Advertisement

Differences in Hydration Structure Around Hydrophobic and Hydrophilic Model Peptides Probed by THz Spectroscopy

  • Hanna Wirtz
  • Sarah Schäfer
  • Claudius Hoberg
  • Martina Havenith
Article
  • 237 Downloads

Abstract

We have recorded the THz spectra of the peptides NALA and NAGA as well as the amino acid leucine as model systems for hydrophobic and hydrophilic hydration. The spectra were recorded as a function of temperature and concentration and were analyzed in terms of a principal component analysis approach. NAGA shows positive absorptions with an increasing effective absorption coefficient for increasing concentrations. We conclude that NAGA due to its polar and hydrophilic structure does not have a significant influence on the surrounding water network, but is instead integrated into the water network forming a supramolecular complex. In contrast, for NALA, one hydrogen atom is substituted by a hydrophobic iso-butyl chain. We observe for NALA a decrease in absorption below 1.5 THz and a nonlinearity with a turning point around 0.75 M. Our measurements indicate that the first hydration shell of NALA is still intact at 0.75 M (corresponding to 65 water molecules per NALA). However, for larger concentrations the hydration shells can overlap, which explains the nonlinearity. For leucine, the changes in the spectrum occur at smaller concentrations. This might indicate that leucine exhibits a long-range effect on the solvating water network.

Keywords

Hydrophobic Hydration Hydrophilic Hydration Model Peptides Solvation science 

1 Introduction

Biological function of proteins requires conformational flexibility and structural dynamics mediated by long-range vibrational dynamics in the protein [1]. These large-amplitude motions of biomolecules occur on the picosecond (ps) time scale and thus fall into the region where librational, diffusional, and other collective motions of hydration water come into play. THz radiation extends from the dielectric regime, i.e., probing motions on a nanosecond time scale down to probing (sub-) picosecond motions.

Recently, we have shown that THz absorption spectroscopy can probe changes in the fast water network dynamics or collective intermolecular vibrations caused by the presence of—or interaction with—biomolecular solutes (e.g., proteins, salts, and sugars) [2, 3, 4, 5, 6, 7].

One of the fundamental questions in solvation science is the understanding of hydration dynamics of proteins. Proteins are characterized by an inhomogeneous surface topology and chemical composition which give rise to inhomogeneous hydration. Experimental as well as simulation studies revealed a blue shift in the vibrational density of states for hydration water at hydrophobic as well as hydrophilic sites, correlated with a retardation of the H-bond dynamics [8, 9]. Conformational fluctuations and local confinement of water molecules lead to a dynamical inhomogeneity in the hydration shell [10]. In this context, the term hydration funnel was established. We demonstrated that dynamics of water molecules in the active site of the catalytic domain of a model enzyme are drastically reduced [11, 12]. Additional calculations revealed a strong substrate-specific coupling between the hydration water and the protein surface.

THz measurements are always ensemble measurements. Investigation of smaller building blocks of the higher homologues enables a deeper understanding of their contribution to the overall change in hydration. Here, we want to focus on individual building blocks of the protein, i.e., hydrophobic and hydrophilic moieties. In a first THz study of the model peptides N-acetyl-leucine-amide (NALA) and N-acetyl-glycine-amide (NAGA), which are derivatives of the amino acid leucine and glycine, we showed that the changes in hydration dynamics are rather dependent on the amount of hydrophobic and hydrophilic moieties than on the size [2].

How is the surrounding water network influenced by the hydrophobic moieties or by the hydrophilic polar groups? In the past, experimental and simulation studies tried to dissect the influence of hydrophobic and hydrophilic parts [13, 14, 15, 16, 17, 18, 19].

Sasisanker and Weingärtner studied NALA using dielectric relaxation spectroscopy [20]. Their results indicated an increase of the 10 ps relaxation time of bulk water with increasing solute concentration. A more pronounced concentration dependence was observed for the 100 ps mode, which was assigned to the dipolar reorientation of the peptide and a coupling between the water and the peptide.

