Search

We have recorded X-ray diffraction patterns at 3.1 A resolution from magnetically aligned fibres of the Pf3 strain of filamentous bacteriophage (Inovirus). The patterns are similar to patterns from the higher-temperature form of the Pf1 strain, indicating that the Pf3 and Pf1 virions have the same helix symmetry and similar protein subunit shape. This is of particular interest, given that the primary structures of the two protein subunits are quite different; and the nucleotide/protein subunit ratio in the Pf3 virion is more than twice that in Pf1, indicating important differences in DNA packaging. We have built a molecular model of the Pf3 protein capsid based on the model of Pf1, and refined it against the diffraction data using simulated annealing. The refinement confirms that the two structures are similar, which may reflect a fundamental motif of α-helix packing. However, there are some differences between the structures: the Pf3 subunit appears to be completely α-helical, beginning at the N terminus, whereas the first few residues of the Pf1 subunit are not helical; and the structure of the C-terminal region of the Pf3 subunit at the inner surface of the tubular capsid indicates that DNA/protein interactions in this virion may involve both aromatic side-chains and positively charged side-chains, whereas those in the Pf1 virion involve predominantly only the latter. In the course of this work, we have developed new approaches to refinement and validation of helical structures with respect to continuous transform fibre diffraction data.

Differential scanning calorimetry is employed at a scan rate of 2 K h−1 to study the effects of sodium chloride and calcium chloride on the main phase transition of dilute aqueous suspensions of dimyristoylphosphatidylcholine. The primary effect of sodium chloride over the concentration range studied, 0–5 molar, is to raise the transition temperature. In the case of calcium chloride, studied over the range 0–2 molar, the increase in the transition temperature is much larger, and the transitions become unsymmetrical and much broadened at concentrations above 0.75 M.

Cytochrome b562 is a four-helix-bundle protein containing a non-covalently bound b-type heme prosthetic group. In the absence of heme, cytochrome b562 remains highly structured under native conditions. Here we report thermodynamic data for the thermal denaturation of the holo- and apoproteins as determined by differential scanning calorimetry. Thermal denaturation of holocytochrome b562 is a highly reversible process, and unexpectedly does not involve dissociation of the heme prosthetic group. Thermal denaturation of the corresponding apoprotein, with the heme group chemically removed, remains a cooperative, reversible process. Apocytochrome b562 is substantially destabilized relative to the holoprotein: the t1/2 is more than ten degrees lower, and enthalpy and heat capacity changes are about one-half of the holoprotein values. However, the energetic parameters of apocytochrome b562 denaturation are within the range of observed values for small proteins.

The adaptability of Escherichia coli thioredoxin to the substitution of a series of non-natural amino acids has been investigated. Different thiosulfonated alkyl groups were inserted into the hydrophobic core of the protein in position 78 via disulfide bonding with a buried cysteine residue as previously described (Wynn R, Richards FM. 1993. Unnatural amino acid packing mutants of Escherichia coli thioredoxin produced by combined mutagenesis/chemical modification techniques. Protein Sci 2:395-403). The side chains added to the cysteine included methyl, ethyl, n-propyl, n-butyl, n-pentyl, and cyclo-pentyl derivatives. The side chains appear to exploit the presence of the large cavities to incorporate these variant forms, enabling the protein to fold and have some activity. Solution structural and kinetic data suggested that these substitutions had little effect on the overall fold of the protein. Thermodynamic data revealed that the entropic effect of restricting the side chains in the folded protein has an effect on the stability. The variant forms of thioredoxin have different propensities to form dimers despite the limited structural perturbations. Molecular modeling studies allow the conformation of the side chains to be assessed.

The study of a wide variety of reversible reactions in solution indicates that the enthalpy, DeltaH(vH), which controls the temperature variation of the equilibrium constant for a reaction, can seldom, if ever, be taken to be independent of the temperature. It is also found in most cases that the values for DeltaH(vH), properly evaluated as varying with the temperature, differ significantly from the values for the enthalpy, DeltaH(cal), determined by direct calorimetry under the same experimental conditions. In a continuing search for reactions which show agreement between DeltaH(vH) and DeltaH(cal), we have studied by isothermal titration calorimetry the reactions of heptylamine with heptanoic acid in dodecane solution and of alpha-cyclodextrin with sodium heptanoate in aqueous solution.

Here we describe how the systematic redesign of a protein's hydrophobic core alters its structure and stability. We have repacked the hydrophobic core of the four-helix-bundle protein, Rop, with altered packing patterns and various side chain shapes and sizes. Several designs reproduce the structure and native-like properties of the wild-type, while increasing the thermal stability. Other designs, either with similar sizes but different shapes, or with decreased sizes of the packing residues, destabilize the protein. Finally, overpacking the core with the larger side chains causes a loss of native-like structure. These results allow us to further define the roles of tight residue packing and the burial of hydrophobic surface area in the construction of native-like proteins.

The N-terminal 200 amino acids of SHC constitute a unique phosphotyrosine (Tyr(P)) interaction (PI) domain that shows no significant sequence similarity to the other Tyr(P)-recognizing module, the SH2 domain. We describe the thermodynamic parameters characterizing PI domain binding to various tyrosyl phosphopeptides, using isothermal titration calorimetry. The PI domain forms 1:1 complexes of similar affinity with a 12-mer peptide (ISLDNPDpYQQDF) derived from Tyr-1148 of the epidermal growth factor receptor (EGFR) (KD = 28 nm) and an 18-mer (LQGHIIENPQpYFSDACVH) derived from Tyr-490 of Trk (KD = 42 nM). Binding of the EGFR-derived peptide was largely enthalpy-driven at 25 degrees C, while Trk490 peptide binding was entropy-driven. Based on the change in heat capacity upon binding, approximately 700 A2 of nonpolar surface was estimated to be buried upon interaction. Alteration of the Asn or Pro to Ala in the NPXpY motif of the EGFR Tyr-1148 peptide increased the KD of PI domain interactions to 238 and 370 nM, respectively. Alteration of a Leu at position -5 (with respect to Tyr(P)) in the EGFR peptide to Gly also reduced the binding affinity (KD = 580 nM). It is proposed that the PI domain recognizes the beta1 turn that is found in NPXpY-containing peptides and also interacts with a larger segment of the peptide than seen for SH2 domains.

