Bimodal sedimenting zones due to ligand-mediated interactions.

Ligand-mediated association-dissociation reactions can give rise to band sedimentation patterns showing bimodal bands despite instantaneous establishment of equilibrium. Weaker interactions result in unimodal bands whose sedimentation coefficients decrease with time of sedimentation and in characteristic patterns of total ligand. The implications of these results for fundamental investigations of protein interactions and for conventional analytical applications of zone sedimentation and molecular sieve chromatography are considered.

A fluorometric study of the interaction of bradykinin with lipids.

The interaction of bradykinin (BK) with lipids has been followed by steady-state fluorescence measurements. Addition of either cerebroside sulfate (CS) or phosphatidylinositol (PI), solubilized with the nonionic surfactant C12E8, to BK or its analogue [Gly6]-BK enhances the relative fluorescence intensity of peptide emission at 288 nm. Fluorometric titration of the peptide with lipid has been used to quantitate the interactions in terms of stoichiometry and equilibrium constant. Job's method of continuous variation for the BK-CS interaction gave a stoichiometry of 1:2 for the complex. The value of the equilibrium constant, K, for the interaction of either BK or [Gly6]-BK with CS is 1.5.10(4) M-1. The BK-PI interaction is weaker; K = 5.0.10(3) M-1. Although electrostatic forces no doubt play a major role in these interactions, measurements on the model peptide Gly-Phe-Gly indicate that the phenylalanine residues of BK are disposed in the hydrophobic environment provided by the lipid-C12E8 mixed micelle. 13C-NMR measurements on [99% 13C alpha-Gly6]-BK show that there is no change in its cis/trans ratio upon interaction with CS. The increase in the relative fluorescence intensity of BK accompanying its cooperative interaction with sodium dodecyl sulfate (SDS) implicates the role of hydrophobic forces in this interaction as well. These results bear on the interpretation of the changes in circular dichroism (CD) of BK caused by SDS.

Analysis of the gel electrophoresis of looped protein-DNA complexes by computer simulation.

The theory of mass transport coupled to reversible interactions under chemical kinetic control forms the basis of a numerical model that has been applied to systems such as lac repressor-lac operator DNA, in which a protein binds in two different modes to linear DNA carrying two specific binding sites. Three complexes may be formed: (1) a linear 1:1 complex with one protein molecule bound to one site on the DNA molecule; (2) a 1:1 complex in which a single protein molecule is bound to both sites simultaneously, thereby inducing a large DNA loop; and (3) a 2:1 linear complex in which two protein molecules are bound in tandem, each occupying a single site. The computational model affords a quantitative numerical simulation of the observed gel electrophoretic patterns produced by titration of the DNA with protein and provides new insights into the shape and nature of the patterns. In particular, the patterns may represent unimodal or bimodal reaction zones. Nevertheless, analysis of the peaks in the patterns obtained at low DNA and high protein concentration provides essential information as to the stoichiometry of the complexes and satisfactory estimates of association constants. The theory thus provides the experimenter with guidelines for quantitative evaluation of the results of gel retardation assays of the particular system under investigation, once protein-induced DNA (or RNA) loops have been established by independent physical or chemical methods. It is suggested that these insights might also find application to systems involving the binding of two or three different proteins to DNA with loop formation.

Boundary spreading in sedimentation velocity experiments on partially liganded aspartate transcarbamoylase. A ligand-mediated isomerization.

Transport theory for rapidly reversible interacting systems was used to analyze boundary spreading in sedimentation velocity experiments on partially liganded aspartate transcarbamoylase. In the presence of sub-stoichiometric amounts of a bisubstrate analog, N-(phosphonacetyl)-L-aspartate, which is bound with high affinity to the enzyme (Kd approximately 100 nM), broad boundaries were observed consistent with the presence of two conformational forms. The theoretical treatment showed that under these conditions, the interconversion between the compact (11.7 S) and swollen (11.3 S) forms of the enzyme appears uncoupled, due to the formation of a gradient of free ligand that is caused by the re-equilibration resulting from the differential sedimentation of the two enzyme forms. Sedimentation velocity patterns for such systems are interpretable in terms of two independent species. When, however, the enzyme is in the presence of a sub-saturating amount of the weakly bound ligand, succinate (Kd approximately 1 mM), the re-equilibration caused by the differential sedimentation does not perturb the large background of free ligand and form a gradient. Instead, the two different forms of the enzyme are in dynamic equilibrium, resulting in a boundary having average sedimentation and diffusion coefficients. The observed boundary spreading experiments with different ligands are satisfactorily interpreted in terms of a ligand-mediated isomerization of aspartate transcarbamoylase from a compact to a swollen conformation.

