Förster Resonance Energy Transfer-Based Biosensing Platform with

Letter pubs.acs.org/ac

Förster Resonance Energy Transfer-Based Biosensing Platform with Ultrasmall Silver Nanoclusters as Energy Acceptors Yan Xiao,† Fan Shu,† Kwok-Yin Wong,‡ and Zhihong Liu*,† †

Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education), College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei, 430072 China ‡ Department of Applied Biology and Chemical Technology, The Hong Kong Polytechnic University, Hunghom, Kowloon, Hong Kong, China S Supporting Information *

ABSTRACT: We studied the energy transfer (ET) property of ultrasmall Ag nanoclusters (Ag NCs) and exploited its biosensing application for the first time. A hybridized DNA duplex model was designed to study the energy transfer process from fluorescent energy donors to Ag NCs. By changing the DNA duplex model and the number of hybridized pairs, the separation distance between the energy donor and Ag NCs was adjusted to investigate the distance dependence and possible mechanisms involved in the ET process, which was assigned to Förster resonance energy transfer (FRET). Using Ag NCs with different photophysical properties as energy acceptors, FRET-based biosensing platforms with two different energy donors were constructed utilizing either the off−on or ratiometric fluorescence signaling. This study will provide the basis for understanding energy transfer properties of Ag NCs and bring to light the universal application of these properties in bio/chemo sensing.

N

containing 12 cytosine bases (C12) as the scaffold according to Dickson’s method5 (named as Str-A in Scheme 1; please see Table S1, Supporting Information, for the sequences of all DNA chains used). The formation of Ag NCs was verified and

oble-metal nanoclusters consisting of several to tens of metal atoms have attracted extensive interest in the past decade because of their unique optical and electronic properties.1 With their dimensions approaching the Fermi wavelength of electrons, the continuous density of states breaks up into discrete energy levels, which leads to molecule-like optical properties and endows the clusters with applicability in chemical/biological labeling and imaging.2 The research and application of Ag nanoclusters (Ag NCs) have recently been accelerated owing to the facile synthesis of Ag NCs using proper scaffolds such as the cytosine-rich DNA oligonucleotides.3 So far, several Ag NC-based biosensors have been built, which all employ the photoluminescence of the clusters.4 Ag NCs have proved to be promising luminescent labels due to their in situ generation on nucleic acid scaffolds and ultrasmall sizes. On the other hand, however, the energy accepting capability of Ag NCs has not yet been illustrated and exploited in bio/chemo sensing, which could have been reasonably expected because of the existence of discrete energy levels. Herein, we report the first fundamental study on energy transfer (ET) from fluorescent energy donors to Ag NCs generated on DNA scaffolds and the construction of ET-based biosensing platforms using Ag NCs as the energy acceptors, aiming at revealing the new photophysical property of Ag NCs and expanding their application in biosensing as well. We first investigated the energy transfer from a fluorescent dye to Ag NCs, which were prepared using an ssDNA chain © 2013 American Chemical Society

Scheme 1. Schematic Illustration of the Construction of DNA Duplex−Ag NCs Complexes and the Energy Transfer in DNA Duplex (1−4)−Ag NCs Complexes

Received: July 12, 2013 Accepted: August 28, 2013 Published: August 28, 2013 8493

