Mapping the active site of CD59.

CD59 is a widely distributed membrane-bound inhibitor of the cytolytic membrane attack complex (MAC) of complement. This small (77 amino acid) glycoprotein is a member of the Ly6 superfamily of proteins and is important in protecting host cells from the lytic and proinflammatory activity of the MAC. CD59 functions by binding to C8 and/or C9 in the nascent MAC and interfering with C9 membrane insertion and polymerization. We present data obtained from a combination of molecular modeling and mutagenesis techniques, which together indicate that the active site of CD59 is located in the vicinity of a hydrophobic groove on the face of the molecule opposite to a "hydrophobic strip" suggested earlier. In addition, removal of the single N-linked glycosylation site at Asn18 of CD59 resulted in an enhancement of complement inhibitory activity.

Spatial chemical distance based on atomic property fields.

Similarity of compound chemical structures often leads to close pharmacological profiles, including binding to the same protein targets. The opposite, however, is not always true, as distinct chemical scaffolds can exhibit similar pharmacology as well. Therefore, relying on chemical similarity to known binders in search for novel chemicals targeting the same protein artificially narrows down the results and makes lead hopping impossible. In this study we attempt to design a compound similarity/distance measure that better captures structural aspects of their pharmacology and molecular interactions. The measure is based on our recently published method for compound spatial alignment with atomic property fields as a generalized 3D pharmacophoric potential. We optimized contributions of different atomic properties for better discrimination of compound pairs with the same pharmacology from those with different pharmacology using Partial Least Squares regression. Our proposed similarity measure was then tested for its ability to discriminate pharmacologically similar pairs from decoys on a large diverse dataset of 115 protein-ligand complexes. Compared to 2D Tanimoto and Shape Tanimoto approaches, our new approach led to improvement in the area under the receiver operating characteristic curve values in 66 and 58% of domains respectively. The improvement was particularly high for the previously problematic cases (weak performance of the 2D Tanimoto and Shape Tanimoto measures) with original AUC values below 0.8. In fact for these cases we obtained improvement in 86% of domains compare to 2D Tanimoto measure and 85% compare to Shape Tanimoto measure. The proposed spatial chemical distance measure can be used in virtual ligand screening.

Stabilization of G-quadruplex DNA with platinum(II) Schiff base complexes: luminescent probe and down-regulation of c-myc oncogene expression.

The interactions of a series of platinum(II) Schiff base complexes with c-myc G-quadruplex DNA were studied. Complex [PtL(1a)] (1 a; H(2)L(1a)=N,N'-bis(salicylidene)-4,5-methoxy-1,2-phenylenediamine) can moderately inhibit c-myc gene promoter activity in a cell-free system through stabilizing the G-quadruplex structure and can inhibit c-myc oncogene expression in cultured cells. The interaction between 1 a and G-quadruplex DNA has been examined by (1)H NMR spectroscopy. By using computer-aided structure-based drug design for hit-to-lead optimization, an in silico G-quadruplex DNA model has been constructed for docking-based virtual screening to develop new platinum(II) Schiff base complexes with improved inhibitory activities. Complex [PtL(3)] (3; H(2)L(3)=N,N'-bis{4-[1-(2-propylpiperidine)oxy]salicylidene}-4,5-methoxy-1,2-phenylenediamine) has been identified with a top score in the virtual screening. This complex was subsequently prepared and experimentally tested in vitro for its ability to stabilize or induce the formation of the c-myc G-quadruplex. The inhibitory activity of 3 (IC(50)=4.4 muM) is tenfold more than that of 1 a. The interaction between 1 a or 3 with c-myc G-quadruplex DNA has been examined by absorption titration, emission titration, molecular modeling, and NMR titration experiments, thus revealing that both 1 a and 3 bind c-myc G-quadruplex DNA through an external end-stacking mode at the 3' terminal face of the G-quadruplex. Such binding of G-quadruplex DNA with 3 is accompanied by up to an eightfold increase in the intensity of photoluminescence at lambda(max)=652 nm. Complex 3 also effectively down-regulated the expression of c-myc in human hepatocarcinoma cells.