An NMR, CD, molecular dynamics, and fluorometric study of the conformation of the bradykinin antagonist B-9340 in water and in aqueous micellar solutions.

A detailed NMR, CD, fluorometry, and molecular modeling study of a novel bradykinin antagonist B-9340, containing a novel amino acid D-Igl (alpha-(2-indanyl)glycine) at position 7, was carried out. The sequence of B-9340 is D-Arg0-Arg1-Pro2-Hyp3-Gly4-Thi5-Ser6-D- Igl7-Oic8-Arg9, where Hyp is hydroxyproline, Thi is beta-(2-thienyl)alanine, and Oic is (3aS,7aS)-octahydroindole-2-carboxylic acid. The CD results exhibit a striking effect of SDS on the spectrum of the BK antagonist, indicating that interaction with the surfactant induces a folded peptide structure. The interaction of this antagonist with phosphatidylinositol was monitored by fluorometry, indicating that the interaction of the peptide with the lipid is cooperative, and gives a Hill coefficient of 2.3. The two-dimensional proton NMR measurements indicate that B-9340 has no stable secondary structure in water solution and contains about 10-15% cis peptide bonds arising from Pro2, Hyp3, and Oic8. In SDS micelles, NMR reveals the existence of two beta-turns based on a number of medium-range connectivities that were useful for molecular modeling. The actual molecular modeling and dynamic runs were performed on B-9340 in an environment consisting of a layer of octyl sulfate anions and water. Ther results indicate that the structure of B-9340 in a micellar environment is characterized by a nonideal betaII-turn comprising residues Pro2 to Thi5, a nonideal betaII'-turn comprising residues Ser6-Arg9, and broad folding in the middle part of the molecule. The structure is stabilized by several hydrogen bonds and by a salt bridge between the guanidine moiety of Arg1 and the carboxyl group of Arg9, whereas the middle part of the peptide is buried in the micelle. The structure is deposited as Brookhaven PDB file 1 BDK.

Probing the role of proline in peptide hormones. NMR studies of bradykinin and related peptides.

The use of NMR methods to study conformational and dynamic aspects of the proline residues in the nonapeptide bradykinin is reviewed. NMR analyses involve considerations of bistable equilibria which include the cis/trans conformational heterogeneity of the imide bond, the cis'/trans' regions of conformational stability which characterize rotation about the C alpha-CO bond (dihedral angle psi), and the interconversion of the pyrrolidine ring of proline between puckered C gamma-endo and C gamma-exo conformations. These conformational features are all characterized by different kinetic behavior, are interdependent with peptide bond conformation, and exhibit sensitivity to amino acid substitutions. Thus, the substitution of Gly6 for Ser6 increases the fractional cis probability of the sixth peptide bond from 0.1 to 0.35. Substitutions of alpha-aminoisobutyric acid (AIB) residues for proline introduce conformational constraints analogous to those in cis' proline. Correlations of pyrrolidine ring conformation and dynamics with the cis/trans ratio of the imide bond have also been observed in model systems. Conformational and activity analyses of [AIB7]-bradykinin provided a stimulus for the development of the first bradykinin antagonist by Stewart and Vavrek (Vavrek RJ and Stewart JM, Competitive antagonists of bradykinin. Peptides 6: 161-164, 1985).

Interaction of bradykinin with sodium dodecyl sulfate and certain acidic lipids.

The striking change in the circular dichroism (CD) of bradykinin (BK) occasioned by its interaction with sodium dodecyl sulfate (SDS) is evidently due in large part to a change in the conformation of the C-terminal tetrapeptide moiety of the hormone. The full change in CD is induced by the binding of two molecules of monomeric SDS per peptide molecule, the complex being aggregated. Formation of the 1:2 BK-SDS complex apparently proceeds via intermediates of stoichiometry 1:1 and 2:1. The cooperative nature of the interaction is attributed to the SDS-promoted aggregation of BK. Electrostatic interactions with the Arg residues appear important for the binding reaction per se. CD reveals that BK also interacts with acidic lipids which bear a net electrical charge (e.g., cerebroside sulfate and phosphatidyl inositol) but not with lipids bearing no net charge (e.g., cerebroside and phosphatidyl choline). The interactions are with particular mixed micelles of the lipid and the nonionic surfactant used for their solubilization, micellar size and structure being examined by surface tensiometry and electron microscopy.

