Development of Orally Efficacious Allosteric Inhibitors of TNFα via Fragment-Based Drug Design.

Tumor necrosis factor α (TNFα) is a soluble cytokine that is directly involved in systemic inflammation through the regulation of the intracellular NF-κB and MAPK signaling pathways. The development of biologic drugs that inhibit TNFα has led to improved clinical outcomes for patients with rheumatoid arthritis and other chronic autoimmune diseases; however, TNFα has proven to be difficult to drug with small molecules. Herein, we present a two-phase, fragment-based drug discovery (FBDD) effort in which we first identified isoquinoline fragments that disrupt TNFα ligand-receptor binding through an allosteric desymmetrization mechanism as observed in high-resolution crystal structures. The second phase of discovery focused on the de novo design and optimization of fragments with improved binding efficiency and drug-like properties. The 3-indolinone-based lead presented here displays oral, in vivo efficacy in a mouse glucose-6-phosphate isomerase (GPI)-induced paw swelling model comparable to that seen with a TNFα antibody.

Development of a high-content screening assay panel to accelerate mechanism of action studies for oncology research.

Efficient elucidation of the biological mechanism of action of novel compounds remains a major bottleneck in the drug discovery process. To address this need in the area of oncology, we report the development of a multiparametric high-content screening assay panel at the level of single cells to dramatically accelerate understanding the mechanism of action of cell growth-inhibiting compounds on a large scale. Our approach is based on measuring 10 established end points associated with mitochondrial apoptosis, cell cycle disruption, DNA damage, and cellular morphological changes in the same experiment, across three multiparametric assays. The data from all of the measurements taken together are expected to help increase our current understanding of target protein functions, constrain the list of possible targets for compounds identified using phenotypic screens, and identify off-target effects. We have also developed novel data visualization and phenotypic classification approaches for detailed interpretation of individual compound effects and navigation of large collections of multiparametric cellular responses. We expect this general approach to be valuable for drug discovery across multiple therapeutic areas.

Identification and preliminary characterization of a potent, safe, and orally efficacious inhibitor of acyl-CoA:diacylglycerol acyltransferase 1.

A high-throughput screen against human DGAT-1 led to the identification of a core structure that was subsequently optimized to afford the potent, selective, and orally bioavailable compound 14. Oral administration at doses ≥0.03 mg/kg significantly reduced postprandial triglycerides in mice following an oral lipid challenge. Further assessment in both acute and chronic safety pharmacology and toxicology studies demonstrated a clean profile up to high plasma levels, thus culminating in the nomination of 14 as clinical candidate ABT-046.

Fragment-based lead discovery: challenges and opportunities.

Fragment-based lead discovery has undergone remarkable changes over the last 15 years. During this time, the pharmaceutical industry has changed dramatically as well, and continued evolution of the industry is assured. These changes present many challenges but also several opportunities for executing fragment-based drug design. This article will explore some of the more significant changes in the industry and how they may affect future discovery efforts related to fragment-based initiatives.

Identification and characterization of mGlu3 ligands using a high throughput FLIPR assay for detection of agonists, antagonists, and allosteric modulators.

When targeting G-protein coupled receptors (GPCRs) in early stage drug discovery, or for novel targets, the type of ligand most likely to produce the desired therapeutic effect may be unknown. Therefore, it can be desirable to identify potential lead compounds from multiple categories: agonists, antagonists, and allosteric modulators. In this study, we developed a triple addition calcium flux assay using FLIPR Tetra to identify multiple ligand classes for the metabotropic glutamate receptor 3 (mGlu3), using a cell line stably co-expressing the human G-protein-coupled mGlu3 receptor, a promiscuous G-protein (G(α16)), and rat Glast, a glutamate transporter. Compounds were added to the cells followed by stimulation with EC(10) and then EC(80) concentration of glutamate, the physiological agonist for mGlu receptors. This format produced a robust assay, facilitating the identification of agonists, positive allosteric modulators and antagonists/negative allosteric modulators. Follow up experiments were conducted to exclude false positives. Using this approach, we screened a library of approximately 800,000 compounds using FLIPR Tetra and identified viable leads for all three ligand classes. Further characterization revealed the selectivity of individual ligands.

