11615
2006, 110, 11615-11619 Published on Web 06/01/2006
Tuning the Cooperativity of the Helix-Coil Transition by Aqueous Reverse Micelles Smita Mukherjee, Pramit Chowdhury, and Feng Gai* Department of Chemistry, UniVersity of PennsylVania, Philadelphia, PennsylVania 19104 ReceiVed: April 17, 2006; In Final Form: May 18, 2006
We show in this letter that the thermodynamic properties of helical peptides can be tuned by varying the degrees of backbone hydration. The latter was achieved by solubilizing peptides in the water pool of sodium bis(2-ethylhexyl) sulfosuccinate (AOT) reverse micelles with different water contents or w0 values. Far-UV circular dichroism measurements on a series of alanine-rich peptides indicate that the helicity of shorter peptides is significantly increased in AOT reverse micelles at low w0 values, as compared to the corresponding helical content in buffer. This result therefore corroborates the previous simulation studies suggesting that desolvation of backbone CO and NH groups increases the stability of monomeric helices. In addition, it was found that the thermal unfolding transition of these peptides can either be very noncooperative or very cooperative, depending on w0 and peptide chain length. A simple model, which considers the heterogeneous distribution of the water molecules inside the polar core of AOT reverse micelles as well as the geometric confinement effect exerted on the peptide by the reverse micelles, was used to interpret these results.
Introduction Alanine-based helical peptides have been widely used in studies aimed at understanding the thermodynamics and kinetics of the helix-coil transition.1-3 Typically, the sequence of such peptides contains XAAAA repeats, with X being a polar or charged amino acid (e.g., lysine and arginine). While Marqusee and Baldwin4,5 originally used such amino acids to increase the solubility of alanine-based peptides in aqueous solution, recent simulation studies nevertheless suggest that the side chain of such residues may play an unexpected role in stabilizing the helical conformation. For example, using conformational energy calculations, Scheraga and co-workers6,7 have shown that a lysine side chain can effectively desolvate several amide NH and CO groups, thereby strengthening the corresponding helical hydrogen bonds. As a result, the R-helix is stabilized. Furthermore, their calculations indicate that lowering the dielectric constant can also promote helix formation. Using an explicit solvent model, Garcia and co-workers8,9 have obtained similar results indicating that such side-chain-shielding effects are indeed stabilizing. Moreover, Jarrold and co-workers have demonstrated that alanine-rich peptides can form very stable helical conformations in vacuo.10 Taken together, these studies thus indicate that water molecules actually act as denaturants toward the helical conformation by competing for hydrogen bonds with the amide CO and NH groups. Therefore, in principle, one should be able to tune the folding stability and cooperativity of a helical peptide by varying the degrees of backbone hydration or the dielectric constant of the environment. While a few experiments have been carried out to test the simulation results stated above,11 a systematic study on how the thermodynamics of alanine-rich peptides vary with the * To whom correspondence should be addressed. E-mail: gai@ sas.upenn.edu.
10.1021/jp062362k CCC: $33.50
properties of water is lacking. To provide further insight into the understanding of the effect of hydration on the helix-coil transition,12 we studied the thermal unfolding transitions of a series of alanine-based peptides in aqueous reverse micelles using circular dichroism (CD) spectroscopy. Many studies13-15 have shown that the properties of water confined in reverse micelles could be quite different from those of bulk water and that it is possible to tune some of the physical characteristics by simply varying the molar ratio of water to surfactant. Thus, reverse micelles with different water contents provide a unique platform for investigating, in a systematic manner, how peptidewater interactions affect the thermodynamics of the helix-coil transition. In addition, the influence of geometrical confinements16 on the structure and stability of a biological molecule of interest encapsulated within the polar core of the reverse micelles could also be studied.17 Specifically, we have investigated the thermodynamic properties of five AKAn peptides used by Wang et al.,2 that were solubilized in the aqueous phase of reverse micelles formed by