The apparent change in heat capacity, delta C(p), accompanying the thermally induced unfolding of lysozyme and of ribonuclease A was determined by means of differential scanning calorimetry in dilute aqueous buffer containing one of the following added solutes: 0.5 M or 1.0 M sucrose, 1.0 M glycine, 0.5 M, 1.0 M, or 2.0 M guanidinium chloride, 10% glycerol, or 0.5 M NaCl over a pH range. In each system the temperature of half-completion, t1/2, of the unfolding transition was varied by varying the pH. The resulting enthalpies of denaturation were linearly dependent on t1/2 for each solvent system. The resulting values of delta C(p) for each protein showed variation of almost 2-fold. Such large variations in the sensitivity of the proteins to temperature changes are not readily interpreted.

The thermal unfolding of eglin c, a small proteinase inhibitor of molecular weight 8.1 kDa, is studied by means of high sensitivity scanning calorimetry over a wide pH range in dilute buffer solutions, and in the presence of varying concentrations of guanidinium chloride at pH 7.00 and 10.55. The temperature of half-completion of the unfolding transition, t1/2, in dilute buffer varies from 41 degrees C at pH 1.1 to 86 degrees C at pH 7.0 to 10.55, with corresponding enthalpy changes of approximately 40 kcal mol-1 and 71 kcal mol-1. This latter enthalpy change, amounting to 8.7 cal g-1, is unusually large for a protein, especially for one of unusually small molecular weight. Addition of 3.3 M guanidinium chloride at pH 10.55 lowered t1/2 from 86 degrees C to 40 degrees C and decreased the enthalpy change from approximately 71 kcal mol-1 to 25 kcal mol-1.

Amino acid replacements of Met73, which is hyper-exposed in the native structure, with Asp, Glu, Lys, Gly, Ala, Val, Leu or Ile changed the stability of Streptomyces subtilism inhibitor by +1.60 to -0.94 kcal/mol-1 in free energy at 82.21 degrees C at pH 7.0, with higher hydrophobicity of the replacing amino acid side-chain showing a correlation with lower stability. Thermodynamic parameters obtained from detailed calorimetric analyses gave a nearly proportional relation between the entropy change and the enthalpy change of denaturation, i.e. T delta delta S degrees = -0.03 (+/- 0.11) + 1.14 (+/- 0.03) delta delta H, with the linear correlation coefficient 0.996 for 18 data points. This proportionality observed uniquely for the substitutions at position 73 was caused primarily by the water/side-chain interaction, or hydration effect, which can account for the majority of the changes in enthalpy and entropy induced by the mutations.

The thermodynamic effects of replacing the Met residue at amino acid position 103 ofStreptomycessubtilisin inhibitor with other non-polar aliphatic residues were studied by means of differential scanning calorimetry. All but the Leu mutant, which is as stable as the wild-type but has different cooperative units in the course of unfolding, showed destabilization in terms of free energy. Similar losses in free energy, however, were caused by different reasons, i.e. by increased entropy for the Ala mutant and by decreased enthalpy for the Ile mutant, with a tendency that increases in entropy are accompanied by increases in enthalpy. The gain in entropy that caused the largest loss in free energy for the Gly mutant was unexpectedly smaller than that for the Ala mutant. The changes in enthalpy and entropy induced by the mutations exhibited some correlations with hydrophobicity, while no clear correlation was found between the changes in free energy and hydrophobicity.

Two site-directed mutants of Escherichia coli thioredoxin (L78K and L78R) were designed to study the effect of placing a charged residue in the hydrophobic core of the protein. Both mutants retain catalytic activity in the assembly of phage M13. Thermal denaturation of both these mutant proteins at pH 7.0 shows a reduction of stability of approximately 4 kcal.mol-1 with respect to the oxidized wild-type form. The thermal denaturation of the protein fits a dimeric state model. A significant reduction in the change in heat capacity (delta Cp) on unfolding is observed compared to oxidized wild-type thioredoxin. We present data to indicate that this reduction in delta Cp is attributable to structural perturbations resulting in localized unfolding of the native protein and exposure to solvent of residues that are buried in the wild-type protein.

Two fragments of pancreatic ribonuclease A, a truncated version of S-peptide (residues 1-15) and S-protein (residues 21-124), combine to give a catalytically active complex. We have substituted the wild-type residue at position 13, methionine (Met), with norleucine (Nle), where the only covalent change is the replacement of the sulfur atom with a methylene group. The thermodynamic parameters associated with the binding of this variant to S-protein, determined by titration calorimetry in the temperature range 10-40 degrees C, are reported and compared to values previously reported [Varadarajan, R., Connelly, P. R., Sturtevant, J. M., & Richards, F. M. (1992) Biochemistry 31, 1421-1426] for other position 13 analogs. The differences in the free energy and enthalpy of binding between the Met and Nle peptides are 0.6 and 7.9 kcal/mol at 25 degrees C, respectively. These differences are slightly larger than, but comparable to, the differences in the values for the Met/Ile and Met/Leu pairs. The structure of the mutant complex was determined to 1.85 Angstrom resolution and refined to an R-factor of 17.4% The structures of mutant and wild-type complexes are practically identical although the Nle side chain has a significantly higher average B-factor than the corresponding Met side chain. In contrast, the B-factors of the atoms of the cage of residues surrounding position 13 are all somewhat lower in the Nle variant than in the Met wild-type. Thus, the large differences in the binding enthalpy appear to reside entirely in the difference in chemical properties or dynamic behavior of the -S- and -CH2- groups and not in differences in the geometry of the side chains or the internal cavity surface. In addition, a novel method of obtaining protein stability data by means of isothermal titration calorimetry is introduced.