Novel anchorage of GluR2/3 to the postsynaptic density by the AMPA receptor-binding protein ABP.

We report the cloning of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid (AMPA) receptor-binding protein (ABP), a postsynaptic density (PSD) protein related to glutamate receptor-interacting protein (GRIP) with two sets of three PDZ domains, which binds the GluR2/3 AMPA receptor subunits. ABP exhibits widespread CNS expression and is found at the postsynaptic membrane. We show that the protein interactions of the ABP/GRIP family differ from the PSD-95 family, which binds N-methyl-D-aspartate (NMDA) receptors. ABP binds to the GluR2/3 C-terminal VKI-COOH motif via class II hydrophobic PDZ interactions, distinct from the class I PSD-95-NMDA receptor interaction. ABP and GRIP also form homo- and heteromultimers through PDZ-PDZ interactions but do not bind PSD-95. We suggest that the ABP/GRIP and PSD-95 families form distinct scaffolds that anchor, respectively, AMPA and NMDA receptors.

NRIF3 is a novel coactivator mediating functional specificity of nuclear hormone receptors.

Many nuclear receptors are capable of recognizing similar DNA elements. The molecular event(s) underlying the functional specificities of these receptors (in regulating the expression of their native target genes) is a very important issue that remains poorly understood. Here we report the cloning and analysis of a novel nuclear receptor coactivator (designated NRIF3) that exhibits a distinct receptor specificity. Fluorescence microscopy shows that NRIF3 localizes to the cell nucleus. The yeast two-hybrid and/or in vitro binding assays indicated that NRIF3 specifically interacts with the thyroid hormone receptor (TR) and retinoid X receptor (RXR) in a ligand-dependent fashion but does not bind to the retinoic acid receptor, vitamin D receptor, progesterone receptor, glucocorticoid receptor, or estrogen receptor. Functional experiments showed that NRIF3 significantly potentiates TR- and RXR-mediated transactivation in vivo but has little effect on other examined nuclear receptors. Domain and mutagenesis analyses indicated that a novel C-terminal domain in NRIF3 plays an essential role in its specific interaction with liganded TR and RXR while the N-terminal LXXLL motif plays a minor role in allowing optimum interaction. Computer modeling and subsequent experimental analysis suggested that the C-terminal domain of NRIF3 directly mediates interaction with liganded receptors through an LXXIL (a variant of the canonical LXXLL) module while the other part of the NRIF3 protein may still play a role in conferring its receptor specificity. Identification of a coactivator with such a unique receptor specificity may provide new insight into the molecular mechanism(s) of receptor-mediated transcriptional activation as well as the functional specificities of nuclear receptors.

Long pentraxins: an emerging group of proteins with diverse functions.

The earliest described pentraxins, C reactive protein (CRP) and serum amyloid P component (SAP), are cytokine-inducible acute phase proteins implicated in innate immunity whose concentrations in the blood increase dramatically upon infection or trauma. The highly conserved family of pentraxins was thought to consist solely of approximately 25 kDa proteins. Recently, several distinct larger proteins have been identified in which only the C-terminal halves show characteristic features of the pentraxin family. One of the recently described "long" pentraxins (TSG-14/PTX3) is inducible by TNF or IL-1 and is produced during the acute phase response. Other newly identified long pentraxins are constitutively expressed proteins associated with sperm-egg fusion (apexin/p50), may function at the neuronal synapse (neuronal pentraxin I, NPI), or may serve yet other, unknown functions (NPII and XL-PXN1). Evidence obtained by molecular modeling and by direct physicochemical analysis suggests that TSG-14 protein retains some characteristic structural features of the pentraxins, including the formation of pentameric complexes.

Soft docking an L and a D peptide to an anticholera toxin antibody using internal coordinate mechanics.

