and Sequence Specificity. To determine the binding stoichiometry of ERR-wtERE complex formation, we incubated a fixed amount of ERR (100 nM) with varied amount of wtERE (25, 50, and 100 nM) before being subjected to testing. The kinetic result (Figure 3A) reveals that the particle stability was low when very little amount of ERE was used (e.g., 25 nM, insufficient to complex with all the 100 nM ERR). When the ERE amount increased (i.e., 50 nM and above), the particles’ stability increased and was maximized. This confirms the notion that the amount of the protein-DNA complex is associated with AuNPs’ stability and the inhibition of nanoparticle aggregation. In addition, that AuNPs’ stability maximized at an ERR-wtERE molar ratio of 2:1 (100 nM: 50 nM) is consistent with the fact that ERR binds to the wtERE as a dimer, as characterized using SPR33,36 and fluorescent anisotropy techniques.37 Increasing the wtERE concentration to 100 nM did not result in further improvement in AuNPs’ stability probably because no further ERR was available to form more complexes. This approach to unravel protein-DNA binding stoichiometry is more convenient than using SPR spectroscopy Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

2761

Figure 3. (A) Aggregation kinetics of AuNPs exposed to an ERR-wtERE incubation mixture of 100 nM ERR with wtERE of 25, 50, or 100 nM. (B) Aggregation kinetics of AuNPs in the presence of (a) ERR-wtERE, (b) ERR-mutERE, (c) ERR-scrDNA, and (d) ERR (control). All the complexes were preincubated at ERR-ERE ratio of 100:50 (nM/nM) before being subjected to colorimetric testing. (C) UV spectrum and color photograph of the corresponding AuNP solutions in (B), taken at 15 min upon salt addition. KCl (20 mM, final concentration) was used to trigger particle aggregation in all experiments.

that relies on titration of a known amount of immobilized DNA with increasing protein concentrations to find the amount of protein that saturates the DNA. 2762

Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

In the following experiments, we further demonstrated that this AuNP assay could be used to screen for the impact of sequence variation on protein-DNA binding affinity. Besides the wtERE that contains a perfect core ERE sequence, a mutant ERE (mutERE) with two base substitutions in the core and a negative control DNA (scrDNA) with the entire core sequence randomly scrambled were used (Table 1). The mutERE is known to have a very low protein binding affinity compared to wtERE, whereas the scrDNA has no sequence-dependent binding to the receptor but exerts weak electrostatic contact.32,33,36 Each individual sequence (50 nM) was preincubated (for 30 min) with the receptor (100 nM). The use of an ERR-ERE ratio of 2:1 was to ensure the formation of fully saturated ERR-DNA complexes according to the binding stoichiometry previously determined. The colorimetric testing was then carried out for each of the samples in the AuNP solution. Figure 3B,C shows the aggregation kinetics and the endpoint (taken at 15 min) UV-vis adsorption spectra together with the solution color of AuNPs exposed to the ERR complexes with the respective DNA sequences (i.e., wtERE, mutERE, and scrDNA). ERR in the absence of target DNA was used as a control. We found that the degree of stabilization exerted by different ERR-DNA complexes is sequence-dependent, following the affinity order of ERR-wtERE > ERR-mutERE > ERR-scrDNA, as observed by the SPR technique.33,36 The correlation between ER-ERE complex affinity and AuNP stability can be reasoned as the higher the binding affinity; the more protein-DNA complex can be formed to exert electrosteric protection. It should be mentioned that the stability order is consistently observed throughout the kinetics measurement over 15 min. This implies that a shorter monitoring period (e.g., 1 min) could be applied for fast colorimetric detection, provided that the sampling time is the same for all the samples tested upon salt addition. Generality to Other Protein Subtype: ERβ-ERE Complex Formation, Stoichiometry, and Sequence Specificity. To demonstrate the generality of current assay, similar experiments were carried out for ERβ, another protein subtype that has a high degree of homology (96% amino acid identity)31 with ERR in their DNA binding domain, yet harbors a distinct DNA binding profile. Specifically estrogen-bound ERβ binds to consensus ERE sequence at a higher order than ERR with a weaker binding affinity.32,33,36,38 We first validated the current assay to detect the formation of ERβ-wtERE complex and to determine the binding ratio of ERβ to wtERE in forming fully saturated complex (Figure 4A). Results show that the formation of an ERβ-wtERE complex (preincubation of 200 nM ERβ with 25, 50, or 100 nM wtERE) is detectable by the improved particle stability, relative to the case with ERβ only. Moreover, at an ERβ-wtERE ratio of 4:1 (200 nM ERβ and 50 nM wtERE), particles’ stability maximized, indicating that ERβ was completely consumed at this condition. This result (33) Teh, H. F.; Peh, W. Y. X.; Su, X.; Thomsen, J. S. Biochemistry 2007, 46, 2127–2135. (34) Mandal, S.; Gole, A.; Lala, N.; Gonnade, R.; Ganvir, V.; Sastry, M. Langmuir 2001, 17, 6262–6268. (35) Liedberg, B.; Carlsson, C.; Lundstrom, I. J. Colloid Interface Sci. 1987, 120, 64–75. (36) Su, X.; Lin, C.-Y.; O’Shea, S. J.; Teh, H. F.; Peh, W. Y. X.; Thomsen, J. S. Anal. Chem. 2006, 78, 5552–5558. (37) Boyer, M.; Poujol, N.; Margeat, E.; Royer, C. A. Nucleic Acids Res. 2000, 28, 2494–2502. (38) Margeat, E.; Bourdoncle, A.; Margueron, R.; Poujol, N.; Cavaille`s, V.; Royer, C. J. Mol. Biol. 2003, 326, 77–92.

