Calorimetric studies of protein--inhibitor interaction. I. Binding of 3'-cytidine monophosphate to ribonuclease A at pH 5.5

Synergistic Inhibition of Protein Fibrillation by Proline and Sorbitol: Biophysical Investigations

We report here interesting synergistic effects of proline and sorbitol, two well-known chemical chaperones, in the inhibition of fibrillation of two proteins, insulin and lysozyme. A combination of many biophysical techniques has been used to understand the structural morphology and modes of interaction of the chaperones with the proteins during fibrillation. Both the chaperones establish stronger polar interactions in the elongation and saturation stages of fibrillation compared to that in the native stage. However, when presented as a mixture, we also see contribution of hydrophobic interactions. Thus, a co-operative adjustment of polar and hydrophobic interactions between the chaperones and the protein surface seems to drive the synergistic effects in the fibrillation process. In insulin, this synergy is quantitatively similar in all the stages of the fibrillation process. These observations would have significant implications for understanding protein folding concepts, in general, and for designing combination therapies against protein fibrillation, in particular.

Conformation of 3'CMP bound to RNase A using TrNOESY

The conditions for accurately determining distance constraints from TrNOESY data on a small ligand (3'CMP) bound to a small protein (RNase A, <14 kDa) are described. For small proteins, normal TrNOESY conditions of 10:1 ligand:protein or greater can lead to inaccurate structures for the ligand-bound conformation due to the contribution of the free ligand to the TrNOESY signals. By using two ligand:protein ratios (2:1 and 5:1), which give the same distance constraints, a conformation of 3'CMP bound to RNase A was determined (glycosidic torsion angle, chi=-166 degrees ; pseudorotational phase angle, 0 degrees < or = P < or =36 degrees ). Ligand-protein NOESY cross peaks were also observed and used to dock 3'CMP into the binding pocket of the apo-protein (7rsa). After energy minimization, the conformation of the 3'CMP:RNase A complex was similar to the X-ray structure (1 rpf) except that a C3'-endo conformation for the ribose ring (rather than C2'-exo conformation) was found in the TrNOESY structure.

The salt-dependence of a protein-ligand interaction: ion-protein binding energetics

Using the binding of a nucleotide inhibitor (guanosine-3'-monophosphate) to a ribonuclease (ribonuclease Sa) as a model system, we show that the salt-dependence of the interaction arises due to specific ion binding at the site of nucleotide binding. The presence of specific ion-protein binding is concluded from a combination of differential scanning calorimetry and NMR data. Isothermal titration calorimetry data are then fit to determine the energetic profile (enthalpy, entropy, and heat capacity) for both the ion-protein and nucleotide-protein interactions. The results provide insight into the energetics of charge-charge interactions, and have implications for the interpretation of an observed salt-dependence. Further, the presence of specific ion-binding leads to a system behavior as a function of temperature that is drastically different from that predicted from Poisson-Boltzmann calculations.

Salt influence on glutathione--Schistosoma japonicum glutathione S-transferase binding

There has been some speculation about the salt independence of Schistosoma japonicum glutathione S-transferase (Sj26GST, EC. 2.5.1.18), but this aspect has not been carefully studied before. To establish the basis for a further development of this dependence, we have performed a methodical study of the influence of some important ions and their concentration on the binding properties of glutathione to Sj26GST by means of isothermal calorimetry and fluorescence quenching. Salts like NaCl, Na(2)SO(4) and MgSO(4) do not change practically the affinity of the protein for its substrate, whilst MgCl(2) has the effect of decreasing the affinity as its concentration rises. However, the enthalpy change is not affected by all the salts studied, and so, the entropy change is the causal factor in dropping the affinity. We also looked at the conformational stability of the protein under different conditions to check the structural changes they provide, and found that the unfolding parameters are practically not affected by the salt concentration. We discuss the results in terms of the chaotropic nature of the ions implied.

Influence of dynamic power compensation in an isothermal titration microcalorimeter

A theoretical analysis in Laplace's transformed domain based on a power balance represents a suitable model for an isothermal titration calorimeter with dynamic power compensation, designed and implemented in our laboratory. A rigorous calibration of the injection system and the calorimetric response was also made. Using electrically generated heat pulses, two different time constants have been determined from the calorimetric transfer function and assigned to the physical parts of the calorimeter. The same was done for a protein-ligand interaction. The binding of 2'-CMP to ribonuclease A at low and high ionic strengths was used to check the apparatus and the results were compared with those obtained by other authors (Wiseman, T.; Williston, S.; Brandts, J.F.; Lung-Nan, L. Anal. Biochem. 1989, 179, 131-137). In this case, the analysis showed a different time constant for the heat source. Independently of the nature of the heat source, the calorimetric time constants obtained while working under compensation are always smaller than those corresponding to a noncompensated system. The improvement of the calorimetric response introduced by dynamic power compensation is thus explained in terms of the reduction of the time constants characteristic of the calorimeter. This theoretical model can be used to predict the shape of the thermogram for any given reaction of either known or supposed thermodynamic parameters. Therefore, the calorimetric study is extended to the other nucleotides, 2'-UMP and 5'-dUMP, which have not hitherto been reported in the literature.