We have measured the heats of formation of the trp repressor/operator complex by direct titration calorimetry over the temperature range 10 degrees C to 40 degrees C. A primary strong mode of binding displays the characteristic large negative heat capacity change observed by other methods in the formation of specific protein/DNA complexes. Unlike most such reactions, however, the formation of the trp repressor/operator complex is enthalpically driven throughout the physiological temperature range. After saturation of this principal mode, we also detected a secondary weaker binding mode, which we ascribe to a now well documented interaction called "half-site" binding. Although weak, this mode also exhibits an unusually large negative heat capacity change. Since the interface of the proposed secondary half-site binding mode has the same complementary stereochemistry as the primary one (due to internal symmetry), we correlate the negative heat capacity change with the formation of a stereospecific interface and not with high affinity. As in similar cases, the empirical correlation between buried non-polar surfaces and reduction of heat capacity does not account for the large negative delta Cp, nor do crystal structures reveal any further reduction in solvent excluded surfaces within the reactants upon complex formation. We attribute the "unaccounted for" decrement in the heat capacity of the complex to the stereospecific restriction of the hydrated polar elements that form the specific interface. We suggest that the "tightening of soft internal modes" at and near the polar interface of the complex is more important than previously recognized because previous considerations did not take into account the highly hydrated nature of these polar elements and the concomitant reduction in the degrees of freedom of the water structure.

Intramolecular disulfide bonds in protein molecules, whether present in the wild-type protein or engineered via site-directed mutagenesis, are capable of significantly increasing the stability. Establishing thermodynamic parameters associated with the redox formation of this linkage is often encumbered by other global structural changes within the protein molecule. The active site of Escherichia coli thioredoxin possesses a disulfide/dithiol in a short loop, oxidation/reduction of which is accompanied by little structural alteration of the protein. Data for the thermal denaturation of the reduced protein are presented, which on comparison to the data obtained for the oxidized form [Ladbury, J.E., Wynn, R., Hellinga, H.W., & Sturtevant, J.M. (1993) Biochemistry 32, 7526-7530] are used to establish thermodynamic parameters for the redox reaction in this molecule. Data for an isosteric double mutation in the active site of thioredoxin (Cys32Ser/Cys35Ser) are also presented. Although the wild-type and mutated proteins show a similar reduction in free energy compared to the oxidized form (-3.0 +/- 0.4 and -3.1 +/- 0.3 kcal mol-1, respectively), the enthalpic and entropic contributions to this destabilization are different for the two proteins.

The denaturation of Escherichia coli acyl carrier protein (ACP) in buffers containing both monovalent and divalent cations was followed by variable-temperature NMR and differential scanning calorimetry. Both high concentrations of monovalent salts (Na+) and moderate concentrations of divalent salts (Ca2+) raise the denaturation temperature, but calorimetry indicates that a significant increase in the enthalpy of denaturation is obtained only with the addition of a divalent salt. NMR experiments in both low ionic strength monovalent buffers and low ionic strength monovalent buffers containing calcium ions show exchange between native and denatured forms to be slow on the NMR time scale. However, in high ionic strength monovalent buffers, where the temperature of denaturation is elevated as it is in the presence of Ca2+, the transition is fast on the NMR time scale. These results suggest that monovalent and divalent cations may act to stabilize ACP in different ways. Monovalent ions may nonspecifically balance the intrinsic negative charge of this protein in a way that is similar for native, denatured, and intermediate forms. Divalent cations provide stability by binding to specific sites present only in the native state.

The development of cloning procedures has made possible a rapid increase in the appearance of reports on the thermodynamic effects of mutations in proteins. It will be helpful if the availability of excellent calorimetric instrumentation leads to an increase in the fraction of these studies directed toward the determination of enthalpies and entropies as well as free energies.

The effects of pH in the range 6.0-8.0 on the thermodynamics of the reversible thermal unfolding of Escherichia coli thioredoxin in the oxidized state have been determined over a range of concentrations using differential scanning calorimetry. The thermal denaturation indicated an inverse temperature dependence on concentration. The data were shown to fit a model based on dimerization of both the native and denatured states of the protein. The degree of dimerization of both states was found to be pH dependent. The previously described importance of protonation of the anomalously titrating aspartic acid 26 residue [Langsetmo, K., Fuchs, J., & Woodward, C. (1991) Biochemistry 30 ,7603-7609] was apparently verified by the agreement between the experimentally determined delta delta Gzerod and the calculated delta delta GzeroH in the pH range 7.0-8.0.

The effects of eight mutations on the thermodynamics of the reversible thermal unfolding of staphylococcal nuclease have been determined over a range of pH and protein concentration by means of differential scanning calorimetry. Variation of the protein concentration was included in our study because we found a significant dependence of the thermodynamics of protein unfolding on concentration. Values for the change in the standard free energy of unfolding, delta delta G0d, produced by the mutations in the pH range 5.0-7.0 varied from 1.9 kcal mol-1 (apparent stabilization) for H124L to -2.8 kcal mol-1 (apparent destabilization) for L25A. As has been observed in numerous other cases, there is no correlation in magnitude or sign between delta delta G0d and the corresponding values for delta delta Hd and T delta delta S0d, the latter quantities being in most cases much larger in magnitude than delta delta G0d. This fact emphasizes the difficulty in attempting to correlate the thermodynamic changes with structural changes observed by X-ray crystallography.

