Letter pubs.acs.org/ac
Triggered Excited-State Intramolecular Proton Transfer Fluorescence for Selective Triplex DNA Recognition Ying Wang, Yuehua Hu, Tao Wu, Xiaoshun Zhou, and Yong Shao* Institute of Physical Chemistry, Zhejiang Normal University, Jinhua 321004, Zhejiang, People’s Republic of China S Supporting Information *
ABSTRACT: The triplex DNA has received much interest due to its various applications in gene regulation, molecular switch, and sensor development. However, realizing a highly selective recognition using a fluorescence probe specific only for the triplex topology is still a great challenge. Herein, we found that relative to the structural analogues of natural robinetin, myricetin, quercetin, kaempferol, morin, rutin, baicalin, luteolin, naringenin, genistein, chrysin, galangin, isorhamnetin, and several synthetic flavonoids, fisetin (FIS) is the brightest emitter when targeting the triplex DNA in contrast to binding with ss-DNA, ds-DNA (with or without an abasic site), i-motif, and DNA/RNA G-quadruplexes. Only the triplex association triggers the FIS green fluorescence that is relaxed from the tautomer favorable for excited-state intramolecular proton transfer (ESIPT). FIS can stabilize the triplex structure and primarily interact with the two terminals of the triplex via a 2:1 binding mode. This work demonstrates the potential of FIS as a DNA structure-selective switch-on ESIPT probe when evolving the triplex-forming oligonucleotides and developing the novel triplex-based sensors and switches. ucleic acid structure-sensitive fluorescence probes have received much attention due to their various applications in the selective identification of specific nucleic acid structures, development of high performance sensors, and improved diagnosis and therapy of gene-related diseases.1−5 The signaling probes, which are nonfluorescent when free in solution but are highly fluorescent upon binding with specific nucleic acid structures, are more straightforward in recognizing nucleic acids and benefit from a nucleic acid label-free manipulation. Thus, the recognition event can be directly read out without special separation steps and complicated detection facilities.6 Triplex DNA structures form via the sequence-selective recognition of ds-DNA at the major groove by a third strand. According to the orientation of the phosphate backbone polarity, the triplex-forming oligonucleotide (TFO) can bind with its partner ds-DNA in a parallel or an antiparallel manner.7,8 Generally, the parallel triplex is more stable than the antiparallel one. Triplexes have been identified to have wide bioactivities including regulation of gene expression,9,10 DNA damage,11 DNA repair,12 and even diseases.13,14 Additionally, benefiting from their unique structures, triplexes have found various applications such as molecular switches,15−18 nanodevices,19−21 and drug release.22 Triplexes have also been employed as sensor elements in the recognition and analysis of DNA methylation,23 SNP,24 proteins,25,26 toxic metal ions,17−19 small biomolecules,27 and even cancer cells.28 Several techniques including colorimetry,16,24 calorimetry, 29,30 NMR,31 surface-enhanced Raman spectroscopy (SERS),19 electrophoresis,20,21 DNA melting,22,23,27,29,32,33 circular dichroism (CD),29 high-pressure liquid chromatog-
N
© XXXX American Chemical Society
raphy−mass spectrometry (HPLC−MS),34 electrochemistry,35 competition dialysis,36,37 etc. have been used to identify the triplex formation and to evolve the triplex-selective ligands. Fluorescence methods such as Förster resonance energy transfer (FRET),15,16,20,21,25,26 chemiluminescence resonance energy transfer (CRET),38 and pyrene excimer,17,18 in which the involved DNAs are fluorophores labeled, have been exploited extensively as efficient tools in the investigation of the triplex folding and unfolding. As an alternative to consider the universal applications, a few fluorescence ligands have been selected out as the triplex binders for developing the label-free methods.29,31,39,40 However, although the label-free methods for the triplex recognition and triplex-based sensor preparation are more convenient in manipulation, developing an ideal triplex ligand is continuously a great challenge because of the following requirements: (1) the ligand must support a high triplex binding selectivity over the parent ds-DNA and other nucleic acid structures; (2) the ligand binding can stabilize the triplex structure; (3) the triplex formation can be easily readout by a switch-on fluorescence response of the ligand that is triggered only by the recognition event; (4) the ligand has a solely qualified binding mode in the triplex. In order to concurrently meet these 4S requirements, a ligand having a particular structure and an extraordinary fluorescence property should be developed. In this work, we identify fisetin (FIS) as Received: July 28, 2015 Accepted: November 10, 2015
