Technologies for High-Throughput Identification of Antibiotic Mechanism of Action

Abstract

There are two main strategies for antibiotic discovery: target-based and phenotypic screening. The latter has been much more successful in delivering first-in-class antibiotics, despite the major bottleneck of delayed Mechanism-of-Action (MOA) identification. Although finding new antimicrobial compounds is a very challenging task, identifying their MOA has proven equally challenging. MOA identification is important because it is a great facilitator of lead optimization and improves the chances of commercialization. Moreover, the ability to rapidly detect MOA could enable a shift from an activity-based discovery paradigm towards a mechanism-based approach. This would allow to probe the grey chemical matter, an underexplored source of structural novelty. In this study we review techniques with throughput suitable to screen large libraries and sufficient sensitivity to distinguish MOA. In particular, the techniques used in chemical genetics (e.g., based on overexpression and knockout/knockdown collections), promoter-reporter libraries, transcriptomics (e.g., using microarrays and RNA sequencing), proteomics (e.g., either gel-based or gel-free techniques), metabolomics (e.g., resourcing to nuclear magnetic resonance or mass spectrometry techniques), bacterial cytological profiling, and vibrational spectroscopy (e.g., Fourier-transform infrared or Raman scattering spectroscopy) were discussed. Ultimately, new and reinvigorated phenotypic assays bring renewed hope in the discovery of a new generation of antibiotics.

Keywords: Mechanism-of-Action (MOA); antibiotic discovery; bacterial cytological profiling; chemical genetics; high-throughput screening; metabolomics; phenotypic screening; proteomics; transcriptomics; vibrational spectroscopy.

Figure 1

Representation of the chemical space (blue), including relevant subsets, such as those have undesirable characteristics for antibiotics (red), those that induce some level of phenotypic modulation but are not detectable with traditional screening (grey), and easily identified compounds with high-antimicrobial activity (green).

Figure 2

High-throughput screening assays based on genome-wide (A) or promoter-wide (B) bacterial mutant libraries that are tested against a chemical library. Ideally, the total number of genes (g) or promoters (p) considered in a screening campaign are minimized to reduce the workload, although this often yields less information. On the other hand, the diversity of all the compounds that constitute the chemical library (c) should be increased, as this improves the likelihood of discovering novelty.

Figure 3

Two commonly employed approaches to predict antibiotic MOA. These are typically based on the high-throughput screening of a chemical library against a high dimension mutant bacteria library, or against a smaller set of model bacteria. Accordingly, different types of assays are applied, which generate specific outputs that are often complementary.