Theory of sedimentation for ligand-mediated heterogeneous association-dissociation reactions.

Theoretical analytical sedimentation patterns have been computed for ligand-mediated heterogeneous association-dissociation reactions between macromolecules. Involvement of either a single kind of ligand or two different ligands acting in a stepwise fashion has been considered. Self-association, mediated in a stepwise fashion by two different ligands, has also been examined. The conclusion reached is that such interactions have the potentiality for exhibiting as many as three or four sedimenting peaks despite rapid rates of reaction. In general, the peaks correspond to different equilibrium compositions and not to individual macromolecular species; that is to say, they constitute a reaction boundary. Their resolution depends upon generation of concentration gradients of ligand(s) along the centrifuge cell by chemical reequilibration during sedimentation of the several macromolecular species. The implications of these findings for fundamental studies on subunit proteins and protein assemblies and for conventional applications of ultracentrifugation are discussed.

Determination of ligand-binding constants by isoelectric focusing. Generalization and extension of theory.

The theory of a proposed isoelectric focusing procedure for the determination of ligand binding constants has been generalized (via numerical calculations) to non-linear, electrophoretic velocity profiles and systems for which the change in velocity of the protein accompanying ligand binding depends upon pH. The theory has also been extended to include binding of ligand to heterogeneous sites. Determination of accurate values of the binding parameters depends upon extrapolation of apparent parameters to infinite dilution of protein.

Isoelectric focusing of interacting systems. IV. interaction of macromolecules with each other and with ligands.

The theoretical isoelectric focusing behavior for rapidly reversible, bimolecular complexing between two macromolecules depends upon the relative value of the isoelectric point of the complex. When it is intermediate in value, the transient patterns exhibit three peaks. As equilibrium is approached the central peak of complex disappears leaving two reactant peaks. When the isoelectric point is acidic or alkaline to both reactants, the equilibrium pattern also shows two peaks; but in this case only one is pure reactant, the other being a reaction zone. The two cases can be distinguished by varying the relative amounts of reactants. Transient patterns for ligand-binding exhibit a peak of unliganded protein and a reaction zone. As the charged ligand is driven out of the focusing column the reaction zone disappears, so that the equilibrium pattern shows only a peak of unliganded protein. In general, the isoelectric point of the complex cannot be determined from the transient patterns.

Isoelectric focusing of interacting systems. V. Determination of ligand-binding constants.

A theory is formulated for an isoelectric focusing procedure which permits determination of intrinsic ligand-binding constants. The protein is first focused in the absence of ligand, after which ligand is added to the appropriate electrode compartment and then driven by the electric field into the focusing column where it complexes with the protein. The band of protein and its complexes moves to the constituent isoelectric point. An equation linearly relates the reciprocal of the overall distance moved to the reciprocal of the local concentration of ligand. The quotient of the intercept and slope gives the intrinsic binding constant. If the concentration of ligand in the electrode compartment is used in lieu of the local concentration, an apparent constant is obtained. Extrapolation of the apparent constant to infinite dilution of protein gives the intrinsic constant. For certain systems, conditions may be realized which give an apparent constant within 4% of the intrinsic constant.

Theory of counterion electrophoresis. Guidelines for determination of ligand-binding parameters.

A theory is formulated to provide guidelines for the quantitative interpretation of steady-state counterion electrophoretic patterns (T.-H. Ueng and F. Bronne, Arch. Biochem. Biophys. 197 (1979) 205) in terms of intrinsic ligand-binding constant and number of binding sites on the protein molecule. Briefly, the prescribed procedure calls for extrapolation of the steady-state binding constant to infinite dilution of protein to obtain a quantity which is the product of a readily evaluated kinetic factor and the intrinsic binding constant. On the other hand, extrapolation of the steady-state number of binding sites to infinite dilution can probably be dispensed with if determined at a reasonably low protein concentration.

Isoelectric focusing of interacting systems. I. Carrier ampholyte-induced macromolecular isomerization.