Pulsed ultrafiltration characterization of binding

We describe a new method for measuring binding constants, pulsed ultrafiltration. In this technique, a single injection or "pulse" of ligand is passed through a cell containing macromolecules confined by a conventional ultrafiltration membrane. Any binding of the ligand to the macromolecule alters the elution profile of the ligand. We describe this interaction by a set of coupled differential equations whose solution allows us to extract the binding density as a function of free ligand concentration eluting from the cell. A method of comparing elution profile areas which leads to values for both binding affinity and stoichiometry is also presented. We show that the pulsed ultrafiltration method can generate an extensive binding isotherm with a dense set of data points over a wide range of binding densities. We apply the method to several model ligand-macromolecule binding systems to demonstrate the measurement of equilibrium association constants and binding stoichiometry, the accuracy and precision of the method, and temperature dependence of binding. In general, our results agree with those from the literature, and they show that the approach is a fast and flexible method for characterizing ligand-macromolecule binding.

Measurement of biochemical affinities with a Gill titration calorimeter

A Gill titration calorimeter is evaluated as an instrument to determine in one experiment the equilibrium constant and the enthalpy change of a biochemical reaction. The dimensionless parameter kc (the product of the association equilibrium constant and the concentration of the reagent to be titrated; Wiseman et al., Anal. Biochem. 179, 131-137, 1989) is used to analyze the instrument performance. The analysis of simulated titration data corresponding to a simple model case shows that association equilibrium constants in the 10(2)-10(7) M-1 range may be determined when the kc parameter is between 1 and 1000. In addition we use a Monte Carlo approach to estimate the precision in the thermodynamic parameters of the reaction under study. The relative precision in the calculated constants ranges from 3 to 80% depending on the macromolecule concentration and kc value in the experiment. These results were checked with the study of the reactions of beta-trypsin with its inhibitor and ribonuclease A with cytidine 2'-monophosphate and cytidine 3'-monophosphate.

Thermodynamics of interaction of the fusion-inhibiting peptide Z-D-Phe-L-Phe-Gly with dioleoylphosphatidylcholine vesicles: direct calorimetric determination

The binding of the fusion-inhibiting peptide Z-D-Phe-L-Phe-Gly to unilamellar lipid vesicles of dioleoylphosphatidylcholine (DOPC) was investigated by isothermal titration calorimetry (ITC). The peptide Z-D-Phe-L-Phe-Gly is known to inhibit fusion of myxo- and paramyxoviruses with cells as well as cell-cell and vesicle-vesicle fusion in model systems. Calorimetric titrations conducted over a range of temperatures permitted characterization of the thermodynamics of the interaction of Z-D-Phe-L-Phe-Gly with model DOPC lipid membranes. Simultaneous global analysis of 15 ITC binding curves acquired at four different temperatures allowed determination of the equilibrium site association constant (K), stoichiometry of binding (n), binding enthalpy change (delta H), and heat capacity change of binding (delta Cp) in a single set of experiments. The binding affinity and enthalpy change per mole of DOPC bound at 25 degrees C was log K = 2.463 +/- 0.075 and delta H = -1.07 +/- 0.12 kcal/mol DOPC while the binding heat capacity change per mole of DOPC bound was delta Cp = -20.3 +/- 2.8 cal/(K.mol DOPC) with a temperature dependence (from 10-45 degrees C) of d(delta Cp)/dT = 0.37 +/- 0.18 cal/(K2.mol DOPC). A temperature-independent binding stoichiometry was determined to be n = 5.56 +/- 0.33 DOPC molecules per Z-D-Phe-L-Phe-Gly. A comparison of these results with previous peptide-lipid binding studies is discussed as is their relevance to a current model of the interaction of fusion-inhibiting peptides with phospholipid membranes.

