Supplementary methods, tables S1-S3, and supplementary figures 1-7

Structures and PC3 viability IC50 data for compounds in Figure 2.

Abstract Cancer cells are highly reliant on NAD+-dependent processes, including glucose metabolism, calcium signaling, DNA repair, and regulation of gene expression. Nicotinamide phosphoribosyltransferase (NAMPT), the rate-limiting enzyme for NAD+ salvage from nicotinamide, has been investigated as a target for anticancer therapy. Known NAMPT inhibitors with potent cell activity are composed of a nitrogen-containing aromatic group, which is phosphoribosylated by the enzyme. Here, we identified two novel types of NAM-competitive NAMPT inhibitors, only one of which contains a modifiable, aromatic nitrogen that could be a phosphoribosyl acceptor. Both types of compound effectively deplete cellular NAD+, and subsequently ATP, and produce cell death when NAMPT is inhibited in cultured cells for more than 48 hours. Careful characterization of the kinetics of NAMPT inhibition in vivo allowed us to optimize dosing to produce sufficient NAD+ depletion over time that resulted in efficacy in an HCT116 xenograft model. Our data demonstrate that direct phosphoribosylation of competitive inhibitors by the NAMPT enzyme is not required for potent in vitro cellular activity or in vivo antitumor efficacy. Mol Cancer Ther; 16(7); 1236–45. ©2017 AACR.

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.

Prioritization of compounds based on human hepatotoxicity potential is currently a key unmet need in drug discovery, as it can become a major problem for several lead compounds in later stages of the drug discovery pipeline. The authors report the validation and implementation of a high-content multiparametric cytotoxicity assay based on simultaneous measurement of 8 key cell health indicators associated with nuclear morphology, plasma membrane integrity, mitochondrial function, and cell proliferation. Compounds are prioritized by (a) computing an in vitro safety margin using the minimum cytotoxic concentration (IC(20)) across all 8 indicators and cell-based efficacy data and (b) using the minimal cytotoxic concentration alone to take into account concentration of drug in tissues. Feasibility data using selected compounds, including quinolone antibiotics, thiazolidinediones, and statins, suggest the viability of this approach. To increase overall throughput of compound prioritization, the authors have identified the higher throughput, plate reader-based CyQUANT assay that is similar to the high-content screening (HCS) assay in sensitivity of measuring inhibition of cell proliferation. It is expected that the phenotypic output from the multiparametric HCS assay in combination with other highly sensitive approaches, such as microarray-based expression analysis of toxic signatures, will contribute to a better understanding and predictivity of human hepatotoxicity potential.

High throughput screening (HTS) has become a pivotal part of drug discovery for identifying potential lead molecules for specific disease targets. The success of screening is measured by several factors including the throughput and cost of the assays, but ultimately by the number of HTS hits that progress forward through the drug discovery pipeline. Miniaturization of assays by conversion from 96- to 384- to 1536- wells to well-less platforms, and the utilization of automated equipments are methodologies implemented to increase the throughput and reduce the cost of the screen. Pooling of test compounds is also an effective, though controversial method to enhance screening throughput while reducing reagent costs. Herein, we describe specific measures implemented at Abbott Laboratories to enhance the efficiency of 10-mixture orthogonal screening, and the strategy developed to reduce the rate of false positives and eliminate false negatives. Results from a comparison study of mixtures versus single compound screening will be discussed for three distinct screening formats: a GPCR functional assay, a GPCR binding assay and a tyrosine kinase assay.

Maturation of the hepatitis C virus (HCV) polyprotein occurs by a series of proteolytic processes catalyzed by host cell proteases and the virally encoded proteases NS2 and NS3. Although several peptidomimetic inhibitors of NS3 protease have been published, only a few small molecule inhibitors have been reported. In an effort to improve screening efficiency by minimizing the spectral interference of various test compounds, a substrate that contains the longer wavelength fluorescence resonance energy transfer (FRET) pair, TAMRA/QSY-7, was devised. For the optimized substrate T-Abu-Q, with sequence Ac-Asp-Glu-Lys(TAMRA)-Glu-Glu-Abu-Ψ(COO)Ala-Ser-Lys(QSY-7)-amide, the kinetic parameters with HCV NS3 protease are Km=30μM, kcat=0.6s−1, and kcat/Km=20,100s−1M−1. We show that this substrate is suitable for inhibitor analysis and mechanistic studies so long as the substrate concentration is low enough (0.5μM) to avoid complications from high inner filter effects. The substrate is especially useful with ultra-high-density screening formats, such as microarrayed compound screening technology, because there is less spectral interference from the compounds being tested than with more traditional (EDANS/DABCYL) FRET protease substrates. The merits of the new substrate, as well as potential applications of this FRET pair to other protease substrates, are discussed.

