Discovering novel ligands for macromolecules using X-ray crystallographic screening

The need to decrease the time scale for clinical compound discovery has led to innovations at several stages in the process, including genomics/proteomics for target identification, ultrahigh-throughput screening for lead identification, and structure-based drug design and combinatorial chemistry for lead optimization. A critical juncture in the process is the identification of a proper lead compound, because a poor choice may generate costly difficulties at later stages. Lead compounds are commonly identified from high-throughput screens of large compound libraries, derived from known substrates/inhibitors, or identified in computational prescreeusing X-ray crystal structures. Structural information is often consulted to efficiently optimize leads, but under the current paradigm, such data require preidentification and confirmation of compound binding. Here, we describe a new X-ray crystallography-driven screening technique that combines the steps of lead identification, structural assessment, and optimization. The method is rapid, efficient, and high-throughput, and it results in detailed crystallographic structure information. The utility of the method is demonstrated in the discovery and optimization of a new orally available class of urokinase inhibitors for the treatment of cancer.     

Determining the structural basis for specificity of ligands using crystallographic screening

Crystallographic screening has been used to identify new inhibitors for potential target for drug development. Here, we describe the application of the crystallographic screening to assess the structural basis of specificity of ligands against a protein target. The method is efficient and results in detailed crystallographic information. The utility of the method is demonstrated in the study of the structural basis for specificity of ligands for human purine nucleoside phosphorylase (PNP). Purine nucleoside phosphorylase catalyzes the phosphorolysis of the N-ribosidic bonds of purine nucleotides and deoxynucleosides. This enzyme is a target for inhibitor development aiming at T-cell immune response modulation and has been submitted to extensive structure-based drug design. This methodology may help in the future development of a new generation of PNP inhibitors.     

T-state hemoglobin with four ligands bound

Flash photolysis kinetics have been measured for ligand recombination to hemoglobin (Hb) in the presence of two effectors: bezafibrate (Bzf) and inositol hexakisphosphate (IHP). The combined influence of the two independent effectors leads to predominantly T-state behavior. Samples equilibrated with 0.1 atm of CO are fully saturated, yet after photodissociation they show only T-state bimolecular recombination rates at all photolysis levels; this indicates that the allosteric transition from R to T occurs before CO rebinding and that the allosteric equilibrium favors the T-state tetramer with up to three ligands bound. Since all four ligands bind at the rate characteristic for the T-state, the return transition from T to R must occur after the fourth ligand was bound. At 1 atm of CO, rebinding to the initial R state competes with the allosteric transition resulting in a certain fraction of CO bound at the rate characteristic for the R state; this fraction is greater the smaller the percentage dissociation. Under 1 atm of oxygen, samples are not more than 93% saturated and show mainly T-state kinetics. The results show that all four hemes can bind oxygen or CO ligands in the T structure. The fraction of the kinetics occurring as geminate is less for partially liganded (T-state) samples than for fully liganded (R-state) Hb.     

Distribution of unselectively bound ligands along DNA

Unselective and reversible adsorption of ligands on DNA for a model of binding proposed by Zasedatelev, Gursky, and Volkenshtein is considered. In this model, the interaction between neighboring ligands located at the distance of i binding centers is characterized by the statistical weight a(i). Each ligand covers L binding centers. For this model, expressions for binding averages are represented in a new simple form. This representation is convenient for the calculation of the fraction of inter-ligand distances of i binding centers f(d)(i) and the fraction of binding centers included in the distances of i binding centers f(bc)(i) for various types of interaction between bound ligands. It is shown that, for non-cooperative binding, contact cooperativity and long-range cooperativity, the fraction of the zero inter-ligand distance f(d)(0) is maximal at any relative concentration of bound ligands (r). Calculations demonstrate that, at low r, f(d)(0) approximately r . a(o), and f(d)(i) approximately r at 1 1/r-L, then f(d)(i) rapidly decreases with i at any r for all types of inter-ligand interaction. At high ligand concentration (r is close to r(max) = L(-1)), f(d)(0) is close to unity and f(d)(i) rapidly decreases with i for any type of inter-ligand interaction. For strong contact cooperativity, f(d)(0) is close to unity in a much lager r interval ((0.5-1) . r(max)), and f(d)(1) approximately a(o)(-1) at r approximately 0.5 . r(max). In the case of long-range interaction between bound ligands, the dependence f(d)(i) is more complex and has a maximum at i approximately (1/r-L)(1/2) for anti-cooperative binding. f(bc)(i) is maximal at i approximately 1/r-L for all types of binding except the contact cooperativity. A strong asymmetry in the influence of contact cooperativity and anticooperativity on the ligand distribution along DNA is demonstrated.     