Decoupling of melting domains in immobilized ribonuclease A

Ribonuclease A has been immobilized on silica beads through glutaraldeyde-mediated chemical coupling in order to improve the stability of the protein against thermal denaturation. The thermodynamic and binding properties of the immobilized enzyme have been studied and compared with those of the free enzyme. The parameters describing the binding of the inhibitor 3'-CMP (Ka and delta H) as monitored by spectrophotometry and calorimetry were not significantly affected after immobilization. Conversely both the stability and unfolding mechanism drastically changed. Thermodynamic analysis of the DSC data suggests that uncoupling of protein domains has occurred as a consequence of the immobilization. The two state approximation of the protein unfolding process is not longer valid for the immobilized RNase. Protein stability strongly depends on the hydrophobicity properties of the support surface as well as on the presence of the inhibitor and pH. For example, after immobilization on a highly hydrophobic surface, the enzyme is partially in the unfolded state. The binding of a ligand is able to reorganize the protein structure into a native-like conformation. The refolding rates are different for the two protein domains and vary as a function of pH and presence of the inhibitor 3'-CMP.

The use of microcalorimetry to measure thermodynamic parameters of the binding of ligands to insulin

Flow microcalorimetry was used to measure the free energies, enthalpies, and entropies of interactions between the hormone insulin and small ligand molecules or ions. Measurable amounts of heat were obtained for binding of four phenolic preservative molecules--phenol, meta-cresol, resorcinol, and methylparaben--to both two-zinc and zinc-free insulin and for binding of zinc ions to zinc-free insulin. All of the reactions were spontaneous, but the phenolic binding was driven by enthalpy, while that of zinc was entropy-driven. A combination of van der Waals interactions, hydrophobic effects, and protein conformational changes appeared to be involved in binding of the phenolic ligands. Zinc ions displayed two types of binding to insulin, both involving ion-dipole interactions.

A calorimetric approach to the study of the interactions of cytidine-3'-phosphate with bovine seminal ribonuclease

A calorimetric study at 25 degrees C is reported on the interaction between allosteric bovine seminal ribonuclease and cytidine-3'-phosphate. The results are compared with those obtained under identical experimental conditions for the interaction of pancreatic ribonuclease A and the same nucleotide. The analysis of the data provides evidence that the binding sites of seminal ribonuclease for cytidine-3'-phosphate are not equivalent, in agreement with previous equilibrium dialysis studies. A model with two sites with different affinities toward the nucleotide, the site with higher affinity resembling the binding site of pancreatic ribonuclease, is proposed. The values calculated for the thermodynamic parameters provide an insight of the forces involved in the interaction of the two enzymes with the nucleotide.

Rapid measurement of binding constants and heats of binding using a new titration calorimeter

A new titration calorimeter is described and results are presented for the binding of cytidine 2'-monophosphate (2'CMP) to the active site of ribonuclease A. The instrument characteristics include very high sensitivity, rapid calorimetric response, and fast thermal equilibration. Convenient software is available for instrument operation, data collection, data reduction, and deconvolution to obtain least-squares estimates of binding parameters n, delta H degree, delta S degree, and the binding constant K. Sample through-put for the instrument is high, and under favorable conditions binding constants as large as 10(8) M-1 can be measured. The bovine ribonuclease A (RNase)/2'CMP system was studied over a 50-fold range of RNase concentration and at two different temperatures. The binding constants were in the 10(5) to 10(6) M-1 range, depending on conditions, and heats of binding ca. -15,000 cal/mol. Repeat determinations suggested errors of only a few percent in n, delta H degree, and K values over the most favorable concentration range.

Error analysis in titration microcalorimetry of biochemical systems

A simplified method for titrations of biochemical systems is described as well as extensive error propagation through the data analysis. This work uses a Tronac Model 450 isoperibol titration calorimeter. Sample volumes of 2 ml are used and total heats of less than 5 mcal can be routinely measured. The binding of 3'-CMP to bovine pancreatic ribonuclease A is used to illustrate the methods. The binding enthalpy can be determined with a standard deviation of 1.5% and the free energy with a standard deviation of 2 to 3%.

The influence of methanol on the interactions of calcium with the pyridine nucleotides

The interactions of calcium with NAD+, NADH, NADP+ and NADPH in a 50% (by volume) methanol/water mixture (pH 7, 25 degrees C) were studied by calorimetry. The association constants for 1:1 complex formation were found to be 6.6 +/- 0.2, 270 +/- 76, 18 +/- 3 and 98 +/- 10 for NAD+, NADH, NADP+ and NADPH, respectively. Comparing these to the association constants for an aqueous system reveals that as the polarity of the solvent system is decreased the interactions involving NAD+, NADP+ and NADPH are all decreased. In contrast, the interaction involving NADH is markedly increased. All the interactions were found to be endothermic.

