Mapping Hydration Dynamics around a β-Barrel Protein - Journal of

Publication Date (Web): March 1, 2017 .... Stack Locally and Act Globally: A Few Nucleotides Make All the Difference in Enterovirus 71 IRES Binding hn...
0 downloads 0 Views 36MB Size
Subscriber access provided by Fudan University

Article

Mapping Hydration Dynamics Around a #-Barrel Protein Jin Yang, Yafang Wang, Lijuan Wang, and Dongping Zhong J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.6b12463 • Publication Date (Web): 01 Mar 2017 Downloaded from http://pubs.acs.org on March 2, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

JACS---Article (2016)

Mapping Hydration Dynamics Around a β-Barrel Protein

Jin Yang, Yafang Wang¶, Lijuan Wang and Dongping Zhong*

Department of Physics, Department of Chemistry and Biochemistry, and Programs of Biophysics, Chemical Physics and Biochemistry, The Ohio State University, Columbus, Ohio 43210, USA

*

Corresponding author. E-mail: [email protected].

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract: Protein surface hydration is fundamental to its structure, flexibility, dynamics and function, but it has been challenging to disentangle their ultimate relationships. Here, we report our systematic characterization of hydration dynamics around a β-barrel protein, rat liver fatty acid-binding protein (rLFABP), to reveal the effect of different protein secondary structures on hydration water. We employed a tryptophan scan to the protein surface one at a time and examined total 17 different sites. We observed three types of hydration water relaxations with distinct time scales, from hundreds of femtoseconds to a hundred of picoseconds. We also examined the anisotropy dynamics of the corresponding tryptophan sidechains and observed two distinct relaxations from tens to hundreds of picoseconds. Integrating our previous findings on α-helical proteins, we conclude that (1) the hydration dynamics is highly heterogeneous around the protein surface of both α-helical and β-sheet proteins. The outer-layers water of the hydration shell behaves bulk-type and relaxes in hundreds of femtoseconds. The inner-layers water collectively relaxes in two time scales, reorientation motions in a few picoseconds and network restructuring in tens-to-a hundred of picoseconds; (2) the hydration dynamics are always faster than local protein relaxations and in fact drive the protein fluctuations on the picosecond time scale; (3) the hydration dynamics in general are more retarded around β-sheet structures than α-helical motifs. A thicker hydration shell and a more rigid interfacial hydration network are observed in the β-sheet protein. Overall, these findings elucidate the intimate relationship of water-protein interactions and dynamics on the ultrafast timescale.

1

ACS Paragon Plus Environment

Page 2 of 46

Page 3 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Introduction Protein hydration water plays significant roles in mediating many biological activities, such as protein misfolding/aggregation, molecular recognition and enzymatic catalysis.1-12 To understand such roles, it is critical to characterize the dynamic nature of hydration water and to unravel their interactions with proteins. A variety of methods including experimental techniques such as NMR,13,14 neutron scattering,15-17 2D-IR18-20 and THz absorption laser spectroscopy21,22 and theoretical approaches including molecular dynamics (MD) simulations23-26 have been used to tackle this problem on different time and length scales. By integrating femtosecond (fs) fluorescence spectroscopy and site-directed mutagenesis, we developed a methodology to study protein hydration dynamics and local protein sidechain motions with femtosecond temporal resolution and single-residue spatial resolution.27-35 Using intrinsic tryptophan (Trp) as a local optical probe, we have extensively studied solvation dynamics around several proteins and observed hydration water relaxations at three different timescales.30,34 Briefly, the first ultrafast relaxation in hundreds of femtoseconds arises from the outer-layers mobile water of the hydration shell. This motion is not detected by buried probes (typically with emission maxima ≤ 338 nm) due to long distances and weak perturbations.33,34 The second relaxation occurring in a few picoseconds is attributed to the reorientation motion of water molecules in the inner-layers of the hydration shell.32-34 The third and slowest relaxation process occurs in tens-to-a hundred of picoseconds, depending on the local environments, and results from the water-network restructuring dynamics of the inner hydration shell,27-35 coupled with local protein fluctuations.

2

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Our recent studies on the proteins of SNase31,33,35 and Dpo434 have clarified the roles of hydration water and the proteins in the observed relaxation dynamics, i.e., the proteins barely contribute to the total Stokes shifts (solvation energy) but are strongly coupled with surrounding water molecules in such a way that the local protein environment alters the mobility of hydration water; however, on the other hand, hydration water motions dominate such coupling and drive local protein sidechain motions on the picosecond timescale.34,35 These findings are significant to uncover the ultimate relationship between hydration water and protein fluctuations. However, some important questions still need to be addressed. As the main machinery of most organisms, proteins are highly diverse from structure to function. The global heterogeneous hydration around protein surfaces not only is correlated with local protein chemical and structural properties, but also can be related to the overall protein structural architecture, for example, β-sheet proteins versus α-helical proteins. The distinct rigidity of protein structural elements, as shown by neutron scattering studies that β-sheet motifs appear more rigid than α-helix motifs,36,37 have invoked interest to examine how protein secondary structures affect the structure and dynamics of hydration water.38,39 Our previous studies of apomyoglobin have provided extensive details about hydration dynamics around a globular α-helical protein29,30 and showed that hydration water network is more flexible in loose loop regions than around α-helices. In this study, we report our global mapping of hydration dynamics and protein sidechain motions around a β-barrel protein, rat liver fatty acid binding protein (rLFABP) in its liganded

3

ACS Paragon Plus Environment

Page 4 of 46

Page 5 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

(holo) form, to evaluate the perturbation of β-sheet protein structures on hydration water networks. rLFABP is a member of the intracellular lipid-binding protein family and consists of 10 anti-parallel β-strands and 2 short α-helices (Figure 1).40,41 The protein can bind various fatty

