Redefinition of the cleavage sites of DNase I on the nucleosome core particle

DNase I has been widely used for the footprinting of DNA-protein interactions including analyses of nucleosome core particle (NCP) structure. Our understanding of the relationship between the footprint and the structure of the nucleosome complex comes mainly from digestion studies of NCPs, since they have a well-defined quasi-symmetrical structure and have been widely investigated. However, several recent results suggest that the established consensus of opinion regarding the mode of digestion of NCPs by DNase I may be based on erroneous interpretation of results concerning the relationship between the NCP ends and the dyad axis. Here, we have used reconstituted NCPs with defined ends, bulk NCPs prepared with micrococcal nuclease and molecular modelling to reassess the mode of DNase I digestion. Our results indicate that DNase I cuts the two strands of the nucleosomal DNA independently with an average stagger of 4 nt with the 3'-ends protruding. The previously accepted value of 2 nt stagger is explained by the finding that micrococcal nuclease produces NCPs not with flush ends, but with approximately 1 nt 5'-recessed ends. Furthermore we explain why the DNA stagger is an even and not an odd number of nucleotides. These results are important for studies using DNase I to probe nucleosome structure in complex with other proteins or any DNA-protein complex containing B-form DNA. We also determine the origin of the 10n +/- 5 nt periodicity found in the internucleosomal ladder of DNase I digests of chromatin from various species. The explanation of the 10n +/- 5 nt ladder may have implications for the structure of the 30 nm fibre.     

Analysis of ordered and disordered protein complexes reveals structural features discriminating between stable and unstable monomers

Most proteins exist in the cell as multi-component assemblies. However, which proteins need to be present simultaneously in order to perform a given function is frequently unknown. The first step toward this goal would be to predict proteins that can function only when in a complexed form. Here, we propose a scheme to distinguish whether the protein components are ordered (stable) or disordered when separated from their complexed partners. We analyze structural characteristics of several types of complexes, such as natively unstructured proteins, ribosomal proteins, two-state and three-state complexes, and crystal-packing dimers. Our analysis makes use of the fact that natively unstructured proteins, which undergo a disorder-to-order transition upon binding their partner, and stable monomeric proteins, which exist as dimers only in their crystal form, provide examples of two vastly different scenarios. We find that ordered monomers can be distinguished from disordered monomers on the basis of the per-residue surface and interface areas, which are significantly smaller for ordered proteins. With this scale, two-state dimers (where the monomers unfold upon dimer separation) and ribosomal proteins are shown to resemble disordered proteins. On the other hand, crystal-packing dimers, whose monomers are stable in solution, fall into the ordered protein category. While there should be a continuum in the distributions, nevertheless, the per-residue scale measures the confidence in the determination of whether a protein can exist as a stable monomer. Further analysis, focusing on the chemical and contact preferences at the interface, interior and exposed surface areas, reveals that disordered proteins lack a strong hydrophobic core and are composed of highly polar surface area. We discuss the implication of our results for de novo design of stable monomeric proteins and peptides.     

Recognition of 5'-YpG-3' sequences by coupled stacking/hydrogen bonding interactions with amino acid residues

The combined biochemical and structural study of hundreds of protein-DNA complexes has indicated that sequence-specific interactions are mediated by two mechanisms termed direct and indirect readout. Direct readout involves direct interactions between the protein and base-specific atoms exposed in the major and minor grooves of DNA. For indirect readout, the protein recognizes DNA by sensing conformational variations in the structure dependent on nucleotide sequence, typically through interactions with the phosphodiester backbone. Based on our recent structure of Ndt80 bound to DNA in conjunction with a search of the existing PDB database, we propose a new method of sequence-specific recognition that utilizes both direct and indirect readout. In this mode, a single amino acid side-chain recognizes two consecutive base-pairs. The 3'-base is recognized by canonical direct readout, while the 5'-base is recognized through a variation of indirect readout, whereby the conformational flexibility of the particular dinucleotide step, namely a 5'-pyrimidine-purine-3' step, facilitates its recognition by the amino acid via cation-pi interactions. In most cases, this mode of DNA recognition helps explain the sequence specificity of the protein for its target DNA.     