Calorimetric study of rheumatoid factor binding to human IgG

The binding of human IgM monoclonal rheumatoid factor (RF) to native human IgG was studied in a flow calorimeter. Thermal titration gave a enthalpy of binding of 5.5 X 10(4) J/mole site for the reaction, with 1 mole of RF binding to 2.5 moles of IgG; this corresponds to one RF subunit per IgG heavy chain. Similar experiments performed with heat-aggregated human IgG gave almost identical results. If the IgG molecule undergoes a conformational change upon heating, this is not such as to affect the enthalpic contribution of binding.

Calorimetric study of manganese binding to concanavalin A

The reaction of concanavalin A with Mn2+ has been studied calorimetrically. The binding enthalpy was measured at two different temperatures, 25 and 30 degrees C, in 10(-3) M acetate buffer; it was found to be constant between pH 4.0 and 5.0, delta H250 = 95 kJ/mol and delta H300 = 65 kJ/mol, respectively. The two S1 binding sites are identical and independent. Experiments at pH 5.6 are distorted by the heat of aggregation, which is several times higher than the heat of binding. Aggregation was demonstrated by spectrophotometric experiments and by light scattering. The presence of Mn2+ increases the stability of the protein molecule.

Calorimetric study of concanavalin A binding to saccharides

The binding of concanavalin A in the dimer form to various saccharides has been studied by calorimetry, and estimates of the binding enthalpy and binding constants have been calculated. Methyl alpha-D-mannoside and methyl alpha-D-glucoside have a -- delta H0 of 21.5 and 11.5 kJ/mol, respectively, at both pH 4 and 4.5. The p-nitrophenyl derivatives react with enthalpic values of 15.6 and 14.6 kJ/mol. The galactosepyranosides show no heat effects during mixing with the protein solutions. The apparent binding enthalpies calculated from the variations of the equilibrium constants with temperature are in good agreement with the values measured experimentally. The two binding sites of the dimer form of concanavalin A are equal and independent, and the low enthalpies obtained do not justify a large conformational change during the reaction. The binding reaction has also been estimated for other sugars normally contained in glycoproteins.

Calorimetric studies of ligand-induced modulation of calcium adenosine 5'-triphosphatase from sarcoplasmic reticulum

The enthalpy change (delta H degrees ') associated with the binding of Mg2+ to the sarcoplasmic reticulum calcium adenosine 5'-triphosphatase [(Ca2+)ATPase] is -76 kcal/mol. The affinity constant for Mg2+ obtained from calorimetric measurements agrees with the Km value for Mg2+ in the phosphorylation of the enzyme by inorganic phosphate (Pi). The delta H degrees ' of binding of Pi to the enzyme is -23.5 kcal/mol, and the affinity constant for Pi obtained from the calorimetry also agrees with the Km value for Pi in the phosphorylation reaction. delta H degrees ' of Mg2+ binding is reduced to -35 kcal/mol in the presence of either 20 mM Pi or 1.2 mM Ca2+ without a significant change in the affinity of the enzyme for Mg2+. delta H degrees ' of Pi binding to the enzyme drops to -8.5 kcal/mol in the presence of 10 mM Mg2+ without a significant change in the affinity of the enzyme for Pi. On the other hand, the presence of Ca2+ does not affect the delta H degrees ' for the binding of the substrate analogue 5'-adenylyl beta,gamma-imidodiphosphate [App(NH)p], and the presence of this analogue does not affect the delta H degrees ' for Ca2+ binding. The results suggest a model in which a conformational change, largely controlled by Mg2+ binding to the enzyme, leads to the formation of the covalent phosphoprotein intermediate.

Thermodynamics and stoicheiometry of the binding of substrate analogues to uricase

The subunit composition, metal content, substrate-analogue binding and thermal stability of Aspergillus flavus uricase were determined. A. flavus uricase is a tetramer and contains no copper, iron or any other common prosthetic group. Analytical-gel-filtration and equilibrium-dialysis experiments showed one binding site per subunit for urate analogues. The free energy of xanthine binding was -30.5 kJ (-7.3 kcal)/mol of subunit by equilibrium dialysis and -30.1 kJ (-7.2 kcal)/mol of subunit by microcalorimetry. The enthalpy change for xanthine binding was -15.9 kJ (-3.8 kcal)/mol of subunit when determined from the temperature-dependence of the equilibrium constant and -18.0 kJ (-4.3 kcal)/mol of subunit when measured microcalorimetrically. The thermal inactivation rate of A. flavus uricase increases as protein concentration is decreased. This concentration-dependent instability is not due to subunit dissociation.