water, sodium bis(2-ethylhexyl) sulfosuccinate (AOT) and isooctane (IO).13-15 It is well-known that the size of the water pool in such AOT reverse micelles depends on w0 ) [H2O]/ [AOT],13 and the entrapped water behaves very differently compared to bulk liquid. For example, the confined water molecules lack the typical hydrogen bonded structures, differ in viscosity, and lack the translational mobility found in bulk water.13-15 In addition, the effective dielectric constant inside the water pool of AOT reverse micelles has also been shown to be lower than that of bulk water.18,19 Thus, the water pools suspended in reverse micelles have been suggested to mimic those found in confined spaces of biological systems20 and have attracted tremendous interest over the years. In particular, numerous studies have been devoted to study the structure and © 2006 American Chemical Society
11616 J. Phys. Chem. B, Vol. 110, No. 24, 2006
Letters
dynamics of the confined water molecules by employing various simulation and spectroscopic techniques and a plethora of optical probes.20-26 Consistent with previous simulation studies,6-9,27 our results indicate that dehydrating the backbone CO and NH groups of a helical peptide results in an increase in its overall helicity. However, the net effect of the AOT reverse micelles on a specific peptide can be quite complex, depending on peptide chain length, w0, and temperature. Experimental Method Materials. AOT and isooctane were purchased from Sigma Chemical Co. (St. Louis, MO). AOT was purified using standard procedures.28 Isooctane was used without further purification. Peptide Synthesis and Purification. The peptides used in the current study have the following sequence, YGAKAAAA(KAAAA)nG (i.e., AKAn peptide with n ) 1, 2, 3, 5, and 6), and were synthesized by employing the standard Fmoc protocol and purified by reverse phase HPLC. The identity of the samples was further verified by electrospray-ionization mass spectroscopy. Preparation of Samples. The purified AOT was dried under vacuum overnight before use. A 50 mM AOT/isooctane solution was prepared by dissolving the appropriate mass of AOT in isooctane, and the resultant solution was vortexed for 3 min and then sonicated for 10 min in a bath sonicator.29 To prepare the peptide-AOT/IO solutions, lyophilized peptide solid was first dissolved in Millipore water and then an appropriate aliquot of this peptide solution was added to the above AOT/IO solution to achieve the desired w0 values. Before use, this solution was further stirred for 30 min and then centrifuged for 15 min to remove any precipitates. Similarly, the peptide/buffer solution was prepared by directly dissolving lyophilized peptide solid into phosphate buffer (50 mM, pH 7). The final peptide concentration of all of the samples was determined optically by the single tyrosine absorbance at 276 nm using 276 ) 1450 cm-1 M-1 and was found to be in the range 30-123 µM. This broad range of concentration was used to verify that the results obtained were not concentration dependent. CD Spectroscopy. The far-UV CD data were collected on an AVIV 62DS spectrometer (Lakewood, NJ) using a 1 mm quartz cell. Results and Discussion Previous studies18 have shown that at low w0 values (e.g., below 10) most of the water molecules are hydrogen bonded to the negatively charged sulfosuccinate headgroups of the AOT molecules and, thus, very few free water molecules are available to interact with any guest molecule entrapped inside the water pool of the reverse micelles. For example, at w0 ) 6, there are about 300-325 water molecules per water cavity,28 of which only a small fraction (∼32%) is considered as free while the rest remains bound to the AOT headgroups.30 Since water plays an essential role in determining the structure, stability, and dynamics of biological molecules, it is expected that this limited supply of water molecules will affect the conformational and/ or thermodynamic properties of proteins and peptides solubilized inside the water pool of AOT reverse micelles.29 For instance, it is expected that AOT reverse micelles with low water content will promote alanine-based peptides to adopt helical conformations because the intrahelical hydrogen bonds in such an environment will be strengthened, owing to the simultaneous decrease in the overall dielectric constant and degrees of backbone hydration.6-9,27 Consistent with this expectation
Figure 1. Far-UV CD spectra of AKA1 peptide in pH 7 phosphate buffer (open triangles) and AOT reverse micelle at w0 ) 6 (open circles). These data were collected at 4 °C.