Figure 1. Solution structure of rLFABP (PDB ID: 2JU8) with two bound oleic acid molecules (green color) shown in both ribbon (A and C) and surface (B and D) representations. acid ligands and is believed to facilitate intracellular lipid transport, uptake and metabolic regulation.42-48 The β-barrel forms an interior ligand-binding cavity while the helix-turn-helix region and neighboring turns serve as a portal, allowing water accessible to the protein cavity for diffusion-controlled ligand transport.42,43 The wild type (WT) rLFABP has no tryptophan residue, making it a perfect system for our study by mutations of tryptophan probe into different positions. As shown in Figure 1, a total of 17 mutants, covering every single β-strand and α-helix of the protein, were finally chosen and studied to map the global hydration dynamics around the β-barrel protein surface. By such systematic characterization, a series of dynamic correlations of water-protein relaxations with protein chemical and structural properties are obtained and compared with our previous studies on α-helical proteins. These results provide important insights into the relationship between hydration dynamics and the rigidity of protein secondary 4

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

structures.

Methods Sample Preparation. The plasmid with rLFABP gene was generously provided by Prof. Judith Storch, Rutgers University. The wild-type protein has no tryptophan and thus can be directly used as the mutant template. More than 20 mutants were initially designed around the protein surface to cover every single secondary structural unit. These mutants were expressed in Escherichia coli cells and then purified according to the protocol described in ref. 46. The proteins purified from Escherichia coli were in its liganded (holo) form with (one or) two endogenous long-chain (typically C16 and C18) fatty acids inside the cavity48 and were directly used for our experiments without any delipidation. The structures of the mutants were confirmed by their circular dichroism (CD) spectra and are similar to the wild type. The mutants were further screened by their fluorescence lifetimes to eliminate quenching ones and 17 were finally selected for the hydration study with the locations shown in Figure 1. The protein concentration used in the femtosecond-resolved experiments was 0.5-1.0 mM and the buffer condition was 10 mM Tris-HCl at pH 7.7. The steady-state fluorescence emission was measured using a SPEX FluoroMax-3 spectrometer at a sample concentration of 5-10 µM. Femtosecond-Resolved Fluorescence Spectroscopy. The fs-resolved fluorescence up-conversion method was used for all ultrafast measurements. The experimental setup has been described elsewhere in detail.49 Briefly, the pump beam was set at 292 nm with the energy of

5

ACS Paragon Plus Environment

Page 6 of 46

Page 7 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

~140 nJ per pulse before been focused into the motor-controlled sample cell. The fluorescence emission was collected by a pair of parabolic mirrors and mixed with a gating pulse at 800 nm in a 0.2-mm β-barium borate (BBO) crystal to generate an up-converted signal. The instrument response time is between 400 and 450 fs as determined from the Raman scattering of water at ~320 nm. For solvation dynamics measurements, the polarization of the pump beam was set at the magic angle (54.7°) with respect to the acceptance axis (vertical) of the BBO crystal. For fluorescence anisotropy measurements, the polarization of the pump beam was adjusted to be either parallel or perpendicular to the acceptance axis to obtain the parallel (I//) or perpendicular (I⊥) signal, respectively. The time-resolved anisotropy can thus be constructed r (t ) =

I / / (t ) − I ⊥ ( t ) I / / (t ) + 2 I ⊥ ( t )

(1)

Data Analysis. A detailed description of the data analysis method has been given elsewhere.27,30,50 Briefly, all femtosecond-resolved fluorescence transients can be best fitted by multiple exponential decays as follows: I λ (t ) = I λsolv (t ) + I λpopul (t ) = ∑ α i e − t /τ i + ∑ β j e i

− t /τ j

(2)

j

where the first term represents solvation processes and the second term describes intrinsic lifetime emissions (population decays). The prefactor αi is positive (decay dynamics) at the blue side of the emission peak and can be negative (rise dynamics) at the red side, while the prefactor βj is always positive representing relative contributions of lifetime emissions. With the steady-state emission intensity Iss(λ), the overall time-resolved fluorescence emission spectra can be constructed as 6

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

I (λ , t ) =

I SS (λ ) I λ (t ) ∑ α iτ i + ∑ β jτ j i

Page 8 of 46

(3)

j

Similarly, the lifetime-associated emission spectra can also be constructed I popul (λ , t ) =

I SS (λ ) I λpopul (t ) ∑ α iτ i + ∑ β jτ j i

(4)

j

The time-resolved spectra I(λ, t) and Ipopul(λ, t) are then converted to wavenumber domain and fitted by log-normal functions to deduce the emission maxima νS(t) and νl(t), respectively. The solvation correlation function can thus be calculated as follows: c (t ) =

ν S (t ) −ν l (t ) ν S (0) −ν l (0)

(5)

For all rLFABP mutants, the solvation correlation functions can be well fitted by triple-exponential decays, c (t ) = a1e − t /τ1S + a2 e − t /τ 2 S + a3 e − t /τ 3 S , where a1 + a2 + a3 = 1

(6)

The three terms in eq. (6) correspond to three different solvation processes and the solvation energy of each process can also be derived with the total Stokes shift ∆E,

∆Ei = ai ∆E , where ∆E = ν S (0) −ν l (0) and i = 1, 2, 3

(7)

We further define an average solvation speed to evaluate how much energy (in cm-1) drops per picosecond for each process by local solvent reorganizations, Si =

∆Ei

τ iS

, i = 1, 2, 3

(8)

The parameters Si (i = 1, 2, 3) are direct measurements of how fast the local environment responds to the external perturbation of sudden changes in the static dipole moments of