Navigating the kinome.

Although it is increasingly being recognized that drug-target interaction networks can be powerful tools for the interrogation of systems biology and the rational design of multitargeted drugs, there is no generalized, statistically validated approach to harmonizing sequence-dependent and pharmacology-dependent networks. Here we demonstrate the creation of a comprehensive kinome interaction network based not only on sequence comparisons but also on multiple pharmacology parameters derived from activity profiling data. The framework described for statistical interpretation of these network connections also enables rigorous investigation of chemotype-specific interaction networks, which is critical for multitargeted drug design.

A unified, probabilistic framework for structure- and ligand-based virtual screening.

We present a probabilistic framework for interpreting structure-based virtual screening that returns a quantitative likelihood of observing bioactivity and can be quantitatively combined with ligand-based screening methods to yield a cumulative prediction that consistently outperforms any single screening metric. The approach has been developed and validated on more than 30 different protein targets. Transforming structure-based in silico screening results into robust probabilities of activity enables the general fusion of multiple structure- and ligand-based approaches and returns a quantitative expectation of success that can be used to prioritize (or deprioritize) further discovery activities. This unified probabilistic framework offers a paradigm shift in how docking and scoring results are interpreted, which can enhance early lead-finding efforts by maximizing the value of in silico computational tools.

Discovery and characterization of non-ATP site inhibitors of the mitogen activated protein (MAP) kinases.

Inhibition of protein kinases has validated therapeutic utility for cancer, with at least seven kinase inhibitor drugs on the market. Protein kinase inhibition also has significant potential for a variety of other diseases, including diabetes, pain, cognition, and chronic inflammatory and immunologic diseases. However, as the vast majority of current approaches to kinase inhibition target the highly conserved ATP-binding site, the use of kinase inhibitors in treating nononcology diseases may require great selectivity for the target kinase. As protein kinases are signal transducers that are involved in binding to a variety of other proteins, targeting alternative, less conserved sites on the protein may provide an avenue for greater selectivity. Here we report an affinity-based, high-throughput screening technique that allows nonbiased interrogation of small molecule libraries for binding to all exposed sites on a protein surface. This approach was used to screen both the c-Jun N-terminal protein kinase Jnk-1 (involved in insulin signaling) and p38α (involved in the formation of TNFα and other cytokines). In addition to canonical ATP-site ligands, compounds were identified that bind to novel allosteric sites. The nature, biological relevance, and mode of binding of these ligands were extensively characterized using two-dimensional (1)H/(13)C NMR spectroscopy, protein X-ray crystallography, surface plasmon resonance, and direct enzymatic activity and activation cascade assays. Jnk-1 and p38α both belong to the MAP kinase family, and the allosteric ligands for both targets bind similarly on a ledge of the protein surface exposed by the MAP insertion present in the CMGC family of protein kinases and distant from the active site. Medicinal chemistry studies resulted in an improved Jnk-1 ligand able to increase adiponectin secretion in human adipocytes and increase insulin-induced protein kinase PKB phosphorylation in human hepatocytes, in similar fashion to Jnk-1 siRNA and to rosiglitazone treatment. Together, the data suggest that these new ligand series bind to a novel, allosteric, and physiologically relevant site and therefore represent a unique approach to identify kinase inhibitors.

Discovery of a potent and selective Bcl-2 inhibitor using SAR by NMR.