BACKGROUND: The tremendous increase in sequential and structural information is a challenge for computer-assisted modelling to predict the binding modes of interacting biomolecules. One important area is the structural understanding of protein-peptide interactions, information that is increasingly important for the design of biologically active compounds. RESULTS: We predicted the three-dimensional structure of a complex between the monoclonal antibody TE33 and its cholera-toxin-derived peptide epitope VPGSQHID. Using the internal coordinate mechanics (ICM) method of flexible docking, the bound conformation of the initially extended peptide epitope to the antibody crystal or modelled structure reproduced the known binding conformation to a root mean square deviation of between 1.9 A and 3.1 A. The predicted complexes are in good agreement with binding data obtained from substitutional analyses in which each epitope residue is replaced by all other amino acids. Furthermore, a de novo prediction of the recently discovered TE33-binding D peptide dwGsqhydp (single-letter amino acid code where D amino acids are represented by lower-case letters) explains results obtained from binding studies with 172 peptide analogues. CONCLUSIONS: Despite the difficulties arising from the huge conformational space of a peptide, this approach allowed the prediction of the correct binding orientation and the majority of essential binding features of a peptide-antibody complex.

The crystal structure of an engineered monomeric triosephosphate isomerase, monoTIM: the correct modelling of an eight-residue loop.

BACKGROUND: The triosephosphate isomerase (TIM) fold is found in several different classes of enzymes, most of which are oligomers; TIM itself always functions as a very tight dimer. It has recently been shown that a monomeric form of TIM ('monoTIM') can be constructed by replacing a 15-residue interface loop, loop-3, with an eight-residue fragment; modelling suggests that this should result in a short strain-free turn, resulting in the subsequent helix, helix-A3, having an additional turn at its amino terminus. RESULTS: The crystal structure of monoTIM shows that it retains the characteristic TIM-barrel (betaalpha)8-fold and that the new loop has a structure very close to that predicted. Two other interface loops, loop-1 and loop-4, which contain the active site residues Lys13 and His95, respectively, show significant changes in structure in monoTIM compared with dimeric wild-type TIM. CONCLUSION: The observed structural differences between monoTIM and wild-type TIM indicate that the dimeric appearance of TIM determines the location and conformation of two of the four catalytic residues.

Three new crystal structures of point mutation variants of monoTIM: conformational flexibility of loop-1, loop-4 and loop-8.

BACKGROUND: Wild-type triosephosphate isomerase (TIM) is a very stable dimeric enzyme. This dimer can be converted into a stable monomeric protein (monoTIM) by replacing the 15-residue interface loop (loop-3) by a shorter, 8-residue, loop. The crystal structure of monoTIM shows that two active-site loops (loop-1 and loop-4), which are at the dimer interface in wild-type TIM, have acquired rather different structural properties. Nevertheless, monoTIM has residual catalytic activity. RESULTS: Three new structures of variants of monoTIM are presented, a double-point mutant crystallized in the presence and absence of bound inhibitor, and a single-point mutant in the presence of a different inhibitor. These new structures show large structural variability for the active-site loops, loop-1, loop-4 and loop-8. In the structures with inhibitor bound, the catalytic lysine (Lys13 in loop-1) and the catalytic histidine (His95 in loop-4) adopt conformations similar to those observed in wild-type TIM, but very different from the monoTIM structure. CONCLUSIONS: The residual catalytic activity of monoTIM can now be rationalized. In the presence of substrate analogues the active-site loops, loop-1, loop-4 and loop-8, as well as the catalytic residues, adopt conformations similar to those seen in the wild-type protein. These loops lack conformational flexibility in wild-type TIM. The data suggest that the rigidity of these loops in wild-type TIM, resulting from subunit-subunit contacts at the dimer interface, is important for optimal catalysis.

High-throughput docking for lead generation.

Recent improvements in flexible docking technology may lead to a bigger role for computational methods in lead discovery. Although fast and accurate computational prediction of binding affinities for an arbitrary molecule is still beyond the limits of current methods, the docking and screening procedures can select small sets of likely lead candidates from large libraries of either commercially or synthetically available compounds.