dx.doi.org/10.1021/ac402125g | Anal. Chem. 2013, 85, 8493−8497

Analytical Chemistry

Letter

characterized with photoluminescence, UV−vis spectra, and transmission electron microscopy (TEM) images. The TEM observation revealed that the as-prepared Ag NCs had an average size of 1.89 ± 0.35 nm and good dispersibility in aqueous solution (Figure S1a, Supporting Information). As further evidence, the UV−vis spectrum exhibited three absorption peaks at 440, 560, and 650 nm (Figure S1b, Supporting Information), which was consistent with the literature results indicating the generation of ultrasmall metal nanoclusters.4b,5 In addition, the silver clusters stabilized by the DNA strand showed maximal fluorescence excitation and emission at 560 and 610 nm, respectively (Figure S1b, Supporting Information). The fairly large absorption of Ag NCs and the excitability at 560 nm implies the possibility to use Ag NCs as energy acceptors of proper donors. To explore the feasibility, we designed a hybridized DNA duplex model to study the energy transfer process (Scheme 1, the left column). In consideration of the spectral overlap of the energy donor−acceptor pair, we chose the organic dye 6-carboxy-4′,5′-dichloro-2′,7′-dimethoxyfluorescein (JOE) as the energy donor. The dye has a maximum emission around 555 nm that matches well with the excitation wavelength of Ag NCs (Figure S2, Supporting Information). The energy donor was tagged at the end of another ssDNA (Str-D in the Scheme). In the presence of a linker DNA strand (Str-L), which contains two segments that hybridize with Str-A (the red part) and Str-D (the blue part), respectively, the JOE dye and Ag NCs were brought to close proximity. In this model, Ag NCs were produced in situ after the hybridization of nucleotides. In an initial attempt using the configuration of Duplex1 (Scheme 1, the uppermost one of the right column), it was observed that the emission of JOE was quenched upon the formation of Ag NCs (Figure 1a), while the control experiment in which no reducing agent (NaHB4) was added indicated that the donor luminescence was not significantly quenched by silver ions. In order to further investigate the distance dependence and possible mechanisms involved in the ET process, we designed three more DNA duplex configurations (the right column of Scheme 1). The separation distance between the energy donor and Ag NCs was adjusted by changing the DNA duplex model and the number of hybridized pairs. In these models, the position of the C12 strand and the dye was separated by 10, 20, and 30 base pairs in the hybridization complexes. According to the facts that doublehelical DNA is generally considered as a rigid-rod structure and that 3 pairs of nucleotides account for approximately 1.0 nm, the distances between the energy donor and acceptor were estimated to be 3.4, 6.8, and 10.2 nm, respectively. Typically, there are three possible mechanisms involved in the energy transfer from organic fluorophores to nanosized structures, which are photoinduced electron transfer (PET) that requires contact of electron cloud, Förster resonance energy transfer (FRET), and metal surface energy transfer (SET).6 Our results showed that the quenching of donor emission occurred at distances as large as 3.4 and 6.8 nm (Figure 1b,c), which is strong evidence of the existence of nonradiative through-space energy transfer (FRET or SET). In some ET systems where metal nanoparticles or films act as energy acceptors, the ET process was attributed to SET,7 which is a dipole−surface resonance mechanism and is effective at distances far larger than 100 Å. Our experiments revealed that the energy transfer no longer occurred (or was totally ineffective) when the donor-

Figure 1. (a−d) The fluorescence spectra of DNA duplexes ((a) DNA duplex 1, (b) DNA duplex 2, (c) DNA duplex 3, (d) DNA duplex 4) before (solid line) and after (dash line) the generation of Ag NCs. Excitation wavelength: 529 nm. (e) The quenching efficiency of JOE emission by Ag NCs in the four duplex models.

to-acceptor distance was increased to 10.2 nm (Figure 1d), which is typically the limitation of detectable distance for FRET. As shown in Figure 1e, the quenching efficiency of JOE emission by Ag NCs in the four duplex models is highly distance dependent. According to the Förster theory, the energy transfer efficiency of FRET exhibits the 1/r6 dependence: E = R 0 6/(R 0 6 + r 6)

(1)

R 0 6 = 8.79 × 10−25κ 2n−4 ΦDJDA

(2)

where R0 is the Förster distance, r is the distance between the donor and the acceptor, κ2 is the dipole orientation factor, n is the refractive index of the medium, ΦD is the quantum yield of the donor, and JDA is the spectral overlap integral between the normalized donor emission and the acceptor extinction coefficient.6d Whereas for the dipole−surface energy transfer model, the energy transfer efficiency of SET exhibits the 1/r4 dependence: E = d0 4 /(d0 4 + r 4)