A hybrid representation approach for modelling complex dynamic bioprocesses

This paper considers the use of hybrid models to represent the dynamic behaviour of biotechnological processes. Each hybrid model consists of a set of non linear differential equations and a neural model. The set of differential equations attempts to describe as much as possible the phenomenology of the process whereas neural networks model predict some key parameters that are an essential part of the phenomenological model. The neural model is obtained indirectly, that is, using the prediction errors of one or more state variables to adjust its weights instead of successive presentations of input-output data of the neural network. This approach allows to use actual measurements to derive a suitable neural model that not only represents the variation of some key parameters but it is also able to partly include dynamic behaviour unaccounted for by the phenomenological model. The approach is described in detail using three test cases: (1) the fermentation of glucose to gluconic acid by the micro-organism Pseudomonas ovalis, (2) the growth of filamentous fungi in a solid state fermenter, and (3) the propagation of filamentous fungi growing on a 2-D solid substrate. Results for the three applications clearly demon- strate that using a hybrid model is a viable alternative for modelling complex biotechnological bioprocesses.     

Conformational diversity of ligands bound to proteins

The phenomenon of molecular recognition, which underpins almost all biological processes, is dynamic, complex and subtle. Establishing an interaction between a pair of molecules involves mutual structural rearrangements guided by a highly convoluted energy landscape, the accurate mapping of which continues to elude us. Increased understanding of the degree to which the conformational space of a ligand is restricted upon binding may have important implications for docking studies, structure refinement and for function prediction methods based on geometrical comparisons of ligands or their binding sites. Here, we present an analysis of the conformational variability exhibited by three of the most ubiquitous biological ligands in nature, ATP, NAD and FAD. First, we demonstrate qualitatively that these ligands bind to proteins in widely varying conformations, including several cases in which parts of the molecule assume energetically unfavourable orientations. Next, by comparing the distribution of bound ligand shapes with the set of all possible molecular conformations, we provide a quantitative assessment of previous observations that ligands tend to unfold when binding to proteins. We show that, while extended forms of ligands are indeed common in ligand-protein structures, instances of ligands in almost maximally compact arrangements can also be found. Thirdly, we compare the conformational variation in two sets of ligand molecules, those bound to homologous proteins, and those bound to unrelated proteins. Although most superfamilies bind ligands in a fairly conserved manner, we find several cases in which significant variation in ligand configuration is observed.     

Affinity surfactants as reversibly bound ligands for high-performance affinity chromatography

Pyridine was coupled covalently to a nonionic ethoxylated alcohol: octaethylene glycol n-hexadecyl ether. This modified surfactant was found to be a reversible, competitive inhibitor of horse serum cholinesterase. The surfactant bound irreversibly, in aqueous media, to octadecyl-bounded reverse phase silica particles commonly used for high-performance liquid chromatography. The amount of ligand bound was found to be 550 mumol/ml of packing, a concentration that is over 100 times higher than what can be normally bound to agarose affinity chromatography supports. With this packing, a 280-fold purification of cholinesterase from horse serum and a 79-fold purification of human serum cholinesterase were accomplished, with yields greater than 80%, using a 2-cm-long column and a 7-min elution time. The affinity surfactant could be eluted from the column using a 6:4 (v/v) mixture of methanol and isopropanol. This technique should be generally applicable in the development of biospecific supports for high-performance affinity chromatography.     