Hydrophobic groups are postulated to enhance the structure of the water, thus becoming “ice-like” [15, 21, 22, 23]. For the amphiphilic trimethylamine-N-oxide (TMAO), it was shown that immobilization of water molecules in the vicinity of the molecule can be attributed to the presence of hydrophobic CH3-groups with each CH3-group influencing four OH-groups of the water [13]. However, the immobilization cannot be explained by the strengthening of the H-bonds as expected for ice-like structures, since experiments revealed H-bond strengths comparable to bulk water [13, 24, 25, 26]. In contrast, various studies propose a mechanism which is based on a sterical effect. The presence of the hydrophobic moiety prevents the formation of a fivefold coordinated water molecule which in turn results in hindrance or retardation of the rotational motions [26]. Furthermore, simulations indicated that hydrophobic groups increase the amount of linear H-bonds in the first hydration layer, whereas hydrophilic and ionic solutes increase the amount of distorted H-bonds due to electrostatic interactions [27, 28].

Here, we use THz time-domain spectroscopy to investigate the solvation dynamics in the vicinity of hydrophobic (NALA) and hydrophilic (NAGA) model peptides and leucine in the frequency region from 0.2 to 2.5 THz.

2 Materials and Methods

2.1 Sample Preparation

N-acetyl-leucine-amide (NALA) was purchased from Bachem (Bachem AG, Switzerland) with a purity of > 99%. N-acetyl-glycine-amide (NAGA) was purchased from Alfa Aesar (Thermo Fisher GmbH, Germany) with 97% purity. NAGA was dried overnight by a lyophilisator to minimize the water content. NALA and NAGA were dissolved in water (HPLC grade). To increase the solubility of both solutes, the solutions were set to an acidic pH of 4 with 1 M HCl. Stock solutions of 2 M NALA and 3 M NAGA were prepared and successively diluted. Leucine was purchased from Fluka BioChemika (Honeywell Fluka, Germany) with a purity of 99% and used without purification. Due to the low solubility, 1 M HCl was used for the preparation of a 0.3 M stock solution. The structures of the investigated model peptides are shown in Fig. 1.
Fig. 1

Bond-line structures of NAGA, NALA, and leucine

2.2 THz Time-Domain Spectrometer

THz absorption measurements were carried out using a THz time-domain (THz TD) spectrometer (Fig. 2). Our custom-built setup comprises a Ti:Sa laser (MaiTai DeepSee eHP SpectraPhysics, USA), which emits mode-locked 800 nm pulses at an 80-MHz repetition rate with a pulse duration of 70 fs and an average output power of 3 W. A beam splitter splits the 800 nm pulse into a pump (96%) and a probe beam (4%). The pump pulse is propagated via a variable delay line onto a GaAs photoconductive THz antenna (TeraSED3, GigaOptics, Germany), which is biased and modulated with 20 V and 92 kHz, respectively. The emitted THz field (0.2–3 THz) is focused onto the sample cell by two off-axis parabolic mirrors. The sample cell consists of two z-cut quartz windows separated by a 100-μm thin Teflon spacer. After passing through the sample, two additional off-axis parabolic mirrors focus the transmitted pulse on a 0.5 mm (110)-ZnTe crystal, where it overlaps with the probe pulse for electro-optical detection. The electric field of the THz pulses induces birefringence in the ZnTe crystal, which is proportional to the electric field of the THz pulse. The net change of polarization of the probe beam is detected by a balanced photodetector (125 kHz, Nirvana Auto-Balanced Photoreceiver Model 2007, New Focus, USA) probing both polarization states and processed by a Lock-in amplifier (HF2-LI, Zurich Instruments, Switzerland). A fast delay stage (2 Hz, ScanDelay 150, APE, Germany) in the pump beam path allows us to observe the THz pulse in real time, covering a time window of about 17 ps. For THz measurements, we used a time constant of 1 μs with a filter of 3 dB. With these parameters, a dynamic range of 50 dB was achieved with an integration time of 30 s. Temperature was kept constant by an external thermostat with 0.1 °C accuracy.
Fig. 2