Biased probability Monte Carlo conformational searches and electrostatic calculations for peptides and proteins.

Two major components are required for a successful prediction of the three-dimensional structure of peptides and proteins: an efficient global optimization procedure which is capable of finding an appropriate local minimum for the strongly anisotropic function of hundreds of variables, and a set of free energy components for a protein molecule in solution which are computationally inexpensive enough to be used in the search procedure, yet sufficiently accurate to ensure the uniqueness of the native conformation. We here found an efficient way to make a random step in a Monte Carlo procedure given knowledge of the energy or statistical properties of conformational subspaces (e.g. phi-psi zones or side-chain torsion angles). This biased probability Monte Carlo (BPMC) procedure randomly selects the subspace first, then makes a step to a new random position independent of the previous position, but according to the predefined continuous probability distribution. The random step is followed by a local minimization in torsion angle space. The positions, sizes and preferences for high-probability zones on phi-psi maps and chi-angle maps were calculated for different residue types from the representative set of 191 and 161 protein 3D-structures, respectively. A fast and precise method to evaluate the electrostatic energy of a protein in solution is developed and combined with the BPMC procedure. The method is based on the modified spherical image charge approximation, efficiently projected onto a molecule of arbitrary shape. Comparison with the finite-difference solutions of the Poisson-Boltzmann equation shows high accuracy for our approach. The BPMC procedure is applied successfully to the structure prediction of 12- and 16-residue synthetic peptides and the determination of protein structure from NMR data, with the immunoglobulin binding domain of streptococcal protein G as an example. The BPMC runs display much better convergence properties than the non-biased simulations. The advantage of a true global optimization procedure for NMR structure determination is its ability to cope with local minima originating from data errors and ambiguities in NMR data.

Optimal protocol and trajectory visualization for conformational searches of peptides and proteins.

Conformational searches by molecular dynamics and different types of Monte Carlo or build-up methods usually aim to find the lowest-energy conformation. However, this is often misleading, as the energy functions used in conformational calculations are imprecise. For instance, though positions of local minima defined by the repulsive part of the Lennard-Jones potential are usually altered only slightly by functional modification, the relative depths of the minima could change significantly. Thus, the purpose of conformational searches and, correspondingly, performance criteria should be reformulated and appropriate methods found to extract different local minima from the search trajectory and allow visualization in the search space. Attempts at convergence to the lowest-energy structure should be replaced with efforts to visit a maximum number of different local energy minima with energies within a certain range. We use this quantitative criterion consistently to evaluate performances of different search procedures. To utilize information generated in the course of simulation, a "stack" of low energy conformations is created and stored. It keeps track of variables and visit numbers for the best representatives of different conformational families. To visualize the search, projection of multidimensional walks onto a principal plane defined by a set of reference structures is used. With Met-enkephalin as a structural example and a Monte Carlo procedure combined with energy minimization (MCM) as a basic search method, we analyzed the influence on search efficiency of different characteristics as temperature schedules, the step size for variable modification, constrained random step and response mechanisms to search difficulties. Simulated annealing MCM had comparable efficiency with MCM at constant and elevated temperature (about 600 K). Constraining the randomized choice of side-chain chi angles to optimal values (rotamers) on every MCM step did not improve, but rather worsened, the search efficiency. Two low-energy Met-enkephalin conformations with parallel Tyr1 and Phe4 rings, a gamma-turn around the Gly2 residue, and Phe4 and Met5 side-chains forming together a compact hydrophobic cluster were found and are suggested as possible structural candidates for interaction with a receptor or a membrane.

Contact area difference (CAD): a robust measure to evaluate accuracy of protein models.

