Source: °Ref. 1 (normalized to zero for glycine).
Source: °Ref. 1 (normalized to zero for glycine).
behavior of amino acids or small peptides. Scales that were formulated on the basis of location or buriedness of residues measured in different proteins attempt to address this problem but are still limited in universality of application, due to the need to extrapolate the behavior of residues to proteins other than those for which data are available. Some researchers (18) have proposed that different hydrophobicity coefficients should be computed for each of four structural classes of proteins (aa, /)/?, a + /?, and a/ft). In the case of food systems, the problem is even more complex, due to the simultaneous occurrence of many different proteins. Determination of a net or average hydrophobicity value for complex systems requires not only the calculation of a hydrophobicity value for each protein, but also some knowledge of how these might be changed through possible interactions between the proteins. Furthermore, calculated values fail to take into account the effects of processing on buriedness or surface exposure of the residues. For these reasons, various methods of measuring parameters that may relate to the hydrophobicity of complex food protein systems are usually favored over calculation of average values or profiles based on the constituent amino acids. Examples of such methods are illustrated in the following sections.
Partition in Aqueous Two-Phase Systems. An indication of the relative hydrophobicity of proteins may be given by their partition coefficients measured as the solubility ratio between a polar and an immiscible nonpolar solvent. Due to the virtual insolubility of most proteins in organic solvents, two-phase aqueous systems containing dextran and polyethylene glycol), or PEG, are used (22,23). Esterifi-cation of PEG with a fatty acid such as palmitate is used to alter nonpolarity of the system. The difference in partition of proteins in phase systems with and without the fatty acyl hydrocarbon group bound to PEG is taken as a measure of hydrophobic interaction. Length of the hydrocarbon chain may be varied to investigate the effects on the partition behavior of different proteins. The method has been shown to yield useful data (22-24) but suffers from the tedious nature of the procedure and difficulties in solubilizing certain proteins.
Reverse-Phase and Hydrophobic Interaction Chromatography. The relative retention times of solutes during chromatography indicate their relative solubility in or affinity for a nonpolar stationary phase versus a mobile phase of differing polarity. According to the solvophobic effect theory (25), the retention time of peptides depends mainly on their nonpolar and polar surfaces, and thus may be a good measure of hydrophobicity. In fact, this hypothesis has proved true for small peptides (9). However, chromatographic behavior of proteins is not as straightforward. Possible denaturation of proteins under the harsh solvent con ditions often required in reverse-phase chromatography has led to recommendations to use the milder conditions of hydrophobic-interaction chromatography, but the relevance of these retention data to hydrophobicity of native protein molecules has still come under question. The hydrophobicity of the stationary phase is an important parameter in maintaining native structure of the proteins upon elution (26). Nevertheless, differences in chromatographic retention between proteins have been demonstrated, and it has been suggested that the differences in binding to aliphatic versus aromatic types of adsorbent (27) may confirm the need to differentiate between these types of interactions in the hydrophobic effect (28-31).
Binding Methods. Various methods have been proposed for quantitating the binding of a nonpolar or hydrophobic ligand to proteins as a measure of the protein hydrophobicity. The ligands used have included aliphatic and aromatic hydrocarbons (32,33), sodium dodecylsulfate (34), Tween 80 (35), simple triglycerides (36), and corn oil (37). Commonly, the mixture of protein solution and ligand are incubated for a specified time to allow interaction, followed by removal of unbound or free ligand by dialysis, extraction, microfiltration, or other such techniques. The protein-bound ligand may then be quantitated by various methods, including gas chromatographic (32,33,36) or radioactive count (36) analysis, colorimetric reaction (34), or by a fluorescence probe method (37). Although the procedures are time-consuming, these binding methods can be useful in determining empirical parameters that may be relevant in assessing interaction of proteins with various nonpolar components of foods, such as flavor compounds, vitamins, pigments, fatty acids, or triglycerides.
Contact Angle Measurement. The quantitative determination of the Lifshitz-van der Waals (LW) and electron donor-acceptor or Lewis acid-base (AB) interactions that contribute to surface tension has been extended to the case of proteins. The method is based on Young's equation describing the relationship of the advancing contact angle of drops of a liquid on a flat solid surface (38). The solid surface in this case consists of protein sample prepared as flat layers. The advancing contact angles of droplets of three different well-characterized liquids (eg, water, glycerol, and a-bromonaphthalene) on the protein surface are measured. Based on Young's equation and the known surface tensions of the three liquids, the contact angles are then used to calculate the contributions of LW and AB interactions to the surface tension of the liquids on the layered protein. This approach has been used to obtain values for native hydrated proteins using only one liquid, namely drops of saline water. These values are correlated with the hydrophobicity of proteins measured as relative retention on hydrophobic chromatography (39,40). However, much higher surface tension values are obtained by contact angle measurement of a drop of protein solution on a solid polymer surface, compared with the values obtained by measuring the angle for the air-dried protein layer, or by other measurements for surface tension at an air-liquid interface such as platinum ring tensiometry, Wilhelmy plate, and pendant drop shape (39,40). This suggests that protein exposed at an interface may reorient to a much more hydrophobic configuration than found in its native hydrated state.
SPECTROSCOPIC METHODS: INTRINSIC FLUORESCENCE.
