Changes in the secondary structure of bovine casein by Fourier transform infrared spectroscopy: effects of calcium and temperature.

Bovine casein submicelles and reformed micelles, produced by addition of Ca2+, were examined by Fourier transform infrared spectroscopy at 15 and 37 degrees C in aqueous salt solutions of K+ and Na+. Previous measurements of caseins, made in D2O and in the solid form, can now be made in a more realistic environment of H2O. When analyzed in detail, data obtained by Fourier transform infrared spectroscopy have the potential to show subtle changes in secondary structural elements that are associated with changes in protein environment. Electrostatic binding of Ca2+ to casein resulted in a redistribution of the components of the infrared spectra. Addition of Ca2+ in salt solutions of K+ and Na+ led to apparent decreases in large loop or helical structures at 37 degrees C with concomitant increases in the percentage of structures having greater bond energy, such as turns and extended helical structures. At 15 degrees C, Na+ and K+ have differential effects on the Ca(2+)-casein complexes. All of these observations are in accordance with the important role of serine phosphate side chains as sites for Ca2+ binding in caseins and the swelling of the casein structure upon incorporation into reformed micelles at 37 degrees C. This new open, hydrated structure is buttressed by a change in backbone as evidenced by a shift in absorbance to higher wave numbers (greater bond energies) as colloidal micelles are reformed.

Salt and pH effects on electrochemistry of myoglobin in thick films of a bilayer-forming surfactant.

Salt concentration and pH of external solutions were shown to control the electrochemistry of the heme protein myoglobin (MbFe(III)-H2O) in stable, ordered films of didodecyldimethylammonium bromide (DDAB). Protonation of aquometmyoglobin (MbFe(III)-H2O) in these films precedes electron transfer from electrodes, causing formal potentials to shift negative as pH increases from 5 to 8. At pH > 8, MbFe(III)-H2O dissociates to MbFe(III)-OH, which is reduced directly at the electrode at higher rates than MbFe(III)-H2O. Correlations of voltammetric data with FT-IR spectra suggested that at pH < 4.6, an unfolded form of Mb resides in the films and is reduced directly. The concentration of salt in solution influences electrochemical properties of Mb-DDAB films by its influence on Mb conformation and by effects on interfacial Donnan potentials. NMR indicated strong binding of anions to Mb within DDAB films. Bound anions may neutralize positive charge on Mb's surface so that it can reside in a partly hydrophobic environment, as postulated on the basis of previous ESR and linear dichroism studies.

Reproducibility of the nonlinear regression fit to the FTIR spectra of lysozyme.

Controversy exists concerning the influence of experimental artifacts on the number of component FTIR vibrational bands which may be resolved from the amide I and II envelopes of proteins in water. Whether these bands represent unique populations of vibrating protein groups in a particular global 2 degrees structure or whether the bands are due to instrumental and environmental fluctuations has been addressed, T.F. Kumosinski and J.J. Unruh, Talanta, 43 (1996) 199-219. The repeatability of the methodology and the apparent uniqueness of the nonlinear regression fits are addressed in this study. We obtained a series of the spectra of lysozyme, and carried-out nonlinear regression analysis of each spectrum. Coefficients of variation (COV) were calculated for the wavenumber and area values of assigned component peaks obtained. Low COVs obtained attest to the precision of the methodology and the apparent uniqueness of the nonlinear regression fits. This methodology for acquisition and analysis of protein FTIR spectra yields results with good precision.

Particle sizes of purified kappa-casein: metal effect and correspondence with predicted three-dimensional molecular models.

kappa-Casein as purified from bovine milk exhibits a rather unique disulfide bonding pattern as revealed by SDS-PAGE. The disulfide-bonded caseins present range from dimer to octamer and above and preparations contain about 10% monomer. All of these heterogenous polymers, however, self-associated into nearly spherical uniform particles with an average radius of 8.9 nm as revealed by negatively stained transmission electron micrographs. Evidence is presented that multivalent cations play a role in the stabilization of these spherical particles. Treatment with EDTA causes disruption of the kappa-casein particles and leads to a broder size distribution as judged by electron microscopy and dynamic light scattering. The size and shape of the particles are in accord with earlier proposed 3D models for kappa-casein that actually predicted participation of divalent cations in the structure.

