Chinese Journal of Chemical Physics  2018, Vol. 31 Issue (4): 477-484

#### The article information

Xin-yue Huang, Min You, Guang-liu Ran, Hao-ran Fan, Wen-kai Zhang

Ester-Derivatized Indoles as Fluorescent and Infrared Probes for Hydration Environments

Chinese Journal of Chemical Physics, 2018, 31(4): 477-484

http://dx.doi.org/10.1063/1674-0068/31/cjcp1805118

### Article history

Accepted on: July 8, 2018
Ester-Derivatized Indoles as Fluorescent and Infrared Probes for Hydration Environments
Xin-yue Huanga, Min Youa, Guang-liu Rana, Hao-ran Fanb, Wen-kai Zhanga
Dated: Received on May 25, 2018; Accepted on July 8, 2018
a. Department of Physics and Applied Optics Beijing Area Major Laboratory, Center for Advanced Quantum Studies, Beijing Normal University, Beijing 100875, China;
b. Department of Chemistry, Beijing Normal University, Beijing 100875, China
*Author to whom correspondence should be addressed. Wen-kai Zhang, E-mail:wkzhang@bnu.edu.cn
Abstract: Tryptophan derivatives have long been used as site-specific biological probes. 4-Cyanotryptophan emits in the visible region and is the smallest blue fluorescent amino acid probe for biological applications. Other indole or tryptophan analogs may emit at even longer wavelengths than 4-cyanotryptophan. We performed FTIR, UV-Vis, and steady-state and time-resolved fluorescence spectroscopy on six ester-derivatized indoles in different solvents. Methyl indole-4-carboxylate emits at 450 nm with a long fluorescence lifetime, and is a promising candidate for a fluorescent probe. The ester-derivatized indoles could be used as spectroscopic probes to study local protein environments. Our measurements provide a guide for choosing esterderivatized indoles to use in practice and data for computational modeling of the effect of substitution on the electronic transitions of indole.
Key words: Infrared probe    Fluorescent probe    Tryptophan derivative    Indole derivative
Ⅰ. INTRODUCTION

Site-specific fluorescent and infrared (IR) probes have been widely used to study protein structure, conformation, function, and dynamics [1-4]. Tryptophan (Trp) is the most popular fluorescent probe because it is frequently found at or near sites that are responsible for protein-protein interactions, ligand binding, protein-DNA interactions, and enzyme catalysis [5-8]. Additionally, Trp has a larger fluorescence quantum yield (QY) than any other natural amino acid and its fluorescence is sensitive to the environment [9]. However, Trp absorbs and emits in the ultraviolet (UV) region and its QY is lower than larger dye molecule probes. Because of these factors, it is difficult to use Trp as a fluorophore for single-molecule measurements and imaging applications, especially under in vivo conditions. Therefore, much effort has been dedicated to the development of synthetic Trp-based amino acids for use as fluorescent and IR probes.

Among the Trp analogs, azatryptophans [10-12] and derivatives prepared by indole ring substitution [13-16] have been studied the most as fluorescent probes. Recently, a series of Trp analogs based on $n$-cyanotryptophan ($n$-CN-Trp, $n$=4-7), where $n$ indicates the carbon position for the CN substituent on the indole ring, have been used as site-specific fluorescent probes and IR probes for the local protein environment [13-19]. Talukder et al. showed that the QYs of 6-CN-Trp and 7-CN-Trp were two times more than that of Trp, and these derivatives could be used to study interactions between proteins and nucleic acids [13]. Markiewicz et al. found that 5-CN-Trp had a large QY change in response to the local hydration environment and could be used to probe local hydration status in proteins [14]. Hilaire et al. showed that 4-CN-Trp emitted in the blue region of the visible spectrum, and could be used as a blue fluorescent amino acid for biological spectroscopy and imaging applications [15]. Hilaire et al. also studied the fluorescence lifetimes of six $n$-cyanoindoles ($n$-CNI, where $n$=2-7) and found that only 4-CNI had a long fluorescence lifetime and high QY in water [16].

