Targeted purification development enabled by computational biophysical modeling.

Chromatographic and non-chromatographic purification of biopharmaceuticals depend on the interactions between protein molecules and a solid-liquid interface. These interactions are dominated by the protein-surface properties, which are a function of protein sequence, structure, and dynamics. In addition, protein-surface properties are critical for in vivo recognition and activation, thus, purification strategies should strive to preserve structural integrity and retain desired pharmacological efficacy. Other factors such as surface diffusion, pore diffusion, and film mass transfer can impact chromatographic separation and resin design. The key factors that impact non-chromatographic separations (e.g., solubility, ligand affinity, charges and hydrophobic clusters, and molecular dynamics) are readily amenable to computational modeling and can enhance the understanding of protein chromatographic. Previously published studies have used computational methods such as quantitative structure-activity relationship (QSAR) or quantitative structure-property relationship (QSPR) to identify and rank order affinity ligands based on their potential to effectively bind and separate a desired biopharmaceutical from host cell protein (HCP) and other impurities. The challenge in the application of such an approach is to discern key yet subtle differences in ligands and proteins that influence biologics purification. Using a relatively small molecular weight protein (insulin), this research overcame limitations of previous modeling efforts by utilizing atomic level detail for the modeling of protein-ligand interactions, effectively leveraging and extending previous research on drug target discovery. These principles were applied to the purification of different commercially available insulin variants. The ability of these computational models to correlate directionally with empirical observation is demonstrated for several insulin systems over a range of purification challenges including resolution of subtle product variants (amino acid misincorporations). Broader application of this methodology in bioprocess development may enhance and speed the development of a robust purification platform.

Molecular umbrella conjugate for the ocular delivery of siRNA.

The synthesis, computer modeling, and biological activity of an octawalled molecular umbrella short interfacing RNA (siRNA) conjugate is described. This molecular umbrella-siRNA conjugate exhibited mRNA knockdown activity in vitro in the absence of a transfection reagent. Evaluation of this molecular umbrella conjugate in vivo, using the rat eye via intravitreal injection, resulted in sequence specific mRNA knockdown in the retina with no obvious signs of toxicity, as judged by ophthalmic examination.

Potent and selective HIV-1 ribonuclease H inhibitors based on a 1-hydroxy-1,8-naphthyridin-2(1H)-one scaffold.

Optimization studies using an HIV RNase H active site inhibitor containing a 1-hydroxy-1,8-naphthyridin-2(1H)-one core identified 4-position substituents that provided several potent and selective inhibitors. The best compound was potent and selective in biochemical assays (IC(50)=0.045 µM, HIV RT RNase H; 13 µM, HIV RT-polymerase; 24 µM, HIV integrase) and showed antiviral efficacy in a single-cycle viral replication assay in P4-2 cells (IC(50)=0.19 µM) with a modest window with respect to cytotoxicity (CC(50)=3.3 µM).

Scoring of KDR kinase inhibitors: using interaction energy as a guide for ranking.

Within a congeneric series of ATP-competitive KDR kinase inhibitors, we determined that the IC(50) values, which span four orders of magnitude, correlated best with the calculated ligand-protein interaction energy using the Merck Molecular Force Field (MMFFs(94)). Using the ligand-protein interaction energy as a guide, we outline a workflow to rank order virtual KDR kinase inhibitors prior to synthesis. When structural information of the target is available, the ability to score molecules a priori can be used to rationally select reagents. Our implementation allows one to select thousands of readily available reagents, enumerate compounds in multiple poses and score molecules in the active site of a protein within a few hours. In our experience, virtual library enumeration is best used when a correlation between computed descriptors/properties and IC(50) or K (i) values has been established.

Reagent Selector: using Synthon Analysis to visualize reagent properties and assist in combinatorial library design.

Reagent Selector is an intranet-based tool that aids in the selection of reagents for use in combinatorial library construction. The user selects an appropriate reagent group as a query, for example, primary amines, and further refines it on the basis of various physicochemical properties, resulting in a list of potential reagents. The results of this selection process are, in turn, converted into synthons: the fragments or R-groups that are to be incorporated into the combinatorial library. The Synthon Analysis interface graphically depicts the chemical properties for each synthon as a function of the topological bond distance from the scaffold attachment point. Displayed in this fashion, the user is able to visualize the property space for the universe of synthons as well as that of the synthons selected. Ultimately, the reagent list that embodies the selected synthons is made available to the user for reagent procurement. Application of the approach to a sample reagent list for a G-protein coupled receptor targeted library is described.

Web enabling technology for the design, enumeration, optimization and tracking of compound libraries.

Motivated by the need to augment Merck's in-house small molecule collection, web-based tools for designing, enumerating, optimizing and tracking compound libraries have been developed. The path leading to the current version of this Virtual Library Tool Kit (VLTK) is discussed in context of the (then) available commercial offerings and the constraints and requirements imposed by the end users. Though the effort was initiated to simplify the tasks of designing novel, drug-like and diverse compound libraries containing between 2K-10K unique entities, it has also evolved into a powerful tool for outsourcing syntheses as well as lead identification and optimization. The web tool includes components that select reagents, analyze synthons, identify backup reagents, enumerate libraries, calculate properties, optimize libraries and finally track the synthesized compounds through biological assays. In addition to accommodating project specific designs and virtual 3D library scanning, the application includes tools for parallel synthesis, laboratory automation and compound registration.

Similarity to molecules in the training set is a good discriminator for prediction accuracy in QSAR.

