Microwave dielectric studies on proteins,tissues, and heterogeneous suspensions

We summarize the results of several of our recent studies on the dielectric properties of protein solutions, tissues, and nonionic microemulsions at microwave frequencies extending to 18 GHz. The data in all cases are analyzed using the Maxwell mixture theory to determine the dielectric properties of the suspending water and the amount and dielectric properties of the water of hydration associated with the suspended phase. The dielectric data from the protein solutions and tissues are broadly consistent with the results of previous studies at UHF frequencies; they indicate hydration values in the range of 0.4–0.6 g water/g protein. There is evidence of a dielectric relaxation process occurring at low-GHz frequencies that can be attributed in part to dielectric relaxation of the “bound” water in the system. The remaining solvent water appears to have dielectric properties close to, if not precisely the same as, those of pure water. The average relaxation frequency of the suspending water in the microemulsions is reduced from that of pure water, evidently reflecting an average of that of the water of hydration (～5–6 GHz) and that of pure water. This reduced average relaxation frequency implies an increased average viscosity of the water and (by Walden's rule) accounts for the unexpectedly low ionic conductivity of the preparations.     

Temperature-dependent dielectric properties of slightly hydrated horn keratin

With an aim to reveal the mechanism of protein-water interaction in a predominantly two phase model protein system this study investigates the frequency and temperature dependence of dielectric constant epsilon' and loss factor epsilon' in cow horn keratin in the frequency range 30 Hz to 3 MHz and temperature range 30-200 degrees C at two levels of hydration. These two levels of hydration were achieved by exposing the sample to air at 50% relative humidity (RH) at ambient temperature and by evacuating the sample for 72 h at 105 degrees C. A low frequency dispersion (LFD) and an intermediate frequency alpha-dispersion were the two main dielectric responses observed in the air-dried sample. The LFD and the high frequency arm of the alpha-dispersion followed the same fractional power law of frequency. Within the framework of percolation cluster model these dispersions, respectively have been attributed to percolation of protons between and within the clusters of hydrogen-bonded water molecules bound to polar or ionizable protein components. The alpha-dispersion peak, which results from intra-cluster charge percolation conformed to Cole-Cole modified Debye equation. Temperature dependence of the dielectric constant in the air-dried sample exhibited peaks at 120 and 155 degrees C which have been identified as temperatures of onset of release of water bound to polar protein components in the amorphous and crystalline regions, respectively. An overall rise in the permittivity was observed above 175 degrees C, which has been identified as the onset of chain melting in the crystalline region of the protein.     

The origin and impact of bound water around intrinsically disordered proteins

Proteins and water couple dynamically over a wide range of time scales. Motivated by their central role in protein function, protein-water dynamics and thermodynamics have been extensively studied for structured proteins, where correspondence to structural features has been made. However, properties controlling intrinsically disordered protein (IDP)-water dynamics are not yet known. We report results of megahertz-to-terahertz dielectric spectroscopy and molecular dynamics simulations of a group of IDPs with varying charge content along with structured proteins of similar size. Hydration water around IDPs is found to exhibit more heterogeneous rotational and translational dynamics compared with water around structured proteins of similar size, yielding on average more restricted dynamics around individual residues of IDPs, charged or neutral, compared with structured proteins. The on-average slower water dynamics is found to arise from excess tightly bound water in the first hydration layer, which is related to greater exposure to charged groups. The more tightly bound water to IDPs correlates with the smaller hydration shell found experimentally, and affects entropy associated with protein-water interactions, the contribution of which we estimate based on the dielectric measurements and simulations. Water-IDP dynamic coupling at terahertz frequencies is characterized by the dielectric measurements and simulations.     

Rotational and translational water diffusion in the hemoglobin hydration shell: dielectric and proton nuclear relaxation measurements.