The thermodynamics of nucleotide binding to proteins

Models describing the interaction of a small molecule with a protein are typically couched in terms of the stoichiometry, cooperativity, and binding free-energy change. These parameters are readily available from equilibrium dialysis experiments (or appropriate variations). With the recent advent of extremely sensitive calorimeters, it is possible to obtain thermal data for the binding reaction and, thus, the entire set of thermodynamic parameters, delta G', delta H', delta S', delta C', become readily available. This review is limited to the binding of nucleotides and nucleotide analogs to proteins for which complete thermal data are available. While the majority of such systems have been characterized by calorimetry, we have not excluded, per se, van't Hoff enthalpy determinations. The systems we have considered include, but are not limited to, thymidylate synthetase, phosphorylase, several dehydrogenases, aldolase, glutamine synthetase, hemoglobin, asparate transcarbamylase, and ribonuclease. A variety of forces contribute to the total free-energy change upon ligand binding. These forces include ionic, van der Waals, hydrogen bond, and hydrophobic. In several cases, properly designed experiments have allowed a partial resolution of the individual contributions of these various forces. Variation of easily accessible conditions such as temperature, pH, ionic strength, or solvent third component produce changes in the set of thermodynamic parameters which lead to the resolution of the forces. The generality of heat effects makes this method very useful for studying the involvement of protons in binding reactions. The variation in the magnitude and direction (release or uptake) of the proton flux is readily studied by determining the apparent heat of reaction at constant pH, ionic strength, and temperature in two or more buffers of differing heat of ionization. This application has been exploited in several cases and is examined in great detail. An overview of the results in these systems to date suggests that several trends observed in the thermodynamic parameters need to be confirmed by further experimentation and, if they hold, an appropriate theoretical basis must be developed to aid in their interpretation.

Structure and thermodynamic properties of the complexes between phospholipase A2 and lipid micelles

The interaction between porcine pancreatic phospholipase A2 and a homogeneous population of micelles of the subtrate analogue n-hexadecylphosphorylcholine containing 155 lipid monomers was studied by light scattering, equilibrium gel filtration, and isothermal calorimetry. From the detergent/protein molar ratio and the equivalent "molecular weight" of the resulting complex it is concluded that insertion of the enzyme into the detergent micelle results in a protein--detergent complex containing two phospholipase A2 molecules and 80 lipid monomers at 25 degrees C. The affinity constants and complex composition have been determined at different temperatures, allowing calculation of the thermodynamic parameters of the binding process. It is concluded that the interaction of phospholipase A2 with micellar lipids is predominantly hydrophobic.

A calorimetric comparison of the interaction of sodium dodecyl sulfate with cytochrome c and erythrocyte glycoproteins

The interactions of sodium dodecyl sulfate with cytochrome c and erythrocyte glycoproteins have been studied by the method of titration calorimetry. It was found that the initial addition of sodium dodecyl sulfate to cytochrome c caused an endothermic unfolding of the protein, detectable by circular dichroism (CD). This was followed by the exothermic binding of sodium dodecyl sulfate to the protein, without further CD-detectable conformational changes. In contrast, sodium dodecyl sulfate bound directly to the erythrocyte glycoproteins in an exothermic reaction without any accompanying CD-detectable conformation changes. This indicates that the glycoproteins solubilized in aqueous media have exposed hydrophobic regions which can interact directly with this detergent. The enthalpy changes and stoichiometries of binding are reported.

Calorimetric analysis of aspartate transcarbamylase from Escherichia coli: binding of cytosine 5'-triphosphate and adenosine 5'-triphosphate

The binding of CTP and ATP to aspartate transcarbamylase at pH 7.8 and 8.5 at 25 degrees has been investigated by equilibrium dialysis and flow microcalorimetry. The binding isotherms for CTP at both pH 7.8 and 8.5 and ATP AT PH 8.5 can be fit by a model which assumes three tight, three moderately tight, and six weak binding sites. The binding isotherms for ATP at pH 7.8 are best fit by a model which assumes six tight and six weaker sites. Both finite differenceH binding and finite differenceS binding are negative for both nucleotides at both pH values, so that the binding is enthalpy driven. For both nucleotides, finite differenceH is the same for the first two classes of binding sites, implying that the difference in the dissociation constants of these two classes of sites is the result of entropic effects. Direct pH measurements and calorimetric measurements in two buffers with very different heats of ionization (Tris and Hepes) indicate that the binding of both nucleotides is accompanied by the binding of protons. In the pH range 6.7-8.4, the number of moles of protons bound per mole of nucleotide increases as the pH decreases.