Thermodynamic investigations of proteins. IV. Calcium binding protein parvalbumin.

The conformational transitions of calcium binding protein parvalbumin III from carp muscle were studied by scanning calorimetry, potentiometric titration and isothermal calorimetric titration. Changes of Gibbs energy, enthalpy and partial heat capacity were determined. The removal of calcium ions by EDTA is accompanied by 1) a heat absorption of 75 +/- 10 kJ per mole of the protein, 2) a decrease in the Gibbs energy of protein structure stabilisation of about 42 kJ mol-1 and 3) a decrease in thermostability by more than 50 K. The protonation of the acidic groups leads to a loss of calcium followed by denaturation, while the pH of the transition strongly depends on calcium activity. The enthalpy and heat capacity changes at denaturation are comparable with the values observed for other compact globular proteins.     

Thermodynamic investigations of proteins. III. Thermodynamic description of lysozyme.

Standard functions of enthalpy, entropy and the Gibbs energy of native and denatured lysozyme in the range of 0-100 degrees C and pH 1.5-7.0 are represented in three-dimensional projections. The denaturational Gibbs energy change reaches 16 kcal mol-1 at conditions of maximal protein stability (0 degrees C, pH 4.5-7.0) and equals 14.5 kcal mol-1 at 25 degrees C and neutral pH. This result was found to be in agreement with the data reported from guanidine hydrochloride denaturation studies. Partial thermodynamic functions of the conformational and ionizational changes of the protein are obtained from entropy and Gibbs-energy changes in denaturation. The conformational partial entropy and Gibbs-energy change are found to be independent of pH. The pH-dependent partial ionizational entropy and Gibbs-energy changes are induced by normalization of the ionization behaviour of buried groups and cause a decrease of protein stability.     

Rapid screening of membrane topology of secondary transport proteins

Limited experimental data may be very useful to discriminate between membrane topology models of membrane proteins derived from different methods. A membrane topology screening method is proposed by which the cellular disposition of three positions in a membrane protein are determined, the N- and the C-termini and a position in the middle of the protein. The method involves amplification of the encoding genes or gene fragments by PCR, rapid cloning in dedicated vectors by ligation independent cloning, and determination of the cellular disposition of the three sites using conventional techniques. The N-terminus was determined by labeling with a fluorescent probe, the central position and the C-terminus by the reporter fusion technique using alkaline phosphatase (PhoA) and green fluorescence protein (GFP) as reporters. The method was evaluated using 16 transporter proteins of known function from four different structural classes. For 13 proteins a complete set of three localizations was obtained. The experimental data was used to discriminate between membrane topology models predicted by TMHMM, a widely used predictor using the amino acid sequence as input and by MemGen that uses hydropathy profile alignment and known 3D structures or existing models. It follows that in those cases where the models from the two methods were similar, the models were consistent with the experimental data. In those cases where the models differed, the MemGen model agreed with the experimental data. Three more recent predictors, MEMSAT3, OCTOPUS and TOPCONS showed a significantly higher consistency with the experimental data than observed with TMHMM.     

Application of inverse regression for estimating molecular masses and Stokes radii of globular proteins by gel filtration chromatography

The aim of this paper is to apply inverse regression as a method for treating experimental data obtained from gel filtration chromatography in order to obtain estimates of hydrodynamic parameters of globular proteins with true confidence intervals. The method is illustrated with the determination, using inverse regression, of molecular mass and Stokes radius for four test proteins (aldolase, chymotrypsinogen A, aldehyde oxidase and xanthine oxidase), from experimental data obtained with a Sephacryl S-300 column. A simple personal computer (PC) program written in standard basic, that is useful for this purpose, is included.     

Proteomics of Synechocystis sp. PCC 6803. Identification of novel integral plasma membrane proteins

The cyanobacterial plasma membrane is an essential cell barrier with functions such as the control of taxis, nutrient uptake and secretion. These functions are carried out by integral membrane proteins, which are difficult to identify using standard proteomic methods. In this study, integral proteins were enriched from purified plasma membranes of Synechocystis sp. PCC 6803 using urea wash followed by protein resolution in 1D SDS/PAGE. In total, 51 proteins were identified by peptide mass fingerprinting using MALDI-TOF MS. More than half of the proteins were predicted to be integral with 1-12 transmembrane helices. The majority of the proteins had not been identified previously, and include members of metalloproteases, chemotaxis proteins, secretion proteins, as well as type 2 NAD(P)H dehydrogenase and glycosyltransferase. The obtained results serve as a useful reference for further investigations of the address codes for targeting of integral membrane proteins in cyanobacteria.     