The Bcl-2 family of proteins plays a major role in the regulation of apoptosis, or programmed cell death. Overexpression of the anti-apoptotic members of this family (Bcl-2, Bcl-x(L), and Mcl-1) can render cancer cells resistant to chemotherapeutic agents and therefore these proteins are important targets for the development of new anti-cancer agents. Here we describe the discovery of a potent, highly selective, Bcl-2 inhibitor using SAR by NMR and structure-based drug design which could serve as a starting point for the development of a Bcl-2 selective anti-cancer agent. Such an agent would potentially overcome the Bcl-x(L) mediated thrombocytopenia observed with ABT-263.

Rational approaches to targeted polypharmacology: creating and navigating protein-ligand interaction networks.

Many successful drugs bind to and modulate multiple targets in vivo. Successfully navigating protein-ligand polypharmacology will be a crucial and increasingly utilized component of pharmaceutical research. As publicly available databases of ligand activity values continue to grow in size and quality, infrastructure is needed to enable scientists to create and interact with these networks to fuel hypothesis-driven science. While most of the individual tools for creating this infrastructure exist, effectively connecting the data to the network to the scientist is very much a work in progress. Standards for publishing network data are also important to facilitate the analysis and comparison of networks from different research groups using different methods.

Biochemical and biophysical characterization of unique switch pocket inhibitors of p38α.

Herein we describe the identification and characterization of a class of molecules that are believed to extend into a region of p38 known as the 'switch pocket'. Although these molecules lack a canonical hinge binding motif, they show K(i) values as low as 100 nM against p38. We show that molecules that interact with this region of the protein demonstrate different binding kinetics than a canonical ATP mimetic, as well as a wide range of kinome profiles. Thus, the switch pocket presents new opportunities for kinome selectivity which could result in unique biochemical responses and offer new opportunities in the field of kinase drug discovery.

Labeled Ligand Displacement: Extending NMR-Based Screening of Protein Targets.

NMR spectroscopy has enjoyed widespread success as a method for screening protein targets, especially in the area of fragment-based drug discovery. However, current methods for NMR-based screening all suffer certain limitations. Two-dimensional methods like "SAR by NMR" require isotopically labeled protein and are limited to proteins less than about 50 kDa. For one-dimensional, ligand-based methods, results can be confounded by nonspecific compound binding, resonance overlap, or the need for a special NMR probe. We present here a ligand-based method that relies on the exchange broadening observed for a (13)C-labeled molecule upon binding to a protein target (labeled ligand displacement). This method can be used to screen both individual compounds and mixtures and is free of the artifacts inherent in other ligand-based methods.

Healthy skepticism: assessing realistic model performance.

Although the development of computational models to aid drug discovery has become an integral part of pharmaceutical research, the application of these models often fails to produce the expected impact on productivity. One reason for this may be that the expected performance of many models is simply not supported by the underlying data, because of often neglected effects of assay and prediction errors on the reliability of the predicted outcome. Another significant challenge to realizing the full potential of computational models is their integration into prospective medicinal chemistry campaigns. This article will analyze the impact of assay and prediction error on model quality, and explore scenarios where computational models can expect to have a significant influence on drug discovery research.

Structural characterization of a soluble amyloid beta-peptide oligomer.

Alzheimer's disease (AD) is a neurodegenerative disorder that is linked to the presence of amyloid beta-peptides that can form insoluble fibrils or soluble oligomeric assemblies. Soluble forms are present in the brains and tissues of Alzheimer's patients, and their presence correlates with disease progression. Long-lived soluble forms can be generated in vitro by using small amounts of aliphatic hydrocarbon chains of detergents or fatty acids in preparations of amyloid beta-peptides. Using NMR, we have characterized soluble oligomers of Abeta preglobulomer and globulomer that are stable and alter synaptic activity. The NMR data indicate that these soluble forms have a mixed parallel and antiparallel beta-sheet structure that is different from fibrils which contain only parallel beta-sheets. Using the structural data, we engineered a disulfide bond into the soluble Abeta globulomer to give a "new" soluble antigen that is stable, homogeneous, and binds with the same affinity to selective antibodies as the parent wt globulomer.