Schematic of the THz time-domain spectrometer

3 Results

3.1 Concentration-Dependent Measurements of NALA

THz spectra of NALA were recorded for concentrations between 0 M (water, pH 4) to 2 M at 20 °C. The transmitted THz pulse was measured for an empty and a filled cell (sample) in order to determine the frequency-dependent absorption coefficient α(ν). The effective absorption αeff(v, c) is obtained by subtraction of the scaled water spectrum (\( \frac{c_w}{c_{w,0}}\ast {\alpha}_{\mathrm{water}}(v) \), with cw,0 being the concentration of pure water) from the total absorption αsolution(v, c) of the solution. The effective absorption contains contributions of the solvated molecule as well as any changes in the absorption of water compared to bulk water

$$ {\alpha}^{\mathrm{eff}}\left(v,c\right)={\alpha}_{\mathrm{solution}}\left(v,c\right)-\frac{c_w}{c_{w,0}}\ast {\alpha}_{\mathrm{water}}(v) $$
(1)
In Eq. 1, cw denotes the concentration of water in the solution, which can be calculated from the measured densities ρ(c), the concentration of the solute c, and the molecular weight of the solute Ms and water Mw, respectively.
$$ {c}_w(c)=\frac{\rho (c)-{c}_s\ast {M}_s}{M_w} $$
(2)
The effective absorption coefficient is then normalized to the total molar concentration yielding the averaged extinction of the solution (Eq. 3).
$$ {\varepsilon}^{\mathrm{avg}}(v)=\frac{\alpha^{\mathrm{eff}}\left(v,\kern0.75em {c}_s\right)\ }{\left({c}_w+{c}_s\right)} $$
(3)
For the hydrophobic NALA molecule, the average extinctions for increasing mole fractions of NALA are displayed in Fig. 3. Below 1.5 THz, we observe negative extinctions εavg(v) < 0, which are more pronounced for increasing mole fractions. Above 1.5 THz, the extinctions are positive and increase with increasing concentrations. Up to a mole fraction of 0.0153, which corresponds to a concentration of 0.75 M, we observe an almost linear increase. However, above 0.75 M, the changes deviate from a linear response.
Fig. 3

Averaged extinction—compared to bulk water—for an aqueous NALA solution in water at pH 4 is plotted as a function of frequency (left) and as a function of concentration at selected frequencies (right)

In order to dissect the spectrum and identify major spectral components, we performed a principal component analysis (PCA) with the average extinction data as described in a previous publication [29]. PCA is a well-known mathematical method of multivariate data analysis [30, 31]. PCA uses an orthogonal transformation to convert a set of observations of correlated variables (here, a data set of THz spectra for a given concentration range) into a set of values of linearly uncorrelated variables called principal components. This transformation is defined in such a way that the first principal component has the largest possible variance (i.e., it accounts for as much of the variability in the data as possible), and each succeeding component in turn has the highest variance possible under the constraint that it is orthogonal to the preceding components. The resulting vectors are an uncorrelated orthogonal basis set. Thus, if each extinction spectrum would just be a (concentration dependent) multiple of a 1-mol solution spectrum, PCA would yield a single uncorrelated principal component. Any higher component would be zero within the experimental error. However, if for increasing concentrations additionally, statistically significant, new spectral features are required to describe the observed spectra—these will be contained in a second principal component. PCA is an eigenvector-based multivariate analysis method. The deduced matrices contain the eigenvectors for each eigenvalue. In PCA, the results are represented by scores, the singular values and the loadings as row or column vectors [29, 32]. Figure 4 shows the two significant principal components for NALA. Each principal component corresponds to the product of loading vector with the corresponding singular value. The first principal component (PC) accounts for 94.9% and represents the averaged spectrum, while PC 2 (5.0%) contains the changes relative to this averaged spectrum. PC 1 increases almost linearly with concentration. In comparison, PC 2 shows a nonlinear concentration response with a turning point at a mole fraction of 0.0153. The loadings of PC 2 contribute to the spectrum as a positive offset in the entire frequency range. Figure 4 (right) shows the recombined spectrum for NALA using only PC 1. The measured extinction spectrum is shown as an inset for comparison.
Fig. 4