The movement of bound ligands over cell surfaces

The long range movements of membrane-bound ligands into surface caps and into the pseudopods of phagocytizing cells, the uropods of motile cells and the cleavage furrow of dividing cells appear to be analogous processes. A common mechanism to explain these movements must take into account several new and central observations: ligand-receptor complexes can migrate to regions of existing microfilament accumulation; laser photobleaching studies with fluorescent Con A indicate that ligand-receptor movement occurs unidirectionally; video computer analyses of Con A redistribution show that movement may exceed the maximum rates measured for protein diffusion in membranes. These observations are not consistent with models in which ligand-receptor movement occurs by diffusion or by direct interaction with contractile microfilaments. However, they can be satisfied by a new model that proposes the entrainment of selected membrane determinants on membrane waves directed towards regions such as caps, pseudopodia, uropods or cleavage furrow. These oriented waves are initiated by tension due to asymmetric microfilamentmembrane interaction.     

Ligand Depot: a data warehouse for ligands bound to macromolecules

Ligand Depot is an integrated data resource for finding information about small molecules bound to proteins and nucleic acids. The initial release (version 1.0, November, 2003) focuses on providing chemical and structural information for small molecules found as part of the structures deposited in the Protein Data Bank. Ligand Depot accepts keyword-based queries and also provides a graphical interface for performing chemical substructure searches. A wide variety of web resources that contain information on small molecules may also be accessed through Ligand Depot. AVAILABILITY: Ligand Depot is available at http://ligand-depot.rutgers.edu/. Version 1.0 supports multiple operating systems including Windows, Unix, Linux and the Macintosh operating system. The current drawing tool works in Internet Explorer, Netscape and Mozilla on Windows, Unix and Linux.     

A simple method for quantitating ligands covalently bound to agarose beads

Agarose derivatives, as used for affinity chromatography, can be dissolved by warming in aqueous media at suitable pH values. Dilute solutions so formed are stable and transparent in regions of the ultraviolet and visible range, depending on the method of solubilization. Covalently bound ligands which possess chromophoric groups, or functional groups which can be converted to chromophores, can be quantitated by direct spectral analysis provided a solubilizing medium is chosen which results in minimal interference by absorbing decomposition products of the matrix.     

NFkappaB selectivity of estrogen receptor ligands revealed by comparative crystallographic analyses

Our understanding of how steroid hormones regulate physiological functions has been significantly advanced by structural biology approaches. However, progress has been hampered by misfolding of the ligand binding domains in heterologous expression systems and by conformational flexibility that interferes with crystallization. Here, we show that protein folding problems that are common to steroid hormone receptors are circumvented by mutations that stabilize well-characterized conformations of the receptor. We use this approach to present the structure of an apo steroid receptor that reveals a ligand-accessible channel allowing soaking of preformed crystals. Furthermore, crystallization of different pharmacological classes of compounds allowed us to define the structural basis of NFkappaB-selective signaling through the estrogen receptor, thus revealing a unique conformation of the receptor that allows selective suppression of inflammatory gene expression. The ability to crystallize many receptor-ligand complexes with distinct pharmacophores allows one to define structural features of signaling specificity that would not be apparent in a single structure.     