A
DOI: 10.1021/acs.analchem.5b02851 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry the attractive candidate from the flavonoid variants (Table S1) for the selective fluorescence recognition of the DNA triplex. FIS belongs to the flavonols and it is almost nonfluorescent in aqueous solution. More interestingly, a suitable aprotic environment will populate its tautomer state with the competency of the excited-state intramolecular proton transfer (ESIPT) between the 3-OH and 4-carbonyl and favor an intensified green emission.41 We believe that the triplet structures in the triplexes will accommodate the FIS molecule considering their dimensionalities and support a strong ESIPT emission. We first compared an intramolecular triplex structure 15GCT with the other intramolecular nucleic acid structures (Table S2). The triplex formation was confirmed by the DNA melting experiment (Figure S1). As shown in Figure 1, the triplex
Subsequently, various sequences of the intramolecular triplexes, named nMN-Y, were tested, where n is the base-pair length in the triplex stem, MN the terminal base pair in the triplex stem, and Y the loop sequence. Interestingly, we found that all of the triplexes (including 15AT-T, 15AT-C, 10AT-C, 10AT-T, 10GC-T, 9AT-T, and 9GC-T, Table S2) can activate the FIS’ ESIPT green emission. Additionally, for the same length n, the FIS fluorescence is almost independent of the sequences of MN and Y. However, shortening in the length n results in a slight decrease in fluorescence (Figure S2). This is caused by the decreased stability and various local dynamics for the stemshortened triplex. However, at the same stem length, the loop length seems to have a minor effect on the FIS fluorescence (Figure S3). Furthermore, we also checked the intermolecular triplexes of the 15-mer 15CpGC and 22-mer 22TpAT (Table S2) and found that the switch-on FIS fluorescence is worked as opposed to the feeble fluorescence response for the random double-stranded DNA (ds-DNA). Besides FIS, we then tested various natural flavonoids (Table S1) for the triplex-activated fluorescence including robinetin (ROB), myricetin (MYR), quercetin (QUE), kaempferol (KAE), morin (MOR), rutin (RUT), baicalin (BAI), luteolin (LUT), naringenin (NAR), genistein (GEN), chrysin (CHR), galangin (GAL), and isorhamnetin (ISO). Among them, FIS, ROB, MYR, QUE, KAE, MOR, GAL, and ISO as flavonols are of the potential ESIPT ligands, while flavones (like BAI, LUT, GEN, CHR, and RUT) and flavanones (like NAR) cannot serve as the efficient ESIPT ligands due to lack of the 3-OH substituent. It is thus reasonable that these flavones and flavanones give an unaffected fluorescence upon binding with triplexes (Figure 2). However, surprisingly, among these natural
Figure 1. Fluorescence intensity of FIS (2 μM, pH 5.9) in 0.1 M K+ (black) and Na+ (red) in the presence of various nucleic acids (1 μM): 15GC-T (1), i-motif (2), rTel22 (3), Tel22 (4), PS2.M (5), 1XAV (6), 2O3M (7), T3TT (8), and T3 (9). Inset: photographs of these solutions under UV illumination and the corresponding spectra in Na+. N* and T* represent the normal and the tautomer ESIPT emission bands, respectively.
15GC-T actually activates the FIS’ tautomer ESIPT emission band (T*) at 538 nm as opposed to the relatively weak normal emission band (N*) at 460 nm,41 suggesting its triplex binding. It is well-known that cytosine- and guanine-rich sequences can form the i-motif and G-quadruplex (G4) structures, respectively. For G4s, herein we used rTel22, Tel22, PS2.M, 1XAV, 2O3M, T3TT, and T3 (Table S2) as the representatives for the parallel-stranded RNA G4 and the DNA G4s with the typical chair, basket, hybrid, and parallel conformations in Na+ and K+.42 However, FIS incubated with these structures exhibits a much weaker ESIPT emission, showing the preference of FIS to the triplex structure. Such the triplex differentiation from the other intramolecular structures can be even observed by the naked eye with the bright green fluorescence under UV illumination (inset of Figure 1). The role of the TFO strand on the FIS preference for the triplex was further confirmed by the negligible fluorescence for the hairpin 15GC (Table S2) that is exactly the duplex substructure in the triplex 15GC-T (Figure S2). Previously we explored that a DNA containing an abasic site can light up the FIS fluorescence.41 An abasic site was herein introduced into the hairpin 15GC in the center of the sequence to generate the hairpin 15GCX (Table S2). Indeed, relative to 15GC, an increase in fluorescence was observed for 15GCX but was still 15 times weaker than that for the triplex 15GC-T (Figure S2).