Quantitation of the global secondary structure of globular proteins by FTIR spectroscopy: Comparison with X-ray crystallographic structure.

Fourier transform infrared spectroscopy (FTIR) is potentially a powerful tool for determining the global secondary structure of proteins in solution, providing the spectra are analyzed using a statistically and theoretically justified methodology. We have performed FTIR experiments on 14 globular proteins and two synthetic polypeptides whose X-ray crystal structures are known to exhibit varying types and amounts of secondary structures. Calculation of the component structural elements of the vibrational bands was accomplished using nonlinear regression analysis, by fitting both the amide I and amide II bands of the Fourier self-deconvoluted spectra, the second-derivative spectra, and the original spectra. The methodology was theoretically justified by comparing (via nonlinear regression analysis) the global secondary structure determined after deconvolving into component bands the vibrational amide I envelopes with the calculated structure determined by first principles from Ramachandran analysis of the X-ray crystallographic structure of 14 proteins from the Brookhaven protein data bank. Justification of the nonlinear regression analysis model with respect to experimental and instrumental considerations was achieved by the decomposition of all the bands of benzene and an aqueous solution of ammonium acetate into component bands while floating the Gaussian/Lorentzian character of the line shapes. The results for benzene yield all pure Lorentzian line shapes with no Gaussian character while the ammonium acetate spectra yielded all Gaussian line shapes with no Lorentzian character. In addition, all-protein spectra yielded pure Gaussian line shapes with no Lorentzian character. Finally, the model was statistically justified by recognizing random deviation patterns in the regression analysis from all fits and by the extra sum of squares F-test which uses the degrees of freedom and the root mean square values as a tool to determine the optimum number of component bands required for the nonlinear regression analysis. Results from this study demonstrate that the globular secondary structure calculated from the amide I envelope for these 14 proteins from FTIR is in excellent agreement with the values calculated from the X-ray crystallographic data using three-dimensional Ramachandran analysis, providing that the proper contribution from GLN and ASN side chains to the 1667 and 1650 cm(-1) component bands has been taken into account. The standard deviation of the regression analysis for the per cent helix, extended, turn and irregular conformations was found to be 3.49%, 2.07%, 3.59% and 3.20%, respectively.

alpha-Crystallin quaternary structure: molecular basis for its chaperone activity.

alpha-Crystallin, the major protein in all vertebrate lenses, functions as a chaperone. In the present analysis, an 'open' micellar structure composed of alpha A subunits is used to simulate chaperoning of partially heat denatured soluble gamma-crystallin. The interaction is both electrostatic and hydrophobic and satisfies experimental evidence for a 1:1 alpha/gamma molar ratio, a doubling of molecular mass and a minimal increase in the dimensions of the complex [J. Biol. Chem. (1994) 269, 13601-13608; Invest. Opthalmol. Vis. Sci. (1995) 36, 311-21]. These data are also in accord with Farahbaksh et al. [Biochemistry (1995) 34, 509-16]; i.e. the bound gamma-crystallin monomers are not in a central cavity, but are separated by alpha A subunits.

Identification of the active-site residues of the L proteinase of foot-and-mouth disease virus.