A phenomenological theory of isoelectric focusing is formulated for rapidly reversible, ampholyte-induced macromolecular isomerization. The calculations reveal that such interactions can give well resolved, bimodal transient and equilibrium isoelectric focusing patterns in which the two peaks correspond to different chemical equilibrium compositions and not to separated isomers. The kinetics of approach to the equilibrium pattern are characteristically biphasic: During the first phase, which is controlled by the rate of migration of the isomers in the electric field, two peaks are positioned in the region between the isoelectric points of the two isomers; one of the peaks then grows slowly at the expense of the other with a diffusion-dominated rate. The kinetics are dependent upon the initial distribution of macromolecule in the isoelectric focusing column, and in certain cases only a single peak is apparent during the first phase. These findings have practical implications for unambiguous interpretation of isoelectric focusing patterns, furnish explanations for hitherto puzzling experimental observations, and provide theoretical insights required for application of isoelectric focusing to the detection and characterization of macromolecular interactions in general.

Isoelectric focusing of interacting systems. II. pH-dependent conformational transitions.

Transient and equilibrium isoelectric focusing patterns have been computed for pH-dependent conformational transitions in the limits of complete cooperativity and instantaneous chemical equilibration. Transitions induced by the binding of a relatively large number of hydrogen ions by the macromolecule give well resolved bimodal equilibrium patterns, provided that the resulting conformer has the lower isoelectric point. The corresponding transient patterns may be either bimodal or virtually unimodal for practical times of operation depending upon the point of insertion of the sample into the pH gradient and the stoichiometry of the interaction. A macromolecule undergoing sequential transitions can give multimodal isoelectric focussing patterns.

Isoelectric focusing of a dimerizing solute in rapid chemical equilibrium: comparison of simulation procedures.

The approach to isoelectric focusing equilibrium of a rapidly dimerizing solute was simulated by two different computing procedures: a stationary-grid model developed by Cann and Stimpson and a distorted-grid technique derived from the method of Cox. The results given by the two models were virtually identical at all times during the approach to equilibrium. Of the two procedures, the distorted-grid method has an advantage in computing time, while the stationary-grid model is applicable to a broader range of transport experiments. The effect on the focusing experiment of varying the electric field was examined by distorted-grid simulations. When the field was increased, the equilibrium distribution sharpened somewhat and the peak of the concentration profile shifted toward the isoelectric position of the dimer. The rate of approach to equilibrium was approximately proportional to the field strength.

High-field NMR and circular dichroism solvent-dependent conformational studies of the bradykinin C-terminal tetrapeptide Ser-Pro-Phe-Arg.

The conformational properties of the tetrapeptide Ser1-Pro2-Phe3-Arg4, the C-terminal fragment of the nonapeptide hormone bradykinin, have been studied by circular dichroism and two-dimensional NMR techniques. Measurements of coupling constants, NH temperature dependence rates and nuclear Overhauser effects (performed with rotating frame nuclear Overhauser spectroscopy, ROESY) in H2O and CD3OH/D2O (80/20, v/v) reveal different conformations in the corresponding solvent. In aqueous solution the molecule exists in a random conformation or as an average of several conformations in rapid exchange. In CD3OH/D2O, however, the conformation is well-defined. The backbone of the peptide is extended, and the side-chains of Phe3 and Arg4 exhibit unusual rigidity for a peptide of this size. Evidently, the secondary structure is stabilized by a charge interaction between the guanidino group of Arg4 and the terminal carboxyl group, since experiments at various pH's show clearly that the definition of conformation decreases strongly upon protonation of the carboxyl function. A NH3+(Ser1)-COO-(Arg4) salt bridge, as well as any form of turn stabilized by hydrogen bonds can be ruled out with certainty.

Conformational analysis of the type II and type III collagen alpha-1 chain C-telopeptides by 1H NMR and circular dichroism spectroscopy.

The type II and type III collagen alpha-1 chain C-telopeptides are a 27 mer with the sequence NAc-GPGIDMSAFAGLGPREKGPDPLQYMRA and a 22mer,NAc-GGGVASLGAGEKGPVGYGYEYR, respectively. Their conformations have been studied in CD3OH/H2O (80/20) solution by means of two-dimensional proton NMR and CD spectroscopy. Based on TOCSY and NOESY experiments, all resonances were assigned and the conformational properties were analyzed in terms of vicinal NH-H alpha coupling constants, sequential and medium range NOEs and amide proton temperature coefficients. The conformation of the type II C-telopeptide is essentially extended. Evidence from CD spectroscopy suggests that a very minor proportion of the peptide might be helical (ca.8%), but the NMR data show no evidence for a non-linear structure. The observation of reduced amide proton temperature dependence coefficients in certain sections of the molecule can, in view of the absence of any other supporting evidence, only be interpreted in terms of local shielding from solvent for sterical reasons (large hydrophobic side-chains). The conformation of the type III C-telopeptide is mostly extended except for a beta-turn ranging from Gly8 to Glu11, which is stabilized by a hydrogen-bond between NH of Glu11 and the carbonyl group of Gly8. The low temperature coefficient of NH(Glu11) and, in particular, the observation of a medium range NOE between H alpha (A9) and NH(E11) corroborate the existence of a beta-turn in this region. Although spectral overlap prevents a precise conclusion with regard to the type of beta-turn present, there is some evidence that it might be type II.