Figure 2. Fluorescence of the investigated flavonoids (2 μM) in 0.1 M Na+ in the absence (blue) and presence of 15AT-T (black), 15AT-C (red), and 15GC-T (green) (1 μM). From 1 to 14 for the flavonoids: FIS, ROB, MYR, QUE, KAE, MOR, RUT, BAI, LUT, NAR, GEN, CHR, GAL, and ISO. Inset: photographs of these solutions under UV illumination and the typical emission spectra.
flavonols, notwithstanding their structural similarities, only the ESIPT green emission of FIS can be activated by the triplex DNAs (Figure 2). This suggests a high structure selectivity of the investigated flavonols in binding with the triplexes and favoring the ESIPT emission. We then used four synthetic flavonols of 3-hydroxyflavone (3HF), 5,7-dimethoxy-3- hydroxyflavone (5,7-dM-3HF), 6,3′-dimethoxy-3-hydroxyflavone (6,3′-dM-3HF), and 7,4′-dimethoxy-3- hydroxyflavone (7,4′dM-3HF) as controls. We found that upon binding with the triplexes these flavonols are slightly brighter than ROB, MYR, QUE, KAE, MOR, GAL, and ISO but are still several times weaker than FIS (Figure S4). Among them, 5,7-dM-3HF shares B
DOI: 10.1021/acs.analchem.5b02851 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry
3 and Figure S6), showing that the intercalative interaction of the coplanar flavonoids with the triplex is impossible and the binding most likely occurs at the triplex extremity via end stacking. The binding specificity of FIS with the triplex structure was further investigated by the Job’s plot analysis. The fluorescence changes give a perfect turning point of about 0.67 for 15AT-T and 15GC-T, as obtained in Figure 4A. Additionally, the
the similar substituent position as GAL but bears the methoxyl groups in place of the hydroxyl groups at the corresponding positions of GAL. The relatively higher emission of 5,7-dM3HF than GAL upon binding with the triplexes should reflect the adverse effect of the 5-OH group on the ESIPT fluorescence, since the flavones and the flavonols containing the 5-OH group are also similarly nonfluorescent.43,44 Thus, in comparison with these synthetic flavonols, the hydroxyl substituents in the natural flavonols play determinant roles in their triplex binding and/or the resultant ESIPT emission. DNA melting was further used to investigate the triplex binding of these flavonoids. The triplexes are well featured in the melting curves by two transitions since increasing the solution temperature will first detach the TFO strand from the duplex and then be followed by the duplex disconnection. As shown in Figure 3, the two processes occur at the melting
Figure 4. (A) Job’s plot analysis of the stoichiometry of FIS binding to 15GC-T and 15AT-T. The total concentration of FIS and the triplex was kept at 2 μM and the concentration ratio was systematically changed. (B) Fluorescence titration of flavonoid (0.5 μM) upon gradual addition of DNA. The solid lines for the triplex + FIS show a 2:1 binding model by the curve fitting, while the lines for 15GC + FIS and 15AT-C + MOR are only for clarity. Inset: typical photographs of these solutions under UV illumination. Figure 3. Change of the 15AT-C (2 μM) melting temperature (Tm) with increasing flavonoid concentrations. Inset: the typical melting curves for FIS with its concentration increasing from 0 to 20 μM.