7

ACS Paragon Plus Environment

Page 9 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

tryptophan through optical excitation and reflect the flexibility of the solvation contributors, mostly hydration water networks. The fluorescence anisotropy obtained in eq. (1) directly measures the rotational dynamics of the tryptophan probe. The anisotropy dynamics usually follow multiple exponential decays due to different molecular mechanisms and can be expressed as follows:

r (t ) = rIC (t ) + r2W (t ) + r3W (t ) + rT (t )

(9)

or r (t ) = rIC0 e − t /τ IC + r20W e − t /τ 2 W + r30W e − t /τ 3W + rT0 e − t /τ T

(10)

The initial ultrafast decay τIC, typically less than 100 fs after deconvolution from the instrument response,51 is due to the internal conversion between the two concurrently excited states of tryptophan, 1La and 1Lb. The following two components (τ2W and τ3W) result from the local wobbling motions of the tryptophan probe. The last decay τT is from the overall tumbling motion of the protein in nanoseconds. The wobbling semi-angle θi can be estimated following an axially symmetric oscillation model,52 2

 3cos 2 θi − 1 riW0 1− 0 =  , i = 2, 3 2 riW + rT0  

(11)

With the semi-angle θi, we can define an average wobbling angular speed ωi, which reflects how mobile the probe is in the constrained cone.

ωi =

θi , i = 2, 3 τ iW

8

ACS Paragon Plus Environment

(12)

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 46

Results and Discussion Ultrafast Fluorescence Transients and Hydration Correlation Functions. Figure 2 shows eight selected fs-resolved fluorescence transients of mutants K80W (Figure 2A) and L50W (Figure 2B) gated from the blue to red side of their emission spectra, respectively. All transients were taken within 3-ns time window and exhibit clear features of solvation with ultrafast decays at the blue side and rises at the red side. The steady-state emission peaks of the two mutants are 342.1 nm and 331.5 nm, respectively. Thus, K80W is exposed to the solvent and L50W is buried

Figure 2. Normalized femtosecond-resolved fluorescence transients of mutants K80W (A) and L50W (B) gated at various wavelengths.

inside the protein,32-34 consistent with the structure shown in Figure 1; the side chain of

K80W points to the outside of the barrel and L50W points to the inside in direct contact with the bulky fatty acid. For K80W, we observed three solvation components with three decays in 0.42-0.67 ps, 3.2-4.3 ps and 97-139 ps at the blue side and two rises in 0.27-0.41 ps and 3.5-3.8 ps at the red side besides two lifetime decays. Comparatively, only two solvation components were observed for L50W with two decays in 3.5-7.8 ps and 68-88 ps at the blue side and one rise 9

ACS Paragon Plus Environment

Page 11 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

in 2.0-2.9 ps at the red side. Using the methodology we recently developed,27,50 the solvation correlation functions c(t) are obtained for all mutants by constructing the fs-resolved emission spectra (FRES) of tryptophan and the fitting results are given in Table S1. Figure 3 shows the

Figure 3. Solvation correlation functions c(t) of two exposed mutants, K80W and L28W, (A) and one buried mutant L50W (B).

derived

results

for

K80W, L50W and another exposed mutant L28W (λpeak = 338.9 nm). Clearly, both K80W and L28W exhibit three distinct solvation timescales: 0.44, 3.9 and 145 ps for K80W and 0.57, 6.3 and 74 ps for L28W. On the other hand, L50W shows only two components, 4.2 and 83 ps. From our extensive studies here and before,34 the first ultrafast component named τ1S in sub-picosecond is absent when the probe is buried inside the protein and becomes detectable as the probe moves to the protein surface. Thus, τ1S is attributed to ultrafast water relaxations from the outer layers of the hydration shell, typically referring to water molecules ≥ 7 Å away from the protein surface (blue water molecules in Figure 3). Such water relaxes more like bulk water 10

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 46

but is still slightly retarded with τ1S between 0.43 and 0.65 ps, compared with bulk water of 0.34 ps.50 The second solvation component named τ2S in a few picoseconds results from the collective reorientation motions of the inner-layers interfacial water, i.e., water molecules within 7 Å to the protein surface (red water molecules in Figure 3). The interfacial water is significantly slowed down by water-protein interactions compared with outer-layers bulk-type water.27-35 The third and slowest component named τ3S, ranging from tens to a hundred of picoseconds, is assigned to the subsequent cooperative water-network restructuring dynamics coupled with local protein fluctuations and is also mainly from the inner hydration shell. The observed timescales of these solvation processes vary widely among mutants, reflecting the heterogeneous nature of hydration dynamics around the protein surface, similar to the observed results of α-helical protein apomyoglobin. These dynamics are closely correlated with local protein properties, as will be discussed below.

Solvation Energies and Water-Protein Relaxations. For each solvation component τiS, the corresponding solvation energy ∆Ei can be calculated from the total Stokes shift (solvation energy) ∆E of tryptophan following eq. (7). Figure 4 shows the derived results of all 17 mutants with respect to the emission peak (λpeak) of tryptophan, an indication of the probe solvent exposure. All the solvation energies increase monotonically along with the increase of λpeak and can be fitted by sigmoid functions. As shown in Figure 4A, the total solvation energy ∆E increases gradually from 403 to 1854 cm-1 with an inflection around 338 nm. The first solvation energy ∆E1 increases from 412 to 927 cm-1 in a similar trend. Note that ∆E1 is not observed for

11

ACS Paragon Plus Environment

Page 13 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

buried mutants (L50W and I59W) and becomes noticeable first for L28W as the probe moves to the protein surface and starts to detect outer-layers hydration water. Thus, the ∆E1 component

terminates

around

338

nm.