The thermotropic phase behavior of three difluorodimyristoylphosphatidylcholines, containing a difluoromethylene group in the 2-acyl chain at either the 4-, 8-, or 12-position, has been studied by differential scanning calorimetry (DSC) and by 31 P nuclear magnetic resonance (NMR) spectroscopy at temperatures at and below their normal gel to liquid crystalline transition. The fluorinated lipids have been found to behave in a manner similar to their protonated counterpart with the exception of the 4,4-difluoro isomer

Results obtained by means of differential scanning calorimetry (DSC) and isothermal titration calorimetry (ITC) with several proteins and mutant forms thereof are reported. DSC has been employed primarily to determine the thermodynamic changes accompanying the thermally induced unfolding of proteins, and ITC to measure the thermodynamics of the interactions of proteins with various ligands. It has become inreasingly evident that we are dealing with very complicated systems in which a single amino acid replacement probably causes numerous small effects which are widely distributed in the molecule, and which add up to an observed effect that is obviously difficult to rationalize in terms of molecular structure. It is nevertheless important to continue to broaden the base of quantitative information in the field.

The binding of mononucleotides to ribonucleases has received much attention in recent years, in part because of the possibility that the different specificities of the ribonucleases might be of interest in connection with specific protein-nucleic acid interactions. Among the most widely studied ribonucleases are ribonuclease A (RNase A), which is specific for pyrimidine nucleotides, RNase U 2 , specific for purine nuclcotides, and RNase T 1 , specific for guanosine nucleotides (Egami, E.; Oshima, T.; Uchida, T. Mol. Biol. Biochem. Biophys. 1980, 32, 250-277)

Aspartate transcarbamylase (EC 2.1.3.2) from E. coli is a multimeric enzyme consisting of two catalytic subunits and three regulatory subunits whose activity is regulated by subunit interactions. Differential scanning calorimetric (DSC) scans of the wild-type enzyme consist of two peaks, each comprised of at least two components, corresponding to denaturation of the catalytic and regulatory subunits within the intact holoenzyme (Vickers et al., J. Biol. Chem. 253 (1978) 8493; Edge et al., Biochemistry 27 (1988) 8081). We have examined the effects of nine single-site mutations in the catalytic chains. Three of the mutations (Asp-100-Gly, Glu-86-Gln, and Arg-269-Gly) are at sites at the C1: C2 interface between c chains within the catalytic subunit. These mutations disrupt salt linkages present in both the T and R states of the molecule (Honzatko et al., J. Mol. Biol. 160 (1982) 219; Krause et al., J. Mol. Biol. 193 (1987) 527). The remainder (Lys-164-Ile, Tyr-165-Phe, Glu-239-Gln, Glu-239-Ala, Tyr-240-Phe and Asp-271-Ser) are at the C1: C4 interface between catalytic subunits and are involved in interactions which stabilize either the T or R state. DSC scans of all of the mutants except Asp-100-Gly and Arg-269-Gly consisted of two peaks. At intermediate concentrations, Asp-100-Gly and Arg-269-Gly had only a single peak near the Tm of the regulatory subunit transition in the holoenzyme, although their denaturational profiles were more complex at high and low protein concentrations. The catalytic subunits of Glu-86-Gln, Lys-164-Ile and Asp-271-Ser appear to be significantly destabilized relative to wild-type protein while Tyr-165-Phe and Tyr-240-Phe appear to be stabilized. Values of delta delta G degree cr, the difference between the subunit interaction energy of wild-type and mutant proteins, evaluated as suggested by Brandts et al. (Biochemistry 28 (1989) 8588) range from -3.7 kcal mol-1 for Glu-86-Gln to 2.4 kcal mol-1 for Tyr-165-Phe.

Two fragments of pancreatic ribonuclease A, a truncated version of S-peptide (residues 1-15) and S-protein (residues 21-124), combine to give a catalytically active complex designated ribonuclease S. Residue 13 in the peptide is methionine. According to the X-ray structure of the complex of S-protein and S-peptide (1-20), this residue is almost fully buried. We have substituted Met-13 with seven other hydrophobic residues ranging in size from glycine to phenylalanine and have determined the thermodynamic parameters associated with the binding of these analogues to S-protein by titration calorimetry at 25 degrees C. These data should provide useful quantitative information for evaluating the contribution of hydrophobic interactions in the stabilization of protein structures.

In this paper we show that the usual assumption in studies of the temperature variation of equilibrium constants for equilibria of the form A+B AB that a plot of ln K vs. 1/T (K = equilibrium constant, T = temperature in degrees kelvin) is a straight line with slope equal to -delta HvH/R (delta HvH = van't Hoff or apparent enthalpy, R = gas constant) is not valid in many cases. In all the cases considered here, delta HvH is temperature dependent and is significantly different from the true or calorimetrically measured enthalpy, and the respective values for delta Cp are also significantly different.

Isothermal titration calorimetric (ITC) studies over a range of temperatures of the binding of maltose, maltotriose, maltotetraose and beta-cyclodextrin to the maltodextrin-binding protein (MBP) of Escherichia coli are reported. The binding constants of maltose, maltotriose and beta-cyclodextrin are not very different, namely 8.7 x 10(5), 13.0 x 10(5) and 2.55 x 10(5) M-1, respectively at 25 degrees C. The calorimetric data obtained with maltotetraose cannot be interpreted in terms of a definite binding constant. The binding of maltose and maltotriose is endothermic with a large entropy increase while that of beta-cyclodextrin is exothermic, with a smaller entropy increase. The binding of maltotetraose was endothermic or exothermic depending on the temperature.