Calculated loadings and scores from principal component analysis of NALA (left) and recombined spectrum from PC 1 (right). The inset shows the measured spectrum for comparison

The scores displayed in Fig. 4 were fitted to a simple quadratic equation. These score functions are used to deduce the so-called dilute limit. For the dilute limit, the derivative of the score function is derived at cs=0 in order to extrapolate the spectrum for c→0. For a detailed description, we refer to our previous publication [29, 33]. The spectrum in the dilute limit, i.e., at very low concentrations of NALA, is given in Fig. 5 in red.
Fig. 5

Spectrum of NALA in the dilute limit

3.2 Concentration-Dependent Measurements of NAGA

The averaged extinction for NAGA at 20 °C is positive in the frequency region between 0 and 2.5 THz and increases with increasing mole fraction. For lower frequencies (0.3 and 0.7 THz), the changes for εavg are very small (Fig. 6, right).
Fig. 6

Extinction for different mole fractions of NAGA in water pH 4 plotted as a function of frequency  from 0.25–2.5 THz (left) and as a function of concentration at discrete frequencies (right)

We find a linear increase of εavg with increasing mole fraction of NAGA. The results from the PCA with the average extinction data are shown in Fig. 7. PC 1 accounts for 99.9% variance. In contrast to NALA, the extinction spectrum of NAGA can be recombined only from PC 1 (not shown).
Fig. 7

Calculated loadings and scores from PCA for NAGA. One principal component is sufficient to reproduce the observed spectrum

3.3 Concentration-Dependent Measurements of Leucine

The concentration-dependent spectrum of the amino acid leucine at 20 °C is displayed in Fig. 8, left. Whereas, for the smallest mole fraction, which corresponds to 0.05 M, εavg is slightly positive (Fig. 8, right); for higher concentrations, we find εavg < 0.
Fig. 8

Averaged extinction of leucine plotted as a function of frequency from 0.25 to 2.5 THz (left) and as a function of concentration for discrete frequencies

PCA of the leucine extinction data yields two principal components with 93.7% (PC 1) and 5.3% (PC 2) variance. Both components show no prominent spectral features. The corresponding scores show systematic, nonlinear dependence on concentration, revealing a maximum at lower mole fractions. For leucine, recombination of the spectrum using PC 1 (Fig. 9, right) is not sufficient to describe the measured spectrum (inset), instead, a second principal component is required for a reproduction of the spectral features above 1.5 THz (Fig. 10).
Fig. 9

Spectrum of leucine in the dilute limit. A nonlinear contribution is observable for very high concentrations of leucine

Fig. 10

Results obtained from PCA for leucine. The scores indicate nonlinear changes with increasing concentration

As for NALA, we calculated the dilute limit from the score functions. In the case of leucine, we used a simple third order polynomial to fit the data. Both, the linear and nonlinear contribution show the same spectral behavior although opposite in sign. A more pronounced absorption feature is found at 1.7 THz.

