Letter pubs.acs.org/JPCL
Observation of the Global Dynamic Collectivity of a Hydration Shell around Apomyoglobin Yangzhong Qin,§ Luyuan Zhang,§ 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, United States S Supporting Information *
ABSTRACT: Protein surface hydration is critical to the protein’s structural properties and biological activities. However, it is still unknown whether the hydration shell is intrinsically connected and how its fluctuations dynamically interact with protein motion. Here, by selecting five site-specific locations with distinctly different environments around the surface of apomyoglobin, we used a tryptophan scan with femtosecond fluorescence spectroscopy and simultaneously detected hydration water dynamics and tryptophan side-chain relaxations with temperature dependence. We observed two types of relaxations for both interfacial hydration water and the tryptophan side chain. The former is always faster than the latter, and both motions show direct linear correlations with temperature changes, indicating one origin of their motions and hydration water driving of sidechain fluctuations. Significantly, we found the relaxation energy barriers are uniform across the entire protein surface, all less than 20 kJ/mol, strongly suggesting highly extended cooperative water networks and the nature of global dynamic collectivity of the entire hydration shell.
T
with drastically different environments and local hydration dynamics. Figure 1 shows the five positions (W7, W14, A71W, A84W, and A144W) in apoMb with eight helices (A−H) by a tryptophan scan, one at a time; their fluorescence emission peaks are 329.8, 327.6, 336.8, 341.0, and 342.7 nm, respectively, at room temperature. The former three probes are buried or partially buried inside the protein surface, and the latter two are exposed to surface solvent (the dividing emission peak is at about 338 nm).13 As shown in Figure 1, both W7 and W14 are buried in a hydrophobic environment. The former, W7, is capped by the dynamic EF loop and near the flexible N terminal. The latter, W14, is at the center of the AGH hydrophobic core and solely interacts with the interior hydrophobic residues. A71W is partially buried inside the protein and located near the hemebinding pocket with a local hydrophobic protein environment. Both A84W and A144W are exposed to surface water but in two vastly different environments. A84W originally belongs to the F helix when the heme is bound in the pocket, but in the apo state, the F helix transforms into a flexible coiled structure.25 A144W is on the H-helix and is surrounded by dense charged residues. These unique environments of five mutants basically represent various typical local properties and lead to dramatically different hydration dynamics in a wide time range. According to our previous studies at room temperature,13 the long dynamics of coupled water−protein
he protein hydration shell, consisting of several layers of water molecules, forms an extended hydrogen bond network, and the inner-layer water molecules directly interact with protein atoms.1−3 The coupled water−protein interactions alter the structures and dynamics of water networks and protein conformations.4−11 Hydration water shows a dynamic ordered structure on the protein surface and its dynamics, compared to that of bulk water, significantly slows and strongly correlates to heterogeneous protein surface properties.8−10,12−15 Similarly, the protein dynamics respond to interfacial water motions and follow hydration shell fluctuations on ultrafast time scales.7,11,13,16−18 As Frauenfelder et al. proposed, proteins exist with two types of fluctuations (α and β relaxations) that are slaved by bulk-water and hydration shell motions, respectively. These water−protein interactions are vital to protein structural stability and flexibility, dynamics, and functions.3,8,9,19−24 However, the dynamic property of the hydration shell is still unknown, especially its collectivity as an entire water shell. In addition, the question remains: how does such a global connected shell, if it exists, interact with the protein’s fluctuations? We have recently reported the dynamics of hydration water and protein side chains with temperature dependence for Dpo4 polymerase (Dpo4) and Staphylococcal nuclease (SNase)17,18 and directly observed water-driven protein side-chain relaxations. Significantly, we found that the hydration shell has uniform fluctuation energy barriers with heterogeneous hydration dynamics at various sites of Dpo4.17 Here, we further examine this significant observation and carefully study the hydration shell dynamics by wide selection of five distinct positions around the protein surface of apomyoglobin (apoMb) © XXXX American Chemical Society
Received: January 26, 2017 Accepted: February 17, 2017 Published: February 17, 2017 1124
DOI: 10.1021/acs.jpclett.7b00205 J. Phys. Chem. Lett. 2017, 8, 1124−1131
Letter
The Journal of Physical Chemistry Letters
three components of 0.5−0.37 ps (τ1S), 5.9−3.4 ps (τ2S), and 67−31 ps (τ3S) for temperatures from 3 to 45 °C (Table S1 in the Supporting Information). From our extensive studies of several protein systems,9,13,17,18,27,29 these solvation dynamics mainly represent water motions around the protein surface, and the protein contributions are minor. The first ultrafast component is from the outer layers of water molecules (beyond 7 Å) in the hydration shell in hundreds of femtoseconds (blue water molecules in Figure 2B and Table S2), and the latter two components on the picosecond time scales are from interfacial inner-layer water relaxations (red and dark-red water molecules in Figure 2B and Table S2), i.e., about 2 inner layers (within 7 Å) in the hydration shell.17,18 Both components are from collective water-network relaxations, mainly reorientational motions in a few picoseconds and rearrangements coupled with the protein in tens of picoseconds.9,10 Similarly, we observed three hydration dynamics for the mutant A144W but having quite different time scales (Table S1). For the buried probes, we cannot observe the dynamics of outer-layer mobile water molecules and detect only inner-layer interfacial ones in the hydration shell. Figure 2C shows the obtained hydration water dynamics