Thermodynamic dissection of the binding energetics of proline-rich peptides to the Abl-SH3 domain: implications for rational ligand design

The inhibition of the interactions between SH3 domains and their targets is emerging as a promising therapeutic strategy. To date, rational design of potent ligands for these domains has been hindered by the lack of understanding of the origins of the binding energy. We present here a complete thermodynamic analysis of the binding energetics of the p41 proline-rich decapeptide (APSYSPPPPP) to the SH3 domain of the c-Abl oncogene. Isothermal titration calorimetry experiments have revealed a thermodynamic signature for this interaction (very favourable enthalpic contributions opposed by an unfavourable binding entropy) inconsistent with the highly hydrophobic nature of the p41 ligand and the Abl-SH3 binding site. Our structural and thermodynamic analyses have led us to the conclusion, having once ruled out any possible ionization events or conformational changes coupled to the association, that the establishment of a complex hydrogen-bond network mediated by water molecules buried at the binding interface is responsible for the observed thermodynamic behaviour. The origin of the binding energetics for proline-rich ligands to the Abl-SH3 domain is further investigated by a comparative calorimetric analysis of a set of p41-related ligands. The striking effects upon the enthalpic and entropic contributions provoked by conservative substitutions at solvent-exposed positions in the ligand confirm the complexity of the interaction. The implications of these results for rational ligand design are discussed.     

Molecular determinants for ATP-binding in proteins: a data mining and quantum chemical analysis

Adenosine 5'-triphosphate (ATP) plays an essential role in all forms of life. Molecular recognition of ATP in proteins is a subject of great importance for understanding enzymatic mechanism and for drug design. We have carried out a large-scale data mining of the Protein Data Bank (PDB) to analyze molecular determinants for recognition of the adenine moiety of ATP by proteins. Non-bonded intermolecular interactions (hydrogen bonding, pi-pi stacking interactions, and cation-pi interactions) between adenine base and surrounding residues in its binding pockets are systematically analyzed for 68 non-redundant, high-resolution crystal structures of adenylate-binding proteins. In addition to confirming the importance of the widely known hydrogen bonding, we found out that cation-pi interactions between adenine base and positively charged residues (Lys and Arg) and pi-pi stacking interactions between adenine base and surrounding aromatic residues (Phe, Tyr, Trp) are also crucial for adenine binding in proteins. On average, there exist 2.7 hydrogen bonding interactions, 1.0 pi-pi stacking interactions, and 0.8 cation-pi interactions in each adenylate-binding protein complex. Furthermore, a high-level quantum chemical analysis was performed to analyze contributions of each of the three forms of intermolecular interactions (i.e. hydrogen bonding, pi-pi stacking interactions, and cation-pi interactions) to the overall binding force of the adenine moiety of ATP in proteins. Intermolecular interaction energies for representative configurations of intermolecular complexes were analyzed using the supermolecular approach at the MP2/6-311 + G* level, which resulted in substantial interaction strengths for all the three forms of intermolecular interactions. This work represents a timely undertaking at a historical moment when a large number of X-ray crystallographic structures of proteins with bound ATP ligands have become available, and when high-level quantum chemical analysis of intermolecular interactions of large biomolecular systems becomes computationally feasible. The establishment of the molecular basis for recognition of the adenine moiety of ATP in proteins will directly impact molecular design of ATP-binding site targeted enzyme inhibitors such as kinase inhibitors.     

Rapid analysis of large protein-protein complexes using NMR-derived orientational constraints: the 95 kDa complex of LpxA with acyl carrier protein

Characterization of protein-protein interactions that are critical to the specific function of many biological systems has become a primary goal of structural biology research. Analysis of these interactions by structural techniques is, however, challenging due to inherent limitations of the techniques and because many of the interactions are transient, and suitable complexes are difficult to isolate. In particular, structural studies of large protein complexes by traditional solution NMR methods are difficult due to a priori requirement of extensive assignments and a large number of intermolecular restraints for the complex. An approach overcoming some of these challenges by utilizing orientational restraints from residual dipolar couplings collected on solution NMR samples is presented. The approach exploits existing structures of individual components, including the symmetry properties of some of these structures, to assemble rapidly models for relatively large protein-protein complexes. An application is illustrated with a 95 kDa homotrimeric complex of the acyltransferase protein, LpxA (UDP-N-acetylglucosamine acyltransferase), and acyl carrier protein. LpxA catalyzes the first step in the biosynthesis of the lipid A component of lipopolysaccharide in Gram-negative bacteria. The structural model generated for this complex can be useful in the design of new anti-bacterial agents that inhibit the biosynthesis of lipid A.     