The aggregation properties of some bradykinin analogs.

Bradykinin (BK) is a peptide hormone with sequence Arg1-Pro2-Pro3-Gly4-Phe5-Ser6-Pro7-Phe8-Arg9 and has been implicated in a multitude of pathophysiological processes such as the ability to lower systemic blood pressure and stimulate pain. Bulky, beta-branched D-aliphatic residues at position 7 combined with bulky L-aliphatic residues at position 8 have now been observed to yield strong antagonists. Nuclear magnetic resonance studies have been carried out on many of these molecules with a view to determining their solution conformations. However, two such analogs, namely DArg-[Hyp3, Thi5, DSer6, DCpg7, Cpg8]-BK [I] and DArg-[Hyp3, DSer6, DCpg7, Cpg8]-BK [II] (Cpg = alpha-cyclopentyl-glycine; Hyp = 4-hydroxy-L-proline, Thi = beta-(2-thienyl)-L-alanine), have exhibited an abnormal, non-linear temperature dependence for the amide NH proton of Cpg8. The NH of Arg9 also shows a slightly non-linear temperature dependence at higher temperatures above 25 degrees C. In addition, a very slow exchange rate for the NH protons of DCpg7, Cpg8 and Arg9 indicated aggregation of these two analogs, which was confirmed using the circular dichroism experiments.

A proton magnetic resonance and a circular dichroism study of the solvent dependent conformation of the synthetic tubulin fragment Ac tubulin, alpha (430-441) amide and its interaction with substance-P.

Proton magnetic resonance techniques were used to study the conformation of the synthetic tubulin fragment Ac-tubulin (430-441) amide in H2O and 80% CD3OH/20% D2O solutions, using water suppression techniques. Proton assignments are based on two-dimensional COSY experiments combined with one-dimensional spin decoupling. A comparison of the NH proton shifts between the two solvents, namely delta(CD3OH/H2O-H2O) shows a small solvent effect for the Lys1 to Val6 region of the molecule, whereas for Gly7 to Glu12 the solvent effect is much larger. The smaller effects in the region of Lys1 to Val6 may be due to some hydrogen bonding as these protons are shielded from the solvent. These conclusions are in agreement with the circular dichroism results in 80% methanol/20% water where the alpha helix is present to the extent of 30%, whereas the peptide is completely unstructured in water with some aggregation. The temperature dependence of the NH proton shifts was also carried out. In water these shifts are of the order of 7-9 X 10(-3) ppm/K indicating that most of the protons are not involved in hydrogen bonding. In CD3OH/H2O, these values range from about 4-6 X 10(-3) ppm/K, which are compatible with the presence of hydrogen bonds. Finally, binding studies were carried out between the tubulin peptide and the undecapeptide neutrotransmitter substance P. The largest shifts are for the Tyr3 NH proton of the tubulin fragment, whereas for substance P it is for the Lys3, Gln5 and Leu10 NH protons, indicating a change in conformation of both peptides on interaction.

A combined spectroscopic study of the solution conformation of bradykinin.

CD, NMR and laser Raman spectroscopy have been used in a complementary fashion to study the conformation of bradykinin and related peptides. The results are interpreted to mean that the secondary structure of bradykinin in aqueous solution is a time average of two interconverting structures - one disordered and the other partially ordered. The peptide evidently spends a maximum of about 20% of its time in the partially ordered conformation which features a hydrogen-bonded structure at the C-terminus, most likely a gamma-turn with Pro7 as the second residue. Bradykinin and its analogs enjoy considerable conformational adaptability in terms of response to changes in temperature and solvent composition. Thus, dioxane favors the hydrogen-bonded structure, while aqueous NaCl0(4) and methanol favor the trans Pro7 isomer. The pronounced CD changes caused by increasing the temperature are attributed to a conformational transition, which is reflected in the CD of the Phe residues and possibly also the Arg-Pro-Pro moiety. The magnitude of the thermodynamic functions for the presumed transition are interpreted to mean that the structure of water plays a dominant role in determining the conformation of the peptide.