A 384-well soft agar assay was developed to identify potential novel anticancer compounds. Normally used to detect cell transformation, the assay is used here to quantitate cell proliferation in a 3-dimensional (3-D) anchorage-independent format. HCC827 cells, which are highly sensitive to epithelial growth factor receptor (EGFR) tyrosine kinase inhibitors, were used to develop the method and a set of 9600 compounds used to validate the assay. Results were compared to a monolayer assay using the same compound set. The assay provides a robust method to discover compounds that could be missed using traditional monolayer formats.

ABSTRACT We report the discovery and characterization of a novel ribosome inhibitor (NRI) class that exhibits selective and broad-spectrum antibacterial activity. Compounds in this class inhibit growth of many gram-positive and gram-negative bacteria, including the common respiratory pathogens Streptococcus pneumoniae , Haemophilus influenzae , Staphylococcus aureus , and Moraxella catarrhalis , and are nontoxic to human cell lines. The first NRI was discovered in a high-throughput screen designed to identify inhibitors of cell-free translation in extracts from S. pneumoniae. The chemical structure of the NRI class is related to antibacterial quinolones, but, interestingly, the differences in structure are sufficient to completely alter the biochemical and intracellular mechanisms of action. Expression array studies and analysis of NRI-resistant mutants confirm this difference in intracellular mechanism and provide evidence that the NRIs inhibit bacterial protein synthesis by inhibiting ribosomes. Furthermore, compounds in the NRI series appear to inhibit bacterial ribosomes by a new mechanism, because NRI-resistant strains are not cross-resistant to other ribosome inhibitors, such as macrolides, chloramphenicol, tetracycline, aminoglycosides, or oxazolidinones. The NRIs are a promising new antibacterial class with activity against all major drug-resistant respiratory pathogens.

A strategy is described for the development of high-throughput screening assays against targets of unknown function that involves the use of nuclear magnetic resonance (NMR) spectroscopy. Using this approach, molecules that bind to the protein target are identified from an NMR-based screen of a library of substrates, cofactors, and other compounds that are known to bind to many proteins and enzymes. Once a ligand has been discovered, a fluorescent or radiolabeled analog of the ligand is synthesized that can be used in a high-throughput screen. The approach is illustrated in the development of a high-throughput screening assay against HI-0033, a conserved protein from Haemophilus influenzae whose function is currently unknown. Adenosine was found to bind to HI-0033 by NMR, and fluorescent analogs were rapidly identified that bound to HI-0033 in the submicromolar range. Using these fluorescent compounds, a fluorescence polarization assay was developed that is suitable for high-throughput screening and obtaining detailed structure-activity relationships for lead optimization.

Streptococcus pneumoniae is a major human pathogen that causes high mortality and morbidity rates and has developed resistance to many antibiotics. The genome of S. pneumoniae has recently been completely sequenced revealing many genes encoding hypothetical proteins of unknown function. We have found that the gene encoding one such conserved protein, SP14.3, is essential for growth of S. pneumonia. Since it is essential, SP14.3 represents a potential target for drug discovery. Here, we describe the three-dimensional solution structure of SP14.3 as determined by NMR spectroscopy. The structure consists of two domains each with an alpha/beta-fold. The N-terminal domain contains two alpha-helices and a three-stranded beta-sheet, while the C-terminal domain is composed of one alpha-helix and a five-stranded beta-sheet. The N-terminal domain of the protein contains a highly negatively charged surface and resembles the fold of the N-terminal domain of Thermus thermophilus ribosomal protein S3. The C-terminal domain has a protein fold similar to human small nuclear ribonucleoprotein Sm D3 and Haloarcula marismortui ribosomal protein L21E. The two domains of the protein tumble in solution overall as a whole with an overall molecular rotational correlation time (tau(m)) of 12.9 ns at 25 degrees C. The relative orientation of the two domains is not defined by the nuclear Overhauser effect data. Indeed, residual dipolar couplings and the structure calculations indicate that the relative orientation of the two domains is not rigidly oriented with respect to one another in solution.