Interestingly, the patterns of the second (∆E2, 203-590 cm-1) and third solvation energies (∆E3, 201-325 cm-1) are slightly different from ∆E and ∆E1, as shown in Figure 4B, in that the increases of ∆E2 and ∆E3 become insignificant after 338 nm. It is known that tryptophan is mainly sensitive to neighboring water molecules within a distance of ~10 Å due to dipole-dipole interactions27,53 and the increase of the solvation energy is directly

Figure 4. The solvation energies (∆E1, ∆E2, ∆E3 and the total ∆E) of all 17 mutants with respect to the steady-state emission peak λpeak.

correlated with more hydration water molecules detected. In Figure 4, we divided the emission

peaks λpeak into two regions: (I) λpeak < 338 nm (buried) and (II) λpeak ≥ 338 nm (exposed). The buried probes in region (I) detect only part of the inner hydration layers, and thus ∆E2 and ∆E3 show noticeable increments with the increase of λpeak as a result of more relaxing interfacial water molecules. When the probe moves to the surface, almost all interfacial water molecules

12

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 46

within 10 Å are probed, resulting in insensitivity of ∆E2 and ∆E3 with λpeak in region (II). Therefore, the feature of the solvation energies also reveals the origin of the corresponding solvation dynamics, indicating the dominance of hydration water contributions and the layered structure of the hydration shell. While the solvation energies reflect the integral effect of hydration water relaxations over time, the timescales directly measure the dynamics of hydrogen-bond water networks. Such interfacial water motions are clearly related to protein structural and chemical properties. Figure 5 gives the time scales (τiS) of three distinct water-network relaxations for all 17 mutants. As shown, the initial outer-layers water relaxation (τ1S, Figure 5A) occurs in 0.43-0.65 ps for all exposed mutants, slightly retarded compared with that of bulk water. The second relaxation, i.e., inner-layers water reorientations (τ2S, Figure 5B), occurs in

Figure 5. Time scales of three water-network relaxations τ1S (A), τ2S (B) and τ3S (C).

3.4-6.3 ps and shows clear relationship with local chemical identities. Specifically, for 13

ACS Paragon Plus Environment

Page 15 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

probes in hydrophobic environments such as T94W, V58W and T110W, the dynamics is fast in 3.4-4.0 ps, indicating relatively flexible water networks around the hydrophobic protein surface; for probes in the dense-charged sites such as V42W, V92W and L28W, the dynamics is much slower in 4.9-6.3 ps, reflecting stronger retardation of hydration water by neighboring charged residues. Such observations are consistent with previous studies in apomyoglobin.30 Note that no obvious correlations were observed between τ2S and the secondary structure of the protein. The three mutants in α-helices, E16W, L28W and K31W, basically spread in between the mutants in β-sheets. This is compatible with the assignment of τ2S as the collective water-network relaxation,30,34 which mostly involves local interfacial water reorientations and is thus unlikely to be influenced by the flexibility of distinct secondary structures (α-helix or β-sheet).24 Instead, local protein topologies and chemical identities are the major factors determining the slowdown of τ2S. The third relaxation, i.e., the subsequent water-network restructuring dynamics (τ3S, Figure 5C), takes much longer times ranging from 52 (K31W) to 156 ps (V92W). After the initial local relaxation, the inner-layers water networks need rearrangement to reach a new equilibrated configuration of the hydration water structure around the probe. Such rearrangement couples to local protein fluctuations and also has dynamic exchange with outer-layers water.30 As a result, the relaxation is not only correlated with local protein chemical identities but also with higher-order structural rigidity. As shown in Figure 5C, we observed: (1) for probes in the same secondary structure, i.e., α-helix (green bar) or β-sheet (orange bar), the dynamics is faster in hydrophobic sites than in partially charged and heavily charged surroundings; (2) for probes in

14

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 46

similar chemical environments, the relaxation is much slower for probes in β-sheet than in α-helix. Specifically, for mutants in the β-sheet structure, K80W and V92W with heavily charged environments show the slowest relaxation of 145 and 156 ps while the six hydrophobic sites, T94W, T110W, I59W, L50W, V58W and E68W, show faster dynamics between 79 and 95 ps. The three mutants in the α-helix, K31W, L28W and E16W, are all surrounded by certain charged residues yet exhibit similar time scales of τ3S as the hydrophobic sites in the β-sheet structure. Among the three, K31W with less surrounding charges shows the fastest dynamics of 52 ps. The results are striking and suggest the intimate coupling between hydration dynamics and protein fluctuations on the picosecond time scale. Such coupled relaxation is faster around the α-helix and slower near the β-sheet.

Anisotropy Dynamics and Protein Sidechain Motions. To examine the local Figure 6. Parallel (I//) and perpendicular (I⊥) fluorescence transients of K80W

rigidity of the

gated at 360 nm (A) and the constructed anisotropy dynamics r(t) (B).

fluorescence

probe, we measured

anisotropy

dynamics

of

the the

tryptophan, r(t). Figure 6 shows a typical result for mutant K80W with the parallel and perpendicular fluorescence transients (Figure 6A) and the

15

ACS Paragon Plus Environment

Page 17 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

derived anisotropy dynamics (Figure 6B). All anisotropy dynamics can be best fitted with four exponential decays,33,34 initial electronic relaxation (τIC),51 two intermediate local tryptophan wobbling motions (τ2W and τ3W) in tens and hundreds of picoseconds, and entire protein tumbling motion (τT) in about 7 ns. The first wobbling motion (τ2W, 8-25 ps) is ultrafast and usually occurs in small amplitude,

representing

the

immediate

shaking of the probe after excitation. Such relaxation is likely to be constrained by the local

topological

structure

as

well

as

electrostatic interactions with neighboring charges. As shown in Figure 7A, τ2W is longest for the two buried mutants, L50W

