Computational methods to estimate drug development parameters.

Computational methods are currently available to estimate oral bioavailability, solubility, metabolism, toxicity, pKa, blood-brain barrier permeability and other ADME and physicochemical parameters. Decisions as to which methods to implement and to employ must be made in accordance with the stated goals of a drug discovery organization, the timeline for these goals, and the budgetary limitations as set forth to accomplish these goals. Certain methods are more attractive to the production environment of a pharmaceutical project team where early ADME and Tox information is sought to aid in drug design decisions and prioritization. Practical limitations of these methods, ease of use, utility of results, as well as their scope and limitations are discussed. Recommendations as to which parameters are best estimated by commercial products, as opposed to those that can be developed in-house, are delineated. Special attention is given to those methods that can be integrated into the current high-throughput paradigms of drug discovery programs. Together, these considerations define a 'zero-infrastructure' approach to provide ADME and Tox information during the early stages of the drug design process.

Simple selection criteria for drug-like chemical matter.

A simple pharmacophore point filter has been developed that discriminates between drug-like and nondrug-like chemical matter. It is based on the observation that nondrugs are often underfunctionalized. Therefore, a minimum count of well-defined pharmacophore points is required to pass the filter. The application of the filter results in 66-69% of subsets of the MDDR database to be classified as drug-like. Furthermore, 61-68% of subsets of the CMC database are classified as drug-like. In contrast, only 36% of the ACD are found to be drug-like. While these results are not quite as good as those obtained with recently described neural net approaches, the method used here has clear advantages. In contrast to a neural net approach and also in contrast to decision tree methods described recently, the pharmacophore filter has been developed by using "chemical wisdom" that is unbiased from fitting the structural content of specific drug databases to prediction models. Similar to decision tree methods, the pharmacophore point filter provides a detailed structural reason for the classification of each molecule as drug or nondrug. The pharmacophore point filter results are compared to neural net filter results. A statistically significant overlap between compounds recognized as drug-like validates both approaches. The pharmacophore point filter complements neural net approaches as well as property profiling approaches used as drug-likeness filters in compound library analysis and design.

"Holistic" in silico methods to estimate the systemic and CNS bioavailabilities of potential chemotherapeutic agents.

A fundamental fact in the drug development process is that physical quantity of chemical substance is needed for experimental determinations. Information that could guide the course of a drug discovery program, in particular the penultimate ADME parameters % oral bioavailability (%F) and CNS permeability (BBB) are not explicitly determined until after large capital, human and time resources have been invested in a particular chemical series to produce the substance. To assure better go/no-go decisions and to protect the risks of a process that necessitates a considerable "front-loading" of resources, project teams are turning to computational methods to estimate these parameters. Herein we provide a detailed review of "holistic" in silico methods toward the estimation of %F and BBB. An unbiased description of the scope and limitations of their installation and application will be given in the context of an on going pharmaceutical project.

Evaluation of docking/scoring approaches: a comparative study based on MMP3 inhibitors.

An increasing number of docking/scoring programs are available that use different sampling and scoring algorithms. A reliable scoring function is the crucial element of such approaches. Comparative studies are needed to evaluate their current capabilities. DOCK4 with force field and PMF scoring as well as FlexX were used to evaluate the predictive power of these docking/scoring approaches to identify the correct binding mode of 61 MMP-3 inhibitors in a crystal structure of stromelysin and also to rank them according to their different binding affinities. It was found that DOCK4/PMF scoring performs significantly better than FlexX and DOCK4/FF in both ranking ligands and predicting their binding modes. Most notably, DOCK4/PMF was the only scoring/docking approach that found a significant correlation between binding affinity and predicted score of the docked inhibitors. However, comparing only those cases where the correct binding mode was identified (scoring highest among sampled poses), FlexX showed the best 'fine tuning' (lowest rmsd) in predicted binding modes. The results suggest that not so much the sampling procedure but rather the scoring function is the crucial element of a docking program.

Simulating proton translocations in proteins: probing proton transfer pathways in the Rhodobacter sphaeroides reaction center.

