Macrocyclic Host-Dye Reporter for Sensitive Sandwich-Type

Letter pubs.acs.org/ac

Macrocyclic Host-Dye Reporter for Sensitive Sandwich-Type Fluorescent Aptamer Sensor Cheng Yang,†,‡ Nicolas Spinelli,‡ Sandrine Perrier,† Eric Defrancq,‡ and Eric Peyrin*,† †

Université Grenoble Alpes, CNRS, DPM UMR 5063, F-38041 Grenoble, France Université Grenoble Alpes, CNRS, DCM UMR 5250, F-38041 Grenoble, France

‡

S Supporting Information *

ABSTRACT: We describe herein a novel approach for the fluorescent detection of small molecules using a sandwich-type aptamer strategy based on a signaling macrocyclic host-dye system. One split adenosine aptamer fragment was 5′conjugated to a β-cylodextrin (CD) molecule while the other nucleic acid fragment was labeled at the 3′-end by a dansyl molecule prone to be included into the macrocycle. The presence of the small target analyte governed the assembly of the two fragments, bringing the dye molecule and its specific receptor in close proximity and promoting the inclusion interaction. Upon the inclusion complex formation, the microenvironment of dansyl was modified in such a way that the fluorescent intensity increased. Concomitantly, this supplementary interaction at the aptamer extremities induced stabilizing effects on the ternary complex. We next proposed a bivalent signaling design where the two extremities of one split aptamer fragment were conjugated to the β-CD molecule while those of the other fragment were tagged by the dansyl dye. The dual reporting dye inclusion promoted an improvement of both the signal-to-background change and the assay sensitivity. Owing to the vast diversity of responsive hostmacrocycle systems available, this aptasensor strategy has potential to be extended to the multiplexed analysis and to other kinds of transducers (such as electrochemical).

T

reported. The second one is related to the signal enhancement by using either enzymatic catalysis15 or an improved detection process.16,17 We describe herein an alternative signaling strategy based on the use of macrocyclic host-dye reporters that provides a sandwich-type aptasensor of high transduction capability. The general principle is depicted in Figure 1A. One split aptamer fragment is 5′-conjugated to a host molecule while the other nucleic acid fragment is labeled at 3′-end by a dye molecule prone to be included into the host. The presence of the small target will govern the assembly of the two fragments, bringing the guest molecule (i.e., the dye) and its specific receptor (i.e., the macrocyclic host) in close proximity and promoting the inclusion interaction. Upon the inclusion complex formation, the microenvironment of the dye is modified in such way that its photophysical properties are changed, leading to the fluorescent transduction. Concomitantly, this supplementary interaction at the aptamer extremities is expected to produce stabilizing effects on the ternary complex. To improve the assay performance, a bivalent host−guest design is also proposed (Figure 1B). In this case, the two extremities of one split aptamer fragment are conjugated to the host molecule while those of the other fragment are tagged by the dye. The closure

he sandwich assays constitute the dominant strategy for the protein determination in bioanalysis because of the high specificity and sensitivity resulting from the dual recognition mechanism. Two receptors such as antibodies or more recently aptamers able to recognize two spatially distant regions of the macromolecule target are simultaneously used for both the capture and the signal generation.1,2 In contrast, because of their small size, low molecular-weight analytes cannot typically be measured by sandwich assays but rather by less sensitive and specific competitive methods.3 To circumvent this drawback, two types of aptamer sensor strategies based on binary probes prone to function as a sandwich-type format have been reported.4,5 Stojanovic et al.4 notably exploited the ability of nucleic acid aptamers to be split into two fragments which are able to specifically form a ternary assembly in the presence of the ligand. This elegant approach has been extensively used in the small target sensing area using different transduction methods including colorimetric, fluorescence, and electrochemical techniques.6−10 However, the major problem of the split aptamer approach is related to the significant decrease in the ternary complex stability upon splitting of the functional oligonucleotide,11 leading to low sensitivity assays. To address this concern, two main strategies have been proposed. The first one focuses on the stabilization of the target-bound complex. In this way, the DNA-templated chemical reaction between the two fragments of the split aptamer11−13 and the multiple aptamer binding14 were © XXXX American Chemical Society

Received: January 27, 2015 Accepted: March 4, 2015

A

DOI: 10.1021/acs.analchem.5b00341 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry

Figure 1. Schematic illustration of the (A) monovalent and (B) bivalent host-dye reporter sandwich-type aptamer sensing platform. The model systems based on the antiadenosine aptamer and the signaling β-CD/dansyl pair are depicted.

adenosine target (200 μM) (Figure 2A). This very likely originates from dye-nucleobase interaction within the doublestranded DNA structure.27 On the other hand, in both cases, a great enhancement of the H2-dansyl signal intensity was obtained upon addition of the CD-H1 conjugate (Figure 2A). Concomitantly, the emission spectra revealed a blue shift in the maximum emission wavelength (Figure 2C). These findings are in accordance with the incorporation of the dansyl moiety into the hydrophobic cavity of the β-CD,18−21 indicating that the signaling properties of the host-dye pair are effective in both the two- and the three-component assemblies. To note, the addition of the adenosine target to the reaction solution did not compete with the dansyl inclusion within the macrocycle (Supporting Information). As a comparison, the CD-aptdansyl (full aptamer) displayed high fluorescence intensity (similar to that retrieved with H2-dansyl at CD-H1 saturating concentration) both in the absence and presence of target. This can be attributed to its stable hairpin structure in both free and bound forms,28,29 allowing full dye inclusion into the β-CD cavity in the two cases. The difference in the response observed between the free and target-bound forms of CD-H1/H2-dansyl (Figure 2A) indicates that the DNA strand assembly is favored by the adenosine binding, consistent with the split aptamer principle. Moreover, by monitoring the circular dichroism signal changes as a function of temperature (Figure 2D), we pointed out an enhancement of the DNA architecture stability upon target binding. Indeed, in the presence of 100 μM adenosine, an increase in the melting temperature (Tm) of ∼6.5 °C can be estimated (24.6 °C vs 18 °C). Similar behavior was also obtained by fluorescence intensity measurements (Supporting Information). Importantly, the ternary complex formed between the CD-H1/H2-dansyl conjugates and the target also displayed a higher Tm value than that of the control adenosine-bound split aptamer with unconjugated H1/H2 strands (ΔTm of 6 °C, Figure 2D). As expected, this result

of the stem structure could determine two reporting dye inclusions for one target-induced assembly event, allowing further signal generation. We investigated the feasibility of this strategy by using the βcylodextrin (CD)/dansyl dye pair and the antiadenosine DNA hairpin aptamer as host−guest and split aptamer (H1 and H2) model systems, respectively (Figure 1 and the Supporting Information). The β-CD/dansyl pair has found previous chemosensing applications.18−21 Notably, when 5′-β-CD and 3′-dansyl DNA conjugates were used, Kuzuya et al.20 and De Tito et al.21 reported that a significant fluorescence enhancement of the dye was observed upon the formation of duplex or G-quadruplex structures. However, to the best of our knowledge, the host−guest reporter principle in general and the β-CD/dansyl signaling system in particular have never been described for a sandwich-type aptamer sensing. First, we explored the ability of the monovalent system to signal the target binding to the split aptamer (Figure 1A). We used either copper(I) catalyzed Huisgen 1,3-cycloaddition of azides and alkynes (CuAAC)22 for conjugating oligonucleotides at their 5′-end with mono-6-azide-deoxy-6-cyclodextrin23 and oxime ligation for conjugating oligonucleotides at their 3′-end using 2-aminooxy-N-[3-(5-dimethylamino-naphtalene-1-sulfonylamino)-propyl]acetamide24,25 (Supporting Information). Using these strategies, we prepared the two fragments of the split aptamer with either a 5′-CD or a 3′-dansyl moiety (respectively, CD-H1 and H2-dansyl, Figure 1). For control, the 5′-CD, 3′-dansyl bis-conjugated full aptamer (CD-aptdansyl) was also synthesized using both CuAAC and oxime in a sequential bis-conjugation protocol.26 We initially evaluated the ability of the DNA-conjugated dansyl group (H2-dansyl, 400 nM) to experience inclusion into the cyclodextrin cavity of the CD-H1 (at 10 °C). The fluorescence intensity of the H2-dansyl probe slightly increased when an unmodified H1 strand was added to the reaction solution, either in the absence or in the presence of the B

DOI: 10.1021/acs.analchem.5b00341 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry

Figure 2. (A) Fluorescence intensity (FI) change of H2-dansyl (400 nM) upon addition of H1 and CD-H1, in the absence and presence of 200 μM adenosine. (B) Fluorescence intensity change of dansyl-H2-dansyl (200 nM) upon addition of H1 and CD-H1-CD, in the absence and presence of 200 μM of adenosine. (C) Emission spectra of H2-dansyl (200 nM) and dansyl-H2-dansyl (200 nM) upon addition of CD-H1 (400 nM) and CDH1-CD (400 nM), respectively, in the absence and presence of 200 μM adenosine. (D) Normalized circular dichroism melting curves for H2-dansyl (1 μM)/CD-H1 (2 μM) and dansyl-H2-dansyl (1 μM)/CD-H1-CD (2 μM) in the absence and presence of 100 μM adenosine. For comparison, the melting curves are also presented for unconjugated H2 (1 μM)/H1 (2 μM).

indicates that the inclusion of the dansyl into the β-CD cavity significantly stabilizes the ternary complex. Once the single-ended β-CD/dansyl dye pair approach was established, we subsequently concentrated our efforts on the development of the bivalent signaling system. Initially, two conjugation designs were apprehended. The first one involved the attachment of the β-CD and dansyl moieties at the respective 5′- and 3′-extremities of the same DNA strand (CDH1-dansyl, CD-H2-dansyl), potentially leading to a “head-totail” configuration of the reporting system (Supporting Information). The second one relied on the 3′,5′ labeling of one DNA fragment by the two dyes whereas the other one is 3′,5′ conjugated to the two hosts (dansyl-H2-dansyl and CDH1-CD, respectively), providing an expected “head-to-head” configuration (Figure 1B). The synthesis of the different bisconjugates was carried out by using the same procedures than those reported above for the single conjugates (Supporting Information). While dansyl-H2-dansyl conjugate displayed fluorescence signal intensity consistent with that of the 3′dansyl conjugate (almost twice for the same concentration), CD-H1-dansyl and CD-H2-dansyl hetero bis-conjugates exhibited an increased fluorescence signal (Supporting Information). This data suggests a significant encapsulation of the dye into the β-CD macrocycle even in absence of the other aptamer fragment. Moreover, a decrease in the fluorescence signal was obtained when a DNA fragment complementary to

the fragment of the split aptamer was added to the reaction solution, reflecting an intramolecular inclusion interaction that is partly disrupted by the formation of a hybridized structure (Supporting Information). Therefore, the “head-to-head” signaling configuration (Figure 1B) was next privileged for the dual sensing design. The fluorescence intensity of the dansyl-H2-dansyl probe was measured upon addition of CD-H1-CD to the reaction solution (in the presence of 200 μM adenosine, temperature = 10 °C). The titration curve revealed that the dansyl-H2-dansyl conjugate (200 nM, two dyes per strand) produces with the CD-H1-CD conjugate a maximal signal close to that observed with the H2-dansyl probe (400 nM, one dye per strand) at saturation concentration of CD-H1 (Figure 2A,B). Furthermore, a significant blue shift in the maximal emission wavelength was also observed (Figure 2C). These results show that, during the target-induced DNA assembly, each dye of the probe is able to form inclusion complex with the host facing it. On the other hand, in contrast with the single-ended reporter approach, the fluorescence intensity of dansyl-H2dansyl in absence of the target showed only a modest variation upon addition of CD-H1-CD (Figure 2B vs Figure 2A). Obviously, the lower probe concentration used (200 vs 400 nM) contributes to this smaller signal change. However, the weaker stability of the two-component assembly (Tm = 14 °C as compared with 18 °C for the monovalent system, Figure 2D) C

DOI: 10.1021/acs.analchem.5b00341 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry

dye reporter serves not only as efficient transducer but also as a stabilizer of the ternary split aptamer complex. Second, the dual inclusion reporter strategy improves assay performances. To the best of our knowledge, the current work represents the first demonstration of a bivalent transduction strategy for split aptamer. Third, the signaling/stabilizing properties could be easily modulated through changes in the reporter couple. Owing to the vast diversity of responsive dye-macrocycle systems available,31 we may reasonably envisage a wide choice of fluorescent signaling pairs (or even the extension to other kind of transducers such as electrochemical probe-macrocycle inclusion systems).32 Finally, one could envision a multiplexed analysis by choosing adequately several reporter couples that target different spectral domains.