Kinase-targeted libraries: the design and synthesis of novel, potent, and selective kinase inhibitors.

Protein kinases continue to hold tremendous promise for therapeutic intervention, and the search for novel, safe and efficacious kinase inhibitors has intensified over the past decade. Given that most kinases are readily inhibited by organic small molecules and that a wealth of structural data exists on kinase-inhibitor complexes, there has been almost universal success in the design and identification of potent kinase inhibitors. The issues of non-selectivity and congested IP space, however, present formidable challenges for the successful clinical development of these compounds. We describe a systematic approach implemented at Abbott to enable the rapid discovery and design of novel and potent kinase inhibitors that provide additional opportunities for targeting new intellectual property space and achieving acceptable selectivity profiles.

Molecular determinants of species-specific activation or blockade of TRPA1 channels.

TRPA1 is an excitatory, nonselective cation channel implicated in somatosensory function, pain, and neurogenic inflammation. Through covalent modification of cysteine and lysine residues, TRPA1 can be activated by electrophilic compounds, including active ingredients of pungent natural products (e.g., allyl isothiocyanate), environmental irritants (e.g., acrolein), and endogenous ligands (4-hydroxynonenal). However, how covalent modification leads to channel opening is not understood. Here, we report that electrophilic, thioaminal-containing compounds [e.g., CMP1 (4-methyl-N-[2,2,2-trichloro-1-(4-nitro-phenylsulfanyl)-ethyl]-benzamide)] covalently modify cysteine residues but produce striking species-specific effects [i.e., activation of rat TRPA1 (rTRPA1) and blockade of human TRPA1 (hTRPA1) activation by reactive and nonreactive agonists]. Through characterizing rTRPA1 and hTRPA1 chimeric channels and point mutations, we identified several residues in the upper portion of the S6 transmembrane domains as critical determinants of the opposite channel gating: Ala-946 and Met-949 of rTRPA1 determine channel activation, whereas equivalent residues of hTRPA1 (Ser-943 and Ile-946) determine channel block. Furthermore, side-chain replacements at these critical residues profoundly affect channel function. Therefore, our findings reveal a molecular basis of species-specific channel gating and provide novel insights into how TRPA1 respond to stimuli.

Application of belief theory to similarity data fusion for use in analog searching and lead hopping.

A wide variety of computational algorithms have been developed that strive to capture the chemical similarity between two compounds for use in virtual screening and lead discovery. One limitation of such approaches is that, while a returned similarity value reflects the perceived degree of relatedness between any two compounds, there is no direct correlation between this value and the expectation or confidence that any two molecules will in fact be equally active. A lack of a common framework for interpretation of similarity measures also confounds the reliable fusion of information from different algorithms. Here, we present a probabilistic framework for interpreting similarity measures that directly correlates the similarity value to a quantitative expectation that two molecules will in fact be equipotent. The approach is based on extensive benchmarking of 10 different similarity methods (MACCS keys, Daylight fingerprints, maximum common subgraphs, rapid overlay of chemical structures (ROCS) shape similarity, and six connectivity-based fingerprints) against a database of more than 150,000 compounds with activity data against 23 protein targets. Given this unified and probabilistic framework for interpreting chemical similarity, principles derived from decision theory can then be applied to combine the evidence from different similarity measures in such a way that both capitalizes on the strengths of the individual approaches and maintains a quantitative estimate of the likelihood that any two molecules will exhibit similar biological activity.

Scaffold oriented synthesis. Part 2: Design, synthesis and biological evaluation of pyrimido-diazepines as receptor tyrosine kinase inhibitors.

We report the discovery of the pyrimido-diazepine scaffolds as novel adenine mimics. Structure-based design led to the discovery of analogs with potent inhibitory activity against receptor tyrosine kinases, such as KDR, Flt3 and c-Kit. Compound 14 exhibited low nanomolar KDR enzymatic and cellular potencies (IC(50)=9 and 52 nM, respectively).