A general method for simulating proton translocations in proteins and for exploring the role of different proton transfer pathways is developed and examined. The method evaluates the rate constants for proton transfer processes using the energetics of the relevant proton configurations. The energies (DeltaG((m))) of the different protonation states are evaluated in two steps. First, the semimicroscopic version of the protein dipole Langevin dipole (PDLD/S) method is used to evaluate the intrinsic energy of bringing the protons to their protein sites, when the charges of all protein ionized residues are set to zero. Second, the interactions between the charged groups are evaluated by using a Coulomb's Law with an effective dielectric constant. This approach, which was introduced in an earlier study by one of the authors of the current report, allows for a very fast determination of any DeltaG((m)) and for practical evaluation of the time-dependent proton population: That is, the rate constants for proton transfer processes are evaluated by using the corresponding DeltaG((m)) values and a Marcus type relationship. These rate constants are then used to construct a master equation, the integration of which by a fourth-order Runge-Kutta method yields the proton population as a function of time. The integration evaluates, 'on the fly,' the changes of the rate constants as a result of the time-dependent changes in charge-charge interaction, and this feature benefits from the fast determination of DeltaG((m)). The method is demonstrated in a preliminary study of proton translocation processes in the reaction center of Rhodobacter sphaeroides. It is found that proton transfer across water chains involves significant activation barriers and that ionized protein residues probably are involved in the proton transfer pathways. The potential of the present method in analyzing mutation experiments is discussed briefly and illustrated. The present study also examines different views of the nature of proton translocations in proteins. It is shown that such processes are controlled mainly by the electrostatic interaction between the proton site and its surroundings rather than by the local bond rearrangements of water molecules that are involved in the proton pathways. Thus, the overall rate of proton transport frequently is controlled by the highest barrier along the conduction pathway. Proteins 1999;36:484-500.

Evaluation of PMF scoring in docking weak ligands to the FK506 binding protein.

A new knowledge-based scoring function (PMF-score), implemented into the DOCK4 program, was used to screen a database of 3247 small molecules for binding to the FK506 binding protein (FKBP). The computational ranking of these compounds was compared to the binding affinities measured by NMR. It was demonstrated that small, weakly binding molecules have, on average, higher computational scores than nonbinders and are enriched in the upper ranks of the computational scoring lists. In addition, the results obtained with the PMF scoring function were superior (by 30-120% larger enrichment factors) to those obtained with the standard force field score of DOCK4. The reliable ranking of small, weakly binding molecules offers new ways of designing building blocks in combinatorial libraries as well as SAR by NMR libraries with the increased chance of identifying suitable lead compounds for drug design.

A general and fast scoring function for protein-ligand interactions: a simplified potential approach.

A fast, simplified potential-based approach is presented that estimates the protein-ligand binding affinity based on the given 3D structure of a protein-ligand complex. This general, knowledge-based approach exploits structural information of known protein-ligand complexes extracted from the Brookhaven Protein Data Bank and converts it into distance-dependent Helmholtz free interaction energies of protein-ligand atom pairs (potentials of mean force, PMF). The definition of an appropriate reference state and the introduction of a correction term accounting for the volume taken by the ligand were found to be crucial for deriving the relevant interaction potentials that treat solvation and entropic contributions implicitly. A significant correlation between experimental binding affinities and computed score was found for sets of diverse protein-ligand complexes and for sets of different ligands bound to the same target. For 77 protein-ligand complexes taken from the Brookhaven Protein Data Bank, the calculated score showed a standard deviation from observed binding affinities of 1.8 log Ki units and an R2 value of 0.61. The best results were obtained for the subset of 16 serine protease complexes with a standard deviation of 1.0 log Ki unit and an R2 value of 0.86. A set of 33 inhibitors modeled into a crystal structure of HIV-1 protease yielded a standard deviation of 0.8 log Ki units from measured inhibition constants and an R2 value of 0.74. In contrast to empirical scoring functions that show similar or sometimes better correlation with observed binding affinities, our method does not involve deriving specific parameters that fit the observed binding affinities of protein-ligand complexes of a given training set. We compared the performance of the PMF score, Böhm's score (LUDI), and the SMOG score for eight different test sets of protein-ligand complexes. It was found that for the majority of test sets the PMF score performs best. The strength of the new approach presented here lies in its generality as no knowledge about measured binding affinities is needed to derive atomic interaction potentials. The use of the new scoring function in docking studies is outlined.

The effect of protein relaxation on charge-charge interactions and dielectric constants of proteins.

The effect of the reorganization of the protein polar groups on charge-charge interaction and the corresponding effective dielectric constant (epsilon(eff)) is examined by the semimicroscopic version of the Protein Dipole Langevin Dipoles (PDLD/S) method within the framework of the Linear Response Approximation (LRA). This is done by evaluating the interactions between ionized residues in the reaction center of Rhodobacter sphaeroides, while taking into account the protein reorganization energy. It is found that an explicit consideration of the protein relaxation leads to a significant increase in epsilon(eff) and that semimicroscopic models that do not take this relaxation into account force one to use a large value for the so-called "protein dielectric constant," epsilon(p), of the Poisson-Boltzmann model or for the corresponding epsilon(in) in the PDLD/S model. An additional increase in epsilon(eff) is expected from the reorganization of ionized residues and from changes in the degree of water penetration. This finding provides further support for the idea that epsilon(in) (or epsilon(p)) represents contributions that are not considered explicitly. The present study also provides a systematic illustration of the nature of epsilon(eff), supporting our previously reported view that charge-charge interactions correspond to a large value of this "dielectric constant," even in protein interiors. It is also pointed out that epsilon(eff) for the interaction between ionizable groups in proteins is very different from the effective dielectric constant, epsilon'(eff), that determines the free energy of ion pairs in proteins (epsilon'(eff) reflects the effect of preoriented protein dipoles). Finally, the problems associated with the search for a general epsilon(in) are discussed. It is clarified that the epsilon(in) that reproduces the effect of protein relaxation on charge-charge interaction is not equal to the epsilon(in) that reproduces the corresponding effect upon formation of individual charges. This reflects fundamental inconsistencies in attempts to cast microscopic concepts in a macroscopic model. Thus one should either use a large epsilon(in) for charge-charge interactions and a small epsilon(in) for charge-dipole interactions or consider the protein relaxation microscopically.

