Skip to main content

Hit to lead optimisation of novel anti-TB scaffolds through an academic-industrial partnership

Final Report Summary - COOPERA-TB (Hit to lead optimisation of novel anti-TB scaffolds through an academic-industrial partnership)

A summary description of the project objectives:

Often associated with poverty, tuberculosis (TB) not only remains rampant in many parts of Africa and Asia but is now also returning to many developed nations. Associated with this resurgence in the number of cases of TB is a growing resistance against currently used antibiotics, which is now one of the most pressing problems affecting the global TB epidemic. Treating TB infections currently requires a drug cocktail, given for at least six months, and extending to two years for infections with multi-drug resistant strains of Mycobacterium tuberculosis, the agent causing TB. A largely successful campaign by the World Health Organisation (WHO) to stem the global rise of TB has not diminished the urgent need for new and better drugs and cellular targets to alleviate the devastating impact of this disease.

To meet these challenges, CooperaTB has established a PhD programme for four Early-Stage Researchers (ESRs) in the area of TB drug discovery. This focused training and research programme seeks to train two biochemists/biologists and two chemists in TB drug discovery. The four ESRs will spend the first 18 months of their PhD registration at The University of Birmingham (UoB), U.K. followed by 18 months at the Tres Cantos campus of GlaxoSmithKline, Diseases of the Developing World (GSK DDW) in Madrid, Spain. In their individual, yet interconnected, research projects, the four ESRs will work on the hit-to-lead optimisation of inhibitors of two drug targets, M. tuberculosis DprE1/2 and AspRS, building on preliminary work done at Birmingham in collaboration with GSK DDW. Upon successful completion of the programme, the ESRs will be awarded PhD degree titles by The University of Birmingham and be equipped with the skills and expertise to play a part in seeking solutions to tackling the growing problem of TB and antibiotic resistance more generally.

A description of the work performed since the beginning of the project:

All four ESRs began their projects at UoB in August 2014 (Month 7). All four were appointed as employees of UoB and enrolled as PhD students of the University, two (ESRs 1 and 3) within the School of Biosciences and two (ESRs 2 and 4) within the School of Chemistry. Although all four ESRs spent over half of their fellowships at GSK DDW in Spain, they remained employees of UoB for the duration of their appointments.

Having completed the first 16 months of their PhD registration at UoB, in December 2015 (M23), all four ESRs transferred to the Industrial Project Partner site at GSK DDW in Tres Cantos, Madrid, Spain. After 18 months at GSK DDW, two of the ESRs returned to UoB to focus on their PhD dissertations. Owing to the direction that their projects had taken, the other two ESRs opted to remain at GSK DDW for a further two months (spending 20 months in total with the industrial partner).

With four ESRs (two chemists and two biologists) and two protein targets, ESRs 1 and 2 have focused on the protein pair, DprE1, a flavin-containing oxidoreductase, and DprE2, an NADH-dependent reductase. These two enzymes work together and are responsible for isomerising decaprenylphosphoryl-beta-D-ribose (DPR) to decaprenylphosphoryl-beta-D-arabinose (DPA). ESR1 has focused on the biology aspects of these enzyme-catalysed processes whilst ESR2 has undertaken a Structure Activity Relationship (SAR) study of a DprE2 inhibitor that was discovered from a high-throughput phenotypic screening campaign undertaken by project partner, GSK DDW. With a similar division of labour, ESRs 3 and 4 focused their efforts on a new TB drug target, AspRS, with ESR3 concentrating on the underlying biology and ESR4 on hit-to-lead optimisation of a hit molecule that had also been discovered from a high-throughput phenotypic screening campaign carried out at GSK DDW.

A description of the main results achieved so far:

ESR1 Szilvia Tóth. PhD project title: Development of new tools for screening and characterisation of lead compounds for TB therapy directed at decaprenylphosphoryl-beta-D-arabinose synthesis in Mycobacterium tuberculosis.

The sugar donor DPA, used by all membrane-embedded arabinosyltransferases, is essential for M. tuberculosis arabinogalactan and lipoarabinomannan AG biosynthesis. DPA is formed from decaprenylphosphoryl-beta-D-ribose (DPR) in a two-step epimerisation reaction, catalysed by Mt-DprE1 (oxidation at C-2) and Mt-DprE2 (reduction at C-2). Benzothiazinones (BTZ) and other compounds inhibit the conversion of DPR to DPA, specifically by inhibiting Mt-DprE1.