The development of a continuous isothermal titration calorimetric method for equilibrium studies

A continuous isothermal titration calorimetry (cITC) method for microcalorimeters has been developed. The method is based on continuous slow injection of a titrant into the calorimetric vessel. The experimental time for a cITC binding experiment is 12-20 min and the number of data points obtained is on the order of 1000. This gives an advantage over classical isothermal titration calorimetry (ITC) binding experiments that need 60-180 min to generate 20-30 data points. The method was validated using two types of calorimeters, which differ in calorimetric principle, geometry, stirring, and way of delivering the titrant into the calorimetric vessel. Two different experimental systems were used to validate the method: the binding of Ba(2+) to 18-crown-6 and the binding of cytidine 2'-monophosphate to RNAse A. Both systems are used as standard test systems for titration calorimetry. Computer simulations show that the dynamic range for determination of equilibrium constants can be increased by three orders of magnitude compared to that of classical ITC, making it possible to determine high affinities. Simulations also show an improved possibility to elucidate the actual binding model from cITC data. The simulated data demonstrate that cITC makes it easier to discriminate between different thermodynamic binding models due to the higher density of data points obtained from one experiment.     

Calculation of translational friction and intrinsic viscosity. II. Application to globular proteins.

The translational friction coefficients and intrinsic viscosities of four proteins (ribonuclease A, lysozyme, myoglobin, and chymotrypsinogen A) are calculated using atomic-level structural details. Inclusion of a 0.9-A-thick hydration shell allows calculated results for both hydrodynamic properties of each protein to reproduce experimental data. The use of detailed protein structures is made possible by relating translational friction and intrinsic viscosity to capacitance and polarizability, which can be calculated easily. The 0.9-A hydration shell corresponds to a hydration level of 0.3-0.4 g water/g protein. Hydration levels within this narrow range are also found by a number of other techniques such as nuclear magnetic resonance spectroscopy, infrared spectroscopy, calorimetry, and computer simulation. The use of detailed protein structures in predicting hydrodynamic properties thus allows hydrodynamic measurement to join the other techniques in leading to a unified picture of protein hydration. In contrast, earlier interpretations of hydrodynamic data based on modeling proteins as ellipsoids gave hydration levels that varied widely from protein to protein and thus challenged the existence of a unified picture of protein hydration.     

Application of long-range order to predict unfolding rates of two-state proteins

Predicting the experimental unfolding rates of two-state proteins and models describing the unfolding rates of these proteins is quite limited because of the complexity present in the unfolding mechanism and the lack of experimental unfolding data compared with folding data. In this work, 25 two-state proteins characterized by Maxwell et al. (Protein Sci 2005;14:602–616) using a consensus set of experimental conditions were taken, and the parameter long-range order (LRO) derived from their three-dimensional structures were related with their experimental unfolding rates ln(k(u)). From the total data set of 30 proteins used by Maxwell et al. (Protein Sci 2005;14:602–616), five slow-unfolding proteins with very low unfolding rates were considered to be outliers and were not included in our data set. Except all beta structural class, LRO of both the all-alpha and mixed-class proteins showed a strong inverse correlation of r = -0.99 and -0.88, respectively, with experimental ln(k(u)). LRO shows a correlation of -0.62 with experimental ln(k(u)) for all-beta proteins. For predicting the unfolding rates, a simple statistical method has been used and linear regression equations were developed for individual structural classes of proteins using LRO, and the results obtained showed a better agreement with experimental results.     

Experimental parameterization of an energy function for the simulation of unfolded proteins

The determination of conformational preferences in unfolded and disordered proteins is an important challenge in structural biology. We here describe an algorithm to optimize energy functions for the simulation of unfolded proteins. The procedure is based on the maximum likelihood principle and employs a fast and efficient gradient descent method to find the set of parameters of the energy function that best explain the experimental data. We first validate the method by using synthetic reference data, and subsequently apply the algorithms to data from nuclear magnetic resonance spin-labeling experiments on the Δ131Δ fragment of Staphylococcal nuclease. A significant strength of the procedure that we present is that it directly uses experimental data to optimize the energy parameters, without relying on the availability of high resolution structures. The procedure is fully general and can be applied to a range of experimental data and energy functions including the force fields used in molecular dynamics simulations.     

The effect of pore-size distribution on the binding of proteins to porous resin beads

The effect of the distribution of pore radii in the resin beads on protein binding was taken into account to analyze the elution profiles of proteins from the polymer-packed column obtained by repetitive injection method. By assuming that the distribution of pore radii in the resin beads is logarithmic Gaussian, the theoretical curves obtained agreed well with the experimental data.     