Cytochrome b562 is a four-helix bundle protein containing a noncovalently bound b-type heme prosthetic group. For the first time, energetics of heme binding to an apocytochrome were measured by isothermal titration calorimetry. The heme is tightly bound to native apocytochrome b562, with a dissociation constant (Kd) of approximately 9 nM (DeltaG degrees = 11 kcal mol-1) at 25 degrees C. Unexpectedly, the thermally denatured apoprotein is also capable of specifically binding heme with modest affinity (Kd = 3 microM, DeltaG degrees = 7.6 kcal mol-1). This interaction results in the dependence of holocytochrome b562 stability on protein concentration in the submicromolar range.

Significantly different m values (1.9-2.7 kcal mol-1 M-1) were observed for point mutations at a single, solvent-exposed site (T53) in a variant of the B1 domain of streptococcal Protein G using guanidine hydrochloride (GuHCl) as a denaturant. This report focuses on elucidating the energetic and structural implications of these m-value differences in two Protein G mutants, containing Ala and Thr at position 53. These two proteins are representative of the high (m+) and low (m-) m-value mutants studied. Differential scanning calorimetry revealed no evidence of equilibrium intermediates. A comparison of GuHCl denaturation monitored by fluorescence and circular dichroism showed that secondary and tertiary structure denatured concomitantly. The rates of folding (286 S-1 for the m+ mutant and 952 S-1 for the m- mutant) and the rates of unfolding (11 S-1 for m+ mutant and 3 S-1 for the m- mutant) were significantly different, as determined by stopped-flow fluorescence. The relative solvation free energies of the transition states were identical for the two proteins (alpha ++ = 0.3). Small-angle X-ray scattering showed that the radius of gyration of the denatured state (Rgd) of the m+ mutant did not change with increasing denaturant concentrations (Rgd approximately 23 A); whereas, the Rgd of the m- mutant increased from approximately 17 A to 23 A with increasing denaturant concentration. The results indicate that the mutations exert significant effects in both the native and GuHCl-induced denatured state of these two proteins.

Isothermal calorimetric titration of 18-crown-6 ether with BaCl2 in pure aqueous solution over the temperature range 7-40 degrees C gives precise binding constants and enthalpy changes. Nonlinear least-squares fitting of the binding constants to the integrated van't Hoff equation, including a temperature-independent change in heat capacity, leads to van't Hoff enthalpies that differ significantly from the observed calorimetric enthalpies. This perplexing discrepancy appears at present to be very widely occurring.

Amino acid replacements in a homodimeric protein, Streptomyces subtilisin inhibitor, at the position of Val13, which is located at the center of the beta-sheet interface of the domain-domain interaction, changed both the overall stability and the denaturational scheme. All the mutant forms lost stability, losses in free energy at 82.21 degrees C at pH 7.0 obtained by calorimetric measurements ranging from 10.3 kcal (mol dimer)-1 for the Gly mutant, which involved a loss in enthalpy without a compensating loss in entropy, to 0.84 kcal (mol dimer)-1 for the Ile mutant, which involved comparable losses in enthalpy and entropy. A gain in enthalpy for the Ala mutant was overcome by a gain in entropy, resulting in a loss in free energy by 6.78 kcal (mol dimer)-1. It was found that decreases in enthalpy and entropy showed good correlation with increasing side-chain hydrophobicity, while no such correlation was present for free energy. Replacements with Gly, Ala, Met and Phe altered the thermal denaturational scheme from N2 reversible 2D, which was the case for the Leu and Ile mutants as well as for the wild-type, to N2 reversible 2N* reversible 2D, where the first step can be regarded as a dissociation of the native dimer followed by a major unfolding in the second step. The free energy acquired by forming the dimer was estimated to be 2.03 kcal mol-1 for the Ala mutant at 66.95 degrees C at pH 9.5 and lower for the other three mutants.

Rationally redesigned variants of the 4-helix-bundle protein Rop are described. The novel proteins have simplified, repacked, hydrophobic cores and yet reproduce the structure and native-like physical properties of the wild-type protein. The repacked proteins have been characterized thermodynamically and their equilibrium and kinetic thermal and chemical unfolding properties are compared with those of wild-type Rop. The equilibrium stability of the repacked proteins to thermal denaturation is enhanced relative to that of the wild-type protein. The rate of chemically induced folding and unfolding of wild-type Rop is extremely slow when compared with other small proteins. Interestingly, although the repacked proteins are more thermally stable than the wild type, their rates of chemically induced folding and unfolding are greatly increased in comparison to wild type. Perhaps as a consequence of this, their equilibrium stabilities to chemical denaturants are slightly reduced in comparison to the wild type.

Isothermal titration calorimetry (ITC) is used to study the thermodynamics of assembly of the three DNA oligonucleotides S1 (5'-GCCTGCCACCGC), S2 (5'-GCGGTGCGTCCG), and S3AA (5'-CGGACGAAGCAGGC) to form a three-way junction (TWJ) complex consisting of three double-helical arms radiating from a junction region having two unpaired adenosines in one strand (S3AA). The thermodynamics of assembly were measured for three different orders of addition of the component oligonucleotides at four temperatures between 10 and 25 degrees C. At each temperature studied, the overall values of delta H, delta S degrees, and delta G degrees for assembly of the complex from the component single strands were found to be independent of the order of addition. The enthalpy of binding, delta H, was found to be linearly dependent on temperature. From the temperature dependence of delta H, the change in heat capacity delta Cp, for the overall assembly of three strands to form the junction complex was calculated and found to be -1.6 kcal mol-1K-1. This work represents the first attempt to evaluate the thermodynamics of DNA TWJ formation by ITC.