Cyanotryptophans can also be used as site-specific IR probes as they can provide effective information about proteins or polypeptides [17-19]. Zhang et al. found a linear dependence between the nitrile stretching frequency of 5-CNI and the Kamlet-Taft parameters ($\sigma$=$\pi^*$+$\beta$-$\alpha$), and that 5-CNI could be used as an IR probe of the local protein environment [17]. Rodgers et al. demonstrated that the Fermi resonance of 4-CNI could be used to determine hydrogen bonding (H-bonding) status in the surrounding region. These developments make CNI a versatile site-specific probe that can be used to study the local environment using both fluorescence and IR spectroscopy [18]. For example, Markiewicz et al. utilized 5-CN-Trp to investigate the hydration status near the Trp gate in the influenza A M2 proton channel using IR and fluorescence spectroscopy. They found that a lack of sufficient water molecules near the Trp gate provided an additional energetic bottleneck for M2 proton conduction [19]. Unfortunately, the dipole strength of the nitrile stretching vibration is low, which makes it difficult to use in two-dimensional IR measurements, and much effort has been dedicated to the development of an amino acid IR probe with stronger dipole strength [20-25]. Moreover, the maximum emission wavelength of CNI is 405 nm for 4-CNI in water and the longest fluorescence lifetime of CNI is 15.7 ns for 7-CNI in formamide. For these probes, the redshift of fluorescence spectrum is not large enough and the fluorescence lifetime is not very long. Consequently, new synthetic Trp-based amino acids with larger red shifts and longer fluorescence lifetimes are required for use as fluorescent probes. Ideally, these new probes would also have large vibrational dipole strengths and could be used as IR probes.

Recently, Pazos et al. found that ester carbonyl stretching frequencies were linearly dependent on the local electric field with a large dynamic range, and demonstrated that ester carbonyl could be used to probe small changes in local electric fields [26]. This prompted us to study ester-derivatized indoles that could combine the IR probe features of ester carbonyls with the fluorescent probe features of Trp derivatives. To the best of our knowledge, no systematic fluorescence and IR spectroscopy study has been performed on indole or tryptophan derivatives with ester carbonyl substituents. In this study, we systematically performed FTIR, UV-Vis, and steady-state and time-resolved fluorescence spectroscopy on six ester-derivatized indoles in different solvents. These derivatives we studied are methyl indole-$n$-carboxylate ($n$-MIC, $n$=4-7, FIG. 1), where $n$ indicates the carbon position for the ester carbonyl substituent on the indole ring. We investigated the spectral properties of the MIC derivatives using Kamlet-Taft empirical parameters, which are related to the properties of solution, such as solubilities, partition coefficients, thermodynamic and kinetic properties of chemical reactions, etc. Three empirical parameters are used to represent the solvent polarity ($\pi^*$), the hydrogen-bond acceptor ($\beta$), and the hydrogen-bond donor ($\alpha$) [27, 28]. We found that the MIC derivatives could be used as spectroscopic probes to study the local protein environment. Our measurements provide guidelines for choosing appropriate MIC derivatives in practice and data for theoretical calculation of substitution effects on indoles.

 FIG. 1 Structures of (a) indole and (b) methyl indole-4-carboxylate.
Ⅱ. EXPERIMENTS A. Materials and sample preparation

The following chemicals were purchased and used without further purification: indole (> 99%, Acros Organics), methyl indole-2-carboxylate (2-MIC, Accela.), methyl indole-3-carboxylate (3-MIC, 98%, Innochem), methyl indole-4-carboxylate (4-MIC, 99%, Acros Organics), methyl indole-5-carboxylate (5-MIC, 98+%, Alfa Aesar), methyl indole-6-carboxylate (6-MIC, Ark), and methyl indole-7-carboxylate (7-MIC, Ark). The following solvents of HPLC grade or higher quality were purchased and used without further purification: 1, 4-dioxane (Alfa Aesar), 2-propanol (Alfa Aesar), acetonitrile (Acros Organics), dimethyl sulfoxide (DMSO, Acros Organics), ethanol (EtOH, Decon Laboratories), methanol (MeOH, Acros Organics), tetrahydrofuran (THF, Acros Organics), dichloromethane (DCM, Acros Organics), and toluene (Alfa Aesar). All the sample solutions were freshly prepared by directly dissolution of the solutes in the solvents before use in spectroscopy measurements. The final concentrations of the solutions were approximately 50 mmol/L for the FTIR measurements and 15 μmol/L for the absorption and fluorescence measurements.