Ligand-induced thermostability in proteins: thermodynamic analysis of ANS-albumin interaction

A comparative thermodynamic study of the interaction of anilinonaphthalene sulfonate (ANS) derivatives with bovine serum albumin (BSA) was performed by using differential scanning calorimetry (DSC) and isothermal titration calorimetry (ITC). The chemically related ligands, 1,8-ANS and 2,6-ANS, present a similar affinity for BSA with different binding energetics. The analysis of the binding driving forces suggests that not only hydrophobic effect but also electrostatic interactions are relevant, even though they have been extensively used as probes for non-polar domains in proteins. Ligand association leads to an increase in protein thermostability, indicating that both dyes interact mainly with native BSA. ITC data show that 1,8-ANS and 2,6-ANS have a moderate affinity for BSA, with an association constant of around 1-9x10(5) M(-1) for the high-affinity site. Ligand binding is disfavoured by conformational entropy. The theoretical model used to simulate DSC data satisfactorily reproduces experimental thermograms, validating this approach as one which provides new insights into the interaction between one or more ligands with a protein. By comparison with 1,8-ANS, 2,6-ANS appears as a more "inert" probe to assess processes which involve conformational changes in proteins.     

Wavelet change-point prediction of transmembrane proteins

MOTIVATION: A non-parametric method, based on a wavelet data-dependent threshold technique for change-point analysis, is applied to predict location and topology of helices in transmembrane proteins. A new propensity scale generated from a transmembrane helix database is proposed. RESULTS: We show that wavelet change-point performs well for smoothing hydropathy and transmembrane profiles generated using different scales. We investigate which wavelet bases and threshold functions are overall most appropriate to detect transmembrane segments. Prediction accuracy is based on the analysis of two data sets used as standard benchmarks for transmembrane prediction algorithms. The analysis of a test set of 83 proteins results in accuracy per segment equal to 98.2%; the analysis of a 48 proteins blind-test set, i.e. containing proteins not used to generate the propensity scales, results in accuracy per segment equal to 97.4%. We believe that this method can also be applied to the detection of boundaries of other patterns such as G + Cisochores and dot-plots. AVAILABILITY: The transmembrane database, TMALN and source code are available upon request from the authors.     

Structural changes and fluctuations of proteins. II. Analysis of the denaturation of globular proteins.

The statistical thermodynamic model of protein structure proposed in paper I is developed with special attention to the hydrophobic interaction. Calorimetric measurements of the thermal denaturation of five globular proteins, ribonuclease A, lysozyme, alpha-chymotrypsin, cytochrome c, and myoglobin, are quantitatively analyzed using the model. The thermodynamic parameters obtained by the least squares method reflect the global, average properties of proteins and are in good agreement with the expected values estimated from experimental and theoretical studies for model peptides. The average bond energy epsilon is well related to the tertiary structure of each protein. However, the difference in the parameters between different proteins is not observed for the cooperative energy ZJ and the chain entropy alpha. The individuality of a protein as far as its structural stability is concerned, is mainly reflected by the parameter gamma specifying the hydrophobic nature of a protein. The model is further applied in the analysis of several aspects of the structural stability of globular proteins. Denaturation induced by denaturants, salts, and pH are also explained by the model in a unified manner.     

Misuse of thermodynamics in the interpretation of isothermal titration calorimetry data for ligand binding to proteins

Isothermal titration calorimetry (ITC) has given a mass of data on the binding of small molecules to proteins and other biopolymers, with particular interest in drug binding to proteins chosen as therapeutic indicators. Interpretation of the enthalpy data usually follows an unsound protocol that uses thermodynamic relations in circumstances where they do not apply. Errors of interpretation include incomplete definitions of ligand binding and equilibrium constants and neglect of the non-ideality of the solutions under study, leading to unreliable estimates of standard free energies and entropies of binding. The mass of reported thermodynamic functions for ligand binding to proteins estimated from ITC enthalpies alone is consequently of uncertain thermodynamic significance and utility. ITC and related experiments to test the protocol assumptions are indicated. A thermodynamic procedure avoiding equilibrium constants or other reaction models and not requiring protein activities is given. The discussion draws attention to the fundamental but neglected relation between the thermodynamic activity and bioactivity of drugs and to the generally unknown thermodynamic status of ligand solutions, which for drugs relates directly to effective therapeutic dosimetry.     