The foot-and-mouth disease virus (FMDV) leader (L) protein is involved in autocatalytic cleavage at the L/P1 junction and in the cleavage of translation initiation factor p220, a subunit of the cap-binding protein complex. It has been suggested that this proteinase has homology to the papain-like family of cysteine proteinases, and from this information, we have investigated the active-site residues by introducing specific mutations into the L gene. Mutations of Cys-23 to Ala or His-120 to Leu resulted in enzymes that lacked cis activity at the L/VP4 cleavage site, trans activity on a truncated L-P1 substrate, and p220 cleavage activity. Mutations of Cys-23 to ser or His-110 to Leu resulted in enzymes that retained some or all cis activity and had reduced p220 cleavage. These mutations were introduced separately into a full-length FMDV cDNA, and RNA transcripts derived from these cDNAs were translated in a cell-free system and transfected into cells. The C23S mutant inefficiently cleaved at the L/P1 junction and within P1, and virus obtained from transfected cells reverted to wild type. The H110L mutant cleaved the L/P1 junction almost as well as the wild-type enzyme, and virus recovered from transfected cells retained the mutation and displayed wild-type viral protein synthesis and host shut-off kinetics.

An energy-minimized casein submicelle working model.

To develop a molecular basis for structure-function relationships of the complex milk protein system, an energy-minimized, three-dimensional model of a casein submicelle was constructed consisting of kappa-casein, four alpha s1-casein, and four beta-casein molecules. The models for the individual caseins were from previously reported energy-minimized, three-dimensional structures. Docking of one kappa-casein and four alpha s1-casein molecules produced a framework structure through the interaction of two hydrophobic antiparallel sheets of kappa-casein with two small hydrophobic antiparallel sheets (residue 163-174) of two preformed alpha s1-casein dimers. The resulting structure is approximately spherically symmetric, with a loose packing density; its external portion is composed of the hydrophilic domains of the four alpha s1-caseins, while the central portion contains two hydrophbic cavities on either side of the kappa-casein central structure. Symmetric and asymmetric preformed dimers of beta-casein formed from the interactions of C-terminal beta-spiral regions as a hinge point could easily be docked into each of the two central cavities of the alpha-kappa framework. This yielded two plausible energy-minimized, three-dimensional structures for submicellar casein, one with two symmetric beta-casein dimers and one with two asymmetric dimers. These refined submicellar structures are in good agreement with biochemical, chemical, and solution structural information available for submicellar casein.

Comparison of the three-dimensional molecular models of bovine submicellar caseins with small-angle X-ray scattering. Influence of protein hydration.

To test the applicability of two energy-minimized, three-dimensional structures of the bovine casein submicelle, theoretical small-angle X-ray scattering curves in the presence and absence of water were compared to experimental data. The published method simulates molecular dynamics of proteins in solution by employing adjustable Debye-Waller temperature factors (B factors) for the protein, for the solvent, and for protein-bound water. The programs were first tested upon bovine pancreatic trypsin inhibitor beginning with its known X-ray crystal structure. To approximate the degree of protein hydration previously determined by NMR relaxation experiments (0.01 g water/g protein), 120 water molecules were docked into the large void of the kappa-casein portion of the structure for both the symmetric and asymmetric casein submicelle models. To approximate hydrodynamic hydration (0.244 g water/g protein), 2703 water molecules were added to each of the above structures using the "droplet" algorithm in the Sybyl molecular modeling package. All structures were then energy-minimized and their solvation energies calculated. Theoretical small-angle X-ray scattering curves were calculated for all unhydrated and hydrated structures and compared with experimentally determined scattering profiles for submicellar casein. Best results were achieved with the 120-bound-water structure for both the symmetric and asymmetric submicelle models. Comparison of results for the protein submicelle models with those for the theoretical and literature values of bovine pancreatic trypsin inhibitor demonstrates the applicability of the methodology.

Regulation of the soluble form of nicotinamide adenine dinucleotide phosphate-specific isocitrate dehydrogenase from lactating bovine mammary gland: effects of metabolites on activity and structure.