(3)

d0 4 = 0.225c 3ΦD/ωD2ωf k f

(4)

where c is the velocity of light in vacuum, ωD is the frequency of the donor electronic transition, ωf and kf are the Fermi frequency and Fermi wavevector of the metal, respectively.6c,8 Using eqs 2 and 4, the Förster radius (R0) and the SET radius (d0) for the JOE−Ag NCs system were calculated to be 4.5 and 7.9 nm, respectively. We then plotted energy transfer efficiency (which is represented with the quenching efficiency of the donor emission) as a function of separation distance and 8494

dx.doi.org/10.1021/ac402125g | Anal. Chem. 2013, 85, 8493−8497

Analytical Chemistry

Letter

are of great importance for clinical diagnostics and drug discovery. Upon the catalytic cleavage, the ssDNA strand functioning as the scaffold of Ag NCs would be broken into small fragments, as the energy acceptors are no longer connected with the donors resulting in the blocking of the energy transfer process (Figure 2a). The hydrolysis reaction

compared it with the theoretical curves for FRET (eq 1) and SET (eq 3) (Figure S3, Supporting Information). The comparison indicated better agreement with a Förster energy transfer mechanism for the JOE−Ag NCs pair. The difference between experimental and predicted energy transfer efficiency may be attributed to errors of the estimation of the donor-toacceptor distance and less effective isotropic distribution of electronic dipoles of energy donor. On the basis of these experimental facts, we can primarily assign the energy transfer from the fluorophore to Ag NCs to Förster resonance energy transfer. To further confirm our deduction, we performed fluorescence lifetime measurements of the DNA duplexes−Ag NCs complexes and control samples (Figure S4, Table S2, Supporting Information). The fluorescence lifetime of the energy donor was decreased by Ag NCs when the donor and the acceptor was in close proximity (in DNA duplex 1), while no obvious change in lifetime was observed when the distance was increased to 10.2 nm (in DNA duplex 4). The lifetime results are consistent with the steady-state fluorescence measurements. On the other hand, SET originates from the interaction of the electromagnetic field of the donor dipole with the nearly free conduction electrons of the accepting metal species.8 In an SET process, the donor dipole interacts with the electronic continuum levels of the metallic system. Since the ultrasmall Ag NCs behave as multielectron artificial atoms with high polarizability9 and the continuous density of states breaks up into discrete energy levels, they are less likely to provide such coupling interactions. While the energy transfer from the fluorophore to Ag NCs has been confirmed, the emission of Ag NCs (at 610 nm) was not observed in the hybridization complexes (Figure 1a−d). In a FRET pair when both the energy donor and the acceptor are luminescent materials, it is normally anticipated to observe both the decrease in donor emission and the appearance of acceptor emission when ET occurs. We attribute the absence of Ag NCs emission to the rather low photoluminescence quantum yield (QY) of Ag NCs.10 Similar phenomena are seen in FRET pairs where the luminescence efficiencies of the energy donor and acceptor differ significantly.11 The room-temperature fluorescence QY of the as-obtained silver clusters was estimated to be 0.11 by comparing with rhodamine B in methanol (with a standard QY of 0.71). Since the QY of Ag NCs is much lower than that of the JOE dye (around 0.75), it is hard to simultaneously detect their luminescence in a single measurement with the same detector. We found that the emission intensity of Ag NCs could be enhanced through increasing their concentration by using more precursors (silver ions), but an over excess of precursor was likely to cause the formation of large-size particles, as was reported in the literature.5 To simultaneously obtain the emission of the energy donor and acceptor in this FRET system, a practical way would be using energy donors with the QY levels comparable to Ag NCs (vide infra). Thus far, we have gained the profile of the energy accepting property of the ultrasmall Ag NCs. To examine the applicability of this property, we then constructed an ET-based sensing platform with S1 nuclease as the proof-of-concept target, using the model of DNA duplex 1 in Scheme 1. S1 nuclease is an endonuclease capable of hydrolyzing the phosphodiester linkages in a single-stranded nucleic acid backbone.12 DNA or RNA cleavage reactions catalyzed by nucleases are very important in biological processes involving the replication, repair, and recombination of DNA, molecular cloning, genotyping, and mapping.13 Hence, assays of nuclease activity

Figure 2. (a) Schematic illustration of S1 nuclease biosensor based on the energy transfer from JOE to Ag NCs using the DNA duplex 1 model. (b) The change of the fluorescence of DNA duplex 1−Ag NCs complex in the presence of varying concentrations (0, 1, 2, 5, 10, 20, 30 U/mL) of S1 nuclease. (c) The linear relationship between the relative fluorescence intensity (F − F0)/F0 and the concentration of S1 nuclease ranging from 1 to 30 U/mL. Excitation wavelength: 529 nm.