(Figure 1), the helicity of the AKA1 peptide, which lacks significant helical content in aqueous solution, undergoes a dramatic increase in helicity when introduced into AOT reverse micelles at w0 ) 6. While geometrical confinement exerted by the finite size of the aqueous cavity of the AOT reverse micelle could also lead to structural changes (see discussion below), we believe in the current case the confinement effect is negligible because the diameter of the water pool at w0 ) 6 is approximately 28 Å,30 much longer than the helix length of AKA1 (∼14 Å estimated by assuming that only the KAAAAKAAAA segment can form a helical conformation). Furthermore, we found that the AKA1 peptide (and also other AKAn peptides used in the current study) is insoluble in both isooctane and AOT-isooctane mixtures (without water) and the increase in helicity is independent of peptide concentration. Therefore, we attributed the increased helicity of the AKA1 peptide in AOT reverse micelles to alleviated peptide backbone hydration. Apparently, results obtained in earlier simulation studies6-9 of alanine-rich peptides in water and reverse micelles27 corroborate such a conclusion. This finding is also consistent with the fact that fluorinated alcohols, such as 2,2,2-trifluoroethanol (TFE)31-38 and 1,1,1,3,3,3-hexafluoro-propan-2-ol (HFIP),38-40 can promote helix formation in aqueous solution. While several microscopic mechanisms have been proposed, this helix-promoting ability of fluorinated alcohols could be regarded as a consequence of alcohol-water and/or alcohol-peptide interactions, which have been suggested to help reduce the degree of hydrogen bonding between peptide amide groups and water molecules.31-40 While the above result suggests that the water pool formed in AOT reverse micelles at w0 ) 6 preferentially stabilizes the helical conformation of short alanine-based peptides, the mean residue ellipticity of AKA1 at 222 nm (i.e., [θ]222, which is a commonly used indicator of the overall helicity of proteins and peptides41) and 4 °C is only about 14 000 deg cm2 dmol-1, indicating that the peptide conformation is far from being fully helical.34 Therefore, to provide a better understanding of how AOT reverse micelles and degrees of hydration affect the thermodynamics of the helix-coil transition, we further studied (a) the helicity of the AKA3 peptide in AOT reverse micelles with different w0 values and (b) the thermal unfolding transitions of a series of AKAn peptides in AOT reverse micelles at w0 ) 6. As shown (Figure 2), the far-UV CD spectra of AKA3 collected at different water concentrations indicate that its overall
Letters
Figure 2. Far-UV CD spectra of AKA3 peptide collected at 4 °C and in pH 7 phosphate buffer and AOT reverse micelles at w0 ) 4, 6, 10, and 20, as indicated.
helical content in AOT reverse micelles is significantly increased compared to its buffer counterpart, as judged by the value of [θ]222. While these results further support the idea that decreasing degrees of backbone hydration stabilizes the helical conformation of alanine-based peptides, the net effect of a reverse micelle is somewhat complicated and depends on w0. For example, the [θ]222 value of AKA3 at w0 ) 4 is smaller than those obtained at larger w0 values, suggesting that a different effect, which is destabilizing in nature, also exists. Quite likely, this destabilization effect arises from the fact that the peptide molecules can only sample a limited and confined space in reverse micelles and, as a result, the longest helical conformation that can exist in a certain water cavity is limited by the size of the latter. For AKA3, this “geometric confinement effect” would become more pronounced for water pools formed in AOT reverse micelles at w0 ) 4 because the diameter of the latter is approximately 20 Å,30 whereas the length of the fully helical structure formed by an AKA3 peptide is ∼29 Å (estimated by assuming that only the KAAAA(KAAAA)n segment can adopt a helical conformation). While it has been shown that confinement or crowding generally leads to an increase in the stability of proteins or even induces unstructured polypeptide chains to fold,17,42,43 the consequence of confinement on a helix is exactly the opposite because it prefers to adopt a linear structure. Consistent with our result, a similar confinement effect has also been observed for DNA molecules entrapped in the water cavities of reverse micelles.44 Taken together, the CD data presented in Figure 2 therefore signify the net result of two competing effects, namely, confinement and backbone dehydration, on the helicity of the AKA3 peptide. It is well established that the helix-coil transition in aqueous solution is broad and less cooperative than the thermal unfolding transition of most proteins, which commonly have a hydrophobic core. Since the above results clearly demonstrate the stabilizing effect of backbone dehydration toward the helical structures, one would expect that the thermal unfolding transition of helical peptides solubilized in the water pool of reverse micelles becomes more cooperative. Indeed, CD thermal melting experiments (Figure 3) indicate that at low w0 values, especially at w0 ) 6, the AKA3 peptide undergoes a very sharp, protein-like thermal unfolding transition, which is distinctly different from its buffer counterpart. Therefore, these results support the conclusion reached by Garcia and co-workers9 who stated that “perturbations from helical state leading to more open structures
J. Phys. Chem. B, Vol. 110, No. 24, 2006 11617
Figure 3. CD thermal melting curves of AKA3 in phosphate buffer and AOT reverse micelles with w0 ) 4, 6, 10, and 20, as indicated.