Orientation of bound ligands in mannose-binding proteins. Implications for multivalent ligand recognition

Mannose-binding proteins (MBPs) are C-type animal lectins that recognize high mannose oligosaccharides on pathogenic cell surfaces. MBPs bind to their carbohydrate ligands by forming a series of Ca(2+) coordination and hydrogen bonds with two hydroxyl groups equivalent to the 3- and 4-OH of mannose. In this work, the determinants of the orientation of sugars bound to rat serum and liver MBPs (MBP-A and MBP-C) have been systematically investigated. The crystal structures of MBP-A soaked with monosaccharides and disaccharides and also the structure of the MBP-A trimer cross-linked by a high mannose asparaginyl oligosaccharide reveal that monosaccharides or alpha1-6-linked mannose bind to MBP-A in one orientation, whereas alpha1-2- or alpha1-3-linked mannose binds in an orientation rotated 180 degrees around a local symmetry axis relating the 3- and 4-OH groups. In contrast, a similar set of ligands all bind to MBP-C in a single orientation. The mutation of MBP-A His(189) to its MBP-C equivalent, valine, causes Man alpha 1-3Man to bind in a mixture of orientations. These data combined with modeling indicate that the residue at this position influences the orientation of bound ligands in MBP. We propose that the control of binding orientation can influence the recognition of multivalent ligands. A lateral association of trimers in the cross-linked crystals may reflect interactions within higher oligomers of MBP-A that are stabilized by multivalent ligands.     

Integrating knowledge representation and quantitative modelling in physiology

A wealth of potentially shareable resources, such as data and models, is being generated through the study of physiology by computational means. Although in principle the resources generated are reusable, in practice, few can currently be shared. A key reason for this disparity stems from the lack of consistent cataloguing and annotation of these resources in a standardised manner. Here, we outline our vision for applying community-based modelling standards in support of an automated integration of models across physiological systems and scales. Two key initiatives, the Physiome Project and the European contribution - the Virtual Phsysiological Human Project, have emerged to support this multiscale model integration, and we focus on the role played by two key components of these frameworks, model encoding and semantic metadata annotation. We present examples of biomedical modelling scenarios (the endocrine effect of atrial natriuretic peptide, and the implications of alcohol and glucose toxicity) to illustrate the role that encoding standards and knowledge representation approaches, such as ontologies, could play in the management, searching and visualisation of physiology models, and thus in providing a rational basis for healthcare decisions and contributing towards realising the goal of of personalized medicine.     

A novel method for the modelling of peptide ligands to their receptors

A knowledge-based approach to the modelling of enzyme-peptide inhibitor complexes is described. Given the structure of an enzyme, and knowledge of its binding site, the method seeks to predict the binding geometry of a peptide ligand. This novel method involves using examples of side-chain packing derived from proteins of known three-dimensional structure to define possible packing arrangements of a peptide inhibitor group to its binding site. A suite of programs, GEMINI, was written and used to predict the packing of pairs of amino acid groups from three inhibitors complexed to their enzymes for which the X-ray structures were available. These included the Phe group of the inhibitor H142 bound to endothiapepsin, the Leu group of CLT complexed to thermolysin and the C-terminus of Gly-L-Tyr bound to carboxypeptidase A. A detailed comparison of the modelled and observed inhibitor coordinates was made. This approach may be extended to modelling other types of protein interactions.     

Physiologically realistic modelling of a mechanism for neural representation of intervals of time

A model for a recurrent network of bistable spiking neurons is examined. Each neuron is described by a leaky-integrate-and-fire formulation with biophysically realistic currents and noise. Specially, neuronal bistability is equipped by after-depolarisation current. Results obtained by computer simulation show that spiking of each neuron starting at an initial time continues for an extended period and then suddenly ceases at around a certain time. We hypothesise that activation of neurons that starts at t = 0 and voluntarily ceases at t = T is a neural underpinning of internal representation of an interval of time T. The above results theoretically support this hypothesis by demonstrating one possible mechanism to generate such time course of neuronal activation.     