A simple unified measure to evaluate the accuracy of three-dimensional atomic protein models is proposed. This measure is a normalized sum of absolute differences of residue-residue contact surface areas calculated for a reference structure and a model. It employs more rigorous quantitative evaluation of a contact than previously used contact measures. We argue that the contact area difference (CAD) number is a robust single measure to evaluate protein structure predictions in a wide range of model accuracies, from ab initio and threading models to models by homology, since it reflects both backbone topology and side-chain packing, is smooth, continuous and threshold-free, is not sensitive to typical crystallographic errors and ambiguities, adequately penalizes domain and/or secondary structure rearrangements and protein plasticity, and has consistent linear and matrix representations for more detailed analysis. The CAD quality of crystallographic structures, NMR structures, models by homology, and unfolded and misfolded structures is evaluated. It is shown that the CAD number discriminates between models better than Cartesian root-mean-square deviation (cRMSD). Structural variability of the NMR structures was found to be three times larger than deformations of crystallographic structures in different packing environments.

Do aligned sequences share the same fold?

Sequence comparison remains a powerful tool to assess the structural relatedness of two proteins. To develop a sensitive sequence-based procedure for fold recognition, we performed an exhaustive global alignment (with zero end gap penalties) between sequences of protein domains with known three-dimensional folds. The subset of 1.3 million alignments between sequences of structurally unrelated domains was used to derive a set of analytical functions that represent the probability of structural significance for any sequence alignment at a given sequence identity, sequence similarity and alignment score. Analysis of overlap between structurally significant and insignificant alignments shows that sequence identity and sequence similarity measures are poor indicators of structural relatedness in the "twilight zone", while the alignment score allows much better discrimination between alignments of structurally related and unrelated sequences for a wide variety of alignment settings. A fold recognition benchmark was used to compare eight different substitution matrices with eight sets of gap penalties. The best performing matrices were Gonnet and Blosum50 with normalized gap penalties of 2.4/0.15 and 2.0/0.15, respectively, while the positive matrices were the worst performers. The derived functions and parameters can be used for fold recognition via a multilink chain of probability weighted pairwise sequence alignments.

A method to configure protein side-chains from the main-chain trace in homology modelling.

Protein homology modelling typically involves the prediction of side-chain conformations in the modelled protein while assuming a main-chain trace taken from a known tertiary structure of a protein with homologous sequence. It is generally believed that the need to examine all possible combinations of side-chain conformations poses the major obstacle to accurate homology modelling. Methods proposed heretofore use only discrete or limited searches of the side-chain torsion angle space to mitigate the combinatorial problem and also rely on simplified energy functions for calculational speed. The configurational constraints are typically based upon use of frequently observed torsion angles, fixed steps in torsion angles, or oligopeptide segments taken from tertiary structural databanks that are similar in sequence and conformation with the target structure. In the present work, a more fundamental approach is explored for several protein structures and it is demonstrated that the combinatorial barrier in side-chain placement hardly exists. Each side-group can be configured individually in the environment of only the backbone atoms using a systematic search procedure combined with extensive local energy minimization. Tests, using the main-chain or both the main-chain and remaining side-chain atoms to calculate low energy geometries for each residue, established the dominance of the main-chain contribution. The final structure is achieved by combining the individually placed side-chains followed by a full energy refinement of the structure. The prediction accuracy of the present homology modelling technique was assessed relative to other automated procedures and was found to yield improved predictions relative to the known side-chain conformations determined by X-ray crystallography.

The hydration of globular proteins as derived from volume and compressibility measurements: cross correlating thermodynamic and structural data.