Electrostatic contributions to protein-protein binding affinities: application to Rap/Raf interaction.

The challenge of evaluating absolute binding free energies of protein-protein complexes is addressed using the scaled Protein Dipoles Langevin Dipoles (PDLD/S) model in combination with the Linear Response Approximation (LRA). This is done by taking the complex between Rap1A (Rap) and the p21ras binding domain of c-Raf (Raf-RBD) (Nassar et al., Nature 375:554-560, 1995) as a model system. Several formulations and different thermodynamic cycles are explored taking advantage of the LRA method and considering the protein reorganization during complex formation. The performance of different approximations is examined by comparing the calculated and observed absolute binding energies for the native complex and some of its mutants. The evaluation of the contributions of individual residues to the binding free energy, which is referred to here as group contributions is also examined. Special attention is paid to the role of the "dielectric constant," epsilon(in) which is in fact a scaling factor that represents the contributions that are treated implicitly. It is found that explicit consideration of protein relaxation is crucial for obtaining reasonable results with small values of epsilon(in), but it is also found that such a treatment of protein-protein interactions is very challenging and does not always give stable results. This indicates that more advanced explicit calculations should be based on experimentally determined structures of both the complex and the isolated proteins. Nevertheless, it is demonstrated that the qualitative trend of the effect of mutations can be reproduced by considering the effect of protein reorganization implicitly, using epsilon(in) approximately 25 for ionized residues and epsilon(in) approximately 4 for polar residues. Thus, it is concluded that an explicit treatment of solvent relaxation (which is common to current continuum models) does not provide sufficient compensation for turning off the charges of ionized residues on the interaction surface of the Raf-RBD/Rap complex. Representing the missing contribution by large epsilon(in) can, of course, reproduce the observed effect of ionized residues, but now the contribution of uncharged residues will be largely underestimated. Regardless of these conceptual problems, it is established that a very simple nonrelaxed approach, where the relaxation of both the protein and the solvent are considered implicitly, can provide an effective qualitative way for evaluating group contributions, using large and small values for epsilon(in) of ionized and neutral residues, respectively. As much as the actual system studied is concerned we find that more residues than generally assumed play a role in Raf-RBD/Rap interaction. This includes residues that are not located at the protein-protein interaction surface. These residues contribute to the binding energy through direct charge-charge interaction without leading to drastic structural changes. The overall contribution of the surface residues is quite significant since Raf and Rap are positively and negatively charged, respectively, and their charges are distributed along the interaction site between the two proteins.

A fast estimate of electrostatic group contributions to the free energy of protein-inhibitor binding.

Dissecting ligand-protein binding free energies in individual contributions of protein residues (which are referred to here as 'group contributions') is of significant importance. For example, such contributions could help in estimating the corresponding mutational effects and in studies of drug resistance problems. However, the meaning of group contributions is not always uniquely defined and the approximations for rapid estimates of such contributions are not well developed. In this paper, the nature of group contributions to binding free energy is examined, focusing particularly on electrostatic contributions which are expected to be well behaved. This analysis examines different definitions of group contributions; the 'relaxed' group contributions that represent the change in binding energy upon mutation of the given residue to glycine, and the 'non-relaxed' group contributions that represent the scaled Coulomb interaction between the given residue and the ligand. Both contributions are defined and evaluated by the linear response approximation (LRA) of the PDLD/ S method. The present analysis considers the binding of pepstatin to endothiapepsin and 23 of its mutants as a test case for a neutral ligand. The 'non-relaxed' group contributions of 15 endothiapepsin residues show significant peaks in the 'electrostatic fingerprint'. The residues that contribute to the electrostatic fingerprint are located in the binding site of endothiapepsin. They include the aspartic dyad (Asp32, Asp215) with adjacent residues and the flap region. Twelve of these 15 residues have a heavy atom distance of <3.75 A to pepstatin. The contributions of 8 (10) of these 12 residues can be reconciled with the calculated 'relaxed' group contributions where one allows the protein and solvent (solvent only) to relax upon mutation of the given residue to glycine. On the other hand, it was found that residues at the second 'solvation shell' can have relaxed contributions that are not captured by the non-relaxed approach. Hence, whereas residues with significant non-relaxed electrostatic contributions are likely to contribute to binding, residues with small non-relaxed contributions may still affect the binding energy. At any rate, it is established here that even in the case of uncharged inhibitors it is possible to use the non-relaxed electrostatic fingerprint to detect 'hot' residues that are responsible for binding. This is significant since some versions of the non-relaxed approximation are faster by several orders of magnitude than more rigorous approaches. The general applicability of this approach is outlined, emphasizing its potential in studies of drug resistance where it is crucial to have a rapid way of anticipating the effect of mutation on both drug binding and catalysis.