Differential scanning calorimetry (DSC) is employed in a study of the effects of the p -nitrophenyl esters of the even numbered fatty acids from C 6 (caproic acid) to C 18 (stearic acid) on the main phase transition of multilamellar suspensions of dimyristoylphosphatidylcholine (DMPC). Mole fractions in the range of 0.02–0.12 were used. Within this concentration range the observed transitions could be well fit on the basis of a model assuming ideal behavior of the esters in both the gel and liquid crystal phases of the lipid, and independently additive van't Hoff and impurity broadening. As expected on the basis of this model, the transition temperatures decreased linearly with increasing ester concentration and the transition enthalpies were independent of ester concentration. Significant differences between the effects of the various esters were observed, but these did not change with changing ester chain length in any regular fashion.

In continuation of our earlier work on the effects of amino acid replacements on the thermodynamics of the thermal unfolding of T4 lysozyme [Kitamura, S.,Sturtevant, J. M. (1989) Biochemistry 28, 3788-3792; Connelly, P., Ghosaini, L., Hu, C.-Q., Kitamura, S., Tanaka, A.,Sturtevant, J. M. (1991) Biochemistry 30, 1887-1891; Hu, C.-Q., Kitamura, S., Tanaka, A.,Sturtevant, J. M. (1992) Biochemistry 31, 1643-1647], we report here a study by differential scanning calorimetry of the effects of five replacements at Ile3. Four of these replacements, those with Glu, Phe, Pro, and Thr, caused apparent destabilizations, while the replacement by Leu led to a small apparent stabilization. The largest observed destabilization (Ile3Pro) amounted to -3.0 kcal mol-1 in free energy at pH 2.00 and 38.8 degrees C (the denaturational temperature of the wild-type protein at this pH), and the largest stabilization amounted to +1.2 kcal mol-1 at pH 3.00 and 53.6 degrees C.

Thermodynamic studies of bacteriorhodopsin (BR) have been undertaken in order to investigate the factors that stabilize the structure of a membrane protein. The stability of the native, intact protein was compared to that of protein with retinal removed, and/or cleaved in one or two of the loops connecting the transmembrane helices. The stability was assessed using differential scanning calorimetry and thermal denaturation curves obtained from ultraviolet circular dichroism and absorption spectroscopy. Retinal binding and the loop connections were each found to make a small contribution to stability, and even a sample that was cleaved twice as well as bleached to remove retinal denatured well above room temperature. Removal of retinal destabilized the protein more than cleaving once, and about as much as cleaving twice. Retinal binding and the connections in the loops were found to stabilize BR in independent ways. Cleavage of the molecule into fragments did not reduce the intermolecular cooperativity of the denaturation. Dilution of the protein by addition of excess lipid in order to eliminate the purple membrane crystal lattice also did not alter the cooperativity. These results are used to compare the relative importance of various contributors to the stability of BR. Because parts of integral membrane proteins must fold in the hydrophobic region of lipid bilayers, the balance of interactions stabilizing the proteins' structure is likely to be both quantitatively and qualitatively different from that found in soluble proteins. The factors contributing to the stability of proteins that contain several transbilayer a-helices can be separated conceptually into those stabilizing the helices themselves and those stabilizing the interactions between the helices in the tertiary structure (Popot & Engelman, 1990). It has been proposed that the stability of the helices in bilayers arises from main-chain hydrogen bonding and from the hydrophobic nature of the side chains (Engelman et al., 1986) and experiments have shown that some individual helices from a polytopic membrane protein can be considered to be independent folding domains (Kahn & Engelman, 1992; Hunt et al., 1991). The relative importance of the factors leading to the association of the helices has not been as thoroughly explored. Possible major factors stabilizing helix-helix association in bacteriorhodopsin (BR)' can be divided roughly into four categories. First, the extramembranous loops connecting the helices may constrain the positions of the helices by being short or by forming secondary and tertiary structures. Second, polar amino acid side chains within the membrane may cause the helices to associate in such a way as to allow the formation of hydrogen bonds and ion pairs in order to minimize the

Differential scanning calorimetry has been used to investigate the thermodynamics of denaturation of ribonuclease T1 as a function of pH over the pH range 2-10, and as a function of NaCl and MgCl2 concentration. At pH 7 in 30 mM PIPES buffer, the thermodynamic parameters are as follows: melting temperature, T1/2 = 48.9 +/- 0.1 degrees C; enthalpy change, delta H = 95.5 +/- 0.9 kcal mol-1; heat capacity change, delta Cp = 1.59 kcal mol-1 K-1; free energy change at 25 degrees C, delta G degrees (25 degrees C) = 5.6 kcal mol-1. Both T1/2 = 56.5 degrees C and delta H = 106.1 kcal mol-1 are maximal near pH 5. The conformational stability of ribonuclease T1 is increased by 3.0 kcal/mol in the presence of 0.6 M NaCl or 0.3 M MgCl2. This stabilization results mainly from the preferential binding of cations to the folded conformation of the protein. The estimates of the conformational stability of ribonuclease T1 from differential scanning calorimetry are shown to be in remarkably good agreement with estimates derived from an analysis of urea denaturation curves.