is also implied in the limited background. Such behavior is presumably attributable to a poor simultaneous alignment of the two β-CD and the two dansyl dyes over the duplex20 that partly impedes the dye inclusion and the target-free structure stabilization. Nevertheless, upon adenosine binding, the DNA architecture reached an overall stability identical to that of the monovalent signaling system (Tm ∼ 25 °C, Figure 2D). The conjugate stoichiometry, buffer composition and temperature were next optimized for the bivalent sensor (see details in the Supporting Information). The dose−response curve, i.e., the signal-to-background ratio (S/B) vs the adenosine concentration, was constructed under optimal conditions (Figure 3). The estimated detection limit (3σ)

■

ASSOCIATED CONTENT

S Supporting Information *

Synthesis of conjugates, experimental parts, and additional data as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

Figure 3. Dose−response curves for the adenosine and inosine ligands using the dansyl-H2-dansyl/CD-H1-CD (stoichiometry 1:2) system. For comparison, the dose−response curve for adenosine is also presented using the dansyl-H2-dansyl/CD-H1 (stoichiometry 1:2) system. Experimental conditions: 10 mM Tris-HCl, pH = 7.5, 5 mM MgCl2, T = 10 °C.

ACKNOWLEDGMENTS This work was supported by Labex ARCANE (Grant ANR-11LABX-0003-01). The Nanobio-ICMG platform (FR 2603) is acknowledged for providing synthesis and purification of oligonucleotides facilities as well as for MALDI-TOF-MS analyses.

was 1 μM with a dynamic range up to 100 μM. This limit of detection was better than or comparable to those of most small analyte-binding split aptamer-based optical methods (Supporting Information). In order to properly evaluate the influence of the two signaling pairs on the assay performance, the dose− response curve was also established using CD-H1 in place of CD-H1-CD. We observed both lower S/B variation and assay sensitivity which in turn determined a 5-fold increase in the limit of detection (Figure 3). These findings emphasize the beneficial effect of the bivalent reporting approach over the monovalent one. The presence of the noncognate ligand (inosine) in the dansyl-H2-dansyl/CD-H1-CD solution did not produce a significant fluorescence signal variation, demonstrating the specificity of the sensing platform (Figure 3). Finally, we evaluated the suitability of the sensing approach for a biological environment using heat-treated diluted human serum samples (with different serum dilution factors ranging from 1/50 to 1/20) spiked with known amounts of adenosine (final concentration = 3 and 5 μM). We observed an increase in the assay response in the 20- and 33-fold diluted serum samples compared with the buffered solution (Supporting Information). This can be attributed to the well-known autofluorescence of serum.30 However, recoveries in the 110%−115% range were still achieved in 2% serum (Supporting Information). In summary, we reported a novel sandwich-like aptamer sensing approach based on a host−guest reporting system. This aptasensor displays several attractive features. First, the host-

(1) Pinto, A.; Bermudo Redondo, M. C.; Ozalp, V. F.; O’Sullivan, C. K. Mol. Biosyst. 2009, 5, 548−553. (2) Pei, X.; Zhang, B.; Tang, J.; Liu, B.; Lai, W.; Tang, D. Anal. Chim. Acta 2013, 758, 1−18. (3) Gosling, J. P. Clin. Chem. 1990, 36, 1408−1427. (4) Stojanovic, M. N.; de Prada, P.; Landry, D. W. J. Am. Chem. Soc. 2000, 122, 11547−11548. (5) Durand, G.; Lisi, S.; Ravelet, C.; Dausse, E.; Peyrin, E.; Toulmé, J. J. Angew. Chem., Int. Ed. 2014, 126, 7062−7065. (6) Zuo, X.; Xiao, Y.; Plaxco, K. W. J. Am. Chem. Soc. 2009, 131, 6944−6945. (7) Lin, Z.; Luo, F.; Liu, Q.; Chen, L.; Qiu, B.; Cai, Z.; Chen, G. Chem. Commun. 2011, 47, 8064−8066. (8) Du, Y.; Guo, S.; Qin, H. S.; Dong, E.; Wang, E. Chem. Commun. 2012, 48, 799−801. (9) Zhu, Z.; Ravelet, C.; Perrier, S.; Guieu, V.; Fiore, E.; Peyrin, E. Anal. Chem. 2012, 84, 7203−7211. (10) Sato, Y.; Zhang, Y.; Nishizawa, S.; Seino, T.; Nakamura, K.; Li, M.; Teramae, N. Chem.Eur. J. 2012, 18, 12719−12724. (11) Sharma, A. K.; Kent, A. D.; Heemstra, J. M. Anal. Chem. 2012, 84, 6104−6109. (12) Sharma, A. K.; Heemstra, J. M. J. Am. Chem. Soc. 2011, 133, 12426−12429. (13) Kent, A. D.; Spiropulos, N. G.; Heemstra, J. M. Anal. Chem. 2013, 85, 9916−9923. (14) Liu, J.; Liu, Y.; Yang, X.; Wang, K.; Wang, Q.; Shi, H.; Li, L. Anal. Chem. 2013, 85, 11121−11128. (15) Freeman, R.; Sharon, E.; Tel-Vered, R.; Willner, I. J. Am. Chem. Soc. 2009, 131, 5028−5029.