The cytosolic form of NADP+: isocitrate dehydrogenase, a primary source of the NADPH required for de novo fatty acid synthesis in lactating bovine mammary gland, was studied to determine possible mechanisms of regulation by metabolites. The enzymatic reduction of NADP+ exhibits lag-burst (hysteretic) kinetics that are eliminated by the noncatalytic binding of the substrate, a complex (1:1) of a metal ion (Mn2+ or Mg2+) and isocitrate. Preincubation of the enzyme with metal-citrate complex also nearly abolished the lag or activation time. In steady-state experiments, analyses of velocity versus metal-citrate complex as a binding isotherm, following the assumptions of Wyman's theory of thermodynamic linkage, showed that binding of metal-citrate complex could both stimulate and inhibit the enzyme. This analysis suggested hyperactivation by binding to sites with an average dissociation constant of .25 mM, inhibition by binding to sites with an average dissociation constant of 3.83 mM, and modulation (reactivation) by binding to sites with an average dissociation constant of 1.54 mM. Conformational changes induced by the binding of ligands were assessed using circular dichroism. The results suggest that binding of metal-isocitrate induces a conformational transition involving tyrosyl residues that is related to the altered kinetic processes. Reexamination of Michaelis-Menten kinetics using non-linear regression analysis also demonstrated hyperactivation of enzyme activity by metal-isocitrate with a dissociation constant equal to 21 microM (which is nearly seven times greater than the Michaelis constant). Concentration ranges observed for these transitions are compatible with physiological conditions, suggesting that complexes of metal-citrate and metal-isocitrate serve to modulate the activity of NADP+: isocitrate dehydrogenase.

Three-dimensional molecular modeling of bovine caseins: a refined, energy-minimized kappa-casein structure.

A refined three-dimensional molecular model of kappa-casein has been produced using energy minimization techniques and a Kollman force field on a previously reported predicted three-dimensional structure. This initial model was constructed via molecular modeling techniques from sequence-based secondary structural prediction algorithms. Both the initial and refined structures agreed with global secondary structure analysis from vibration spectroscopy. The refined structure contained many of the features of the initial model, including two sets of antiparallel beta-sheet structures containing predominantly hydrophobic side chains, which could form interaction sites with alpha s1-casein. Two types of energy-minimized dimer and tetramer models are presented: 1) using Cys as potential intermolecular disulfide binding sites and 2) using the two sheets as possible hydrophobic self-association sites, without Cys interactions. All structures yielded good stabilization energies and are in agreement with chemical, biochemical, and physical chemical results obtained for kappa-casein.

Three-dimensional molecular modeling of bovine caseins: an energy-minimized beta-casein structure.

To obtain a molecular basis for the similarities and dissimilarities in the functional, chemical, and biochemical properties between beta-casein and the other caseins, a predicted three-dimensional model is presented. The predicted structure was assembled using molecular modeling techniques, as well as secondary structural prediction algorithms, in conjunction with global secondary structural information from Raman spectroscopy. To add validity to this model, the structure was refined using energy minimization techniques to diminish the likelihood of structural overlaps and energetically unfavorable van der Waals contacts arising from the large number of proline residues present in the beta-casein sequence. The refined model overall showed a loosely packed, asymmetrical structure with an axial ratio of 2:1. Hydrophobic side chains were uniformly dispersed over one end (C terminal) and the center surface of the structure; the other end (N terminal) was hydrophilic. The hydrophobic section also possessed a large loop through which water could easily travel. Such a suprasurfactant structure could account for the micellar type of hydrophobically driven self-association exhibited by beta-casein. Other chemical and biochemical data are in good agreement with the refined structure.

P relaxation responses associated with n(2)/o(2) diffusion in soybean nodule cortical cells and excised cortical tissue.