3.4 Temperature-Dependent Measurements of NALA and NAGA

Temperature-dependent measurements were carried out for temperatures between 273.15 and 313.15 K for the highest soluble concentration (2 M NALA, 3 M NAGA). For the deduction of averaged extinctions, the temperature-dependent densities were used. The results for NALA and NAGA are displayed in Fig. 11. As can be seen, the spectral shape of NALA is not influenced by temperature. However, slight differences are observed: with increasing temperature, the averaged extinction is increasing. This effect is more pronounced at higher frequencies.
Fig. 11

Extinction for 2 M NALA (upper left) and 3 M NAGA (lower, left) plotted as a function of frequency for 0.25–2.25 THz and as a function of temperature at discrete frequencies

For all frequencies (except at 1.7 THz), a more pronounced change in extinction is observed between 293.15 and 303.15 K (Fig. 11, upper right). In contrast, NAGA shows almost equal extinctions for all temperatures except for 0 °C (Fig. 11, lower left).

The temperature dependence can be described by two principal components with partial contributions of PC 1 = 98.8% and PC 2 = 0.93% (Fig. 12, left). The two PCA components are very similar to the PCA of NALA (compare Fig. 4, left). The corresponding scores for PC 1 are increasing linearly with temperature. However, the scores for PC 2 show a strong nonlinear behavior. In order to determine the melting temperature, assuming two-state thermodynamics, a Boltzmann function was used to fit the scores for PC 2 [34]. The fit yielded a melting temperature of Tm = (300.13 ± 0.93) K for NALA. In contrast, for NAGA, we find a linear dependence on temperature for T > 273.15 K for PC 1, which accounts for 99.9% of the total spectrum.
Fig. 12

Results obtained from PCA for temperature-dependent NALA (left) and NAGA (right). The loadings of PC 1 (and PC 2 for NALA) are comparable to what was found for concentration dependence of NALA and NAGA. The scores (inset) indicate nonlinear contributions of PC 2 for NALA

4 Discussion

A comparison of the two molecules NALA and NAGA in respect to hydrophobicity reveals that NALA is strongly hydrophobic due to its leucine side chain (iso-butyl moiety); thus, NALA belongs to the group of amphiphilic molecules [20].

For the following discussion, we will dissect the effective absorption into xs ∗ αS, describing the absorption of the solute, and xHW ∗ (αHW − αBW) corresponding to the difference in absorption of hydration and bulk water in regard to the volume fraction x.
$$ {\alpha}^{\mathrm{eff}}={x}_s\ast {\alpha}_S+{x}_{HW}\ast \left({\alpha}_{HW}-{\alpha}_{BW}\right) $$
(4)

The hydrophobic NALA molecule exhibits negative extinctions below 1.5 THz, similar as is observed for leucine, but not for hydrophilic and polar NAGA. Above 1.5 THz, NALA shows positive extinctions like NAGA, however not for leucine.

We propose here that we can attribute the low frequency spectrum of NALA to the presence of both hydrophilic and hydrophobic moieties in the NALA molecule. According to Eq. 4, negative absorption corresponds to a decreased absorption of hydration water compared to bulk water while assuming a negligible contribution of the intramolecular solute spectrum.

NAGA shows positive absorptions with an increasing αeff(v) for increasing concentrations. This is similar as for glycine, which is the building block of NAGA. THz measurements in the frequency region from 50 to 400 cm−1 (1.5–12 THz) revealed a linear concentration dependence [18, 35]. This linearity was attributed to the presence of a very small dynamical hydration shell, implying that the molecule glycine fits very well into the water network, similar as for the polar molecule urea [14].

The influence of urea on the surrounding water network is very weak, i.e., the solvation of urea does not lead to changes in the hydrogen-bond network [27]. Neutron scattering experiments, which probe the static hydration layer around solutes, revealed no change or displacement in the distance of the so-called water ring due to solvation of NAGA [23]. Thus, we conclude that NAGA due to its polar and hydrophilic structure does not have a significant influence on the surrounding water network, but is instead very well integrated into the water network.