intermolecular triplex (with 22TpAT as an example) follows the same result (Figure S7) despite the increased length. Therefore, the FIS binding mode is independent of the triplex molecularity (intramolecular and intermolecular), the triplex sequence, and the triplex length, and a universal binding stoichiometry of 2:1 for the FIS preference with the triplex can be predicted. The 2:1 binding mode is further supported by the fluorescence titration experiments with good curve fittings to give a binding constant of about 3.6 ± 0.3 × 106 M−1 (Kd = 0.3 μM, with 15AT-C and 15GC-T as typical examples), as shown in Figure 4B. Thus, the end stacking is the determinant binding mode for FIS interacting with the triplex structure. We rationalize that in comparison with the previously reported hydrophobic intercalator triplex binders,29,31,39,40 the presence of multiple hydroxyl groups in FIS makes the molecule more hydrophilic and the triplex end stacking is the solely qualified binding mode, as opposed to the variant types of the triplex binding sites for the intercalator derivatives.29,31,36,37 This end stacking mode should prefer to the intercalative one due to the less interference from the intrinsic ds-DNA binding, although intentionally designed intercalator-groove-binder conjugates can improve the triplex selectivity.33,37 As for the fluorescent intercalators, 2-(2-naphthyl)quinoline intercalator derivative39 only exhibited two times difference in fluorescence responses upon binding with the triplex and ds-DNA. Relative to the triplex binding and the overall ds-DNA unbinding reported herein using FIS as the selector, only 3−8 times higher binding constant was obtained for the ethidium bromide analogues on binding with the triplexes in comparison to the ds-DNA binding.40 Note that FIS shows a triplex binding affinity that is comparable to the strong indoloquinoline intercalator29 and is at least 3 times higher than ethidium bromide analogues.40
temperatures (Tm) of 29.4 and 57.5 °C for 15AT-C. However, the FIS addition results in stabilizing of the triplex structure with the ΔTm increasing up to 14 °C, while the duplex stability is almost unaffected, suggesting the specific binding of FIS with the triplex structure. However, ROB and MYR also only stabilize the triplex structure as FIS does, in spite of their difference from FIS in the resultant ESIPT fluorescence. Interestingly, we also found that MOR and RUT are ineffective in stabilizing either the triplex or the duplex structure. This structure-selective binding was also confirmed by the fluorescence experiments of FIS alone and the coexistence of FIS with MOR or RUT, respectively (Figure S5). We believe that these differences in the triplex binding are most likely caused by the spatial structures of the flavonoids. For example, because of the bulky substituent of the rutinoside at the 3position and the steric hindrance arising from the 2′-OH substituent, the ring B in RUT and MOR, respectively, adopts a significant noncoplanar conformation relative to the ring A/C plane,45,46 thus predicting a poor binding with the triplex structure. Additionally, the strong binding of FIS, ROB, and MYR with the triplexes suggests a more coplanar conformation for these flavonoids, as predicted for such a conformation in the literature.45,46 This conformation assumption was also confirmed by the melting experiments for the coplanar QUE, KAE, and LUT and noncoplanar NAR,45,46 respectively (Figure S6). Thus, the triplex design will also provide an excellent platform for experimentally verifying the calculated flavonoid conformations. Furthermore, we note that all of the investigated flavonoids exhibit no influence on the duplex melting (Figure C
DOI: 10.1021/acs.analchem.5b02851 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry Such the end-stacking mode (see abstract graphic) and the switch-on fluorescence can provide a practical label-free method to evaluate the binding affinity of the TFO strand with its ds-DNA partner. As shown in Figure 5, the titrations of
■
Experimental procedures, flavonoid structures, nucleic acid sequences, Tm curves, fluorescence spectra, Job’s plot analysis, nanoswitch triggered by pH, ATP sensor, and peptide interaction (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Fax: 86 579 82282595. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the Zhejiang Provincial Natural Science Foundation of China (Grants LR12B05001, LR15B030002), the Zhejiang Provincial Public Welfare Project (Grant 2014C31150), and the National Natural Science Foundation of China (Grants 21545009 and 21075112).
■
Figure 5. Affinity analysis of the TFO binding with the corresponding ds-DNA (0.5 μM). An excess of FIS (6 times higher than the ds-DNA) was used to ensure the formed triplex in the FIS-bound state.