for W7 at three different temperatures. We observed only the two water relaxation dynamics (red and dark-red water molecules within 7 Å in Figure 2C and Table S2) in 7.7−3.6 ps (τ2S) and 189− 65 ps (τ3S) with temperature increase (Table S1). Also, the mutants of W14 and A71W show only two hydration water dynamics with different relaxation times (Table S1). Clearly, these hydration dynamics at five different environments show drastically different time scales, indicating the very heterogeneous nature of the hydration dynamics around the entire protein surface. These dynamics are well-correlated with local protein chemical identities and structural properties.12,13,17 We have recently studied the tryptophan side-chain motions of Dpo4 polymerase by measuring its anisotropy dynamics and have found that all side-chain motions at different positions are slower than the corresponding hydration water relaxations. With temperature dependence, both the hydration water and tryptophan side chain have the same relaxation barriers, leading us to conclude that the hydration shell slaves protein side-chain fluctuations on the picosecond time scales.17 Here, we measured the anisotropy dynamics of a buried probe W7 with temperature changes. Figure 2D shows the parallel and perpendicular fluorescence transients, and Figure 2E is the derived anisotropy dynamics in three temperatures. These relaxations have to be fit by four exponential components: the initial component (τIC < 100 fs) results from the ultrafast internal conversion between 1La and 1Lb states of tryptophan through conical intersection,30−32 and the slowest component (τT) in nanoseconds reflects the whole protein tumbling motion. Two other components represent the protein sidechain relaxations in 12.4−5.8 ps (τ2W) and 254−86 ps (τ3W) from 1 to 40 °C (Table S1), respectively, and similarly, they are slaved by the two corresponding hydration dynamics observed above (also see below). Figure 3A shows the obtained time scales of the hydration dynamics for the five mutants with temperature dependence. For the three buried probes of W7, W14, and A71W, we detect only two interfacial water dynamics on the picosecond time scales, and for the two exposed probes of A84W and A144W, we can also detect outer-layer, bulk-type mobile water motions in hundreds of femtoseconds besides the two interfacial innerlayer water dynamics. The A144W mutant is surrounded by
Figure 1. Probing the global heterogeneous surface hydration dynamics of apoMb. (Top) Structure of the native state of apoMb at pH6, sharing a structure similar to that of holoMb (PDB ID code 1MBD) consisting of eight helices A−H, with the F helix becoming a random coil. Five mutants (yellow spheres), each with a special environment, detect the local hydration dynamics. The right five panels show the surface map of each mutant within 12 Å from tryptophan. The charged residues are red (negative) and blue (positive), and the hydrophobic residues are white. (Bottom) The steady-state emission spectra at room temperature confirm distinct local environments of the five mutants, which are separated into two groups (exposed and buried) according to their emission peaks with the dividing line at ∼338 nm (gray dashed line).
relaxations are 38, 58, 87, 103, and 209 ps for A84W, A71W, W7, W14, and A144W, respectively. Here, we will study these dynamics with temperature changes to elucidate the nature of the dynamic hydration shell and the consequent interactions with protein fluctuations. For each of the five mutants, we measured the solvation dynamics at five different temperatures from ∼1 to ∼45 °C. A series of femtosecond-resolved fluorescence transients were taken at various wavelengths from the blue side to the red end of the emission spectra. Figure 2A gives an example of the gated fluorescence transients of A84W mutant at 22 °C, showing a typical solvation signature of the decay at the shorter wavelengths and the rise at the longer wavelengths. Because the A84W mutant is exposed to solvent, besides two long lifetime components, we observed three decay components of the solvation dynamics in 0.35−0.45, 3.5−4.6, and 38−52 ps at the blue side of the emission. At the red side, we observed two initial rise components in 0.4−0.45 and 3−5 ps. According to the methodology we developed,26−28 we can construct the femtosecond-resolved emission spectra (FRES) and thus obtain the energy relaxation correlation function of C(t), a direct measurement of the solvation dynamics. Figure 2B shows the derived correlation functions for A84W mutant at three different temperatures. Clearly, the solvation dynamics exhibit 1125
DOI: 10.1021/acs.jpclett.7b00205 J. Phys. Chem. Lett. 2017, 8, 1124−1131
Letter
The Journal of Physical Chemistry Letters
Figure 2. Dynamics of hydration water relaxations and protein side-chain motions. (A) Normalized fluorescence transients of A84W mutant gated at various wavelengths ranging from 305 to 360 nm of the emission spectrum at 22 °C with excitation at 290 nm. (B and C) Solvation correlation functions of exposed A84W and buried W7 mutants at three selected temperatures. The scattered points indicate the peak shifts of the emission spectra constructed from the fluorescence transients, and the solid lines are the best multiexponential fits. The dashed lines indicate the corresponding solvation components from subpicosecond (τ1S) to a few picoseconds (τ2S) and tens-to-hundreds of picoseconds (τ3S). Note that only the exposed mutants detect the ultrafast component (τ1S). The corresponding local protein structures (white) and hydration water molecules are also shown from MD simulations. Protein surface hydration water can be categorized into three groups: first layer 7 Å (blue) from the protein surface. Only the exposed mutants detect the outer-layer mobile water. (D) Two fluorescence transients (parallel and perpendicular) were measured to construct the anisotropy dynamics of the tryptophan side chain for W7 mutant at 1 °C. The inset shows a close-up view on a shorter time scale. (E) Constructed anisotropy dynamics for W7 at three temperatures, well-fitted by four exponentials (dashed lines) and indicating multiple relaxations of the tryptophan side chain, ultrafast internal conversion (τIC) in