The dynamic properties of water in the hydration shell of hemoglobin have been studied by means of dielectric permittivity measurements and nuclear magnetic resonance spectroscopy. The temperature behavior of the complex permittivity of hemoglobin solutions has been measured at 3.02, 3.98, 8.59, and 10.80 GHz. At a temperature of 298 K the average rotational correlation time tau of water within a hydration shell of 0.5-nm thickness is determined from the activation parameters to be 68 +/- 10 ps, which is 8-fold the corresponding value of bulk water. Solvent proton magnetic relaxation induced by electron-nuclear dipole interaction between hemoglobin bound nitroxide spin labels and water protons is used to determine the translational diffusion coefficient D(T) of the hydration water. The temperature dependent relaxation behavior for Lamor frequencies between 3 and 90 MHz yields an average value D(298K) = (5 +/- 2) x 10(-10)m2 s-1, which is about one-fifth of the corresponding value of bulk water. The decrease of the water mobility in the hydration shell compared to the bulk is mainly due to an enhanced activation enthalpy.     

The molecular stoichiometric hydration model (SHM) as applied to tendon/collagen, globular proteins and cells

This report describes and documents the presence of multiple water-of-hydration fractions on proteins and in cells. Initial studies of hydration fractions in g of water/g of DM (dry mass) for tendon/collagen led to the development of the molecular SHM (stoichiometric hydration model) and the development of methods for calculating the size of hydration fractions on a number of different proteins of known amino acid composition. The water fractions have differences in molecular motion and other physical properties due to electrostatic interactions of polar water molecules with electric fields generated by covalently bound pairs of opposite partial charge on the protein backbone. The methods allow calculation of the size of four hydration fractions: single water bridges, double water bridges, dielectric water clusters over polar-hydrophilic surfaces and water clusters over hydrophobic surfaces. These four fractions provide monolayer water coverage. The predicted SHM hydration fractions match closely measured hydration fraction values for collagen and for globular proteins. This report also presents water sorption findings that support the SHM. The SHM is applicable for cell systems where it has been studied. In seven cell systems studied, more than half of all of the cell water had properties unlike those of bulk water. The SHM predicts and explains the commonly cited and measured bound water fraction of 0.2-0.4 g of water/g of DM on proteins. The commonly accepted concept that water beyond this bound water fraction can be considered bulk-like water in its physical properties is unwarranted.     

The influence of the molecular structure of lipid membranes on the electric field distribution and energy absorption

We consider the influence of the molecular structure of phospholipid membranes on their dielectric properties in the radio frequency range. Membranes have a stratified dielectric structure on the submolecular level, with the lipid chains forming a central hydrophobic layer enclosed by the polar headgroups (HGs) and bound water layers. In our numerical model, isotropic permittivities of 2.2 and 48.8 were assigned to the lipid chain and bound water layers, respectively. The HG region was assumed to possess an anisotropic static permittivity with 142.2 and 30.2 in the tangential and normal directions, respectively. The permittivities of the HG and bound water regions have been assumed to disperse at frequencies around 51 and 345 MHz to become 2.2 and 1.8, respectively, in both the normal and tangential directions. Electric field distribution and absorption were calculated for phospholipid vesicles with 75 nm radius as an example. Significant absorption has been obtained in the HG and bound water regions. Averaging the membrane absorption over the layers resulted in a decreased absorption below 1 GHz but a more than 10-fold increase above 1 GHz, compared to a model with a homogeneous membrane of averaged properties. We propose single particle dielectric spectroscopy by AC electrokinetics at low-bulk medium conductivities for an experimental verification of our model.     

Measuring enzyme hydration in nonpolar organic solvents using NMR

A very sensitive NMR method has been developed for measuring deuterated water bound to proteins suspended in nonpolar solvents. This has been used to determine the amount of bound water as a function of water activity for subtilisin Carlsberg suspended in hexane, benzene, and toluene and for alpha-chymotrypsin in hexane. The adsorption isotherms for subtilisin in the three solvents are very similar showing that water activity can be usefully employed to predict the amount of water bound to proteins in nonpolar organic media. Comparison of the degree of enzyme hydration reached in nonpolar solvents with that obtained in air shows that adsorption of strongly bound water is hardly affected by the low dielectric medium, but adsorption of loosely bound water is significantly reduced. This suggests that the hydrophobic regions of the protein surface are preferentially solvated by solvent molecules, and that in a nonpolar environment formation of a complete monolayer of water over the protein surface is thermodynamically unfavorable. (c) 1995 John Wiley & Sons, Inc.     