In two recent papers, we reported the effects of several point mutations on the thermodynamics of the thermal unfolding of the lysozyme of phage T4 as determined by differential scanning calorimetry. The mutants studied were R96H [Kitamura, S.,Sturtevant, J.M. (1989) Biochemistry 28, 3788-3792] and T157 replaced by A, E, I, L, N, R, and V [Connelly, P., Ghosaini, L., Hu, C.-Q., Kitamura, S., Tanaka, A.,Sturtevant, J.M. (1991) Biochemistry 30, 1887-1891]. Here we report the results of a similar study of the single mutations A82P, A93P, and G113A and the double mutation C54T:C97A. The three single mutants all show small apparent stabilization at pH 2.5 and 46.2 degrees C (the denaturational temperature of the wild-type protein), amounting to -0.5 +/- 0.4 kcal mol-1 in free energy, whereas the double mutant shows a weak apparent destabilization, +0.8 +/- 0.4 kcal mol-1. As in all our previous studies of mutant proteins, the enthalpy changes produced by these mutations are in general of much larger magnitude than the corresponding free energy changes and frequently of opposite sign.

Two fragments of pancreatic ribonuclease A, a truncated version of S-peptide (residues 1-15) and S-protein (residues 21-124), combine to give a catalytically active complex designated ribonuclease S. We have substituted the wild-type residue Met-13 with six other hydrophobic residues ranging in size from alanine to phenylalanine and have determined the thermodynamic parameters associated with binding of these analogues to S-protein by titration calorimetry in the temperature range 5-25 degrees C. The heat capacity change (delta Cp) associated with binding was obtained from a global analysis of the temperature dependences of the free energies and enthalpies of binding. The delta Cp's were not correlated in any simple fashion with the nonpolar surface area (delta Anp) buried upon binding.

Amino acid substitutions at a site in the center of the bacteriophage protein P22 tailspike polypeptide chain suppress temperature-sensitive folding mutations at many sites throughout the chain. Characterization of the intracellular folding and chain assembly process reveals that the suppressors act in the folding pathway, inhibiting the aggregation of an early folding intermediate into the kinetically trapped inclusion body state. The suppressors alone increase the folding efficiency of the otherwise wild-type polypeptide chain without altering the stability or activity of the native state. These amino acid substitutions identify an unexpected aspect of the protein folding grammar--sequences within the chain that carry information inhibiting unproductive off-pathway conformations. Such mutations may serve to increase the recovery of protein products of cloned genes.

The reversible thermally induced unfolding of various forms of tropomyosin, a two-chain alpha-helical coiled coil, has been studied by high-sensitivity differential scanning calorimetry (DSC). Included in the study are the reduced and oxidized (disulfide cross-linked) forms of alpha alpha- and beta beta-tropomyosin, and the forms of alpha alpha-tropomyosin in which all sulfhydryl groups have been blocked by carboxymethylation or carboxyamidomethylation. Oxidation or blocking of the sulfhydryl groups of tropomyosin strongly affect the thermotropic behavior of the protein in unpredictable ways. The empirical results presented here are in qualitative agreement with those from an earlier DSC study of the oxidized and carboxymethylated forms of alpha alpha-tropomyosin [S.A. Potekhin and P.L. Privalov (1982) Journal of Molecular Biology, Vol. 159, pp. 519-535], but we find that a different decomposition into subtransitions is possible. Comparison of the alpha alpha and beta beta species indicates, in agreement with extant CD studies, that the noncross-linked beta beta species is somewhat less stable than its alpha alpha counterpart, but that cross-linking enhances the stability of the beta beta doubly cross-linked species by a greater amount and does not lead to the small low-temperature transition ("pretransition") seen in the singly cross-linked alpha alpha species.

High-sensitivity differential scanning calorimetry has been applied to the study of the reversible thermal unfolding of the lysozyme of T4 bacteriophage in which the threonine residue at position 157 has been replaced by seven different residues. High-resolution structures derived from X-ray crystallography have been reported for these and six other mutants by Alber et al. [Alber, T., Dao-Pin, S., Wilson, K., Wozniak, J. A., Cook, S. P., & Matthews, B. W. (1987) Nature 330, 41-46]. At pH 2.5 the changes relative to the wild-type protein in the standard free energy of unfolding produced by these mutations indicate apparent destabilizations of 0.6 kcal mol-1 (T157R) to 1.9 kcal mol-1 (T157I), whereas the changes in enthalpy of unfolding range from -5.8 kcal mol-1 (T157N) to 11.9 kcal mol-1 (T157E). Since the denaturations are in all cases accompanied by large changes in heat capacity amounting to 2.5 kcal K-1 mol-1, both the free energies and enthalpies are functions of temperature. An intriguing feature of the present results is the relatively large enthalpy changes and the corresponding compensating entropy changes. Our present understanding of the intramolecular energetics of proteins is insufficient to account for these changes.

The main phase transitions of aqueous dispersions of four synthetic phosphatidylcholines (PCs) containing sulfur atoms in thioester and thioether linkages have been studied by high sensitivity differential scanning calorimetry (DSC). The transition enthalpies ranged from 7.4 to 10.3 kcal mol-1, with values for tm, the temperature of maximal excess heat capacity, in the range 38.0 to 40.8 degrees C. The corresponding values for dipalmitoyl PC (DPPC) are 8.5 kcal mol-1 and 41.7 degrees C. Curve fitting required the sum of from one to four two-state components to give an accurate representation of the observed DSC curves. Comparison of the results given here with those reported by B.Z. Chowdhry, G. Lipka, J. Hajdu and J.M. Sturtevant, (1984) Biochemistry 23, 2044-2049, for PCs containing amide, ether or carbamoyl linkages in place of the usual ester bonds shows that small changes in organic structure can result in large changes in thermotropic behavior. The complexity in the cases showing more than a single two-state component is presumably due to a series of sequential cooperative transitions the character of which is at present unknown.