the ds-DNA partners in the triplex 22TpAT and 15CpGC with the corresponding TFOs gradually increase the FIS fluorescence. At the excess of FIS, the fluorescence will actually follow the resultant triplex product. The titration data are well agreed with the 1:1 fitting for the TFO binding with the ds-DNA and give the binding constants of 2.1 × 107 and 4.5 × 106 M−1, which mean the binding free energies are −9.8 and −8.9 kcal/ mol for 22TpAT and 15CpGC, respectively. These values fall in the range of the previously reported ones for the TFO affinity30,47 and are in agreement with the prediction of the strengthened TFO binding for the lengthened DNA within a series of the sequences having the comparable base compositions. We also note that at the excess of the TFOs, 22TpAT and 15CpGC result in the same FIS fluorescence responses, in spite of the difference in the triplex lengths, again reflecting the end-stacking binding mode. Thus, FIS will find wide applications in the TFO evolution and the triplex-based elements.15−21 As examples, FIS can serve as a promising reporter for the triplex sequence-tuned pH nanoswitch15 (Figure S8), highly selective triplex-based ATP sensor using a more universal molecular beacon design19 (Figure S9), and the triplex regulation by an HIV-1 Tat cell-penetrating protein peptide (Figure S10) for the potency in protein−triplex interactions. In conclusion, the high selective recognition of the triplex structure over the other types of nucleic acids including ssDNA, ds-DNA, i-motif, and DNA/RNA G-quadruplexes is achieved using FIS as the fluorescence probe. The triplextriggered ESIPT fluorescence is most efficient for FIS over the other flavonoids in spite of the similarities in their molecular structures, suggesting the pivotal role of the flavonoids’ substituent pattern on the triplex recognition. The 2:1 triplex terminal binding mode provides promising applications for FIS in evolving the triplex-forming oligonucleotides and developing the novel triplex-based sensors and switches.
■
REFERENCES
(1) Liu, J.; Cao, Z.; Lu, Y. Chem. Rev. 2009, 109, 1948−1998. (2) Li, D.; Song, S.; Fan, C. Acc. Chem. Res. 2010, 43, 631−641. (3) Iliuk, A. B.; Hu, L.; Tao, W. A. Anal. Chem. 2011, 83, 4440−4452. (4) Zhu, G.; Ye, M.; Donovan, M. J.; Song, E.; Zhao, Z.; Tan, W. Chem. Commun. 2012, 48, 10472−10480. (5) Deigan, K. E.; Ferré-DÁ maré, A. R. Acc. Chem. Res. 2011, 44, 1329−1338. (6) Nutiu, R.; Li, Y. J. Am. Chem. Soc. 2003, 125, 4771−4778. (7) Arya, D. P. Acc. Chem. Res. 2011, 44, 134−146. (8) Gowers, D. M.; Fox, K. R. Nucleic Acids Res. 1999, 27, 1569− 1577. (9) Hegarat, N.; Novopashina, D.; Fokina, A. A.; Boutorine, A. S.; Venyaminova, A. G.; Praseuth, D.; Francois, J. C. FEBS J. 2014, 281, 1417−1431. (10) Jain, A.; Wang, G.; Vasquez, K. M. Biochimie 2008, 90, 1117− 1130. (11) Rogers, F. A.; Tiwari, M. K. Yale J. Biol. Med. 2013, 86, 471− 478. (12) Chin, J. Y.; Glazer, P. M. Mol. Carcinog. 2009, 48, 389−399. (13) Bissler, J. J. Front. Biosci., Landmark Ed. 2007, 12, 4536−4546. (14) Mukherjee, A.; Vasquez, K. M. Biochimie 2011, 93, 1197−1208. (15) Idili, A.; Vallée-Bélisle, A.; Ricci, F. J. Am. Chem. Soc. 2014, 136, 5836−5839. (16) Antony, T.; Thomas, T.; Sigal, L. H.; Shirahata, A.; Thomas, T. J. Biochemistry 2001, 40, 9387−9395. (17) Zheng, J.; Li, J.; Jiang, Y.; Jin, J.; Wang, K.; Yang, R.; Tan, W. Anal. Chem. 2011, 83, 6586−6592. (18) Zheng, J.; Nie, Y.; Hu, Y.; Li, J.; Li, Y.; Jiang, Y.; Yang, R. Chem. Commun. 2013, 49, 6915−6917. (19) Zheng, J.; Jiao, A.; Yang, R.; Li, H.; Li, J.; Shi, M.; Ma, C.; Jiang, Y.; Deng, L.; Tan, W. J. Am. Chem. Soc. 2012, 134, 19957−19960. (20) Rusling, D. A.; Nandhakumar, I. S.; Brown, T.; Fox, K. R. ACS Nano 2012, 6, 3604−3613. (21) Chen, Y.; Lee, S.-H.; Mao, C. Angew. Chem., Int. Ed. 2004, 43, 5335−5338. (22) Kroner, C.; Gockel, A.; Liu, W. J.; Richert, C. Chem. - Eur. J. 2013, 19, 15879−15887. (23) Johannsen, M. W.; Gerrard, S. R.; Melvin, T.; Brown, T. Chem. Commun. 2014, 50, 551−553. (24) Zhu, X.; Liu, Y.; Yang, J.; Liang, Z.; Li, G. Biosens. Bioelectron. 2010, 25, 2135−2139. (25) Zhu, D.; Zhu, J.; Zhu, Y.; Wang, L.; Jiang, W. Chem. Commun. 2014, 50, 14987−14990. (26) Altevogt, D.; Hrenn, A.; Kern, C.; Clima, L.; Bannwarth, W.; Merfort, I. Org. Biomol. Chem. 2009, 7, 3934−3939. (27) Huang, H.; Tlatelpa, P. C. Chem. Commun. 2015, 51, 5337− 5339.