B. Spectroscopic measurements

FTIR spectra were measured on a Bruker VERTEX 70 spectrometer with a resolution of 2 cm$^{-1}$. The sample solution was placed between two 3-mm CaF$_2$ windows separated by either a 50-μm or 100-μm Teflon spacer. Measurements were conducted at room temperature. The solvent background was subtracted from each spectrum.

C. Absorption measurements

UV-Vis spectra were collected on an Agilent Technologies Cary 60 UV-Vis spectrophotometer at room temperature using a 1-cm quartz cuvette.

D. Static and time-resolved fluorescence measurements

Static fluorescence spectra were collected on a fluorometer (F-4600, HITACHI) at room temperature in a 1-cm quartz cuvette with a resolution of 0.2 nm and integration time of 4 nm/s. Time-resolved measurements were collected on a commercial time-correlated single photon counting system (FLS980 spectrometer, Edinburgh Instruments) in a 1-cm quartz cuvette at room temperature. The sample solutions were prepared with an OD of 0.05-0.2 at an excitation wavelength ($\lambda_{\text{ex}}$) of 285 nm. To minimize any inner filter effect, the excitation beam was positioned near the edge of the sample cuvette that faced the fluorescence collection optics.

Ⅲ. RESULTS AND DISCUSSION A. Static FTIR measurements

FTIR spectra were obtained of the ester carbonyl stretching vibrations of all six MIC derivatives at room temperature in DMSO (FIG. 2(a)). For these derivatives, the stretching vibration of the ester carbonyl was located between 1680 and 1730 cm$^{-1}$. Among the MIC derivatives, 2-MIC and 3-MIC showed the largest differences in the frequency of the ester carbonyl stretching vibration. The frequency for 2-MIC showed the largest blueshift, and that for 3-MIC showed the largest redshift. The frequencies for the ester carbonyl stretching vibrations of all the other MIC derivatives in DMSO were similar to each other, even though these derivates had the ester carbonyl group located at different positions on the indole ring. In order to understand these observation, we calculated the natural charges of indole. As shown in FIG. S1 in supplementary materials, the carbon atom at the position 2 has positive charge, the carbon at the position 3 has the most negative charge, while the carbon atom at the position 4-7 has the negative charge in between. We know that the frequency of the carbonyl group stretching vibration is correlated with the electronegativity of its neighboring carbon atom. So our observations are consistent with the calculations.

 FIG. 2 FTIR spectra of $n$-MIC derivatives ($n$=2-7) in (a) DMSO and (b) 2-propanol.

All the MIC derivatives had only one obvious ester carbonyl stretching vibration band in DMSO (FIG. 2(a)). However, in 2-propanol and MeOH, all the MIC derivatives showed more than one vibration band between 1680 and 1725 cm$^{-1}$ (FIG. 2(b) and FIG. S2 in supplementary materials). Qualitatively, these results could be explained by H-bonding between the solute and solvent [26]. The ester carbonyl group is a strong H-bonding acceptor, whereas 2-propanol ($\beta$=0.95, $\alpha$=0.76) is both a strong H-bonding acceptor and a strong H-bonding donor. Therefore, the ester carbonyl group can form zero, one, or two H-bonds with 2-propanol [29]. We can assign the higher frequency peak as the ester carbonyl stretching vibration which does not form any H-bonds with 2-propanol [26, 29]. However, DMSO ($\beta$=0.76, $\alpha$=0) is a strong H-bonding acceptor and a very weak H-bonding donor. So the ester carbonyl group cannot form H-bonds with DMSO, and consequently, there is only one band in the FTIR results. We found that the two peaks of 2-MIC have similar intensities but not for other MIC derivatives. We suspect that the steric effect may cause the population difference of H-bond between the solvent and different MIC derivatives. For example, our calculation found that the 7-MIC can form stable intramolecular H-bond while other MIC derivatives cannot. Quantitative analysis of the H-bonds effect of these MIC derivatives will be presented in a future publication.