Mixed lipid samples containing dimyristoylglycerophosphocholine and small amounts of myristic acid were examined calorimetrically. Examination of multilamellar and small vesicle samples indicated that upon heating small vesicles combine to form more extended structures. An exothermic peak (at 19.5°C) can be associated with the structural transformation. The enthalpy for this process, which may be interpreted as vesicle-vesicle fusion, is found to be approx. -2 kcal/mol.

In the concentration range of 10 −5 to 10 −1 M Ca 2+ modulates the thermotropic properties of several neutral and anionic glycosphingolipids (galactosylceramide, asialo-G M1 , sulfatide, G M1 , G D1a , G T1b ) and of their mixtures with dipalmitoylphosphatidylcholine. The transition temperature of gangliosides is not appreciably changed while the transition enthalpy increases by 20% in the presence of Ca 2+ . The more marked effect of Ca 2+ is on the thermotropic behavior of systems containing sulfatide. Increasing concentrations of Ca 2+ between 10 −5 and 10 −3 M (up to a molar ratio of Ca 2+ /sulfatide 1:2) induce a progressive increase of both the transition temperature and enthalpy. Further increases up to 10 −1 M Ca 2+ induce a new phase transition at a lower temperature. No evidence is found for induction of phase separation of pure glycosphingolipid-Ca 2+ domains in mixtures of any of the glycosphingolipids with dipalmitoylphosphatidylcholine. The modification of the phase behavior of anionic glycosphingolipids by Ca 2+ does not involve detectable variations of the intermolecular packing but is accompanied by marked modifications of the dipolar properties of the polar head group region.

The apparent latent heat of fusion (ΔH) of thick pastes ofHalobacterium halobium was determined by differential scanning calorimetry and compared with values obtained for a control paste of identical composition but containing bacteria already lysed by freezing and thawing. The rationale of the experiment was that, if there were any physico-chemical “abnormality” in the state of intracellular water or K+, it might be reflected in the freezing behavior of the two types of preparation. Differences were observed, the paste of whole cells consistently giving higher values of ΔH than the lysed bacteria. Furthermore, the differences between the two types of preparation increased with the age of the culture. Scans of whole-cell pastes apparently distinguished between intra- and extracellular water and enabled ΔH to be calculated for those two aqueous compartments. Supplementary measurements were also made on solutions of NaCl, KCl, and mixtures of the two salts. Sodium chloride lowered ΔH whereas KCl raised it. Differences in ΔH between internal and external solutions and between whole and lysed bacteria could be attributed to the different effects of the two salts on freezing behavior. Changing values of ΔH with the age of a culture correlated with changing ratios of K+/Na+ in the bacteria. It is concluded that there is no evidence at this level of any physicochemical abnormality in the halobacterial cytoplasm and no reason for supposing that retention of KCl by these bacteria is achieved by any agency other than the plasma membrane.

The binding of nicotinamide adenine dinucleotide (NAD+) to yeast glyceraldehyde-3-phosphate dehydrogenase (GPDH) has been studied at pH 6.5 and 8.5, at 5,25, and 40 degrees C, by calorimetry, fluorometry, spectrophotometry, equilibrium dialysis, and flow dialysis. As reported earlier for pH 7.3 (Velick S.F., Baggott, J.P., and Sturtevant, J.M. (1971), Biochemistry 10, 779), the binding is accompanied by enthalpy changes which become rapidly more negative as the temperature increases, with delta Cp = -500 to -750 cal deg-1 (mole of NAD+ bound)-1, and by entropy changes which also, as required by the large negative delta Cp, become rapidly more negative with increasing temperature. The binding data at pH 6.5 can be fitted on the basis of either four identical noninteracting sites, or of four sites showing a small degree of negative cooperativity. The data at pH 8.5, particularly at 40 degrees C, require the introduction of positive cooperativity, as was previously shown by Kirschner et al. (Kirschner, K., Eigen, M., Bittman, R., and Voigt, B. (1966), Proc. Natl. Acad. Sci. U.S.A. 56, 1661), and can be equally well fitted on the basis of a sequential model (Adair, G.S. (1925), J. Biol. Chem. 63, 529) or a concerted model (Monod, J., Wyman, J., and Changeux, J.P. (1965), J. Mol. Biol. 12, 88). It is proposed that the observed thermodynamic changes are largely the result of a hydrophobic effect due to a decrease in the exposure of nonpolar groups to the solvent, and of a tightening of the protein structure when the coenzyme is bound with concomitant decrease in the number of easily excitable internal degrees of freedom.

Dispersions of pure, spin-label phosphatidylcholines in aqueous buffer have been investigated with the Privalov high-sensitivity differential scanning calorimeter. The lipids studied are mixed-chain ones in which C-2 of glycerol bears a spin-label derivative of stearic acid and the fatty acid group at C-1 is palmitate. A well-defined phase transition is observed at 30.3-30.7 degrees C for the phosphatidylcholine labeled near the polar end of the stearate chain (label at C-5). A sharp transition (32-34 degrees C) is also observed for the lipid spin-labeled near the terminal methyl of stearate (label at C-16), but the thermodynamic parameters for this lipid depend strongly on the history of the sample. Calorimetric evidence for hysteresis in the phase transition of the C-16-labeled lipid is presented. In contrast to the above spin-label lipids, the lipid labeled at C-12 does not show a sharp transition in the region 5-35 degrees C. In general, therefore, the thermal behavior of the spin-label phosphatidylcholines resembles that of phosphatidylcholines bearing double bonds or branched methyl groups at similar locations on acyl chains. During synthesis of mixed-chain lipids, migration of acyl chains occurs. Methyl esterification procedures which are compatible with the acid-labile spin-label group are described. Gas chromatographic analysis of methyl esters shows that chain migration during synthesis gives 15-20% of the spin-label fatty acid at the glycerol C-1 position.