N(2)-fixing Bradyrhizobium japonicum nodules and cortical tissue derived from these nodules were examined in vivo by (31)P nuclear magnetic resonance (NMR) spectroscopy. Perfusion of the viable nodules and excised cortical tissue with O(2) followed by N(2) or Ar caused a loss of orthophosphate (Pi) resonance magnetization associated with the major portion of acidic Pi (delta 0.9 ppm, pH 5.5) residing in the cortical cells. Resumption of O(2) perfusion restored approximately 80% of the intensity of this peak. Detailed examination of the nuclear relaxation processes, spin-lattice relaxation time (T(1)), and spin-spin relaxation time (T(2)), under perfusion with N(2) or Ar as opposed to O(2), indicated that loss of signal was due to T(1) saturation of the acidic Pi signal under the rapid-pulsed NMR recycling conditions. In excised cortical tissue, Pi T(1), values derived from biexponential relaxation processes under perfusing O(2) were 59% 3.72 +/- 0.93 s and 41% 0.2 +/- 0.08 s, whereas under N(2) these values were 85% 7.07 +/- 1.36 s and 15% 0.39 +/- 0.07 s. The T(1) relaxation behavior of whole nodule vacuolar Pi showed the same trend, but the overall values were somewhat shorter. T(2) values for cortical tissue were also biexponential but were essentially the same under O(2) (38% 0.066 +/- 0.01 s and 63% 0.41 +/- 0.08 s) and N(2) (39% 0.07 +/- 0.01 s and 61% 0.37 +/- 0.01 s) perfusion. Soybean (Glycine max) root tissue as well as Pi solutions exhibited single exponential T(1) decay values that were not altered by changes in the perfusing gas. These data indicate that oxygen induces a change in the physical environment of phosphate in the cortical cell tissue. Although under certain conditions oxygen has been observed to act as a paramagnetic relaxation agent, model T(1) experiments demonstrate that O(2) does not significantly influence Pi relaxation in this manner. Alternatively, we suggest that an increase in solution viscosity brought on by the production of an occlusion glycoprotein (under O(2) perfusion) is responsible for the observed relaxation changes.

Quantification of alpha s1-casein in goat milk from French-Alpine and Anglo-Nubian breeds using reversed-phase high performance liquid chromatography.

Samples of isoelectrically precipitated goat casein from the milks of French-Alpine and Anglo-Nubian breeds were separated into four components in a single run by reversed-phase HPLC. The proportion of alpha s1-casein thus resolved was determined quantitatively. The method uses a reversed-phase C-4 column and a linear gradient from 30 to 50% acetonitrile in 30 min with trifluoroacetic acid constant at .1%. Sodium dodecyl sulfate-PAGE was carried out to establish the identity of the isolated components. By a comparison with previously published results for caprine and bovine milk caseins, the four peaks were identified as kappa-, alpha s2-, alpha s1-, and beta-casein. Quantitative variations in the chromatographically resolved alpha s1-casein fraction of goat milk were evident. Some individual goat milks contained high levels of alpha s1-casein (2.70 g/L), but others contained significantly low levels (.12 g/L). There was no statistical difference in the overall means between breeds in alpha s1-casein composition, but cluster analysis statistics showed three distinct categories of alpha s1-producers: high, medium, and low. Interestingly, 6 of 15 French-Alpine goats and only one Anglo-Nubian goat fell into the "low" producer category (.38 +/- .2 g/L). Thus, expression of the alpha s1-component may be genetically regulated but may not be a breed-specific trait.

Three-dimensional molecular modeling of bovine caseins: kappa-casein.

Three-dimensional structures derived from X-ray crystallography are extremely important in elucidating relationships between structure and function for many proteins. However, not all proteins can be crystallized. The caseins of bovine milk are one class of noncrystallizable proteins. The complete primary and partial secondary structures of these proteins are known, but homologous proteins with known crystallographic structure are not available. In this report, sequence-based predictions of secondary structure were made and adjusted to conform with global secondary structures derived from Fourier transform infrared spectroscopy. With this information, a three-dimensional structure for kappa-casein was constructed using molecular modeling computer programs. The constructed model contains two unstranded beta-sheets; both are predominantly hydrophobic and capable of forming quaternary structural interaction sites with alpha s1-casein. This unrefined structure is in good agreement with much of the biochemical information available for kappa-casein.

Three-dimensional molecular modeling of bovine caseins: alpha s1-casein.