We noticed that the ester carbonyl stretching vibration bands of 2-MIC and 7-MIC had shoulders at around 1680 and 1685 cm$^{-1}$, respectively in 2-propanol, and these might be caused by formation of two H-bonds with the solvent [26]. These results indicate that the ester carbonyl stretching vibrations of MIC derivatives are sensitive to the H-bonding status. Therefore, they could be used to study the local H-bonding environments in biological systems. In THF and acetonitrile, 7-MIC had two vibration bands, whereas the other MIC derivatives had only one vibration band (FIG. S2 in supplementary materials). The FTIR spectrum of 7-MIC in DCM was more complicated than the spectra of the other MIC derivatives (FIG. S2 in supplementary materials). Furthermore, the ester carbonyl stretching vibration bands of all MIC derivatives in toluene were very complicated; 2-MIC and 7-MIC had three vibration bands, 5-MIC had two vibration bands, and 3-MIC, 4-MIC, and 6-MIC each had one vibration band with a shoulder (FIG. S2 in supplementary materials). We suspect that this complexity is caused by $\pi$-$\pi$ interactions between the benzene ring and indoles [30, 31].

We then investigated the FTIR spectra of the MIC derivatives further. FTIR spectra of 4-MIC in various solvents were obtained (FIG. 3(a)). More than one vibrational band was observed for 4-MIC in protic solvents (e.g., 2-propanol and MeOH), but only one vibrational band was observed in aprotic solvents (e.g., DMSO). Therefore, 4-MIC can be used to investigate the hydration status of the local environment. When dissolved in different solvents, 4-MIC, 5-MIC, and 6-MIC showed similar responses to the change in solvent in their vibrational spectra (FIG. S3 in supplementary materials and FIG. 3(a)). The only difference was that 5-MIC had two vibrational bands in toluene and the others did not. These results suggest that 4-MIC, 5-MIC, and 6-MIC can be used as an IR probe like other ester studied in Ref.[26]. The situation was different for the other MIC derivatives. The FTIR spectra of 3-MIC in different solvents were similar to those of 4-MIC, 5-MIC, and 6-MIC, but it could not be used as a protein side chain probe, instead, it could possibly be used as a probe for drugs interactions with proteins since it is the precursor of many drugs [32, 33]. The FTIR spectrum of 7-MIC was very complicated (FIG. 3(b)), and both the number and positions of its vibrational bands changed in different solvents. In DMSO, there was only one vibrational band for 7-MIC, whereas there was more than one vibrational band in other solvents and the spectra were relatively complicated. We performed a quick density functional theory calculation and found that 7-MIC could form intramolecular H-bonds, which may explain the complexity of its FTIR spectra. Quantitative analysis of the 7-MIC IR spectroscopy will be presented in a future publication. The spectra of 2-MIC were also complicated, with only one band in DMSO and THF but more than one band in other solvents.

 FIG. 3 FTIR spectra of (a) 4-MIC and (b) 7-MIC in different solvents.

Since these MIC derivatives have different properties for use as probes, we could use different MIC derivatives to label different Trp sites and investigate different aspects of the local environment simultaneously. In addition, the dipole strength of the stretching vibration of the ester carbonyl is stronger than that of nitrile, and we suspect that some of the MIC derivatives could be better IR probes than cyanotryptophan derivatives for site-specific monitoring of the local environment near Trp residues.

