Apparent activity in high-throughput screening: origins of compound-dependent assay interference
Introduction
High-throughput screening (HTS) involves testing hundreds of thousands, if not millions, of chemical compounds for activity in a biological assay. Originally developed as a drug discovery methodology by pharmaceutical companies for identification of chemical entities with potential therapeutic value, HTS has branched into the academic sector as a technique to discover chemical probes, which are applied as research tools to study a biological target or pathway [1, 2, 3].
Libraries used for HTS consist of chemicals with a high degree of structural heterogeneity (compared to nucleic acid or peptide libraries, for example) that often have unknown and undefined properties and biological activities. One challenge in HTS is to successfully differentiate between compounds that demonstrate genuine activity against the biological target, target family, or pathway of interest from compounds that interfere with elements of the assay format or technique. Generally, genuine activity of a compound against its target is characterized by a high affinity, non-covalent interaction that is reversible [4]. Activity in an assay due to compound interference can result in a “false positive” (see Box 1) [5••], and as compounds that are genuinely active against the target are rare (∼0.01–0.1% of library), they are easily obscured by a high incidence of false-positives in a screen (Figure 1A) [6••, 7, 8]. In this review we will focus on a discussion of compound activity that reproducibly interferes with HTS assays. Such compound interference can arise solely from the compounds themselves [5••, 9•], as is the case for fluorescent compounds (Figure 1B), or can be the result of their interaction with biological components of the assay system. The latter is the case for compounds that directly inhibit a biological reporter, such as firefly luciferase (Figure 1C), or compounds that indirectly inhibit or activate proteins in an assay, for example, by undergoing redox cycling to generate hydrogen peroxide. Non-specific chemical reactivity against target proteins, such as covalent binding or metal ion chelation [4, 10], is another source of false positives in HTS [11••, 12•, 13•], but is out of the scope of this Review and thus will not be discussed in detail here except in the context of redox reactivity.
Section snippets
Introduction to artifacts typical of light-based detection methods in HTS
Current HTS technologies rely heavily on sensitive light-based detection methods, such as fluorescence or luminescence, to quantify the effect of a compound on a target molecule or signaling pathway [5••]. While advantages to light-based detection technologies include a desirable balance between sensitivity and ease of automation for HTS, they are also susceptible to a wide-range of different types of assay interference (see Table 1) [9•]. Whereas most light-based assay interference is due to
Compound fluorescence
Whether or not a compound fluoresces in an assay depends upon its structural properties and the excitation and emission wavelengths used in the experiment. Generally, conjugated bonds within the compound confer fluorescent character, and the greater the degree of conjugation within a compound, the longer the wavelength at which it fluoresces [14••]. Compound libraries tend to contain a greater percentage of heterocyclic compounds and compounds with low levels of conjugation [5••, 15, 16], and
Compounds that inhibit firefly luciferase: concentration-dependent inhibition or activation in cell-based assays
Firefly luciferase (FLuc)-based bioluminescence assays are highly favored in HTS due to their sensitivity (extremely low endogenous background signal leads to superior signal to background ratio relative to fluorescence methods) [21] and dynamic response in cell-based reporter gene assays, due to a relatively short FLuc protein half-life [22]. FLuc itself, however, is an enzyme, and is thus susceptible to inhibition by small molecules used in screening. A profiling effort of a 70K compound
Promiscuous enzymatic inhibition in biochemical assays: compound aggregation
Compound aggregation was recently discovered to be one of the main causes for promiscuous enzyme inhibition [12•, 29, 30•, 31, 32]. Under certain conditions, above certain concentrations, some compounds self-associate to form an aggregate structure, which, at 50–400 nm in size, can be visualized by transmission electron microscopy (TEM) [31]. Evidence suggests that enzymes are sequestered on the surface of the aggregate particles, where their function is non-specifically inhibited [30•, 33].
Compounds capable of redox cycling and direct oxidation
Some compounds, such as certain quinones, undergo redox-dependent cycling (redox cycling compounds, or RCCs) in the presence of strong reducing agents such as dithiothreitol (DTT) and tris(2-carboxyethyl)phosphine (TCEP), which results in generation of reactive oxygen species (ROS) [37, 38, 39, 40••] (Supplemental Figure S2). DTT and other reducing agents are commonly added to buffer systems of enzymatic assays as a means to keep catalytic and key structural amino acids of enzymes in a reduced
Conclusion
Although compound-dependent assay interference in HTS cannot, at this time, be entirely eliminated, it is possible to significantly reduce the probability of its occurrence, in addition to making ‘off-target’ activity significantly easier to differentiate from activity that is target- and pathway-specific. Acknowledging the potential interference that a given assay may be susceptible to and designing appropriate orthogonal assays to confirm compound activity are the best means of identifying
Conflict of interest
The authors have no conflicts of interest relating to this publication.
References and recommended reading
Papers of particular interest, published within the period of review, have been highlighted as:
• of special interest
•• of outstanding interest
Acknowledgements
We would like to thank Ruili Huang for preparation and use of the SOM presented in Figure 2. We would also like to sincerely thank Anton Simeonov for data on aggregation-based inhibition used in Figure 2, as well as for critically reading the manuscript. The NIH Chemical Genomics Center is supported by funding from the Molecular Libraries Initiative of the NIH Roadmap for Medical Research.
References (62)
- et al.
HTS and hit finding in academia- from chemical genomics to drug discovery
Drug Discovery Today
(2009) Nonleadlikeness and leadlikeness in biochemical screening
Drug Discov Today
(2003)- et al.
Quantitative analyses of aggregation, autofluorescence, and reactivity artifacts in a screen for inhibitors of thiol protease
J Med Chem
(2010) - et al.
New substructure filters for removal of pan assay interference compounds (PAINS) from screening libraries and for their exclusion in bioassays
J Med Chem
(2010) - et al.
Reply to Peltz et al: Post-translational stabilization of the firefly luciferase reporter by PTC124 (Ataluren)
Proc Natl Acad Sci U S A
(2009) - et al.
Identification and prediction of promiscuous aggregating inhibitors among known drugs
J Med Chem
(2003) - et al.
Effect of detergent on “promiscuous” inhibitors
J Med Chem
(2003) - et al.
Identification of aminothienopyridazine inhibitors of tau assembly by quantitative high-throughput screening
Biochemistry
(2009) - et al.
Cellular platforms for HTS: three case studies
Drug Discov Today
(2002) - et al.
A bioluminescent cytotoxicity assay for assessment of membrane integrity using a proteolytic biomarker
Toxicol In Vitro
(2008)