The intrinsic fluorescence spectra of proteins is primarily attributed to the aromatic amino acid residues of tryptophan, tyrosine, and phenylalanine. In practice, fluorescence from tryptophan is the most commonly studied aspect of the spectrum, because phenylalanine has a low quantum yield while tyrosine fluorescence is frequently weakened due to quenching by tryptophan residues and the protein backbone itself (41). The fluorescence of both tryptophan and tyrosine residues depends substantially on their environment, with the magnitude of fluorescence intensity as well as the wavelength of maximum fluorescence emission being sensitive to the polarity of the environment. Three spectral classes of protein tryptophan residues have been reported: the residues that are completely buried in nonpolar regions of the molecule, those that are completely exposed to the surrounding water, and those that have limited contact with water and are probably immobilized at the protein surface. The typical wavelengths of maximum emission for these three groups of tryptophan residues are 330 to 332 nm, 350 to 353 nm and 340 to 342 nm, respectively (42). More detailed information may be obtained by recording intrinsic emission spectra at different excitation wavelengths (43). Measurements of intrinsic fluorescence can give information on buriedness of aromatic residues and on the effect of interactions with other molecules (44). However, it is often difficult to directly relate this information to hydrophobic interactions as the fluorescence characteristics may also be altered by general changes in conformation.
Spectroscopic Methods: Derivative Spectroscopy. The ultraviolet absorption spectrum of proteins depends on the chromophoric properties of the constituent aromatic amino acids. Solvent perturbation and folded-unfolded difference spectra can be used to monitor the number of exposed tyrosine and tryptophan residues (41). Although the spectra of the individual chromophores are different, it is difficult to resolve quantitatively their contributions in the resulting broad absorption spectrum. Derivative spectroscopy, particularly of the second order (d2AAU2) and fourth order (d4A/dA4j, has the ability to resolve overlapping bands in the original spectrum (45-47). Moderately turbid samples can be analyzed by derivative spectroscopy since a horizontal baseline is obtained even with appreciable nonselective light scattering (45). In addition to quantitative determination of the aromatic amino acid contents, derivative spectroscopy has been proposed as a means to determine changes in polarity of the microenvironment around the chromophores. The maximum spectral shift observed in near ultraviolet second derivative spectra between solvent-exposed model compounds and side chains completely buried in an apolar protein core was found to be 5, 4, and 2 nm for tyrosine, tryptophan, and phenylalanine, respectively (46). However, for all three aromatic residues in proteins, there was no consistent correlation between absolute spectral band positions and average solvent accessibility, implying the influence of other local effects such as electrostatic interactions on the near ultraviolet spectra of proteins.
Spectroscopic Methods: Fluorescence Probes. Compounds whose quantum yields of fluorescence and wavelength of maximum emission depend on the polarity of their environment have been used to probe the hydrophobic or nonpolar nature of proteins. Figure 1 shows the chemical structures of several such fluorescence probes. The most popular types used include anionic probes of the aromatic sulfonic class, such as the amphiphilic 1-anilinonaphthalene-8-sulfonate (ANS), or its dimeric form, bis-ANS. These probes have high quantum yields of fluorescence in organic solvents but not in water; they thus fluoresce when bound to membranes or hydrophobic cavities in proteins (48,49). However, ANS fluorescence cannot always be directly correlated with hydrophobicity, as enhancement of fluorescence and a blue shift of the emission maximum have been observed in strong aqueous MgCl2 solutions (50). It has been suggested that solvents or environments that are not necessarily nonpolar but that favor the rigid, planar configuration of the ANS molecule may influence fluorescence (51).
Another category of anionic fluorescence probes is the fatty acid analogue type, including cis-parinaric acid (CPA), which has been used as a probe for proteins and biological membranes (52-54). The parinaric acids are among the few nonaromatic fluorophores known. Their similarity to natural fatty acids, nonfluorescence in water, and good Stokes shift characteristics are among the advantages of their use as probes for hydrophobic regions that may be relevant to protein-lipid interactions in food systems. Good correlation was obtained between the relative hydrophobicity values of proteins determined by CPA fluorescence, and properties related to protein-lipid interactions such as interfacial tension and emulsifying activity (55).
Titration of protein solutions with increasing concentrations of the fluorescence probe can provide information on both the number and binding constants of the hydrophobic sites (56,57). Alternatively, proteins may be compared by calculation of an index of surface hydrophobicity SD based on the initial slope of the measured relative fluorescence intensity of excess fluorescence probe as a function of increasing concentrations of protein (55). Transfer efficiency of the excitation energy from aromatic chromophores to bound probe at an adjacent hydrophobic site has also been proposed as an index of surface hydrophobicity (55).
Limitations in using anionic probes such as ANS and CPA include the possibility that electrostatic as well as hydrophobic interactions may contribute to the binding of the probes. The use of neutral or uncharged probes such as diphenylhexatriene (37), 6-propionyl-2-(dimethyl-amino)naphthalene or PRODAN (58) and Nile Red (59) may circumvent this problem.
Other Spectroscopic Methods. Most of the spectroscopic methods outlined previously provide information on the contribution to hydrophobic interactions from aromatic amino acids. Only a few methods have been proposed that can assess the involvement of aliphatic amino acids in hydrophobic interactions, including the CPA probe assay or ligand binding involving aliphatic hydrocarbons. A number of other spectroscopic methods can be used to investigate protein structure and environment of constituent amino acid residues in general. Each of these methods has particular advantages and limitations. However, consid
Was this article helpful?