On the importance of being zwitterionic: enzymatic catalysis of decarboxylation and deprotonation of cationic carbon

Carbanion ylides are strongly stabilized by electrostatic interactions between opposing charges at neighboring atoms and this stabilizing electrostatic interaction increases with decreasing dielectric constant of the medium through which the charges interact. Consequently, there is a large increase in the thermodynamic driving force, with decreasing dielectric constant of the reaction medium, for deprotonation of cationic carbon acids and decarboxylation to form related ylides. This favors catalysis of the formation of unstable ylides at enzyme active sites of low dielectric constant. A brief survey of enzymes that catalyze deprotonation of cationic carbon acids and related decarboxylation reactions shows catalysis generally occurs for substrates that are bound in a deep pocket on the protein, with an apparent dielectric constant that is much lower than for the solvent water. In several cases, proton transfer is to a catalytic residue that is relatively weakly solvated in water. We suggest that there is a strong advantage for evolution of protein catalysts that utilize weakly solvated basic side chains which are relatively easily buried in nonpolar active sites that are favorable for zwitterion formation.     

Dielectric behavior of water in biological solutions: studies on myoglobin, human low-density lipoprotein, and polyvinylpyrrolidone

The dielectric behavior of the aqueous solutions of three widely differing macromolecules has been investigated: myoglobin, polyvinylpyrrolidone (PVP), and human serum low-density lipoprotein (LDL). It was not possible to interpret unambiguously the dielectric properties of the PVP solution in terms of water structure. The best interpretation of the dielectric data on the myoglobin and LDL solutions was that, in both cases, the macromolecule attracts a layer of water of hydration one or two water molecules in width. For LDL, this corresponds to a hydration factor of only 0.05 g/g, whereas for myoglobin the figure is nearer 0.6 g/g. With myoglobin, part of the water of hydration exhibits its dispersion at frequencies of a few GHz, and the rest disperses at lower frequencies, perhaps as low as 10-12 MHz. The approximate constancy of the width of the hydration shell for two molecules as dissimilar in size as LDL and myoglobin confirms that the proportion of water existing as water of hydration in a biological solution depends critically on the size of the macromolecules as well as on their concentration.     

α‐chymotrypsin in water–acetone and water–dimethyl sulfoxide mixtures: Effect of preferential solvation and hydration

We investigated water/organic solvent sorption and residual enzyme activity to simultaneously monitor preferential solvation/hydration of protein macromolecules in the entire range of water content at 25°C. We applied this approach to estimate protein destabilization/stabilization due to the preferential interactions of bovine pancreatic α‐chymotrypsin with water‐acetone (moderate‐strength H‐bond acceptor) and water‐DMSO (strong H‐bond acceptor) mixtures. There are three concentration regimes for the dried α‐chymotrypsin. α‐Chymotrypsin is preferentially hydrated at high water content. The residual enzyme activity values are close to 100%. At intermediate water content, the dehydrated α‐chymotrypsin has a higher affinity for acetone/DMSO than for water. Residual enzyme activity is minimal in this concentration range. The acetone/DMSO molecules are preferentially excluded from the protein surface at the lowest water content, resulting in preferential hydration. The residual catalytic activity in the water‐poor acetone is ～80%, compared with that observed after incubation in pure water. This effect is very small for the water‐poor DMSO. Two different schemes are operative for the hydrated enzyme. At high and intermediate water content, α‐chymotrypsin exhibits preferential hydration. However, at intermediate water content, in contrast to the dried enzyme, the initially hydrated α‐chymotrypsin possesses increased preferential hydration parameters. At low water content, no residual enzyme activity was observed. Preferential binding of DMSO/acetone to α‐chymotrypsin was detected. Our data clearly demonstrate that the hydrogen bond accepting ability of organic solvents and the protein hydration level constitute key factors in determining the stability of protein–water–organic solvent systems.     