would be highly unfavorable in vacuum or low dielectric medium”. In addition, it is noticeable that the AKA3 peptide exhibits a broader transition and also a smaller [θ]222 value at low temperatures for w0 ) 4 than those observed at w0 ) 6, which may well be due to the confinement effect discussed above. While the current study does not allow us to determine the structure of the AKA3 peptide in these reverse micelles, it is conceivable that the increased confinement effect at w0 ) 4 forces the peptide to adopt conformations that contain more than one helical stretch. Interestingly, the CD data collected at w0 ) 20 indicate that the AKA3 peptide unfolds in a highly noncooperative manner under this condition, suggestive of a mechanism that leads to a gradual loss of the helical structure with increasing temperature. Since similar results were also observed for other AKAn peptides used in the current study (data not shown), this behavior is not peptide specific. While we cannot exclude other possibilities, a probable explanation of such noncooperative helix-coil transition is that it arises from a complex interplay between the inhomogeneous hydration of the peptide backbone and location of the peptide molecule inside the water pool of the reverse micelles. It has been widely suggested that three types of water populations exist in reverse micelles at higher w0 values and are often referred to as (a) bound water (i.e., that bound to AOT headgroups), (b) trapped water (i.e., trapped between polar headgroups of surfactant molecules at the interface), and (c) free water (i.e., that locates in the center region of the cavity).30 Since the diameter of the water pool (∼70 Å)30 at w0 ) 20 is roughly 2.4 times larger than the length of the fully helical conformation of the AKA3 peptide, the latter can in principle sample different kinds of water molecules or locations inside the reverse micelles, thereby leading to inhomogenity in backbone hydration. As a result, the local stability of the helix formed under such conditions becomes position dependent. For instance, the peptide may orient itself in such a way that its N-terminus is positioned near a negatively charged AOT headgroup which stabilizes the helical conformation through the mechanism of charge-helical-dipole interaction, whereas its C-terminus is positioned toward the center of the water pool where the water molecules are more bulklike. Therefore, the N-terminal region of the helix becomes more stable than the C-terminal region because of the stabilizing effect arising from both dehydration and favorable electrostatic interaction. As a result, the unfolding of the helical structure could begin from the C-terminus and incrementally propagate toward the N-
11618 J. Phys. Chem. B, Vol. 110, No. 24, 2006
Figure 4. CD thermal melting curves of AKAn peptides (n ) 1, 2, 3, 5, and 6) in AOT reverse micelles at w0 ) 6.
terminus with increasing temperature, leading to a noncooperative thermal unfolding CD profile. Consistent with this hypothesis, a recent simulation study45 suggests that the reaction free energy of a model phenol-amine proton transfer system confined in a nanocavity is dependent on the distance of the reaction complex from the cavity wall. Because the total number of water molecules decreases with decreasing w0, it is therefore expected that the inhomogeneous hydration effect discussed above becomes less important for AOT reverse micelles with smaller w0 values, as observed (Figure 3). Nonetheless, the steep pretransition CD baselines observed for w0 ) 4 and 6 indicate that there is a premelting phase before the onset of the cooperative transition. Taken together, these results demonstrate the complex effects of AOT reverse micelles on the conformation, stability, and cooperativity of the thermal unfolding transition of the AKA3 peptide. In light of these findings, it would be interesting to use an experimental technique that offers single-residue resolution, such as the infrared isotope-editing method,46,47 to examine the local conformation as well as stabilities of the peptide. Such studies should help provide further insights regarding the factors that control the cooperativity of the helix-coil transition in reverse micelles. To gain a better understanding of the confinement effect discussed above, the thermal unfolding transitions of five AKAn peptides were compared at a fixed w0 value (i.e., w0 ) 6). As shown (Figure 4), the CD thermal unfolding curves of these peptides exhibit rather interesting but different thermal melting behaviors when encapsulated inside the water pool of AOT reverse micelles at w0 ) 6. First, all peptides, except AKA1, exhibit a major and also rather sharp unfolding transition. However, the