We report the first thermodynamic characterization of protein hydration that does not depend on model compound data but rather is based exclusively on macroscopic (volumetric) and microscopic (X-ray) measurements on protein molecules themselves. By combining these macroscopic and microscopic characterizations, we describe a quantitative model that allows one for the first time to predict the partial specific volumes, v(zero), and the partial specific adiabatic compressibilities, ks(zero), of globular proteins from the crystallographic coordinates of the constituent atoms, without using data derived from studies on low-molecular-mass model compounds. Specifically, we have used acoustic and densimetric techniques to determine v(zero) and ks(zero) for 15 globular proteins over a temperature range from 18 to 55 degrees C. For the subset of the 12 proteins with known three-dimensional structures, we calculated the molecular volumes as well as the solvent-accessible surface areas of the constituent charged, polar and nonpolar atomic groups. By combining these measured and calculated properties and applying linear regression analysis, we determined, as a function of temperature, the average hydration contributions to v(zero) and ks(zero) of 1 A2 of the charged, polar, and nonpolar solvent-accessible protein surfaces. We compared these results with those derived from studies on low-molecular-mass compounds to assess the validity of existing models of protein hydration based on small molecule data. This comparison revealed the following features: the hydration contributions to v(zero) and ks(zero) of charged protein surface groups are similar to those of charged groups in small organic molecules. By contrast, the hydration contributions to v(zero) and ks(zero) of polar protein surface groups are qualitatively different from those of polar groups in low-molecular-mass compounds. We suggest that this disparity may reflect the presence of networks of water molecules adjacent to polar protein surface areas, with these networks involving waters from second and third coordination spheres. For nonpolar protein surface groups, we find the ability of low-molecular-mass compounds to model successfully protein properties depends on the temperature domain being examined. Specifically, at room temperatures and below, the hydration contribution to ks(zero) of protein nonpolar surface atomic groups is close to that of nonpolar groups in small organic molecules. By contrast, at higher temperatures, the hydration contribution to ks(zero) of protein nonpolar surface groups becomes more negative than that of nonpolar groups in small organic molecules. We suggest that this behaviour may reflect nonpolar groups on protein surfaces being hydrated independently at low temperatures, while at higher temperatures some of the solvating waters become influenced by neighboring polar groups. We discuss the implications of our aggregate results in terms of various approaches currently being used to describe the hydration properties of globular proteins, particularly focusing on the limitations of existing additive models based on small molecule data.

Aminoglycoside binding in the major groove of duplex RNA: the thermodynamic and electrostatic forces that govern recognition.

We use a combination of spectroscopic, calorimetric, viscometric and computer modeling techniques to characterize the binding of the aminoglycoside antibiotic, tobramycin, to the polymeric RNA duplex, poly(rI).poly(rC), which exhibits the characteristic A-type conformation that is conserved among natural and synthetic double-helical RNA sequences. Our results reveal the following significant features: (i) CD-detected binding of tobramycin to poly(rI).poly(rC) reveals an apparent site size of four base-pairs per bound drug molecule; (ii) tobramycin binding enhances the thermal stability of the host poly(rI).poly(rC) duplex, the extent of which decreases upon increasing in Na(+) concentration and/or pH conditions; (iii) the enthalpy of tobramycin- poly(rI).poly(rC) complexation increases with increasing pH conditions, an observation consistent with binding-induced protonation of one or more drug amino groups; (iv) the affinity of tobramycin for poly(rI).poly(rC) is sensitive to both pH and Na(+) concentration, with increases in pH and/or Na(+) concentration resulting in a concomitant reduction in binding affinity. The salt dependence of the tobramycin binding affinity reveals that the drug binds to the host RNA duplex as trication. (v) The thermodynamic driving force for tobramycin- poly(rI).poly(rC) complexation depends on pH conditions. Specifically, at pH< or =6.0, tobramycin binding is entropy driven, but is enthalpy driven at pH > 6.0. (vi) Viscometric data reveal non-intercalative binding properties when tobramycin complexes with poly(rI).poly(rC), consistent with a major groove-directed mode of binding. These data also are consistent with a binding-induced reduction in the apparent molecular length of the host RNA duplex. (vii) Computer modeling studies reveal a tobramycin-poly(rI). poly(rC) complex in which the drug fits snugly at the base of the RNA major groove and is stabilized, at least in part, by an array of hydrogen bonding interactions with both base and backbone atoms of the host RNA. These studies also demonstrate an inability of tobramycin to form a stable low-energy complex with the minor groove of the poly(rI).poly(rC) duplex. In the aggregate, our results suggest that tobramycin-RNA recognition is dictated and controlled by a broad range of factors that include electrostatic interactions, hydrogen bonding interactions, drug protonation reactions, and binding-induced alterations in the structure of the host RNA. These modulatory effects on tobramycin-RNA complexation are discussed in terms of their potential importance for the selective recognition of specific RNA structural motifs, such as asymmetric internal loops or hairpin loop-stem junctions, by aminoglycoside antibiotics and their derivatives.