Dielectric properties of hydrated collagen

Dielectric measurements have been carried out on partially hydrated collagen in the frequency ranges 100 kHz–5 MHz, 100 MHz–1 GHz, and 8–23 GHz. In the low-frequency range, a dispersion was observed around 100 kHz which results from inhomogeneous conductivity of the samples. A dielectric relaxation was observed aroud 0.3 GHz using time-domain-spectroscopy techniques. This relaxation can be considered to originate from mobile side chains. Microwave measurements indicate that the water relaxation may extend into the 10-GHz region. An apparent discrepancy between the main water relaxation time and the average rotational correlation time of water as measured by nmr line widths was resolved by the assumption that a fraction of the water molecules is bound to the collagen with residence times on the order of 10^?6 sec, whereas the remainder of the water is only weakly bound and exhibits rotational rates on the order of 10^?10 sec.     

Dielectric dispersion in aqueous solutions of oxyhaemoglobin and carboxyhaemoglobin

The relative permittivity and conductivity of aqueous solutions of oxyhaemoglobin and carboxyhaemoglobin were measured over the frequency range 150kHz-100MHz. To minimize errors of measurement the investigations were carried out with three different samples of each type of haemoglobin, independent apparatus being used in two different laboratories. The dielectric increment and relaxation time were calculated at each of several temperatures from the results. These lead to a dipole moment of 400 Debyes and an activation enthalpy of 17.6+/-1.4kJ.mol(-1), both of which were found to be independent of temperature to within experimental error over the range 3-35 degrees C. The value of the dipole moment shows that the distribution of charge throughout the haemoglobin molecule is nearly symmetrical with respect to the centre of charge. The magnitude of the activation enthalpy is similar to that of the viscosity of water, in accord with the common observation that dielectric relaxation and viscosity are related phenomena. No significant differences are found between the dielectric parameters of oxyhaemoglobin and carboxyhaemoglobin. Combining the results with those obtained from X-ray diffraction of the solid a hydration value of 0.45g of water/g of protein is suggested, subject to the limitations of the model used. Finally, the results indicate the presence of a subsidiary dispersion, which could be attributed to the above quantity of bound water having a static permittivity of about 100 and a relaxation frequency in the region 100-200MHz.     

Microwave dielectric study on bound water of globule proteins in aqueous solution

A dielectric relaxation peak due to bound water of globule proteins in aqueous solution was observed at first by the use of a time domain reflectometry. This peak locates around 100 MHz as well as that of the aqueous DNA solution and the moist collagen, and has a relaxation strength in proportion to surface of the globule protein except for trypsin and pepsin of hydrolase. It is suggested that this peak is caused by orientation of bound water molecules on the protein surface. The number of bound water molecules estimated is in good agreement with that obtained by other method such as x-ray analysis. The solution exhibits another peak below 100 MHz, which is caused by the rotation of globule protein supplemented by migration of the counterion. Its relaxation time is completely proportional to the molecular weight of the protein. © 1994 John Wiley & Sons, Inc.     

Dielectric properties of developing rabbit brain at 37 degrees C

The dielectric properties of developing rabbit brain were measured at 37 degrees C between 10 MHz and 18 GHz using time domain and frequency domain systems. The results show a variation with age of the dielectric properties of brain. An analysis of the data suggests that the water dispersion in the brain of newly born animals can be represented by a Debye equation. This dispersion increases in complexity with age, and there is evidence of a smaller additional relaxation process centered around 1 GHz. It is concluded that the principal contribution to this subsidiary dispersion region arises from water of hydration.     

Hydration Water, Charge Transport and Protein Dynamics

The hydration water of proteins is essential to biological activity but its properties are not yet fully understood. A recent study of dielectric relaxation of hydrated proteins [A. Levstik et al., Phys. Rev E.60 7604 (1999)] has found a behavior typical of a proton glass, with a glass transition of about 268 K. In order to analyze these results, we investigate the statistical mechanics and dynamics of a model of `two-dimensional water' which describes the hydrogen bonding scheme of bounded water molecules. We discuss the connection between the dynamics of bound water and charge transport on the protein surface as observed in the dielectric measurements.     