A simple physical model for the prediction and design of protein-DNA interactions

Protein-DNA interactions are crucial for many biological processes. Attempts to model these interactions have generally taken the form of amino acid-base recognition codes or purely sequence-based profile methods, which depend on the availability of extensive sequence and structural information for specific structural families, neglect side-chain conformational variability, and lack generality beyond the structural family used to train the model. Here, we take advantage of recent advances in rotamer-based protein design and the large number of structurally characterized protein-DNA complexes to develop and parameterize a simple physical model for protein-DNA interactions. The model shows considerable promise for redesigning amino acids at protein-DNA interfaces, as design calculations recover the amino acid residue identities and conformations at these interfaces with accuracies comparable to sequence recovery in globular proteins. The model shows promise also for predicting DNA-binding specificity for fixed protein sequences: native DNA sequences are selected correctly from pools of competing DNA substrates; however, incorporation of backbone movement will likely be required to improve performance in homology modeling applications. Interestingly, optimization of zinc finger protein amino acid sequences for high-affinity binding to specific DNA sequences results in proteins with little or no predicted specificity, suggesting that naturally occurring DNA-binding proteins are optimized for specificity rather than affinity. When combined with algorithms that optimize specificity directly, the simple computational model developed here should be useful for the engineering of proteins with novel DNA-binding specificities.     

Structural basis for the carbohydrate specificities of artocarpin: variation in the length of a loop as a strategy for generating ligand specificity

Artocarpin, a tetrameric lectin of molecular mass 65 kDa, is one of the two lectins extracted from the seeds of jackfruit. The structures of the complexes of artocarpin with mannotriose and mannopentose reported here, together with the structures of artocarpin and its complex with Me-alpha-mannose reported earlier, show that the lectin possesses a deep-seated binding site formed by three loops. The binding site can be considered as composed of two subsites; the primary site and the secondary site. Interactions at the primary site composed of two of the loops involve mainly hydrogen bonds, while those at the secondary site comprising the third loop are primarily van der Waals in nature. Mannotriose in its complex with the lectin interacts through all the three mannopyranosyl residues; mannopentose interacts with the protein using at least three of the five mannose residues. The complexes provide a structural explanation for the carbohydrate specificities of artocarpin. A detailed comparison with the sugar complexes of heltuba, the only other mannose-specific jacalin-like lectin with known three-dimensional structure in sugar-bound form, establishes the role of the sugar-binding loop constituting the secondary site, in conferring different specificities at the oligosaccharide level. This loop is four residues longer in artocarpin than in heltuba, providing an instance where variation in loop length is used as a strategy for generating carbohydrate specificity.     

Binding sub-site dissection of a carbohydrate-binding module reveals the contribution of entropy to oligosaccharide recognition at "non-primary" binding subsites

The optimal ligands for many carbohydrate-binding proteins are often oligosaccharides comprising two, three, or more monosaccharide units. The binding affinity for these sugars is increased incrementally by contributions from binding subsites on the protein that accommodate the individual monosaccharide residues of the oligosaccharide. Here, we use CsCBM6-1, a xylan-specific type B carbohydrate-binding module (CBM) from Clostridium stercorarium falling into amino acid sequence family CBM6, as a model system to investigate the structural and thermodynamic contributions of binding subsites in this protein to carbohydrate recognition. The three-dimensional structures of uncomplexed CsCBM6-1 (at 1.8 A resolution) and bound to the oligosaccharides xylobiose, xylotriose, and xylotetraose (at 1.70 A, 1.89 A, and 1.69 A resolution, respectively) revealed the sequential occupation of four subsites within the binding site in the order of subsites 2, 3, 4 then 1. Overall, binding to all of the xylooligosaccharides tested was enthalpically favourable and entropically unfavourable, like most protein-carbohydrate interactions, with the primary subsites 2 and 3 providing the bulk of the free energy and enthalpy of binding. In contrast, the contributions to the changes in entropy of the non-primary subsites 1 and 4 to xylotriose and xylotetraose binding, respectively, were positive. This observation is remarkable, in that it shows that the 10-20-fold improvement in association constants for oligosaccharides longer than a disaccharide is facilitated by favourable entropic contributions from the non-primary binding subsites.