The 1.8 A crystal structure of the dimeric peroxisomal 3-ketoacyl-CoA thiolase of Saccharomyces cerevisiae: implications for substrate binding and reaction mechanism.

The dimeric, peroxisomal 3-ketoacyl-CoA thiolase catalyses the conversion of 3-ketoacyl-CoA into acyl-CoA, which is shorter by two carbon atoms. This reaction is the last step of the beta-oxidation pathway. The crystal structure of unliganded peroxisomal thiolase of the yeast Saccharomyces cerevisiae has been refined at 1.8 A resolution. An unusual feature of this structure is the presence of two helices, completely buried in the dimer and sandwiched between two beta-sheets. The analysis of the structure shows that the sequences of these helices are not hydrophobic, but generate two amphipathic helices. The helix in the N-terminal domain exposes the polar side-chains to a cavity at the dimer interface, filled with structured water molecules. The central helix in the C-terminal domain exposes its polar residues to an interior polar pocket. The refined structure has also been used to predict the mode of binding of the substrate molecule acetoacetyl-CoA, as well as the reaction mechanism. From previous studies it is known that Cys125, His375 and Cys403 are important catalytic residues. In the proposed model the acetoacetyl group fits near the two catalytic cysteine residues, such that the oxygen atoms point towards the protein interior. The distance between SG(Cys125) and C3(acetoacetyl-CoA) is 3.7 A. The O2 atom of the docked acetoacetyl group makes a hydrogen bond to N(Gly405), which would favour the formation of the covalent bond between SG(Cys125) and C3(acetoacetyl-CoA) of the intermediate complex of the two-step reaction. The CoA moiety is proposed to bind in a groove on the surface of the protein molecule. Most of the interactions of the CoA molecule are with atoms of the loop domain. The three phosphate groups of the CoA moiety are predicted to interact with side-chains of lysine and arginine residues, which are conserved in the dimeric thiolases.

Evaluating the energetics of empty cavities and internal mutations in proteins.

The energetics of cavity formation in proteins is evaluated with two different approaches and results are analyzed and compared to experimental data. In the first approach, free energy of cavity formation is extracted by RMS fitting from the distribution of numbers of cavities, N, with different volumes, Vcav, in 80 high-resolution protein structures. It is assumed that the distribution of number of cavities according to their volume follows the Boltzmann law, N(Vcav) = exp [(-a.Vcav-b)/kT], or its simplified form. Specific energy cost of cavity formation, a, extracted by RMS fitting from these distributions is compared to a values extracted from experimental free energies of cavity formation in T4 lysozyme fitted to similar expressions. It is found that fitting of both sets of data leads to similar magnitudes and uncertainties in the calculated free energy values. It is shown that Boltzmann-like distribution of cavities can be derived for a simple model of an equilibrium interconversion between mutants in an extracellular system. We, however, suggest that a partitioning into cavity-dependent and cavity-independent terms may lose meaning when one attempts to describe mutation effects on protein stability in terms of specific free energy contributions. As an alternative approach, a direct molecular mechanics evaluation is attempted of T4 lysozyme destabilization by five single cavity-creating mutations. The calculations are based on the approach used in calculations of the energetics of packing defects in crystals. For all mutations calculated destabilizations agree with the corresponding experimental values within +/-0.6 kcal/mol. A computational relaxation of the mutant was most difficult to achieve for the mutation producing the smallest cavity. However, calculations do not always reproduce crystallographically observed contraction/expansion of cavities. It is suggested that this may be related to usually observed large RMS differences (> 1 A) between crystallographic and energy-minimized protein structures, and thus correct energetics might be easier to calculate than the correct geometry.