Microwave dielectric study on hydration of moist collagen

Two dielectric relaxation peaks were found in moist collagen by the time domain reflectometry. The low-frequency peak around 100 MHz moves little as the water content is varied. Its relaxation strength depends on the content and vanishes for completely dried collagen. This process is concluded to be due to water molecules strongly bound to the tropocollagen. Amount of the bound water is estimated as 0.12 g water/g collagen. Twenty-one water molecules are bound to one repeat of the triple helix. The existence of stringlike water chains is suggested. If the water content is less than 0.5 g water/g collagen, the high frequency peak locates between those of bound and bulk water. Water among the tropo-collagen is weakly bound to the collagen. In the higher region it does not change much with the content, being close to that of bulk water. The bulk water appears in this region.     

The dielectric method of investigating bound water in biological material: An appraisal of the technique

Three independent dielectric methods for the measurement of water of hydration (bound water) in a biological material are described and discussed comparatively. For well-defined aqueous solutions of biological molecules, hydration can be obtained from direct observations made on the δ dispersion or from measurement of the dielectric values of the β dispersion. For whole tissue, however, neither of these two methods is applicable, and to deduce the hydration, it is necessary to use the third technique in which the volume of the hydrated biological particle is obtained by measuring the effect of it on the known dielectric properties of pure water. The hydration can then be calculated by deducting the volume of the anhydrous particle from the experimentally determined volume of the hydrated particle. Owing to possible systemmatic errors the uncertainty in the absolute hydration value associated with this technique is rather larger than that obtained with the other two dielectric methods. For studying the differences between hydration in similar tissues, however, this objection disappears.     

Hydration of gluten: a dielectric, calorimetric, and fourier transform infrared study

The nature of the hydration of proteins and the subsequent implications for functionality is a matter of importance in both pharmaceutical and food applications. Most published studies rely on the use of one technique and attempt to characterize the system. Few studies have used combinations of techniques. In this paper we report on the use of infrared, dielectric, and calorimetric methods to examine the hydration process of wheat gluten. This has been the subject of considerable study by other techniques and has been well characterized by our group. Results show that in both the infrared and dielectric measurements there is a change in behavior at about 35% water content. This is also the water content below which lowering the temperature of the sample does not result in ice formation. We suggest that at this water content the protein amide groups are fully hydrated, and beyond this point addition of water results in protein dilution rather than further hydration.     

Water in keratin. Piezoelectric, dielectric, and elastic experiments.

To investigate actions of water in keratin, the piezoelectric, dielectric, and elastic constants are measured at 10 Hz, at temperatures between -160 and 150 degrees C, and at various hydration levels. From changes in the piezoelectric, dielectric, and dynamic mechanical parameters with moisture content (m.c.), we have identified three regimes (I, II, and III) in the hydration of water for keratin. At high hydration (21% m.c.) around 0 degree C, the piezoelectric constants for keratin steeply decrease with increasing temperature. This may be attributed to interfacial polarization which is strongly related to self-associated water molecules (particularly regime III water) just around crystalline helical regions which can exhibit the stress-induced, i.e., piezoelectric, polarization and may be attributed to electrode polarization induced by the increase of mobile ions in the amorphous matrix region, some of which would be released from their trapped states just around the piezoelectric phase by the regime III water. With increasing hydration, the elastic constants for keratin are found to increase below -70 degrees C and decrease above -70 degrees C. This suggests a viscoelastic transition of the keratin structure due to bound water (regime II water). The piezoelectric, dielectric, and elastic loss peaks are found at around -120 degrees C for hydrated keratin, believed to be due to tightly bound water (regime I water), which acts only to stiffen the keratin structure. The adsorption regions of water in keratin are discussed by a piezoelectric two-phase model, which consists of piezoelectric and nonpiezoelectric phases. It is proposed that water molecule would at least adsorb in the nonpiezoelectric phase.