Szilvia has focused primarily on developing a biochemical assay for the DprE2 target. She has investigated a number of approaches to reduce the tendency for this protein to aggregate. Of those studied, only His-SUMO-DprE2 currently produces enough protein for meaningful investigation in size-exclusion chromatography and enzymatic assays. Co-expression of DprE2 with DprE1 also provides active protein but in low concentrations when expressed from the same plasmid. Even these best preparations of DprE2 still contain substantial amounts of aggregated particles.

Rv3789 (GtrA), which lies immediately upstream of Mt-DprE1 in the chromosome, has been proposed in conflicting reports to be either a DPA flippase and/or a scaffold protein that recruits arabinosyltransferases to the site of DPA biosynthesis. We have shown that only by co-expressing Mt-DprE1HIS-tagged and Mt-DprE2 on a pET-duet plasmid alongside pET23-Mt-GtrA in E. coli BL21, are we able to purify soluble and active Mt-DprE2 by virtue of its affinity for Mt-DprE1. We have recently developed a robust in vitro assay to measure Mt-DprE2 enzymatic activity directly. In brief, geranylgeranyl-phospho-betaβ-D-ribose (GGPR) is converted to geranylgeranyl-phospho-2'-keto-β-D-arabinose (GGPX) by excess Mt-DprE1. Mt-DprE2 activity is directly assayed by measuring the reduction of NADPH. Interestingly, NADPH (rather than NADH) is the preferred cofactor for the reduction of GGPX to geranylgeranyl-phospho-betaβ-D-arabinose (GGPA). These new data provide two key findings: (i) Mt-DprE1 and Mt-DprE2 display a strong in vitro protein-protein interaction, and (ii) Mt-GtrA is necessary for the correct folding of Mt-DprE2. Whilst DprE1 inhibitors have been successfully tested, the DprE2 inhibitor, which has been the focus of SAR investigations by ESR 2 Giacomo Chiodarelli, is inactive in this in vitro assay. Our current hypothesis is that this compound (a 2-nitrofuran) is a pro-drug and needs to be pre-treated with a reductase to release the active DprE2 inhibitor.

Szilvia has also performed a high-throughput whole-cell assay of GSK's TB compound library using a DprE2 over-expressing M. bovis BCG strain, and identified 340 compounds, which have been submitted to more detailed testing in dose-response experiments in order to identify new, selective DprE2 inhibitors.

ESR2 Giacomo Chiodarelli. PhD project title: Development of new synthetic inhibitors against decaprenylphosphoryl--D-arabinose synthesis.

DprE1 and DprE2 are both essential enzymes for Mtb DPA synthesis. We have focused on DprE2, which has received far less attention than DprE1. The BCG-active hit, GSK3731247A, and some relevant BCG-inactive derivatives of this hit, prepared at UoB, were shipped to GSK for testing in H37Rv Mtb, and for determination of physicochemical properties. An SAR study around the hydrazide of the hit GSK3731247A, started in Birmingham, was completed at GSK with the synthesis and testing of two monoacyl hydrazide derivatives, and two derivatives presenting an amide instead of the hydrazide. One of these amide derivatives proved to be six times more potent than the starting hit. A set of 2-nitrofuran amides with ideal predicted physicochemical properties were therefore prepared and tested. Unfortunately, all of these derivatives proved inactive. It was later shown that the potent amide derivative of the starting hit is not DprE2-selective. A series of derivatives of a structurally related hit, GSK347301A, with modifications to the naphthalene part of the molecule have also been prepared. These compounds were designed to display sufficient solubility for in vitro safety evaluation. One derivative has been progressed to in vitro safety evaluation and shown not to be a MAO-B inhibitor, which had been the main safety alert raised during the in silico evaluation of these compounds.

Giacomo postulated that the nitro functionality in his initial hit is processed in vivo to provide a nitroso metabolite, which functions as a covalent inhibitor of DprE2. To evaluate this hypothesis, he synthesised a series of derivatives possessing electrophilic substituents (aldehyde, acetals, epoxide, nitrile and Michael acceptors) in place of the nitro group. All of these derivatives proved inactive. Giacomo has since postulated an alternative mechanism of action in which the 2-nitrofuran moiety in GSK3731247A undergoes reduction by a Type I nitroreductase in analogy with the mechanism of action of the anti-trypanosomal compound Nifurtimox, where the active compound is an unsaturated open-chain nitrile metabolite. The synthesis of an analogous unsaturated open-chain nitrile derivative of GSK3731247A and of a new series of compounds with alternative linkers to the hydrazide has been a focus of study.