For NALA, one H atom is substituted by a hydrophobic iso-butyl chain. We observe for NALA a decrease in absorption below 1.5 THz and a nonlinearity with a turning point around 0.75 M. When varying the solute concentration, we cover low (2 M correspond to ~ 19 water molecules per solute) and high (0.25 M corresponds to 213 water molecules per solute) hydration levels. A concentration of 0.75 M corresponds to roughly 65 water molecules per NALA. Our measurements indicate that the first hydration shell of NALA is still intact at 0.75 M. However, for larger concentrations, the hydration shells can overlap, which explains the nonlinear contribution of PC 2. Depolarizing light scattering showed that water molecules around NALA are retarded by a factor of 8 and that this retardation is effecting roughly 57 water molecules [19]. Interestingly, this number corresponds to a concentration of 0.875 M, which is in good agreement with the turning point observed in our measurements.

Molecular dynamics simulation revealed that hydrophobic residues induce a blue-shift in the vibrational density of states which is correlated with a retardation of the hydration dynamics. This blue-shift leads to a decrease (increase) in absorption of hydration water below (above) 1.5 THz compared to bulk water as observed for NALA and leucine, but not for NAGA [8].

In our THz study of solvated alcohols, we observed two distinct hydration water peaks at 5.846 THz (195 cm−1) and 4.917 THz (164 cm−1) which we attributed to more tetrahedral and more disordered interstitial water in the vicinity of the solute [34]. We observed two-state thermodynamics with Tm = 311 K for the more disordered hydration water in the first hydration shell. Following this approach, for NALA, we deduce a transition temperature of Tm = (300.13 ± 0.93) K, which is slightly lower. This means that less energy is needed to observed in a reorganization of the hydrogen-bond network in the vicinity of the solute until a bulk-like response is observed.

In summary, the second principal component found for NALA is attributed to a spectral contribution, which is sensitive to both concentration and temperature. We propose that the wing of a spectral feature at higher frequencies contributes to the low-frequency spectra as a constant offset, which is supported by the fact that we do not observe a second principal component or nonlinearity for NAGA, but for leucine.

In the case of leucine, we observe negative extinction in the frequency region from 0.2–2.5 THz. Similar results were found for leucine for frequencies around 2.7 THz [18]. We find secondary principal components for NALA and leucine, both revealing strong nonlinearities. For leucine, the changes in the spectrum occur at smaller concentrations. This indicates that leucine exhibits a longrange effect on the solvating water network.

We attribute the negative contribution in extinction spectrum to the presence of the iso-butyl side chain of leucine. This is supported by neutron-scattering experiment that observed a shift in the scattering peak for NALA to smaller angles compared to bulk water [23]. The authors showed that the iso-butyl side chain is responsible for this shift, which was taken as a signature of a change to more linear H-bonds.

For both NALA and leucine, a nearly featureless positive contribution to αeff(v) was found which scales nonlinear with concentration. This additional contribution is attributed to hydration water in the second shell. For NALA, we can describe the temperature dependence with Tm = (300.13 ± 0.93) K.

Notes

Acknowledgments

We thank J. Savolainen for helpful discussions when setting up the THz—time-domain laser system and G. Schwaab for his support when analysing the data.

Funding Information

This work is part of the Cluster of Excellence Ruhr Explores Solvation (RESOLV) (EXC 1069) funded by the Deutsche Forschungsgemeinschaft.