Figure 7. Correlations between τ2W and τ2S (A), τ3W and τ3S (B), and n3 and n2 (C).

and

I59W, mostly due to the spatial

constraints in the interior of the protein by the bound long-chain fatty acid. The second wobbling relaxation (τ3W, 91-632 ps) is relatively slow and spreads over a wide time range, indicating the local structural relaxation of the tryptophan

16

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 46

indole ring. The two wobbling relaxations of tryptophan have recently been found to be coupled with the two hydration water motions from inner hydration layers in the two proteins of Dpo4 and SNase using temperature-dependent studies.33,34 The timescales of the two wobbling motions (τ2W and τ3W) are shown with respect to the two hydration relaxation times (τ2S and τ3S) in Figure 7A and Figure 7B, respectively. Clearly, the sidechain relaxations are much slower than the hydration dynamics for all mutants (the dashed lines are equal for both relaxations), similar to the observations in many protein systems.30,33,34 Furthermore, our recent simultaneous measurements of hydration water and protein sidechain relaxations in Dpo4 and SNase with temperature change,34,35 have elucidated the coupling mechanism of the two dynamics, i.e., both are from the same origin with similar energy barriers, indicating that protein sidechain motions are driven by the hydration shell fluctuations. Here, we define the ratio ni = τiW/τiS (i =2, 3) as the slowing factor, similar to the concept of slowing coefficient n(T) proposed by Frauenfelder,54,55 to represent how many times the protein sidechain relaxations are slower than the hydration dynamics, or in terms of the energy landscape, how many steps of hydration water fluctuations are required to change the protein conformational substate. ni depends weakly on temperature34 but varies widely among different mutants, as shown in Figure 7C, due to protein intrinsic heterogeneity. Note that n2 and n3 are not necessarily to be identical for the same mutant because they represent transitions between different tiers of protein conformational substates. It should be noted that all ni values are always equal to or larger than 1, the nature of slaving to local protein fluctuations by hydration water networks on the picosecond time scale.

17

ACS Paragon Plus Environment

Page 19 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Solvation Speeds and Water-Protein Coupling. With the solvation energies ∆Ei and the relaxation times τiS, we can calculate the average solvation speeds Si to evaluate the rate of energy relaxation in cm-1/ps. The solvation speeds reflect the local water-network flexibility regardless of specific time and energy. The derived solvation speeds, S2 and S3, of all 17 mutants of the protein are given in Figure 8 and Table S2. As shown in Figure 8A, S2 is smallest for the two buried

Figure 8. Solvation speeds of (A) the second relaxation S2 and (B) the water-protein restructuring S3 are shown on the left column. mutants, L50W (48 cm-1/ps) and I59W (83 cm-1/ps), since only the first and most retarded hydration layer (< 5 Å from the protein surface) is detected by the tryptophan probe. On the

18

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 46

contrary, the fully exposed mutants with hydrophobic surroundings, V58W, T110W and T94W, display the largest S2 from 157 to 183 cm-1/ps. Overall, S2 increases from charged sites to hydrophobic sites, and from buried sites to solvent-exposed sites. As a result, S2 is a good indication of the local collective water-network mobility (Figure 8A) and shows strong correlations with the electrostatic interactions between interfacial hydration water and the associated protein surface. However, the relationship between S3 and the chemical properties of surrounding residues is a little complicated. The higher-order structural characteristics including α-helix and β-sheet have to be considered when evaluating S3. As shown in Figure 8B, S3 ranges from 1.9 to 3.0 cm-1/ps for the mutants with charged surroundings in β-sheets. However, with similar chemical environments, S3 has a much larger value for mutants in α-helices from 3.0 to 6.3 cm-1/ps, close to the mutants with hydrophobic surroundings in β-sheets (3.2-4.4 cm-1/ps). Clearly, both the chemical properties such as surrounding charged residues and the secondary structure rigidity are critical to S3, which evaluates the water-network restructuring dynamics from an initial nonequilibrated configuration to a final equilibrated one. Such water-network rearrangements are coupled with local protein fluctuations and thus correlated with local protein flexibility. To reflect the local protein flexibility, we calculated the angular wobbling speeds ωi of the tryptophan sidechain with the wobbling angles θi and the local wobbling times τiW. The derived results for the two intermediate wobbling motions, ω2 and ω3, are shown in Figure 8C and Figure 8D, respectively. As mentioned above, the fast relaxation of tryptophan (τ2W, 8-25 ps) represent the immediate shiver of the probe driven by local water-network reorientations (τ2S).

19

ACS Paragon Plus Environment

Page 21 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Such relaxation is typically restricted to a small semi-angle of 7-18°. ω2 examines the local mobility of the tryptophan probe, which is determined largely by local structural constraints and electrostatic interactions. As shown in Figure 8C, the buried mutants, L50W and I59W, show the smallest angular speed ω2 from 0.41 to 0.52 deg/ps, due to interior compactness of rLFABP in its holo form with bound fatty acids. For the exposed mutants, ω2 shows a gradual increase from 0.58 to 1.29 deg/ps when the electrostatic environment of the probe changes from charged to hydrophobic. On the other hand, ω3 is a measurement of the tryptophan sidechain relaxation in a relatively longer time scale (τ3W, 91-632 ps), and is related with more extended protein structural rigidity in addition to local constraints. As shown in Figure 8D, ω3 is relatively larger for the mutants in the α-helices (0.06-0.11 deg/ps) and those close to the end of a β-sheet (0.06-0.10 deg/ps). For the mutants in the middle of a β-sheet, ω3 has smaller values from 0.03 to 0.05 deg/ps. The observation is consistent with the intrinsic higher rigidity of the β-sheet structure compared with α-helix.36,39 Finally, the global mapping of inner-layers hydration dynamics and protein sidechain motions around rLFABP is given in Figure 9 with the solvation speeds, S2 and S3, and tryptophan angular speeds, ω2 and ω3. Overall, as observed in α-helical proteins,30-34 both hydration water motions and protein fluctuations show high heterogeneity among mutants and the water-driven water-protein relaxations strongly correlate with local protein chemical and structural properties.