Modification of alkoxo ligands of BINOL-Ti ladder: Isolation and X-ray crystallographic analysis

The stable binaphthol-titanium ladder complexes have been successfully prepared by using bulky alkoxo ligands. From the secondary OR ligand (cyclohexyloxo, 2,4-dimethyl-3-pentyloxo or 2-adamantyloxo) and terially OR ligand (tert-butyloxo, 1-adamantyloxo), partial hydrolysis proceeded to give the μ³-oxo titanium complexes. The use of [Ti(BINOLato)(OEt)₂]_n made it possible to prepare the Ti(BINOLato)(OR)₂ complexes using alcohols (ROH) of high boiling point (R = cyclohexyl, 2-adamantyl, 1-adamantyl). X-ray analyses of [(R)-1,1′-bi-2-naphtholato]bis(O-2,4-dimethyl-3-pentyloxo)titanium and [(R)-3,3′-dimethyl-1,1′-bi-2-naphtholato]bis(2-adamantyloxo)titanium showed a good agreement with the estimated ladder complexes. The catalytic activity of BINOL-Ti catalyst analogues, obtained by partial hydrolysis of Ti(BINOLato)(OR)₂ with wet MS 4A was studied in asymmetric glyoxylate-ene reaction by two methods. Moderate to good chemical yields and enantioselectivities were obtained.     

Covariance density estimation for autoregressive spectral modelling of point processes

The use of autoregressive modelling has acquired great importance in time series analysis and in principle it may also be applicable in the spectral analysis of point processes with similar advantages over the nonparametric approach. Most of the methods used for autoregressive spectral analysis require positive semidefinite estimates for the covariance function, while current methods for the estimation of the covariance density function of a point process given a realization over the interval [0,T] do not guarantee a positive semidefinite estimate. This paper discusses methods for the estimation of the covariance density and conditional intensity function of point processes and present alternative computational efficient estimation algorithms leading always to positive semidefinite estimates, therefore adequate for autoregressive spectral analysis. Autoregressive spectral modelling of point processes from Yule-Walker type equations and Levinson recursion combined with the minimum AIC or CAT principle is illustrated with neurobiological data.     

Lysozyme revisited: crystallographic evidence for distortion of an N-acetylmuramic acid residue bound in site D.

A structure of the trisaccharide 2-acetamido-2-deoxy-D-muramic acid-beta (1----4)-2-acetamido-2-deoxy-D-glucose-beta (1----4)-2-acetamido-2-deoxy-D-muramic acid (NAM-NAG-NAM), bound to subsites B, C and D in the active-site cleft of hen egg-white lysozyme has been determined and refined at 1.5 A resolution. The resulting atomic co-ordinates indicate that the NAM residue in site D is distorted from the full 4C1 chair conformation to one in which the ring atoms C-1, C-2, O-5 and C-5 are approximately coplanar, and the hydroxymethyl group is positioned axially (a conformation best described as a sofa). This finding supports the original proposals that suggested the ground-state conformation of the sugar bound in site D is strained to one that more closely resembles the geometry required for the oxocarbonium-ion transition state, the next step along the reaction pathway. Additionally, detailed analysis at 1.5 A resolution of the environments of the catalytic residues Glu35 and Asp52 provides new information on the properties that may allow lysozyme to promote the stabilization of an unusually long-lived oxocarbonium-ion transition state. Intermolecular interactions between the N-acetylmuramic acid residue in site D and the lysozyme molecule that contribute to the saccharide ring distortion include: close packing of the O-3' lactyl group with a hydrogen-bonded "platform" of enzyme residues (Asp52, Asn46, Asn59, Ser50 and Asp48), a close contact between the hydroxymethyl group of ring D and the 2'-acetamido group of ring C and a strong hydrogen-bonded interaction between the NH group of Val109 and O-6 of ring D that stabilizes the observed quasi-axial orientation of the -CH2OH group. Additionally, the structure of this complex shows a strong hydrogen bond between the carboxyl group of Glu35 and the beta-anomeric hydroxyl group of the NAM residue in site D. The hydrogen-bonded environment of Asp52 in the native enzyme and in the complex coupled with the very unfavorable direction of approach of the potential carboxylate nucleophile makes it most unlikely that there is a covalent glycosylenzyme intermediate on the hydrolysis pathway of hen egg-white lysozyme.