References

  1. 1.
    Acbas, G., Niessen, K. A., Snell, E. H. & Markelz, A. G. Optical measurements of long-range protein vibrations. Nat. Commun. 5, 1–7 (2014).CrossRefGoogle Scholar
  2. 2.
    Born, B., Weingärtner, H., Bründermann, E. & Havenith, M. Solvation dynamics of model peptides probed by terahertz spectroscopy. observation of the onset of collective network motions. J. Am. Chem. Soc. 131, 3752–3755 (2009).CrossRefGoogle Scholar
  3. 3.
    Born, B. & Havenith, M. Terahertz dance of proteins and sugars with water. in Journal of Infrared, Millimeter, and Terahertz Waves 30, 1245–1254 (2009).Google Scholar
  4. 4.
    Heugen, U. et al. Solute-induced retardation of water dynamics probed directly by terahertz spectroscopy. Proc. Natl. Acad. Sci. U. S. A. 103, 12301–6 (2006).CrossRefGoogle Scholar
  5. 5.
    Born, B., Kim, S. J., Ebbinghaus, S., Gruebele, M. & Havenith, M. The terahertz dance of water with the proteins: the effect of protein flexibility on the dynamical hydration shell of ubiquitin. Faraday Discuss. 141, 161–173 (2009).CrossRefGoogle Scholar
  6. 6.
    Xu, Y. & Havenith, M. Perspective: watching low-frequency vibrations of water in biomolecular recognition by THz spectroscopy. J. Chem. Phys. 143, 170901 (2015).CrossRefGoogle Scholar
  7. 7.
    Heyden, M. et al. Long-range influence of carbohydrates on the solvation dynamics of water-answers from terahertz absorption measurements and molecular modeling simulations. J. Am. Chem. Soc. 130, 5773–5779 (2008).CrossRefGoogle Scholar
  8. 8.
    Heyden, M. & Havenith, M. Combining THz spectroscopy and MD simulations to study protein-hydration coupling. Methods 52, 74–83 (2010).CrossRefGoogle Scholar
  9. 9.
    Fogarty, A. C. & Laage, D. Water dynamics in protein hydration shells: the molecular origin of the dynamical perturbation. J. Phys. Chem. B 118, 7715–7729 (2014).CrossRefGoogle Scholar
  10. 10.
    Laage, D., Elsaesser, T. & Hynes, J. T. Perspective: structure and ultrafast dynamics of biomolecular hydration shells. Struct. Dyn. 4, 44018 (2017).CrossRefGoogle Scholar
  11. 11.
    Nibali, V. C. & Havenith, M. New insights into the role of water in biological function: studying solvated biomolecules using terahertz absorption spectroscopy in conjunction with molecular dynamics simulations. J. Am. Chem. Soc. 136, 12800–12807 (2014).CrossRefGoogle Scholar
  12. 12.
    Heyden, M. & Tobias, D. J. Spatial dependence of protein-water collective hydrogen-bond dynamics. Phys. Rev. Lett. 111, 218101 (2013).CrossRefGoogle Scholar
  13. 13.
    Rezus, Y. L. A. & Bakker, H. J. Observation of immobilized water molecules around hydrophobic groups. Phys. Rev. Lett. 99, 148301 (2007).CrossRefGoogle Scholar
  14. 14.
    Funkner, S., Havenith, M. & Schwaab, G. Urea, a structure breaker? Answers from THz absorption spectroscopy. J. Phys. Chem. B 116, 13374–13380 (2012).CrossRefGoogle Scholar
  15. 15.
    Murarka, R. K. & Head-Gordon, T. Dielectric relaxation of aqueous solutions of hydrophilic versus amphiphilic peptides. J. Phys. Chem. B 112, 179–186 (2008).CrossRefGoogle Scholar
  16. 16.
    Head-Gordon, T., Sorenson, J. M., Pertsemlidis, A. & Glaeser, R. M. Differences in hydration structure near hydrophobic and hydrophilic amino acids. Biophys. J. 73, 2106–2115 (1997).CrossRefGoogle Scholar
  17. 17.
    Murarka, R. K. & Head-Gordon, T. Single particle and collective hydration dynamics for hydrophobic and hydrophilic peptides. J. Chem. Phys. 126, 215101 (2007).CrossRefGoogle Scholar