Protein Higher-Order Structure and Hydration. We compare the global water motions around rLFABP with our previous extensive studies on an α-helical protein

20

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 9. Color scale representation of hydration dynamics (S2 and S3) and local protein sidechain relaxations (ω2 and ω3) at 17 positions of rLFABP.

21

ACS Paragon Plus Environment

Page 22 of 46

Page 23 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

apomyoglobin30 and an α/β-mixture protein Dpo434 in terms of solvation speeds and the results are shown in Figure 10. Specifically, for the outer-layers water relaxation (Figure 10A), we observed: (i) the solvation speed S1 increases with the solvent exposure of the probe, i.e., the emission peak. For example, S1 is 728 cm-1/ps for L28W with λpeak = 338.9 nm and 2156 cm-1/ps for a fully exposed mutant of V58W with λpeak = 349.5 nm; (ii) globally, the mutants on α-helix structure (Dpo4) show higher S1 than those on a β-barrel protein (rLFABP). Note that S1 for tryptophan in bulk water (λpeak = 349 nm), about 3800 cm-1/ps,34 is much larger than that of the fully exposed mutants in proteins, compatible with the protein hydration layer extending to more than 10 Å (the probe distance of tryptophan) from the surface. A thicker hydration shell around the β-barrel protein can also be inferred from the global slowdown of the outer-layers hydration dynamics. The molecular reason for such extension of hydration layers could be related to the

22

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 46

hydrogen-bond network pattern of β-barrel proteins. The solvation speed of interfacial water via reorientation relaxation S2, as shown in Figure 10B, shows an intimate

correlation

with

the

surface exposure of the probe. Such observation is a signature of the gradient retardation of the interfacial water network from the immediate vicinity of the protein to the third or fourth hydration layers. The solvation speed of the water-network

restructuring

dynamics S3 is also examined among the three proteins in

Figure 10. Comparison of the hydration dynamics around three proteins with distinct secondary structures.

Figure 10C. Significantly, S3 is

observed to be much larger around α-helix structures than around β-sheet structures except for a few mutants of apomyoglobin, showing that the water-protein network is more flexible for α-helix than β-sheet. Such observation should be correlated with the intrinsic rigidity of distinct secondary structures (α or β) since the water rearrangements on a long timescale (tens to a hundred of picoseconds) is intimately coupled with local protein fluctuations and thus influenced

23

ACS Paragon Plus Environment

Page 25 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

by more rigid protein network. The exceptional mutants in apomyoglobin, on the other hand, are found to have a dense-charged environment that also results in significant slowdown of the water-network restructuring.

Conclusion We report here our extensive studies of global hydration dynamics and related protein sidechain fluctuations around a β-barrel protein rLFABP and evaluate the hydration dynamics and water-network flexibility around different protein secondary structures. With site-directed mutagenesis, a total of 17 mutants covering every single secondary structural element of the protein were designed and studied. With femtosecond resolution, we observed three distinct relaxation dynamics from hydration water, similar to the dynamic patterns we observed for α-helical proteins. The first relaxation (τ1S) in the sub-picosecond range results from the outer-layers water of the hydration shell and is not detected by buried probes due to the long distances and weak perturbations. The outer-layers hydration water behaves like bulk-type water. Both the second and third relaxations are from the collective motions of inner-layers hydration water. The second relaxation (τ2S) in a few picoseconds represents the immediate reorientation motions of the inner-layers hydration water and is more retarded near heavily charged sites than around hydrophobic sites. The third relaxation (τ3S) in tens-to-a hundred of picoseconds is attributed to the subsequent water-network restructuring process, coupled with local protein fluctuations, and is observed to be correlated with local chemical properties and structural

24

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 46

flexibility of the protein. The protein sidechain flexibility is examined by the anisotropy dynamics of the tryptophan probe. Two relaxations of the tryptophan probe were observed, fast shaking and slow relaxation. The immediate fast shaking dynamics (τ2W) is observed to be highly restricted for mutants with strong structural constraints, such as buried inside the protein, and for mutants with charged surroundings. The longer relaxation of tryptophan sidechain (τ3W) directly reflects the flexibility of the protein, showing slower dynamics for mutants in the middle of a β-sheet and faster for mutants in a α-helix or close to the end of a β-sheet. The two relaxation dynamics of tryptophan sidechain are correlated with the two types of water motions from inner hydration layers34,35 and are always slower than the corresponding hydration water motions regardless of the local environment of the probe, indicating that the hydration water is likely to dominate the water-protein coupled relaxation and to drive local protein fluctuations on the picosecond time scale. Together with our previous studies on α-helical proteins, the characterization of hydration dynamics around the β-barrel protein rLFABP shows two significant findings, the thicker hydration shell and the more rigid water networks around the β-sheet proteins than near the α-helical proteins. These observations contribute to the fundamental understanding of the coupling between hydration water and protein in terms of secondary structures on the picosecond timescale. The observed relationship between hydration water-network inflexibility and protein higher-order structure rigidity provides insights of changes of hydration-water dynamics during

25

ACS Paragon Plus Environment

Page 27 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

protein transition from α-helical globular structures to β-sheet configurations.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publication website at DOI: Detailed information on the fitting results of hydration water dynamics and tryptophan sidechain relaxations for all 17 mutant proteins of rLFABP (PDF).