ESR3 Ramón Soto Garcia. PhD project title: Development of new tools for screening and characterisation of lead compounds for TB therapy directed to AspRS in Mycobacterium tuberculosis.

Ramon's work has focused on the identification and characterisation of inhibitors of Mtb AspRS, an essential enzyme involved in protein biosynthesis. Biochemical assays against the M. tuberculosis and human AspRS enzymes and whole-cell activity assays in M. bovis BCG have been used to determine the Minimum Inhibitory Concentration (MIC) of a series of analogues derived from an initial hit, which was identified from a high-throughput phenotypic screening campaign carried out at GSK DDW. In vitro profiling of a series of these compounds containing a thiazolidinone core [synthesised by Bogdan Duma (ESR4)] are implicating an embedded Michael acceptor for binding and therefore activity in both a biochemical assay and a ligand binding experiment.

Alongside this work, Ramon has developed an AspRS-target-based screening methodology with the aim of identifying new AspRS inhibitors. He has used a target-based whole-cell assay to identify compounds that would appear to inhibit AspRS based on a resistance phenotype upon target over-expression in a replicative pMV261 plasmid. Out of 11000 compounds screened, 14 have undergone validation in a biochemical assay, which Ramon has transferred to 96-well plate format. He has also worked on further miniaturising this assay to a 384-well-format for HTS applications in order to increase throughput and save on reagents. Early results have already uncovered three new and structurally distinct compounds, which display potent anti-tubercular activity and binding to AspRS.

In a parallel project Ramon has generated a number of spontaneous resistant isolates against selected compounds from the TB box set, with a view to identifying their protein target as a starting point for more detailed mode-of-action studies. Future studies will include gene over-expression or biochemical assays that could help to uncover new biological functions of unexplored pathways.

ESR4 Bogdan Duma. PhD project title: Development of new synthetic inhibitors against AspRS.

A high-throughput phenotypic screening campaign carried out by the industrial Project Partner led to the disclosure of a range of hits of which one was shown to act against AspRS. The initial hit displays poor solubility and a rhodanine core that is a known PAIN (Pan-Assay INterference) compound scaffold. Bogdan's experimental work focused first on the preparation of analogues of the initial hit against AspRS. Here, he developed robust synthetic routes to a range of analogues to explore structure-activity relationships whilst also generating analogues that display improved physicochemical properties based on in silico predictive tools.

In a second SAR study, Bogdan developed a synthesis of a second hit, which is structurally distinct from the initial compound. He successfully confirmed the inhibitory activity of this compound in an over-expressor assay against AspRS, and generated analogues, which, according to in silico studies, should display improved physicochemical properties and better safety profiles.

Given the initial hit is now unlikely to advance further owing to perceived intractable safety issues, there still remained a number of unanswered questions around its mode of action. In a third study, Bogdan therefore prepared two types of tool compound based on his AspRS inhibitor. In the first of these, he incorporated a photoactivatable diazirine function into the molecule, which he has used to generate an AspRS–hit conjugate. Mass spectrometric degradation studies are now planned to gain insight into the location of binding of this hit. His second tool compound is a biotinylated analogue of the hit, which he used in pull-down experiments to show that his hit is surprisingly (given it contains a PAIN scaffold) selective for its AspRS target.

To complement the synthesis part of his project, Bogdan also undertook training on how to analyse compounds from the GSK database, search for analogues of hits identified in HTS campaigns, interpret physicochemical properties and prioritise molecules based on their properties. And to broaden his skills base further, he helped to design the biological experiments, which used his tool compounds, and received training on how to grow bacteria and perform biology experiments in a semi-autonomous manner.

Expected final results and potential impact and use: Throughout the project, the guiding principle of this joint academic-industrial PhD programme was to provide the four ESRs with a comprehensive perspective on the drug-discovery process, with a specific focus on TB drug discovery. By the time they graduate, they will not only have acquired in-depth knowledge of drug discovery and development but will also have made significant research contributions to a globally relevant disease area. Finally, through their exposure to both academic and industrial research environments, the ESRs will have become effective advocates for future knowledge transfer and collaboration between academia and industry within Europe. The ESRs have contributed to dissemination activities in terms of poster / oral presentations to specialised and more general audiences at a wide range of meetings and workshops. A project website [] has been set up as a central contact point, which has helped to publicise this ITN project to the wider scientific community as well as to the general public. ESR4 has also set up Twitter [@Coopera_TB] and LinkedIn accounts and Pod-Casts are available on the project web-site, which we are using to publicise our own work and just as importantly, emerging news stories in the broader field of TB and antimicrobial resistance.