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.5b02851. D
DOI: 10.1021/acs.analchem.5b02851 Anal. Chem. XXXX, XXX, XXX−XXX
Letter
Analytical Chemistry (28) Xi, D.; Wang, X.; Ai, S.; Zhang, S. Chem. Commun. 2014, 50, 9547−9549. (29) Riechert-Krause, F.; Autenrieth, K.; Eick, A.; Weisz, K. Biochemistry 2013, 52, 41−52. (30) Eick, A.; Riechert-Krause, F.; Weisz, K. Bioconjugate Chem. 2010, 21, 1105−1114. (31) Dickerhoff, J.; Riechert-Krause, F.; Seifert, J.; Weisz, K. Biochimie 2014, 107, 327−337. (32) Reddy, P. M.; Bruice, T. C. J. Am. Chem. Soc. 2004, 126, 3736− 3747. (33) Xue, L.; Charles, I.; Arya, D. P. Chem. Commun. 2002, 70−71. (34) Xu, N.; Yang, H.; Cui, M.; Wan, C.; Liu, S. Anal. Chem. 2012, 84, 2562−2568. (35) Patterson, A.; Caprio, F.; Vallée-Bélisle, A.; Moscone, D.; Plaxco, K. W.; Palleschi, G.; Ricci, F. Anal. Chem. 2010, 82, 9109− 9115. (36) Holt, P. A.; Ragazzon, P.; Strekowski, L.; Chaires, J. B.; Trent, J. O. Nucleic Acids Res. 2009, 37, 1280−1287. (37) Arya, D. P.; Xue, L.; Tennant, P. J. Am. Chem. Soc. 2003, 125, 8070−8071. (38) Zhang, S.; Yan, Y.; Bi, S. Anal. Chem. 2009, 81, 8695−8701. (39) Lu, E.; Peng, X.; Song, F.; Fan, J. Bioorg. Med. Chem. Lett. 2005, 15, 255−257. (40) Tam, V. K.; Liu, Q.; Tor, Y. Chem. Commun. 2006, 2684−2686. (41) Xu, S.; Shao, Y.; Ma, K.; Cui, Q.; Liu, G.; Wu, F. Sens. Actuators, B 2012, 171−172, 666−671. (42) Liu, L.; Shao, Y.; Peng, J.; Huang, C.; Liu, H.; Zhang, L. Anal. Chem. 2014, 86, 1622−1631. (43) Ash, S.; De, S. P.; Pyne, S.; Misra, A. J. Mol. Model. 2010, 16, 831−839. (44) Hofener, S.; Kooijman, P. C.; Groen, J.; Ariese, F.; Visscher, L. Phys. Chem. Chem. Phys. 2013, 15, 12572−12581. (45) Van Acker, S. A. B. E.; De Groot, M. J.; Van den Berg, D. J.; Tromp, M. N. J. L.; Den Kelder, G. D. O.; Van der Vijgh, W. J. F.; Bast, A. Chem. Res. Toxicol. 1996, 9, 1305−1312. (46) Á lvarez-Diduk, R.; Ramirez-Silva, M. T.; Galano, A.; Merkoçi, A. J. Phys. Chem. B 2013, 117, 12347−12359. (47) Sugimoto, N.; Wu, P.; Hara, H.; Kawamoto, Y. Biochemistry 2001, 40, 9396−9405.
E
DOI: 10.1021/acs.analchem.5b02851 Anal. Chem. XXXX, XXX, XXX−XXX