Structures derived from X-ray crystallography are extremely important in elucidating functional relationships for many proteins. However, the caseins of bovine milk are one class of noncrystallizable proteins. The complete primary and partial secondary structures of these proteins are known, but homologous proteins of known crystallographic structure cannot be found. Therefore, sequence-based predictions of secondary structure were made and adjusted to conform with global secondary structures determined by Raman spectroscopy. With this information, a three-dimensional structure for alpha s1-casein was constructed using molecular modeling programs. The predicted structure of alpha s1-casein contains a hydrophobic and a hydrophilic domain, which are connected by a segment of alpha-helix. This unrefined structure shows good agreement with global biochemical and chemical information concerning alpha s1-caseins A, B, and C.

Calcium-induced associations of the caseins: thermodynamic linkage of calcium binding to colloidal stability of casein micelles.

The caseins occur in milk as colloidal complexes of protein aggregates, calcium, and inorganic phosphate. As determined by electron microscopy, these particles are spherical and have approximately a 650 A radius (casein micelles). In the absence of calcium, the protein aggregates themselves (submicelles) have been shown to result from mainly hydrophobic interactions. The fractional concentration of stable colloidal casein micelles can be obtained in a calcium caseinate solution by centrifugation at 1500 g. Thus, the amount of stable colloid present with varying Ca2+ concentrations can be determined and then analyzed by application of equations derived from Wyman's Thermodynamic Linkage Theory. Ca(2+)-induced colloid stability profiles were obtained experimentally for model micelles consisting of only alpha s1- (a calcium insoluble casein) and the stabilizing protein kappa-casein, eliminating the complications arising from beta- and minor casein forms. Two distinct genetic variants alpha s1-A and B were used. Analysis of alpha s1-A colloid stability profiles yielded a precipitation (salting-out) constant k1, as well as colloid stability (salting-in) parameter k2. No variations of k1 or k2 were found with increasing amounts of kappa-casein. From the variation of the amount of colloidal casein capable of being stabilized vs. amount of added kappa-casein an association constant of 4 L/g could be calculated for the complexation of alpha s1-A and kappa-casein. For the alpha s1-B and kappa-casein micelles, an additional Ca(2+)-dependent colloidal destabilization parameter, k3, was added to the existing k1 and k2 parameters in order to fully describe this more complex system. Furthermore, the value of k3 decreased with increasing concentration of kappa-casein. These results were analyzed with respect to the specific deletion which occurs in alpha s1-casein A in order to determine the sites responsible for these Ca(2+)-induced quaternary structural effects.

Tertiary and quaternary structural differences between two genetic variants of bovine casein by small-angle X-ray scattering.

The casein complexes of bovine milk consist of four major protein fractions, alpha s1, alpha s2, beta, and kappa. Colloidal particles of casein (termed micelles) contain inorganic calcium and phosphate; they are very roughly spherical with an average radius of 650 A. Removal of Ca2+ leads to the formation of smaller protein aggregates (submicelles) with an average radius of 94 A. Two genetic variants, A and B, of the predominant fraction, alpha s1-casein, result in milks with markedly different physical properties, such as solubility and heat stability. To investigate the molecular basis for these differences, small-angle X-ray scattering was performed on the respective colloidal micelles and submicelles. Scattering curves for submicelles of both variants showed multiple Gaussian character; data for the B variant were previously interpreted in terms of two concentric regions of different electron density, i.e., a "compact" core and a relatively "loose" shell. For the submicelle of A, there was a third Gaussian, reflecting a negative contribution due to interparticle interference. Molecular parameters for submicelles of both A and B are in agreement with hydrodynamic data in the literature. Data for the micelles, for which scattering yields cross-sectional information, were fitted by a sum of three Gaussians for both variants; for these, the corresponding two lower radii of gyration represent the two concentric regions of the submicelles, while the third reflects the average packing of submicelles within the micellar cross section. Most of the molecular parameters obtained showed small but consistent differences between A and B, but for submicelles within the micelle several differences were particularly notable: A has a greater molecular weight for the "compact" region of the constituent submicelle (82,000 vs 60,000) and a much greater submicellar packing number (6:1 vs 3:1). Reasons for these and other differences are to be sought in sequence differences and in differences in calcium-binding sites and charge distribution.