Effect of interdomain dynamics on the structure determination of modular proteins by small-angle scattering

Multidomain proteins in which consecutive globular regions are connected by linkers are prevalent in nature (Levitt in Proc Natl Acad Sci USA 106:11079–11084, 2009). Some members of this family have largely resisted structural characterization as a result of challenges associated with their inherent flexibility. Small-angle scattering (SAS) is often the method of choice for their structural study. An extensive set of simulated data for both flexible and rigid multidomain systems was analyzed and modeled using standard protocols. This study clearly shows that SAXS profiles obtained from highly flexible proteins can be wrongly interpreted as arising from a rigid structure. In this context, it would be important to identify features from the SAXS data or from the derived structural models that indicate interdomain motions to differentiate between these two scenarios. Features of SAXS data that identify flexible proteins are: (1) general attenuation of fine structure in the scattering profiles, which becomes more dramatic in Kratky representations, and (2) a reduced number of interdomain correlation peaks in p(r) functions that also present large D _max values and a smooth decrease to 0. When modeling this dynamically averaged SAXS data, the structures obtained present characteristic trends: (1) ab initio models display a decrease in resolution, and (2) rigid-body models present highly extended conformations with a lack of interdomain contacts. The ensemble optimization method represents an excellent strategy to identify interdomain motions unambiguously. This study provides information that should help researchers to select the best modeling strategy for the structural interpretation of SAS experiments of multidomain proteins.     

CDF it all: Consensus prediction of intrinsically disordered proteins based on various cumulative distribution functions

Many biologically active proteins are intrinsically disordered. A reasonable understanding of the disorder status of these proteins may be beneficial for better understanding of their structures and functions. The disorder contents of disordered proteins vary dramatically, with two extremes being fully ordered and fully disordered proteins. Often, it is necessary to perform a binary classification and classify a whole protein as ordered or disordered. Here, an improved error estimation technique was applied to develop the cumulative distribution function (CDF) algorithms for several established disorder predictors. A consensus binary predictor, based on the artificial neural networks, NN-CDF, was developed by using output of the individual CDFs. The consensus method outperforms the individual predictors by 4-5% in the averaged accuracy.     

In silico chaperonin-like cycle helps folding of proteins for structure prediction

Currently, one of the most serious problems in protein-folding simulations for de novo structure prediction is conformational sampling of medium-to-large proteins. In vivo, folding of these proteins is mediated by molecular chaperones. Inspired by the functions of chaperonins, we designed a simple chaperonin-like simulation protocol within the framework of the standard fragment assembly method: in our protocol, the strength of the hydrophobic interaction is periodically modulated to help the protein escape from misfolded structures. We tested this protocol for 38 proteins and found that, using a certain defined criterion of success, our method could successfully predict the native structures of 14 targets, whereas only those of 10 targets were successfully predicted using the standard protocol. In particular, for non-α-helical proteins, our method yielded significantly better predictions than the standard approach. This chaperonin-inspired protocol that enhanced de novo structure prediction using folding simulations may, in turn, provide new insights into the working principles underlying the chaperonin system.     

Annotation of hypothetical proteins orthologous in Pongo abelii and Sus scrofa

A hypothetical protein is predicted to be expressed from an open reading frame without known experimental evidence of translation. They constitute a substantial fraction of proteomes. Domain extraction from these hypothetical sequences helps to search for protein coding genes for protein structural and functional annotation. We describe the analysis of prediction data in a sequence dataset of hypothetical protein orthologs of Pongo abelii (orangutan) and Sus scrofa (pig). It should be noted that these orangutan-pig orthologs are also non-homologous to human proteins. These predicted data find application in the genome wide annotation of proteins in poorly understood genomes.

Abbreviations

PDB - Protein Data Bank, DEG - Database of Essential Genes, CDD - Conserved Domain Database, IUCN - International Union for Conservation of Nature.     

Computation of the electrophoretic mobility of proteins.

A scheme is presented for computing the electrophoretic mobility of proteins in free solution, accounting for the details of the protein shape and charge distribution. The method of Teubner is implemented using a boundary integral formulation within which the velocity distribution, the equilibrium electrical potential around the molecule, and the potential distribution due to the applied field are solved for numerically using the boundary element method. Good agreement of the numerical result is obtained for spheres with the corresponding semi-analytical specialization of Henry's analysis. For protein systems, the method is applied to lysozyme and ribonuclease A. In both cases, the predicted mobility tensors are fairly isotropic, with the resulting scalar mobilities being significantly smaller than for spheres of equal volume and net charge. Comparisons with previously published experimental results for ribonuclease show agreement to be excellent in the presence of a net charge, but poorer at the point of zero charge. The approach may be useful for evaluating approximate methods for estimating protein electrophoretic mobilities and for using electrophoretic measurements to obtain insight into charge distributions on proteins.