B. Absorption spectra

The different MIC derivatives dissolved in water had distinct absorption spectra, as manifested by their spectral shapes, maximum absorption wavelengths ($\lambda_{\text{max}}$), and molar extinction coefficients ($\varepsilon$) (FIG. 4, Table Ⅰ). Compared to the indole, the $\lambda_{\text{max}}$ of all the MIC derivatives showed redshifts and spanned from 9 nm (3-MIC) to 44 nm (7-MIC). Qualitatively, these results agree with the literature, where an electron withdrawing group (EWG) on the indole ring leads to a redshift of $\lambda_{\text{max}}$ and an electron donating group results in a blueshift of $\lambda_{\text{max}}$ [34]. Since the ester is a moderate EWG when connected through a carbon and the cyano is a strong EWG, it is surprised that the $\lambda_{\text{max}}$ shifts of the MIC derivatives were larger than those of the CNIs with the same substitution positions. This result indicates that another mechanism besides EWG plays a role which will be investigated in a future publication. Notably, the absorption spectrum of 6-MIC had two different maxima, centered at 289 and 320 nm, which likely arose from a large separation between the 1L$_\text{a}$ and 1L$_\text{b}$ states in this molecule [16]. Among the MIC derivatives, 4-MIC and 7-MIC showed the largest redshifts in the $\lambda_{\text{max}}$, which were above 310 nm. The $\varepsilon$ of the MIC derivatives in the tested solvents were about two-to-six times higher than those obtained in water (FIG. S4 in supplementary materials).

 FIG. 4 UV-Vis absorption spectra of indole and MIC derivatives in water at room temperature.
Table Ⅰ The absorption ($\lambda_{\text{max}}$) and emission ($\lambda_{\text{em}}$) maxima and the molar extinction coefficient ($\varepsilon$ in the unit of (mol/L)$^{-1}$cm$^{-1}$) at $\lambda_{\text{max}}$ and quantum yield (QY) in water for indole and $n$-MIC. Quantum yield of indole in water is cited from Ref.[35].
C. Fluorescence spectra

The fluorescence spectra of the MIC derivatives in water were dependent on the substitution position of the ester carbonyl on the indole ring (FIG. 5). The emission peak position in the 3-MIC fluorescence spectrum was similar to that in the indole fluorescence spectrum, whereas the fluorescence spectra of all the other MIC derivatives were redshifted by 60-100 nm. The shifts for 4-MIC and 7-MIC were about 100 nm, and were larger than those for any of the CNI derivatives in water. In all the studied MIC derivatives, 6-MIC had the strongest fluorescence intensity and 5-MIC had the weakest. Interestingly, the fluorescence bandwidth of 4-MIC was the broadest and extended to 560 nm, which is nearly light green (FIG. 6). To the best of our knowledge, 4-MIC has the largest redshift among all the indole derivatives reported, which indicates that 4-MIC might be an excellent fluorescence probe. One can synthesize a tryptophan-based 4-MIC analog fluorophore probe and use it in various biological studies, especially in spectroscopic measurements where an amino acid-sized fluorescence reporter is required. Since 4-MIC is a small molecule and sensitive to hydrogen bonding, so it is a convenient and sensitive probe. It is possible to find the correlation between the central frequency of ester carbonyl stretching vibration and the local electrostatic field by measuring the frequency of the ester carbonyl stretching vibration of 4-MIC. In addition, the sensitivity of 4-MIC for hydrogen bonding and non-hydrogen bonding environments is different. We can find more and more useful information from the difference.

 FIG. 5 Fluorescence spectra of indole and MIC derivatives in water at room temperature. The excitation wavelength ($\lambda_{\text{ex}}$) was 285 nm. The fluorescence spectra of different MIC derivatives have been "normalized" against the maximum fluorescence intensity of 6-MIC in each case.
 FIG. 6 Pictures of a 5 mmol/L 4-MIC (a), a 5 mmol/L 4-CNI (b), and a 5 mmol/L indole (c) in water under illumination of 365-nm light, which was taken using a Canon EOS M10 camera.

Therefore, we believe that it is possible to develop a green fluorescent amino acid based on an indole derivative with substitution at the 4 position. We found that the emission peak position of 3-MIC in EtOH showed a larger redshift than in any other solvents, and all of the other MIC derivatives showed their largest redshifts in water (Table S1 in supplementary materials). We also found that the fluorescence intensity of 7-MIC in DMSO was the stronger than in any other studied solvents.