  18. 18.
    Niehues, G., Heyden, M., Schmidt, D. A. & Havenith, M. Exploring hydrophobicity by THz absorption spectroscopy of solvated amino acids. Faraday Discuss. 150, 193–207 (2011).CrossRefGoogle Scholar
  19. 19.
    Comez, L. et al. More is different: experimental results on the effect of biomolecules on the dynamics of hydration water. J. Phys. Chem. Lett. 4, 1188–1192 (2013).CrossRefGoogle Scholar
  20. 20.
    Sasisanker, P. & Weingärtner, H. Hydration dynamics of water near an amphiphilic model peptide at low hydration levels: a dielectric relaxation study. ChemPhysChem 9, 2802–2808 (2008).CrossRefGoogle Scholar
  21. 21.
    Head-Gordon, T. Is water structure around hydrophobic groups clathrate-like? Proc. Natl. Acad. Sci. U. S. A. 92, 8308–8312 (1995).CrossRefGoogle Scholar
  22. 22.
    Russo, D., Murarka, R. K., Copley, J. R. D. & Head-Gordon, T. Molecular view of water dynamics near model peptides. J. Phys. Chem. B 109, 12966–75 (2005).CrossRefGoogle Scholar
  23. 23.
    Pertsemlidis, A., Saxena, A. M., Soper, A. K., Head-Gordon, T. & Glaeser, R. M. Direct evidence for modified solvent structure within the hydration shell of a hydrophobic amino acid. Proc. Natl. Acad. Sci. U. S. A. 93, 10769–10774 (1996).CrossRefGoogle Scholar
  24. 24.
    Gallagher, K. R. & Sharp, K. A. A new angle on heat capacity changes in hydrophobic solvation. J. Am. Chem. Soc. 125, 9853–9860 (2003).CrossRefGoogle Scholar
  25. 25.
    Laage, D. & Hynes, J. T. On the molecular mechanism of water reorientation. J. Phys. Chem. B 112, 14230–14242 (2008).CrossRefGoogle Scholar
  26. 26.
    Laage, D. & Hynes, JT. A molecular jump mechanism of water reorientation. Science (80). 311, 832–835 (2006).CrossRefGoogle Scholar
  27. 27.
    Sharp, K. A., Madan, B., Manas, E. & Vanderkooi, J. M. Water structure changes induced by hydrophobic and polar solutes revealed by simulations and infrared spectroscopy. J. Chem. Phys. 114, 1791–1796 (2001).CrossRefGoogle Scholar
  28. 28.
    Sharp, K. A. Water: structure and properties. in eLS (2001).Google Scholar
  29. 29.
    Sharma, V., Böhm, F., Schwaab, G. & Havenith, M. The low frequency motions of solvated Mn(II) and Ni(II) ions and their halide complexes. Phys. Chem. Chem. Phys. 16, 25101–25110 (2014).CrossRefGoogle Scholar
  30. 30.
    Wold, S., Esbensen, K. & Geladi, P. Principal component analysis. Chemom. Intell. Lab. Syst. 2, 37–52 (1987).CrossRefGoogle Scholar
  31. 31.
    Abdi, H. & Williams, L. J. Principal component analysis. WIREs Comp Stat 2, 433–459 (2010).CrossRefGoogle Scholar
  32. 32.
    Kessler, W. Multivariate Datenanalyse. (2007).Google Scholar
  33. 33.
    Decka, D., Schwaab, G. & Havenith, M. A THz/FTIR fingerprint of the solvated proton: evidence for Eigen structure and Zundel dynamics. Phys. Chem. Chem. Phys. 17, 11898–11907 (2015).CrossRefGoogle Scholar
  34. 34.
    Böhm, F., Schwaab, G. & Havenith, M. Mapping hydration water around alcohol chains by THz calorimetry. Angew. Chemie - Int. Ed. 56, 9981–9985 (2017).CrossRefGoogle Scholar
  35. 35.
    Sun, J. et al. Understanding THz spectra of aqueous solutions: glycine in light and heavy water. J. Am. Chem. Soc. 136, 5031–5038 (2014).CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  • Hanna Wirtz
    • 1
  • Sarah Schäfer
    • 1
  • Claudius Hoberg
    • 1
  • Martina Havenith
    • 1
  1. 1.Lehrstuhl für Physikalische Chemie IIRuhr Universität BochumBochumGermany

Personalised recommendations