was confirmed by polyacrylamide gel electrophoresis analysis (Figure S5, Supporting Information). After the Str-A−Ag NCs complex was treated by nuclease, no obvious band for the DNA template of Ag NCs was observed, indicating the hydrolysis of the DNA scaffold in the presence of Ag NCs. With the introduction of increasing amount of S1 nuclease into the hybridization complex containing Ag NCs, the emission intensity of JOE at 555 nm recovered gradually (Figure 2b), which as expected, was a result of the weakened energy transfer from JOE to Ag NCs. The relative fluorescence intensity of JOE ((F − F0)/F0, in which F and F0 represent the emission intensity of JOE in the presence and in the absence of target molecules, respectively) was dependent on the concentration of nuclease, and a linear calibration was built up within 1−30 U/ mL (Figure 2c), enabling quantification of the target. Following the verification of the applicability of Ag NC-based energy transfer systems, we sought to further improve the sensing efficiency and validate the universality of Ag NCs as energy acceptors. As mentioned above, the JOE−Ag NCs pair did not present the typical dual emission signals. It is known that, in a FRET-based assay with both the donor and acceptor emission detectable, the ratiometric determination of donor-toacceptor emission can provide more reliable results and better performance. To achieve such ratiometric sensing, we proposed another FRET model using an energy donor with the QY level comparable to Ag NCs. Taking advantage of the adjustability of Ag NCs photophysics by using different ssDNA as templates, it is possible to adapt Ag NCs to different energy donors emitting at different regions. We therefore used another ssDNA (Str-A′, see Supporting Information for the sequence) as the scaffold for the synthesis of Ag NCs, which can be excited at around 320 nm (Figure S6, Supporting Information). Accordingly, we employed tryptophan, a fluorophore with a quantum yield of 0.13, and an emission band overlapping with the excitation of 8495

dx.doi.org/10.1021/ac402125g | Anal. Chem. 2013, 85, 8493−8497

Analytical Chemistry

Letter

ratio F355/F490 increased with S1 nuclease concentration, and a linear calibration was obtained with the target concentration ranging from 2 to 70 U/mL (Figure 3c). The limit of detection, at a signal-to-noise ratio of 3.0, was estimated to be 0.14 U/mL, which was quite competitively sensitive among the reported optical methods for S1 nuclease. In a control experiment where some other species including proteins, amino acids, and metal ions were added in place of S1 nuclease, the F355/F490 ratio showed no obvious change (Figure S10, Supporting Information), which precluded possible influence of these species on the fluorescence ratio and confirmed that the ratiometric signal was a result of the enzymatic cleavage-induced alteration of the energy transfer efficiency. In summary, we have demonstrated the energy accepting capability of ultrasmall silver nanoclusters generated on ssDNA scaffolds. The quenching of the donor emission by Ag NCs is distance dependent and can be assigned to Förster resonance energy transfer. By using different ssDNA chains as the synthetic templates, Ag NCs with different photophysical properties can be produced with a diverse number of energy donors. Taking Ag NCs as energy acceptors, energy transferbased biosensing platforms can be constructed utilizing either the alteration of the donor emission or the ratio of donor-toacceptor emission. This work will provide the basis for future applications of the energy transfer properties of Ag NCs in bio/ chemo sensing.

Ag NCs (Figure S6, Supporting Information) as the energy donor. To realize our design, we prepared a short peptide chain (NH2−GWCWG−COOH) composed of two Trp residues as the energy donor and a Cys residue as the connection site for conjugation with the amine group tagged at the 5′ end of the Str-A′ (Figure 3a). After covalently linking the peptide with the

Figure 3. (a) Schematic illustration of S1 nuclease biosensor based on the tryptophan−Ag NCs energy transfer system. (b) The change of the fluorescence of tryptophan and Ag NCs in the presence of varying concentrations (0, 2, 10, 20, 30, 70 U/mL) of S1 nuclease. (c) The linear relationship between the ratios of the donor-to-acceptor emission and the concentration of S1 nuclease ranging from 2 to 70 U/mL. Excitation wavelength: 280 nm.