midpoint of the transition occurs at different temperatures for different peptides. Second, the overall helicity of these peptides at low temperatures (e.g., 4 °C), as judged by their mean residue ellipticity, first increases with an increase of the peptide chain length (i.e., from AKA1 to AKA3) and then decreases as the peptide chain is further lengthened (i.e., from AKA3 to AKA6). Since a monotonic increase in helicity with increasing chain length has been observed for this set of peptides in aqueous solution,2 these results therefore reinforce the idea that the effect of reverse micelles at a specific w0 value on helical peptides is a balance of two opposing effects, namely, (a) stabilization resulting from backbone dehydration and (b) destabilization arising from geometrical confinement. Since the number of water molecules in the water pool of reverse micelles at a fixed w0 value is constant, the number of water molecules
Letters per amide group therefore decreases with increasing peptide chain length. Thus, on average, the degrees of backbone hydration decrease with increasing peptide chain length and, as a result, longer peptides should exhibit a higher thermal stability. On the contrary, the destabilizing effect due to confinement increases with increasing peptide chain length. Therefore, for a given w0 value there should be an optimum peptide chain length at which the peptide shows maximum helicity and also thermal stability. In conclusion, we have studied the helix-coil transition of a series of alanine-based peptides in AOT reverse micelles using CD spectroscopy. Our results corroborate the notion suggested by prior simulation studies6-9,27 that dehydration of the peptide backbone increases the stability of the helical conformation. While decreasing w0 helps stabilize the helical conformation through the mechanism of backbone desolvation, the concomitant decrease in pool size nevertheless forces the peptide to adopt a conformation with a shorter helical stretch. On the other hand, increasing w0 helps alleviate the confinement effect, but the resultant increase in the water content leads to an increase in the degree of backbone hydration and consequently induces a decrease in helicity. Therefore, for a certain peptide, an optimum w0 value exists at which the peptide exhibits maximum helicity. Taken together, this study provides direct experimental evidence indicating that hydration of the amide groups of a helical peptide decreases its helicity and stability. While more studies, perhaps simulations, are needed to provide a better understanding of the helix-coil transition in reverse micelles, these results nevertheless suggest that AOT reverse micelles can have a rather complex effect on the stability of helical peptides. In addition, the very cooperative thermal unfolding transition observed for some of the AKAn peptides in reverse micelles is not only interesting but also useful for other applications; for example, it may be used as a host system to systematically examine the “helical propensity” of individual amino acids. Acknowledgment. We gratefully acknowledge financial support from the National Science Foundation (CHE-0094077 and DMR05-20020). References and Notes (1) Huang, C.-Y.; Getahun, Z.; Zhu, Y.; Klemke, J. W.; DeGrado, W. F.; Gai, F. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2788. (2) Wang, T.; Zhu, Y.; Getahun, Z.; Du, D.; Huang, C.-Y.; DeGrado, W. F.; Gai, F. J. Phys. Chem. B 2004, 108, 15301. (3) van Giessen, A. E.; Straub, J. E. J. Chem. Phys. 2005, 122, 024904. (4) Marqusee, S.; Baldwin, R. L. Proc. Natl. Acad. Sci. U.S.A. 1987, 84, 8898. (5) Marqusee, S.; Baldwin, R. L. Proc. Natl. Acad. Sci. U.S.A. 1989, 86, 5286. (6) Vila, J. A.; Ripoll, D. R.; Scheraga, H. A. Biopolymers 2000, 58, 235. (7) Vila, J. A.; Ripoll, D. R.; Scheraga, H. A. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 13075. (8) Garcia, A. E.; Sanbonmatsu, K. Y. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2782. (9) Ghosh, T.; Garde, S.; Garcia, A. E. Biophys. J. 2003, 85, 3187. (10) Hudgins, R. R.; Jarrold, M. F. J. Am. Chem. Soc. 1999, 121, 3494. (11) Starzyk, A.; Barber-Armstrong, W.; Sridharan, M.; Decatur, S. M. Biochemistry 2005, 44, 369. (12) Levy, Y.; Jortner, J.; Becker, O. M. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 2188. (13) Luisi, L. P. Angew. Chem., Int. Ed. Engl. 1985, 24, 439. (14) Jain, T. K.; Varshney, M.; Maitra, A. J. Phys. Chem. 1989, 93, 7409. (15) Wong, M.; Thomas, J. K.; Gratzel, M. J. Am. Chem. Soc. 1976, 98, 2391. (16) Klimov, D. K.; Newfield, D.; Thirumalai, D. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 8019. (17) Peterson, R. W.; Anbalagan, K.; Tommos, C.; Wand, A. J. J. Am. Chem. Soc. 2004, 126, 9498.