Shift of the special pair redox potential: electrostatic energy computations of mutants of the reaction center from Rhodobacter sphaeroides.

Shifts of the special pair redox potential of the photosynthetic reaction center of Rhodobacter sphaeroides are considered for several point mutations [Lin. X., Murchison, H. A., Nagarijan, V., Parson, W. W., Allen, J. P., & Williams, J. C. (1994) Proc. Natl. Acad. Sci. U.S.A. 91, 10265-10269] in the neighborhood of the special pair. The shifts are calculated from electrostatic energies by solving Poisson's equation for energy-minimized structures of the reaction center. Different conditions for the evaluation of the electrostatic energy are probed. To test the influence of the hydrogen bonding at the acetyl groups of the special pair, the orientation and torsion potential of the acetyl groups are varied. The calculated shifts of the midpoint potential of double and triple mutants can approximately be obtained from the corresponding shifts of the single point mutations. The calculated shifts agree with the measured values for all single and double mutants considered. However, a clear decision between different acetyl group conformations was only possible for the mutants HF(L168) and HF(L168) + LH(L131) where the calculated shifts of the redox potential agree with experiments only if the acetyl oxygen atom at DM points toward the Mg2+ ion of DL. This is corroborated by computations of the interaction energy of the acetyl group at DM, which adopts a lower value in the wild-type reaction center if its oxygen atom is bonded to the Mg2+ ion of DL.

Electrostatic control of GTP and GDP binding in the oncoprotein p21ras.

BACKGROUND: p21ras is one of the GTP-binding proteins that act as intercellular molecular switches. The GTP-bound form of p21ras sends a growth-promoting signal that is terminated once the protein is cycled back into its GDP-bound form. The interaction of guanine-nucleotide-exchange factors (GEFs) with p21ras leads to activation of the protein by promoting GDP --> GTP exchange. Oncogenic mutations of p21ras trap the protein in its biological active GTP-bound form. Other mutations interfere with the activity of GEF. Thus, it is important to explore the structural basis for the action of different mutations. RESULTS: The crystal structures of p21ras are correlated with the binding affinities of GTP and GDP by calculating the relevant electrostatic energies. It is demonstrated that such calculations can provide a road map to the location of 'hot' residues whose mutations are likely to change functional properties of the protein. Furthermore, calculations of the effect of specific mutations on GTP and GDP binding are consistent with those observed. This helps to analyze and locate functionally important parts of the protein. CONCLUSIONS: Our calculations indicate that the protein main chain provides a major contribution to the binding energies of nucleotides and probably plays a key role in relaying the effect of GEF action. Analysis of p21ras mutations in residues that are important for the proper function of GEFs suggests that the region comprising residues 62-67 in p21ras is the major GEF-binding site. This analysis and our computer simulations indicate that the effect of GEF is probably propagated to the P-loop (residues 10-17) through interaction between Gly60 and Gly12. This then reduces the interaction between the main-chain dipoles of the P-loop and the nucleotide. Finally, the results also suggest a possible relationship between the GTP --> GDP structural transition and the catalytic effect of the GTPase-activating protein.

Coexpression of HER-2/neu and p53 is associated with a shorter disease-free survival in node-positive breast cancer patients.

Breast cancer tissue was examined for overexpression of HER-2/neu and p53 oncogene proteins. Samples from 105 breast cancer patients were investigated by Western-blot analysis and their relationship to other established markers and clinical outcome was examined. In 21.0% of the cases HER-2/neu was overexpressed, and in 46.7% the p53 protein level was increased. Expression of these two oncogene products was closely correlated. Overexpression of both oncogenes was associated with larger tumour size and negative hormone receptor. The percentage of HER-2/neu and p53 overexpression was higher in node-positive patients, although statistical evaluation was not significant. While overexpression of HER-2/neu as well as p53 in node-positive patients was associated insignificantly with shorter disease-free survival, a significant difference could be documented when the disease-free survival of patients with overexpression of both oncogene proteins was compared to that of patients with no overexpression.