AUTHOR INFORMATION Corresponding Author *

[email protected]

Author Address ¶

Present address: School of Science, China University of Geosciences, Beijing, China 100083.

Notes The authors declare no competing financial interest.

ACKNOWLEDGEMENTS We thank Prof. Judith Storch (Rutgers University) for generously providing us the rLFABP plasmid, Dr. Yangzhong Qin for the helpful discussion, and Pearson Maugeri, Thomas J. Haver, and Spencer Heidotting for the initial help in experiment. This work was supported in part by the

26

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 46

National Institute of Health Grants GM095997 and GM118332. The MD simulations were supported in part by an allocation of computing time through the Ohio Supercomputer Center.

REFERENCES

1.

Pal, S. K.; Zewail, A. H. Chem. Rev. 2004, 104, 2099-2124.

2.

Levy, Y.; Onuchic, J. N. Annu. Rev. Biophys. Biomol. Struct. 2006, 35, 389-415.

3.

Pal, S. K.; Peon, J.; Bagchi, B.; Zewail, A. H. J. Phys. Chem. B 2002, 106, 12376-12395.

4.

Fernandez-Escamilla, A. M.; Cheung M. S.; Vega, M.C.; Wilmanns, M.; Onuchic, J. N.; Serrano, L. Proc. Natl. Acad. Sci. U. S. A. 2004, 101, 2834-2839.

5.

Kim, S. J.; Born, B.; Havenith, M.; Gruebele, M. Angew. Chem. Int. Ed. 2008, 47, 6486-6489.

6.

Zhou, R.; Huang, X.; Margulis, C. J.; Berne, B. J. Science 2004, 305, 1605-1609.

7.

Garczarek, F.; Gerwert, K. Nature 2006, 439, 109-112.

8.

Lin, J.; Balabin, I. A.; Beratan, D. N. Science 2005, 310, 1311-1313.

9.

Helms, V. ChemPhysChem 2007, 8, 23-33.

10.

Shrimpton, P.; Allemann, R. K. Protein Sci. 2002, 11, 1442-1451.

11.

Pocker, Y. Cell. Mol. Life Sci. 2000, 57, 1008-1017.

12.

Bellissent-Funel, M.-C.; Hassanali, A.; Havenith, M.; Henchman, R.; Pohl, P.; Sterpone, F.; van der Spoel, D.; Xu, Y.; Garcia, A. E. Chem. Rev. 2016, 116, 7673–7697.

13.

Nucci, N. V.; Pometun, M. S.; Wand, A. J. Nat. Struct. Mol. Biol. 2011, 18, 245-249.

14.

Armstrong, B. D.; Choi, J.; Lopez, C.; Wesener, D. A.; Hubbell, W.; Cavagnero, S.; Han, S. J. Am. Chem. Soc. 2011, 133, 5987-5995. 27

ACS Paragon Plus Environment

Page 29 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

15.

Schiro, G.; Fichou, Y.; Gallat, F. X.; Wood, K.; Gabel, F.; Moulin, M.; Hartlein, M.; Heyden, M.; Colletier, J. P.; Orecchini, A.; Paciaroni, A.; Wuttke, J.; Tobias, D. J.; Weik, M. Nat. Commun. 2015, 6, 6490.

16.

Nickels, J. D.; O’Neill, H.; Hong, L.; Tyagi, M.; Ehlers, G.; Weiss, K.L.; Zhang, Q.; Yi, Z.; Mamontov, E.; Smith, J. C.; Sokolov, A. P. Biophys. J. 2012, 103, 1566-1575.

17.

Russo, D.; Murarka, R. K.; Copley, J. R. D.; Head-Gordon, T. J. Phys. Chem. B 2005, 109, 12966-12975.

18.

Thielges, M. C.; Fayer, M. D. Acc. Chem. Res. 2012, 45, 1866-1874.

19.

King, J. T.; Kubarych, K. J. J. Am. Chem. Soc. 2012, 134, 18705-18712.

20.

Waegele, M. M.; Culik, R. M.; Gai, F. J. Phys. Chem. Lett. 2011, 2, 2598-2609.

21.

Nibali, V. C.; Havenith, M. J. Am. Chem. Soc. 2014, 136, 12800-12807.

22.

He, Y. F.; Chen, J. Y.; Knab, J. R.; Zheng, W. J.; Markelz, A. G. Biophys. J. 2011, 100, 1058-1065.

23.

Ghosh, R.; Banerjee, S.; Hazra, M.; Roy, S.; Bagchi, B. J. Chem. Phys. 2014, 141, 22D531.

24.

Sterpone, F.; Stirnemann, G.; Laage, D. J. Am. Chem. Soc. 2012, 134, 4116-4119.

25.

Li, T.; Hassanali, A. A. P.; Kao, Y. T.; Zhong, D.; Singer, S. J. J. Am. Chem. Soc. 2007, 129, 3376-3382.

26.

Nilsson, L.; Halle, B. Proc. Natl. Acad. Sci. U. S. A. 2005, 102, 13867-13872.

27.

Zhong, D. Adv. Chem. Phys. 2009, 143, 83-149.

28.

Zhong, D.; Pal, S. K.; Zewail, A. H. Chem. Phys. Lett. 2011, 503, 1-11.

29.

Zhang, L.; Wang, L.; Kao, Y. T.; Qiu, W.; Yang, Y.; Okobiah, O.; Zhong, D. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 18461-18466.

30.

Zhang, L.; Yang, Y.; Kao, Y. T.; Wang, L.; Zhong, D. J. Am. Chem. Soc. 2009, 131, 10677-10691. 28

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 46

31.