Figure 4. (A) Aggregation kinetics of AuNPs exposed to an ERβ-wtERE incubation mixture of 200 nM ERβ with wtERE of 0, 25, 50, or 100 nM. (B) Aggregation kinetics of AuNPs in the presence of (a) ERβ-wtERE, (b) ERβ-mutERE, (c) ERβ-scrDNA, and (d) ERβ (control). All the complexes were preincubated at ERβ-ERE ratio of 200:50 (nM/nM) before being subjected to colorimetric testing. (C) UV spectrum and color photograph of the corresponding AuNP solutions in (B), taken at 15 min upon salt addition. KCl (20 mM, final concentration) was used to trigger particle aggregation in all experiments.

suggests that ERβ binds to wtERE as a tetramer. The fact that ERβ binds wtERE at a higher order than ERR (dimer) has been previously reported using SPR spectroscopy33,36 and fluorescence

anisotropy38 and validates the utility of this AuNP-based colorimetric assay. Thisexperimentallydeterminedbindingratio(4:1ofERβ-wtERE) was then used as an incubation condition in the next series of experiments, where the sequence impact on ERβ-DNA interactions (involving the same set of DNA as used in the ERR experiments, i.e., wtERE, mutERE, and scrERE) was investigated. As expected, AuNPs protected by ERβ-DNA complexes are more stable than those protected by ERβ alone (Figure 4B,C). The stabilization effect is in the order of ERβ-wtERE > ERβ-mutERE > ERβ-scrDNA, corresponding well with the results obtained through the SPR measurement.33,36 This result, together with that for ERR, evinced that our newly developed assay is reliable and sensitive to determine the relative binding affinities of a protein to different DNA sequences. The fact that the incubation mixtures of ERR and ERβ with the scrambled DNA (no specific core sequence for the receptors) can stabilize the nanoparticles (curves c in Figure 3B and Figure 4B) better than the receptors alone proves that this assay is capable of measuring weak and transient protein-DNA binding, which is usually difficult to detect in EMSA.1,39 Determination of Sequence Specificity with Single BasePair Resolution. To further challenge the capability of the current assay to detect subtle differences in protein binding affinity introduced by a single base substitution, we extended the study to another natural-occurring ERE sequence (i.e., human pS2 ERE), which differs from the wtERE by only one base pair (bold, see Table 1) in its core sequence.40,41 Using solid-liquid phase SPR measurements, we have determined that the binding affinity of ERR to pS2 ERE is only 18% less than its binding to wtERE.33 This difference is much smaller than that seen using the mutERE (∼60%) containing a two-base pair substitution.33,36 Using the previous aggregation condition (20 mM KCl) determined for ERE sequences bearing more base substitutions, we were unable to differentiate the ERR binding affinity between the very similar wtERE and pS2 ERE sequences (Figure 5A). However, when the final salt concentration from 20 to 50 mM KCl was increased to provide a more stringent condition for AuNP aggregation, a difference was seen against the wild-type ERE sequence (Figure 5B). The particle solution containing the ERR-pS2 ERE incubation mixture is 19% less stable than that containing ERR-wtERE, as measured by the absorbance difference at wavelength A650. These results demonstrate that upon a proper optimization of assay conditions, the current AuNP-based colorimetric assay is amendable for the detection of protein-DNA binding specificity up to a single nucleotide resolution. To address the concern of the compatibility of the complex formation and colorimetric detection conditions, we have conducted solid-liquid phase SPR experiments (Figure S2 in the Supporting Information). The SPR results confirm that ERR-DNA complexes performed under physiological salt conditions (80 mM KCl; through multiple hydrogen bonds and van der Waals forces) remain stable when exposed to the diluted buffer containing a low concentration of KCl (20 or 50 mM KCl). (39) Petz, L. N.; Nardulli, A. M. Mol. Endocrinol. 2000, 14, 972–985. (40) Hyder, S. M.; Chiappetta, C.; Stancel, G. M. Biochem. Pharmacol. 1999, 57, 597–601. (41) Loven, M. A.; Wood, J. R.; Nardulli, A. M. Mol. Cell. Endocrinol. 2001, 181, 151–163.

Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

2763

Figure 5. UV-vis spectrum and color photograph of AuNPs exposed to ERR-wtERE and ERR-pS2 incubation mixtures at ERR-ERE ratio of 2:1. A final salt concentration of (A) 20 mM and (B) 50 mM KCl was used to induce particle aggregation.

CONCLUSION We have developed an AuNP-based colorimetric assay for studying nuclear hormone receptor-DNA interactions, exemplified by human estrogen receptors and their classical estrogen response elements. With this assay, one can measure protein-DNA complex formation, screen nucleotide composition impact on binding affinity with single base resolution, determine binding stoichiometry, and measure sequence-independent transient binding. The use of unmodified AuNPs and noncrosslinking aggregation enables a quick display of assay results with minimal sample preparation. The control of particle aggregation/dispersion through electrosteric effect of biopolymers should be applicable for other protein-DNA binding systems where the resulted complexes possess significantly larger molecular size and/or distinguishable charge properties relative to their individual counterparts (i.e., DNA or protein) before forming complexes. Although this assay does not support realtime measurement of protein-DNA binding kinetics, the benefits of simplicity, fast detection, and low cost screening of relative affinity between purified proteins and a large amount of DNA sequences could enable this assay to complement the conventional technique for high throughput screening applications. EXPERIMENTAL SECTION Materials. HAuCl4 · 3H2O (99.99%) and trisodium citrate dihydrate (99.9%) were obtained from Aldrich Pte Ltd. Purified recombinant human estrogen receptors (ERR and ERβ) were purchased from Pan Vera Corp (Madison, WI). Estrogen response element (ERE) oligos used in this study (see the sense strand sequences in Table 1) were purchased from Sigma Pte Ltd. (Singapore). Prior to use, the sense and antisense strands of DNAs were annealed in buffer solution (10 mM Tris-HCl, pH 7.4; 100 mM NaCl; 1 mM EDTA) to form double-stranded ERE. All chemicals and materials were used as received without further purification. Ultrapure water (18 MΩ, prepared from Millipore Elix 3 purification system) was used as the solvent unless indicated otherwise. AuNP Preparation. Gold nanoparticles (AuNPs) of 13 nm in diameter were prepared by the citrate reduction of HAuCl4. 2764

Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

An aqueous solution of sodium citrate (5 mL, 40 mM) was added rapidly to a boiling solution (100 °C) of HAuCl4 (50 mL, 1 mM). Within several minutes, the color of the solution changed from pale yellow to red. The mixture was allowed to heat under reflux for 30 min to ensure complete reduction. The stirring was continued for an additional 15 min after removing the heating mantle and allowed to cool to room temperature. Colorimetric Assay Procedure. The assay was performed by first incubating ERR and ERE in a molar ratio of 2:1 at room temperature for 30 min, in 10 mM Tris-HCl buffer solution (pH 7.4), containing 80 mM KCl, 0.15 mM EDTA, 0.3 mM dithiothreitol (DTT), and 1% of glycerol. Twenty-five microliters of the complex solutions was then mixed with 75 µL of the as-synthesized AuNPs to make up a final KCl concentration of 20 mM. The final concentrations of ERR and ERE in the AuNP solution mixture were 100 nM and 50 nM, respectively. Similar experiments were carried out to study ERβ-ERE interactions, except that they were preincubated in a molar ratio of 4:1 to form complexes in the same buffer used for ERR studies. Twenty-five microliters of the complex solutions were then mixed with 75 µL of the as-synthesized AuNPs to make up a final concentration of 20 mM KCl, 200 nM ERβ, and 50 nM ERE. Characterization. Clear flat bottom UV-transparent microplates (96 well; Corning Incorporated, USA) were used as a reaction carrier. A TECAN infinite M200 plate reader (Tecan Trading AG, Switzerland) was used to measure the UV-vis adsorption spectrum; a normal digital camera was used to record the color of the solution. Zeta potential measurements were performed on citrate-ion capped AuNPs in water and wtERE-, ERR-, and ERR-ERE complex-coated AuNPs in buffer solution containing 20 mM KCl, using a ZETA PLUS zeta potential analyzer (Brookhaven Instruments, USA). ACKNOWLEDGMENT We would like to acknowledge The Agency for Science, Technology and Research (A*STAR), Singapore, for the financial support under the Grant CCOG01-005-2008. This work was partially supported by the EC sixth framework program Grant

CRESCENDO (FP6-018652). We also thank Dr. Ralf Jauch (Genome Institute of Singapore) for providing the amino acid sequence of ERR.

colorimetric detection buffer conditions. This material is available free of charge via the Internet at http://pubs.acs. org.

SUPPORTING INFORMATION AVAILABLE Negative control results with bovine serum albumin (BSA), amino acids sequence of human ERR, and SPR measurement of the compatibility of ERR-wtERE complex formation and

Received for review November 19, 2009. Accepted February 17, 2010. AC9026498

Analytical Chemistry, Vol. 82, No. 7, April 1, 2010

2765