Metabolic Fingerprinting with Fourier-Transform Infrared (FTIR) Spectroscopy: Towards a High-Throughput Screening Assay for Antibiotic Discovery and Mechanism-of-Action Elucidation

Abstract

The discovery of antibiotics has been slowing to a halt. Phenotypic screening is once again at the forefront of antibiotic discovery, yet Mechanism-Of-Action (MOA) identification is still a major bottleneck. As such, methods capable of MOA elucidation coupled with the high-throughput screening of whole cells are required now more than ever, for which Fourier-Transform Infrared (FTIR) spectroscopy is a promising metabolic fingerprinting technique. A high-throughput whole-cell FTIR spectroscopy-based bioassay was developed to reveal the metabolic fingerprint induced by 15 antibiotics on the Escherichia coli metabolism. Cells were briefly exposed to four times the minimum inhibitory concentration and spectra were quickly acquired in the high-throughput mode. After preprocessing optimization, a partial least squares discriminant analysis and principal component analysis were conducted. The metabolic fingerprints obtained with FTIR spectroscopy were sufficiently specific to allow a clear distinction between different antibiotics, across three independent cultures, with either analysis algorithm. These fingerprints were coherent with the known MOA of all the antibiotics tested, which include examples that target the protein, DNA, RNA, and cell wall biosynthesis. Because FTIR spectroscopy acquires a holistic fingerprint of the effect of antibiotics on the cellular metabolism, it holds great potential to be used for high-throughput screening in antibiotic discovery and possibly towards a better understanding of the MOA of current antibiotics.Metabolites 2020, 10, 145 2 of 15 of finding compounds with unique biological and/or chemical properties and limited insight in the pharmacological target. Additionally, phenotypic screening does not explore the chemical grey matter, i.e., compounds capable of inducing some level of phenotypic modulation, but without sufficient potency to induce cell death or growth inhibition, which can be a source of compounds suitable for lead optimization with medicinal chemistry techniques [5,6]. Antibiotic discovery is a very challenging task, but identifying the MOA has proven equally challenging [7]. Currently, determining the MOA of antibiotics is still a bottleneck of the phenotypic screening discovery process, for which metabolomics holds great potential. As such, the ability to rapidly infer MOA and, if possible, the biomolecular target of antibiotics is increasingly important given the pressing need for new antibiotics. Currently, screening hundreds of thousands of compounds is a reasonable throughput of a drug discovery program, in part due to the ease in synthetizing bioactive compounds, and in part given the increasing availability of natural product libraries [8]. Two concepts are relevant when discussing MOA identification. One is determining the molecular pathways affected by a given compound:the drug effects. The second is the specific compound-substrate interactions:the drug target [9]. Although both concepts are very important in antibiotic discovery, given the exploratory purpo...

Towards the sustainable discovery and development of new antibiotics

“…Since universally applicable methods for characterizing the MoA(s) of antibiotics do not exist, a full suite of expertise in genetics, genomics, microbiology, chemical biology and biophysics is required. Identification of the molecular target can be achieved by targeted screens of indicator or mutant strains, whole-genome sequencing upon focused resistance development 224 , 225 , pattern recognition techniques based on transcriptomics 226 , imaging 227 , 228 , metabolomics 229 , macromolecular synthesis 230 , 231 or mutant fitness profiles 232 , 233 , which can be coupled with machine learning approaches for directed predictions 225 , 233 , or chemoproteomics 234 , 235 . The latter is specifically useful in the case of non-essential target inhibitors like pathoblockers, since these may not generate resistant mutants (at least under standard laboratory conditions).…”

Fast identification of off‐target liabilities in early antibiotic discovery with Fourier‐transform infrared spectroscopy

“…FTIRS has been shown to detect antibiotic‐induced profiles that elucidate their MOA (Huleihel et al, 2009; Ribeiro da Cunha et al, 2019, 2020; Xuan Nguyen et al, 2017). In addition to the prediction of both known and simulated novel MOA, we have shown that FTIRS can be used to estimate potency and to probe the gray chemical matter (Ribeiro da Cunha et al, 2021).…”

“…Fourier‐transform infrared (FTIR) spectroscopy (FTIRS) is a technique with such characteristics that reflects the vibrational energy of molecular bonds after their interaction with infrared radiation, thereby revealing the atoms participating in said bonds, which in turn conveys the molecular composition of biological samples regarding nucleic acids, proteins, carbohydrates, lipids, phospholipids, among others (Ribeiro da Cunha et al, 2020). FTIRS conveys the molecular composition of samples with sufficient detail that it is suitable for metabolic fingerprinting in general (Goodacre et al, 2004), and specifically for identifying antibiotic‐induced profiles (Huleihel et al, 2009; Ribeiro da Cunha et al, 2019, 2020; Xuan Nguyen et al, 2017). Recently, FTIRS has also been shown to be sufficiently sensitive for screening animal DNA polymorphism (Gomes Rios et al, 2021) as well as for detecting single mtDNA deletions, which can enable the diagnosis of mitochondrial myopathy in clinical practice (Gervasoni et al, 2020).…”

Technologies for High-Throughput Identification of Antibiotic Mechanism of Action

“…Because FTIRS has been underexplored but holds great potential, we dwelled into its application towards MOA identification. Firstly, we applied a macro-cultivation assay, from which it became clear that metabolic fingerprints reflect the MOA of antibiotics [ 102 ]. We then refined our assay into a high-throughput micro-cultivation protocol, from which we successfully predicted the MOA of antibiotics at the level of the major biosynthetic pathway, class, and individual antibiotics.…”