How well can a QSAR model predict the activity of a molecule not in the training set used to create the model? A set of retrospective cross-validation experiments using 20 diverse in-house activity sets were done to find a good discriminator of prediction accuracy as measured by root-mean-square difference between observed and predicted activity. Among the measures we tested, two seem useful: the similarity of the molecule to be predicted to the nearest molecule in the training set and/or the number of neighbors in the training set, where neighbors are those more similar than a user-chosen cutoff. The molecules with the highest similarity and/or the most neighbors are the best-predicted. This trend holds true for narrow training sets and, to a lesser degree, for many diverse training sets and does not depend on which QSAR method or descriptor is used. One may define the similarity using a different descriptor than that used for the QSAR model. The similarity dependence for diverse training sets is somewhat unexpected. It appears to be greater for those data sets where the association of similar activities vs similar structures (as encoded in the Patterson plot) is stronger. We propose a way to estimate the reliability of the prediction of an arbitrary chemical structure on a given QSAR model, given the training set from which the model was derived.

Nonpeptide alphavbeta3 antagonists. Part 11: discovery and preclinical evaluation of potent alphavbeta3 antagonists for the prevention and treatment of osteoporosis.

3-(S)-Pyrimidin-5-yl-9-(5,6,7,8-tetrahydro-[1,8]naphthyridin-2-yl)-nonanoic acid (5e) and 3-(S)-(methylpyrimidin-5-yl)-9-(5,6,7,8-tetrahydro-[1,8]naphthyridin-2-yl)-nonanoic acid (5f) were identified as potent and selective antagonists of the alpha(v)beta(3) receptor. These compounds have excellent in vitro profiles (IC(50) = 0.07 and 0.08 nM, respectively), significant unbound fractions in human plasma (6 and 4%), and good pharmacokinetics in rat, dog, and rhesus monkey. On the basis of the efficacy shown in an in vivo model of bone turnover following once-daily oral administration, these two compounds were selected for clinical development for the treatment of osteoporosis.

Molecular model of the alpha(IIb)beta(3) integrin.

A molecular model of the alpha(IIb)beta(3) integrin has been developed utilizing (i). the crystal structure of alpha(v)beta(3), (ii). homology model of the alpha(IIb) subdomain, and (iii). the docking of alpha(IIb)beta(3)/alpha(v)beta(3) dual and selective inhibitors into the putative binding sites of alpha(IIb)beta(3) and alpha(v)beta(3). Since the binding sites of these integrins are located at the interface between the two heads of the individual subunits, only the alpha(IIb)beta(3) head region is modeled. The 3D conformations of two loops in alpha(IIb), whose residues have been implicated in non-peptide ligand binding, could not be determined from homology with alpha(v) alone. Mutagenesis data and the modeling of small ligand binding contributed to the rational design of these loop conformations. The final energy minimized loop conformations exhibit permissible phi/psi angles and contribute to a binding site model of alpha(IIb)beta(3) that is consistent with both the known mutagenesis studies and in-house structure-activity relationships. The charged residues alpha(IIb):E117 and beta(3):R214 are found to dominate the ligand-protein binding interaction. The previously identified "exosite" is also identified as a hydrogen bond, hydrophobic or pi-pi interaction with Y190, similar to the recently proposed binding model of alpha(v)beta(3).

Random forest: a classification and regression tool for compound classification and QSAR modeling.

A new classification and regression tool, Random Forest, is introduced and investigated for predicting a compound's quantitative or categorical biological activity based on a quantitative description of the compound's molecular structure. Random Forest is an ensemble of unpruned classification or regression trees created by using bootstrap samples of the training data and random feature selection in tree induction. Prediction is made by aggregating (majority vote or averaging) the predictions of the ensemble. We built predictive models for six cheminformatics data sets. Our analysis demonstrates that Random Forest is a powerful tool capable of delivering performance that is among the most accurate methods to date. We also present three additional features of Random Forest: built-in performance assessment, a measure of relative importance of descriptors, and a measure of compound similarity that is weighted by the relative importance of descriptors. It is the combination of relatively high prediction accuracy and its collection of desired features that makes Random Forest uniquely suited for modeling in cheminformatics.

Binding model for nonpeptide antagonists of alpha(v)beta(3) integrin.

A binding model for nonpeptide antagonists of integrin alpha(v)beta(3) has been developed through docking analyses utilizing the MMFFs force field and the recently published crystal structure, 1JV2. Results of this docking study have led to the identification of a novel binding model for selective antagonists of alpha(v)beta(3) over alpha(IIb)beta(3) integrins. Four different chemical classes are shown to bind in a similar fashion providing a measure of confidence in the proposed model. All alpha(v)beta(3) and alpha(IIb)beta(3) antagonists have a basic nitrogen separated some distance from a carboxylic acid to mimic RGD. For the alpha(v)beta(3) antagonists under present consideration, these charged ends are separated by twelve bonds. The basic nitrogen of the active alpha(v)beta(3) ligands are shown to interact with D150 of alpha(v) and the ligands' carboxylic acid interact with R214 of beta(3) while adopting an extended conformation with minimal protein induced internal strain. In addition, an energetically favorable interaction is found with all of the active alpha(v)beta(3) molecules with Y178 of alpha(v) when docked to the crystallographically determined structure. This novel interaction may be characterized as pi-pi stacking for the most active of the alpha(v)beta(3) selective antagonists. The proposed model is consistent with observed activity as well as mutagenicity and photoaffinity cross-linking studies of the alpha(v)beta(3) integrin.