D. Time-resolved fluorescence measurements

Next, we measured the fluorescence decay kinetics of the MIC derivatives in the different solvents. The fluorescence decay kinetics of some $n$-MIC ($n$=4, 6, 7) in water fit reasonably well to a single-exponential function, which was used to determine the lifetime (Table Ⅱ). However, some other $n$-MIC ($n$=2, 3, 5) were better described by a double-exponential function, and the lifetimes ($\tau$, Table Ⅱ) are the intensity-weighted averages for each time constant, $\tau$. Among the derivatives, 2-MIC had the shortest fluorescence lifetime and 4-MIC had the longest. Combined with the large redshift in the fluorescence spectrum of 4-MIC, these results suggest that 4-MIC could be a good site-specific amino acid probe for the local protein environment.

Table Ⅱ Fluorescence lifetime ($\tau$) of $n$-MIC in different solvents, as indicated.

Similar to other indole derivatives [30-34], the fluorescence lifetimes of these MIC derivatives were dependent on the solvent. A similar procedure was applied for lifetime measurements in other solvents (Table Ⅱ). The longest lifetime of a MIC derivative (16.6 ns for 7-MIC in DMSO) was longer than that of a CNI derivative (15.7 ns for 7-CNI in formamide) [16]. Except for 4-CNI, most CNI derivatives have very short fluorescence lifetime in water, and 4-CNI is the only CNI derivative that could be used as a fluorescent probe. However, the fluorescence lifetimes of multiple $n$-MICs ($n$=4, 6, 7) in water are longer than that they are in other solvents, and most of these derivatives could be used as fluorescent probes.

Traditionally, to study how a solvent affects the fluorescence properties of a fluorophore, empirical solvent parameters are used to correlate between the solvent and a fluorescence property. Recently, Hilaire et al. [16] showed that the fluorescence lifetime of CNI exhibited a strong correlation with $\sigma$=$\pi^*$+$\beta$-$\alpha$, which is a combination of Kamlet-Taft parameters (Table S2 in supplementary materials). Specifically, the $\pi^*$ parameter denotes nonspecific electrostatic interactions between the solute and solvent molecules, whereas the $\beta$ and $\alpha$ parameters signify specific H-bonding interactions between the solvent and the carbonyl ($\beta$) and NH ($\alpha$) groups. Here, we tried to study the influence of the solvent on the fluorescence lifetimes of MIC derivatives using a similar approach. The fluorescence lifetime of 3-MIC exhibited a negative linear correlation with $\sigma$ in all solvents except for EtOH (FIG. 7). The fluorescence lifetimes of the other MIC derivatives also exhibited negative linear correlations with $\sigma$ in all solvents except for water and DMSO. These correlations suggest that increasing either the solvent's polarizability (i.e., $\pi^*$)or its H-bonding interactions (i.e., $\beta$) with the ester carbonyl group would decrease the fluorescence lifetimes of MIC derivatives. By contrast, increasing the H-bonding interactions with the ester carbonyl group (i.e., $\alpha$) would have the opposite effect. Notably, the slope of 5-MIC was steeper than that of any other MIC derivative, which indicates that 5-MIC is the most sensitive probe to the local environment.

 FIG. 7 Dependence of the fluorescence lifetime of (a) 2-MIC, (b) 3-MIC, (c) 4-MIC, (d) 5-MIC, (e) 6-MIC, and (f) 7-MIC on the empirical solvent parameter $\sigma$. Line corresponds to a linear fit of the data to the following equation: $\tau$=$A$+$B$$\sigma$ (the value of parameters $A$ and $B$ are listed in Table S3 in supplementary materials).
Ⅳ. CONCLUSION

Spectroscopic studies of protein structure, conformation, function, and dynamics require site-specific probes with spectroscopic signatures that undergo changes in response to their environment. Herein, we investigated a series of MIC derivatives in different solvents using FTIR and fluorescence spectroscopy. These derivatives could be used as sensitive IR and fluorescent probes of the local environment. The derivative 4-MIC, which emits at 450 nm with a fairly long fluorescence lifetime, is a promising candidate for development of a fluorescent probe. We believe that this study of MIC derivatives will expand their utility as novel spectroscopic probes in biological applications.