■

ASSOCIATED CONTENT

* Supporting Information S

Experimental details, DNA sequences, the spectral and TEM characterization of Ag nanoclusters, and the results of fluorescence lifetime measurements and control experiments. This material is available free of charge via the Internet at http://pubs.acs.org.

Str-A′ following an amine-to-sulfhydryl coupling protocol, Ag NCs were then produced in situ on the peptide−ssDNA complex with an average size of 1.78 ± 0.14 nm (Figure S7, Supporting Information). After the formation of Ag NCs, the fluorescence spectra of the peptide−ssDNA−Ag NCs complex clearly showed a dual-emissive signal (Figure S8, Supporting Information, solid line); i.e., the emission of the donor (at 355 nm) and the acceptor (at 490 nm) were observed when exciting the complex at the excitation wavelength of the donor (280 nm). One should note that the excitation at 280 nm might cause direct excitation of the acceptor through absorption by the DNA nucleobases followed by energy transfer to Ag NCs.14 In order to confirm the occurrence of energy transfer from tryptophan to Ag NCs, we investigated the fluorescence of the ssDNA−Ag NCs complex in the absence of the energy donor. It was found that the emission of the ssDNA−Ag NCs complex excited by 280 nm light was much lower than that of the peptide−ssDNA−Ag NCs complex under the same concentration of ssDNA−Ag NCs (Figure S8, dash line, Supporting Information), suggesting that direct excitation of acceptor was moderate and that the energy transfer from tryptophan to Ag NCs did occur. Furthermore, the FRET efficiency can be adjusted by changing the concentration of Ag NCs. At a fixed amount of the energy donor, the fluorescence intensity of the donor at 355 nm decreased while that of the acceptor at 490 nm increased gradually in accord with the increase of the Ag NCs concentration (Figure S9, Supporting Information). Subsequently, a ratiometric assay of S1 nuclease was performed with the tryptophan−Ag NCs energy transfer system (Figure 3a). The introduction of S1 nuclease resulted in the recovery of the donor emission at 355 nm and the decrease of the acceptor emission at 490 nm (Figure 3b). The fluorescence intensity

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel: 86-27-8721-7886. Fax: 8627-6875-4067. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors acknowledge the financial support from the National Natural Science Foundation of China (No. 21075094) and the National Basic Research Program of China (973 Program, No. 2011CB933600).

■

REFERENCES

(1) (a) Shang, L.; Dong, S.; Nienhaus, G. U. Nano Today 2011, 6, 401−418. (b) Wilcoxon, J. P.; Abrams, B. L. Chem. Soc. Rev. 2006, 35, 1162−1194. (c) Petty, J. T.; Fan, C.; Story, S. P.; Sengupta, B.; Iyer, A. S.; Prudowsky, Z.; Dickson, R. M. J. Phys. Chem. Lett. 2010, 1, 2524− 2529. (d) Diez, I.; Pusa, M.; Kulmala, S.; Jiang, H.; Walther, A.; Goldmann, A. S.; Muller, A. H.; Ikkala, O.; Ras, R. H. Angew. Chem., Int. Ed. 2009, 48, 2122−2125. (2) (a) Latorre, A.; Somoza, A. ChemBioChem 2012, 13, 951−958. (b) Choi, S.; Dickson, R. M.; Yu, J. Chem. Soc. Rev. 2012, 41, 1867− 1891. (c) Diez, I.; Ras, R. H. Nanoscale 2011, 3, 1963−1970. (d) Makarava, N.; Parfenov, A.; Baskakov, I. V. Biophys. J. 2005, 89, 572−580. (e) Muhammed, M. A.; Verma, P. K.; Pal, S. K.; Kumar, R. C.; Paul, S.; Omkumar, R. V.; Pradeep, T. Chem.Eur. J. 2009, 15, 10110−10120. (f) Yeh, H. C.; Sharma, J.; Yoo, H.; Martinez, J. S.; Werner, J. H. Proc. SPIE 2010, 7576, 75760N1−75760N9. 8496