Qiu, W.; Kao, Y. T.; Zhang, L.; Yang, Y.; Wang, L.; Stites, W. E.; Zhong, D.; Zewail, A. H. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 13979-13984.

32.

Qin, Y.; Yang, Y.; Zhang, L.; Fowler, J. D.; Qiu, W.; Wang, L.; Suo, Z.; Zhong, D. J. Phys. Chem. A 2013, 117, 13926-13934.

33.

Jia, M.; Yang, J.; Qin, Y.; Wang, D.; Pan, H.; Wang, L.; Xu, J.; Zhong, D. J. Phys. Chem. Lett. 2015, 6, 5100-5105.

34.

Qin, Y.; Wang, L.; Zhong, D. Proc. Natl. Acad. Sci. U. S. A. 2016, 113, 8424-8429.

35.

Qin, Y.; Jia, M.; Yang, J.; Wang, D.; Wang, L.; Xu, J.; Zhong, D. J. Phys. Chem. Lett. 2016, 7, 4171-4177.

36.

Nickels, J. D.; Ehlers, G.; O'Neill, H.; Zhang, Q.; Sokolov, A. P. Soft Matt. 2013, 9, 9548-9556.

37.

Perticaroli, S.; Nickels, J. D.; Ehlers, G.; Sokolov, A. P. Biophys. J. 2014, 106, 2667-2674.

38.

Sinha, S. K.; Bandyopadhyay, S. J. Chem. Phys. 2011, 134, 115101.

39.

Nucci, N. V.; Pometun, M. S.; Wand, A. J. Am. Chem. Soc. 2011, 133, 12326-12329.

40.

Thompson, J.; Winter, N.; Terwey, D.; Bratt, J.; Banaszak, L. J. Biol. Chem. 1997, 272, 7140-7150.

41.

He, Y.; Yang, X.; Wang, H.; Estephan, R.; Francis, F.; Kodukula, S.; Storch, J.; Stark, R. E. Biochemistry 2007, 46, 12543-12556.

42.

Honma, Y.; Niimi, M.; Uchiumi, T.; Takahashi, Y.; Odani, S. J. Biochem. 1994, 116, 1025-1029.

43.

Li, M.; Ishibashi, T. Biomed. Res. 1992, 13, 335-341.

44.

Richieri, G. V.; Ogata, R. T.; Kleinfeld, A. M. J. Biol. Chem. 1996, 271, 31068-31074.

45.

Thompson, J.; Reese-Wagoner, A.; Banaszak, L. Biochim. Biophys. Acta. Mol. Cell Biol. Lipids 1999, 1441, 117-130.

46.

Wang, H.; He, Y.; Kroenke, C. D.; Kodukula, S.; Storch, J.; Palmer, A. G.; Stark, R. E. Biochemistry 2002, 41, 5453-5461. 29

ACS Paragon Plus Environment

Page 31 of 46

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

47.

Wang, H.; He, Y.; Hsu, K. T.; Magliocca, J. F.; Storch, J.; and Stark, R. E. J. Biomol. NMR 1998, 12, 197-199.

48.

Lowe, J. B.; Sacchettini, J. C.; Laposata, M.; McQuillan, J. J.; Gordon, J. I. J. Biol. Chem. 1987, 262, 5931-5937.

49.

Saxena, C.; Sancar, A.; Zhong, D. J. Phys. Chem. B 2004, 108, 18026-18033.

50.

Lu, W. Y.; Kim, J.; Qiu, W.; Zhong, D. Chem. Phys. Lett. 2004, 388, 120-126.

51.

Yang, J.; Zhang, L.; Wang, L.; Zhong, D. J. Am. Chem. Soc. 2012, 134,16460-16463.

52.

Steiner, R. F. In Topics in Fluorescence Spectroscopy; Lakowicz, J. R., Ed.; Plenum: New York, 1991, Vol.2, 1-51.

53.

Pal, S. K.; Peon, J.; Zewail, A. H. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 1763-1768.

54.

Fenimore, P. W.; Frauenfelder, H.; McMahon, B. H.; Parak, F. G. Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 16047-16051.

55.

Frauenfelder, H.; Chen, G.; Berendzen, J.; Fenimore, P. W.; Jansson, H.; McMahon, B. H.; Stroe, I. R.; Swenson, J.; Young, R. D. Proc. Natl. Acad. Sci. U. S. A. 2009, 106, 5129-5134.

30

ACS Paragon Plus Environment

Journal of the American Chemical Society

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 46

Figure Captions

Figure 1. Solution structure of rLFABP (PDB ID: 2JU8) with two bound oleic acid molecules (green color) shown in both ribbon (A and C) and surface (B and D) representations. rLFABP consists of 10 anti-parallel β-strands and two short α-helices. The mutation sites studied are shown in yellow color. The red and blue colors represent negatively and positively charged residues, respectively.

Figure 2. Normalized femtosecond-resolved fluorescence transients of mutants K80W (A) and L50W (B) gated at various wavelengths. Rainbow colors from purple to dark red are used to indicate the wavelengths from the blue to red side of the emission spectra. Note that the delay time is shown on a linear scale before 10 ps and a logarithmic scale thereafter.

Figure 3. Solvation correlation functions c(t) of two exposed mutants, K80W and L28W, (A) and one buried mutant L50W (B). The symbols represent the derived experimental data and the solid lines are the best exponential fits. The dashed lines show each exponential decay component (τ1S, τ2S and τ3S). Snapshots of MD simulations for K80W and L50W are shown in the right column with water molecules within 10 Å from the tryptophan indole ring. These water molecules are colored by their distances to the protein surface: blue for ≥7 Å and red for