■

D

REFERENCES

DOI: 10.1021/acs.analchem.5b00341 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry (16) Wu, C.; Yan, L.; Wang, C.; Lin, H.; Wang, C.; Chen, X.; Yang, C. J. Biosens. Bioelectron. 2010, 25, 2232−2237. (17) Jin, F.; Lian, Y.; Li, J.; Zheng, J.; Hu, Y.; Liu, J.; Huang, J.; Yang, R. Anal. Chim. Acta 2013, 799, 44−50. (18) Ikeda, H.; Li, Q.; Ueno, A. Biorg. Med. Chem. Lett. 2006, 16, 5420−5423. (19) Ihara, T.; Uemura, A.; Futamura, A.; Shimizu, M.; Baba, N.; Nishizawa, S.; Teramae, N.; Jyo, A. J. Am. Chem. Soc. 2009, 131, 1386− 1387. (20) Kuzuya, A.; Ohnishi, T.; Wasano, T.; Nagaoka, S.; Sumaoka, J.; Ihara, T.; Jyo, A.; Koiyama, M. Bioconjugate Chem. 2009, 20, 1643− 1649. (21) De Tito, S.; Morvan, F.; Meyer, A.; Vasseur, J. J.; Cummaro, A.; Petraccone, L.; Pagano, B.; Novellino, E.; Randazzo, A.; Giancola, C.; Montesarchio, D. Bioconjugate Chem. 2013, 24, 1917−1927. (22) Himo, F.; Lovell, T.; Hilgraf, R.; Rostovtsev, V. V.; Noodleman, L.; Sharpless, K. B.; Fokin, V. V. J. Am. Chem. Soc. 2005, 127, 210− 216. (23) Tang, W.; Ng, S. C. Nat. Protoc. 2008, 3, 691−697. (24) Boturyn, D.; Boudali, A.; Constant, J. F.; Defrancq, E.; Lhomme, J. Tetrahedron 1997, 53, 5485−5492. (25) Forget, D.; Renaudet, O.; Boturyn, D.; Defrancq, E.; Dumy, P. Tetrahedron Lett. 2001, 42, 9171−9174. (26) Meyer, A.; Spinelli, N.; Dumy, P.; Vasseur, J. J.; Morvan, F.; Defrancq, E. J. Org. Chem. 2010, 75, 3927−3930. (27) Nazarenko, I.; Pires, R.; Lowe, B.; Obaidy, M.; Rashtchian, A. Nucleic Acids Res. 2002, 30, 2089−2195. (28) Nishihira, A.; Ozaki, H.; Wakabayashi, M.; Kuwahara, M.; Sawai, H. Nucleic Acids Symp. Ser. 2004, 48, 135−136. (29) Perrier, S.; Ravelet, C.; Guieu, V.; Fize, J.; Roy, B.; Perigaud, C.; Peyrin, E. Biosens. Bioelectron. 2010, 25, 1652−1657. (30) Li, L. L.; Ge, P.; Selvin, P. R.; Lu, Y. Anal. Chem. 2012, 84, 7852−7856. (31) Ghale, G.; Nau, W. M. Acc. Chem. Res. 2014, 47, 2150−2159. (32) Casas-Solvas, J. M.; Ortiz-Salmeron, E.; Fernandez, I.; GarciaFuentes, L.; Santoyo-Gonzales, F.; Vargas-Berenguel, A. Chem.Eur. J. 2009, 15, 8146−8162.

E

DOI: 10.1021/acs.analchem.5b00341 Anal. Chem. XXXX, XXX, XXX−XXX