Letters (18) Freda, M.; Onori, G.; Paciaroni, A.; Santucci, A. J. Mol. Liq. 2002, 101, 55. (19) Senapati, S.; Chandra, A. J. Phys. Chem. B 2001, 105, 5106. (20) Levinger, N. E. Science 2002, 29, 1722. (21) Nucci, N. V.; Vanderkooi, J. M. J. Phys. Chem. B 2005, 109, 18301. (22) Abel, S.; Sterpone, F.; Bandyopadhyay, S.; Marchi, M. J. Phys. Chem. B 2004, 108, 19458. (23) Venables, S. D.; Huang, K.; Schmuttenmaer, C. A. J. Phys. Chem. B 2001, 105, 9132. (24) Piletic, I. R.; Moilanen, D. E.; Spry, D. B.; Levinger, N. E.; Fayer, M. D. J. Phys. Chem. A 2006, 110, 4985. (25) Sarkar, N.; Das, K.; Datta, A.; Das, S.; Bhattacharyya, K. J. Phys. Chem. 1996, 100, 10523. (26) Faeder, J.; Ladanyi, B. M. J. Phys. Chem. B 2001, 105, 11148. (27) Abel, S.; Waks, M.; Urbach, W.; Marchi, M. J. Am. Chem. Soc. 2006, 128, 382. (28) Chowdhury, P. K.; Ashby, K. D.; Datta, A.; Petrich, J. W. Photochem. Photobiol. 2000, 72, 612. (29) Raghuraman, H.; Chattopadhyay, A. Langmuir 2003, 19, 10332. (30) Maitra, A. J. Phys. Chem. 1984, 88, 5122. (31) Cammers-Goodwin, A.; Allen, T. J.; Oslick, S. L.; McClure, K. F.; Lee, J. H.; Kemp, D. S. J. Am. Chem. Soc. 1996, 118, 3082. (32) Kentsis, A.; Sosnick, T. R. Biochemistry 1998, 37, 14613.
J. Phys. Chem. B, Vol. 110, No. 24, 2006 11619 (33) Roccatano, D.; Colombo, G.; Fioroni, M.; Mark, A. E. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 12179. (34) Luo, P.; Baldwin, R. L. Biochemistry 1997, 36, 8413. (35) Buck, M. Q. ReV. Biophys. 1998, 31, 297. (36) Rajan, R.; Balaram, P. Int. J. Pept. Protein Res. 1996, 48, 328. (37) Guo, H.; Karplus, M. J. Phys. Chem. 1994, 98, 7104. (38) Hong, D.-P.; Hoshino, M.; Kuboi, R.; Goto, Y. J. Am. Chem. Soc. 1999, 121, 8427. (39) Andersen, N. H.; Dyer, R. B.; Fesinmeyer, R. M.; Gai, F.; Liu, Z.; Neidigh, J. W.; Tong, H. J. Am. Chem. Soc. 1999, 121, 9879. (40) Roccatano, D.; Fioroni, M.; Zacharias, M.; Colombo G. Protein Sci. 2005, 14, 2582. (41) Circular Dichroism: Principles and Applications, 2nd ed.; Berova, N., Nakanishi, K., Woody, R. W., Eds.; Wiley-VCH: New York, 2000; p 604. (42) Zhou, H. X.; Dill, K. A. Biochemistry 2001, 40, 11289. (43) Eggers, D. K.; Valentine, J. S. Protein Sci. 2001, 10, 250. (44) Budker, V. G.; Slattum, P. M.; Monahan, S. D.; Wolff, J. A. Biophys. J. 2002, 82, 1570. (45) Li, S.; Thompson, W. H. J. Phys. Chem. B 2005, 109, 4941. (46) Getahun, Z.; Huang, C.-Y.; Wang, T.; De Leon, B.; DeGrado, W. F.; Gai, F. J. Am. Chem. Soc. 2003, 125, 405. (47) Decatur, S. M. Acc. Chem. Res. 2006, 39, 169.