Supplementary materials: Emission maxima of MIC derivatives molecules in the indicated solvents; the Kamlet-Taft parameters: polarizability ($\pi^*$), H-bond accepting ability ($\beta$), H-bond donating ability ($\alpha$) and $\sigma$($\pi^*$+$\beta$-$\alpha$) of the indicated solvents; the fitting parameters of FIG. 7; the calculated natural charges of indole; FTIR of MIC derivatives in a special solvent and in all solvents; absorption spectra of MIC derivatives molecules in the indicated solvents; fluorescence spectra of 7-MIC molecules in the indicated solvent; fluorescence decay kinetics of MIC derivatives in water are given.

Ⅴ. ACKNOWLEDGMENTS

This work was supported by Beijing Natural Science Foundation (L172028), the National Natural Science Foundation of China (No.21773012 and No.91753118), the Recruitment Program of Global Youth Experts and the Fundamental Research Funds for Central Universities.

Reference
 [1] J. Ma, I. M. Pazos, W. Zhang, R. M. Culik, and F. Gai, Annu. Rev. Phys. Chem. 66 , 357 (2015). DOI:10.1146/annurev-physchem-040214-121802 [2] H. Kim, and M. Cho, Chem. Rev. 113 , 5817 (2013). DOI:10.1021/cr3005185 [3] E. A. Specht, E. Braselmann, and A. E. Palmer, Annu. Rev. Physiol. 79 , 93 (2017). DOI:10.1146/annurev-physiol-022516-034055 [4] R. Adhikary, J. Zimmermann, and F. E. Romesberg, Chem. Rev. 117 , 1927 (2017). DOI:10.1021/acs.chemrev.6b00625 [5] A. A. Bogan, and K. S. Thorn, J. Mol. Biol. 280 , 1 (1998). DOI:10.1006/jmbi.1998.1843 [6] C. H. Hsu, C. P. Chen, M. L. Jou, A. Y. L. Lee, Y. C. Lin, Y. P. Yu, W. T. Huang, and S. H. Wu, Nucleic Acids Res. 33 , 4053 (2005). DOI:10.1093/nar/gki725 [7] G. J. Bartlett, C. T. Porter, N. Borkakoti, and J. M. Thornton, J. Mol. Biol. 324 , 105 (2002). DOI:10.1016/S0022-2836(02)01036-7 [8] Y. Xue, A. V. Davis, G. Balakrishnan, J. P. Stasser, B. M. Staehlin, P. Focia, T. G. Spiro, J. E. Penner-Hahn, and T. V. O'Halloran, Nat. Chem. Biol. 4 , 107 (2008). DOI:10.1038/nchembio.2007.57 [9] C. A. Royer, Chem. Rev. 106 , 1769 (2006). DOI:10.1021/cr0404390 [10] S. Lepthien, M. G. Hoesl, L. Merkel, and N. Budisa, Proc. Natl. Acad. Sci. USA 105 , 16095 (2008). DOI:10.1073/pnas.0802804105 [11] P. Talukder, S. Chen, C. T. Liu, E. A. Baldwin, S. J. Benkovic, and S. M. Hecht, Bioorg. Med. Chem. 22 , 5924 (2014). DOI:10.1016/j.bmc.2014.09.015 [12] A. V. Smirnov, D. S. English, R. L. Rich, J. Lane, L. Teyton, A. W. Schwabacher, S. Luo, R. W. Thornburg, and J. W. Petrich, J. Phys. Chem. B 101 , 2758 (1997). DOI:10.1021/jp9630232 [13] P. Talukder, S. Chen, B. Roy, P. Yakovchuk, M. M. Spiering, M. P. Alam, M. M. Madathil, C. Bhattacharya, S. J. Benkovic, and S. M. Hecht, Biochemistry 54 , 7457 (2015). DOI:10.1021/acs.biochem.5b01085 [14] B. N. Markiewicz, D. Mukherjee, T. Troxler, and F. Gai, J. Phys. Chem. B 120 , 936 (2016). DOI:10.1021/acs.jpcb.5b12233 [15] M. R. Hilaire, I. A. Ahmed, C. W. Lin, H. Jo, W. F. DeGrado, and F. Gai, Proc. Natl. Acad. Sci. USA 114 , 6005 (2017). DOI:10.1073/pnas.1705586114 [16] M. R. Hilaire, D. Mukherjee, T. Troxler, and F. Gai, Chem. Phys. Lett. 685 , 133 (2017). DOI:10.1016/j.cplett.2017.07.038 [17] W. Zhang, B. N. Markiewicz, R. S. Doerksen, A. B. Smith Ⅲ, and F. Gai, Phys. Chem. Chem. Phys. 18 , 7027 (2016). DOI:10.1039/C5CP04413H [18] J. M. Rodgers, R. M. Abaskharon, B. Ding, J. Chen, W. Zhang, and F. Gai, Phys. Chem. Chem. Phys. 19 , 16144 (2017). DOI:10.1039/C7CP02442H [19] B. N. Markiewicz, T. Lemmin, W. Zhang, I. A. Ahmed, H. Jo, G. Fiorin, T. Troxler, W. F. DeGrado, and F. Gai, Phys. Chem. Chem. Phys. 18 , 28939 (2016). DOI:10.1039/C6CP03426H [20] K. L. Koziol, P. J. Johnson, B. Stucki-Buchli, S. A. Waldauer, and P. Hamm, Curr. Opin. Struct. Biol. 34 , 1 (2015). [21] W. K. Zhang, Chin. J. Chem. Phys. 29 , 1 (2016). DOI:10.1063/1674-0068/29/cjcp1512246 [22] M. You, L. Liu, and W. Zhang, Phys. Chem. Chem. Phys. 19 , 19420 (2017). DOI:10.1039/C7CP01867C [23] M. Maj, C. Ahn, D. Kossowska, K. Park, K. Kwak, H. Han, and M. Cho, Phys. Chem. Chem. Phys. 17 , 11770 (2015). DOI:10.1039/C5CP00454C [24] G. Lee, D. Kossowska, J. Lim, S. Kim, H. Han, K. Kwak, and M. Cho, J. Phys. Chem. B 122 , 4035 (2018). DOI:10.1021/acs.jpcb.8b00887 [25] S. Dutta, Y. L. Li, W. Rock, J. C. Houtman, A. Kohen, and C. M. Cheatum, J. Phys. Chem. B 116 , 542 (2012). DOI:10.1021/jp208677u [26] I. M. Pazos, A. Ghosh, M. J. Tucker, and F. Gai, Angew Chem. Int. Ed. Engl. 53 , 6080 (2014). DOI:10.1002/anie.201402011 [27] M. J. Kamlet, C. Dickinson, and R. W. Taft, Chem. Phys. Lett. 77 , 69 (1981). DOI:10.1016/0009-2614(81)85602-3 [28] M. J. Kamlet, J. L. M. Abboud, M. H. Abraham, and R. W. Taft, J. Org. Chem. 48 , 2877 (1983). DOI:10.1021/jo00165a018 [29] L. Chuntonov, I. M. Pazos, J. Ma, and F. Gai, J. Phys. Chem. B 119 , 4512 (2015). DOI:10.1021/acs.jpcb.5b00745 [30] J. Braun, H. J. Neusser, and P. Hobza, J. Phys. Chem. A 107 , 3918 (2003). DOI:10.1021/jp027217v [31] Y. Geng, T. Takatani, E. G. Hohenstein, and C. D. Sherrill, J. Phys. Chem. A 114 , 3576 (2010). DOI:10.1021/jp9099495 [32] T. V. Sravanthi, and S. L. Manju, Eur. J. Pharm. Sci. 91 , 1 (2016). DOI:10.1016/j.ejps.2016.05.025 [33] V. Sharma, P. Kumar, and D. Pathak, J. Heterocycl. Chem. 47 , 491 (2010). [34] X. Meng, T. Harricharran, and L. J. Juszczak, Photochem. Photobiol. 89 , 40 (2013). DOI:10.1111/j.1751-1097.2012.01219.x [35] E. P. Kirby, and R. F. Steiner, J. Phys. Chem. 74 , 4480 (1970). DOI:10.1021/j100720a004

a. 北京师范大学物理学系, 应用光学北京市重点实验室, 高等量子研究中心, 北京 100875;
b. 北京师范大学化学学院, 北京 100875