dx.doi.org/10.1021/ac402125g | Anal. Chem. 2013, 85, 8493−8497

Analytical Chemistry

Letter

(3) (a) Yeh, H.-C.; Sharma, J.; Shih, I.-M.; Vu, D. M.; Martinez, J. S.; Werner, J. H. J. Am. Chem. Soc. 2012, 134, 11550−11558. (b) Guo, W.; Yuan, J.; Dong, Q.; Wang, E. J. Am. Chem. Soc. 2010, 132, 932− 934. (c) Petty, J. T.; Zheng, J.; Hud, N. V.; Dickson, R. M. J. Am. Chem. Soc. 2004, 126, 5207−5212. (d) Lan, G. Y.; Chen, W. Y.; Chang, H. T. Biosens. Bioelectron. 2011, 26, 2431−2435. (e) Liu, X.; Wang, F.; Niazov-Elkan, A.; Guo, W.; Willner, I. Nano Lett. 2013, 13, 309−314. (4) (a) Sharma, J.; Yeh, H. C.; Yoo, H.; Werner, J. H.; Martinez, J. S. Chem. Commun. 2011, 47, 2294−2296. (b) Han, B.; Wang, E. Biosens. Bioelectron. 2011, 26, 2585−2589. (c) Yang, S. W.; Vosch, T. Anal. Chem. 2011, 83, 6935−6939. (d) Zhang, L.; Zhu, J.; Guo, S.; Li, T.; Li, J.; Wang, E. J. Am. Chem. Soc. 2013, 135, 2403−2406. (5) Ritchie, C. M.; Johnsen, K. R.; Kiser, J. R.; Antoku, Y.; Dickson, R. M.; Petty, J. T. J. Phys. Chem. C 2007, 111, 175−181. (6) (a) Ramakrishna Matte, H. S. S.; Subrahmanyam, K. S.; Venkata Rao, K.; George, S. J.; Rao, C. N. R. Chem. Phys. Lett. 2011, 506, 260− 264. (b) Sapsford, K. E.; Berti, L.; Medintz, I. L. Angew. Chem., Int. Ed. 2006, 45, 4562−4589. (c) Jennings, T. L.; Singh, M. P.; Strouse, G. F. J. Am. Chem. Soc. 2006, 128, 5462−5467. (d) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.; Springer Academic: New York, 2006. (7) (a) Singh, A. K.; Lu, W.; Senapati, D.; Khan, S. A.; Fan, Z.; Senapati, T.; Demeritte, T.; Beqa, L.; Ray, P. C. Small 2011, 7, 2517− 2525. (b) Swathi, R. S.; Sebastian, K. L. J. Chem. Phys. 2009, 130, 086101. (8) Yun, C. S.; Javier, A.; Jennings, T.; Fisher, M.; Hira, S.; Peterson, S.; Hopkins, B.; Reich, N. O.; Strouse, G. F. J. Am. Chem. Soc. 2005, 127, 3115−3119. (9) Zheng, J.; Nicovich, P. R.; Dickson, R. M. Annu. Rev. Phys. Chem. 2007, 58, 409−431. (10) Huang, S.; Pfeiffer, C.; Hollmann, J.; Friede, S.; Chen, J. J.-C.; Beyer, A.; Haas, B.; Volz, K.; Heimbrodt, W.; Montenegro Martos, J. M.; Chang, W.; Parak, W. J. Langmuir 2012, 28, 8915−8919. (11) Zhang, P.; Rogelj, S.; Nguyen, K.; Wheeler, D. J. Am. Chem. Soc. 2006, 128, 12410−12411. (12) Butour, J. L.; Mazard, A. M.; Vieussens, C.; Johnson, N. P. Chem.-Biol. Interact. 1990, 73, 195−205. (13) (a) Grindley, N. D. F.; Whiteson, K. L.; Rice, P. A. Annu. Rev. Biochem. 2006, 75, 567−605. (b) Marti, T. M.; Fleck, O. Cell. Mol. Life Sci. 2004, 61, 336−354. (c) Linn, S. M.; Lloyd, R. S.; Roberts, R. J. Nuclease, 2nd ed.; Cold Spring Harbor Laboratory Press: Cold Spring Harbor, NY, 1993. (14) O’Neill, P. R.; Gwinn, E. G.; Fygenson, D. K. J. Phys. Chem. C 2011, 115, 24061−24066.

8497

dx.doi.org/10.1021/ac402125g | Anal. Chem. 2013, 85, 8493−8497